article
stringlengths 0
745k
| abstract
stringlengths 0
92.7k
|
|---|---|
a charge placed in a polarizable medium is screened .
dielectric theory describes the phenomenon by the induction of a polarization around the charge carrier .
the induced polarization will follow the charge carrier when it is moving through the medium .
the carrier together with the induced polarization is considered as one entity ( see fig.[fig_scheme1 ] ) .
it was called a _ polaron _ by l. d. landau @xcite .
the physical properties of a polaron differ from those of a band - carrier .
a polaron is characterized by its _ binding ( or self- ) energy _
@xmath0 , an _ effective mass _ @xmath1 and by its characteristic _ response _ to external electric and magnetic fields ( e. g. dc mobility and optical absorption coefficient ) .
if the spatial extension of a polaron is large compared to the lattice parameter of the solid , the latter can be treated as a polarizable continuum .
this is the case of a _ large ( frhlich ) _ polaron .
when the self - induced polarization caused by an electron or hole becomes of the order of the lattice parameter , a _ smll ( holstein ) _ polaron can arise . as distinct from large polarons ,
small polarons are governed by short - range interactions .
[ [ the - polaron - radius .- large - polarons - vs - small - polarons ] ] the polaron radius .
large polarons vs small polarons + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + consider the lo phonon field with frequency @xmath2 interacting with an electron .
denote by @xmath3 the quadratic mean square deviation of the electron velocity . in the electron - phonon interaction
is weak , the electron can travel a distance @xmath4 during a time @xmath5characteristic for the lattice period , because it is the distance within which the electron can be localized using the phonon field as measuring device . from the uncertainty
relations it follows @xmath6 at weak coupling @xmath7 is a measure of the polaron radius @xmath8 . to be consistent
, the polaron radius @xmath8 must be considerably larger than the lattice parameter @xmath9(this is a criterion of a large polaron ) .
experimental evaluation of the polaron radius leads to the follwing typical values : @xmath10 for alkali halides , @xmath11 for silver halides , @xmath12 for ii - vi , ii - v semiconductors .
the continuum approximation is not satisfied for transition metal oxides ( nio , cao , mno ) , in other oxides ( uo@xmath13nbo@xmath14 ) . for those solids the small polaron concept is used . in some substances ( e.g. perovskites ) some intermediate region between large and small polarons
is realized .
[ [ the - coupling - constant ] ] the coupling constant @xcite + + + + + + + + + + + + + + + + + + + + + + + + + + + + + consider the case of _ strong electron - phonon interaction _ in a polar crystal .
the electron of mass @xmath15 is then localized and can - to a first approximation - be considered as a static charge distribution within a sphere with radius @xmath16 the medium is characterized by an average dielectric constant @xmath17which will be defined below .
the potential energy of a sphere of radius @xmath18uniformly charged with the charge @xmath19 in a vacuum is ( see eq . ( 8.6 ) of ref .
@xcite)@xmath20 the potential energy of a uniformly charged sphere in a medium with the high - frequency dielectric constant @xmath21 is@xmath22 this is the potential energy of the _ self - interaction _ of the charge @xmath19 uniformly spread over the sphere of radius @xmath18 in a medium with the dielectric constant @xmath21 . in a medium with an inertial polarization field ( due to lo phonons ) , the potential energy of the uniformly charged sphere is@xmath23 where @xmath24 is the static dielectric constant .
the polaron effect is then related to the change of the potential energy of the interaction of the charged sphere due to the inertial polarization field .
this change is the potential energy @xmath25 of the uniformly charged sphere in the presence of the inertial polarization field minus the potential energy of the _ self - interaction _ @xmath26of the charge @xmath19 uniformly spread over the sphere in a medium without the inertial polarization:@xmath27 with@xmath28 the electron distribution in a sphere may be non - uniform , what may influence the numerical coefficient in eqs .
( [ u1 ] ) to ( [ upol ] ) . in this connection
one can use the estimate @xcite@xmath29 the restriction of the electron in space requires its de broglie wave length to be of the order @xmath30so that its kinetic energy is of the order @xmath31minimizing the total energy with respect to @xmath18leads to @xmath32 wherefrom the binding energy is @xmath33 for _ weak coupling _ , one can neglect the kinetic energy of the electron .
taking the polaron radius according to ( [ i2 ] ) , @xmath34 the binding energy is@xmath35 we note that @xmath36 following the conventions of the field theory , the self energy at weak coupling is written as @xmath37 therefore the so - called frhlich polaron coupling constant is @xmath38 for the average dielectric constant one shows that @xmath39 where @xmath21 and @xmath24 are , respectively , the electronic and the static dielectric constant of the polar crystal .
the difference @xmath40 arises because the ionic vibrations occur in the infrared spectrum and the electrons in the shells can follow the conduction electron adiabatically .
[ [ polaron - mobility ] ] polaron mobility + + + + + + + + + + + + + + + + here we give a simple derivation leading to the gross features of the mobility behaviuor , especially its temperature dependence .
the key idea is that the mobility will change because the number of phonons in the lattice , with which the polaron interacts , is changing with temperature .
the phonon density is given by @xmath41 the mobility for large polaron is proportional to the inverse of the number of phonons : @xmath42 and for low temperatures @xmath43 @xmath44 the mobility of continuum poarons decreases with increasing temperature following an exponential law .
the slope of the straight line in @xmath45 vs @xmath46 is characterized by the lo phonon frequesncy .
systematic study performed , in particular , by frhlich and kadanoff , gives @xmath47 the small polaron will jump from ion to ion under the influence of optical phonons .
the lerger the numver of phonons , the lerger the mobility .
the behaviuor of the small polaron is the opposite of that of the large polaron .
one expects : @xmath48 for low temperatures @xmath43 one has : @xmath49 the mobility of small polaron is thermally activated .
systematic analysis within the small - polaron theory shows that @xmath50 with @xmath51 frhlich proposed a model hamiltonian for the large polaron through which its dynamics is treated quantum mechanically ( frhlich hamiltonian ) .
the polarization , carried by the longitudinal optical ( lo ) phonons , is represented by a set of quantum oscillators with frequency @xmath52 , the long - wavelength lo - phonon frequency , and the interaction between the charge and the polarization field is linear in the field @xcite:@xmath53 where @xmath54 is the position coordinate operator of the electron with band mass @xmath55 , @xmath56 is its canonically conjugate momentum operator ; @xmath57 and @xmath58 are the creation ( and annihilation ) operators for longitudinal optical phonons of wave vector @xmath59 and energy @xmath60 .
the @xmath61 are fourier components of the electron - phonon interaction @xmath62 the strength of the electron
phonon interaction is expressed by a dimensionless coupling constant @xmath63 , which is defined as : @xmath64 in this definition , @xmath21 and @xmath24 are , respectively , the electronic and the static dielectric constant of the polar crystal . in table
[ table1 ] the frhlich coupling constant is given for a few solids are not well established . ] .
[ c]@llllllmaterial & @xmath63 & ref . & material & @xmath63 & ref .
+ insb & 0.023 & @xcite & agcl & 1.84 & @xcite + inas & 0.052 & @xcite & ki & 2.5 & @xcite + gaas & 0.068 & @xcite & tlbr & 2.55 & @xcite + gap & 0.20 & @xcite & kbr & 3.05 & @xcite + cdte & 0.29 & @xcite & bi@xmath65sio@xmath66 & 3.18 & @xcite + znse & 0.43 & @xcite & cdf@xmath67 & 3.2 & @xcite + cds & 0.53 & @xcite & kcl & 3.44 & @xcite + @xmath63-al@xmath67o@xmath68 & 1.25 & @xcite & csi & 3.67 & @xcite + agbr & 1.53 & @xcite & srtio@xmath68 & 3.77 & @xcite + @xmath63-sio@xmath67 & 1.59 & @xcite & rbcl & 3.81 & @xcite + in deriving the form of @xmath61 , expressions ( [ eq_1b ] ) and ( [ eq_1c ] ) , it was assumed that ( i ) the spatial extension of the polaron is large compared to the lattice parameter of the solid ( continuum approximation ) , ( ii ) spin and relativistic effects can be neglected , ( iii ) the band - electron has parabolic dispersion , ( iv ) in line with the first approximation it is also assumed that the lo - phonons of interest for the interaction , are the long - wavelength phonons with constant frequency @xmath52 .
the model , represented by the hamiltonian ( [ eq_1a ] ) ( which up to now could not been solved exactly ) has been the subject of extensive investigations , see , e. g. , refs . @xcite . in what follows the key approaches of the frhlich - polaron theory
are briefly reviewed with indication of their relevance for the polaron problems in nanostructures . here
some insight will be given in the type of transformation that might be useful to study the frhlich hamiltonian ( [ eq_1a ] ) .
for this purpose the hamiltonian will be treated for a particle with infinite mass @xmath69 ( which is at @xmath70@xmath71 which can be transformed into the following expression with shifted phonon operators:@xmath72 to determine the eigenstates of this hamiltonian , one can perform a unitary transformation which produces the following shift of the phonon operators:@xmath73 the transformation@xmath74 \label{s}\ ] ] is canonical : @xmath75 = s^{-1}\ ] ] and has the desired property:@xmath76 the transformed hamiltonian is now:@xmath77 the eigenstates of the hamiltonian contain an integer number of phonons @xmath78the eigenenergies are evidently:@xmath79 this expression is divergent at it is often the case in field theory of point charges are considered .
a transformation of the type @xmath80 has been of great interest in developing weak coupling theory as shown below . for actual crystals
@xmath63-values typically range from @xmath81 ( insb ) up to @xmath82 to @xmath83 ( alkali halides , some oxides ) , see table 1 .
a weak - coupling theory of the polaron was developed originally by frhlich @xcite .
he derived the first weak - coupling perturbation - theory results : @xmath84 and @xmath85 expressions ( [ eq_5a ] ) and ( [ eq_5b ] ) are rigorous to order @xmath63 .
inspired by the work of tomonaga on quantum electrodynamics ( q. e. d. ) , lee , low and pines ( llp ) @xcite analyzed the properties of a weak - coupling polaron starting from a formulation based on canonical transformations ( cp .
the results of the subsection [ sec - shift]).as hown by them , the unitary transformation@xmath86 where @xmath87 is a `` c''-number representing the _ total system momentum _ allows to eliminate the electron co - ordinates from the system .
intuitively one might guess this transformation by writing the exact wave function in the form@xmath88 it is plausible that the bloch factor @xmath89 attaches the system to the electron as origin of the co - cordinates .
after this transformation the hamiltonian ( [ eq_1a ] ) becomes:@xmath90 if , for the sake of simplicity , the case of total momentum equal to zero is considered , this expression becomes:@xmath91 the first term of this hamiltonian is the correlation energy term involving different values for @xmath59 and @xmath92if one diagonalizes the second and the trird term of the hamiltonian ( [ hcorr ] ) ( this can be done exactly by means of the `` shifted - oscillator canonical transformation '' @xmath80 ( [ s ] ) ) , the result of llp is found .
the expectation value of the first term is zero for the wave function @xmath93 therefore one is sure to obtain a variational result .
it is remarkable that merely extracting the @xmath94 term from the expression@xmath95 eliminates the divergency from the problem ( cp . with the case @xmath96 ) and is equivalent to the sophisticated theory by lee , low and pines ( llp ) , which corresponds thus to neglect of the term ( [ corr ] ) .
the details of the llp theory are given in appendix 1 .
the explicit form for the energy is now@xmath97 this self energy is no longer divergent .
the divergence is elmininated by the quantum cut - off occurring at @xmath98 for the self energy the llp result is equivalent to the perturbation result .
the effective mass however is now given by@xmath99 a result , which follows if one considers the case @xmath100 and which is also exact for @xmath101 .
however , the llp effective mass is different from the perturbation result if @xmath63 insreases . the llp approximation has often been called intermediate - coupling approximation .
however its range of validity is the same as that of perturbation theory to order @xmath63 .
the significance of the llp approximation consists of the flexibility of the canonical transformations together with the fact that it puts the frhlich result on a variational basis . to order @xmath102 , the analytical expressions for the coefficients are @xmath102 : @xmath103 for the energy and @xmath104 for the polaron mass @xcite . at present
the following weak - coupling expansions are known : for the energy @xcite @xmath105 and for the polaron mass @xcite @xmath106 historically , the strong coupling limit was studied before all other treatments ( landau , pekar @xcite ) .
although it is only a formal case because the actual crystals seems to have @xmath63 values smaller than 5 , it is very interesting because it contains some indication of the intermediate coupling too : approach the excitations from the strong coupling limit and extrapolate to intermediate coupling is interesting because it is expected that some specific strong coupling properties survive at intermediate coupling . in
what follows , a treatment , equivalent to that of pekar , but in second quantization and written with as much analogy to the llp treatment as possible is given .
we start from the frhlich hamiltonian ( [ eq_1a ] ) . at strong
coupling one makes the assumption ( a produkt - ansatz ) for the polaron wave - function@xmath107 where @xmath108 is the electron - component of the wave function ( @xmath109the field - component of the wave function @xmath110 ( @xmath111 parametrically depends on @xmath112 .
the produkt - ansatz ( [ eq_2a ] ) or born - oppenheimer approximation implies that the electron adiabatically follows the motion of the atoms , while the field can not follow the instantaneous motion of the electron .
frhlich showed that the approximation ( [ eq_2a ] ) leads to results , which are only valid for sufficiently large @xmath113 , i. e. in the strong - coupling regime .
a more systematic analysis of strong - coupling polarons based on canonical transformations applied to the hamiltonian ( [ eq_1a ] ) was performed in refs .
@xcite .
the expectation value for the energy is now:@xmath114 \left\vert f\right\rangle\ ] ] with @xmath115 we wish to minimize @xmath116 but also @xmath117 \left\vert f\right\rangle\ ] ] has to be minimized .
this expression will be minimized if @xmath118 is the ground state wave function of the shifted oscullator - type hamiltonian .
as we can diagonalize this hamiltonian exactly:@xmath119 we can apply a canonical transformation similar to ( [ s]):@xmath120 , \ ] ] which has the property : @xmath121 the transformed hamiltonian is now : @xmath122
s\\ & = \sum_{\mathbf{k}}\hbar\omega_{\mathrm{lo}}a_{\mathbf{k}}^{\dag } a_{\mathbf{k}}-\sum_{\mathbf{k}}\frac{\left\vert v_{k}\right\vert ^{2}\left\vert \rho_{\mathbf{k}}\right\vert ^{2}}{\hbar\omega_{\mathrm{lo}}}.\end{aligned}\ ] ] the phonon vacuum @xmath123 provides a minimum:@xmath124 s\left\vert 0\right\rangle = -\sum_{\mathbf{k}}\frac{\left\vert v_{k}\right\vert ^{2}\left\vert \rho_{\mathbf{k}}\right\vert ^{2}}{\hbar\omega_{\mathrm{lo}}}.\ ] ] hence , the hamiltonian ( [ sc1 ] ) is minimized by the ground state wave function@xmath125 \left\vert 0\right\rangle . \label{scwf}\ ] ] it gives the ground state energy@xmath126 which is still a functional of @xmath127 .
the functionals @xmath128 are different for differerent excitations .
[ [ ground - state - of - strong - coupling - polarons ] ] ground state of strong - coupling polarons + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + for the ground state one considers a gaussian wave function : @xmath129 with a variational parameter @xmath130.@xmath131
^{3}\\ & = c^{2}\left ( \frac{\sqrt{\pi}}{\frac{m_{b}\omega_{0}}{\hbar}}\right ) ^{3}=c^{2}\left ( \frac{\pi\hbar}{m_{b}\omega_{0}}\right ) ^{3/2}=1\rightarrow c^{2}=\left ( \frac{m_{b}\omega_{0}}{\pi\hbar}\right ) ^{3/2}\]]@xmath132 for the further use , we introduce a notation @xmath133 such a wave function is consistent with the localization of the electron , which we expect for large @xmath63 . the kinetic energy in ( [ sc2 ] ) for this function
is calculated using the representation of the operator @xmath134 @xmath135:@xmath136@xmath137 \\ & = 3\frac{\hbar^{2}}{2m_{b}}c_{1}^{2}\frac{m_{b}\omega_{0}}{\hbar}\int_{-\infty}^{\infty}dx\left ( 1-\frac{m_{b}\omega_{0}}{\hbar}x^{2}\right ) \exp\left ( -\frac{m_{b}\omega_{0}}{\hbar}x^{2}\right ) \\ & = = 3\frac{\hbar^{2}}{m_{b}}c_{1}^{2}\frac{m_{b}\omega_{0}}{\hbar}\int _ { 0}^{\infty}dx\left ( 1-\frac{m_{b}\omega_{0}}{\hbar}x^{2}\right ) \exp\left ( -\frac{m_{b}\omega_{0}}{\hbar}x^{2}\right ) \\ & = 3\frac{\hbar^{2}}{m_{b}}c_{1}^{2}\frac{m_{b}\omega_{0}}{\hbar}\left [ \frac{\sqrt{\pi}}{2\sqrt{\frac{m_{b}\omega_{0}}{\hbar}}}-\frac{m_{b}\omega _ { 0}}{\hbar}\frac{\sqrt{\pi}}{4\sqrt{\left ( \frac{m_{b}\omega_{0}}{\hbar } \right ) ^{3}}}\right ] \\ & = 3\frac{\hbar\omega_{0}}{4}c_{1}^{2}\sqrt{\frac{\pi\hbar}{m_{b}\omega_{0}}}=3\frac{\hbar\omega_{0}}{4}\sqrt{\frac{m_{b}\omega_{0}}{\pi\hbar}}\sqrt { \frac{\pi\hbar}{m_{b}\omega_{0}}}=\frac{3}{4}\hbar\omega_{0}.\end{aligned}\ ] ] the functional@xmath138 \right ) \nonumber\\ & = c^{2}\exp\left ( -\frac{m_{b}\omega_{0}}{\hbar}\frac{\hbar^{2}k^{2}}{4m_{b}^{2}\omega_{0}^{2}}\right ) \int d^{3}r\exp\left ( -\frac{m_{b}\omega_{0}}{\hbar}\left [ \mathbf{r}+i\frac{\hbar}{2m_{b}\omega_{0}}\mathbf{k}\right ] ^{2}\right ) \rightarrow\nonumber\\ \rho_{\mathbf{k}"1s " } & = \exp\left ( -\frac{\hbar k^{2}}{4m_{b}\omega_{0}}\right ) . \label{scrho}\ ] ] the second term in ( [ sc2 ] ) is then @xmath139 the variational energy ( [ sc2 ] ) thus becomes@xmath140 putting @xmath141 one obtains@xmath142@xmath143 substituting ( [ sc4 ] ) in ( [ sc3 ] ) , we find the ground state energy of the polaron @xmath0 ( calculated with the energy of the uncoupled electron - phonon system as zero energy ) : @xmath144@xmath145 the strong - coupling mass of the polaron , resulting again from the approximation ( [ eq_2a ] ) , is given @xcite as : @xmath146 more rigorous strong - coupling expansions for @xmath0 and @xmath1 have been presented in the literature @xcite : @xmath147@xmath148 the strong - coupling ground state energy ( [ sc5 ] ) is lower than the llp ground state energy for @xmath149 [ [ the - excited - states - of - the - polaron - ss - fc - res ] ] the excited states of the polaron : ss , fc , res + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + in principle , excited states of the polaron exist at all coupling . in the general case , and for simplicity for * * @xmath150 , a continuum of states starts at @xmath60 above the ground state of the polaron .
this continuum physically corresponds to the scattering of free phonons on the polaron .
those scattering.states ( ss ) were studied in @xcite anf for the first time more generally in @xcite are not the only excitations of the polaron .
there are also _ internal _ excitation states corresponding to the excitations of the electron in the potential it created itself .
by analogy with the excited states of colour centers , the following terminology is used .
\(i ) the states where the electron is excited in the potential belonging to the ground state configuration of the lattice are called _ franck - condon _ ( fc ) states \(ii ) excitations of the electron in which the lattice polarization is adapted to the electronic configuration of the excited electron ( which itself then adapts its wave function to the new potential , etc . leading to a self - consistent final state ) , are called _ relaxed excited state _ ( res ) @xcite . [ [ calculation - of - the - lowest - fc - state ] ] calculation of
the lowest fc state + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + the formalism used until now is well adapted to treat the polaron excitations at strong coupling .
the field dependence of the wave function is ( [ scwf ] ) . for the fc state
the @xmath128 are the same as for the ground state ( [ scrho ] ) .
physically @xmath128 tells us , to what electronic distribution the field is adapted .
the electronic part of the excited wave function is @xmath151-like:@xmath152 with a parameter @xmath153 , which is equal to @xmath154@xmath155 ^{2}\\ & = c_{"2p"}^{2}\left ( \frac{\sqrt{\pi}}{\frac{m_{b}\omega_{p}}{\hbar}}\right ) ^{2}\frac{\sqrt{\pi}}{2\left ( \frac{m_{b}\omega_{p}}{\hbar}\right ) ^{3/2}}=c_{"2p"}^{2}\left ( \frac{\pi\hbar}{m_{b}\omega_{p}}\right ) ^{3/2}\frac{\hbar}{2m_{b}\omega_{p}}=1\rightarrow\\ c_{"2p"}^{2 } & = \left ( \frac{m_{b}\omega_{p}}{\pi\hbar}\right ) ^{3/2}\frac{2m_{b}\omega_{p}}{\hbar}\]]@xmath156 we introduce still a notation @xmath157 the fc state energy is , similarly to ( [ sc2]),@xmath158 the kinetic energy term is @xmath159 \nonumber\\ & = -\frac{\hbar^{2}}{2m_{b}}c_{_{"2p"}}^{2}\int_{-\infty}^{\infty}dzz\exp\left ( -\frac{m_{b}\omega_{p}}{2\hbar}z^{2}\right ) \nabla_{z}^{2}\left [ z\exp\left ( -\frac{m_{b}\omega_{p}}{2\hbar}z^{2}\right ) \right ] \nonumber\\ & \times\int_{-\infty}^{\infty}dy\exp\left ( -\frac{m_{b}\omega_{p}}{\hbar } y^{2}\right ) \int_{-\infty}^{\infty}dz\exp\left ( -\frac{m_{b}\omega_{p}}{\hbar}z^{2}\right ) + \frac{2}{4}\hbar\omega_{p}\nonumber\\ & = -\frac{\hbar^{2}}{2m_{b}}c_{2}^{2}\int_{-\infty}^{\infty}dzz\exp\left ( -\frac{m_{b}\omega_{p}}{2\hbar}z^{2}\right ) \nonumber\\ & \times\nabla_{z}\left [ \left ( 1-\frac{m_{b}\omega_{p}}{\hbar}z^{2}\right ) \exp\left ( -\frac{m_{b}\omega_{p}}{2\hbar}z^{2}\right ) \right ] + \frac{1}{2}\hbar\omega_{p}\nonumber\\ & = \frac{\hbar^{2}}{2m_{b}}c_{2}^{2}\int_{-\infty}^{\infty}dzz\exp\left ( -\frac{m_{b}\omega_{p}}{2\hbar}z^{2}\right ) \nonumber\\ & \times\left [ \frac{2m_{b}\omega_{p}}{\hbar}z+\left ( 1-\frac{m_{b}\omega_{p}}{\hbar}z^{2}\right ) \frac{m_{b}\omega_{p}}{\hbar}z\right ] \exp\left ( -\frac{m_{b}\omega_{p}}{2\hbar}z^{2}\right ) \nonumber\\ + \frac{1}{2}\hbar\omega_{p } & = \frac{\hbar^{2}}{2m_{b}}c_{2}^{2}\int_{-\infty}^{\infty}dz\left ( 3-\frac{m_{b}\omega_{p}}{\hbar}z^{2}\right ) \frac{m_{b}\omega_{p}}{\hbar}z^{2}\exp\left ( -\frac{m_{b}\omega_{p}}{\hbar } z^{2}\right ) \nonumber\\ + \frac{1}{2}\hbar\omega_{p } & = \frac{\hbar\omega_{p}}{2}c_{2}^{2}\left [ \frac{3\sqrt{\pi}}{2\left ( \frac{m_{b}\omega_{p}}{\hbar}\right ) ^{3/2}}-\frac{m_{b}\omega_{p}}{\hbar}\frac{3\sqrt{\pi}}{4\left ( \frac{m_{b}\omega_{p}}{\hbar}\right ) ^{5/2}}\right ] + \frac{1}{2}\hbar\omega _ { p}\nonumber\\ & = \frac{\hbar\omega_{p}}{2}c_{2}^{2}\frac{3\sqrt{\pi}}{4\left ( \frac { m_{b}\omega_{p}}{\hbar}\right ) ^{3/2}}+\frac{1}{2}\hbar\omega_{p}\nonumber\\ & = \frac{\hbar\omega_{p}}{2}\frac{2}{\sqrt{\pi}}\left ( \frac{m_{b}\omega _ { p}}{\hbar}\right ) ^{3/2}\frac{3\sqrt{\pi}}{4\left ( \frac{m_{b}\omega_{p}}{\hbar}\right ) ^{3/2}}+\frac{1}{2}\hbar\omega_{p}\nonumber\\ & = \frac{3}{4}\hbar\omega_{p}+\frac{1}{2}\hbar\omega_{p}=\frac{5}{4}\hbar\omega_{p}. \label{sckin}\ ] ] for the fc state , @xmath160 the second term in ( [ sc6 ] ) is precisely ( [ scfield ] ) , @xmath161 and the fc energy ( [ sc6 ] ) becomes@xmath162 the energy of the lowest fc state is , within the produkt - ansatz @xcite : @xmath163 the fact that this energy is positive , is presumably due to the choice of a harmonic potential . the real potential the electron sees is anharmonic , and a bound state may be expected . [
[ calculation - of - res ] ] calculation of res + + + + + + + + + + + + + + + + + + the electronic part of the excited wave function is ( [ scwf2p ] ) with a variational parameter @xmath164which is determined below .
the variational res energy is , similarly to ( [ sc2 ] ) , @xmath165 here the kinetic energy term is given by eq .
( [ sckin ] ) .
the functional , which is now needed , is @xmath166 \right ) \\ & = c_{"2p"}^{2}\ exp\left ( -\frac{m_{b}\omega_{p}}{\hbar}\frac{\hbar ^{2}k^{2}}{4m_{b}^{2}\omega_{p}^{2}}\right ) \int d^{3}rz^{2}\exp\left ( -\frac{m_{b}\omega_{p}}{\hbar}\left [ \mathbf{r}+i\frac{\hbar}{2m_{b}\omega_{p}}\mathbf{k}\right ] ^{2}\right ) \rightarrow\\end{aligned}\]]@xmath167 ^{2}\right ) \nonumber\\ & = .\exp\left ( -\frac{\hbar k^{2}}{4m_{b}\omega_{p}}\right ) c_{2}^{2}\int_{-\infty}^{\infty}dz\left ( z - i\frac{\hbar}{2m_{b}\omega_{p}}k_{z}\right ) ^{2}\exp\left ( -\frac{m_{b}\omega_{p}}{\hbar}z^{2}\right ) \nonumber\\ & = \exp\left ( -\frac{\hbar k^{2}}{4m_{b}\omega_{p}}\right ) c_{2}^{2}\int_{-\infty}^{\infty}dz\left [ z^{2}-\left ( \frac{\hbar}{2m_{b}\omega_{p}}k_{z}\right ) ^{2}\right ] ^{2}\exp\left ( -\frac{m_{b}\omega_{p}}{\hbar } z^{2}\right ) \nonumber\\ & = \exp\left ( -\frac{\hbar k^{2}}{4m_{b}\omega_{p}}\right ) c_{2}^{2}\left\ { \frac{\sqrt{\pi}}{2\left ( \frac{m_{b}\omega_{p}}{\hbar}\right ) ^{3/2}}-\left ( \frac{\hbar}{2m_{b}\omega_{p}}k_{z}\right ) ^{2}\frac { \sqrt{\pi}}{\left ( \frac{m_{b}\omega_{p}}{\hbar}\right ) ^{1/2}}\right\ } \nonumber\\ & = \left ( 1-\frac{\hbar k_{z}^{2}}{2m_{b}\omega_{p}}\right ) \exp\left ( -\frac{\hbar k^{2}}{4m_{b}\omega_{p}}\right ) .
\label{sc8}\ ] ] further , we substitute ( [ sc8 ] ) in the second term in the r.h.s . of eq .
( [ sc7a]):@xmath168\end{aligned}\ ] ] @xmath169 \\ & = -\frac{\alpha\hbar}{\sqrt{\pi}}\sqrt{\omega_{\mathrm{lo}}\omega_{p}}\frac{60 - 20 + 9}{60}=-\frac{49}{60}\frac{\alpha\hbar}{\sqrt{\pi}}\sqrt { \omega_{\mathrm{lo}}\omega_{p}}\ ] ] the variational energy ( [ sc7a ] ) becomes@xmath170 putting @xmath171 one obtains@xmath172 @xmath173 the energy of the res is ( see refs .
@xcite ) : @xmath174 the effective mass of the polaron in the res is given @xcite as : @xmath175 the structure of the energy spectrum of the strong - coupling polaron is shown in fig.[fig_statesa ] . the ground state , @xmath176 the ( first ) relaxed excited state ; the franck - condon states ( @xmath177 ) .
in fact , both the franck - condon states and the relaxed excited states lie in the continuum and , strictly speaking , are resonances . ]
the significance of the strong - coupling large polaron theory is formal only : it allows to test all - coupling theories in the limit @xmath178 .
remarkably , the effective electron - phonon coupling strength significantly increases in systems of low dimension and low dimensionality .
feynman developed a superior all - coupling polaron theory using his path - integral formalism@xcite .
he studied first the self - energy @xmath0 and the effective mass @xmath1 of polarons@xcite .
feynman got the idea to formulate the polaron problem into the lagrangian form of quantum mechanics and then eliminate the field oscillators , in exact analogy to q. e. d. ( resulting in ) a sum over all trajectories .
the resulting path integral ( here limited to the ground - state properties ) is of the form ( ref .
@xcite ) : @xmath179 , \label{eq_8a}\ ] ] where @xmath180 .
( [ eq_8a ] ) gives the amplitude that an electron found at a point in space at time zero will appear at the same point at the ( imaginary ) time @xmath181 .
this path integral ( [ eq_8a ] ) has a great intuitive appeal : it shows the polaron problem as an equivalent one - particle problem in which the interaction , non - local in time or retarded , occurs between the electron and itself .
subsequently feynman showed how the variational principle of quantum mechanics could be adapted to the path - integral formalism and he introduced a quadratic trial action ( non - local in time ) to simulate ( [ eq_8a ] ) . applying the variational principle for path integrals then results in an upper bound for the polaron self - energy at all @xmath63 , which at weak and strong coupling gives accurate expressions .
feynman obtained smooth interpolation between a weak and strong coupling ( for the ground state energy ) .
the weak - coupling expansions of feynman for the ground - state energy and the effective mass of the polaron are : @xmath182@xmath183 in the strong - coupling limit feynman found for the ground - state energy energy : @xmath184 and for the polaron mass : @xmath185 over the years the feynman model for the polaron has remained the most successful approach to this problem .
the analysis of an exactly solvable ( symmetrical ) 1d - polaron model @xcite , monte carlo schemes @xcite and other numerical schemes @xcite demonstrate the remarkable accuracy of feynman s path - integral approach to the polaron ground - state energy .
experimentally more directly accessible properties of the polaron , such as its mobility and optical absorption , have been investigated subsequently . within the path - integral approach , feynman et al . studied later the mobility of polarons@xcite .
subsequently the path - integral approach to the polaron problem was generalized and developed to become a powerful tool to study optical absorption , magnetophonon resonance and cyclotron resonance @xcite . in ref .
@xcite , a self - consistent treatment for the polaron problem at all @xmath63 was presented , which is based on the heisenberg equations of motion starting from a trial expression for the electron position .
it was used to derive the effective mass and the optical properties of the polaron at arbitrary coupling . a variational justification of the approximation used in ref .
@xcite ( through a stiltjes continuous fraction ) is reproduced in appendix 2 . in ref .
@xcite , using a monte carlo calculation , the ground - state energy of a polaron was derived as @xmath186 where @xmath187 with @xmath188 the free energy per polaron and @xmath189 \ln\left ( 2\pi\beta\right ) $ ] the free energy per electron .
the value @xmath190 , used for the actual computation in ref .
@xcite , corresponds to @xmath191 ( @xmath192 @xmath193 is the lo phonon energy ) .
so , as pointed out in ref .
@xcite , the authors of ref .
@xcite actually calculated the _ free energy _
@xmath194 , rather than the polaron _ ground - state energy_. to investigate the importance of temperature effects on @xmath195 the authors of ref .
@xcite considered the polaron energy as obtained by osaka @xcite , who generalized the feynman @xcite polaron theory to nonzero temperatures:@xmath196 \int_{0}^{\beta_{0}}du\frac{e^{-u}}{\sqrt{d\left ( u\right ) } } , \label{e1}\ ] ] where @xmath197 @xmath198 and@xmath199 this result is variational , with variational parameters @xmath200 and @xmath201 and gives an upper bound to the exact polaron free energy .
the results of a numerical - variational calculation of eq .
( [ e1 ] ) are shown in fig .
[ pd316826 ] , where the free energy @xmath202 is plotted ( in units of @xmath193 ) as a function of @xmath63 for different values of the lattice temperature . as seen from fig .
[ pd316826 ] , ( i ) @xmath202 increases with increasing temperature and ( ii ) the effect of temperature on @xmath194 increases with increasing @xmath63 .
[ h ] part1fig3.eps in table [ tablepd316826 ] , the monte carlo results @xcite , @xmath203 , are compared with the free energy of the feynman polaron , @xmath204 calculated in @xcite .
the values for the free energy obtained from the feynman polaron model are _ _ lower _ _ than the mc results for @xmath205 and @xmath206 ( but lie within the 1% error of the monte carlo results ) . since the feynman result for the polaron free energy is an upper bound to the exact result
, we conclude that for @xmath205 and @xmath206 the results of the feynman model are closer to the exact result than the mc results of @xcite .
[ t]|c|c|c|c|@xmath63 & @xmath207 & @xmath208 & @xmath209 ( % ) + @xmath210 & @xmath211 & @xmath212 & @xmath213 + @xmath214 & @xmath215 & @xmath216 & @xmath217 + @xmath218 & @xmath219 & @xmath220 & @xmath221 + @xmath222 & @xmath223 & @xmath224 & @xmath225 + @xmath226 & @xmath227 & @xmath228 & @xmath229 + @xmath230 & @xmath231 & @xmath232 & @xmath233 + @xmath234 & @xmath235 & @xmath236 & @xmath237 + @xmath238 & @xmath239 & @xmath240 & @xmath241 + the analysis of an exactly solvable ( symmetric ) 1d - polaron model was performed in refs .
the model consists of an electron interacting with two oscillators possessing the opposite wave vectors : * * * * @xmath59 and -@xmath243the parity operator , which changes @xmath58 and @xmath244 ( and also @xmath245 and @xmath246 ) , commutes with the hamiltonian of the system .
hence , the polaron states are classified into the even and odd ones with the eigenvalues of the parity operator @xmath2471 and @xmath2481 , respectively . for the lowest even and odd states , the phonon distribution functions
@xmath249 are plotted in fig . 1 , upper panel , at some values of the effective coupling constant @xmath250 of the symmetric model .
the value of the parameter@xmath251 for these graphs was taken 1 , while the total polaron momentum @xmath150 . in the weak - coupling case
( @xmath252 ) @xmath249 is a decaying function of @xmath242 .
when increasing @xmath250 , @xmath249 acquires a maximum , e.g. at @xmath253 for the lowest even state at @xmath254 .
the phonon distribution function @xmath255has the same character for the lowest even and the lowest odd states at all values of the number of the virtual phonons in the ground state .
( as distinct from the higher states ) .
this led to the conclusion that the lowest odd state is an internal excited state of the polaron . in ref @xcite , the structure of the polaron cloud
was investigated using the diagrammatic quantum monte carlo ( dqmc ) method .
in particular , partial contributions of @xmath242-phonon states to the polaron ground state were found as a function of @xmath242 for a few values of the coupling constant @xmath256 see fig .
1 , lower panel . it was shown to gradually evolve from the weak - coupling case ( @xmath257 ) into the strong - coupling regime ( @xmath258 ) .
comparion of the lower panel to the upper panel in fig .
[ npolarons ] clearly shows that the evolution of the shape and the scale of the distribution of the @xmath242-phonon states with increasing @xmath63 as derived for a large polaron within dqmc method @xcite is _ in remarkable agreement _ with the results obtained within the `` symmetric '' 1d polaron model @xcite .
[ h ] part1fig4.eps the mobility of large polarons was studied within various theoretical approaches(see ref .
@xcite for the detailed references ) .
frhlich @xcite pointed out the typical behavior of the large - polaron mobility @xmath259 which is characteristic for weak coupling . here ,
@xmath260 , @xmath261 is the temperature . within the weak - coupling regime , the mobility of the polaron
was then derived , e. g. , using the boltzmann equation in refs .
@xcite and starting from the llp - transformation in ref .
@xcite .
a nonperturbative analysis was embodied in the feynman polaron theory , where the mobility @xmath262 of the polaron using the path - integral formalism was derived by feynman et al .
( usually referred to as fhip ) as a static limit starting from a frequency - dependent impedance function . for sufficiently low temperature @xmath261
the mobility then takes the form @xcite @xmath263 where @xmath200 and @xmath264 are ( variational ) functions of @xmath63 obtained from the feynman polaron model .
using the boltzmann equation for the feynman polaron model , kadanoff @xcite found the mobility , which for low temperatures can be represented as follows : @xmath265 the weak - coupling perturbation expansion of the low - temperature polaron mobility as found using the green s function technique @xcite has confirmed that the mobility derived from the boltzmann equation is exceedingly exact for weak coupling ( @xmath266 ) and at low temperatures ( @xmath267 ) . as shown in ref .
@xcite , the mobility of eq .
( [ eq : p24 - 1 ] ) differs by the factor of @xmath268 from that derived using the polaron boltzmann equation as given by eq .
( [ eq : p24 - 1-kadanoff ] ) . in the limit of weak electron - phonon coupling and low temperature , the fhip polaron mobility of eq .
( [ eq : p24 - 1 ] ) differs by the factor of @xmath269 from the previous result @xcite , which , as pointed out in ref .
@xcite and in later publications ( see , e.g. , refs.@xcite ) , is correct for @xmath270 . as follows from this comparison , the result of ref .
@xcite is not valid when @xmath271 . as argued in ref .
@xcite and later confirmed , in particular , in ref .
@xcite the above discrepancy can be attributed to an interchange of two limits in calculating the impedance . in fhip , for weak electron - phonon coupling , one takes @xmath272 , whereas the correct order is @xmath273 ( @xmath274 is the frequency of the applied electric field ) .
it turns out that for the correct result the mobility at low temperatures is predominantly limited by the absorption of phonons , while in the theory of fhip it is the emission of phonons which gives the dominant contribution as @xmath261 goes to zero @xcite .
the analysis based on the boltzmann equation takes into account the phonon emission processes whenever the energy of the polaron is above the emission threshold .
the independent - collision model , which underlies the boltzmann - equation approach , however , fails in the strong coupling regime of the large polaron , when the thermal mean free path becomes less than the de broglie wavelength ; in this case , the boltzmann equation can not be expected to be adequate @xcite .
in fact , the expression ( [ eq : p24 - 1 ] ) for the polaron mobility was reported to adequately describe the experimental data in several polar materials ( see , e.g. , refs .
@xcite ) . experimental work on alkali halides and silver halides indicates that the mobility obtained from eq .
( [ eq : p24 - 1 ] ) describes the experimental results quite accurately @xcite .
measurements of mobility as a function of temperature for photoexcited electrons in cubic @xmath275-type bi@xmath65sio@xmath66 are explained well in terms of large polarons within the feynman approach @xcite .
the experimental findings on electron transport in crystalline tio@xmath67 ( rutile phase ) probed by thz time - domain spectroscopy are quantitatively interpreted within the feynman model @xcite .
one of the reasons for the agreement between theory based on eq .
( [ eq : p24 - 1 ] ) and experiment is that in the path - integral approximation to the polaron mobility , a maxwellian distribution for the electron velocities is assumed , when applying the adiabatic switching on of the frhlich interaction .
although such a distribution is not inherent in the frhlich interaction , its incorporation tends to favor agreement with experiment because other mechanisms ( interaction with acoustic phonons etc . )
cause a gaussian distribution .
at zero temperature and in the weak - coupling limit , the optical absorption is due to the elementary polaron scattering process , schematically shown in fig.[fig_scheme ] . in the weak - coupling limit ( @xmath276 )
the polaron absorption coefficient was first obtained by v. gurevich , i. lang and yu .
firsov @xcite , who started from the kubo formula .
their optical - absorption coefficient is equivalent to a particular case of the result of j. tempere and j. t. devreese ( ref .
@xcite ) , with the dynamic srtucture factor @xmath277 corresponding to the hartree - fock approximation ( see also @xcite , p. 585 ) .
at zero temperature , the absorption coefficient for absorption of light with frequency @xmath274 can be expressed in terms of elementary functions in two limiting cases : in the region of comparatively high polaron densities ( @xmath278 ) @xmath279 and in the low - concentration region ( @xmath280 ) @xmath281 where @xmath282 , @xmath283 is the dielectric permittivity of the vacuum , @xmath275 is the refractive index of the medium , @xmath242 is the concentration of polarons and @xmath284 is the fermi level for the electrons .
a step function @xmath285{lll}1 & \mbox{if } & \omega>1,\\ 0 & \mbox{if } & \omega<1 \end{array } \right.\ ] ] reflects the fact that at zero temperature the absorption of light accompanied by the emission of a phonon can occur only if the energy of the incident photon is larger than that of a phonon ( @xmath286 ) . in the weak - coupling limit , according to eqs.([weak ] ) , ( [ weak1 ] ) , the absorption spectrum consists of a one - phonon line . at nonzero temperature , the absorption of a photon can be accompanied not only by emission , but also by absorption of one or more phonons .
a simple derivation in ref.@xcite using a canonical transformation method gives the absorption coefficient of free polarons , which coincides with the result ( [ weak1 ] ) of ref .
@xcite .
the optical absorption of large polarons as a function of the frequency of the incident light is calculated using the canonical - transformation formalism by devreese , huybrechts and lemmens ( dhl ) ref .
@xcite . a simple calculation , which is developed below in full detail , gives a result for the absorption coefficient , which is exact to order @xmath287 we start from the hamiltonian of the electron - phonon system interacting with light is written down using the vector potential of an electromagnetic field @xmath288:@xmath289 the electric field is related to the vector potential as@xmath290 within the electric dipole interaction the electric field with frequency @xmath274 is @xmath291@xmath292 when expanding @xmath293 in the hamiltonian , we find@xmath294 where the first term is the kinetic energy of the electron , and the second term describes the interaction of the electron - phonon system with light@xmath295@xmath296 since @xmath288 does not depend on the electron coordinates , the term @xmath297 in ( [ dhl-4a ] ) does not play a role in our description of the optical absorption .
the total hamiltonian for the system of a continuum polaron interacting with light is thus@xmath298 where @xmath299 is frhlich s hamiltonian ( [ eq_1a ] ) .
the absorption coefficient for absorption of light with frequency @xmath274 by free polarons is proportional to the probability @xmath300 that a photon is absorbed by these polarons in their ground state,@xmath301 here @xmath242 is number of polarons , which are considered as independent from each other , @xmath24 is the permittivity of vacuum , @xmath302 is the velocity of light , @xmath275 is the refractive index of the medium in which the polarons move , @xmath303 is the modulus of the electric field vector of the incident photon .
if the incident light can be treated as a perturbation , the transition probability @xmath300 is given by the golden rule of fermi:@xmath304 @xmath305 is the amplitude of the time - dependent perturbation given by ( [ dhl-7 ] ) . * * the ground state wave function of a free polaron is @xmath306 and its energy is @xmath307 the wave functions of all possible final states are @xmath308 and the corresponding energies are @xmath309 .
the possible final states are all the excited states of the polaron .
the main idea of the present calculation is to avoid the explicit summation over the final polaron states , which are poorly known , by eliminating all the excited state wave functions @xmath118 from the expression ( [ dhl2a ] ) . with this aim ,
the representation of the @xmath310-function is used:@xmath311 .\ ] ] this leads to:@xmath312 \\ & = 2\operatorname{re}{\displaystyle\sum\nolimits_{f } } { \displaystyle\int\nolimits_{-\infty}^{0 } } dt\exp\left [ -i(\omega+i\varepsilon)t\right ] \left\langle \phi_{0}\left\vert v\right\vert f\right\rangle \left\langle f\left\vert e^{iht}ve^{-iht}\right\vert \phi_{0}\right\rangle .\end{aligned}\ ] ] using the fact that@xmath313 and the notation@xmath314\end{aligned}\ ] ] we find@xmath315 \left\langle \phi_{0}\left\vert v(0)v(t)\right\vert \phi_{0}\right\rangle .
\label{dhl8}\ ] ] defining@xmath316 \left\langle \phi_{0}\left\vert v(0)v(t)\right\vert \phi_{0}\right\rangle , \label{dhl-10}\ ] ] one has@xmath317 substituting ( [ dhl-7 ] ) to ( [ dhl-10 ] ) ,
we find that@xmath318 and hence@xmath319 it is convenient to apply the first llp transformation @xmath320([eq_6a ] ) , which eliminates the electron operators from the polaron hamiltonian:@xmath321 where @xmath322 can be diagonalized exactly and gives rise to the self - energy @xmath323and @xmath324 contains the correlation effects between the phonons . the optical absorption will be calculated here for the total momentum of the system @xmath325 in the llp approximation the explicit form of the matrix element in ( [ dhl-11 ] ) is@xmath326 where @xmath320and @xmath327 are the first ( [ eq_6a ] ) and the second ( [ eq_6b ] ) llp transformations .
the application of @xmath320 gives:@xmath328 using @xmath329 we arrive at @xmath330further we recall @xmath331 where @xmath332 and @xmath56 is set 0 ( see appendix 1 ) .
this results in @xmath333 then ( [ dhl15 ] ) takes the form@xmath334 here the second llp transformation is given by ( [ eq_6b ] ) with@xmath335 and the vacuum is defined by @xmath336 the calculation of the matrix element ( [ dhl17a ] ) proceeds as follows:@xmath337 further on , we calculate@xmath338 where@xmath339 s_{2}\ ] ] further we use @xmath340:@xmath341 \\ & = \sum_{\mathbf{k}}\left ( \omega_{\mathrm{lo}}+\frac{k^{2}}{2m_{b}}\right ) a_{\mathbf{k}}^{\dag}a_{\mathbf{k}}+\sum_{\mathbf{k}}\frac{\left\vert v_{k}\right\vert ^{2}}{\left ( \omega_{\mathrm{lo}}+\frac{k^{2}}{2m_{b}}\right ) } -2\sum_{\mathbf{k}}\frac{\left\vert v_{k}\right\vert ^{2}}{\left ( \omega_{\mathrm{lo}}+\frac{k^{2}}{2m_{b}}\right ) } \\ & = \sum_{\mathbf{k}}\left ( \omega_{\mathrm{lo}}+\frac{k^{2}}{2m_{b}}\right ) a_{\mathbf{k}}^{\dag}a_{\mathbf{k}}-\sum_{\mathbf{k}}\frac{\left\vert v_{k}\right\vert ^{2}}{\left ( \omega_{\mathrm{lo}}+\frac{k^{2}}{2m_{b}}\right ) } .\end{aligned}\ ] ] the last term can be calculated analytically:@xmath342 @xmath343 the term@xmath344 will be neglected:@xmath345 neglecting @xmath346 , consistent with the llp description , introduces no error in order @xmath63 . therefore ( [ dhl18a ] ) becomes@xmath347{c}\sum_{\mathbf{k}}\mathbf{e}\cdot\mathbf{k}\left ( a_{\mathbf{k}}^{\dagger } a_{\mathbf{k}}+f_{\mathbf{k}}a_{\mathbf{k}}^{\dagger}+f_{\mathbf{k}}^{\ast } a_{\mathbf{k}}+f_{\mathbf{k}}f_{\mathbf{k}}^{\ast}\right ) e^{ih_{0}t}\\ \times\sum_{\mathbf{k}}\mathbf{e}\cdot\mathbf{k}\left ( a_{\mathbf{k}}^{\dagger}a_{\mathbf{k}}+f_{\mathbf{k}}a_{\mathbf{k}}^{\dagger}+f_{\mathbf{k}}^{\ast}a_{\mathbf{k}}+f_{\mathbf{k}}f_{\mathbf{k}}^{\ast } \right ) e^{-ih_{0}t}\end{array } \right\vert 0\right\rangle . \label{dhl19}\ ] ] for * * @xmath150 there is no privileged direction and @xmath348 ( [ dhl19 ] ) reduces to:@xmath349 from the equation of motion for @xmath350@xmath351 = i\left ( \omega_{\mathrm{lo}}+\frac{k^{2}}{2m_{b}}\right ) a_{\mathbf{k}}^{\dag},\ ] ] it is easy now to calculate@xmath352 .\ ] ] the matrix element ( [ dhl17a ] ) now becomes@xmath353 + o(\alpha^{2}).\ ] ] the transition probability ( [ dhl8 ] ) is then given by the expression@xmath354 \\ & = 2\pi\frac{e^{2}}{m_{b}^{2}\omega^{2}}\sum_{\mathbf{k}}\left ( \mathbf{e}\cdot\mathbf{k}\right ) ^{2}\left\vert f_{\mathbf{k}}\right\vert ^{2}\delta\left ( \omega-\omega_{\mathrm{lo}}-\frac{k^{2}}{2m_{b}}\right ) .\end{aligned}\ ] ] using ( [ dhl17c ] ) , we obtain@xmath355 where@xmath356{cc}1 & \text{if\qquad}\frac{\omega}{\omega_{\mathrm{lo}}}>1\\ 0 & \text{if\qquad}\frac{\omega}{\omega_{\mathrm{lo}}}<1 \end{array } \right . .\ ] ] the absorption coefficient ( [ dhl1 ] ) for absorption by free polarons for @xmath357 finally takes the form@xmath358 the behaviuor of @xmath359 ( [ dhl24 ] ) as a function of @xmath274 is as follows . for @xmath360
there is no absorption .
the threshold for absorption is at @xmath361from @xmath362 up to @xmath363 increases to a maximum and for @xmath364 the absorption coefficient decreases slowly with increasing @xmath365 experimentally , this one - phonon line has been observed for free polarons in the infrared absorption spectra of cdo - films , see fig.[fig_abscdo ] . in cdo , which is a weakly polar material with @xmath366 , the polaron absorption band is observed in the spectral region between 6 and 20 @xmath262 m ( above the lo phonon frequency ) .
the difference between theory and experiment in the wavelength region where polaron absorption dominates the spectrum is due to many - polaron effects .
@xmath367 at @xmath368 k. the experimental data ( solid dots ) of ref .
@xcite are compared to different theoretical results : with ( solid curve ) and without ( dashed line ) the one - polaron contribution of ref .
@xcite and for many polarons ( dash - dotted curve ) of ref . @xcite .
the absorption of light by free large polarons was treated in ref .
@xcite using the polaron states obtained wihtin the adiabatic strong - coupling approximation , which was considered above in subsection [ strong ] .
it was argued in ref .
@xcite , that for sufficiently large @xmath63 ( @xmath369 ) , the ( first ) res of a polaron is a relatively stable state , which can participate in optical absorption transitions .
this idea was necessary to understand the polaron optical absorption spectrum in the strong - coupling regime .
the following scenario of a transition , which leads to a _ zero - phonon peak _ in the absorption by a strong - coupling polaron , can then be suggested .
if the frequency of the incoming photon is equal to @xmath370 then the electron jumps from the ground state ( which , at large coupling , is well - characterized by `` @xmath371''-symmetry for the electron ) to an excited state ( `` @xmath151 '' ) , while the lattice polarization in the final state is adapted to the `` @xmath151 '' electronic state of the polaron . in ref .
@xcite considering the decay of the res with emission of one real phonon it is demonstrated , that the zero - phonon peak can be described using the wigner - weisskopf formula valid when the linewidth of that peak is much smaller than @xmath372 for photon energies larger than @xmath373 a transition of the polaron towards the first scattering state , belonging to the res , becomes possible .
the final state of the optical absorption process then consists of a polaron in its lowest res plus a free phonon .
a one - phonon sideband then appears in the polaron absorption spectrum .
this process is called _ one - phonon sideband absorption_. the one- , two- , ...
@xmath374- , ...
phonon sidebands of the zero - phonon peak give rise to a broad structure in the absorption spectrum .
it turns that the _
first moment _ of the phonon sidebands corresponds to the fc frequency @xmath375 : @xmath376 to summarize , the polaron optical absorption spectrum at strong coupling is characterized by the following features ( @xmath377 ) : 1 .
an intense absorption peak ( `` zero - phonon line '' ) appears , which corresponds to a transition from the ground state to the first res at @xmath378 .
2 . for @xmath379 ,
a phonon sideband structure arises . this sideband structure peaks around @xmath380 .
the qualitative behaviour predicted in ref.@xcite , namely , an intense zero - phonon ( res ) line with a broader sideband at the high - frequency side , was confirmed after an all - coupling expression for the polaron optical absorption coefficient at @xmath381 had been studied @xcite . in what precedes ,
the low - frequency end of the polaron absorption spectrum was discussed ; at higher frequencies , transitions to higher res and their scattering states can appear .
the two - phonon sidebands in the optical absorption of free polarons in the strong - coupling limit were numerically studied in ref .
@xcite .
the study of the optical absorption of polarons at large coupling is mainly of formal interest because all reported coupling constants of polar semiconductors and ionic crystals are smaller than 5 ( see table 1 ) .
[ [ definitions ] ] definitions + + + + + + + + + + + we derive here the linear response of the frhlich polaron , described by the hamiltonian@xmath382 to a spatially uniform , time - varying electric field@xmath383 this field induces a current in the @xmath384-direction @xmath385 the complex function @xmath386 is called the impedance function .
the frequency - dependent mobility is defined by @xmath387 for nonzero frequencies ( in the case of polarons the frequencies of interest are in the infrared ) one defines the absorption coefficient @xcite@xmath388 where @xmath283 is the dielectric constant of the vacuum , @xmath275 the refractive index of the crystal , and @xmath302 the velocity of light . in the following the amplitude of the electric field @xmath0 is taken sufficiently small so that linear - response theory can be applied . the impedance function can be expressed via a frequency - dependent conductivity of a single polaron in a unit volume @xmath389 using _ the standard kubo formula _ ( cf .
( 3.8.8 ) from ref .
\right\rangle dt . \label{ifkubo}\ ] ] in order to introduce a convenient representation of the impedance function , we give in the next subsection a definition and discuss properties of a scalar product of two operators [ cf .
@xcite , chapter 5 ] .
[ [ definition - and - properties - of - the - scalar - product ] ] definition and properties of the scalar product + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + for two operators @xmath391 and @xmath392 ( i.e. , elements of the hilbert space of operators ) the scalar product is defined as @xmath393 the notation @xmath394 is used in order to indicate that the operator @xmath395 acts on the operator @xmath396 .
the time evolution of the operator @xmath391 is determined by the liouville operator @xmath397:@xmath398 \label{if9}\ ] ] with a commutator @xmath399 $ ] , wherefrom@xmath400 the expectation value in ( [ if8 ] ) is taken over the gibbs ensemble:@xmath401 with the equilibtium density matrix when the electric field is absent @xmath402 one can show that ( [ if8 ] ) defines a positive definite scalar product with the following properties @xmath403 \right\rangle , \label{if11c}\ ] ] and [ cf .
( 5.11 ) of @xcite ] @xmath404 [ [ demonstration - of - the - property - if11a . ] ] demonstration of the property ( [ if11a ] ) . starting from the definition ( [ if8 ] ) and using ( [ if10 ] )
, we obtain@xmath405 substituting here ( [ if10a ] ) with ( [ if10b ] ) , one finds@xmath406 /\mathrm{tr}(e^{-\beta h}).\ ] ] change of the variable @xmath407 allows us to represnt this integral as@xmath408 /\mathrm{tr}(e^{-\beta h}).\ ] ] further , a cyclic permutation of the operators under the trace @xmath409 sign gives @xmath410 /\mathrm{tr}(e^{-\beta h})\\ & = { \displaystyle\int\nolimits_{0}^{\beta } } d\lambda\left\langle e^{\lambda h}be^{-\lambda h}a^{\dag}\right\rangle = { \displaystyle\int\nolimits_{0}^{\beta } } d\lambda\left\langle \left ( e^{\lambda\hbar l}b\right ) a^{\dag}\right\rangle \\ & = { \displaystyle\int\nolimits_{0}^{\beta } } d\lambda\left\langle \left [ e^{\lambda\hbar l}\left ( b^{\dag}\right ) ^{\dag}\right ] a^{\dag}\right\rangle .\end{aligned}\ ] ] according to the definition ( [ if8 ] ) , this finalizes the demonstration of ( [ if11a ] ) .
[ [ demonstration - of - the - property - if11b . ] ] demonstration of the property ( [ if11b ] ) .
starting from ( [ if11e ] ) and using ( [ if9 ] ) , we obtain@xmath411 a cyclic permutation of the operators under the average @xmath412 sign gives@xmath413 using the commutation of @xmath414 and @xmath415 , one finds@xmath416\right ) ^{\dag } e^{-\lambda h}b\right\rangle .\end{aligned}\ ] ] with the definition ( [ if9 ] ) , this gives @xmath417 b\right\rangle .\ ] ] according to the definition ( [ if8 ] ) , this finalizes the demonstration of ( [ if11b ] ) .
[ [ demonstration - of - the - property - if11c . ] ] demonstration of the property ( [ if11c ] ) .
starting from ( [ if11 g ] ) and performing a cyclic permutation of the operators under the average @xmath418 we find @xmath419 further we notice that @xmath420 consequently , @xmath421 /\mathrm{tr}(e^{-\beta h})\nonumber\\ & = \frac{1}{\hbar}\mathrm{tr}\left [ e^{-\beta h}a^{\dag}b - a^{\dag}e^{-\beta h}b\right ] /\mathrm{tr}(e^{-\beta h}).\end{aligned}\ ] ] further , a cyclic permutation of the operators in the second term under the trace @xmath409 sign gives@xmath422 /\mathrm{tr}(e^{-\beta h})\\ & = \frac{1}{\hbar}\mathrm{tr}\left [ e^{-\beta h}\left ( a^{\dag}b - ba^{\dag } \right ) \right ] /\mathrm{tr}(e^{-\beta h})\\ & = \frac{1}{\hbar}\left\langle a^{\dag}b - ba^{\dag}\right\rangle = \frac { 1}{\hbar}\left\langle \left [ a^{\dag},b\right ] \right\rangle .\end{aligned}\ ] ] thus , the property ( [ if11c ] ) has been demonstrated .
[ [ demonstration - of - the - property - if11d . ] ] demonstration of the property ( [ if11d ] ) .
we start from the represntation of the scalar product ( [ if11e ] ) and take a complex conjugate:@xmath423 a cyclic permutation of the operators under the average @xmath412 sign gives then@xmath424 according to the definition ( [ if8 ] ) , this finalizes the demonstration of ( [ if11d ] ) .
the above scalar product allows one to represent different dynamical quantities in a rather simple way .
for example , let us consider a scalar product @xmath425 b\right\rangle \label{if12a}\\ & = -i{\displaystyle\int\nolimits_{0}^{\infty } } dte^{izt}{\displaystyle\int\nolimits_{0}^{\beta } } d\lambda\left\langle e^{\lambda\hbar l}a^{\dag}e^{-ilt}b\right\rangle \nonumber\\ & = -i{\displaystyle\int\nolimits_{0}^{\infty } } dte^{izt}{\displaystyle\int\nolimits_{0}^{\beta } } d\lambda\left\langle e^{\lambda h}a^{\dag}e^{-\lambda h}e^{-iht/\hbar } be^{iht/\hbar}\right\rangle \nonumber\\ & = -i{\displaystyle\int\nolimits_{0}^{\infty } } dte^{izt}{\displaystyle\int\nolimits_{0}^{\beta } } d\lambda\left\langle e^{iht/\hbar+\lambda h}a^{\dag}e^{-iht/\hbar-\lambda h}b\right\rangle \nonumber\\ & = -i{\displaystyle\int\nolimits_{0}^{\infty } } dte^{izt}{\displaystyle\int\nolimits_{0}^{\beta } } d\lambda\left\langle e^{ih\left ( t - i\lambda\hbar\right ) /\hbar}a^{\dag } e^{-ih\left ( t - i\lambda\hbar\right ) /\hbar}b\right\rangle \nonumber\\ & = -i{\displaystyle\int\nolimits_{0}^{\infty } } dte^{izt}{\displaystyle\int\nolimits_{0}^{\beta } } d\lambda\left\langle e^{il\left ( t - i\lambda\hbar\right ) } a^{\dag } b\right\rangle \label{if12b}\\ & = -i{\displaystyle\int\nolimits_{0}^{\infty } } dte^{izt}{\displaystyle\int\nolimits_{0}^{\beta } } d\lambda\left\langle a^{\dag}(t - i\hbar\lambda)b(0)\right\rangle .
\label{if13}\ ] ] [ [ representation - of - the - impedance - function - in - terms - of - the - relaxation - function ] ] representation of the impedance function in terms of the relaxation function + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + the impedance function is related to the relaxation function@xmath426 where @xmath427 is the velocity operator , by the following expression : @xmath428 ( @xmath429 [ [ demonstration - of - the - representation - if7 . ] ] demonstration of the representation ( [ if7 ] ) .
_ _ apply ( [ if12b ] ) to the relaxation function entering ( [ if6]):@xmath430 and perform the integration by parts using the formula@xmath431@xmath432 this allows us to represent the relaxation function in the form@xmath433 where the expression ( [ if15a ] ) is inserted in the first term . further on , the integral over @xmath250 is taken as follows:@xmath434 \\
& = \frac{1}{\hbar\mathrm{tr}e^{-\beta h}}\mathrm{tr}\left [ \left ( \dot { x}\left ( t\right ) e^{-\beta h}-e^{-\beta h}\dot{x}\left ( t\right ) \right ) \dot{x}\right ] \\ & = \frac{1}{\hbar\mathrm{tr}e^{-\beta h}}\mathrm{tr}\left [ e^{-\beta h}\left ( \dot{x}\dot{x}\left ( t\right ) -\dot{x}\left ( t\right ) \right ) \dot{x}\right ] \\ & = \frac{1}{\hbar}\left\langle \dot{x}\dot{x}\left ( t\right ) -\dot { x}\left ( t\right ) \dot{x}\right\rangle = -\frac{1}{\hbar}\left\langle \dot{x}\left ( t\right ) , \dot{x}\right\rangle .\end{aligned}\ ] ] hence , @xmath435 \right\rangle .
\label{if13a}\ ] ] when setting @xmath436 with @xmath437 , we have@xmath438 \right\rangle dt . \label{k}\ ] ] for @xmath439 , we can set@xmath440 \right\rangle dt.\ ] ] multiplying @xmath441 by @xmath442 , we find that@xmath443 \right\rangle dt\right ) \nonumber\\ & = i\frac{e^{2}}{m_{b}\omega}+\frac{e^{2}}{\hbar\omega}\int_{0}^{\infty } e^{i\left ( \omega+i\varepsilon\right ) t}\left\langle \left [ \dot{x}\left ( t\right ) , \dot{x}\right ] \right\rangle dt\nonumber\\ & = i\frac{e^{2}}{m_{b}\omega}+\frac{1}{\hbar\omega}\int_{0}^{\infty } e^{i\left ( \omega+i\varepsilon\right ) t}\left\langle \left [ j_{x}\left ( t\right ) , j_{x}\right ] \right\rangle dt , \label{if13c}\ ] ] where the electric current density is @xmath444 substituting further ( [ if13c ] ) in ( [ if6 ] ) , we arrive at@xmath445 \right\rangle dt,\ ] ] what coincides with the expression of the impedance function ( [ ifcond ] ) through a frequency - dependent conductivity given by the kubo formula ( [ ifkubo ] ) , q.e.d . [ [ application - of - the - projection - operator - technique ] ] application of the projection operator technique + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + using the mori - zwanzig projection operator technique ( cf .
@xcite , chapter 5 ) , the relaxation function ( [ if7 ] ) @xmath446 can be represented in a form , which is especially convenient for the application in the theory of the optical absorption of polarons .
the projection operator @xmath447 ( @xmath448 ) is defined as @xmath449 with @xmath391 an operator and@xmath450 the projection operator @xmath448 projects an operator onto the space orthogonal to the space containing @xmath451 here we give some examples of the action of the projection operators:@xmath452
\right\rangle = 0,\label{if14f}\\ qx & = ( 1-p)x = x.\label{if14ff}\\ pa_{\mathbf{k } } & = \frac{\dot{x}\left ( \dot{x},a_{\mathbf{k}}\right ) } { \chi}=\frac{\dot{x}\left ( ilx , a_{\mathbf{k}}\right ) } { \chi}=-\frac { i\dot{x}\left ( a_{\mathbf{k}},lx\right ) } { \chi}=-\frac{i\dot{x}}{\chi } \left\langle \left [ a_{\mathbf{k}}^{\dag},x\right ] \right\rangle = 0,\label{if14g}\\ qa_{\mathbf{k } } & = ( 1-p)a_{\mathbf{k}}=a_{\mathbf{k } } \label{if14gg}\ ] ] the projection operators @xmath447 and @xmath453 are idempotent : @xmath454 the liouville operator can be identiaclly represented as @xmath455 . then the operator @xmath456 in the relaxation function ( [ if7 ] ) can be represented as follows:@xmath457 we use the algebraic operator identity:@xmath458 with @xmath459 and @xmath460@xmath461 consequently , the relaxation function ( [ if7 ] ) takes the form@xmath462 the first term in the r.h.s . of ( [ ifmori1 ] ) simplifies as follows:@xmath463 \dot{x}\right ) = \left ( \dot{x},\frac{1}{z}\dot{x}\right)\ ] ] because @xmath464 using the quantity ( [ if14a ] ) we obtain : @xmath465 the second term in the r.h.s . of ( [ ifmori1 ] )
contains the operator@xmath466 which according to the definition of the projection operator @xmath447 ( [ if14d ] ) can be transformed as@xmath467 it is remarkable that this term is exactly expressed in terms of the sought relaxation function ( [ if7 ] ) . substituting ( [ ifmori2 ] ) and ( [ ifmori3 ] ) in ( [ ifmori1 ] )
, we find @xmath468 l\frac{\dot{x}}{\chi}\right ) \phi\left ( z\right ) \rightarrow\nonumber\\ z\phi\left ( z\right ) & = \chi+\left [ \frac{\left ( \dot{x},l\dot { x}\right ) } { \chi}+\frac{1}{\chi}\left ( \dot{x},\frac{lq}{z - lq}l\dot { x}\right ) \right ] \phi\left ( z\right ) .\nonumber\end{aligned}\ ] ] introducring the quantity@xmath469 and the function called the
_ memory function _ @xmath470 we represent ( [ ifmori3a ] ) in the form of the equation@xmath471 \phi\left ( z\right ) = \chi.\ ] ] a solition of this equation gives the relaxation function @xmath472 represented within the mori - zwanzig projection operator technique : @xmath473 the memory function ( [ ifmori4 ] ) can be still transformed to another useful form.first of all , we apply the property of a scalar product ( [ if11b ] ) : @xmath474 for any two operators @xmath391 and @xmath392 @xmath475 \right ) \\ & = \frac{\left ( \dot{x},a\right ) } { \chi}\left [ \left ( \dot{x},b\right ) -\frac{\left ( \dot{x},\dot{x}\right ) \left ( \dot{x},b\right ) } { \chi } \right ] \\ & = \frac{\left ( \dot{x},a\right ) } { \chi}\left [ \left ( \dot{x},b\right ) -\left ( \dot{x},b\right ) \right ] = 0,\end{aligned}\ ] ] therefore the first term on the r.h.s . in ( [ ifmori5 ] ) vanishes , and we obtain@xmath476 in this expression , the operator @xmath477 can be represented in the following form , using the fact that @xmath453 is the idempotent operator:@xmath478 \nonumber\\ & = \frac{1}{z}q+\frac{1}{z^{2}}qlq+\frac{1}{z^{3}}qlqlq+ ... \nonumber\\ & = \frac{1}{z}q+\frac{1}{z^{2}}qlq^{2}+\frac{1}{z^{3}}qlq^{2}lq^{2}+ ... \nonumber\\ & = \left [ \frac{1}{z}+\frac{1}{z^{2}}qlq+\frac{1}{z^{3}}qlqqlq+ ... \right ] q\rightarrow\nonumber\\ q\frac{1}{z - lq } & = \frac{1}{z - qlq}q . \label{ifmori7}\ ] ] a new liouville operator can be defined , @xmath479 , which describes the time evolution in the hilbert space of operators , which is orthogonal complement of @xmath451 substituting then ( [ ifmori7 ] ) with the operator @xmath480 into ( [ ifmori6 ] ) , we bring it to the form , which will be used in what follows.@xmath481 for the hamiltonian ( [ eq_1a ] ) we obtain the following quantities:@xmath482 using ( [ if11c ] ) , we find@xmath483 \right\rangle = \frac{i}{m_{b}\hbar}\left\langle \left ( -i\hbar\right ) \right\rangle = \frac{1}{m_{b } } \label{if15a}\ ] ] and@xmath484
\right\rangle = 0 . \label{if15b}\ ] ] substituting ( [ if15a ] ) and ( [ if15b ] ) in ( [ if14 ] ) , one obtains@xmath485 the operator @xmath486 = \nonumber\\ & = -\frac{1}{m_{b}\hbar}\left [ p_{x},\sum_{\mathbf{k}}(v_{k}a_{\mathbf{k}}e^{i\mathbf{k\cdot r}}+v_{k}^{\ast}a_{\mathbf{k}}^{\dag}e^{-i\mathbf{k\cdot r}})\right ] \nonumber\\ & = \frac{i}{m_{b}}\sum_{\mathbf{k}}ik_{x}(v_{k}a_{\mathbf{k}}e^{i\mathbf{k\cdot r}}-v_{k}^{\ast}a_{\mathbf{k}}^{\dag}e^{-i\mathbf{k\cdot r}})\rightarrow\nonumber\\ l\dot{x } & = -\frac{1}{m_{b}}\sum_{\mathbf{k}}k_{x}(v_{k}a_{\mathbf{k}}e^{i\mathbf{k\cdot r}}-v_{k}^{\ast}a_{\mathbf{k}}^{\dag}e^{-i\mathbf{k\cdot r } } ) \label{if15d}\ ] ] does not depend on the velocities . therefore ,
multiplying both parts of ( [ if15d ] ) with @xmath453 and taking into account ( [ if14ff ] ) and ( [ if14gg ] ) , we obtain@xmath487 what allows us to represent the memory function in the form @xmath488{c}k_{x}(v_{k}a_{\mathbf{k}}e^{i\mathbf{k\cdot r}}-v_{k}^{\ast}a_{\mathbf{k}}^{\dag}e^{-i\mathbf{k\cdot r}}),\\ \frac{1}{z-\mathcal{l}}k_{x}^{\prime}(v_{k^{\prime}}a_{\mathbf{k}^{\prime}}e^{i\mathbf{k}^{\prime}\mathbf{\cdot r}}-v_{k^{\prime}}^{\ast}a_{\mathbf{k}^{\prime}}^{\dag}e^{-i\mathbf{k}^{\prime}\mathbf{\cdot r } } ) \end{array } \right ) \nonumber\\ & = \frac{1}{m_{b}}\sum_{\mathbf{k}}\sum_{\mathbf{k}^{\prime}}k_{x}k_{x}^{\prime}v_{k}v_{k^{\prime}}^{\ast}\left ( \begin{array } [ c]{c}(a_{\mathbf{k}}e^{i\mathbf{k\cdot r}}+a_{\mathbf{k}}^{\dag}e^{-i\mathbf{k\cdot r}}),\\ \frac{1}{z-\mathcal{l}}(a_{\mathbf{k}^{\prime}}e^{i\mathbf{k}^{\prime } \mathbf{\cdot r}}+a_{\mathbf{k}^{\prime}}^{\dag}e^{-i\mathbf{k}^{\prime } \mathbf{\cdot r } } ) \end{array } \right ) .
\label{if16}\ ] ] in transition to ( [ if16 ] ) we have used the property of the amplitude ( [ eq_1b ] ) : @xmath489 and taken into account that according to the definition ( [ if8 ] ) , the first operator enters a scalar product in the hermitian conjugate form . introducing the operators
@xmath490 we represent the memory function as@xmath491 we notice that @xmath492 .
it will be represented through the four relaxation functions:@xmath493 , \label{if18a}\\ \phi_{\mathbf{kk}^{\prime}}^{+\,+}(z ) & = \left ( b_{\mathbf{k}}^{\dag},\frac{1}{z-\mathcal{l}}b_{\mathbf{k}^{\prime}}^{\dag}\right ) , \label{if18b}\\ \phi_{\mathbf{kk}^{\prime}}^{-\,-}(z ) & = \left ( b_{\mathbf{k}},\frac { 1}{z-\mathcal{l}}b_{\mathbf{k}^{\prime}}\right ) , \label{if18c}\\ \phi_{\mathbf{kk}^{\prime}}^{+\,-}(z ) & = \left ( b_{\mathbf{k}}^{\dag},\frac{1}{z-\mathcal{l}}b_{\mathbf{k}^{\prime}}\right ) , \label{if18d}\\ \phi_{\mathbf{kk}^{\prime}}^{-\,+}(z ) & = \left ( b_{\mathbf{k}},\frac { 1}{z-\mathcal{l}}b_{\mathbf{k}^{\prime}}^{\dag}\right ) .
\label{if18e}\ ] ] there exist relations between the above relaxation functions .
for example , the relaxation function ( [ if18c ] ) , takes the form@xmath494 then the complex conjugate of this relaxation function : @xmath495 ^{\ast } & = i{\displaystyle\int\nolimits_{0}^{\infty } } dte^{-iz^{\ast}t}\left ( e^{i\mathcal{l}t}b_{\mathbf{k}}(0),b_{\mathbf{k}^{\prime}}(0)\right ) ^{\ast}\\ & = i{\displaystyle\int\nolimits_{0}^{\infty } } dte^{-iz^{\ast}t}\left ( b_{\mathbf{k}^{\prime}}(0),e^{i\mathcal{l}t}b_{\mathbf{k}}(0)\right ) , \end{aligned}\ ] ] where the property ( [ if11d ] ) has been used . the property ( [ if11a ] )
gives @xmath495 ^{\ast } & = i{\displaystyle\int\nolimits_{0}^{\infty } } dte^{-iz^{\ast}t}\left ( b_{\mathbf{k}^{\prime}}(0),e^{i\mathcal{h}t/\hbar } b_{\mathbf{k}}(0)e^{-i\mathcal{h}t/\hbar}\right ) \\ & = i{\displaystyle\int\nolimits_{0}^{\infty } } dte^{-iz^{\ast}t}\left ( e^{i\mathcal{h}t/\hbar}b_{\mathbf{k}}^{\dag } ( 0)e^{i\mathcal{h}t/\hbar},b_{\mathbf{k}^{\prime}}^{\dag}(0)\right ) \\ & = i{\displaystyle\int\nolimits_{0}^{\infty } } dte^{-iz^{\ast}t}\left ( e^{i\mathcal{l}t}b_{\mathbf{k}}^{\dag}(0),b_{\mathbf{k}^{\prime}}^{\dag}(0)\right ) \\ & = i{\displaystyle\int\nolimits_{0}^{\infty } } dte^{-iz^{\ast}t}\left ( b_{\mathbf{k}}^{\dag}(t),b_{\mathbf{k}^{\prime}}^{\dag}(0)\right ) = -\phi_{\mathbf{kk}^{\prime}}^{+\,+}(-z^{\ast}),\end{aligned}\ ] ] wherefrom it follows that @xmath496 ^{\ast}. \label{if19c}\ ] ] similarly , the relation @xmath497
^{\ast } \label{if19d}\ ] ] is proven .
[ [ memory - function ] ] memory function + + + + + + + + + + + + + + + in this subsection we indicate which approximations must be made in the calculation of the relaxation functions in order to obtain the fhip results for the impedance function .
consider the relaxation function ( [ if18b]):@xmath498 where @xmath499 and perform a partial integration:@xmath500 here we supposed that @xmath501 in the second term in ( [ if19]),@xmath502 because @xmath503 and @xmath504 further on , we have@xmath505 so , we find from ( [ if19])@xmath506 \right\rangle . \label{if19a}\ ] ] the first term in the r.h.s .
of this expression can be represented as follows : @xmath507 \right\ } \\ & = \frac{1}{z\hbar}\frac{1}{\mathrm{tr}e^{-\beta h}}\mathrm{tr}\left\ { b_{\mathbf{k}^{\prime}}^{\dag}\frac{1}{l}b_{\mathbf{k}}e^{-\beta h}-e^{-\beta h}\frac{1}{l}b_{\mathbf{k}}b_{\mathbf{k}^{\prime}}^{\dag}\right\ } \\ & = \frac{1}{z\hbar}\left\langle b_{\mathbf{k}^{\prime}}^{\dag}\frac{1}{l}b_{\mathbf{k}}-\frac{1}{l}b_{\mathbf{k}}b_{\mathbf{k}^{\prime}}^{\dag } \right\rangle \\ & = \frac{i}{z\hbar}\left\langle b_{\mathbf{k}^{\prime}}^{\dag}\frac{1}{il}b_{\mathbf{k}}-\frac{1}{il}b_{\mathbf{k}}b_{\mathbf{k}^{\prime}}^{\dag } \right\rangle \\ & = \frac{i}{z\hbar}\left\langle { \displaystyle\int\nolimits_{0}^{\infty } } dte^{ilt}b_{\mathbf{k}}b_{\mathbf{k}^{\prime}}^{\dag}-{\displaystyle\int\nolimits_{0}^{\infty } } dtb_{\mathbf{k}^{\prime}}^{\dag}e^{ilt}b_{\mathbf{k}}\right\rangle \\ & = \frac{i}{z\hbar}\left\langle { \displaystyle\int\nolimits_{0}^{\infty } } dtb_{\mathbf{k}}(t)b_{\mathbf{k}^{\prime}}^{\dag}-{\displaystyle\int\nolimits_{0}^{\infty } } dtb_{\mathbf{k}^{\prime}}^{\dag}b_{\mathbf{k}}(t)\right\rangle \\ & = \frac{i}{z\hbar}{\displaystyle\int\nolimits_{0}^{\infty } } dt\left\langle \left [ b_{\mathbf{k}}(t),b_{\mathbf{k}^{\prime}}^{\dag } ( 0)\right ] \right\rangle .\end{aligned}\ ] ] substituting it in the r.h.s . of ( [ if19a ] ) , we find@xmath508 \right\rangle -\frac{i}{z\hbar}{\displaystyle\int\nolimits_{0}^{\infty } } dte^{izt}\left\langle \left [ b_{\mathbf{k}}(t),b_{\mathbf{k}^{\prime}}^{\dag } ( 0)\right ] \right\rangle \nonumber\\ & = \frac{i}{z\hbar}{\displaystyle\int\nolimits_{0}^{\infty } } dt\left ( 1-e^{izt}\right ) \left\langle \left [ b_{\mathbf{k}}(t),b_{\mathbf{k}^{\prime}}^{\dag}(0)\right ] \right\rangle .
\label{if19b}\ ] ] in a similar way one obtains @xmath509 \right\rangle . \label{if19e}\ ] ] inserting the relaxation functions ( [ if19b ] ) , ( [ if19e ] ) , ( [ if19c ] ) and ( [ if19d ] ) , we find the memory function ( [ if18a])@xmath510{c}\left\langle \left [ b_{\mathbf{k}}(t),b_{\mathbf{k}^{\prime}}^{\dag } ( 0)\right ] \right\rangle + \left\langle \left [ b_{\mathbf{k}}(t),b_{\mathbf{k}^{\prime}}(0)\right ] \right\rangle \\
\left [ b_{\mathbf{k}}(t),b_{\mathbf{k}^{\prime}}^{\dag } ( 0)\right ] \right\rangle ^{\ast}-\left\langle \left [ b_{\mathbf{k}}(t),b_{\mathbf{k}^{\prime}}(0)\right ] \right\rangle ^{\ast}\end{array } \right ] \\ & = -\frac{1}{m_{b}}\sum_{\mathbf{k}}\sum_{\mathbf{k}^{\prime}}k_{x}k_{x}^{\prime}v_{k}v_{k^{\prime}}^{\ast}\frac{2}{z\hbar}{\displaystyle\int\nolimits_{0}^{\infty } } dt\left ( 1-e^{izt}\right ) \\ & \times\operatorname{im}\left [ \left\langle \left [ b_{\mathbf{k}}(t),b_{\mathbf{k}^{\prime}}^{\dag}(0)\right ] \right\rangle + \left\langle \left [ b_{\mathbf{k}}(t),b_{\mathbf{k}^{\prime}}(0)\right ] \right\rangle \right ] , \end{aligned}\ ] ] wherefrom @xmath511 with @xmath512 \right\rangle + \left\langle \left [ b_{\mathbf{k}}(t),b_{\mathbf{k}^{\prime}}(0)\right ] \right\rangle \right\ } .
\label{if20b}\ ] ] [ [ derivation - of - the - memory - function ] ] derivation of the memory function + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + to calculate the expectation values in eq .
( [ if20b ] ) , we shall make the following approximations ( cf .
the liouville operator @xmath480 , which determines the time evolution of the operator @xmath513 , is replaced by @xmath514 where @xmath515 is the liouville operator for free phonons and @xmath516 is the liouville operator for the feynman polaron model @xcite .
the frhlich hamiltonian appearing in the statistical average @xmath517 is imilarly replaced by @xmath518with @xmath519 the hamiltonian of free phonons and @xmath520 the hamiltonian of the feynman polaron model . with this approximation , e.g. , the average @xmath521 the time evolution of the free - phonon annihilation operator ( [ if10]),@xmath522 accorting to ( [ if9 ] ) is @xmath523 \exp\left ( -i\omega_{\mathbf{k}}a_{\mathbf{k}}^{+}a_{\mathbf{k}}t\right ) \\ & = \omega_{\mathbf{k}}\exp\left ( i\omega_{\mathbf{k}}a_{\mathbf{k}}^{+}a_{\mathbf{k}}t\right ) \left [ a_{\mathbf{k}}^{+},a_{\mathbf{k}}\right ] a_{\mathbf{k}}\exp\left ( -i\omega_{\mathbf{k}}a_{\mathbf{k}}^{+}a_{\mathbf{k}}t\right ) \\ & = -\omega_{\mathbf{k}}\exp\left ( i\omega_{\mathbf{k}}a_{\mathbf{k}}^{+}a_{\mathbf{k}}t\right ) a_{\mathbf{k}}\exp\left ( -i\omega_{\mathbf{k}}a_{\mathbf{k}}^{+}a_{\mathbf{k}}t\right ) \rightarrow\\ a_{\mathbf{k}}(t ) & = \exp\left ( -i\omega_{\mathbf{k}}t\right ) a_{\mathbf{k}}.\end{aligned}\ ] ] similarly , @xmath524 hence , we have @xmath525 , \ ] ] where @xmath526 ^{-1}$ ] is the average number of phonons with energy @xmath527 . the calculation of the fourier component of the electron density - density correlation function @xmath528 in eq .
( [ if21 ] ) for an electron described by the feynman polaron model is given below following the approach of ref .
@xcite .
we calculate the correlation function @xmath529 with the feynman trial hamiltonian@xmath530 here , @xmath531 denotes the operator in the heisenberg representation@xmath532 we show that the correlation function @xmath533 depends on @xmath534 rather than on @xmath535 and @xmath536 independently:@xmath537 where @xmath538 .
the normal coordinates are the center - of - mass vector @xmath539 and the vector of the relative motion @xmath540:@xmath541{c}\mathbf{r}=\frac{m\mathbf{r}+m_{f}\mathbf{r}_{f}}{m+m_{f}}\\ \mathbf{\rho}=\mathbf{r}-\mathbf{r}_{f}\end{array } \right .
\label{tf - eq2}\ ] ] the inverse transformation is:@xmath541{c}\mathbf{r}=\mathbf{r}+\frac{m_{f}}{m+m_{f}}\mathbf{\rho}\\ \mathbf{r}_{f}=\mathbf{r}-\frac{m}{m+m_{f}}\mathbf{\rho}\end{array } \right .
\label{tf - eq3}\ ] ] the same transformation as ( [ tf - eq2 ] ) takes place for velocities:@xmath541{c}\mathbf{\dot{r}}=\mathbf{\dot{r}}+\frac{m_{f}}{m+m_{f}}\mathbf{\dot{\rho}}\\ \mathbf{\dot{r}}_{f}=\mathbf{\dot{r}}-\frac{m}{m+m_{f}}\mathbf{\dot{\rho}}\end{array } \right.\ ] ] from ( [ tf - eq2 ] ) we derive the transformation for moments@xmath541{c}\frac{\mathbf{p}}{m}=\frac{\mathbf{p}}{m+m_{f}}+\frac{m_{f}}{m+m_{f}}\frac{\mathbf{p}_{\rho}}{\frac{mm_{f}}{m+m_{f}}}\\ \frac{\mathbf{p}_{f}}{m_{f}}=\frac{\mathbf{p}}{m+m_{f}}-\frac{m}{m+m_{f}}\frac{\mathbf{p}_{\rho}}{\frac{mm_{f}}{m+m_{f}}}\end{array } \right.\]]@xmath542@xmath541{c}\mathbf{p}=\frac{m}{m+m_{f}}\mathbf{p}+\mathbf{p}_{\rho}\\ \mathbf{p}_{f}=\frac{m_{f}}{m+m_{f}}\mathbf{p}-\mathbf{p}_{\rho}\end{array } \right .
\label{tf - eq4}\ ] ] the hamiltonian ( [ tf - eq1 ] ) then takes the form@xmath543 with the masses@xmath544 and with the frequency@xmath545 the cartesian coordinates and moments corresponding to the relative motion can be in the standard way expressed in terms of the second quantization operators:@xmath546 in these notations , the hamiltonian ( [ tf - eq5 ] ) takes the form@xmath547 using ( [ tf - eq9 ] ) , we find the operators in the heisenberg representation , ( i ) for the center - of mass coordinates@xmath548 \\ & = x_{j}+i\frac{\sigma}{2m\hbar}\left ( p_{j}^{2}x_{j}-p_{j}x_{j}p_{j}+p_{j}x_{j}p_{j}-x_{j}p_{j}^{2}\right ) \\ & = x_{j}+i\frac{\sigma}{2m\hbar}\left ( p_{j}\left [ p_{j},x_{j}\right ] + \left [ p_{j},x_{j}\right ] p_{j}\right ) \\ & = x_{j}+i\frac{\sigma}{2m\hbar}p_{j}\left ( -2i\hbar\right ) = x_{j}+\frac{\sigma}{m}p_{j},\end{aligned}\]]@xmath549 ( ii ) for the operators @xmath550 and @xmath551@xmath552 using the first formula of ( [ tf - eq3 ] ) we find@xmath553@xmath542@xmath554 we denote@xmath555@xmath556 therefore , we obtain@xmath557 @xmath558 \nonumber\\ & = \exp\left ( -i\mathbf{k\cdot r}-i\frac{\sigma}{m}\mathbf{k\cdot p}\right ) \exp\left ( -ia\mathbf{k\cdot c}e^{-i\bar{\omega}\sigma}-ia\mathbf{k\cdot c}^{\dag}e^{i\bar{\omega}\sigma}\right ) \nonumber\\ & = \prod_{j=1}^{3}\exp\left ( -ik_{j}x_{j}-i\frac{\sigma}{m}k_{j}p_{j}\right ) \exp\left ( -iak_{j}c_{j}e^{-i\bar{\omega}\sigma}-iak_{j}c_{j}^{\dag}e^{i\bar{\omega}\sigma}\right ) .
\label{tf - eq14}\ ] ] the disentangling of the exponents is performed using the formula@xmath559 in the case when @xmath560 $ ] commutes with both @xmath391 and @xmath392 , this formula is reduced to@xmath561 } .
\label{tf - eq16}\ ] ] we perform the necessary commutations:@xmath562 = \frac{1}{2}k_{j}^{2}\frac{\sigma}{m}\left [ x_{j},p_{j}\right ] = i\frac{\hbar k_{j}^{2}}{2m}\sigma,\]]@xmath563 = \frac{1}{2}a^{2}k_{j}^{2}\left [ c_{j}^{\dag},c_{j}\right ] = -\frac{1}{2}a^{2}k_{j}^{2}\]]@xmath564 = \frac{1}{2}a^{2}k_{j}^{2}\left [ c_{j}^{\dag } , c_{j}\right ] = -\frac{1}{2}a^{2}k_{j}^{2}\]]@xmath542@xmath565@xmath542@xmath566 it follows from eq .
( [ tf - eq16 ] ) that when @xmath560 $ ] commutes with both @xmath391 and @xmath392,@xmath567 } .
\label{tf - eq17}\ ] ] using ( [ tf - eq17 ] ) , we find@xmath568 } \\ & = e^{-iak_{j}\cdot c_{j}^{\dag}e^{i\bar{\omega}\sigma}}e^{iak_{j}\cdot c_{j}}e^{a^{2}k_{j}^{2}e^{i\bar{\omega}\sigma}}.\end{aligned}\ ] ] herefrom , we find @xmath569 the correlation function then is@xmath570 since the variables of the center - of mass motion and of the relative motion are averaged independently.@xmath571 with@xmath572 let us consider the auxiliary expectation value@xmath573@xmath574 where @xmath575 are the eigenstates of @xmath576 .
the operators @xmath577 act on these states as follows:@xmath578 therefore , we find@xmath579@xmath542@xmath580 and@xmath581 in particular , for @xmath582 we have@xmath583 as a result , the expectation value @xmath584 is@xmath585 , \label{tf - eq19}\ ] ] with@xmath586 using this result , we obtain the expression@xmath587 \\ & = \exp\left [ -n\left ( \omega\right ) a^{2}k^{2}\left ( 1-e^{i\bar{\omega } \sigma}\right ) \left ( 1-e^{-i\bar{\omega}\sigma}\right ) \right ] \\ & = \exp\left [ -4n\left ( \omega\right ) a^{2}k^{2}\sin^{2}\left ( \frac { 1}{2}\omega\sigma\right ) \right ] .\end{aligned}\ ] ] the expectation value @xmath588 is@xmath589 collecting all factors in eq .
( [ tf - eq20 ] ) together , we find@xmath590 \rightarrow.\]]@xmath591 with the function@xmath592 , \label{if22a}\ ] ] where @xmath593 according to ( [ tf - time ] ) , @xmath594 taking @xmath595 in ( [ tf - eq21 ] ) we finally find the fourier component of the electron density - density correlation function @xmath596 which enters eq .
( [ if21 ] ) : @xmath597 \label{if22}\ ] ] finally , the correlation functions in ( [ if20b ] ) reduce to@xmath598 \exp\left ( -i\omega_{\mathbf{k}}t\right ) \exp\left [ -k^{2}d(-t)\right ] , \\
\left\langle b_{\mathbf{k}}^{\dag}(0)b_{\mathbf{k}}(t)\right\rangle & = n(\omega_{\mathbf{k}})\exp\left ( -i\omega_{\mathbf{k}}t\right ) \exp\left [ -k^{2}d(t)\right ] , \\
\left\langle b_{\mathbf{k}}(t)b_{\mathbf{k}}(0)\right\rangle & = 0,\\ \left\langle b_{\mathbf{k}}(0)b_{\mathbf{k}}(t)\right\rangle & = 0.\end{aligned}\ ] ] inserting these equations into eqs .
( [ if20a ] ) and ( [ if20b ] ) , one obtains@xmath599 with@xmath600{c}-\left [ 1+n(\omega_{\mathbf{k}})\right ] \exp\left ( -i\omega_{\mathbf{k}}t\right ) \exp\left [ -k^{2}d(-t)\right ] \\ + n(\omega_{\mathbf{k}})\exp\left ( -i\omega_{\mathbf{k}}t\right ) \exp\left [ -k^{2}d(t)\right ] \end{array } \right\ } .\ ] ] using the property @xmath601 for real vaues of @xmath602one obtains@xmath603 \ } & = \operatorname{im}\exp\left ( i\omega_{\mathbf{k}}t\right ) \exp\left [ -k^{2}d(-t)^{\ast}\right ] \}\\ & = \operatorname{im}\exp\left ( i\omega_{\mathbf{k}}t\right ) \exp\left [ -k^{2}d(t)\right ] \}\end{aligned}\ ] ] and consequently@xmath604 \left\ { \left [ 1+n(\omega_{\mathbf{k}})\right ] \exp\left ( i\omega_{\mathbf{k}}t\right ) + n(\omega_{\mathbf{k}})\exp\left ( -i\omega_{\mathbf{k}}t\right ) \right\ } .
\label{if23b}\ ] ] owing to the rotational invariance of @xmath605 and @xmath606we can substitute in ( [ if23b])@xmath607 the resulting expression for @xmath608 \left\ { \left [ 1+n(\omega_{\mathbf{k}})\right ] \exp\left ( i\omega_{\mathbf{k}}t\right ) + n(\omega_{\mathbf{k}})\exp\left ( -i\omega_{\mathbf{k}}t\right ) \right\ } \label{if23c}\ ] ] is identical with eq .
( 35 ) of fhip @xcite . in the case of frhlich polarons , taking into account ( [ eq_1b ] ) , eq .
( [ if23c ] ) simplifies to @xmath609{c}\left [ 1+n(\omega_{\mathrm{lo}})\right ] \exp\left ( i\omega_{\mathrm{lo}}t\right )
\\ + n(\omega_{\mathrm{lo}})\exp\left ( -i\omega_{\mathrm{lo}}t\right ) \end{array } \right\ } \left [ d(t)\right ] ^{-3/2}. \label{if23d}\ ] ] upon substituion of ( [ if6 ] ) and ( [ if15c ] ) into eq . ( [ if5 ] ) , we find the absorption coefficient@xmath610 = -\frac { 1}{n\epsilon_{0}c}\frac{e^{2}}{m_{b}}\operatorname{im}\left [ \frac{1}{\omega-\sigma(\omega)}\right ] \\ & = -\frac{1}{n\epsilon_{0}c}\frac{e^{2}}{m_{b}}\operatorname{im}\left\ { \frac{\omega-\sigma^{\ast}(\omega)}{\left [ \omega-\operatorname{re}\sigma(\omega)\right ] ^{2}+\left [ \operatorname{im}\sigma(\omega)\right ] ^{2}}\right\ } \\ & = \frac{1}{n\epsilon_{0}c}\frac{e^{2}}{m_{b}}\frac{\operatorname{im}\sigma^{\ast}(\omega)}{\left [ \omega-\operatorname{re}\sigma(\omega)\right ] ^{2}+\left [ \operatorname{im}\sigma(\omega)\right ] ^{2}}\rightarrow\end{aligned}\ ] ] @xmath611 ^{2}+\left [ \operatorname{im}\sigma(\omega)\right ] ^{2}}\ .
\label{eq : p24 - 2}\ ] ] this general expression was the strating point for a derivation of the theoretical optical absorption spectrum of a single large polaron , at _ all electron - phonon coupling strengths _ by devreese et al . in ref.@xcite . the memory function @xmath612 as given by eq .
( [ if23a ] ) with ( [ if23d ] ) contains the dynamics of the polaron and depends on @xmath63 , temperature and @xmath365following the notation , introduced in ref .
@xcite , @xmath613 we reresent eq .
( [ eq : p24 - 2 ] ) in the form used in ref.@xcite:@xmath614 ^{2}+\left [ \operatorname{im}\chi(\omega)\right ] ^{2}}. \label{eq : p24 - 2a}\ ] ] according to ( [ if23a ] ) and ( [ chi0]),@xmath615 s(t ) . \label{if24}\ ] ] in the present notes we limit our attention to the case @xmath377 @xmath616 it was demonstrated in ref.@xcite that the exact zero - temperature limit arises if the limit @xmath617 is taken directly in the expressions ( [ if24 ] ) ( see appendices a and b of ref.@xcite ) . as follows from ( [ if23d ] ) , @xmath618 ^{-3/2}\qquad(\beta\rightarrow\infty ) .
\label{if25}\ ] ] accorting to ( [ if22a])@xmath619 \qquad(\beta\rightarrow\infty).\ ] ] using the feynman units ( where @xmath620 and @xmath621 ) , we obtain from ( [ if22b]):@xmath622 and consequently@xmath623 with @xmath624 and according to ( [ if25 ] ) @xmath625 ^{-3/2}\qquad(\beta\rightarrow\infty).\ ] ] from ( [ if24 ] ) one obtains immediately@xmath626 ^{3/2}},\label{if26a}\\ \operatorname{re}\chi(\omega ) & = \frac{2\alpha}{3\sqrt{\pi}}\left ( \frac { v}{w}\right ) ^{3}\operatorname{im}{\displaystyle\int\nolimits_{0}^{\infty } } dt\frac{\left [ 1-\cos(\omega t)\right ] e^{it}}{\left [ r(1-e^{ivt})-it\right ] ^{3/2}}. \label{if26b}\ ] ] in the limit @xmath617 the function @xmath627 was calculated by fhip @xcite
. however , to study the optical absorption to the same approximation as fhip s treatment of the impedance , we have also to calculate @xmath628 and use this result in ( [ eq : p24 - 2a ] ) .
the calculation of @xmath629 which is a kramers - kronig - type transform of @xmath627 , is a key ingredient in ref.@xcite .
the details of those calculations are presented in the appendices a , b and c to ref.@xcite .
developing the denominator of both integrals on the right - hand side of ( [ if26a ] ) and ( [ if26b ] ) , the calculations are reduced to the evaluation of a sum of integrals of the type@xmath630 in appendix b to ref.@xcite it is shown how such integrals are evaluated using a recurrence formula .
for @xmath627 a very convenient result was found in @xcite:@xmath631 this expression is a finite sum and not and infinite series .
fhip gave the first two terms of ( [ if27 ] ) explicitly .
using the same recurrence relation it is seen the analytical expression ( see appendix b to ref.@xcite ) , which was found for @xmath628 is far more complicated . to circumwent the difficulty with the numerical treatment of @xmath628 , the corresponding integrals in ( [ if26c ] )
have been transformed in @xcite to integrals with rapildy convergent integrands:@xmath632 e^{i(1+nv)t}}{\left ( r - it\right ) ^{3/2+n}}\nonumber\\ & = -\frac{1}{\gamma(n+\frac{3}{2})}{\displaystyle\int\nolimits_{0}^{\infty } } dx\left [ ( n+\frac{1}{2})x^{n-1/2}e^{-rx}-rx^{n+1/2}e^{-rx}\right ] \nonumber\\ & \times\ln\left\vert \left ( \frac{\left ( 1+nv+x\right ) ^{2}}{\omega ^{2}-\left ( 1+nv+x\right ) ^{2}}\right ) \right\vert ^{1/2}. \label{if28}\ ] ] the integral on the right - hand side of ( [ if28 ] ) is adequate for computer calculations . in appendix c to ref.@xcite some supplementary details of the computation of ( [ if28 ] ) are given . another analytical representation for the memory function ( [ if23a ] )
was derived in ref .
@xcite . at weak coupling , the optical absorption spectrum ( [ eq : p24 - 2 ] ) of the polaron
is determined by the absorption of radiation energy , which is reemitted in the form of lo phonons . for @xmath633 ,
the polaron can undergo transitions toward a relatively stable res ( see fig .
[ fig_3 ] ) .
the res peak in the optical absorption spectrum also has a phonon sideband - structure , whose average transition frequency can be related to a fc - type transition .
furthermore , at zero temperature , the optical absorption spectrum of one polaron exhibits also a zero - frequency central peak [ @xmath634 . for non - zero temperature , this central peak smears out and gives rise to an anomalous drude - type low - frequency component of the optical absorption spectrum .
for example , in fig .
[ fig_3 ] from ref .
@xcite , the main peak of the polaron optical absorption for @xmath635 5 at @xmath636 is interpreted as due to transitions to a res . a shoulder at the low - frequency side of the main peak is attributed to one - phonon transitions to polaron - scattering states .
the broad structure centered at about @xmath637 is interpreted as a fc band . as seen from fig .
[ fig_3 ] , when increasing the electron - phonon coupling constant to @xmath63=6 , the res peak at @xmath638 stabilizes .
it is in ref .
@xcite that the all - coupling optical absorption spectrum of a frhlich polaron , together with the role of res - states , fc - states and scattering states , was first presented . and @xmath639 .
the res peak is very intense compared with the fc peak .
the frequency @xmath640 is indicated by the dashed lines.),scaledwidth=70.0% ] recent interesting numerical calculations of the optical conductivity for the frhlich polaron performed within the diagrammatic quantum monte carlo method @xcite , see fig .
[ fig_4 ] , fully confirm the essential analytical results derived by devreese et al . in ref .
@xcite for @xmath641 in the intermediate coupling regime @xmath642 the low - energy behavior and the position of the res - peak in the optical conductivity spectrum of ref .
@xcite follow closely the prediction of ref .
there are some minor qualitative differences between the two approaches in the intermediate coupling regime : in ref .
@xcite , the dominant ( res ) peak is less intense in the monte - carlo numerical simulations and the second ( fc ) peak develops less prominently .
there are the following qualitative differences between the two approaches in the strong coupling regime : in ref.@xcite , the dominant peak broadens and the second peak does not develop , giving instead rise to a flat shoulder in the optical conductivity spectrum at @xmath643 this behavior has been tentatively attributed to the optical processes with participation of two @xcite or more phonons . the above differences can arise also due to the fact that , within the feynman polaron model , one - phonon processes are assigned more oscillator strength and the res tends to be more stable as compared to the monte - carlo result .
the nature of the excited states of a polaron needs further study .
an independent numerical simulation might be called for . and @xmath257 and to the analytical dsg calculations @xcite ( solid lines ) .
_ right - hand panel _ : monte carlo optical conductivity spectra for the intermediate coupling regime ( open circles ) compared to the analytical dsg approach @xcite ( solid lines ) .
arrows point to the two- and three - phonon thresholds .
( from ref.@xcite.),scaledwidth=80.0% ] in fig
. [ fig_4a ] , monte - carlo optical conductivity spectrum of one polaron for @xmath257 compares well with that obtained in ref . @xcite within the canonical - transformation formalism taking into account correlation in processes involving two lo phonons .
the difference between the results of these two approaches becomes less pronounced when decreasing the value of @xmath257 and might be indicative of a possible precision loss , which requires an independent check . for @xmath257 calculated within the monte carlo approach @xcite ( open circles ) and
derived using the expansion in powers of @xmath63 up to @xmath102 @xcite ( solid curve).,scaledwidth=70.0% ] the coupling constant @xmath63 of the known ionic crystals is too small ( @xmath644 ) to allow for the experimental detection of sharp res peaks , and the resonance condition @xmath645 can not be satisfied for @xmath646 as shown in ref , @xcite . nevertheless , for @xmath647 the development of res is already reflected in a broad optical absorption peak .
such a peak , predicted in ref .
@xcite , was identified , e. g. , in the optical absorption of pr@xmath67nio@xmath648 in ref . @xcite .
also , the resonance condition can be fulfilled if an external magnetic field is applied ; the magnetic field stabilizes the res , which then can be detected in a cyclotron resonance peak . in this section
, we analyze the sum rules for the optical conductivity spectra obtained within the dsg approach @xcite with those obtained within the diagrammatic monte carlo calculation @xcite .
the values of the polaron effective mass for the monte carlo approach are taken from ref .
@xcite . in tables 3 and 4
, we represent the polaron ground - state @xmath0 and the following parameters calculated using the optical conductivity spectra:@xmath649 where @xmath650 is the upper value of the frequency available from ref .
@xcite,@xmath651 where @xmath1 is the polaron mass , the optical conductivity is calculated in units @xmath652 @xmath1 is measured in units of the band mass @xmath55 , and the frequency is measured in units of @xmath52 .
the values of @xmath650 are : @xmath653 for @xmath654 1 and 3 , @xmath655 for @xmath656 5.25 and 6 , @xmath657 for @xmath658 7 and 8 . * table 3 .
polaron parameters obtained within the diagrammatic monte carlo approach * [ c]|c|c|c|c|@xmath63 & @xmath659 & @xmath660 & @xmath661 + @xmath662 & @xmath663 & @xmath664 & @xmath665 + @xmath666 & @xmath667 & @xmath668 & @xmath669 + @xmath670 & @xmath671 & @xmath672 & @xmath673 + @xmath674 & @xmath675 & @xmath676 & @xmath677 + @xmath678 & @xmath679 & @xmath680 & @xmath681 + @xmath639 & @xmath682 & @xmath683 & @xmath684 + @xmath685 & @xmath686 & @xmath687 & @xmath688 + @xmath689 & @xmath690 & @xmath691 & @xmath692 + @xmath693 & @xmath694 & @xmath695 & @xmath696 + [ c]|c|c|@xmath697 & @xmath698 + @xmath699 & @xmath225 + @xmath700 & @xmath701 + @xmath702 & @xmath703 + @xmath704 & @xmath705 + @xmath706 & @xmath707 + @xmath708 & @xmath709 + @xmath710 & @xmath711 + @xmath712 & @xmath713 + @xmath714 & @xmath715 + * table 4 .
polaron parameters obtained within the path - integral approach * [ c]|c|c|c|c|@xmath63 & @xmath716 & @xmath717 & @xmath718 + @xmath662 & @xmath719 & @xmath664 & @xmath665 + @xmath666 & @xmath720 & @xmath721 & @xmath722 + @xmath670 & @xmath723 & @xmath724 & @xmath725 + @xmath674 & @xmath726 & @xmath727 & @xmath728 + @xmath678 & @xmath729 & @xmath730 & @xmath731 + @xmath639 & @xmath732 & @xmath733 & @xmath734 + @xmath685 & @xmath735 & @xmath736 & @xmath737 + @xmath689 & @xmath738 & @xmath739 & @xmath740 + @xmath693 & @xmath741 & @xmath742 & @xmath743 + [ c]|c|c|@xmath744 & @xmath745 + @xmath746 & @xmath225 + @xmath747 & @xmath748 + @xmath749 & @xmath750 + @xmath751 & @xmath752 + @xmath753 & @xmath754 + @xmath755 & @xmath756 + @xmath757 & @xmath758 + @xmath759 & @xmath760 + @xmath761 & @xmath762 + the parameters corresponding to the monte carlo calculation are obtained using the numerical data kindly provided by a. mishchenko .
the comparison of the zero frequency moments @xmath661 and @xmath718 with each other and with the value @xmath763 corresponding to the sum rule @xcite@xmath764 shows that @xmath765 is smaller than each of the differences @xmath766 , @xmath767 which appear due to a finite interval of the integration in ( [ 1 ] ) , ( [ 2 ] ) .
we analyze also the fulfilment of the ground - state theorem @xcite@xmath768 using the first - frequency moments @xmath769 and @xmath770 .
the results of this comparison are presented in fig .
[ moments2-f1 ] .
the dots indicate the polaron ground - state energy calculated using the feynman variational principle .
the solid curve is the value of @xmath771 calculated numerically using the optical conductivity spectra and the ground - state theorem with the dsg optical conductivity @xcite for a polaron,@xmath772 the dashed and the dot - dashed curves are the values obtained using @xmath773 and @xmath774 , respectively:@xmath775 as seen from the figure , @xmath776 to a high degree of accuracy coincides with the variational polaron ground - state energy . both @xmath777 and @xmath778 differ from @xmath779 due to the integration over a finite interval of frequencies .
however , @xmath780 and @xmath781 are very close to each other .
herefrom , a conclusion follows that for integrals over the finite frequency region characteristic for the polaron optical absorption ( i. e. , except the tails ) , the function @xmath782 ( [ e3 ] ) reproduces very well the function @xmath783 .
the form of the frhlich hamiltonian in @xmath275 dimensions is the same as in 3d,@xmath784 except that now all vectors are @xmath275-dimensional . in this subsection ,
dispersionless longitudinal phonons are considered , i.e. , @xmath785 , and units are chosen such that @xmath786 .
the electron - phonon interaction is a representation in second quantization of the electron interaction with the lattice polarization , which in 3d is essentially the coulomb potential @xmath787 @xmath788 is proportional to the fourier transform of this potential , and as a consequence we have in @xmath275 dimensions@xmath789 where @xmath790 is the volume of the @xmath275-dimensional crystal .
note that @xmath791 , where @xmath792 is an ( @xmath793)-dimensional vector , can be obtained from @xmath794 , where @xmath795 @xmath796 is an @xmath275-dimensional vector , by summing out one of the dimensions explicitly:@xmath797 inserting eq .
( [ sr2 ] ) into eq .
( [ sr3 ] ) , we have@xmath798 replacing the sum in eq .
( [ sr4 ] ) by an integral , i.e.,@xmath799 we obtain @xmath800 since @xmath801 we have@xmath802 where @xmath803 is eh @xmath804 function .
inserting eq .
( [ sr6 ] ) into eq .
( [ sr5 ] ) , we obtain@xmath805 in 3d the interaction coefficient is well known , @xmath806 so that @xmath807 inserting eq .
( [ sr8 ] ) into eq .
( [ sr7 ] ) , we immediately obtain@xmath808 applying eq .
( [ sr7 ] ) @xmath809 times , we further obtain for @xmath810@xmath811 so , the interaction coefficient in @xmath275 dimensions becomes @xcite @xmath812 following the feynman approach @xcite , the upper bound for the polaron ground - state energy can be written down as@xmath813 where @xmath80 is the exact action functional of the polaron problem , while @xmath814 is the trial action functional , which corresponds to a model system where an electron is coupled by an elastic force to a fictitious particle ( i.e. , the model system describes a harmonic oscillator ) .
@xmath0 is the ground - state energy of the above model system , and@xmath815 as indicated above , the frhlich hamiltonian in @xmath275 dimensions is the same as in 3d , except that now all vectors are @xmath275-dimensional [ and the coupling coefficient @xmath816 is modified in accordance with eq .
( [ srv ] ) ] .
similarly , the only difference of the model system in @xmath275 dimensions from the model system in 3d is that now one deals with an @xmath275-dimensional harmonic oscillator .
so , directly following @xcite , one can represent @xmath817 as @xmath818 where@xmath819 } \right\rangle _ { 0}e^{-t}dt , \label{sr14}\]]@xmath820 ^{2}\right\rangle _ { 0}e^{-wt}dt , \label{sr15}\ ] ] @xmath264 and @xmath200 are variational parameters , which should be determined by minimizing @xmath303 of eq .
( [ sr11 ] ) . since the averaging @xmath821 in eq .
( [ sr14 ] ) is performed with the trial action , which corresponds to a harmonic oscillator , components of the electron coordinates , @xmath822 ( @xmath823 ) , in @xmath824 } \right\rangle _ { 0}$ ] separate @xcite:@xmath825 } \right\rangle _ { 0}=\prod\limits_{j=1}^{n}\left\langle e^{ik_{j}\left [ r_{j}\left ( t\right ) -r_{j}\left ( 0\right ) \right ] } \right\rangle _ { 0}. \label{sr16}\ ] ] for the average @xmath826 } \right\rangle _ { 0}$ ] , feynman obtained @xcite@xmath827 } \right\rangle _ { 0}=e^{-k_{j}^{2}d_{0}\left ( t\right ) } , \label{sr17}\ ] ] where@xmath828 inserting eq .
( [ sr16 ] ) with eq .
( [ sr17 ] ) into eq .
( [ sr14 ] ) , we obtain@xmath829 inserting expression ( [ srv ] ) for @xmath788 into eq .
( [ sr19 ] ) and replacing the sum over @xmath59 by an integral [ see ( [ sr4a ] ) ] , we have @xmath830 where @xmath831 is the elemental solid angle in @xmath275 dimensions . since the integrand in eq .
( [ sr20 ] ) depends only on the modulus @xmath832 of @xmath59 , one have simply @xmath833 with@xmath834 so , we obtain for @xmath391 the result@xmath835 like in ref .
@xcite , @xmath392 can be easily calculated by noticing that @xmath836 ^{2}\right\rangle _ { 0 } & = \sum_{j=1}^{n}\left\langle \left [ r_{j}\left ( t\right ) -r_{j}\left ( 0\right ) \right ] ^{2}\right\rangle _ { 0}=\sum_{j=1}^{n}\left .
\left [ -\frac{\partial^{2}}{\partial k_{j}^{2}}\left\langle e^{i\mathbf{k}\cdot\left [ \mathbf{r}\left ( t\right ) -\mathbf{r}\left
( 0\right ) \right ] } \right\rangle _ { 0}\right ] \right\vert _ { \mathbf{k}=0}\nonumber\\ & = \sum_{j=1}^{n}2d_{0}\left ( t\right ) = 2nd_{0}\left ( t\right ) , \label{ssr1}\ ] ] so that@xmath837 dt\nonumber\\ & = \frac{nw\left ( v^{2}-w^{2}\right ) } { 2}\left [ \frac{w^{2}}{2v^{2}}\frac{1}{v^{2}}+\frac{v^{2}-w^{2}}{2v^{3}}\left ( \frac{1}{w}-\frac{1}{v+w}\right ) \right ] \nonumber\\ & = \frac{n\left ( v^{2}-w^{2}\right ) } { 4v}. \label{ssr3}\ ] ] inserting eq .
( [ srv ] ) with @xmath391 and @xmath392 , given by eqs .
( [ sr22 ] ) and ( [ ssr3 ] ) , together with the ground - state energy of the model system @xcite ( an isotropic @xmath275-dimensional harmonic oscillator),@xmath838 into eq .
( [ sr11 ] ) , we obtain @xmath839 in order to make easier a comparison of @xmath303 for @xmath275 dimensions with the feynman result @xcite for 3d,@xmath840 it is convenient to rewrite eq .
( [ ssr5 ] ) in the form@xmath841 .
\label{ssr7}\ ] ] it is worth recalling that the parameters @xmath264 and @xmath200 must be determined by minimizing @xmath303 .
thus , in the case of eq .
( [ ssr7 ] ) one has to minimize the expression in the square brackets .
the only difference of this expression from the r.h.s . of eq .
( [ ssr6 ] ) is that @xmath63 is multiplied by the factor @xmath842 this means that the minimizing parameters @xmath264 and @xmath200 in @xmath275d at a given @xmath63 will be exactly the same as those calculated in 3d for the frhlich constant as large as @xmath843 : @xmath844 therefore , comparing eq .
( [ ssr7 ] ) to eq .
( [ ssr6 ] ) , we obtain the scaling relation @xcite@xmath845 where @xmath846 is given by eq .
( [ sran ] ) . as discussed in ref .
@xcite , the above scaling relation is not an exact relation .
it is valid for the feynman polaron energy and also for the ground - state energy to order @xmath63 . the next - order term ( i.e. , @xmath102 ) no longer satisfies eq .
( [ ssr8 ] ) .
the reason is that in the exact calculation ( to order @xmath102 ) the electron motion in the different space directions is coupled by the electron - phonon interaction .
no such a coupling appears in the feynman polaron model [ see , e.g. , eq . ( [ sr16 ] ) ] ; and this is the underlying reason for the validity of the scaling relation for the feynman approximation . in refs .
@xcite , scaling relations are obtained also for the impedance function , the effective mass and the mobility of a polaron .
the inverse of the impedance function @xmath847 is given by @xmath848 where the memory function @xmath849 can be expressed as @xcite@xmath850 with @xmath851 and@xmath852@xmath853 e^{t}+n\left ( 1\right ) e^{-it},\]]@xmath854 .\ ] ] here , @xmath181 is the inverse temperature and @xmath855 is the occupation number of phonons with frequency @xmath856 ( recall that in our units @xmath857 ) . as implied from eqs .
( [ ssr9 ] ) and ( [ ssr10 ] ) , scaling of @xmath858 and @xmath847 is determined by scaling of @xmath859 . for
an isotropic crystal , since @xmath794 , @xmath860 and @xmath861 do not depend on the direction of @xmath59 , one can write @xmath862 , so that@xmath863 inserting expression ( [ srv ] ) for @xmath788 and replacing the sum over @xmath59 by an integral , we have@xmath864 in particular , for 3d one has from eq .
( [ ssr13])@xmath865 for @xmath275d , eq .
( [ ssr13 ] ) can be rewritten is the form @xmath866 so , the only difference of the expression for @xmath867 from @xmath868 is that @xmath63 is multiplied by @xmath846 . since for the minimizing parameters @xmath264 and @xmath200 , which enter @xmath860 , scaling is determined by the same product @xmath63 with @xmath846 [ see eq .
( [ srvw ] ) ] , we can write @xmath869 so that @xcite@xmath870 and@xmath871 the polaron mass at zero temperature can be obtained from the impedance function as @xcite@xmath872 so that from the scaling relation ( [ ssr16 ] ) for the memory function we also have a scaling relation for the polaron mass @xcite:@xmath873 since the mobility can be obtained from the memory function as @xcite@xmath874 fulfilment of the scaling relation ( [ ssr16 ] ) implies also a scaling relation for the mobility @xcite:@xmath875 in the important particular case of 2d , the above scaling relations take the form @xcite:@xmath876@xmath877@xmath878@xmath879 the fulfilment of the pd - scaling relation @xcite is checked for the path integral monte carlo results @xcite for the polaron free energy .
the path integral monte carlo results of ref.@xcite for the polaron free energy in 3d and in 2d are given for a few values of temperature and for some selected values of @xmath287 for a check of the scaling relation , the values of the polaron free energy at @xmath880 are taken from ref .
@xcite in 3d ( table i , for 4 values of @xmath63 ) and in 2d ( table ii , for 2 values of @xmath63 ) and plotted in fig .
[ sccomp ] , upper panel , with squares and open circles , correspondingly .
the pd - scaling relation for the polaron ground - state energy as derived in ref .
@xcite reads:@xmath881 in fig .
[ sccomp ] , lower panel , the available data for the free energy from ref @xcite are plotted in the following form _ inspired by the l.h.s . and the r.h.s parts of eq .
( 1 ) _ : @xmath882 ( squares ) and @xmath883(open triangles ) .
as follows from the figure , t__he path integral monte carlo results for the polaron free energy in 2d and 3d very closely follow the pd - scaling relation of the form given by eq .
( [ e]):__@xmath884 [ h ] part1fig11.eps inspired by the work of tomonaga on quantum electrodynamics ( q. e. d. ) , lee , low and pines ( l.l.p . )
@xcite derived ( [ eq_5a ] ) and @xmath885 from a canonical transformation formulation , which establishes ( [ eq_5a ] ) as a variational upper bound for the ground - state energy .
the wave equation corresponding to the frhlich hamiltonian ( [ eq_1a ] ) is @xmath886 we shall take advantage of the fact that the total momentum of the system @xmath887 ( where @xmath888 * * is the momentum of the electron ) is a constant of motion because it commutes with the hamiltonian ( [ eq_1a ] ) indeed,@xmath889 & = [ \mathbf{p},\sum_{\mathbf{k}}(v_{k}a_{\mathbf{k}}e^{i\mathbf{k\cdot r}}+v_{k}^{\ast}a_{\mathbf{k}}^{\dag } e^{-i\mathbf{k\cdot r}})]=\sum_{\mathbf{k}}(v_{k}a_{\mathbf{k}}\left [ \mathbf{p},e^{i\mathbf{k\cdot r}}\right ] + v_{k}^{\ast}a_{\mathbf{k}}^{\dag } \left [ \mathbf{p},e^{-i\mathbf{k\cdot r}}\right ] ) \\ & = \sum_{\mathbf{k}}\hbar\mathbf{k}(v_{k}a_{\mathbf{k}}e^{i\mathbf{k\cdot r}}-v_{k}^{\ast}a_{\mathbf{k}}^{\dag}e^{-i\mathbf{k\cdot r}});\end{aligned}\ ] ] @xmath890 & = \left [ \sum_{\mathbf{k}}\hbar\mathbf{k}a_{\mathbf{k}}^{\dag } a_{\mathbf{k}},\sum_{\mathbf{k}^{\prime}}(v_{k^{\prime}}a_{\mathbf{k}^{\prime}}e^{i\mathbf{k}^{\prime}\mathbf{\cdot r}}+v_{k^{\prime}}^{\ast } a_{\mathbf{k}^{\prime}}^{\dag}e^{-i\mathbf{k}^{\prime}\mathbf{\cdot r}})\right ] = \\ & = \sum_{\mathbf{k}}\hbar\mathbf{k}\left [ a_{\mathbf{k}}^{\dag}a_{\mathbf{k}},(v_{k}a_{\mathbf{k}}e^{i\mathbf{k\cdot r}}+v_{k}^{\ast } a_{\mathbf{k}}^{\dag}e^{-i\mathbf{k\cdot r}})\right ] = \\ & = \sum_{\mathbf{k}}\hbar\mathbf{k}\left\ { v_{k}\left [ a_{\mathbf{k}}^{\dag}a_{\mathbf{k}},a_{\mathbf{k}}\right ] e^{i\mathbf{k\cdot r}}+v_{k}^{\ast}\left [ a_{\mathbf{k}}^{\dag}a_{\mathbf{k}},a_{\mathbf{k}}^{\dag } \right ] e^{-i\mathbf{k\cdot r}}\right\ } = \\ & = -\sum_{\mathbf{k}}\hbar\mathbf{k}\left ( v_{k}a_{\mathbf{k}}e^{i\mathbf{k\cdot r}}-v_{k}^{\ast}a_{\mathbf{k}}^{\dag}e^{-i\mathbf{k\cdot r}}\right ) ; \end{aligned}\ ] ] @xmath891 = \left [ \sum_{\mathbf{k}}\hbar \mathbf{k}a_{\mathbf{k}}^{\dag}a_{\mathbf{k}}+\mathbf{p},h\right ] = 0 . \label{phcomm}\ ] ] because of the commutation ( [ phcomm ] ) , the operators @xmath414 and @xmath892 have a common set of basis functions : @xmath893 and @xmath894 it is possible to transform to a representation in which @xmath892 becomes a c number @xmath87 , and in which the hamiltonian no longer contains the electron coordinates .
the unitary ( canonical ) transformation required is @xmath895 , where @xmath896 .
\label{eq_6a}\ ] ] derivation of the transformations of the operators .
\mathbf{p}\exp\left [ \frac{i}{\hbar}(\mathbf{p}-\sum_{\mathbf{k}}\hbar\mathbf{k}a_{\mathbf{k}}^{\dagger}a_{\mathbf{k}})\mathbf{\cdot r}\right ] \nonumber\\ & = \exp\left [ -\frac{i}{\hbar}(\mathbf{p}-\sum_{\mathbf{k}}\hbar \mathbf{k}a_{\mathbf{k}}^{\dagger}a_{\mathbf{k}})\mathbf{\cdot r}\right ] \left ( -i\hbar\nabla\right ) \exp\left [ \frac{i}{\hbar}(\mathbf{p}-\sum_{\mathbf{k}}\hbar\mathbf{k}a_{\mathbf{k}}^{\dagger}a_{\mathbf{k}})\mathbf{\cdot r}\right ] \nonumber\\ & = \exp\left [ -\frac{i}{\hbar}(\mathbf{p}-\sum_{\mathbf{k}}\hbar \mathbf{k}a_{\mathbf{k}}^{\dagger}a_{\mathbf{k}})\mathbf{\cdot r}\right ] \left\ { \begin{array } [ c]{c}(\mathbf{p}-\sum_{\mathbf{k}}\hbar\mathbf{k}a_{\mathbf{k}}^{\dagger } a_{\mathbf{k}})\exp\left [ \frac{i}{\hbar}(\mathbf{p}-\sum_{\mathbf{k}}\hbar\mathbf{k}a_{\mathbf{k}}^{\dagger}a_{\mathbf{k}})\mathbf{\cdot r}\right ] \\ + \exp\left [ \frac{i}{\hbar}(\mathbf{p}-\sum_{\mathbf{k}}\hbar\mathbf{k}a_{\mathbf{k}}^{\dagger}a_{\mathbf{k}})\mathbf{\cdot r}\right ] \left ( -i\hbar\nabla\right ) \end{array }
\right\ } \nonumber\\ & = \mathbf{p}-\sum_{\mathbf{k}}\hbar\mathbf{k}a_{\mathbf{k}}^{\dagger } a_{\mathbf{k}}+\mathbf{p , } \label{llp3a}\ ] ] @xmath898 \left ( \sum_{\mathbf{k}}\hbar\mathbf{k}a_{\mathbf{k}}^{\dag}a_{\mathbf{k}}+\mathbf{p}\right ) \exp\left [ \frac{i}{\hbar}(\mathbf{p}-\sum_{\mathbf{k}}\hbar\mathbf{k}a_{\mathbf{k}}^{\dagger}a_{\mathbf{k}})\mathbf{\cdot r}\right ] \nonumber\\ & = \exp\left [ \frac{i}{\hbar}\sum_{\mathbf{k}}\hbar\mathbf{k}a_{\mathbf{k}}^{\dagger}a_{\mathbf{k}}\mathbf{\cdot r}\right ] \sum_{\mathbf{k}}\hbar\mathbf{k}a_{\mathbf{k}}^{\dag}a_{\mathbf{k}}\exp\left [ -\frac{i}{\hbar } \sum_{\mathbf{k}}\hbar\mathbf{k}a_{\mathbf{k}}^{\dagger}a_{\mathbf{k}}\mathbf{\cdot r}\right ] + s_{1}^{-1}\mathbf{p}s_{1}\nonumber\\ & = \sum_{\mathbf{k}}\hbar\mathbf{k}a_{\mathbf{k}}^{\dag}a_{\mathbf{k}}+\mathbf{p}-\sum_{\mathbf{k}}\hbar\mathbf{k}a_{\mathbf{k}}^{\dagger } a_{\mathbf{k}}+\mathbf{p = p}+\mathbf{p , } \label{llp3b}\ ] ] @xmath899 a_{\mathbf{k}}\exp\left [ \frac{i}{\hbar}(\mathbf{p}-\sum_{\mathbf{k}}\hbar\mathbf{k}a_{\mathbf{k}}^{\dagger}a_{\mathbf{k}})\mathbf{\cdot r}\right ] \nonumber\\ & = \exp\left [ \frac{i}{\hbar}\sum_{\mathbf{k}}\hbar\mathbf{k}a_{\mathbf{k}}^{\dagger}a_{\mathbf{k}}\mathbf{\cdot r}\right ] a_{\mathbf{k}}\exp\left [ -\frac{i}{\hbar}\sum_{\mathbf{k}}\hbar\mathbf{k}a_{\mathbf{k}}^{\dagger } a_{\mathbf{k}}\mathbf{\cdot r}\right ] \nonumber\\ & = \exp\left [ i\mathbf{k}a_{\mathbf{k}}^{\dagger}a_{\mathbf{k}}\mathbf{\cdot r}\right ] a_{\mathbf{k}}\exp\left [ -i\mathbf{k}a_{\mathbf{k}}^{\dagger } a_{\mathbf{k}}\mathbf{\cdot r}\right ] \nonumber\\ & = \exp\left [ i\mathbf{k}a_{\mathbf{k}}^{\dagger}a_{\mathbf{k}}\mathbf{\cdot r}\right ] a_{\mathbf{k}}{\displaystyle\sum\nolimits_{n=0}^{\infty } } \frac{1}{n!}\left ( -i\mathbf{k}a_{\mathbf{k}}^{\dagger}a_{\mathbf{k}}\mathbf{\cdot r}\right ) ^{n}\nonumber\\ & = \exp\left [ i\mathbf{k}a_{\mathbf{k}}^{\dagger}a_{\mathbf{k}}\mathbf{\cdot r}\right ] { \displaystyle\sum\nolimits_{n=0}^{\infty } } \frac{1}{n!}a_{\mathbf{k}}\left ( -i\mathbf{k}a_{\mathbf{k}}^{\dagger } a_{\mathbf{k}}\mathbf{\cdot r}\right ) ^{n}\nonumber\\ & = \exp\left [ i\mathbf{k\cdot r}a_{\mathbf{k}}^{\dagger}a_{\mathbf{k}}\right ] { \displaystyle\sum\nolimits_{n=0}^{\infty } } \frac{1}{n!}\left ( -i\mathbf{k\cdot r}\right ) ^{n}a_{\mathbf{k}}\left ( a_{\mathbf{k}}^{\dagger}a_{\mathbf{k}}\right ) ^{n}\overset{\text{see}(\ast ) } { = } \nonumber\\ & = \exp\left [ i\mathbf{k\cdot r}a_{\mathbf{k}}^{\dagger}a_{\mathbf{k}}\right ] { \displaystyle\sum\nolimits_{n=0}^{\infty } } \frac{1}{n!}[-i\mathbf{k\cdot r}\left ( a_{\mathbf{k}}^{\dagger}a_{\mathbf{k}}+1\right ) ] ^{n}a_{\mathbf{k}}\nonumber\\ & = \exp\left [ i\mathbf{k\cdot r}a_{\mathbf{k}}^{\dagger}a_{\mathbf{k}}\right ] \exp\left [ -i\mathbf{k\cdot r(}a_{\mathbf{k}}^{\dagger } a_{\mathbf{k}}+1)\right ] a_{\mathbf{k}}\nonumber\\ & = a_{\mathbf{k}}\exp\left ( -i\mathbf{k\cdot r}\right ) .
\label{llp3c1}\ ] ] here the property was used : @xmath900 it is evident for @xmath901for @xmath902 it is demonstrated as follows : @xmath903 then for @xmath904 the validity of ( [ * ] ) is straightforwardly demonstrated by induction . finally , @xmath905^{\dag}=a_{\mathbf{k}}^{\dag}\exp\left ( i\mathbf{k\cdot r}\right ) .
\label{llp3d}\ ] ] using ( [ llp3a ] ) , ( [ llp3b ] ) , ( [ llp3c1 ] ) and ( [ llp3d ] ) , one arrives at @xmath906 where @xmath56 is set @xmath907 . leads to @xmath908at the same time , applying eq .
( [ llp3b ] ) , we obtain @xmath909setting the gauge @xmath910 eliminates the operator @xmath56 from the problem . ] the wave equation ( [ llp1 ] ) takes the form @xmath911 our aim is to calculate for a given momentum @xmath87 the lowest eigenvalue @xmath912 of the hamiltonian ( [ llp3e ] ) . for the low - lying energy levels of the electron @xmath912
may be well represented by the first two terms of a power series expansion in @xmath913where @xmath914 is the effective mass of the polaron .
the canonical transformation ( [ eq_6a ] ) formally eliminates the electron operators from the hamiltonian .
llp use further a variational method of calculation .
the trial wave function is chosen as @xmath915 where @xmath916 is the eigenstate of the unperturbed hamiltonian with no phonons present ( vacuum state ) .
specifically , @xmath916 is defined by @xmath917 and the second canonical transformation : @xmath918 , \label{eq_6b}\ ] ] where @xmath919 are treated as variational functions and will be chosen to minimize the energy .
the physical significance of eq.([eq_6b ] ) is that it dresses the electron with the virtual phonon field , which describes the polarization .
viewed as a unitary transformation , @xmath327 is a _ displacement _ operator on @xmath58 and @xmath350 @xmath920 a_{\mathbf{k}}\exp\left [ \sum_{\mathbf{k}}(a_{\mathbf{k}}^{\dagger}f_{\mathbf{k}}-a_{\mathbf{k}}f_{\mathbf{k}}^{\ast})\right ] \nonumber\\ & = \exp\left [ -(a_{\mathbf{k}}^{\dagger}f_{\mathbf{k}}-a_{\mathbf{k}}f_{\mathbf{k}}^{\ast})\right ] a_{\mathbf{k}}\exp\left [ ( a_{\mathbf{k}}^{\dagger}f_{\mathbf{k}}-a_{\mathbf{k}}f_{\mathbf{k}}^{\ast})\right ] \overset{\text{see}(\ast\ast)}{=}\nonumber\\ & = a_{\mathbf{k}}+\left [ a_{\mathbf{k}},(a_{\mathbf{k}}^{\dagger } f_{\mathbf{k}}-a_{\mathbf{k}}f_{\mathbf{k}}^{\ast})\right ] + \frac{1}{2}\left [ \left [ a_{\mathbf{k}},(a_{\mathbf{k}}^{\dagger}f_{\mathbf{k}}-a_{\mathbf{k}}f_{\mathbf{k}}^{\ast})\right ] , ( a_{\mathbf{k}}^{\dagger } f_{\mathbf{k}}-a_{\mathbf{k}}f_{\mathbf{k}}^{\ast})\right ] + ... \nonumber\\ & = a_{\mathbf{k}}+f_{\mathbf{k } } , \label{llp6a}\ ] ] @xmath921 here the relation was used@xmath922 a\exp\left [ v\right ] = a+\left [ a , v\right ] + \frac { 1}{2}\left [ \left [ a , v\right ] , v\right ] + \frac{1}{3!}\left [ \left [ \left [ a , v\right ] , v\right ] , v\right ] + ... \tag{**}\label{**}\ ] ] further we seek to minimize the expression for the energy , @xmath923 in virtue of ( [ llp6a ] ) and ( [ llp6b ] ) , we obtain : @xmath924 ^{2}}{2m_{b}}\\ & + \sum_{\mathbf{k}}\hbar\omega_{\mathrm{lo}}\left ( a_{\mathbf{k}}^{\dag } + f_{\mathbf{k}}^{\ast}\right ) \left ( a_{\mathbf{k}}+f_{\mathbf{k}}\right ) + \sum_{\mathbf{k}}\left [ v_{k}\left ( a_{\mathbf{k}}+f_{\mathbf{k}}\right ) + v_{k}^{\ast}\left ( a_{\mathbf{k}}^{\dag}+f_{\mathbf{k}}^{\ast}\right ) \right ] \\ & = \frac{{\left [ ( \mathbf{p}-\sum_{\mathbf{k}}\hbar\mathbf{k}a_{\mathbf{k}}^{\dag}a_{\mathbf{k}})-\sum_{\mathbf{k}}\hbar\mathbf{k}\left\vert f_{\mathbf{k}}\right\vert ^{2}-\sum_{\mathbf{k}}\hbar\mathbf{k}\left ( a_{\mathbf{k}}^{\dag}f_{\mathbf{k}}+a_{\mathbf{k}}f_{\mathbf{k}}^{\ast } \right ) \right ] ^{2}}}{2m_{b}}\\ & + \sum_{\mathbf{k}}\hbar\omega_{\mathrm{lo}}\left ( a_{\mathbf{k}}^{\dag } a_{\mathbf{k}}+\left\vert f_{\mathbf{k}}\right\vert ^{2}+a_{\mathbf{k}}^{\dag}f_{\mathbf{k}}+a_{\mathbf{k}}f_{\mathbf{k}}^{\ast}\right ) \\ & + \sum_{\mathbf{k}}\left [ v_{k}\left ( a_{\mathbf{k}}+f_{\mathbf{k}}\right ) + v_{k}^{\ast}\left ( a_{\mathbf{k}}^{\dag}+f_{\mathbf{k}}^{\ast } \right ) \right ] \\ & = h_{0}+h_{1},\end{aligned}\ ] ] where@xmath925 ^{2}+\left [ \sum_{\mathbf{k}}\hbar\mathbf{k}\left\vert f_{\mathbf{k}}\right\vert ^{2}\right ] ^{2}}}{2m_{b}}+\sum_{\mathbf{k}}\left [ v_{k}f_{\mathbf{k}}+v_{k}^{\ast } f_{\mathbf{k}}^{\ast}\right ] \nonumber\\ & + \sum_{\mathbf{k}}\left\vert f_{\mathbf{k}}\right\vert ^{2}\left\ { \hbar\omega_{\mathrm{lo}}-\frac{\hbar\mathbf{k\cdot p}}{m_{b}}+\frac{\hbar ^{2}k^{2}}{2m_{b}}\right\ } + \frac{\hbar^{2}}{m_{b}}\sum_{\mathbf{k}}\mathbf{k}a_{\mathbf{k}}^{\dag}a_{\mathbf{k}}\cdot\sum_{\mathbf{k}^{\prime}}\mathbf{k}^{\prime}\left\vert f_{\mathbf{k}^{\prime}}\right\vert ^{2}\nonumber\\ & + \sum_{\mathbf{k}}a_{\mathbf{k}}\left\ { v_{k}+f_{\mathbf{k}}^{\ast}\left [ \hbar\omega_{\mathrm{lo}}-\frac{\hbar\mathbf{k\cdot p}}{m_{b}}+\frac{\hbar ^{2}k^{2}}{2m_{b}}+\frac{\hbar^{2}\mathbf{k}}{m_{b}}\cdot\sum_{\mathbf{k}^{\prime}}\mathbf{k}^{\prime}\left\vert f_{\mathbf{k}^{\prime}}\right\vert ^{2}\right ] \right\ } \nonumber\\ & + \sum_{\mathbf{k}}a_{\mathbf{k}}^{\dag}\left\ { v_{k}^{\ast}+f_{\mathbf{k}}\left [ \hbar\omega_{\mathrm{lo}}-\frac{\hbar\mathbf{k\cdot p}}{m_{b}}+\frac{\hbar^{2}k^{2}}{2m_{b}}+\frac{\hbar^{2}\mathbf{k}}{m_{b}}\cdot \sum_{\mathbf{k}^{\prime}}\mathbf{k}^{\prime}\left\vert f_{\mathbf{k}^{\prime } } \right\vert ^{2}\right ] \right\ } \nonumber\\ & + \sum_{\mathbf{k}}\hbar\omega_{\mathrm{lo}}a_{\mathbf{k}}^{\dag } a_{\mathbf{k } } ; \label{llp7a}\ ] ] @xmath926 using ( [ llp6 ] ) , we obtain from ( [ llp7 ] ) that@xmath927 ^{2}}}{2m_{b}}+\sum_{\mathbf{k}}\left [ v_{k}f_{\mathbf{k}}+v_{k}^{\ast}f_{\mathbf{k}}^{\ast}\right ] \nonumber\\ & + \sum_{\mathbf{k}}\left\vert f_{\mathbf{k}}\right\vert ^{2}\left\ { \hbar\omega_{\mathrm{lo}}-\frac{\hbar\mathbf{k\cdot p}}{m_{b}}+\frac{\hbar ^{2}k^{2}}{2m_{b}}\right\ } . \label{llp8}\ ] ] we minimize ( [ llp8 ] ) by imposing @xmath928 this results in @xmath929 \cdot}\mathbf{k}\right\ } = 0 \label{llp9}\ ] ] and the appropriate complex conjugate equation for @xmath919 . upon comparing ( [ llp9 ] ) and
( [ llp7a ] ) , we see that the linear terms in @xmath245 and @xmath58 are identically zero if ( [ llp9 ] ) is satisfied , and hence that @xmath930 is diagonal in a representation in which @xmath931 is diagonal .
so , the variational calculation by llp is equivalent to the use of ( [ llp7a ] ) as the total hamiltonian provided @xmath932 satisfies ( [ llp9 ] ) .
an estimate of the accuracy of the llp variational procedure was obtained by an estimate of the effect of @xmath346using a perturbation theory .
now we evaluate the energy of the ground state of the system , which is given by eq .
( [ llp8 ] ) with @xmath932 satisfying eq .
( [ llp9 ] ) .
the only preferred direction in this problem is @xmath87 .
therefore one may introduce the parameter @xmath933 defined as@xmath934 then eq . ( [ llp9 ] ) leads to @xmath935 \right . , \label{fk}\ ] ] and we obtain the following implicit equation for @xmath933:@xmath936 ^{2}\right . \\ & = \frac{v}{\left ( 2\pi\right ) ^{3}}\int d^{3}k\hbar\mathbf{k}\left ( \frac{\hbar\omega_{\mathrm{lo}}}{k}\right ) ^{2}\frac{4\pi\alpha}{v}\left ( \frac{\hbar}{2m_{b}\omega_{\mathrm{lo}}}\right ) ^{\frac{1}{2}}\left/ \left [ \hbar\omega_{\mathrm{lo}}-\frac{\hbar\mathbf{k\cdot p}}{m_{b}}(1-\eta ) + \frac{\hbar^{2}k^{2}}{2m_{b}}\right ] ^{2}\right .
.\end{aligned}\ ] ] let us introduce spherical coordinates with a polar axis along @xmath87 and denote @xmath937:@xmath938 ^{2}\right . \\ & = \frac{\alpha\hbar}{2\pi^{2}}\left ( \frac{\hbar}{2m_{b}\omega _ { \mathrm{lo}}}\right ) ^{\frac{1}{2}}2\pi{\displaystyle\int\nolimits_{-1}^{1 } } dxx{\displaystyle\int\nolimits_{0}^{\infty } } dkk\left/ \left [ 1 - 2\frac{\hbar kpx}{2m_{b}\hbar\omega_{\mathrm{lo}}}(1-\eta)+\frac{\hbar^{2}k^{2}}{2m_{b}\hbar\omega_{\mathrm{lo}}}\right ] ^{2}\right .
.\end{aligned}\ ] ] further , we introduce the parameter @xmath939 and a new variable@xmath940 this gives@xmath941 ^{2}\right .
\\ & = \frac{\alpha}{\pi}\frac{\left ( 2m_{b}\hbar\omega_{\mathrm{lo}}\right ) ^{1/2}}{p}{\displaystyle\int\nolimits_{-1}^{1 } } dxx{\displaystyle\int\nolimits_{0}^{\infty } } d\kappa\kappa\left/ \left [ ( \kappa - qx)^{2}+(1-q^{2}x^{2})\right ] ^{2}\right . \\ & = \frac{\alpha}{\pi}\frac{\left ( 2m_{b}\hbar\omega_{\mathrm{lo}}\right ) ^{1/2}}{p}{\displaystyle\int\nolimits_{-1}^{1 } } dxx{\displaystyle\int\nolimits_{-qx}^{\infty } } d\kappa\left ( \kappa+qx\right ) \left/ \left [ \kappa^{2}+(1-q^{2}x^{2})\right ] ^{2}\right .
\\ & = \frac{\alpha}{\pi}\frac{\left ( 2m_{b}\hbar\omega_{\mathrm{lo}}\right ) ^{1/2}}{p}{\displaystyle\int\nolimits_{-1}^{1 } } dxx\left\ { \begin{array } [ c]{c}-\frac{1}{2\left [ \kappa^{2}+(1-q^{2}x^{2})\right ] } \\ + qx\left [ \frac{\kappa}{2(1-q^{2}x^{2})\left [ \kappa^{2}+(1-q^{2}x^{2})\right ] } + \frac{1}{2(1-q^{2}x^{2})^{3/2}}\arctan\left ( \frac{\kappa } { \left [ 1-q^{2}x^{2}\right ] ^{1/2}}\right ) \right ] \end{array } \right\ } _ { -qx}^{\infty}\\ & = \frac{\alpha}{\pi}\frac{\left ( 2m_{b}\hbar\omega_{\mathrm{lo}}\right ) ^{1/2}}{p}{\displaystyle\int\nolimits_{-1}^{1 } } dxx\left\ { \frac{1}{2}+\frac{qx\pi}{4(1-q^{2}x^{2})^{3/2}}+\frac{q^{2}x^{2}}{2(1-q^{2}x^{2})}+\frac{qx}{2(1-q^{2}x^{2})^{3/2}}\arcsin\left ( qx\right ) \right\ } .\end{aligned}\]]@xmath542@xmath942 so , we have arrived at the equation@xmath943 or equivalently , using the definition ( [ q]),@xmath944 using eqs .
( [ etap ] ) and ( [ fk ] ) , we can simplify the energy ( [ llp8 ] ) as follows:@xmath945@xmath946@xmath947@xmath948@xmath949@xmath950@xmath542@xmath951 the sum over @xmath59 in eq .
( [ en ] ) is calculated as follows:@xmath952 } \\ & = \frac{m_{b}\omega_{\mathrm{lo}}^{2}\alpha}{\pi^{2}}\left ( \frac{\hbar } { 2m_{b}\omega_{\mathrm{lo}}}\right ) ^{\frac{1}{2}}\int d\mathbf{k}\frac { 1}{k^{2}\left ( k^{2}-2\frac{\mathbf{k\cdot p}}{\hbar}(1-\eta)+\frac { 2m_{b}\omega_{\mathrm{lo}}}{\hbar}\right ) } .\end{aligned}\ ] ] for the calculation of this integral , we can use the auxiliary identity@xmath953 ^{2}}. \label{id}\ ] ] setting@xmath954 we find@xmath955 ^{2}}\\ & = \frac{m_{b}\omega_{\mathrm{lo}}^{2}\alpha}{\pi^{2}}\left ( \frac{\hbar } { 2m_{b}\omega_{\mathrm{lo}}}\right ) ^{\frac{1}{2}}\int_{0}^{1}dx\int d\mathbf{k}\frac{1}{\left ( k^{2}-2\frac{\mathbf{k\cdot p}}{\hbar}(1-\eta)x+\frac{2m_{b}\omega_{\mathrm{lo}}}{\hbar}x\right ) ^{2}}\\ & = \frac{m_{b}\omega_{\mathrm{lo}}^{2}\alpha}{\pi^{2}}\left ( \frac{\hbar } { 2m_{b}\omega_{\mathrm{lo}}}\right ) ^{\frac{1}{2}}\int_{0}^{1}dx\int d\mathbf{k}\frac{1}{\left ( \left ( \mathbf{k-}\frac{\mathbf{p}}{\hbar}(1-\eta)x\right ) ^{2}+\frac{2m_{b}\omega_{\mathrm{lo}}}{\hbar}x-\frac{p^{2}}{\hbar^{2}}(1-\eta)^{2}x^{2}\right ) ^{2}}\\ & = \frac{m_{b}\omega_{\mathrm{lo}}^{2}\alpha}{\pi^{2}}\left ( \frac{\hbar } { 2m_{b}\omega_{\mathrm{lo}}}\right ) ^{\frac{1}{2}}\int_{0}^{1}dx\int d\mathbf{k}\frac{1}{\left ( k^{2}+\frac{2m_{b}\omega_{\mathrm{lo}}}{\hbar } x-\frac{p^{2}}{\hbar^{2}}(1-\eta)^{2}x^{2}\right ) ^{2}}\ ] ] as long as @xmath956 is sufficiently small so that no spontaneous emission can occur ( roughly @xmath957 ) , the quantity @xmath958 is supposed to be positive for @xmath959 .
therefore , we can use the integral@xmath960 what gives@xmath961 we change the variable @xmath962 what gives@xmath963 and hence@xmath964 as a result , the energy ( [ en ] ) is expressed in a closed form@xmath965 further , we expand the r.h.s . of eq .
( [ llp27-a ] ) to the second order in powers of @xmath447 ( or , what is the same , in powers of @xmath966 ) using the relation@xmath967 what results in@xmath968 = \frac{\alpha}{6}+o\left ( q^{2}\right ) .\ ] ] solving this equation for @xmath933 , we find@xmath969 we also apply the expansion in powers of @xmath966 up to @xmath970 to the energy ( [ llp29]):@xmath971 finally , we have arrived at the llp polaron energy@xmath972 in this derivation it is shown that the approximation used in the evaluation of the function , which determines the polaron mass [ see eqs .
( 40 ) and ( b1 ) from ref .
@xcite ] @xmath973 \exp\left [ k^{2}c(\tau)\right ] \label{b1}\ ] ] with @xmath974 is equivalent to an expansion in a continued fraction limited to the first step .
moreover , it is proved that the choice of the coefficients of the continued fraction can be justified by a variational principle , at least when @xmath975 is real and positive .
expanding the last exponential of eq .
( [ b1 ] ) in a power series leads to @xmath976 \int_{-\infty}^{0}d\tau e^{iz\tau}{\displaystyle\sum\limits_{n=0}^{\infty } } \frac{1}{n!}\left ( \frac{1}{3m^{2}}\right ) ^{n}\nonumber\\ & \times{\displaystyle\sum\limits_{\vec{k}_{1}, ... ,\vec{k}_{n } } } \frac{k_{1}^{2}k_{2}^{2} ... k_{n}^{2}\left\vert f_{k_{1}}\right\vert ^{2}\left\vert f_{k_{2}}\right\vert ^{2} ... \left\vert f_{k_{n}}\right\vert ^{2}}{\gamma_{k_{1}}^{2}\gamma_{k_{2}}^{2} ... \gamma_{k_{n}}^{2}}\nonumber\\ & \times\exp\left [ i(\gamma_{k_{1}}+\gamma_{k_{2}}+ ... \gamma_{k_{n}}^{2})\tau\right ] \nonumber\\ & = -i\exp\left [ -k^{2}c(0)\right ] { \displaystyle\sum\limits_{n=0}^{\infty } } \frac{(3m^{2})^{-n}}{n!}\label{b3}\\ & \times{\displaystyle\sum\limits_{\vec{k}_{1}, ...
,\vec{k}_{n } } } \frac{k_{1}^{2}k_{2}^{2} ... k_{n}^{2}\left\vert f_{k_{1}}\right\vert ^{2}\left\vert f_{k_{2}}\right\vert ^{2} ... \left\vert f_{k_{n}}\right\vert ^{2}}{\gamma_{k_{1}}^{2}\gamma_{k_{2}}^{2} ... \gamma_{k_{n}}^{2}}\frac { 1}{\gamma_{k_{1}}+\gamma_{k_{2}}+ ... +\gamma_{k_{n}}+z}.\nonumber\end{aligned}\ ] ] the multiple sum over the @xmath832 s is in fact an integral with 3n variables .
it is possible to change the variables in that one of the new variables is @xmath977 then the multiple sum which appears in the last term of eq .
( [ b3 ] ) is of the following type : @xmath978 where @xmath979 is the result of the integration over the @xmath793 other variables .
an expansion of integrals of the type ( [ b5 ] ) into stieltjes continued fractions is known to give good results when @xmath975 is real and not located on the cut of @xmath980 , i.e. , when @xmath981 the first nontrivial step in the continued fraction expansion is @xmath982 with @xmath983 @xmath984 a _ variational principle _ can be established , which gives a rather strong argument in favour of the approximation ( [ b7 ] ) .
let us introduce a variational parameter @xmath985 writing @xmath986 performing two steps of the division , this relation becomes @xmath987 with @xmath988 the approximation consists of neglecting the term @xmath989 in eq .
( [ b11 ] ) . as this term is positive [ cf .
( [ b6 ] ) ] , the best approximation is obtained when it is minimum .
therefore let us use the freedom in the choice of @xmath985 to minimize the expression ( [ b12 ] ) , @xmath990 which gives @xmath991 or @xmath992 this provides us with the best value of the variational parameter @xmath993 which is @xmath994[cf .
( [ b9 ] ) ] . with this value of @xmath985 and neglecting @xmath989 ,
the expression ( [ b11 ] ) of the calculated quantity @xmath980 becomes @xmath995 which is the first step ( [ b7 ] ) of a stieltjes continued fraction . to prove that this value of @xmath985 gives a minimum of @xmath989 ,
let us calculate the second derivative @xmath996 now the parameter @xmath985 is replaced by its expression ( [ b16 ] ) .
the relation ( [ b18 ] ) becomes @xmath997 which is positive of @xmath998 , since it follows from relation ( [ b16 ] ) that @xmath999 our approximation is related to that used by feynman which is based on the following inequality : @xmath1000 where the brackets denote the expectation value of the random variable @xmath384 .
for instance , @xmath1001 where l(x ) is the non - normalized probability density of @xmath384 .
the laplace transform of eq .
( [ b19 ] ) gives @xmath1002 which after integration becomes @xmath1003 the last inequality shows the relation with our procedure .
the derivations in the present section are based on ref .
the hamiltonian of a system of @xmath242 interacting continuum polarons is given by : @xmath1004 where @xmath1005 represent the position and momentum of the @xmath242 constituent electrons ( or holes ) with band mass @xmath55 ; @xmath1006 denote the creation and annihilation operators for longitudinal optical ( lo ) phonons with wave vector @xmath1007 and frequency @xmath2 ; @xmath1008 describes the amplitude of the interaction between the electrons and the phonons ; and @xmath19 is the elementary electron charge .
this hamiltonian can be subdivided into the following parts : @xmath1009 where@xmath1010 is the kinetic energy of electrons,@xmath1011 is the potential energy of the coulomb electron - electron interaction,@xmath1012 is the hamlitonian of phonons , and@xmath1013 is the hamiltonian of the electron - phonon interaction .
further on , we use the second quantization formalism for electrons , in which the terms @xmath1014 , @xmath1015 and @xmath1016 are@xmath1017 where @xmath1018 is the symbol of the normal product of operators,@xmath1019 is the fourier component of the coulomb potential , and@xmath1020 is the fourier component of the electron density .
the ground state energy of the many - polaron hamiltonian ( [ hmpol ] ) has been studied by l. lemmens , j. t. devreese and f. brosens ( ldb ) @xcite , for weak and intermediate strength of the electron - phonon coupling .
they introduce a variational wave function : @xmath1021 where @xmath1022 represents the ground - state many - body wave function for the electron ( or hole ) system and @xmath1023 is the phonon vacuum , @xmath1024 is a many - body unitary operator which determines a canonical transformation for a fermion gas interacting with a boson field : @xmath1025 in ref .
@xcite , this canonical transformation was used to establish a many - fermion theory .
the @xmath1026 were determined variationally @xcite resulting in @xmath1027 for a system with total momentum @xmath1028 . in this expression
, @xmath1029 represents the static structure factor of the constituent interacting many electron or hole system : @xmath1030 the angular brackets @xmath1031 represent the expectation value with respect to the ground state .
the many - polaron optical conductivity is the response of the current - density , in the system described by the hamiltonian ( [ hmpol ] ) , to an applied electric field ( along the @xmath384-axis ) with frequency @xmath856 .
this applied electric field introduces a perturbation term in the hamiltonian ( [ hmpol ] ) , which couples the vector potential of the incident electromagnetic field to the current - density . within linear response theory ,
the optical conductivity can be expressed through the kubo formula as a current - current correlation function : @xmath1032 \right\rangle dt.\ ] ] in this expression , @xmath305 is the volume of the system , and @xmath1033 is the @xmath384-component of the current operator @xmath1034 which is related to the momentum operators of the charge carriers : @xmath1035 with @xmath966 the charge of the charge carriers ( @xmath1036 for holes , @xmath1037 for electrons ) and @xmath87 the total momentum operator of the charge carriers .
the real part of the optical conductivity at temperature zero , which is proportional to the optical absorption coefficient , can be written as a function of the total momentum operator of the charge carriers as follows : @xmath1038 \right\rangle dt\right\ } .
\label{k1}\ ] ] let us integrate over time in ( [ k1 ] ) twice by parts as follows : @xmath1039 \right\rangle e^{i\omega t-\delta t}\\ & = \frac{1}{i\omega-\delta}\left\ { \left .
\left\langle \left [ p_{x}\left ( t\right ) , p_{x}\right ] \right\rangle e^{i\omega t-\delta t}\right\vert _ { t=0}^{\infty}-\int_{0}^{\infty}dt\left\langle \left [ \frac{d}{dt}p_{x}\left ( t\right ) , p_{x}\right ] \right\rangle e^{i\omega t-\delta t}\right\ } \\ &
= -\frac{1}{i\omega-\delta}\int_{0}^{\infty}dt\left\langle \left [ \frac { d}{dt}\left ( e^{\frac{it}{\hbar}h}p_{x}e^{-\frac{it}{\hbar}h}\right ) , p_{x}\right ] \right\rangle e^{i\omega t-\delta t}\\ & = -\frac{1}{i\omega-\delta}\int_{0}^{\infty}dt\left\langle \left [ \left ( e^{\frac{it}{\hbar}h}\frac{i}{\hbar}\left [ h , p_{x}\right ] e^{-\frac { it}{\hbar}h}\right ) , p_{x}\right ] \right\rangle e^{i\omega t-\delta t}\\ & = -\frac{1}{i\omega-\delta}\int_{0}^{\infty}dt\left\langle \left [ \frac { i}{\hbar}\left [ h , p_{x}\right ] , e^{-\frac{it}{\hbar}h}p_{x}e^{\frac { it}{\hbar}h}\right ] \right\rangle e^{i\omega t-\delta t}\\ & = -\frac{1}{i\omega-\delta}\int_{0}^{\infty}dt\left\langle \left [ f_{x}\left ( 0\right ) , e^{-\frac{it}{\hbar}h}p_{x}e^{\frac{it}{\hbar}h}\right ] \right\rangle e^{i\omega t-\delta t}\\ & = -\left ( \frac{1}{i\omega-\delta}\right ) ^{2}\left\ { \left . \left\langle \left [ f_{x}\left ( 0\right ) , e^{-\frac{it}{\hbar}h}p_{x}e^{\frac{it}{\hbar}h}\right ] \right\rangle e^{i\omega t-\delta t}\right\vert _ { t=0}^{\infty}\right . \\ & \left .
-\int_{0}^{\infty}dt\left\langle \left [ f_{x}\left ( 0\right ) , \frac{d}{dt}e^{-\frac{it}{\hbar}h}p_{x}e^{\frac{it}{\hbar}h}\right ] \right\rangle e^{i\omega t-\delta t}\right\}\end{aligned}\ ] ] @xmath1040 \right\rangle + \int_{0}^{\infty}dt\left\langle \left [ f_{x}\left ( 0\right ) , e^{-\frac{it}{\hbar}h}\left ( \frac{i}{\hbar } \left [ h , p_{x}\right ] \right ) e^{\frac{it}{\hbar}h}\right ] \right\rangle e^{i\omega t-\delta t}\right\ } \\ & = -\left ( \frac{1}{i\omega-\delta}\right ) ^{2}\left\ { -\left\langle \left [ f_{x},p_{x}\right ] \right\rangle + \int_{0}^{\infty}dt\left\langle \left [ f_{x}\left ( 0\right ) , e^{-\frac{it}{\hbar}h}f_{x}\left ( 0\right ) e^{\frac{it}{\hbar}h}\right ] \right\rangle e^{i\omega t-\delta t}\right\ } \\ & = -\left ( \frac{1}{i\omega-\delta}\right ) ^{2}\left\ { -\left\langle \left [ f_{x},p_{x}\right ] \right\rangle + \int_{0}^{\infty}dt\left\langle \left [ e^{\frac{it}{\hbar}h}f_{x}\left ( 0\right ) e^{-\frac{it}{\hbar}h},f_{x}\left ( 0\right ) \right ] \right\rangle e^{i\omega t-\delta t}\right\ } \\ & = \frac{1}{\left ( \omega+i\delta\right ) ^{2}}\left\ { -\left\langle \left [ f_{x},p_{x}\right ] \right\rangle + \int_{0}^{\infty}dt\left\langle \left [ f_{x}\left ( t\right ) , f_{x}\left ( 0\right ) \right ] \right\rangle e^{i\omega t-\delta t}\right\ } , \end{aligned}\ ] ] where @xmath1041 $ ] is the operator of the force applied to the center of mass of the electrons .
since both @xmath1042 and @xmath1043 are hermitian operators , the average @xmath1044 \right\rangle $ ] is imaginary .
hence , for @xmath1045 this term does not give a contribution into @xmath1046 as a result , integrating by parts twice , the real part of the optical conductivity of the many - polaron system is written with a force - force correlation function : @xmath1047 \right\rangle dt\right\ } .
\label{ff}\ ] ] the force operator is determined as @xmath1048 = \frac{i}{\hbar}\left [ h_{e}+h_{e - e}+h_{ph}+h_{e - ph},p_{x}\right ] .\ ] ] further , we use the commutators : @xmath1049 = \sum_{\mathbf{k}^{\prime}}\hbar k_{x}^{\prime}\left [ a_{\mathbf{k+q},\sigma } ^{+}a_{\mathbf{k},\sigma},a_{\mathbf{k}^{\prime},\sigma}^{+}a_{\mathbf{k}^{\prime},\sigma}\right ] \\
= \sum_{\mathbf{k}^{\prime}}\hbar k_{x}^{\prime}\left ( \begin{array } [ c]{c}a_{\mathbf{k+q},\sigma}^{+}a_{\mathbf{k},\sigma}a_{\mathbf{k}^{\prime},\sigma } ^{+}a_{\mathbf{k}^{\prime},\sigma}+a_{\mathbf{k+q},\sigma}^{+}a_{\mathbf{k}^{\prime},\sigma}^{+}a_{\mathbf{k},\sigma}a_{\mathbf{k}^{\prime},\sigma}\\ -a_{\mathbf{k}^{\prime},\sigma}^{+}a_{\mathbf{k+q},\sigma}^{+}a_{\mathbf{k}^{\prime},\sigma}a_{\mathbf{k},\sigma}-a_{\mathbf{k}^{\prime},\sigma}^{+}a_{\mathbf{k}^{\prime},\sigma}a_{\mathbf{k+q},\sigma}^{+}a_{\mathbf{k},\sigma}\end{array } \right )
\\ = \sum_{\mathbf{k}^{\prime}}\hbar k_{x}^{\prime}\left ( \delta_{\mathbf{kk}^{\prime}}a_{\mathbf{k+q},\sigma}^{+}a_{\mathbf{k}^{\prime},\sigma}-\delta_{\mathbf{k}^{\prime},\mathbf{k+q}}a_{\mathbf{k}^{\prime},\sigma}^{+}a_{\mathbf{k},\sigma}\right ) \\
= \sum_{\mathbf{k}^{\prime}}\hbar k_{x}^{\prime}\left ( \delta_{\mathbf{kk}^{\prime}}a_{\mathbf{k+q},\sigma}^{+}a_{\mathbf{k},\sigma}-\delta _ { \mathbf{k}^{\prime},\mathbf{k+q}}a_{\mathbf{k+q},\sigma}^{+}a_{\mathbf{k},\sigma}\right )
\\ = a_{\mathbf{k+q},\sigma}^{+}a_{\mathbf{k},\sigma}\sum_{\mathbf{k}^{\prime}}\hbar k_{x}^{\prime}\left ( \delta_{\mathbf{kk}^{\prime}}-\delta _ { \mathbf{k}^{\prime},\mathbf{k+q}}\right ) = -\hbar q_{x}a_{\mathbf{k+q},\sigma}^{+}a_{\mathbf{k},\sigma},\end{gathered}\ ] ] @xmath1050 = -\hbar q_{x}\rho_{\mathbf{q}}.\ ] ] hence , @xmath1051 = 0,$ ] @xmath1052 = 0,$]@xmath1053 & = \sum_{\mathbf{q}}\left ( v_{\mathbf{q}}b_{\mathbf{q}}\left [ \rho_{\mathbf{q}},p_{x}\right ] + v_{\mathbf{q}}^{\ast } b_{\mathbf{q}}^{+}\left [ \rho_{-\mathbf{q}},p_{x}\right ] \right ) \\ & = -\hbar\sum_{\mathbf{q}}q_{x}\left ( v_{\mathbf{q}}b_{\mathbf{q}}\rho_{\mathbf{q}}-v_{\mathbf{q}}^{\ast}b_{\mathbf{q}}^{+}\rho_{-\mathbf{q}}\right ) , \end{aligned}\ ] ] so , the commutator of the hamiltonian ( [ hmpol ] ) with the total momentum operator of the charge carriers leads to the expression for the force@xmath1054 this result for the force operator clarifies the significance of using the force - force correlation function rather than the momentum - momentum correlation function .
the operator product @xmath1055 is proportional to @xmath1056 , the charge carrier - phonon interaction strength .
this will be a distinct advantage for any expansion of the final result in the charge carrier - phonon interaction strength , since one power of @xmath1056 is factored out beforehand . substituting ( [ force ] ) into ( [ ff ] ) ,
the real part of the optical conductivity then takes the form : @xmath1057 \rho_{\mathbf{q}}\left ( t\right ) , \left ( v_{-\mathbf{q}^{\prime}}b_{-\mathbf{q}^{\prime}}+v_{\mathbf{q}^{\prime}}^{\ast}b_{\mathbf{q}^{\prime } } ^{+}\right ) \rho_{-\mathbf{q}^{\prime}}\right ] \right\rangle .
\label{ff2}\ ] ] up to this point , no approximations other than _ linear response theory _ have been made .
the expectation value appearing in the right hand side of expression ( [ ff2 ] ) for the real part of the optical conductivity is calculated now with respect to the ldb many - polaron wave function ( [ psildb ] ) .
the unitary operator ( [ u ] ) can be written as @xmath1058 the transformed hamiltonian ( [ h ] ) is denoted as@xmath1059 the momentum operator of an electron @xmath1060 the operator of the total momentum of electrons @xmath87 and the phonon creation and annihilation operators are transformed by the unitary transformation ( [ unit ] ) as follows : @xmath1061 as a result , after the transformation ( [ unit ] ) , the hamiltonian takes the form ( see ref .
@xcite ) : @xmath1062 where the terms are@xmath1063@xmath1064 \nonumber\\ & + \frac{\hbar^{2}}{2m_{b}}\sum_{\mathbf{q}}a_{\mathbf{q}}\sum_{\mathbf{k},\sigma}\left ( \mathbf{q}^{2}+2\mathbf{k\cdot q}\right ) a_{\mathbf{k+q},\sigma}^{+}a_{\mathbf{k},\sigma } , \label{h1b}\]]@xmath1065@xmath1066 the exact expression for the real part of the conductivity ( [ ff2 ] ) after the replacement of @xmath1067 by @xmath1068 is transformed to the approximate one @xmath1069{c}e^{\frac{it}{\hbar}h}\left [ v_{\mathbf{q}}b_{\mathbf{q}}+v_{-\mathbf{q}}^{\ast}b_{-\mathbf{q}}^{+}\right ] \rho_{\mathbf{q}}e^{-\frac{it}{\hbar}h},\\ \left ( v_{-\mathbf{q}^{\prime}}b_{-\mathbf{q}^{\prime}}+v_{\mathbf{q}^{\prime}}^{\ast}b_{\mathbf{q}^{\prime}}^{+}\right ) \rho_{-\mathbf{q}^{\prime}}\end{array } \right ] u\right\vert \phi\right\rangle \right\vert \psi_{el}^{\left ( 0\right ) } \right\rangle\end{aligned}\]]@xmath1070{c}e^{\frac{it}{\hbar}\tilde{h}}u^{-1}\left [ v_{\mathbf{q}}b_{\mathbf{q}}+v_{-\mathbf{q}}^{\ast}b_{-\mathbf{q}}^{+}\right ] u\rho_{\mathbf{q}}e^{-\frac{it}{\hbar}\tilde{h}},\\ u^{-1}\left ( v_{-\mathbf{q}^{\prime}}b_{-\mathbf{q}^{\prime}}+v_{\mathbf{q}^{\prime}}^{\ast}b_{\mathbf{q}^{\prime}}^{+}\right ) u\rho_{-\mathbf{q}^{\prime}}\end{array } \right ] \right\vert \phi\right\rangle \right\vert \psi_{el}^{\left ( 0\right ) } \right\rangle\end{aligned}\]]@xmath1071 \right\vert \phi\right\rangle \right\vert \psi_{el}^{\left ( 0\right ) } \right\rangle .\end{aligned}\ ] ] so , we have arrived at the expression@xmath1072 \right\vert \phi\right\rangle \right\vert \psi_{el}^{\left ( 0\right ) } \right\rangle .\end{gathered}\ ] ] since @xmath1073 and @xmath1074 the terms proportional to @xmath1075 vanish after the summation over @xmath1007 : @xmath1076 hence we obtain the real part of the optical conductivity in the form@xmath1077 \right\vert \phi \right\rangle \right\vert \psi_{el}^{\left ( 0\right ) } \right\rangle .
\label{res}\ ] ] introducing the factor@xmath1078 \right\vert \phi \right\rangle \right\vert \psi_{el}^{\left ( 0\right ) } \right\rangle , \end{aligned}\ ] ] the optical conductivity can be written as@xmath1079 in the case of a weak electron - phonon coupling , we can neglect in the exponent @xmath1080 of ( [ res ] ) the terms @xmath1081 and @xmath1082 [ i. e. , the renormalized hamiltonian of the electron - phonon interaction ( [ h1a ] ) and ( [ hppe ] ) ] .
namely , we replace @xmath1083 in eq .
( [ res ] ) by the hamiltonian @xmath1084 in this case , we find@xmath1085 \right\vert \phi \right\rangle \right\vert \psi_{el}^{\left ( 0\right ) } \right\rangle\end{aligned}\]]@xmath1086 .\end{aligned}\ ] ] the time - dependent phonon operators are@xmath1087 so that we have@xmath1088 \\ & = 2i|v_{\mathbf{q}}|^{2}\delta_{\mathbf{qq}^{\prime}}\operatorname{im}\left [ \left\langle \psi_{el}^{\left ( 0\right ) } \left\vert e^{i\tilde { h}_{e}t/\hbar}\rho_{\mathbf{q}}e^{-i\tilde{h}_{e}t/\hbar}\rho_{-\mathbf{q}}\right\vert \psi_{el}^{\left ( 0\right ) } \right\rangle \left\langle \phi\left\vert b_{\mathbf{q}}b_{\mathbf{q}}^{+}\right\vert \phi\right\rangle \right ] , \end{aligned}\ ] ] where @xmath1089 . taking the expectation value with respect to the phonon vacuum ,
we find @xmath1090 .\ ] ] the optical conductivity ( [ res ] ) then takes the form:@xmath1091 \label{res2}\ ] ] for an isotropic electron - phonon system , @xmath1092 in 3d can be replaced by @xmath1093 what gives us the result@xmath1094 , \label{resig1}\ ] ] where the two - point correlation function is @xmath1095 the same derivation for the 2d case , provides the expression@xmath1096 , \label{resig1 - 2d}\ ] ] where @xmath391 is the surface of the 2d system . to find the formula for the real part of the optical conductivity in its final form , we introduce the standard expression for the dynamic structure factor of the system of charge carriers interacting through a coulomb potential,@xmath1097 the dynamic structure factor is expressed in terms of the two - point correlation function as follows:@xmath1098@xmath542@xmath1099 where @xmath1100 is the fourier image of @xmath1101:@xmath1102 the function @xmath1103 obeys the following property:@xmath1104 where @xmath1105 is the eigenvalue of the hamiltonian @xmath1106:@xmath1107 herefrom , we find that@xmath1108 from ( [ sp ] ) , for the function@xmath1109 \label{b}\ ] ] the following equality is derived:@xmath1110 \\ & = \operatorname{im}\left [ e^{i\omega_{\mathrm{lo}}t}f^{\ast}\left ( \mathbf{q},t\right ) \right ] \\ & = -\operatorname{im}\left [ e^{-i\omega_{\mathrm{lo}}t}f\left ( \mathbf{q},t\right ) \right ] = -b\left ( \mathbf{q},t\right ) , \end{aligned}\]]@xmath1111 the integral in eq .
( [ resig1 ] ) @xmath1112 = \operatorname{im}\int_{0}^{\infty}dte^{i\omega t-\delta t}b\left ( \mathbf{q},t\right)\ ] ] is then transformed as follows:@xmath1113 \\ & = \frac{1}{2i}\left [ \int_{0}^{\infty}dte^{i\omega t-\delta t}b\left ( \mathbf{q},t\right ) -\int_{-\infty}^{0}dte^{i\omega t+\delta t}b\left ( \mathbf{q},-t\right ) \right ] \\ & = \frac{1}{2i}\left [ \int_{0}^{\infty}dte^{i\omega t-\delta t}b\left ( \mathbf{q},t\right ) + \int_{-\infty}^{0}dte^{i\omega t+\delta t}b\left ( \mathbf{q},t\right ) \right ] \\ & = \frac{1}{2i}\int_{-\infty}^{\infty}dte^{i\omega t-\delta\left\vert t\right\vert } b\left ( \mathbf{q},t\right ) \\ & = \frac{1}{2i}\int_{-\infty}^{\infty}dte^{i\omega t-\delta\left\vert t\right\vert } \frac{1}{2i}\left [ e^{-i\omega_{\mathrm{lo}}t}f\left ( \mathbf{q},t\right ) -e^{i\omega_{\mathrm{lo}}t}f^{\ast}\left ( \mathbf{q},t\right ) \right ] \\ & = -\frac{1}{4}\int_{-\infty}^{\infty}dte^{i\omega t-\delta\left\vert t\right\vert } \left [ e^{-i\omega_{\mathrm{lo}}t}f\left ( \mathbf{q},t\right ) -e^{i\omega_{\mathrm{lo}}t}f\left ( \mathbf{q},-t\right ) \right ] .\end{aligned}\ ] ] we can show that , as far as @xmath1114 is the _ ground _ state , the integral @xmath1115 for positive @xmath856 is equal to zero .
let @xmath1116 is the total basis set of the eigenfunctions of the hamiltonian @xmath1117 using these functions we expand @xmath1103:@xmath1118@xmath1119 because for @xmath1120 @xmath1121 is never equal to zero .
so , rewriting expression ( [ resig1 ] ) with the dynamic structure factor of the electron ( or hole ) gas results in:@xmath1122 and we obtain @xmath1123 where @xmath1124{c}n / v\;\text{in 3d,}\\ n / a\;\text{in 2d}\end{array } \right.\ ] ] is the density of charge carriers . for an isotropic medium ,
the dynamic structure factor does not depend on the direction of @xmath1007 , so that @xmath1125 , where @xmath1126 .
let us simplify the expression ( [ opticabs ] ) using explicitly the amplitudes of the frhlich electron - phonon interaction .
the modulus squared of the frhlich electron - phonon interaction amplitude is given by @xmath1127{l}\dfrac{(\hbar\omega_{\mathrm{lo}})^{2}}{q^{2}}\dfrac{4\pi\alpha}{v}\left ( \dfrac{\hbar}{2m_{b}\omega_{\mathrm{lo}}}\right ) ^{1/2}\text { in 3d}\\ \dfrac{(\hbar\omega_{\mathrm{lo}})^{2}}{q}\dfrac{2\pi\alpha}{a}\left ( \dfrac{\hbar}{2m_{b}\omega_{\mathrm{lo}}}\right ) ^{1/2}\,\text{in 2d,}\end{array } \right.\ ] ] where @xmath63 is the ( dimensionless ) frhlich coupling constant determining the coupling strength between the charge carriers and the longitudinal optical phonons @xcite . in 3d and 2d , respectively , the sums over @xmath1007 is transformed to the integrals as follows:@xmath1128@xmath542 @xmath1129 @xmath542@xmath1130 in the same way , we transform @xmath1131 : @xmath1132 using the feynman units ( @xmath1133 , @xmath621 , @xmath1134 ) , @xmath1135 is@xmath1136 from these expressions , it is clear that the scaling relation @xmath1137 which holds for the one - polaron case introduced in ref .
@xcite , is also valid for the many - polaron case if the corresponding 2d or 3d dynamic structure factor is used .
the dynamic structure factor @xmath1138 is expressed through the two - point correlation function by eq .
( [ sthroughf ] ) .
the correlation function @xmath1100 can be found using the retarded green s function of the density operators @xmath1139 \right\vert \psi_{el}^{\left ( 0\right ) } \right\rangle , \ ] ] where @xmath1140 is the step function .
let us consider the more general case of a finite temperature,@xmath1141 \right\rangle , \ ] ] where the average is@xmath1142 the fourier image @xmath1143 of the retarded green s function @xmath1144 is@xmath1145 \right\rangle e^{i\omega t}dt\\ & = -i\int_{0}^{\infty}\left ( \left\langle e^{\frac{it}{\hbar}\tilde{h}_{0}}\rho_{\mathbf{q}}e^{-\frac{it}{\hbar}\tilde{h}_{0}}\rho_{-\mathbf{q}}\right\rangle -\left\langle \rho_{-\mathbf{q}}e^{\frac{it}{\hbar}\tilde { h}_{0}}\rho_{\mathbf{q}}e^{-\frac{it}{\hbar}\tilde{h}_{0}}\right\rangle \right ) e^{i\omega t}dt\end{aligned}\ ] ] the imaginary part of @xmath1143 then is@xmath1146 in the second integral here , we replace @xmath535 by @xmath1147:@xmath1148 as far as the integral over @xmath535 converges ( i. e. , @xmath1149 tends to zero at @xmath1150 ) , we can shift the integration contour to the real axis , what gives us the result@xmath1151 herefrom , we find that@xmath1152@xmath542@xmath1153 so , the equation follows from the analytical properties of the green s functions : @xmath1154 the formula ( [ rel1 ] ) is related to arbitrary temperatures .
in the zero - temperature limit ( @xmath617 ) , the factor @xmath1155 ( [ rel1 ] ) turns into the heavicide step function @xmath1156 , what leads to the formula@xmath1157@xmath542@xmath1158 the retarded green function is related to the dielectric function of the electron gas by the following equation : @xmath1159 .
\label{diel}\ ] ] within the random phase approximation ( rpa ) , following @xcite , the expression for @xmath1143 is @xmath1160 ^{-1}\hbar p\left ( \mathbf{q},\omega\right ) , \label{dr1}\ ] ] where the polarization function @xmath1161 is ( see , e. g. , p. 434 of @xcite ) @xmath1162 with the fermi distribution function of electrons @xmath1163 . for a finite temperature , the explicit analytic expression for the imaginary part of the structure factor @xmath1164
is obtained ( see @xcite ) , @xmath1165 \right\ } } { 1+\exp\left\ { \beta\left [ \mu - e^{\left ( -\right ) } \left
( q,\omega \right ) \right ] \right\ } } , \nonumber\\ e^{\left ( \pm\right ) } \left ( q,\omega\right ) & \equiv\frac{\left ( \hbar\omega\pm\frac{\hbar^{2}q^{2}}{2m_{b}}\right ) ^{2}}{4\frac{\hbar ^{2}q^{2}}{2m_{b}}},\end{aligned}\ ] ] with the chemical potential @xmath262 .
the real part of the structure factor is obtained using the kramers - kronig dispersion relation : @xmath1166 analytical expressions for both real and imaginary parts of @xmath1167 can be written down for the zero temperature ( see @xcite ) , @xmath1168{c}\left [ k_{f}^{2}-\frac{m_{b}^{2}}{\hbar^{2}q^{2}}\left ( \omega-\frac{\hbar q^{2}}{2m_{b}}\right ) ^{2}\right ] \ln\left\vert \frac{\omega-\frac{\hbar q^{2}}{2m_{b}}-\frac{\hbar k_{f}q}{m_{b}}}{\omega-\frac{\hbar q^{2}}{2m_{b}}+\frac{\hbar k_{f}q}{m_{b}}}\right\vert
\\ + \left [ k_{f}^{2}-\frac{m_{b}^{2}}{\hbar^{2}q^{2}}\left ( \omega+\frac{\hbar q^{2}}{2m_{b}}\right ) ^{2}\right ] \ln\left\vert \frac{\omega+\frac{\hbar q^{2}}{2m_{b}}+\frac{\hbar k_{f}q}{m_{b}}}{\omega+\frac{\hbar q^{2}}{2m_{b}}-\frac{\hbar k_{f}q}{m_{b}}}\right\vert
\\ + 2k_{f}q \end{array } \right\ } , \nonumber\\ \operatorname{im}p_{\mathrm{3d}}\left ( q,\omega\right ) & = -\frac{vm_{b}}{4\pi\hbar^{2}q}\left\ { \begin{array } [ c]{c}\left [ k_{f}^{2}-\frac{m_{b}^{2}}{\hbar^{2}q^{2}}\left ( \omega-\frac{\hbar q^{2}}{2m_{b}}\right ) ^{2}\right ] \theta\left ( k_{f}^{2}-\frac{m_{b}^{2}}{\hbar^{2}q^{2}}\left ( \omega-\frac{\hbar q^{2}}{2m_{b}}\right ) ^{2}\right ) \\ -\left [ k_{f}^{2}-\frac{m_{b}^{2}}{\hbar^{2}q^{2}}\left ( \omega+\frac{\hbar q^{2}}{2m_{b}}\right ) ^{2}\right ] \theta\left ( k_{f}^{2}-\frac{m_{b}^{2}}{\hbar^{2}q^{2}}\left ( \omega+\frac{\hbar q^{2}}{2m_{b}}\right ) ^{2}\right ) \end{array } \right\ } , \label{pri}\ ] ] where @xmath1169 is the fermi wave number . after substituting into eq .
( [ rel1 ] ) the retarded green s function ( [ dr1 ] ) in terms of the polarization function we arrive at the formula@xmath1170@xmath542@xmath1171 ^{2}+\left [ v_{\mathbf{q}}\operatorname{im}p\left ( \mathbf{q},\omega\right ) \right ] ^{2}},\label{sqw}\\ \left .
s\left ( \mathbf{q},\omega\right ) \right\vert _ { t=0 } & = -\frac{\hbar}{n}\theta\left ( \omega\right ) \frac{\operatorname{im}p\left ( \mathbf{q,}\omega\right ) } { \left [ 1-v_{\mathbf{q}}\operatorname{re}p\left ( \mathbf{q},\omega\right ) \right ] ^{2}+\left [ v_{\mathbf{q}}\operatorname{im}p\left ( \mathbf{q},\omega\right ) \right ] ^{2}}.\end{aligned}\ ] ] with this dynamic structure factor , the optical conductivity ( [ resigmaff1 ] ) ( in the feynman units ) takes the form@xmath1172 ^{2}+\left [ v_{q}\operatorname{im}p_{\mathrm{3d}}\left ( q,\omega-1\right ) \right ] ^{2}}q^{2}dq . \label{rs3d}\ ] ] correspondingly , in the 2d case we obtain the expression@xmath1173 ^{2}+\left [ v_{q}\operatorname{im}p_{\mathrm{2d}}\left ( q,\omega-1\right ) \right ] ^{2}}q^{2}dq .
\label{rs2d}\ ] ] the rpa dynamic structure factor for the electron ( or hole ) system can be separated in two parts , one related to continuum excitations of the electrons ( or holes ) @xmath1174 , and one related to the undamped plasmon branch : @xmath1175 where @xmath1176 is the wave number dependent plasmon frequency and @xmath1177 is the strength of the undamped plasmon branch . in eqs .
( [ rs3d ] ) , ( [ rs2d ] ) , the contribution of the continuum excitations corresponds to the region @xmath1178 where @xmath1179 .
the contribution related to the undamped plasmons is provided by a region of @xmath1180 where the equations @xmath541{c}\operatorname{im}p\left ( q\mathbf{,}\omega\right ) = 0\\ 1-v_{q}\operatorname{re}p\left ( q,\omega\right ) = 0 \end{array } \right . .
\label{c}\ ] ] are fulfilled simultaneously . using ( [ diel ] ) , we find that eqs .
( [ b ] ) are equivalent to @xmath1181 in the region where @xmath1182 the expression @xmath1183 ^{2}+\left [ v_{q}\operatorname{im}p\left ( q,\omega\right ) \right ] ^{2}}$ ] is proportional to the delta function , which gives a finite contribution to the memory function after the integration over @xmath966 : @xmath1184 ^{2}+\left [ v_{q}\operatorname{im}p\left ( q,\omega\right ) \right ] ^{2}}\right\vert _ { \operatorname{im}p\left ( q\mathbf{,}\omega\right ) = 0}\nonumber\\ & = \frac{1}{\pi v_{q}}\delta\left ( 1-v_{q}\operatorname{re}p\left ( q,\omega\right ) \right ) . \label{vv}\ ] ] using eq .
( [ vv ] ) , the coefficients @xmath1185 in eq .
( [ srpa ] ) can be expressed in terms of the polarization function @xmath1186 as follows:@xmath1184 ^{2}+\left [ v_{q}\operatorname{im}p\left ( q,\omega\right ) \right ] ^{2}}\right\vert _ { \operatorname{im}p\left ( q\mathbf{,}\omega\right ) = 0}\\ & = \left .
\frac{1}{\pi v_{q}^{2}\left\vert \frac{\partial}{\partial\omega } \operatorname{re}p\left ( q,\omega\right ) \right\vert } \right\vert _ { \omega=\omega_{\mathrm{pl}}\left ( q\right ) } \delta\left ( \omega -\omega_{\mathrm{pl}}\left ( q\right ) \right)\end{aligned}\]]@xmath542@xmath1187 the contribution derived from the undamped plasmon branch @xmath1188 is denoted in ref .
@xcite as the _ ` plasmon - phonon ' contribution_. the physical process related to this contribution is the emission of both a phonon and a plasmon in the scattering process
calvani and collaborators have performed doping - dependent measurements of the infrared absorption spectra of the high - t@xmath1192 material nd@xmath1189ce@xmath1190cuo@xmath1191 ( ncco ) .
the region of the spectrum examined by these authors ( 50 - 10000 @xmath1193 ) is very rich in absorption features : they observe is a drude - like component at the lowest frequencies , and a set of sharp absorption peaks related to phonons and infrared active modes ( up to about 1000 @xmath1193 ) possibly associated to small ( holstein ) polarons .
three distinct absorption bands can be distinguished : the ` _ _
d__-band ' ( around 1000 @xmath1193 ) , the mid - infrared band ( mir , around 5000 @xmath1193 ) and the charge - transfer band ( around 10@xmath1194 @xmath1193 ) .
of all these features , the _ d_-band and , at a higher temperatures , the drude - like component have ( hypothetically ) been associated with large polaron optical absorption @xcite .
[ h ] part2fig1.eps for the lowest levels of ce doping , the _ d_-band can be most clearly distinguished from the other features .
the experimental optical absorption spectrum ( up to 3000 @xmath1193 ) of nd@xmath67cuo@xmath1195 ( @xmath1196 ) , obtained by calvani and co - workers @xcite , is shown in fig .
[ p2fig1 ] ( shaded area ) together with the theoretical curve obtained by the present method ( full , bold curve ) and , for reference , the one - polaron optical absorption result ( dotted curve ) . at lower frequencies ( 600 - 1000 @xmath1193 )
a marked difference between the single polaron optical absorption and the many - polaron result is manifest .
the experimental _ d_-band can be clearly identified , rising in intensity at about 600 @xmath1193 , peaking around 1000 @xmath1193 , and then decreasing in intensity above that frequency . at a density of @xmath1197 @xmath367
, we found a remarkable agreement between our theoretical predictions and the experimental curve . in refs .
@xcite , the experimental results on the optical spectroscopy of la@xmath1198sr@xmath1199mno@xmath68 ( lsmo ) and la@xmath1198ca@xmath1199mno@xmath68 ( lcmo ) thin films in the mid - infrared frequency region are presented .
the optical conductivity spectra of lcmo films are interpreted in @xcite in terms of the optical response of small polarons , while the optical conductivity spectra of lsmo films are explained using the large - polaron picture ( see fig .
[ p2fig2 ] ) .
[ h ] part2fig2.eps the real part of the optical conductivity @xmath1200 is expressed in ref .
@xcite by the formula @xmath1201 where @xmath63 is the electron - phonon coupling constant , @xmath1202 is the polaron density , @xmath15 is the electron band mass , @xmath1203 is the lo - phonon frequency , and @xmath1204 is the dynamic structure factor , determined in ref . @xcite through the dielectric function of an electron gas @xmath1205 : @xmath1206 .
\label{s1}\ ] ] in ref .
@xcite , the other definition for the dynamic structure factor is used , which is equivalent to that given by ( [ s1 ] ) ( see ref .
@xcite ) with the factor @xmath242 ( the number of electrons)@xmath1207 here , @xmath1208 denotes the ground state of the electron subsystem ( without the electron - phonon interaction ) , @xmath1209 is the fourier component of the electron density . in ref .
@xcite , the dynamic structure factor is calculated within the random - phase approximation ( rpa ) taking into account the coulomb interaction between electrons with the fourier component of the coulomb potential@xmath1210 where @xmath21 is the high - frequency dielectric constant of the crystal . in refs .
@xcite , the dynamic structure factor is calculated taking into account the local hubbard electron - electron interaction instead of the coulomb interaction .
the local hubbard interaction is used in the small - polaron formalism and describes the potential energy of two electrons on one and the same site ( see , e. g. , refs .
@xcite ) . in its simplest form
the hubbard interaction is ( see eq . ( 1 ) of ref .
@xcite)@xmath1211 where @xmath1024 is the coupling constant of the hubbard interaction , @xmath1212 is the electron occupation number for the @xmath1213-th site . in refs .
@xcite , there are no details of the calculation using the interaction term ( [ hub ] ) .
the following procedure can be supposed .
the transition from the summation over the lattice sites to the integral over the crystal volume @xmath305 is performed taking into account the normalization condition@xmath1214 where the density @xmath1215 is to be determined through @xmath1216 .
as far as the lattice cell volume @xmath1217 , the integral @xmath1218 can be written as the sum over the lattice sites:@xmath1219 where @xmath1220 are the vectors of the lattice .
therefore , from the equality@xmath1221 we find that@xmath1222 the potential ( [ hub ] ) is then transformed from the sum over sites to the integral:@xmath1223 consequently , in the continuum approach the hubbard model is described by the @xmath310-like interparticle potential @xmath1224 .
this development of the approach @xcite performed in refs .
@xcite seems to be contradictory by the following reason . for a many - polaron system , both
the electron - phonon and electron - electron interactions are provided by the electrostatic potentials .
therefore , it would be consistent to consider them both within one and the same approach .
namely , the coulomb electron - electron interaction with the potential ( [ c ] ) is relevant for large and small polarons with the frhlich electron - phonon interaction , while the hubbard electron - electron interaction is relevant for small holstein polarons . nevertheless , as recognized in ref .
@xcite , this model reproduces the observed shape of the polaron peak quite convincingly and provides a better agreement with the experiment @xcite than the phenomenological approach @xcite and the one - polaron theory @xcite .
we consider a system of @xmath242 electrons with mutual coulomb repulsion and interacting with the lattice vibrations following ref .
the system is assumed to be confined by a parabolic potential characterized by the frequency parameter @xmath130 .
the total number of electrons is represented as @xmath1225 where @xmath1226 is the number of electrons with the spin projection @xmath1227 .
the electron 3d ( 2d ) coordinates are denoted by @xmath1228 with @xmath1229 the bulk phonons ( characterized by 3d wave vectors @xmath59 and frequencies @xmath1230 ) are described by the complex coordinates @xmath1231 which possess the property @xcite @xmath1232 the full set of the electron and phonon coordinates are denoted by @xmath1233 and @xmath1234 throughout the present treatment , the euclidean time variable @xmath1235 is used , where @xmath535 is the real time variable . in this representation
the lagrangian of the system is @xmath1236 where @xmath1237 is the lagrangian of an electron with band mass @xmath55 in a quantum dot : @xmath1238 @xmath130 is the confinement frequency , @xmath1239 is the potential of a background charge ( supposed to be static and uniformly distributed with the charge density @xmath1240 in a sphere of a radius @xmath1241),@xmath1242 , \label{vb}\ ] ] where @xmath24 is the static dielectric constant of a crystal , @xmath1243 is the potential energy of the electron - electron coulomb repulsion in the medium with the high - frequency dielectric constant @xmath21 : @xmath1244 @xmath1245 is the lagrangian of free phonons : @xmath1246 further , @xmath1247 is the lagrangian of the electron - phonon interaction : @xmath1248 where @xmath128 is the fourier transform of the electron density operator:@xmath1249 @xmath1250 is the amplitude of the electron - phonon interaction . here , we consider electrons interacting with the long - wavelength longitudinal optical ( lo ) phonons with a dispersionless frequency @xmath1251 , for which the amplitude @xmath1250 is @xcite@xmath1252 where @xmath63 is the electron - phonon coupling constant and @xmath305 is the volume of the crystal .
we consider a _ canonical _
ensemble , where the numbers @xmath1226 are fixed .
the partition function @xmath1253 of the system can be expressed as a path integral over the electron and phonon coordinates : @xmath1254 } , \label{z}\ ] ] where @xmath1255 $ ] is the action functional : @xmath1256 = -\frac{1}{\hbar}\int_{0}^{\hbar\beta}l\left ( \mathbf{\dot{\bar{x}}},\dot{\bar{q}};\mathbf{\bar{x}},\bar{q}\right ) d\tau . \label{sn}\ ] ] the parameter @xmath1257 is inversely proportional to the temperature @xmath261 . in order to take the fermi - dirac statistics into account ,
the integral over the electron paths @xmath1258 in eq .
( [ z ] ) contains a sum over all permutations @xmath447 of the electrons with the same spin projection , and @xmath1259 denotes the parity of a permutation @xmath447 .
the action functional ( [ sn ] ) is quadratic in the phonon coordinates @xmath1260 therefore , _ the path integral over the phonon variables _ in @xmath1261 _ can be calculated analytically _ following ref .
let us describe this path integration in detail .
first , we introduce the real phonon coordinates through the real and imaginary parts of the complex phonon coordinates @xmath1262 , @xmath1263 .
according to the symmetry property ( [ spn ] ) , they obey the equalities@xmath1264@xmath1265{c}\sqrt{2}q_{\mathbf{k}}^{\prime},\;k_{x}\geq0,\\ \sqrt{2}q_{\mathbf{k}}^{\prime\prime},\;k_{x}<0 .
\end{array } \right . \label{det1}\ ] ] in this representation , the sum over phonon coordinates @xmath1266 is transformed in the following way using the symmetry property ( [ sp1]):@xmath1267 \\ & = 2\sum_{\substack{\mathbf{k}\\\left ( k_{x}\geq0\right ) } } \left ( q_{\mathbf{k}}^{\prime}\right ) ^{2}+2\sum_{\substack{\mathbf{k}\\\left ( k_{x}<0\right ) } } \left ( q_{\mathbf{k}}^{\prime\prime}\right ) ^{2}\\ & = \sum_{\substack{\mathbf{k}\\\left ( k_{x}\geq0\right ) } } q_{\mathbf{k}}^{2}+\sum_{\substack{\mathbf{k}\\\left ( k_{x}<0\right ) } } q_{\mathbf{k}}^{2}=\sum_{\mathbf{k}}q_{\mathbf{k}}^{2}.\end{aligned}\ ] ] therefore , the phonon lagrangian ( [ lph ] ) with the real phonon coordinates is@xmath1268 the lagrangian of the electron - phonon interaction ( [ leph ] ) with the real phonon coordinates is transformed in the following way using ( [ sp1]):@xmath1269 let us introduce the real forces:@xmath1270{c}\frac{1}{\sqrt{2}}\left ( \frac{2\omega_{\mathbf{k}}}{\hbar}\right ) ^{1/2}\left ( v_{\mathbf{k}}\rho_{\mathbf{k}}+v_{-\mathbf{k}}\rho _ { -\mathbf{k}}\right ) , \;k_{x}\geq0,\\ \frac{1}{i\sqrt{2}}\left ( \frac{2\omega_{\mathbf{k}}}{\hbar}\right ) ^{1/2}\left ( v_{\mathbf{k}}\rho_{\mathbf{k}}-v_{-\mathbf{k}}\rho _ { -\mathbf{k}}\right ) , \;k_{x}>0 .
\end{array } \right .
\label{realforces}\ ] ] this gives us the lagrangian of the electron - phonon interaction in terms of the real forces and real phonon coordinates:@xmath1271 so , the sum of the lagrangians of phonons and of the electron - phonon interaction is expressed through ordinary real oscillator variables:@xmath1272 the path integration for each phonon mode with the coordinate @xmath1273 is performed independently as described in sec . 2 of ref .
@xcite and gives the result@xmath1274 \\ & = \frac{1}{2\sinh\left ( \frac{\beta\hbar\omega_{\mathbf{k}}}{2}\right ) } \\ & \times\exp\left\ { \frac{1}{4\hbar}\int_{0}^{\hbar\beta}d\tau\int _ { 0}^{\hbar\beta}d\tau^{\prime}\frac{\cosh\left [ \omega_{\mathbf{k}}\left ( \left\vert \tau-\tau^{\prime}\right\vert -\hbar\beta/2\right ) \right ] } { \omega_{\mathbf{k}}\sinh\left ( \beta\hbar\omega_{\mathbf{k}}/2\right ) } \gamma_{\mathbf{k}}\left ( \tau\right ) \gamma_{\mathbf{k}}\left ( \tau^{\prime}\right ) , \right\}\end{aligned}\ ] ] where the exponential is the influence functional of a driven oscillator \{@xcite , eq .
( 3.43)}. therefore , the path integral over all phonon modes is@xmath1275 \\ & = \left ( \prod_{\mathbf{k}}\frac{1}{2\sinh\left ( \frac{\beta\hbar \omega_{\mathbf{k}}}{2}\right ) } \right ) \\ & \times\exp\left\ { \frac{1}{4\hbar}\int_{0}^{\hbar\beta}d\tau\int _ { 0}^{\hbar\beta}d\tau^{\prime}\sum_{\mathbf{k}}\frac{\cosh\left [ \omega_{\mathbf{k}}\left ( \left\vert \tau-\tau^{\prime}\right\vert -\hbar\beta/2\right ) \right ] } { \omega_{\mathbf{k}}\sinh\left ( \beta \hbar\omega_{\mathbf{k}}/2\right ) } \gamma_{\mathbf{k}}\left ( \tau\right ) \gamma_{\mathbf{k}}\left ( \tau^{\prime}\right ) .\right\}\end{aligned}\ ] ] here , the product @xmath1276 is the partition function of free phonons , and the exponential is the influence functional of the phonon subsystem on the electron subsystem .
this influence functional results from the above described _ elimination of the phonon coordinates _ and is usually written down as @xmath1277 , where @xmath1278 is@xmath1279 } { \omega_{\mathbf{k}}\sinh\left ( \beta\hbar\omega_{\mathbf{k}}/2\right ) } \gamma_{\mathbf{k}}\left ( \tau\right ) \gamma_{\mathbf{k}}\left ( \tau^{\prime}\right ) .
\label{phi}\ ] ] the sum over the phonon wave vectors @xmath59 can be simplified as follows:@xmath1280 } { \omega _ { \mathbf{k}}\sinh\left ( \beta\hbar\omega_{\mathbf{k}}/2\right ) } \gamma_{\mathbf{k}}\left ( \tau\right ) \gamma_{\mathbf{k}}\left ( \tau^{\prime}\right ) \\ & = \frac{1}{\hbar}\sum_{\substack{\mathbf{k}\\\left ( k_{x}\geq0\right ) } } \frac{\cosh\left [ \omega_{\mathbf{k}}\left ( \left\vert \tau-\tau^{\prime } \right\vert -\hbar\beta/2\right ) \right ] } { \sinh\left ( \beta\hbar \omega_{\mathbf{k}}/2\right ) } \\ & \times\left [ v_{\mathbf{k}}\rho_{\mathbf{k}}\left ( \tau\right ) + v_{-\mathbf{k}}\rho_{-\mathbf{k}}\left ( \tau\right ) \right ] \left [ v_{\mathbf{k}}\rho_{\mathbf{k}}\left ( \tau^{\prime}\right ) + v_{-\mathbf{k}}\rho_{-\mathbf{k}}\left ( \tau^{\prime}\right ) \right ] \\
& -\frac{1}{\hbar}\sum_{\substack{\mathbf{k}\\\left ( k_{x}<0\right ) } } \frac{\cosh\left [ \omega_{\mathbf{k}}\left ( \left\vert \tau-\tau^{\prime } \right\vert -\hbar\beta/2\right ) \right ] } { \sinh\left ( \beta\hbar \omega_{\mathbf{k}}/2\right ) } \\ & \times\left [ v_{\mathbf{k}}\rho_{\mathbf{k}}\left ( \tau\right ) -v_{-\mathbf{k}}\rho_{-\mathbf{k}}\left ( \tau\right ) \right ] \left [ v_{\mathbf{k}}\rho_{\mathbf{k}}\left ( \tau^{\prime}\right ) -v_{-\mathbf{k}}\rho_{-\mathbf{k}}\left ( \tau^{\prime}\right ) \right ] \\ & = \frac{2}{\hbar}\sum_{\substack{\mathbf{k}\\\left ( k_{x}\geq0\right ) } } \frac{\cosh\left [ \omega_{\mathbf{k}}\left ( \left\vert \tau-\tau^{\prime } \right\vert -\hbar\beta/2\right ) \right ] } { \sinh\left ( \beta\hbar \omega_{\mathbf{k}}/2\right ) } v_{\mathbf{k}}v_{-\mathbf{k}}\\ & \times\left [ \rho_{\mathbf{k}}\left ( \tau\right ) \rho_{-\mathbf{k}}\left ( \tau^{\prime}\right ) + \rho_{-\mathbf{k}}\left ( \tau\right ) \rho_{\mathbf{k}}\left ( \tau^{\prime}\right ) \right ] \\ & = \frac{2}{\hbar}\sum_{\mathbf{k}}\frac{\cosh\left [ \omega_{\mathbf{k}}\left ( \left\vert \tau-\tau^{\prime}\right\vert -\hbar\beta/2\right ) \right ] } { \sinh\left ( \beta\hbar\omega_{\mathbf{k}}/2\right ) } \left\vert v_{\mathbf{k}}\right\vert ^{2}\rho_{\mathbf{k}}\left ( \tau\right ) \rho_{-\mathbf{k}}\left ( \tau^{\prime}\right ) .\end{aligned}\ ] ] herefrom , we find that@xmath1281 } { \sinh\left ( \frac { \beta\hbar\omega_{\mathbf{k}}}{2}\right ) } \rho_{\mathbf{k}}\left ( \tau\right ) \rho_{-\mathbf{k}}\left ( \tau^{\prime}\right ) .
\label{phi1}\ ] ] as a result , the partition function of the electron - phonon system ( [ z ] ) factorizes into a product @xmath1282 of the partition function of free phonons with a partition function @xmath1283 of interacting polarons , which is a path integral over the electron coordinates only:@xmath1284 } .
\label{zp}\ ] ] the functional @xmath1285 = -\frac{1}{\hbar } \int_{0}^{\hbar\beta}\left [ l_{e}\left ( \mathbf{\dot{\bar{x}}}\left ( \tau\right ) , \mathbf{\bar{x}}\left ( \tau\right ) \right ) -v_{c}\left ( \mathbf{\bar{x}}\left ( \tau\right ) \right ) \right ] d\tau+\phi\left [ \mathbf{\bar{x}}\left ( \tau\right ) \right ] \label{sp}\ ] ] describes the phonon - induced retarded interaction between the electrons , including the retarded self - interaction of each electron . using ( [ le ] ) and ( [ vc ] ) we write down @xmath1286 $ ] explicitly:@xmath1287 & = \frac{1}{\hbar}\int_{0}^{\hbar\beta}\left [ \sum_{\sigma}\sum_{j=1}^{n_{\sigma}}\frac{m_{b}}{2}\left ( \mathbf{\dot{x}}_{j,\sigma}^{2}+\omega_{0}^{2}\mathbf{x}_{j,\sigma}^{2}\right ) \underset{\left ( j,\sigma\right ) \neq\left ( l,\sigma^{\prime}\right ) } { + \sum_{\sigma,\sigma^{\prime}}\sum_{j=1}^{n_{\sigma}}\sum_{l=1}^{n_{\sigma^{\prime}}}}\frac{e^{2}}{2\varepsilon_{\infty}\left\vert \mathbf{x}_{j,\sigma}-\mathbf{x}_{l,\sigma^{\prime}}\right\vert } \right ] d\tau\nonumber\\ & -\sum_{\mathbf{q}}\frac{\left\vert v_{\mathbf{q}}\right\vert ^{2}}{2\hbar^{2}}\int\limits_{0}^{\hbar\beta}d\tau\int\limits_{0}^{\hbar\beta}d\tau^{\prime}\frac{\cosh\left [ \omega_{\mathrm{lo}}\left ( \left\vert \tau-\tau^{\prime}\right\vert -\hbar\beta/2\right ) \right ] } { \sinh\left ( \beta\hbar\omega_{\mathrm{lo}}/2\right ) } \rho_{\mathbf{q}}\left ( \tau\right ) \rho_{-\mathbf{q}}\left ( \tau^{\prime}\right ) .
\label{sp1}\ ] ] the free energy of a system of interacting polarons @xmath1288 is related to their partition function ( [ zp ] ) by the equation : @xmath1289 at present no method is known to calculate the non - gaussian path integral ( [ zp ] ) analytically . for _ distinguishable _ particles , the jensen - feynman variational principle @xcite provides a convenient approximation technique .
it yields a lower bound to the partition function , and hence an upper bound to the free energy .
it can be shown @xcite that the path - integral approach to the many - body problem for a fixed number of identical particles can be formulated as a feynman - kac functional on a state space for @xmath242 indistinguishable particles , by imposing an ordering on the configuration space and by the introduction of a set of boundary conditions at the boundaries of this state space .
the resulting variational inequality for identical particles takes the same form as the jensen - feynman variational principle : @xmath1290 where @xmath814 is a model action with the corresponding free energy @xmath1291 .
the angular brackets mean a weighted average over the paths@xmath1292 } } { \sum_{p}\frac{\left ( -1\right ) ^{\mathbf{\xi}_{p}}}{n_{1/2}!n_{-1/2}!}\int d\mathbf{\bar{x}}\int _ { \mathbf{\bar{x}}}^{p\mathbf{\bar{x}}}d\mathbf{\bar{x}}\left ( \tau\right ) e^{-s_{0}\left [ \mathbf{\bar{x}}\left ( \tau\right ) \right ] } } .
\label{aver}\ ] ] in the zero - temperature limit , the polaron ground - state energy @xmath1293 obeys the inequality following from ( [ jf ] ) with ( [ fvar]):@xmath1294 with@xmath1295 we consider a model system consisting of @xmath242 electrons with coordinates @xmath1233 and @xmath1296 fictitious particles with coordinates @xmath1297 in a harmonic confinement potential with elastic interparticle interactions as studied in refs .
the lagrangian of this model system takes the form@xmath1298 the frequencies @xmath1299 @xmath1300 @xmath1301 the mass of a fictitious particle @xmath1302 and the force constant @xmath832 are variational parameters . clearly , this lagrangian is symmetric with respect to electron permutations . performing the path integral over the coordinates of the fictitious particles in the same way
as described above for phonons , the partition function @xmath1303 of the model system of interacting polarons becomes a path integral over the electron coordinates:@xmath1304 } , \label{z0}\ ] ] with the action functional @xmath1305 $ ] given by@xmath1306 & = \frac{1}{\hbar}\int_{0}^{\hbar\beta}\sum_{\sigma}\sum_{j=1}^{n_{\sigma}}\frac{m_{b}}{2}\left [ \mathbf{\dot{x}}_{j,\sigma}^{2}\left ( \tau\right ) + \omega ^{2}\mathbf{x}_{j,\sigma}^{2}\left ( \tau\right ) \right ] d\tau\nonumber\\ & -\frac{1}{\hbar}\int_{0}^{\hbar\beta}\sum_{\sigma,\sigma^{\prime}}\sum_{j=1}^{n_{\sigma}}\sum_{l=1}^{n_{\sigma^{\prime}}}\frac{m_{b}\omega^{2}}{4}\left [ \mathbf{x}_{j,\sigma}\left ( \tau\right ) -\mathbf{x}_{l,\sigma^{\prime}}\left ( \tau\right ) \right ] ^{2}d\tau\nonumber\\ & -\frac{k^{2}n^{2}n_{f}}{4m_{f}\hbar\omega_{f}}\int\limits_{0}^{\hbar\beta } d\tau\int\limits_{0}^{\hbar\beta}d\tau^{\prime}\frac{\cosh\left [ \omega _ { f}\left ( \left\vert \tau-\tau^{\prime}\right\vert -\hbar\beta/2\right ) \right ] } { \sinh\left ( \beta\hbar\omega_{f}/2\right ) } \mathbf{x}\left ( \tau\right ) \cdot\mathbf{x}\left ( \tau^{\prime}\right ) , \label{s0}\ ] ] where @xmath1307 is the center - of - mass coordinate of the electrons,@xmath1308 the partition function @xmath1309 [ eq . ( [ z0 ] ) ] for the model system of interacting polarons can be expressed in terms of the partition function @xmath1310 of the model system of interacting electrons and fictitious particles with the lagrangian @xmath1311 [ eq . ( [ lm ] ) ] as follows:@xmath1312 where @xmath1313 is the partition function of fictitious particles,@xmath1314 with the frequency@xmath1315 and d=3(2 ) for 3d(2d ) systems .
the partition function @xmath1310 is the path integral for both the electrons and the fictitious particles:@xmath1316 } \label{zm}\ ] ] with the action functional @xmath1317 = -\frac{1}{\hbar}\int_{0}^{\hbar\beta}l_{m}\left ( \mathbf{\dot{\bar{x}}},\mathbf{\dot{\bar{y}}};\mathbf{\bar{x}},\mathbf{\bar { y}}\right ) d\tau , \label{sm}\ ] ] where the lagrangian is given by eq .
( [ lm ] ) .
let us consider an auxiliary ghost subsystem with the lagrangian @xmath1318 with two frequencies @xmath264 and @xmath1319 where @xmath264 is given by@xmath1320 the partition function @xmath1321 of this subsystem @xmath1322 \right\ } , \label{zg1}\ ] ] with the action functional@xmath1323 = -\frac{1}{\hbar}\int\limits_{0}^{\hbar\beta}l_{g}\left ( \mathbf{\dot{x}}_{g},\mathbf{x}_{g},\mathbf{\dot{y}}_{g},\mathbf{y}_{g}\right ) \,d\tau\label{sg}\ ] ] is calculated in the standard way , because its lagrangian ( [ ghost ] ) has a simple oscillator form .
consequently , the partition function @xmath1321 is@xmath1324 ^{d}}\frac{1}{\left [ 2\sinh\left ( \frac{\beta\hbar w_{f}}{2}\right ) \right ] ^{d}}. \label{zg2}\ ] ] the product @xmath1325 of the two partition functions @xmath1321 and @xmath1326 is a path integral in the state space of @xmath242 electrons , @xmath1296 fictitious particles and two ghost particles with the coordinate vectors @xmath1327 and @xmath1328 the lagrangian @xmath1329 of this system is a sum of @xmath1311 and @xmath1330@xmath1331 the ghost subsystem is introduced because the center - of - mass coordinates in @xmath1329 can be explicitly separated much more transparently than in @xmath1311 .
this separation is realized by the linear transformation of coordinates,@xmath541{l}\mathbf{x}_{j,\sigma}=\mathbf{x}_{j,\sigma}^{\prime}+\mathbf{x}-\mathbf{x}_{g},\\ \mathbf{y}_{j\sigma}=\mathbf{y}_{j\sigma}^{\prime}+\mathbf{y}-\mathbf{y}_{g } , \end{array } \right .
\label{trans}\ ] ] where @xmath1307 and @xmath1332 are the center - of - mass coordinate vectors of the electrons and of the fictitious particles , correspondingly:@xmath1333 before the transformation ( [ trans ] ) , the independent variables are @xmath1334 with the center - of - mass coordinates @xmath1307 and @xmath1332 determined by eq .
( [ xy ] ) .
when applying the transformation ( [ trans ] ) to the centers of mass ( [ xy ] ) , we find that@xmath1335 as seen from eqs .
( [ t1 ] ) , ( [ t2 ] ) , after the transformation ( [ trans ] ) the independent variables are @xmath1336 while the coordinates
@xmath1337 obey the equations@xmath1338 in order to find the explicit form of the lagrangian ( [ lmt ] ) after the transformation ( [ trans ] ) , we use the following relations for the quadratic sums of coordinates : @xmath1339 the substitution of eq .
( [ xy ] ) into eq .
( [ lmt ] ) then results in the following 3 terms:@xmath1340 where @xmath1341 and @xmath1342 are lagrangians of non - interacting identical oscillators with the frequencies @xmath264 and @xmath1319 respectively,@xmath1343 , \label{lwa}\\ l_{w_{f}}\left ( \mathbf{\dot{\bar{y}}}^{\prime},\mathbf{\bar{y}}^{\prime } \right ) & = -\frac{m_{f}}{2}\sum_{j=1}^{n_{f}}\left [ \left ( \mathbf{\dot{y}}_{j,\sigma}^{\prime}\right ) ^{2}+w_{f}^{2}\left ( \mathbf{y}_{j,\sigma}^{\prime}\right ) ^{2}\right ] .
\label{lwb}\ ] ] the lagrangian @xmath1344 describes the combined motion of the centers - of - mass of the electrons and of the fictitious particles,@xmath1345 with@xmath1346 the lagrangian ( [ lc ] ) is reduced to a diagonal quadratic form in the coordinates and the velocities by a unitary transformation for two interacting oscillators using the following replacement of variables : @xmath1347 with the coefficients @xmath1348 ^{1/2},\quad a_{2}=\left [ \frac{1-\chi}{2}\right ] ^{1/2},\label{a12}\\ \chi & \equiv\frac{\tilde{\omega}^{2}-\tilde{\omega}_{f}^{2}}{\left [ \left ( \tilde{\omega}^{2}-\tilde{\omega}_{f}^{2}\right ) ^{2}+4\gamma^{2}\right ]
^{1/2}},\quad\gamma\equiv k\sqrt{\frac{nn_{f}}{m_{b}m_{f}}}. \label{ch}\ ] ] the eigenfrequencies of the center - of - mass subsystem are then given by the expression @xmath541{c}\omega_{1}=\sqrt{\frac{1}{2}\left [ \tilde{\omega}^{2}+\tilde{\omega}_{f}^{2}+\sqrt{\left ( \tilde{\omega}^{2}-\tilde{\omega}_{f}^{2}\right ) ^{2}+4\gamma^{2}}\right ] } , \\
\omega_{2}=\sqrt{\frac{1}{2}\left [ \tilde{\omega}^{2}+\tilde{\omega}_{f}^{2}-\sqrt{\left ( \tilde{\omega}^{2}-\tilde{\omega}_{f}^{2}\right ) ^{2}+4\gamma^{2}}\right ] } .
\end{array } \right .
\label{o12}\ ] ] as a result , four independent frequencies @xmath1349 @xmath1350 @xmath264 and @xmath1351 appear in the problem .
three of them ( @xmath1349 @xmath1350 @xmath264 ) are the eigenfrequencies of the model system .
@xmath1352 is the frequency of the relative motion of the center of mass of the electrons with respect to the center of mass of the fictitious particles ; @xmath1353 is the frequency related to the center of mass of the model system as a whole ; @xmath264 is the frequency of the relative motion of the electrons with respect to their center of mass .
the parameter @xmath1351 is an analog of the second variational parameter @xmath264 of the one - polaron feynman model .
further , the lagrangian ( [ lc ] ) takes the form @xmath1354 leading to the partition function corresponding to the combined motion of the centers - of - mass of the electrons and of the fictitious particles @xmath1355 ^{d}}\frac{1}{\left [ 2\sinh\left ( \frac{\beta\hbar\omega_{2}}{2}\right ) \right ] ^{d}}. \label{zc}\ ] ] taking into account eqs .
( [ zg2 ] ) and ( [ zc ] ) , we obtain finally the partition function of the model system for interacting polarons @xmath1356 ^{d}\mathbb{\tilde{z}}_{f}\left ( \left\ { n_{\sigma}\right\ } , w,\beta\right ) . \label{z02}\ ] ] here@xmath1357 is the partition function of @xmath1358 non - interacting fermions in a parabolic confinement potential with the frequency @xmath1359 the analytical expressions for the partition function of @xmath1226 spin - polarized fermions @xmath1360 were derived in ref .
@xcite . in order to obtain an upper bound to the free energy @xmath1361
, we substitute the model action functional ( [ s0 ] ) into the right - hand side of the variational inequality ( [ jf ] ) and consider the limit @xmath1362:@xmath1363 \nonumber\\ & + \lim_{\beta\rightarrow\infty}\frac{k^{2}n^{2}n_{f}}{4m_{f}\beta\hbar \omega_{f}}\int\limits_{0}^{\hbar\beta}d\tau\int\limits_{0}^{\hbar\beta}d\tau^{\prime}\frac{\cosh\left [ \omega_{f}\left ( \left\vert \tau -\tau^{\prime}\right\vert -\hbar\beta/2\right ) \right ] } { \sinh\left ( \beta\hbar\omega_{f}/2\right ) } \left\langle \mathbf{x}\left ( \tau\right ) \cdot\mathbf{x}\left ( \tau^{\prime}\right ) \right\rangle _ { s_{0}}\nonumber\\ & -\lim_{\beta\rightarrow\infty}\sum_{\mathbf{q}}\frac{\left\vert v_{\mathbf{q}}\right\vert ^{2}}{2\hbar^{2}\beta}\int\limits_{0}^{\hbar\beta } d\tau\int\limits_{0}^{\hbar\beta}d\tau^{\prime}\frac{\cosh\left [ \omega_{\mathrm{lo}}\left ( \left\vert \tau-\tau^{\prime}\right\vert -\hbar\beta/2\right ) \right ] } { \sinh\left ( \beta\hbar\omega_{\mathrm{lo}}/2\right ) } \mathcal{g}\left ( \mathbf{q},\tau-\tau^{\prime}|\left\ { n_{\sigma}\right\ } , \beta\right ) .
\label{varfun}\ ] ] here , @xmath1364 is the energy of @xmath242 non - interacting fermions in a parabolic confinement potential with the confinement frequency @xmath264,@xmath1365 where @xmath1366 is the spin of an electron , @xmath1367 is the lower partly filled or empty level for @xmath1226 electrons with the spin projection @xmath1366 . the first term in the curly brackets of eq .
( [ 2n ] ) ( the upper line ) is the number of electrons at fully filled energy levels , while the second term ( square brackets ) is the number of electrons at the next upper level ( which can be empty or filled partially ) .
the energy levels of a 3d oscillator are degenerate , so that @xmath1368 is the degeneracy of the @xmath275-th energy level .
the parameter @xmath1369 is the number of electrons at all fully filled levels .
the summation in eq .
( [ ew ] ) is performed explicitly , what gives us the result@xmath1370 .\ ] ] in eq .
( [ varfun ] ) , @xmath1371 is the two - point correlation function for the electron density operators:@xmath1372 the averages @xmath1373 are calculated using the generating function method : @xmath1374 \right\rangle _ { s_{0}}\right\vert _ { \substack{\mathbf{\xi}=0,\\\mathbf{\eta}=0 } } , \label{b2}\]]@xmath1375 } { \omega_{i}\sinh\left ( \hbar\beta\omega_{i}/2\right ) } .
\label{xtxs}\ ] ] substituting this expression into eq .
( [ varfun ] ) and performing integrations over @xmath536 and @xmath1366 analytically , we obtain the result@xmath1376 } { \sinh\left ( \beta\hbar\omega_{f}/2\right ) } \left\langle \mathbf{x}\left ( \tau\right ) \cdot\mathbf{x}\left ( \tau^{\prime}\right ) \right\rangle _ { s_{0}}\nonumber\\ & = \frac{3\hbar\gamma}{4}\sum_{i=1}^{2}\frac{a_{i}^{2}}{\omega_{f}^{2}-\omega_{i}^{2}}\left [ \frac{\coth\left ( \beta\omega_{i}/2\right ) } { \omega_{i}}-\frac{\coth\left ( \beta\omega_{f}/2\right ) } { \omega_{f}}\right ] , \end{aligned}\ ] ] and in the zero - temperature limit we have@xmath1377 } { \sinh\left ( \beta\hbar\omega_{f}/2\right ) } \left\langle \mathbf{x}\left ( \tau\right ) \cdot\mathbf{x}\left ( \tau^{\prime}\right ) \right\rangle _ { s_{0}}\nonumber\\ & = \frac{3\hbar\gamma}{4}\sum_{i=1}^{2}\frac{a_{i}^{2}}{\omega_{f}^{2}-\omega_{i}^{2}}\left ( \frac{1}{\omega_{i}}-\frac{1}{\omega_{f}}\right ) .\end{aligned}\ ] ] the average @xmath1378 is transformed , using the described above operations with the ghost subsystem , @xmath1379 and taking into account the first of equations ( [ tosum ] ) @xmath1380 consequently , averaging the left - hand side of eq .
( [ b5 ] ) on the model action functional @xmath1381 one obtains @xmath1382 the term @xmath1383 is expressed using the virial theorem through the ground - state energy @xmath1364 of @xmath242 independent 3d fermion oscillators with the frequency @xmath264 and with the mass @xmath55 , @xmath1384 two other terms in eq .
( [ trr ] ) are [ cf .
( [ xtxs ] ) ] : @xmath1385
so , we obtain @xmath1386 the averaging of the background - charge potential gives us the result@xmath1387 \right\ } , \\
\eta & \equiv\varepsilon_{\infty}/\varepsilon_{0},\;\;a\equiv\frac{\hbar } { 4m_{b}n}\left ( \sum\limits_{i=1}^{2}\frac{a_{i}^{2}}{\omega_{i}}+\frac { n-1}{w}\right ) , \nonumber\end{aligned}\ ] ] where @xmath1388 is the one - particle distribution function of fermions ( the distribution functions are considered in more details in the next subsection ) .
collecting all terms together , we arrive at the variational functional@xmath1389 + \frac{3}{2}\left ( \omega_{1}+\omega_{2}-\omega_{f}\right ) \right .
\nonumber\\ + \left .
\frac{3}{4}\left ( \omega_{0}^{2}-\omega_{1}^{2}-\omega_{2}^{2}+\omega_{f}^{2}\right ) \sum_{i=1}^{2}\frac{a_{i}^{2}}{\omega_{i}}+\frac{3\gamma^{2}}{4\omega_{f}}\sum_{i=1}^{2}\frac{a_{i}^{2}}{\omega _ { i}\left ( \omega_{i}+\omega_{f}\right ) } \right\ } \nonumber\\ + \left\langle u_{b}\left ( \mathbf{\bar{x}}\right ) \right\rangle _ { s_{0}}+e_{c}+e_{e - ph } , \label{evar}\ ] ] where @xmath1390 and @xmath1391 are the coulomb and polaron contributions , respectively : @xmath1392 , \label{ec}\]]@xmath1393 the correlation function ( [ ciuchif ] ) is calculated analytically in the next subsection . with this correlation function
, the variational ground - state energy is calculated and minimized numerically .
the two - point correlation function ( [ ciuchif ] ) is represented as the following path integral:@xmath1394 } \rho_{\mathbf{q}}\left ( \tau\right ) \rho_{-\mathbf{q}}\left ( 0\right ) .
\label{g1}\ ] ] we observe that @xmath1395 can be rewritten as an average within the model action @xmath1396 $ ] of interacting electrons and fictitious particles:@xmath1397 } \nonumber\\ & \times\rho_{\mathbf{q}}\left ( \tau\right ) \rho_{-\mathbf{q}}\left ( 0\right ) .
\label{g2}\ ] ] indeed , one readily derives that the elimination of the fictitious particles in ( [ g2 ] ) leads to ( [ g1 ] ) .
the representation ( [ g2 ] ) allows one to calculate the correlation function @xmath1398 in a much simpler way than through eq .
( [ g1 ] ) , using the separation of the coordinates of the centers of mass of the electrons and of the fictitious particles .
this separation is performed for the two - point correlation function ( [ g2 ] ) by the same method as it has been done for the partition function ( [ zm ] ) . as a result , one obtains@xmath1399 \right\rangle _ { s_{c}}}{\left\langle \exp\left [ i\mathbf{q\cdot}\left ( \mathbf{x}_{g}\left ( \tau\right ) -\mathbf{x}_{g}\left ( \sigma\right ) \right ) \right ] \right\rangle _ { s_{g } } } , \label{fact2}\ ] ] where @xmath1400 is the time - dependent correlation function of @xmath242 non - interacting electrons in a parabolic confinement potential with the frequency @xmath264 , @xmath1401 the action functional @xmath1402 $ ] is related to the lagrangian @xmath1403 [ eq . ( [ lwa])]@xmath1404 = \frac{1}{\hbar}\int \limits_{0}^{\hbar\beta}l_{w}\left ( \mathbf{\dot{\bar{x}}},\mathbf{\bar{x}}\right ) \,d\tau . \label{sw}\ ] ] the averages in ( [ fact2 ] ) are calculated using feynman s method of generating functions @xcite .
namely , according to @xcite , the average @xmath1405 \equiv\left\langle \exp\left\ { \frac{i}{\hbar}\int\limits_{0}^{\beta}f\left ( \tau\right ) x_{\tau}\,d\tau\right\ } \right\rangle _ { s_{\omega } } , \label{gosc}\ ] ] where @xmath1406 is the action functional of a one - dimensional harmonic oscillator with the frequency @xmath856 and with the mass @xmath15 , results in @xmath1405 = \exp\left\ { -\frac{1}{4m\hbar\omega } \int\limits_{0}^{\beta}d\tau\int\limits_{0}^{\beta}d\sigma\frac{\cosh\left [ \omega\left ( \left\vert \tau-\sigma\right\vert -\beta/2\right ) \right ] } { \sinh\left ( \beta\omega/2\right ) } f\left ( \tau\right ) f\left ( \sigma\right ) \right\ } .
\label{gosc1}\ ] ] the diagonalization procedure for the lagrangian @xmath1407 ( [ lc ] ) allows us to represent that lagrangian as a sum of lagrangians of independent harmonic oscillators , what gives the following explicit expressions for averages in eq .
( [ fact2]):@xmath1408 \right\rangle _ { s_{c}}\\ & = \exp\left\ { -\frac{\hbar q^{2}}{nm_{b}}\left [ \sum_{i=1}^{2}a_{i}^{2}\frac{\sinh\left ( \frac{\omega_{i}\left\vert \tau-\sigma\right\vert } { 2}\right ) \sinh\left ( \frac{\omega_{i}\left ( \hbar\beta-\left\vert \tau-\sigma\right\vert \right ) } { 2}\right ) } { \omega_{i}\sinh\left ( \frac{\beta\hbar\omega_{i}}{2}\right ) } \right ] \right\ } , \\ & \left\langle \exp\left [ i\mathbf{q\cdot}\left ( \mathbf{x}_{g}\left ( \tau\right ) -\mathbf{x}_{g}\left ( \sigma\right ) \right ) \right ] \right\rangle _ { s_{g}}\\ & = \exp\left [ -\frac{\hbar q^{2}}{nm_{b}}\frac{\sinh\left ( \frac { w\left\vert \tau-\sigma\right\vert } { 2}\right ) \sinh\left ( \frac{w\left ( \hbar\beta-\left\vert \tau-\sigma\right\vert \right ) } { 2}\right ) } { w\sinh\left ( \frac{\beta\hbar w}{2}\right ) } \right ] .\end{aligned}\ ] ] as seen from the formula ( [ gqtild ] ) , @xmath1409 is the time - dependent correlation function of @xmath242 non - interacting fermions in a parabolic confinement potential with the frequency @xmath264 .
let us consider first of all a system of @xmath242 identical spin - polarized oscillators with the lagrangian @xmath1410 the corresponding hamiltonian is @xmath1411 @xmath1412 a set of eigenfunctions of the one - particle hamiltonian @xmath1413 is determined as follows : @xmath1414 where @xmath1415 @xmath1416 is the @xmath275-th eigenfunction of a one - dimensional oscillator with the frequency @xmath1417 the hamiltonian ( [ 2n ] ) can be written down in terms of the annihilation @xmath1418 and creation @xmath1419 operators : @xmath1420 the many - particle quantum states in the representation of `` occupation numbers '' are written down as @xmath1421 where @xmath1422 is the number of particles in the @xmath1423-th one - particle quantum state .
the states @xmath1424 are defined as the eigenstates of the operator of the number of particles in the @xmath1423-th state @xmath1425 : @xmath1426 let us determine a set of quantum states with a _
finite _ total number of particles @xmath1427 as follows : @xmath1428 further on , we use the basis set of quantum states ( [ 8 ] ) for the derivation of the partition function , of the density function and of the two - point correlation function .
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ * partition function * _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ the density matrix of the _ canonical _ hibbs ensemble is @xmath1429 the partition function of this ensemble is the trace of the density matrix on the set of quantum states ( [ 8 ] ) : @xmath1430 _ { \sum_{\mathbf{n}}n_{\mathbf{n}}=n}. \label{9}\ ] ] this expression can be written down also in the form @xmath1431 where @xmath1432{c}1,\quad j = k\\ 0,\quad j\neq k \end{array } \right.\ ] ] is the delta symbol .
let us introduce the generating function for the partition function in the same way as in ref .
@xcite : @xmath1433 @xmath1434 ^{n_{\mathbf{n}}}\right\ } .
\label{gf1}\ ] ] _ _ _ _ _ _ _ _ _ _ * fermions * _ _ _ _ _ _ _ _ _ _ for fermions , the number @xmath1422 can take only values @xmath1435 and @xmath1436 hence , for fermions ( denoted by the index @xmath1437 ) ,
we obtain : @xmath1438 .\ ] ] since the @xmath275-th level of a 3d oscillator is degenerate with the degeneracy @xmath1439 we find that the generating function @xmath1440 is given by@xmath1441 ^{g\left ( n\right ) } .
\label{equiv1}\ ] ] _ _ _ _ _ _ _ _ * bosons * _ _ _ _ _ _ _ _ for bosons ( denoted by the index @xmath392 ) , @xmath1442 the summations over @xmath1443 in eq .
( [ gf1 ] ) gives : @xmath1444 ^{g\left ( n\right ) } .
\label{gf1a}\ ] ] the results ( [ gf1 ] ) and ( [ gf1a ] ) prove ( for the partition function ) the equivalence of the path - integral approach for identical particles @xcite and of the second - quantization method .
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ * integral representation * _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ let us use the fourier representation for the delta symbol : @xmath1445 d\theta , \label{delta}\ ] ] where @xmath284 is an arbitrary constant . substituting eq .
( [ delta ] ) into eq .
( [ 10 ] ) we obtain @xmath1446 d\theta\\ & = \frac{1}{2\pi}\int\limits_{0}^{2\pi}d\theta\exp\left [ -in\left ( \theta - i\zeta\right ) \right ] \sum_{\left\ { n_{\mathbf{n}}\right\ } } \exp\left ( -\beta\sum_{\mathbf{n}}\varepsilon_{\mathbf{n}}n_{\mathbf{n}}+i\sum_{\mathbf{n}}n_{\mathbf{n}}\left ( \theta - i\zeta\right ) \right ) \\ & = \frac{1}{2\pi}\int\limits_{0}^{2\pi}d\theta\exp\left ( -in\theta -n\zeta\right ) \xi\left ( \beta , e^{i\theta+\zeta}\right ) \quad \longrightarrow\end{aligned}\]]@xmath1447 .
\label{qq}\ ] ] the partition function for a finite number of particles can be obtained from the generation function also by the inversion formula @xcite @xmath1448 } e^{-in\theta}d\theta . \label{211}\ ] ] let us denote in eq .
( [ qq ] ) : @xmath1449 in these notations , eqs .
( [ qq ] ) and ( [ 211 ] ) are _
identical_. for the numerical calculation , it is more convenient to choose in eq .
( [ delta ] ) the interval of the integration over @xmath1450 as @xmath1451 $ ] instead of @xmath1452 , $ ] what gives : @xmath1453 with the function @xmath1454 .
\label{fin}\ ] ] the aforesaid method of derivation of the partition function [ eqs .
( [ delta ] ) to ( [ qq ] ) ] is heuristically useful , because it allows a simple generalization to spin - mixed systems with various polarization distributions . the two - point density - density correlation function in the operator formalism is@xmath1455 where @xmath1456 is the density operator in the heisenberg representation : @xmath1457 in the second - quantization representation , @xmath1456 is @xmath1458 .
\label{rhoqt}\ ] ] after substituting eq .
( [ rhoqt ] ) into ( [ g ] ) , we find that @xmath1459 \left\langle \hat{a}_{\mathbf{n}}^{+}\hat { a}_{\mathbf{n}^{\prime}}\hat{a}_{\mathbf{m}}^{+}\hat{a}_{\mathbf{m}^{\prime}}\right\rangle . \label{gt1}\ ] ] the operator @xmath1460 has non - zero diagonal matrix elements in the basis of quantum states @xmath1461 only in the cases @xmath541{c}\mathbf{n}=\mathbf{n}^{\prime}\\ \mathbf{m}=\mathbf{m}^{\prime}\end{array } \right .
\quad\mathrm{or}\quad\left\ { \begin{array } [ c]{c}\mathbf{n}=\mathbf{m}^{\prime}\\ \mathbf{m}=\mathbf{n}^{\prime}\end{array } \right . .
\label{cond}\ ] ] hence , the average @xmath1462 is not equal to zero only when the condition ( [ cond ] ) is fulfilled .
this allows us to write down the average ( [ av2 ] ) as @xmath1463 here , the notation is used for the average occupation number @xmath1464 : @xmath1465@xmath1466 _ { \sum_{\mathbf{n}}n_{\mathbf{n}}=n}.\end{aligned}\ ] ] in the same way as eq . ( [ 9 ] ) , the average ( [ nav ] ) can be written down in the form @xmath1467 _ { \sum _ { \mathbf{n}^{\prime}}n_{\mathbf{n}^{\prime}}=n}.\ ] ] @xmath1468@xmath1469 since @xmath1470 @xmath1471 depends only on @xmath275 .
( [ zu ] ) , we can write @xmath1472 as @xmath1473 } { \frac{1}{u}\exp\left ( \beta\varepsilon_{n}-i\theta\right ) + 1}d\theta\nonumber\\ & = \frac{1}{2\pi\mathbb{z}_{i}\left ( \beta|n\right ) } \int\limits_{-\pi } ^{\pi}\frac{\xi\left ( \beta , ue^{i\theta}\right ) } { u^{n-1}}\frac{\exp\left [ -i\theta\left ( n-1\right ) -\beta\varepsilon_{n}\right ] } { 1+u\exp\left ( i\theta-\beta\varepsilon_{n}\right ) } d\theta . \label{co}\ ] ] the averages @xmath1474 for @xmath1475 can be also expressed in terms of the integral representation : @xmath1476 _ { \sum_{\mathbf{n}^{\prime}}n_{\mathbf{n}^{\prime}}=n,\quad\mathbf{m}\neq\mathbf{n}}\\ & = \frac{1}{\mathbb{z}_{i}\left ( \beta|n\right ) \beta^{2}}\frac{\delta ^{2}\mathbb{z}_{i}\left ( \beta|n\right ) } { \delta\varepsilon_{\mathbf{m}}\delta\varepsilon_{\mathbf{n}}}\\ & = \frac{1}{\mathbb{z}_{i}\left ( \beta|n\right ) \beta^{2}}\frac{\delta^{2}}{\delta\varepsilon_{\mathbf{m}}\delta\varepsilon_{\mathbf{n}}}\left ( \frac{1}{2\pi}\int\limits_{0}^{2\pi}d\theta\exp\left [ \ln\xi\left ( \beta , e^{i\theta+\zeta}\right ) -n\zeta - in\theta\right ] \right ) \quad\rightarrow\end{aligned}\ ] ] we obtain the integral representation for the average of the product of operators @xmath1477 for @xmath1478 : @xmath1479 \left [ \exp\left ( \beta\varepsilon_{\mathbf{m}}-\zeta -i\theta\right ) + 1\right ] } d\theta . \label{num2}\ ] ] let us introduce the notation @xmath1480 which formally coincides with the fermi distribution function of the energy @xmath1481 with the chemical potential @xmath1482 using this notation , the averages ( [ num1 ] ) and ( [ num2 ] ) can be written down in the form @xmath1483 we can develop the aforesaid procedure for the average of a product of any number of operators @xmath1484 where all quantum numbers @xmath1485 are _ different_. the result is : @xmath1486 it should be emphasized , that all expressions above [ including eq .
( [ gen ] ) ] are derived for a _
canonical _ hibbs ensemble ( i. e. , for a fixed number of particles ) and for both closed - shell and open - shell systems .
let us substitute the average ( [ av3 ] ) into the correlation function @xmath1487 : @xmath1488 \\ & \times\left ( \delta_{\mathbf{n}^{\prime}\mathbf{n}}\delta_{\mathbf{m}^{\prime}\mathbf{m}}\left\langle \hat{n}_{\mathbf{n}}\hat{n}_{\mathbf{m}}\right\rangle + \delta_{\mathbf{m}^{\prime}\mathbf{n}}\delta_{\mathbf{n}^{\prime}\mathbf{m}}\left\langle \hat{n}_{\mathbf{n}}\left ( 1-\hat { n}_{\mathbf{m}}\right ) \right\rangle \right)\end{aligned}\ ] ] @xmath1489@xmath1490 \left\langle \hat{n}_{\mathbf{n}}\left ( 1-\hat{n}_{\mathbf{m}}\right ) \right\rangle . \label{gq}\ ] ] the matrix elements @xmath1491 has the following form @xmath1492 where @xmath1493 is the matrix element of a one - dimensional oscillator with the frequency @xmath264 : @xmath1494{l}n_{>}\equiv\max\left ( n , m\right ) ; \\
n_{<}\equiv\min\left ( n , m\right ) , \end{array } \right . \label{mel1}\ ] ] @xmath1495 , @xmath1496 are the quantum states of the one - dimensional oscillator with the frequency @xmath264 , @xmath1497 is the generalized laguerre polynomial . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ * system with mixed spins * _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ the correlation functions for a system with mixed spins can be explicitly derived by the generalization of eqs .
( [ zi5 ] ) and ( [ gen ] ) to the case of the particles with the non - zero spin .
we use the fact , that the derivation of eqs .
( [ zi5 ] ) and ( [ gen ] ) , performed in this section , does not depend on the concrete form of the energy spectrum @xmath1498 hence , in the formulae , derived above , the replacement should be made : @xmath1499 where @xmath1366 is the electron spin projection .
consequently , the matrix elements @xmath1500 are replaced by @xmath1501 taking into account eqs .
( [ repl ] ) and ( [ repl1 ] ) , the two - point correlation function ( [ gq ] ) becomes @xmath1502 \left\langle \hat{n}_{\mathbf{n},\sigma}\left ( 1-\hat{n}_{\mathbf{m},\sigma } \right ) \right\rangle .
\label{gq1}\ ] ] the averages @xmath1503 and @xmath1504 are , respectively , one - particle and two - particle distribution functions,@xmath1505 the one - electron distribution function @xmath1506 is the average number of electrons with the spin projection @xmath1366 at the @xmath275-th energy level , while the two - electron distribution function @xmath1507 is the average product of the numbers of electrons with the spin projections @xmath1366 and @xmath1508 at the levels @xmath275 and @xmath1509 these functions are expressed through the following integrals [ see ( [ num1a ] ) , ( [ num2a])]:@xmath1510{c}\frac{1}{2\pi\mathbb{z}_{f}\left ( n_{\sigma},w,\beta\right ) } \int \limits_{-\pi}^{\pi}f\left ( \varepsilon_{n},\theta\right ) f\left ( \varepsilon_{n\mathbf{^{\prime}}},\theta\right ) \phi\left ( \theta , \beta , n_{\sigma}\right ) d\theta,\;\text{if } \sigma^{\prime}=\sigma;\\ f_{1}\left ( n,\sigma|n_{\sigma},\beta\right ) f_{1}\left ( n,\sigma^{\prime } |n_{\sigma^{\prime}},\beta\right ) , \;\text{if } \sigma^{\prime}\neq\sigma \end{array } \right .
\label{num2a-1}\ ] ] with the notations@xmath1511 , \label{finn}\]]@xmath1512 the function @xmath1513 formally coincides with the fermi - dirac distribution function of the energy @xmath1481 with the chemical potential @xmath1514 here we consider the zero - temperature limit , for which the integrals ( [ num1a-1 ] ) and ( [ num2a-1 ] ) can be calculated analytically .
the result for the one - electron distribution function is@xmath1515{cc}1 , & n < l_{\sigma};\\ 0 , & n >
l_{\sigma};\\ \frac{n_{\sigma}-n_{l_{\sigma}}}{g_{l_{\sigma } } } , & n = l_{\sigma}. \end{array } \right . \label{p1}\ ] ] according to ( [ p1 ] ) , @xmath1367 is the number of the lowest open shell , and @xmath1516{ccc}\frac{1}{2}\left ( n+1\right ) \left ( n+2\right ) & & \left ( 3d\right ) , \\
n+1 & & \left ( 2d\right ) . \end{array } \right.\ ] ] is the degeneracy of the @xmath275-th shell .
@xmath1517 is the number of electrons in all the closed shells with the spin projection @xmath1518@xmath1519{ccc}\frac{1}{6}l_{\sigma}\left ( l_{\sigma}+1\right ) \left ( l_{\sigma}+2\right ) & & \left ( 3d\right ) , \\
\frac{1}{2}l_{\sigma}\left ( l_{\sigma}+1\right ) & & \left ( 2d\right ) .
\end{array } \right . \label{nls}\ ] ] the two - electron distribution function @xmath1520 at @xmath377 takes the form@xmath1521{cc}\left .
f_{1}\left ( n,\sigma|\beta , n_{\sigma}\right ) \right\vert _ { \beta\rightarrow\infty}\left .
f_{1}\left ( n^{\prime},\sigma^{\prime}|\beta , n_{\sigma^{\prime}}\right ) \right\vert _ { \beta\rightarrow\infty } , & n\neq n^{\prime}\;\mathrm{or}\;\sigma\neq\sigma^{\prime}\\ 1 , & \sigma=\sigma^{\prime}\;\mathrm{and}\;n = n^{\prime}<l_{\sigma};\\ 0 , & \sigma=\sigma^{\prime}\;\mathrm{and}\;n = n^{\prime}>l_{\sigma};\\ \frac{n_{\sigma}-n_{l_{\sigma}}}{g_{l_{\sigma}}}\frac{n_{\sigma}-n_{l_{\sigma } } -1}{g_{l_{\sigma}}-1 } , & \sigma=\sigma^{\prime}\;\mathrm{and}\;n = n^{\prime } = l_{\sigma}. \end{array } \right .
\label{p2}\ ] ] in summary , we have obtained the following expression for @xmath1522:@xmath1523 \nonumber\\ & \times\left [ f_{1}\left ( n,\sigma|\left\ { n_{\sigma}\right\ } , \beta\right ) -f_{2}\left ( n,\sigma;n^{\prime},\sigma|\left\ { n_{\sigma } \right\ } , \beta\right ) \right ] .
\label{g4}\ ] ] this formula is valid for both closed and open shells .
the correlation functions derived in this subsection are used both for the calculation of the ground - state energy and , a shown below , for the calculation of the optical conductivity of an @xmath242-polaron system in a quantum dot . the correlation function given by eq .
( [ g4 ] ) can be subdivided as@xmath1524
with@xmath1525 \nonumber\\ & \times\left [ f_{1}\left ( n,\sigma|\left\ { n_{\sigma}\right\ } , \beta\right ) -f_{2}\left ( n,\sigma;n^{\prime},\sigma|\left\ { n_{\sigma } \right\ } , \beta\right ) \right ] , \end{aligned}\]]@xmath1526 in accordance with the subdivision ( [ subd ] ) of the correlation function , we subdivide the coulomb and polaron contributions:@xmath1527 we have numerically checked whether the polaron contribution per particle @xmath1528 tends to a finite value at @xmath1529 . in figs .
[ polcon3 ] and [ polcon4 ] , we have plotted the polaron contributions @xmath1528 as a function of @xmath242 for a quantum dot in zno and in a polar medium with @xmath1530 , @xmath1531 , respectively . has a physical meaning , the plots for @xmath1532 in figs .
3 and 4 start from @xmath1533 .
the total polaron contribution @xmath1391 for @xmath1534 is plotted below , in fig .
9 . ] as seen from fig .
[ polcon3 ] , the polaron contribution @xmath1535 in zno as a function of @xmath242 oscillates taking expressed maxima for @xmath242 corresponding to the closed shells @xmath1536 .
there exist kinks of @xmath1528 at @xmath242 corresponding to the half - filled shells , but these kinks are extremely small . in the case of the medium with @xmath1537 @xmath1538 for @xmath1539 ( what corresponds to the density @xmath1540 @xmath367 ) , the polaron contribution @xmath1528 oscillates taking maximal values at the numbers of fermions , which correspond to the closed shells for a spin - polarized system with parallel spins . in figs .
[ polcon3 ] and [ polcon4 ] , the dashed curves are the envelopes for local maxima ( closed shells ) and local minima of @xmath1541we see that when these envelopes are _ extrapolated _ to larger number of fermions , the distance between the envelopes decreases .
therefore , the magnitude of the variations of @xmath1542 related to the shell filling diminishes with increasing @xmath242 , and it is safe to suppose that in the limit of large @xmath1543 the envelopes tend to one and the same value . that value corresponds to the _ homogeneous _
( bulk ) limit @xmath1544 .
[ h ] part2fig3.eps [ h ] part2fig4.eps in fig .
[ polcon ] , we compare the polaron contribution @xmath1535 calculated within our variational path - integral method for different numbers of fermions with the polaron contribution to the ground - state energy per particle for a polaron gas in bulk , calculated ( i ) in ref .
@xcite within an intermediate - coupling approach ( the thin solid curve ) , ( ii ) in ref .
@xcite , using a variational approach developed first in ref . @xcite .
as seen from this figure , our all - coupling variational method provides lower values for the polaron contribution than those obtained in refs .
the difference between the polaron contribution calculated within our method and that of ref .
@xcite is smaller at low densities and increases in magnitude with increasing density .
the difference between the polaron contribution calculated within our method and that of ref .
@xcite very slightly depends on the density .
the result of ref .
@xcite becomes closer to our result only at high densities .
[ h ] part2fig5.eps in ref .
@xcite we have extended the memory - function approach to a system of arbitrary - coupling interacting polarons confined to a parabolic confinement potential .
the optical conductivity relates the current @xmath1545 per electron to a time - dependent uniform electric field @xmath1546 in the framework of linear response theory .
further on , the fourier components of the electric field are denoted by @xmath1547 @xmath1548 and the similar denotations are used for other time - dependent quantities .
the electric current per electron @xmath1545 is related to the mean electron coordinate response @xmath1549 by @xmath1550 and hence@xmath1551 within the linear - response theory , both the electric current and the coordinate response are proportional to @xmath1552:@xmath1553 where @xmath1554 is the conductivity per electron . because we treat an isotropic electron - phonon system
, @xmath1555 is a scalar function .
it is determined from the time evolution of the center - of - mass coordinate : @xmath1556 the symbol @xmath1557 denotes an average in the _ real - time _ representation for a system with action functional @xmath80:@xmath1558 } \left ( \bullet\right ) \left .
\left\langle \mathbf{\bar{x}}_{0}\left\vert \hat{\rho}\left ( t_{0}\right ) \right\vert \mathbf{\bar{x}}_{0}^{^{\prime}}\right\rangle \right\vert _ { t_{0}\rightarrow-\infty } , \label{average}\ ] ] where @xmath1559 is the density matrix before the onset of the electric field in the infinite past @xmath1560 .
the corresponding action functional is@xmath1561 = \int\limits_{-\infty}^{t}\left [ l_{e}\left ( \mathbf{\dot{\bar{x}}}\left ( t\right ) , \mathbf{\bar{x}}\left ( t\right ) , t\right ) -l_{e}\left ( \mathbf{\dot{\bar{x}}}^{\prime}\left ( t\right ) , \mathbf{\bar{x}}^{\prime}\left ( t\right ) , t\right ) \right ] dt^{\prime } -i\hbar\phi\left [ \mathbf{\bar{x}}\left ( t\right ) , \mathbf{\bar{x}}^{\prime}\left ( t\right ) \right ] , \label{s1}\ ] ] where @xmath1562 is the lagrangian of @xmath242 interacting electrons in a time - dependent uniform electric field @xmath1546 @xmath1563 the influence phase of the phonons @xmath1564 & = -\sum_{\mathbf{q}}\frac{\left\vert v_{\mathbf{q}}\right\vert ^{2}}{\hbar^{2}}\int\limits_{-\infty}^{t}ds\int\limits_{-\infty}^{s}ds^{\prime}\left [ \rho_{\mathbf{q}}\left ( s\right ) -\rho_{\mathbf{q}}^{\prime}\left ( s\right ) \right ] \nonumber\\ & \times\left [ t_{\omega_{\mathbf{q}}}^{\ast}\left ( s - s^{\prime}\right ) \rho_{\mathbf{q}}\left ( s^{\prime}\right ) -t_{\omega_{\mathbf{q}}}\left ( s - s^{\prime}\right ) \rho_{\mathbf{q}}^{\prime}\left ( s^{\prime}\right ) \right]\end{aligned}\ ] ] describes both a retarded interaction between different electrons and a retarded self - interaction of each electron due to the elimination of the phonon coordinates .
this functional contains the free - phonon green s function : @xmath1565 the equation of motion for @xmath1549 is@xmath1566 where @xmath1567 is the average force due to the electron - phonon interaction , @xmath1568 the two - point correlation function @xmath1569 should be calculated from eq .
( [ average ] ) using the exact action ( [ s1 ] ) , but like for the free energy above , this path integral can not be calculated analytically .
instead , we perform an approximate calculation , replacing the two - point correlation function in eq .
( [ fph ] ) by @xmath1570 where @xmath1571 $ ] is the action functional with the optimal values of the variational parameters for the model system considered in the previous section in the presence of the electric field @xmath1546 .
the functional @xmath1572 $ ] is quadratic and describes a system of coupled harmonic oscillators in the uniform electric field @xmath1573 .
this field enters the term @xmath1574 in the lagrangian , which only affects the center - of - mass coordinate .
hence , a shift of variables to the frame of reference with the origin at the center of mass @xmath541{l}\mathbf{x}_{n}\left ( t\right ) = \mathbf{\tilde{x}}_{n}\left ( t\right ) + \mathbf{r}\left ( t\right ) , \\ \mathbf{x}_{n}^{\prime}\left ( t\right ) = \mathbf{\tilde{x}}_{n}^{\prime } \left ( t\right ) + \mathbf{r}\left ( t\right ) , \end{array } \right .
\label{tr1}\ ] ] results in@xmath1575 } .
\label{tr2}\ ] ] this result ( [ tr2 ] ) is valid for any _ quadratic _ model action @xmath1576 the applicability of the parabolic approximation is confirmed by the fact that a self - induced polaronic potential , created by the polarization cloud around an electron , is rather well described by a parabolic potential whose parameters are determined by a variational method . for weak coupling ,
our variational method is at least of the same accuracy as the perturbation theory , which results from our approach at a special choice of the variational parameters . for strong coupling , an interplay of the electron - phonon interaction and the coulomb correlations within a confinement potential can lead to the assemblage of polarons in multi - polaron systems .
our choice of the model variational system is reasonable because of this trend , apparently occurring in a many - polaron system with arbitrary @xmath242 for a finite confinement strength .
the correlation function @xmath1577 corresponds to the model system in the absence of an electric field . for @xmath1578 this function
is related to the imaginary - time correlation function @xmath1579 described in the previous section:@xmath1580 using the transformation ( [ tr1 ] ) and the relation ( [ tr3 ] ) , we obtain from eq .
( [ fph])@xmath1581 } \mathcal{g}\left ( \mathbf{q},i\left ( t - s\right ) |\left\ { n_{\sigma}\right\ } , \beta\right ) ds . \label{f1}\ ] ] within the linear - response theory , we expand the function @xmath1582 } $ ] in eq .
( [ f1 ] ) as a taylor series in @xmath1583 $ ] up to the first - order term .
the zeroth - order term gives no contribution into @xmath1567 due to the symmetry of @xmath1584 and of @xmath1585 with respect to the inversion @xmath1586 . in this approach
, the cartesian coordinates of the force @xmath1587 become @xmath1588 \nonumber\\ & \times\operatorname{im}\left [ t_{\omega_{\mathrm{lo}}}^{\ast}\left ( t - s\right ) \mathcal{g}\left ( \mathbf{q},i\left ( t - s\right ) |\left\ { n_{\sigma}\right\ } , \beta\right ) \right ] ds . \label{aa}\ ] ] further on , we perform the fourier expansion:@xmath1589 in eq .
( [ aa ] ) , we make the replacement@xmath1590 what gives@xmath1591 \operatorname{im}\left [ t_{\omega _ { \mathrm{lo}}}^{\ast}\left ( \tau\right ) \mathcal{g}\left ( \mathbf{q},it|\left\ { n_{\sigma}\right\ } , \beta\right ) \right ] \\ & = \sum_{k=1}^{3}\sum_{\mathbf{q}}\frac{2\left\vert v_{\mathbf{q}}\right\vert ^{2}q_{j}q_{k}}{n\hbar}\frac{1}{2\pi}\int_{-\infty}^{\infty}d\omega r_{k}\left ( \omega\right ) e^{-i\omega t}\int\limits_{0}^{\infty}d\tau\;\left ( 1-e^{i\omega\tau}\right ) \operatorname{im}\left [ t_{\omega_{\mathrm{lo}}}^{\ast}\left ( \tau\right ) f_{\mathbf{q}}\left ( \tau\right ) \right ] \\ & = \frac{1}{2\pi}\int_{-\infty}^{\infty}d\omega f_{j}\left ( \omega\right ) e^{-i\omega t},\end{aligned}\ ] ] where the fourier component of the force is@xmath1592 r_{k}\left ( \omega\right ) .
\label{bb}\ ] ] the expression ( [ bb ] ) can be written down as@xmath1593 where @xmath1594 are components of the tensor@xmath1595 .
\label{t}\ ] ] in the abstract tensor form , eq .
( [ tt ] ) is@xmath1596 in particular , for the isotropic electron - phonon interaction and in the absence of the magnetic field , the tensor @xmath1597 is proportional to the unity tensor @xmath1598,@xmath1599 where @xmath1600 is the scalar memory function:@xmath1601 .\ ] ] let us perform the fourier transformation of the equation of motion ( [ eqmotion]):@xmath1602 with eq .
( [ tt2 ] ) , this equation takes the form@xmath1603@xmath542@xmath1604 \mathbf{r}\left ( \omega\right ) = e\mathbf{e}\left ( \omega\right ) .
\label{eqm3}\ ] ] comparing eqs .
( [ j2 ] ) and ( [ eqm3 ] ) between each other , we find that@xmath1604 \frac{\sigma\left ( \omega\right ) } { ie\omega } \mathbf{e}\left ( \omega\right ) = e\mathbf{e}\left ( \omega\right ) , \ ] ] so that@xmath1605 ^{-1}.\ ] ] in the case when eq .
( [ scalar ] ) is valid , we obtain the conductivity in the scalar form@xmath1606 } .\ ] ] the real part of the conductivity is@xmath1607 } { m_{b}\left\ { \left [ \left ( \omega^{2}-\omega _ { 0}^{2}\right ) -\operatorname{re}\chi\left ( \omega\right ) \right ] ^{2}+\left [ \operatorname{im}\chi\left ( \omega\right ) \right ] ^{2}\right\ } } \\ & = -\frac{e^{2}\omega}{m_{b}}\frac{\operatorname{im}\chi\left ( \omega\right ) } { \left [ \left ( \omega^{2}-\omega_{0}^{2}\right ) -\operatorname{re}\chi\left ( \omega\right ) \right ] ^{2}+\left [ \operatorname{im}\chi\left ( \omega\right ) \right ] ^{2}}.\end{aligned}\ ] ] in summary , the optical conductivity can be expressed in terms of the memory function @xmath1600 ( cf .
^{2}+\left [ \operatorname{im}\chi\left ( \omega\right ) \right ] ^{2 } } , \label{kw}\ ] ] where @xmath1600 is given by @xmath1609 .
\label{hi}\ ] ] it is worth noting that the optical conductivity ( [ kw ] ) differs from that for a translationally invariant polaron system both by the explicit form of @xmath1600 and by the presence of the term @xmath1610 in the denominator . for @xmath1611
the optical conductivity tends to a @xmath310-like peak at @xmath1612@xmath1613 for a translationally invariant system @xmath1614 , and this weak - coupling expression ( [ limit ] ) reproduces the central peak of the polaron optical conductivity @xcite .
the further simplification of the memory function ( [ hi ] ) is performed in the following way . with the frhlich amplitudes of the electron - phonon interaction , we transform the summation over @xmath1007 to the integral and use the feynman units ( @xmath1133 , @xmath1134 , @xmath621 ) , in which @xmath1615 .
we also use the fact that in an isotropic crystal , @xmath1616 . as a result
, we find @xmath1617 \\ & = \frac{2\sqrt{2}\alpha}{3\pi n}\int_{0}^{\infty}q^{2}dq\int\limits_{0}^{\infty}dt\left ( e^{i\omega t}-1\right ) \operatorname{im}\left [ t_{\omega_{\mathrm{lo}}}^{\ast}\left ( t\right ) \mathcal{g}\left ( q , it|\left\ { n_{\sigma}\right\ } , \beta\right ) \right ] .\end{aligned}\ ] ] in the zero - temperature case , @xmath1618 and we arrive at the expression@xmath1619 .
\label{hi1}\ ] ] substituting the two - point correlation function @xmath1620 with the one - electron ( [ p1 ] ) and the two - electron ( [ p2 ] ) distribution functions into the memory function ( [ hi1 ] ) and expanding @xmath1621 in powers of @xmath1622 , @xmath1623 and @xmath1624 , the integrations over @xmath966 and @xmath535 in eq . ( [ hi1 ] ) are performed analytically .
the similar transformations are performed also in the 2d case . as a result
, the memory function ( [ hi ] ) is represented in the unified form for 3d and 2d interacting polarons : @xmath1625 \right\vert _ { \beta\rightarrow\infty } \right .
\right . \nonumber\end{aligned}\]]@xmath1626{c}\frac{1}{\omega-\omega_{\mathrm{lo}}-\left
[ p_{1}\omega_{1}+p_{2}\omega _ { 2}+\left ( p_{3}-m+n\right ) w\right ] + \mathrm{i}\varepsilon}-\frac { 1}{\omega+\omega_{\mathrm{lo}}+p_{1}\omega_{1}+p_{2}\omega_{2}+\left ( p_{3}-m+n\right ) w+\mathrm{i}\varepsilon}\\ + \mathcal{p}\left ( \frac{2}{\omega_{\mathrm{lo}}+p_{1}\omega_{1}+p_{2}\omega_{2}+\left ( p_{3}-m+n\right ) w}\right ) \end{array } \right ) \nonumber\\ & \times\sum\limits_{l=0}^{m}\sum\limits_{k = n - m+l}^{n}\frac{\left ( -1\right ) ^{n - m+l+k}\gamma\left ( p_{1}+p_{2}+p_{3}+k+l+\frac{3}{2}\right ) } { k!l!}\left ( \frac{1}{wa}\right ) ^{l+k}\nonumber\\ & \left .
\times\binom{n+d-1}{n - k}\binom{2k}{k - l - n+m}\right ] \nonumber\end{aligned}\]]@xmath1627{c}\frac{1}{\omega-\omega_{\mathrm{lo}}-\left ( p_{1}\omega_{1}+p_{2}\omega _ { 2}+p_{3}w\right ) + \mathrm{i}\varepsilon}-\frac{1}{\omega+\omega _ { \mathrm{lo}}+p_{1}\omega_{1}+p_{2}\omega_{2}+p_{3}w+\mathrm{i}\varepsilon}\\ + \mathcal{p}\left ( \frac{2}{\omega_{\mathrm{lo}}+p_{1}\omega_{1}+p_{2}\omega_{2}+p_{3}w}\right ) \end{array } \right ) \right .
\nonumber\\ & \times\sum\limits_{m=0}^{\infty}\sum\limits_{n=0}^{\infty}\sum \limits_{\sigma,\sigma^{\prime}}\left .
f_{2}\left ( n,\sigma;m,\sigma ^{\prime}|\left\ { n_{\sigma}\right\ } , \beta\right ) \right\vert _ { \beta\rightarrow\infty}\nonumber\\ & \times\sum\limits_{k=0}^{n}\sum\limits_{l=0}^{m}\frac{\left ( -1\right ) ^{k+l}\gamma\left ( p_{1}+p_{2}+p_{3}+k+l+\frac{3}{2}\right ) } { k!l!}\left ( \frac{1}{wa}\right ) ^{k+l}\nonumber\\ & \left .
\times\binom{n+d-1}{n - k}\binom{m+d-1}{m - l}\right ] \right\ } , \label{mf}\ ] ] where @xmath1628 is the dimensionality of the space , @xmath1629 denotes the principal value , @xmath391 is defined as @xmath1630 /n$ ] , @xmath1631 and @xmath264 are the eigenfrequencies of the model system , @xmath1632 and @xmath1633 are the coefficients of the canonical transformation which diagonalizes the model lagrangian ( [ lm ] ) .
the shell filling schemes for an @xmath242-polaron system in a quantum dot can manifest themselves in the spectra of the optical conductivity . in fig.[spectra ] , optical conductivity spectra for @xmath1634 polarons are presented for a quantum dot with the parameters of cdse : @xmath1635 @xmath1636 @xcite and with different values of the confinement energy @xmath1637 .
the electron band mass @xmath55 and @xmath1638 have the numerical value of 1 .
this means that the unit of length is the effective bohr radius @xmath1639 , while the unit of energy is the effective hartree @xmath1640 . ] in this case , the spin - polarized ground state changes to the ground state satisfying hund s rule with increasing @xmath1637 in the interval @xmath1641 .
[ h ] part2fig6.eps in the inset to fig.[spectra ] , the first frequency moment of the optical conductivity @xmath1642 as a function of @xmath1637 shows a _ discontinuity _
, at the value of the confinement energy corresponding to the change of the shell filling schemes from the spin - polarized ground state to the ground state obeying hund s rule . this discontinuity might be observable in optical measurements .
the shell structure for a system of interacting polarons in a quantum dot is clearly revealed when analyzing the addition energy and the first frequency moment of the optical conductivity in parallel . in fig [ moments ] , we show both the function @xmath1643 and the addition energy@xmath1644 for interacting polarons in a 3d cdse quantum dot .
[ h ] part2fig7.eps as seen from fig [ moments ] , distinct peaks appear in @xmath1645 and @xmath1646 at the magic numbers corresponding to closed - shell configurations at @xmath1647 and to half - filled - shell configurations at @xmath1648 .
we see that each of the peaks of @xmath1649 corresponds to a peak of the addition energy .
the filling patterns for a many - polaron system in a quantum dot can be therefore determined from the analysis of the first moment of the optical absorption for different numbers of polarons .
in the present section , the ground - state properties of a translation invariant @xmath242-polaron system are theoretically studied in the framework of the variational path - integral method for identical particles , using a further development @xcite of the model introduced in refs .
@xcite . in order to describe a many - polaron system ,
we start from the translation invariant @xmath242-polaron hamiltonian @xmath1650 where @xmath15 is the band mass , @xmath19 is the electron charge , @xmath52 is the longitudinal optical ( lo ) phonon frequency , and @xmath61 are the amplitudes of the frhlich electron - lo - phonon interaction@xmath1651 with the electron - phonon coupling constant @xmath1652 the high - frequency dielectric constant @xmath1653 and the static dielectric constant @xmath1654 and consequently @xmath1655 in the expression ( [ units ] ) , @xmath1656 is the effective hartree @xmath1657 where @xmath1658 is the effective bohr radius . the partition function of the system can be expressed as a path integral over all electron and phonon coordinates .
the path integral over the phonon variables can be calculated analytically @xcite .
feynman s phonon elimination technique for this system is well known and leads to the partition function , which is a path integral over the electron coordinates only:@xmath1659 where @xmath1660 denotes the set of electron coordinates , and @xmath1661 denotes the path integral over all the electron coordinates , integrated over equal initial and final points , i.e. @xmath1662 throughout this paper , imaginary time variables are used .
the effective action for the @xmath242-polaron system is retarded and given by@xmath1663 note that the electrons are fermions .
therefore the path integral for the electrons with parallel spin has to be interpreted as the required antisymmetric projection of the propagators for distinguishable particles . we below use units in which @xmath1133 , @xmath1664 , and @xmath1134 .
the units of distance and energy are thus the effective polaron radius @xmath1665 ^{1/2}$ ] and the lo - phonon energy @xmath60 . for distinguishable particles
, it is well known that the jensen - feynman inequality @xcite provides a lower bound on the partition function @xmath1666 ( and consequently an upper bound on the free energy @xmath1437)@xmath1667@xmath1668 for a system with real action @xmath80 and a real trial action @xmath1576the many - body extension ( ref.@xcite ) of the jensen - feynman inequality , requires that the potentials are symmetric with respect to all particle permutations , and that the exact propagator as well as the model propagator are defined on the same state space . within this interpretation
we consider the following generalization of feynman s trial action@xmath1669 with the variational frequency parameters @xmath1670 .
using the explicit forms of the exact ( [ eq : spol ] ) and the trial ( [ eq : s0 ] ) actions , the variational inequality ( [ varineq ] ) takes the form@xmath1671 in the zero - temperature limit ( @xmath617 ) , we arrive at the following upper bound for the ground - state energy @xmath1672 of a translation invariant @xmath242-polaron system@xmath1673 with@xmath1674 where @xmath1675 is the energy of @xmath242 spin - polarized fermions confined to a parabolic potential with the confinement frequency @xmath856 , @xmath1676 is the coulomb energy of the electrons with parallel spins , @xmath1677 is the coulomb energy of the electrons with opposite spins , @xmath1678 is the electron - phonon energy of the electrons with parallel spins , and @xmath1679 is the electron - phonon energy of the electrons with opposite spins . here , we discuss some results of the numerical minimization of @xmath1680 with respect to the three variational parameters @xmath200 , @xmath264 , and @xmath856 .
the frhlich constant @xmath63 and the coulomb parameter @xmath1681 characterize the strength of the electron - phonon and of the coulomb interaction , obeying the physical condition @xmath1682 [ see ( [ units ] ) ] .
the optimal values of the variational parameters @xmath1683@xmath201and @xmath856 are denoted @xmath1684@xmath1685and @xmath1686 , respectively .
the optimal value of the total spin was always determined by choosing the combination @xmath1687 for fixed @xmath1688which corresponds to the lowest value @xmath1689 of the variational functional@xmath1690 [ h ] part2fig8.eps in fig .
[ trinv - pd ] , the phase diagrams analogous to the bipolaron phase diagram of ref .
@xcite are plotted for an @xmath242-polaron system in bulk with @xmath1691 and @xmath1692 .
the area where @xmath1693 is the non - physical region . for @xmath1694 , each sector between a curve corresponding to a well defined @xmath242 and the line indicating @xmath1695 shows the stability region where @xmath1696 , while the white area corresponds to the regime with @xmath1697 . when comparing the stability region for @xmath1533 from fig .
[ trinv - pd ] with the bipolaron phase diagram of ref .
@xcite , the stability region in the present work starts from the value @xmath1698 ( instead of @xmath1699 in ref .
the width of the stability region within the present model is also larger than the width of the stability region within the model of ref . @xcite .
also , the absolute values of the ground - state energy of a two - polaron system given by the present model are smaller than those given by the approach of ref .
@xcite .
the difference between the numerical results of the present work and of ref .
@xcite is due to the following distinction between the used model systems .
the model system of ref .
@xcite consists of two electrons interacting with two fictitious particles and with each other through quadratic interactions .
but the trial hamiltonian given by eq .
( 6 ) of ref .
@xcite is not symmetric with respect to the permutation of the electrons .
it is only symmetric under the permutation of the pairs electron + fictitious particle . as a consequence ,
this trial system is only applicable if the electrons are distinguishable , i.e. have opposite spin .
in contrast to the model of ref .
@xcite , the model used in the present paper is described by the trial action ( 9 ) , which is fully symmetric with respect to the permutations of the electrons , as is required to describe identical particles .
the phase diagrams for @xmath1700 demonstrate the existence of stable multipolaron states ( see ref .
@xcite ) . as distinct from ref .
@xcite , here the ground state of an @xmath242-polaron system is investigated supposing that the electrons are fermions .
as seen from fig .
[ trinv - gse ] , for @xmath1700 , the stability region for a multipolaron state is narrower than the stability region for @xmath1533 , and its width decreases with increasing @xmath242 .
[ h ] part2fig9.eps a consequence of the fermi statistics is the dependence of the polaron characteristics and of the total spin of an @xmath242-polaron system on the parameters ( @xmath1701@xmath242 ) . in fig .
[ trinv - gse ] , we present the ground - state energy per particle , the confinement frequency @xmath1686 and the total spin @xmath80 as a function of the coupling constant @xmath63 for @xmath1702 and for a different numbers of polarons .
the ground - state energy turns out to be a continuous function of @xmath63 , while @xmath1686and @xmath80 reveal jumps . for @xmath1533 ( the case of a bipolaron ) , we see from fig .
[ trinv - gse ] that the ground state has a total spin @xmath1703 for all values of @xmath63 , i. e. , the ground state of a bipolaron is a singlet .
this result is in agreement with earlier investigations on the large - bipolaron problem ( see , e. g. , @xcite ) . in summary , using the extension of the jensen - feynman variational principle to the systems of identical particles , we have derived a rigorous upper bound for the free energy of a translation invariant system of @xmath242 interacting polarons .
the developed approach is valid for an arbitrary coupling strength .
the resulting ground - state energy is obtained taking into account the fermi statistics of electrons .
spherical shells of charged particles appear in a variety of physical systems , such as fullerenes , metallic nanoshells , charged droplets and neutron stars .
a particularly interesting physical realization of the spherical electron gas is found in multielectron bubbles ( mebs ) in liquid helium-4 .
these mebs are 0.1 @xmath262 m 100 @xmath262 m sized cavities inside liquid helium , that contain helium vapor at vapor pressure and a nanometer - thick electron layer anchored to the surface of the bubble @xcite .
they exist as a result of equilibrium between the surface tension of liquid helium and the coulomb repulsion of the electrons @xcite .
recently proposed experimental schemes to stabilize mebs @xcite have stimulated theoretical investigation of their properties .
we describe the dynamical modes of an meb by considering the motion of the helium surface ( ripplons ) and the vibrational modes of the electrons together .
in particular , we analyze the case when the ripplopolarons form a wigner lattice @xcite .
first , we derive the lagrangian of interacting ripplons and phonons within a continuum approach .
the shape of the surface of a bubble is described by the function @xmath1704 where @xmath1705 is the deformation of the surface from a sphere with radius @xmath1706 the deformation can be expanded in a series of spherical harmonics @xmath1707 with amplitudes @xmath1708 @xmath1709 we suppose that the amplitudes are small in such a way that @xmath1710 the ripplon contribution ( @xmath1711 ) to the kinetic energy of an meb , and the contributions to the potential energy due to the surface tension ( @xmath1712 ) and due to the pressure ( @xmath1713 ) were described in ref .
@xcite : @xmath1714{l}t_{\text{r}}=\dfrac{\rho}{2}r_{\text{b}}^{3}{\displaystyle\sum\limits_{l=1}^{\infty } } { \displaystyle\sum\limits_{m =- l}^{l } } \dfrac{1}{l+1}\left| \dot{q}_{lm}\right| ^{2},\\ u_{\sigma}=4\pi\sigma r_{\text{b}}^{2}+\dfrac{\sigma}{2}{\displaystyle\sum\limits_{l=1}^{\infty } } { \displaystyle\sum\limits_{m =- l}^{l } } \left ( l^{2}+l+2\right ) \left| q_{lm}\right| ^{2},\\ u_{\text{v}}=\dfrac{4\pi}{3}pr_{\text{b}}^{3}+pr_{\text{b}}{\displaystyle\sum\limits_{l=1}^{\infty } } { \displaystyle\sum\limits_{m =- l}^{l } } \left| q_{lm}\right| ^{2}. \end{array } \label{usp}\ ] ] here @xmath1715 kg / m@xmath1716 is the density of liquid helium , @xmath1717 j / m@xmath1718 is its surface tension , and @xmath1719 is the difference of pressures outside and inside the bubble . expanding the surface electron density @xmath1720 in a series of spherical harmonics with amplitudes @xmath1721 @xmath1722 the kinetic energy of the motion of electrons can be written as @xmath1723 where @xmath1724 is the bare electron mass and @xmath242 is the number of electrons .
finally , the electrostatic energy ( @xmath1725 ) of the deformed meb with a non - uniform surface electron density ( [ n1 ] ) is calculated using the maxwell equations and the electrostatic boundary conditions at the surface .
the result is : @xmath1726 with the dielectric constant of liquid helium @xmath1727 .
the last term in eq .
( [ uc ] ) describes the ripplon - phonon mixing . only ripplon and phonon modes which have the same angular momentum couple to each other . after the diagonalization of the lagrangian of this ripplon - phonon system
, we arrive at the eigenfrequencies : @xmath1728 ^{2}+4\gamma^{2}\left ( l\right ) } \right ] \right\ } ^{1/2 } , \label{fab}\ ] ] where @xmath1729 is the bare ripplon frequency , @xmath1730 \right\ } ^{1/2 } , \label{wr}\ ] ] while @xmath1731 is the bare phonon frequency , @xmath1732 and @xmath1733 describes the ripplon - phonon coupling : @xmath1734 the interaction energy between the ripplons and the electrons in the multielectron bubble can be derived from the following considerations : ( i ) the distance between the layer electrons and the helium surface is fixed ( the electrons find themselves confined to an effectively 2d surface anchored to the helium surface ) and ( ii ) the electrons are subjected to a force field , arising from the electric field of the other electrons . for a spherical bubble ,
this electric field lies along the radial direction and equals @xmath1735 a bubble shape oscillation will displace the layer of electrons anchored to the surface .
the interaction energy which arises from this , equals the displacement of the electrons times the force @xmath1736 acting on them .
thus , we get for the interaction hamiltonian @xmath1737 here @xmath1738 is the radial displacement of the surface in the direction given by the spherical angle @xmath274 ; and @xmath1739 is the ( angular ) position operator for electron @xmath1740 . the displacement can be rewritten using ( [ radius ] ) and we find @xmath1741 using the relation @xmath1742 the interaction hamiltonian can be written in the suggestive form @xmath1743 with the electron - ripplon coupling amplitude for a meb given by @xmath1744 substituting @xmath1745 into ( [ hint ] ) , we get @xmath1746 ( \hat{a}_{\ell , m}+\hat{a}_{\ell ,-
m}^{+}).\nonumber\end{aligned}\ ] ] in this expression , we consider the limit of a bubble so large that the surface becomes flat on all length scales of interest .
hence we let @xmath1747 but keep @xmath1748 a constant .
this means we have to let @xmath1749 as well . in this limit , @xmath1750,\ ] ] and @xmath1751 varies locally as a plane wave with wave vector @xmath1752 .
the wave function @xmath1753 is furthermore normalized with respect to integration over the surface ( with total area @xmath1754 ) .
thus , we get in the locally flat approximation @xmath1755 or @xmath1756 this corresponds in the limit of large bubbles to the interaction hamiltonian expected for a flat surface . in their treatment of the electron wigner lattice embedded in a polarizable medium such as a semiconductors or an ionic solid , fratini and qumerais @xcite described the effect of the electrons on a particular electron through a mean - field lattice potential .
the ( classical ) lattice potential @xmath1757 is obtained by approximating all the electrons acting on one particular electron by a homogenous charge density in which a hole is punched out ; this hole is centered in the lattice point of the particular electron under investigation and has a radius given by the lattice distance @xmath1758 . within this particular mean - field approximation , the lattice potential can be calculated from classical electrostatics and we find that for a 2d electron gas it can be expressed in terms of the elliptic functions of first and second kind , @xmath1759 and @xmath1760 , @xmath1761 \right .
\nonumber\\ & \left .
+ \left ( d+r\right ) \mathop{\rm sgn}\left ( d - r\right ) k\left [ -\frac{4rd}{\left ( d - r\right ) ^{2}}\right ] \right\ } .
\label{potential}\ ] ] here , @xmath54 is the position vector measured from the lattice position .
we can expand this potential around the origin to find the small - amplitude oscillation frequency of the electron lattice : @xmath1762 with the confinement frequency @xmath1763 in the mean - field approximation , the hamiltonian for a ripplopolaron in a lattice on a _ locally flat _
helium surface is given by @xmath1764 where @xmath1765 is the electron position operator .
now that the lattice potential has been introduced , we can move on and include effects of the bubble geometry . if we restrict our treatment to the case of large bubbles ( with @xmath1766 electrons ) , then both the ripplopolaron radius and the inter - electron distance @xmath1758 are much smaller than the radius of the bubble @xmath1767 .
this gives us ground to use the locally flat approximation using the auxiliary model of a ripplonic polaron in a planar system described by ( [ h1ri ] ) , but with a modified ripplon dispersion relation and an modified pressing field .
we find for the modified ripplon dispersion relation in the meb : @xmath1768 where @xmath1767 is the equilibrium bubble radius which depends on the pressure and the number of electrons .
the bubble radius is found by balancing the surface tension and the pressure with the coulomb repulsion .
the modified electron - ripplon interaction amplitude in an meb is given by @xmath1769 the effective electric pressing field pushing the electrons against the helium surface and determining the strength of the electron - ripplon interaction is @xmath1770 to study the ripplopolaron wigner lattice at finite temperature and for any value of the electron - ripplon coupling , we use the variational path - integral approach @xcite .
this variational principle distinguishes itself from rayleigh - ritz variation in that it uses a trial action functional @xmath1771 instead of a trial wave function .
the action functional of the system described by hamiltonian ( [ h1ri ] ) , becomes , after elimination of the ripplon degrees of freedom , @xmath1772\right\ } + \sum_{\mathbf{q}}\left\vert m_{q}\right\vert ^{2}\nonumber\\ & \times\displaystyle\int\limits_{0}^{\hbar\beta}d\tau\displaystyle\int \limits_{0}^{\hbar\beta}d\sigma g_{\omega(q)}(\tau-\sigma)e^{i\mathbf{q}\cdot\lbrack\mathbf{r}(\tau)-\mathbf{r}(\sigma ) ] } , \label{sri}\ ] ] with @xmath1773}{\sinh(\beta\hbar\nu/2)}}}.\ ] ] in preparation of its customary use in the jensen - feynman inequality , the action functional ( [ sri ] ) is written in imaginary time @xmath1774 with @xmath180 where @xmath261is the temperature .
we introduce a quadratic trial action of the form @xmath1775 \nonumber\\ & -{\displaystyle{\frac{mw^{2}}{4\hbar}}}\displaystyle\int\limits_{0}^{\hbar\beta}d\tau\displaystyle\int\limits_{0}^{\hbar\beta}d\sigma g_{w}(\tau-\sigma)\mathbf{r}(\tau)\cdot\mathbf{r}(\sigma ) .
\label{s0ri}\ ] ] where @xmath1776 and @xmath274 are the variationally adjustable parameters .
this trial action corresponds to the lagrangian @xmath1777 from which the degrees of freedom associated with @xmath539 have been integrated out .
this lagrangian can be interpreted as describing an electron with mass @xmath1778 at position @xmath54 , coupled through a spring with spring constant @xmath1779 to its lattice site , and to which a fictitious mass @xmath1780 at position @xmath539 has been attached with another spring , with spring constant @xmath374 .
the relation between the spring constants in ( [ l0 ] ) and the variational parameters @xmath1781 is given by @xmath1782 based on the trial action @xmath1771 , feynman s variational method allows one to obtain an upper bound for the free energy @xmath1437 of the system ( at temperature @xmath261 ) described by the action functional @xmath80 by minimizing the following function : @xmath1783 with respect to the variational parameters of the trial action . in this expression
, @xmath1291 is the free energy of the trial system characterized by the lagrangian @xmath1784 , @xmath1785 is the inverse temperature , and the expectation value @xmath1786 is to be taken with respect to the ground state of this trial system .
the evaluation of expression ( [ jfri ] ) is straightforward though lengthy .
we find @xmath1787 + \displaystyle{2 \over\beta}\ln\left [ 2\sinh\left ( \displaystyle{\beta\hbar\omega_{2 } \over2}\right ) \right ] } \nonumber\\ & -{\displaystyle{\frac{2}{\beta}}}\ln\left [ 2\sinh\left ( { \displaystyle{\frac{\beta\hbar w}{2}}}\right ) \right ] -{\displaystyle{\frac { \hbar}{2}}}\sum_{i=1}^{2}a_{i}^{2}\omega_{i}\coth\left ( { \displaystyle{\frac { \beta\hbar\omega_{i}}{2}}}\right ) \nonumber\\ & -{\displaystyle{\frac{\sqrt{\pi}e^{2}}{d}}}e^{-d^{2}/(2d)}\left [ i_{0}\left ( { \displaystyle{\frac{d^{2}}{2d}}}\right ) + i_{1}\left ( { \displaystyle{\frac{d^{2}}{2d}}}\right ) \right ] \label{fri}\\ & -{\displaystyle{\frac{1}{2\pi\hbar\beta}}}\int_{1/r_{b}}^{\infty}dqq|m_{q}|^{2}\int_{0}^{\hbar\beta/2}d\tau{\displaystyle{\frac{\cosh [ \omega(q)(\tau-\hbar\beta/2)]}{\sinh[\beta\hbar\omega(q)/2]}}}\nonumber\\ & \times\exp\left [ -{\textstyle{\frac{\hbar q^{2}}{2m_{e}}}}\sum_{j=1}^{2}a_{j}^{2}{\textstyle{\frac{\cosh(\hbar\omega_{j}\beta/2)-\cosh[\hbar \omega_{j}(\tau-\beta/2)]}{\omega_{j}\sinh(\hbar\omega_{j}\beta/2)}}}\right ] .\nonumber\end{aligned}\ ] ] in this expression , @xmath1788 and @xmath1789 are bessel functions of imaginary argument , and @xmath1790@xmath1791 finally , @xmath1352 and @xmath1353 are the eigenfrequencies of the trial system , given by @xmath1792 .\ ] ] optimal values of the variational parameters are determined by the numerical minimization of the variational functional @xmath1437 as given by expression ( [ fri ] ) .
the lindemann melting criterion @xcite states in general that a crystal lattice of objects ( be it atoms , molecules , electrons , or ripplopolarons ) will melt when the average motion of the objects around their lattice site is larger than a critical fraction @xmath1793 of the lattice parameter @xmath1758 .
it would be a strenuous task to calculate from first principles the exact value of the critical fraction @xmath1793 , but for the particular case of electrons on a helium surface , we can make use of an experimental determination .
grimes and adams @xcite found that the wigner lattice melts when @xmath1794 , where @xmath804 is the ratio of potential energy to the kinetic energy per electron . at temperature
@xmath261 the average kinetic energy in a lattice potential @xmath1757 is @xmath1795 and the average distance that an electron moves out of the lattice site is determined by @xmath1796 from this we find that for the melting transition in grimes and adams experiment @xcite , the critical fraction equals @xmath1797 .
this estimate is in agreement with previous ( empirical ) estimates yielding @xmath1798 @xcite , and we shall use it in the rest of this section . within the approach of fratini and qumerais @xcite , the wigner lattice of ( ripplo)polarons melts when at least one of the two following lindemann criteria are met : @xmath1799@xmath1800 where @xmath540 and @xmath1801 are , respectively , the relative coordinate and the center of mass coordinate of the model system ( [ l0 ] ) : if @xmath54 is the electron coordinate and @xmath539 is the position coordinate of the fictitious ripplon mass @xmath1780 , this is @xmath1802 the appearance of two lindemann criteria takes into account the composite nature of ( ripplo)polarons . as follows from the physical sense of the coordinates @xmath540 and @xmath1801 , the first criterion ( [ lind1 ] )
is related to the melting of the ripplopolaron wigner lattice towards a ripplopolaron liquid , where the ripplopolarons move as a whole , the electron together with its dimple .
the second criterion ( [ lind2 ] ) is related to the dissociation of ripplopolarons : the electrons shed their dimple .
the path - integral variational formalism allows us to calculate the expectation values @xmath1803 and @xmath1804 with respect to the ground state of the variationally optimal model system .
we find @xmath1805 \left ( \omega_{1}^{2}-\omega_{2}^{2}\right ) } } } \nonumber\\ & \times\left [ \omega_{2}^{4}(\omega_{1}^{2}-w^{2})\coth(\hbar\omega _ { 1}\beta/2)/\omega_{1}\right .
\nonumber\\ & \left .
+ \omega_{1}^{4}(w^{2}-\omega_{2}^{2})\coth(\hbar\omega_{2}\beta/2)/\omega_{2}\right ] , \label{rcms}\]]@xmath1806 .
\label{rhori}\ ] ] numerical calculation shows that for ripplopolarons in an meb the inequality @xmath1807 is fulfilled ( @xmath1808 to @xmath1809 ) so that the strong - coupling regime is realized . owing to this inequality
, we find from eqs .
( [ rcms]),([rhori ] ) that @xmath1810 so , the destruction of the ripplopolaron wigner lattice in an meb occurs through the dissociation of ripplopolarons , since the second criterion ( [ lind2 ] ) will be fulfilled before the first ( [ lind1 ] ) .
the results for the melting of the ripplopolaron wigner lattice are summarized in the phase diagram shown in fig .
[ phased1 ] .
for every value of @xmath242 , pressure @xmath1719 and temperature @xmath261 in an experimentally accessible range , this figure shows whether the ripplopolaron wigner lattice is present ( points above the surface ) or molten ( points below the surface ) . below a critical pressure ( on the order of 10@xmath1194 pa ) the ripplopolaron solid will melt into an electron liquid .
this critical pressure is nearly independent of the number of electrons ( except for the smallest bubbles ) and is weakly temperature dependent , up to the helium critical temperature 5.2 k. this can be understood since the typical lattice potential well in which the ripplopolaron resides has frequencies of the order of thz or larger , which correspond to @xmath1811 k. the new phase that we predict , the ripplopolaron wigner lattice , will not be present for electrons on a flat helium surface . at the values of the pressing field necessary to obtain a strong enough electron - ripplon coupling , the flat helium surface is no longer stable against long - wavelength deformations @xcite .
multielectron bubbles , with their different ripplon dispersion and the presence of stabilizing factors such as the energy barrier against fissioning @xcite , allow for much larger electric fields pressing the electrons against the helium surface .
the regime of @xmath242 , @xmath1719 , @xmath261 parameters suitable for the creation of a ripplopolaron wigner lattice lies within the regime that would be achievable in recently proposed experiments aimed at stabilizing multielectron bubbles @xcite .
the ripplopolaron wigner lattice and its melting transition might be detected by spectroscopic techniques @xcite probing for example the transverse phonon modes of the lattice @xcite .
j. t. devreese , in _ lectures on the physics of highly correlated electron systems vii , _ edited a. avella and f. mancini proceedings of the 7th training course in the physics of correlated electron systems&high - t@xmath1812 superconductors , vietri sul mare , italy , october 14 - 16 , 2002 , aip , melville ( 2003 ) , pp .
3 - 56 .
the spectra of the infrared - active lo ( and to ) phonons in @xmath63-al@xmath67o@xmath68 ( sapphire ) contain six modes .
the values of the lo and to phonon frequencies and of the high - frequency dielectric constants @xmath1813 , @xmath1814 are taken from ref .
@xcite . using these parameters and the electron band mass @xmath1815 as estimated in ref .
@xcite , the effective value of the electron - phonon coupling constant @xmath63 in al@xmath67o@xmath68 has been calculated as @xmath1816 where @xmath1817 is the polarization vector of the @xmath1740-th lo - phonon branch , @xmath59 is the phonon wave vector , @xmath1818 denote the angular averaging , and the coupling constants @xmath1819 for each branch are obtained using the method @xcite .
the resulting value of the polaron coupling constant is @xmath1820 the spectra of the lo ( and to ) phonons in @xmath63-sio@xmath67 contain ten modes .
the values of the lo and to phonon frequencies are taken from refs .
@xcite . using these frequencies and the value @xmath1821 from ref .
@xcite for the high - frequency dielectric constant , the effective value of the electron - phonon coupling constant @xmath63 in sio@xmath67 has been calculated using the method of ref . @xcite as indicated in ref .
we use the estimated value of the electron band mass @xmath1822 as in refs . @xcite . the resulting value of the polaron coupling constant is @xmath1823 .
s. i. pekar , _ issledovanija po ekektronnoj teorii kristallov _ ,
gostekhizdat , moskva , 1951 ( in russian ) [ german translation : _ untersuchungen ber die elektronentheorie der kristalle _ , akademie verlag , berlin , 1951 ] .
j. t. devreese and r. evrard , in _ proceedings of the british ceramic society _ * 10 * , 151 ( 1968 ) .
reprinted in : _ path integrals and their applications in quantum , statistical , and solid state physics _ , edited by g. j. papadopoulos and j. t. devreese , nato asi series b , physics , vol .
34 , plenum , new york , 1977 , pp .
344 - 357 . | based on a course presented at the international school of physics enrico fermi , clxi course,.``polarons in bulk materials and systems with reduced dimensionality '' , varenna , italy , 21.6 .
- 1.7.2005 .
in the present course , an overview is presented of the fundamentals of continuum - polaron physics , which provide the basis of the analysis of polaron effects in ionic crystals and polar semiconductors .
these lecture notes deal with large , or continuum , polarons , as described by the frhlich hamiltonian .
the emphasis is on the polaron optical absorption , with detailed mathematical derivations . |
the origin of cosmic radiation and a mechanism trough which it could acquire enormous energies has intrigued many physicists and mathematicians along the last decades . in particular
, in the year of 1949 enrico fermi @xcite proposed a very simple model that qualitatively describes a process in which charged particles were _ bounced _ via interaction with moving magnetic fields .
this heuristic idea was then later modified to encompass in a suitable model that could give quantitative results on the original fermi s model .
after that , many different versions of the model were proposed @xcite and studied in different context and considering many approaches and modifications @xcite .
one of them is the well known fermi - ulam model ( fum ) .
such model consists of a classical point - like particle moving between two walls in the total absence of any external field .
one of the walls is considered to be fixed while the other one moves periodically in time . despite this simplicity ,
the structure of the phase space is rather complex and it includes a large chaotic sea surrounding kam islands and a set of invariant spanning curves limiting the energy of a bouncing particle .
however , the introduction of innelastic collisions on this model is enough to destroy such a mixed structure and the system exhibits attractors .
depending on the initial conditions and control parameters , one can observe a chaotic attractor characterised by a positive lyapunov exponent . by a suitable control parameter variation
, the chaotic attractor might be destroyed via a crisis event @xcite .
after the destruction , the chaotic attractor is replaced by a chaotic transient @xcite . in this paper , we revisit a dissipative version of the fermi - ulam model seeking to understand and describe the behaviour of the dynamics for the regime of high dissipation .
moreover , we investigate a phenomenon quite common observed for a large variety of dissipative and nonlinear systems known as _ period doubling route to chaos_. this route shows a sequence of doubling bifurcations connecting regular - periodic to chaotic behaviour .
the main goal of this paper is then to examine and show that the fermi - ulam model obey the same convergence ratio as that obtained by feigenbaum @xcite for the dynamics of the logistic map .
such a constant value has been further called as `` the feigenbaum s @xmath0 '' .
the organisation of this paper is as follows : in section [ sec2 ] we present all the details needed for the construction of the nonlinear mapping .
our numerical results are discussed in section [ sec3 ] and our conclusions and final remarks are drawn in section [ sec4 ] .
let us describe the model and obtain the equations of the mapping .
the model we are dealing with consists of a classical point - like particle confined in and bouncing betwen two rigid walls .
one of them is assumed to be fixed at the position @xmath1 and the other one moves periodically in time according to @xmath2 .
thus , the moving wall velocity is given by @xmath3 where @xmath4 denotes the amplitude of oscillation and @xmath5 is the angular frequency of the moving wall , respectively .
aditionally , the motion of the particle does not suffer influence of any external field .
we assume that all the collisions with both walls are inelastic .
thus , we introduce a restitution coefficient for the fixed wall as @xmath6 $ ] while for the moving wall we consider @xmath7 $ ] .
the completely inelastic collision happens when @xmath8 and that a single collision is enough to terminate all the particle s dynamics . on the other hand , when @xmath9 , corresponding to a complete elastic collision , all the results for the nondissipative case are recovered @xcite .
we describe the dynamics of the particle using a map @xmath10 where the dynamical variables are @xmath11 and @xmath12 is the velocity of the particle at the instant @xmath13 .
the index @xmath14 denotes the @xmath15 collision of the particle with the moving wall .
starting then with an initial condition @xmath11 , with initial position of the particles given by @xmath16 with @xmath17 , the dynamics is evolved and the map @xmath10 gives a new pair of @xmath18 at the @xmath19 collision .
it is important to say that we have three control parameters , namely : @xmath4 , @xmath20 and @xmath5 and that the dynamics does not depend on all of them .
thus it is convenient to define dimensionless and more appropriated variables .
therefore , we define @xmath21 , @xmath22 and measure the time in terms of the number of oscillations of the moving wall , consequently @xmath23 . incorporating this set of new variables into the model ,
the map @xmath10 is written as @xmath24 where the expressions for both @xmath25 and @xmath26 depend on what kind of collision occurs .
there are two different possible situations , namely : ( i ) multiple collisions with the moving wall and , ( ii ) single collision with the moving wall .
considering the first case , the expressions are @xmath27 and @xmath28 , where @xmath29 is obtained as the smallest solution of @xmath30 with @xmath31 $ ] .
a solution for @xmath30 is equivalent to have that the position of the particle is the same as that of the moving wall at the instant of the impact .
the function @xmath32 is given by @xmath33 if the function @xmath32 does not have a root in the interval @xmath31 $ ] , we can conclude that the particle leaves the collision zone ( the collision zone is defined as the interval @xmath34 $ ] ) without suffering a successive collision .
considering now the case ( ii ) , i.e. the case where the particle leaves the collision zone , the corresponding expressions are @xmath35 and @xmath36 , with the auxiliary terms given by @xmath37 finally , the term @xmath29 is obtained as the smallest solution of @xmath38 for @xmath39 .
the expression of @xmath40 is given by @xmath41 both , eq .
( [ g ] ) and eq .
( [ f ] ) must be solved numerically . after some algebra it is easy to shown that the determinant of the jacobian matrix for the case ( i ) is given by @xmath42~ , \label{det_s}\ ] ] while for the case ( ii ) we obtain @xmath43~. \label{det_ns}\ ] ]
we therefore can conclude based on the above result that area preservation is obtained only for the case of @xmath9 .
in this section we discuss some numerical results considering the case of high dissipation .
we call as high dissipation the situation in which a particle loses more than @xmath44 of its energy upon colision with the moving wall . moreover , we have considered as fixed the values of the damping coefficients @xmath45 and @xmath46 . to illustrate that the dynamics of the system exhibits doubling bifurcation cascade ,
it is shown in fig .
[ fig1](a ) and ( b ) @xmath47 both plotted against @xmath48 . in
( c ) it is shown the lyapunov exponent associated to ( a ) and ( b ) .
the damping coefficients used for the construction of the figures ( a ) , ( b ) and ( c ) were @xmath45 and @xmath46 . ]
the behaviour of the asymptotic velocity plotted against the control parameter @xmath48 , where the sequence of bifurcations is evident .
a similar sequence is also observed for the asymptotic variable @xmath49 , as it is shown in fig . [ fig1](b ) .
note that all the bifurcations of same period in ( a ) and ( b ) happen for the same value of the parameter @xmath48 . at the point of bifurcations
, one can also observe that the positive lyapunov exponent shows null value , as it is shown in fig .
[ fig1](c ) .
it is well known that the lyapunov exponents are important tool that can be used to classify orbits as chaotic . as discussed in @xcite ,
the lyapunov exponents are defined as @xmath50 where @xmath51 are the eigenvalues of @xmath52 and @xmath53 is the jacobian matrix evaluated over the orbit @xmath54 . however , a direct implementation of a computational algorithm to evaluate eq .
( [ eq4 ] ) has a severe limitation to obtain @xmath55 . even in the limit of short @xmath14
, the components of @xmath55 can assume very different orders of magnitude for chaotic orbits and periodic attractors , yielding impracticable the implementation of the algorithm . in order to avoid such problem
, we note that @xmath56 can be written as @xmath57 where @xmath58 is an ortoghonal matrix and @xmath10 is a right up triangular matrix .
thus we rewrite @xmath55 as @xmath59 , where @xmath60 .
a product of @xmath61 defines a new @xmath62 . in a next step , it is easy to show that @xmath63 .
the same procedure can be used to obtain @xmath64 and so on . using this procedure ,
the problem is reduced to evaluate the diagonal elements of @xmath65 .
finally , the lyapunov exponents are given by @xmath66 if at least one of the @xmath67 is positive then the orbit is classified as chaotic .
we can see in fig .
( [ fig1 ] ) ( c ) the behavior of the lyapunov exponents corresponding to both figs .
( [ fig1])(a , b ) .
it is also easy to see that when the bifurcations happen , the exponent @xmath68 assumes the zero value at same values of the control parameter @xmath48 where the bifurcation hold .
another interesting observation is that the lyapunov exponent , in some regions , assumes a constant and negative values , such behavior occurs because the eigenvalues of the jacobian matrix become complex numbers ( imaginary numbers ) ) .
let us now use the points where the lyapunov exponent assumes the null value to characterise a very interesting property of the model .
as it was shown by feigenbaum @xcite , along the bifurcations the dynamics exhibits an universal feature .
it implies that all the bifurcations happen at the same rate of convergence for the bifurcation diagram changing from periodic to chaotic behavior . the procedure used to obtain the feigenbaum value
@xmath0 consists of : let @xmath69 represents the control parameter value at which period-1 gives birth to a period-2 orbit , @xmath70 is the value where period-2 changes to period-4 and so on . in general
the parameter @xmath71 corresponds to the control parameter value at which a period-@xmath72 orbit is born .
thus , we write the feigenbaum s @xmath0 as @xmath73 the numerical value for the constant number @xmath0 obtained by eq .
( [ fei ] ) and considering a sufficient large values for @xmath14 is @xmath74 .
considering the numerical data obtained through the lyapunov exponents calculation , the feigenbaum s @xmath0 obtained for the fermi - ulam model is @xmath75 .
we have considered in our simulations only the bifurcations of fourth to eighth order since the numerical results are very hard to be obtained for higher orders in the bifurcations .
our result is in a good agreement with the feigenbaum s universal @xmath0 .
as a summary , we have studied a dissipative version of the fermi - ulam model .
we introduce dissipation into the model through inelastic collisions with both walls .
we have shown that for regimes of high dissipation , the model exhibits a sequence of doubling bifurcation cascade .
for this cascade , we obtained the so called feigenbaum s number as @xmath75 .
dfmo gratefully acknowledges conselho nacional de desenvolvimento cientfico e tecnolgico cnpq .
edl is grateful to fapesp , cnpq and fundunesp , brazilian agencies . | we have studied a dissipative version of a one - dimensional fermi accelerator model .
the dynamics of the model is described in terms of a two - dimensional , nonlinear area - contracting map .
the dissipation is introduced via innelastic collisions of the particle with the walls and we consider the dynamics in the regime of high dissipation .
for such a regime , the model exhibits a route to chaos known as period doubling and we obtain a constant along the bifurcations so called the feigenbaum s number @xmath0 . |
the results presented in this paper are obtained within the framework of hartree - fock based semi - empirical methods .
this methodology is well tested and reliable to compute the electronic and optical properties of c - conjugated low - dimensional systems @xcite , including pahs themselves @xcite .
the am1 model @xcite is adopted for structural optimization [ 0.4 @xmath3/(mol @xmath4 ) threshold for the forces ] , whereas the zindo / s model @xcite , based on configuration interaction ( ci ) including single excitations only , is employed for the optical spectra . to better characterize the optical excitations ,
we introduce the _ transition density _
@xcite , which is defined as : @xmath5 where @xmath6 ( @xmath7 ) indicates the occupied ( unoccupied ) molecular orbitals ( mos ) and @xmath8 are the ci coefficients of the @xmath9 excited - state configuration @xcite .
the transition density is directly related to the transition dipole moment of the @xmath9 excitation @xcite : @xmath10 where @xmath11 is the momentum operator , @xmath12 is the ground state hartree - fock wave function and @xmath13 the singly excited configuration , including all the possible slater determinants obtained by coupling @xmath14 transitions in the given energy window . from [ eq : trans - dip ]
it is thus clear that the transition density provides a representation of the polarization of the excitation , which can be visualized through a three dimensional isosurface representation @xcite .
finally , as the oscillator strength is proportional to the square modulus of the transition dipole moment , the transition density also carries information about the excitation intensity . point group symmetry , while in the latter case ( below ) the hexagonal symmetry is broken and elongated pahs , with @xmath1 point group symmetry are formed . ] starting from coronene ( ) as multifunctional seed , the effects of size and symmetry modifications on the optical properties are investigated by comparing two families of pahs , characterized by @xmath0 and @xmath1 point group symmetry ( see [ fig1 ] ) .
the first class of hexagonal molecules is a kekul series of fully - benzenoid pahs with purely armchair edges @xcite , obtained by progressively incrementing the number of aromatic rings at the periphery . as a result ,
the molecules belonging to this group are characterized by an increasing number of armchair - shaped rings per edge , ranging from 2 in hexa - benzo - coronene ( hbc , ) up to 4 in ( see [ fig1 ] ) . the second class of molecules is built by anisotropically elongating coronene in one direction only ( @xmath15 axis ) , as indicated in [ fig1 ] .
this gives rise to ribbon - like graphene nanoflakes ( gnfs ) with armchair - shaped edges .
these structures belong to the @xmath1 point group and can be identified through their length ( @xmath16 ) and width parameters ( @xmath17 ) @xcite .
the latter is defined as for periodic armchair graphene nanoribbons as the number of c - dimers along the @xmath18 axis of the molecule ( see [ fig1 ] ) @xcite , while @xmath16 is identified as the number of zigzag chains in the @xmath15 direction , excluding the ends .
the coronene - derived gnfs considered here have fixed width of about 7.4 ( @xmath17=7 ) and increasing length , ranging from about 11 ( @xmath16=4 , circumbiphenyl ) to almost 20 ( @xmath16=8 ) .
note that in both cases we focus on fully armchair - edged molecules to avoid any spurious open - shell effects , typically associated with the presence of zigzag edges @xcite .
point group symmetry ) of different sizes .
the first intense peak p1 ( clar s @xmath19 band ) is given by doubly degenerate excitations .
the lowest energy dipole - forbidden excitations ( d1 and d2 , @xmath20 and @xmath21 bands , according to clar s notation , see also [ table1 ] ) are indicated in the spectra by solid and dashed arrows , respectively .
the spectra are obtained by using a lorentzian broadening of 100 mev . ] the optical absorption spectrum of coronene ( [ fig2 ] , top panel ) is characterized by an intense peak in the near - uv region ( p1 , at 4.26 ev , see [ table1 ] ) and by two low energy dark excitations in the visible band ( d1 and d2 at about 3 and 3.5 ev , respectively ) .
the lowest energy excited states of hexagonal pahs , corresponding to transitions between the doubly degenerate frontier orbitals ( see [ fig4]a ) and here labeled as d1 , d2 and p1 , are commonly identified as @xmath20 , @xmath22 ( @xmath21 ) and @xmath19 bands , respectively , according to clar s notation ( see e.g. ref . @xcite and references therein ) .
these excitations correspond to symmetry forbidden @xmath23 and @xmath24 states ( d1 and d2 in [ fig2 ] ) , and to a higher energy @xmath25 state , characterized by a twofold degeneracy and by a large oscillator strength ( p1 peak ) . within the ci approach
, we find that p1 stems from the combined transitions of doubly degenerate frontier orbitals ( highest occupied mo , homo , and lowest unoccupied mo , lumo , see [ table1 ] ) .
also d1 and d2 arise from frontier orbital transitions that combine here with opposite sign , hence giving rise to dipole - forbidden transitions .
two additional excitations
one of them doubly degenerate appear in the spectrum between d2 and p1 : they are symmetry - forbidden and come from combinations of transitions between the four occupied and unoccupied mos closest to the gap . the features of d1 , d2 and p1 can be clearly captured from the transition density plots in [ fig3 ] .
the modulation of the positive and negative domains in d1 and d2 provides evidence of the zero dipole moment associated to them , while p1 is characterized by a remarkable polarity .
these results are in agreement with the available experimental data @xcite within about 150 mev , as well as with other predictions obtained within different theoretical approaches @xcite .
the lowest energy peak is typically identified in experiments as the strongest absorption peak in the uv / vis region . moreover , as a matter of fact , the lowest dipole - forbidden excitations are also detected in experiments @xcite as associated to vibrational modes @xcite or , in general , with structural deviations from the ideal symmetry , generally adopted in theoretical simulations . .
c|c|c|c|c|c molecule & excitation & mo transitions ( weight ) & @xmath26 & energy [ ev ] & os + & & h @xmath27 l+1 ( 0.47 ) & -0.683 & & + & & h-1 @xmath27 l ( 0.47 ) & 0.683 & & + & & h @xmath27 l ( 0.49 ) & -0.700 & & + & & h-1 @xmath27 l+1 ( 0.49 ) & -0.700 & & + & & h @xmath27 l ( 0.48 ) & 0.690 & & + & & h-1 @xmath27 l+1 ( 0.48 ) & -0.690 & & + & & h @xmath27 l+1 ( 0.48 ) & 0.690 & & + & & h-1 @xmath27 l ( 0.48 ) & 0.690 & & + & & h @xmath27 l+1 ( 0.44 ) & -0.665 & & + & & h-1 @xmath27 l ( 0.44 ) & 0.665 & & + & & h @xmath27 l ( 0.47 ) & -0.684 & & + & & h-1 @xmath27 l+1 ( 0.47 ) & -0.684 & & + & & h @xmath27 l ( 0.18 ) & 0.428 & & + & & h @xmath27 l+1 ( 0.26 ) & -0.514 & & + & & h-1 @xmath27 l ( 0.26 ) & -0.514 & & + & & h-1 @xmath27 l+1 ( 0.18 ) & -0.428 & & + & & h @xmath27 l+1 ( 0.26 ) & -0.514 & & + & & h @xmath27 l ( 0.18 ) & -0.428 & & + & & h-1 @xmath27 l ( 0.18 ) & -0.428 & & + & & h-1 @xmath27 l+1 ( 0.26 ) & 0.514 & & + & & h @xmath27
l+1(0.42 ) & 0.649 & & + & & h-1@xmath27 l ( 0.42 ) & -0.649 & & + & & h @xmath27
l ( 0.44 ) & 0.666 & & + & & h-1 @xmath27 l+1 ( 0.44 ) & 0.666 & & + & & h @xmath27 l ( 0.41 ) & 0.638 & & + & & h @xmath27 l+1 ( 0.05 ) & 0.228 & & + & & h-1 @xmath27 l ( 0.05 ) & 0.228 & & + & & h-1 @xmath27 l+1 ( 0.41 ) & -0.638 & & + & & h @xmath27 l+1 ( 0.41 ) & -0.638 & & + & & h @xmath27 l ( 0.05 ) & 0.228 & & + & & h-1 @xmath27 l ( 0.41 ) & -0.638 & & + & & h-1 @xmath27 l+1 ( 0.05 ) & -0.228 & & + & & h @xmath27 l+1 ( 0.40 ) & 0.629 & & + & & h-1 @xmath27 l ( 0.40 ) & -0.629 & & + & & h @xmath27 l ( 0.42 ) & 0.646 & & + & & h-1 @xmath27 l+1 ( 0.42 ) & 0.646 & & + & & h @xmath27 l ( 0.40 ) & 0.635 & & + & & h-1 @xmath27 l+1 ( 0.40 ) & -0.635 & & + & & h @xmath27
l+1 ( 0.40 ) & 0.635 & & + & & h-1 @xmath27 l ( 0.40 ) & 0.635 & & + for this series of hexagonal pahs , [ fig2 ] , from the top to the bottom , indicates that the full optical spectrum , including dark excitations , is red - shifted with increasing size , as expected from the higher delocalization of the relevant molecular orbitals , as recently reported for other pah series @xcite .
specifically , p1 is brought from the uv to the visible range ( green ) , while the lowest - energy dark excitations move from the visible ( violet ) to the infrared band . as such , p1 undergoes a larger red shift than d1 and d2 , in comparison with the spectrum of coronene .
the lowest energy excitation of hbc ( d1 ) , corresponding to the clar s @xmath20 band , is found in our study at 2.84 ev ( [ table1 ] ) .
this value is in excellent agreement with recent experimental results of hbc in gas phase @xcite and in rare gas matrix @xcite : these studies assign the lowest energy excitation of hbc , at 2.86 ev and at 2.85 ev , respectively . in ref .
@xcite also the position of the @xmath21 band ( d2 , 3.22 ev , see [ table1 ] ) is reported at 3.30 3.35 ev , depending on the embedding gas matrix .
also for this excitation our results compare very well with the experiment within an error of 3@xmath28 .
the first active peak of hbc ( p1 ) stems from our calculations at 3.82 ev ( [ table1 ] ) , also in very good agreement with the experimental data at about 3.7 ev ( error of 5@xmath28 ) @xcite . as a consequence of the preserved symmetry ,
the main optical features in all the examined hexagonal pahs remain unchanged throughout the series ( see [ fig2 ] and [ table1 ] ) .
this is further confirmed by the resulting transition densities shown in [ fig3 ] .
in fact , the mo s contributing to the excitations are the same in all the structures , as their character is maintained by the molecular symmetry . in the low energy
spectral region , between d2 and p1 , only one doubly degenerate excitation appears in the spectra of the larger molecules of this set ( hbc , and ) .
as discussed for d1 and d2 , also this dark excitation can be activated when coupled to vibrational modes ( see e.g. ref .
@xcite for the case of hbc ) .
higher energy excitations than p1 fall beyond the spectral range of interest ( visible and near - uv bands ) for coronene and hbc , while this is not the case for and .
as shown in [ fig2 ] , higher energy peaks stem from transitions between deeper occupied and higher unoccupied mos , carrying much weaker intensity than p1 . , of the examined optical excitations of hexagonal pahs ( @xmath0 point group symmetry ) .
] proceeding now with the results for the elongated gnfs , the drastic effect on the electronic and optical properties induced by symmetry lowering from @xmath0 to @xmath1 point group is evident already from the comparison between coronene and circumbiphenyl ( see [ fig4 ] ) .
the doubly degenerate frontier orbitals of coronene ( both homo and lumo ) are split in circumbiphenyl ( homo and homo-1 , lumo and lumo+1 , [ fig4]a ) : the existence of two inequivalent axes ( @xmath15 and @xmath18 ) in the molecular plane of the @xmath16=4 gnf induces a different parity for homo and homo-1 , as well as for lumo and lumo+1 . in view of this reduced symmetry , the two lowest energy excitations ( t1 and l1 ) are no longer dipole forbidden , as in the case of the @xmath20 and @xmath21 bands in the hexagonal pahs ( [ fig4]b ) .
moreover , the main peak p1 of the @xmath0 series ( clar s @xmath19 band ) splits into two distinct absorption lines , l2 and t2 , with 0.1 ev energy separation .
these excitations are here labelled according to their polarization , as visualized in the transition density plots in [ fig5 ] : l1 and l2 are polarized along the longitudinal ( @xmath15 ) axis of the flake , while t1 and t2 along the transverse direction ( @xmath18 axis ) . by looking at the composition of the excitations ( [ table2 ] ) ,
one notices that t1 and t2 stem from the homo @xmath27 lumo+1 and homo-1 @xmath27 lumo transitions , whereas l1 and l2 are due to the homo @xmath27 lumo and homo-1 @xmath27 lumo+1 transitions .
however , while in t1 and t2 the weight of the two transitions is almost equivalent , the dominant contribution for l1 ( l2 ) is given by homo @xmath27 lumo ( homo-1 @xmath27 lumo+1 ) .
the anisotropy of circumbiphenyl is reflected also in the different oscillator strength of the longitudinally and transversally polarized excitations .
the degeneracy of homo and lumo of coronene is broken in the @xmath1 pahs .
b ) calculated optical absorption spectra of ribbon - like pahs , with fixed width @xmath17=7 ( about 7.4 ) and variable length parameter ranging form @xmath16=4 to @xmath16=8 .
the main optical excitations , polarized either in the longitudinal ( l , @xmath15 axis ) or transverse ( t , @xmath18 axis ) direction are indicated in the spectrum of circumbiphenyl ( @xmath16=4 ) .
the dashed - dotted arrows indicate the position of the lowest energy , weak excitation t1 . for comparison ,
the spectrum of coronene is also shown ( shaded red curve , top , see also [ fig2 ] ) .
the spectra are obtained through a lorentzian broadening of 100 mev . ] the relative intensity of the excitations is connected with the anisotropy
_ i.e. _ length / width ratio of the @xmath1 pahs .
this phenomenon , which was discussed for longer flakes of the same family @xcite , can be definitely clarified here for these short structures . as quantified in [ table2 ] and illustrated in [ fig4]b , going from the @xmath16=4 to the @xmath16=8 ribbon - like pahs , the oscillator strengths increases by almost one order of magnitude , while the intensity of l2 is incremented only by a factor of 1.5
( see also [ table2 ] ) . a visual picture of this feature is provided by the transition densities in [ fig5 ] , which indicates a similar modulation of l1 and l2 in the @xmath16=8 pah , compared to the stronger dipolar character of l2 for circumbiphenyl .
in the case of the l1 excitation , the weight of the homo @xmath27 lumo increases with the length , at the expense of the homo-1 @xmath27 lumo+1 transition : in circumbiphenyl the ratio between these two single - particle transitions is about 1/3 , while it decreases to less than 1/8 in the @xmath16=8 pah .
the same occurs for l2 , with a reversed contributions ( [ table2 ] ) . in the limit of very long ribbon - like pahs with @xmath17=7 and length parameter up to @xmath16=20 , corresponding to a length of about 88 , l1 and l2 present an equal oscillator strength @xcite . also here
the increasing length of the flakes induces a red shift of the spectra ( see [ fig4]b ) , as observed in hexagonal pahs ( [ fig2 ] ) . on the other hand ,
since the flake width is held fixed , the excitations with transverse polarization , t1 and t2 , are very similar in all the considered @xmath1 structures : their energy is only slightly modified , and the oscillator strength is preserved along with the relative weight of the dominant mo transitions , so that on longer flakes the direct excitons will be lower than the forbidden ones @xcite . finally , when comparing the energy of the first active excitation for symmetric ( @xmath0 ) and elongated pahs ( @xmath1 ) , one notices that a peak at @xmath292.9 ev is found for both the hexagonal molecule and for the flake ( @xmath16=6 ) , which present a striking ratio of molecular masses with respect to each other . c|c|c|c|c|c molecule & excitation & mo transitions ( weight ) & @xmath26 & energy [ ev ] & os + & & h @xmath27 l+1 ( 0.46 ) & -0.676 & & + & & h-1 @xmath27 l ( 0.44 ) & 0.666 & & + & & h @xmath27 l ( 0.71 ) & 0.842 & & + & & h-1 @xmath27 l+1 ( 0.25 ) & 0.502 & & + & & h @xmath27 l ( 0.26 ) & 0.507 & & + & & h-1 @xmath27 l+1 ( 0.69 ) & -0.834 & & + & & h @xmath27 l+1 ( 0.42 ) & 0.649 & & + & & h-1 @xmath27 l ( 0.43 ) & 0.660 & & + & & h
@xmath27 l+1 ( 0.44 ) & 0.663 & & + & & h-1 @xmath27 l ( 0.42 ) & -0.646 & & + & & h @xmath27 l ( 0.78 ) & 0.884 & & + & & h-1 @xmath27 l+1 ( 0.15 ) & 0.387 & & + & & h @xmath27 l ( 0.16 ) & 0.403 & & + & & h-1 @xmath27 l ( 0.76 ) & -0.873 & & + & & h @xmath27 l+1 ( 0.43 ) & 0.654 & & + & & h-1 @xmath27 l ( 0.45 ) & 0.671 & & + & & h
@xmath27 l+1 ( 0.41 ) & 0.643 & & + & & h-1 @xmath27 l ( 0.39 ) & 0.627 & & + & l1 & h @xmath27 l ( 0.80 ) & 0.895 & 2.80 & 1.04 + & & h @xmath27 l ( 0.10 ) & 0.323 & & + & & h-1 @xmath27 l ( 0.77 ) & 0.880 & & + & & h @xmath27 l+1 ( 0.41 ) & -0.639 & & + & & h-1 @xmath27 l ( 0.43 ) & 0.658 & & + , of the examined optical excitations of ribbon - like gnfs with @xmath1 point group symmetry . ]
in conclusion , we have presented a detailed quantum - chemical analysis of the effects of size and anisotropy on the optical properties of two prototypical classes of pahs , with @xmath0 and @xmath1 symmetry .
having ruled out spurious effects with the choice of structures with purely armchair edges , the optical properties of these systems appear strongly dependent both on the size and on the anisotropy of the molecule .
most importantly , our findings indicate that the energy of the lowest energy active peak depends more crucially on the molecular symmetry , than on the molecular weight , related in this case to the number of c atoms in the structure . in gnfs with @xmath1 symmetry
optically active peaks appear in the visible band already in the spectrum of circumbiphenyl ( ) . on the other hand , only hexagonal molecules with a few hundreds of c atoms can absorb a photon in the same spectral region .
the results of this work provide significant insights into the interpretation of the experimental spectra of pahs and consequently into the physical mechanisms driving their optical absorption .
moreover , they confirm the large tunability of this class of compounds upon size and symmetry variations , regarding the optical properties . in the field of optoelectronics
these outcomes represent a valuable indication in view of designing molecular nanodevices .
finally , the detailed investigation carried out in the present study on the size and anisotropy effects on the optical spectra of representative classes of pahs can serve as a guide for the astronomical observation of these carbon compounds . specifically , the symmetry analysis of the main excitations in the visible region with respect of the molecular weight is a valuable indication for determining the relative abundance of medium - size aromatic molecules , and hence clarifying the composition of the ism .
the authors are grateful to stefano corni and franco gianturco for stimulating discussions .
cineca is acknowledged for computational support .
part of this research was supported by : the italian ministry of research through the national projects prin - graf ( grant no .
20105zztse ) ,
firb - flashit ( grant no .
rbfr12swoj ) , and the program
progetto premiale 2012 " - project abnanotech ; the italian ministry of foreign affairs through the grant no .
m. j. c. acknowledges support from fapesp and cnpq ( brazil ) .
arfsten , d. p. ; schaeffer , d. j. ; mulveny , d. c. the effects of near ultraviolet radiation on the toxic effects of polycyclic aromatic hydrocarbons in animals and plants : a review .
_ ecotox .
environ .
safe _ * 1996 * , _ 33 _ , 124 clar , e. ; schoental , r. _ polycyclic hydrocarbons _ ; academic press new york , 1964 ; vol
. 2 wu , j. ; pisula , w. ; muellen , k. graphenes as potential material for electronics .
_ * 2007 * , _ 107 _ , 718747 tielens , a. interstellar polycyclic aromatic hydrocarbon molecules . _
astrophys . _ * 2008 * , _ 46 _ , 289337 leger , a. ; puget , j. identification of the unidentified ir emission features of interstellar dust ? _ astronomy and astrophysics _ * 1984 * , _ 137 _ , l5l8 allamandola , l. ; tielens , a. ; barker , j. polycyclic aromatic hydrocarbons and the unidentified infrared emission bands - auto exhaust along the milky way . _
astrophys .
j. _ * 1985 * , _ 290 _ , l25l28 lger , a. ; dhendecourt , l. are polycyclic aromatic hydrocarbons the carriers of the diffuse interstellar bands in the visible ?
_ astron .
astrophys . _ * 1985 * , _ 146 _ , 8185 van der zwet , g. ; allamandola , l. polycyclic aromatic hydrocarbons and the diffuse interstellar bands .
_ astron .
astrophys . _
* 1985 * , _ 146 _ , 7680 salama , f. ; galazutdinov , g. ; kreowski , j. ; allamandola , l. ; musaev , f. polycyclic aromatic hydrocarbons and the diffuse interstellar bands : a survey . _
the astrophysical journal _ * 1999 * , _ 526 _ , 265 allamandola , l. ; tielens , a. ; barker , j. interstellar polycyclic aromatic hydrocarbons : the infrared emission bands , the excitation / emission mechanism , and the astrophysical implications .
_ astrophys .
j. suppl .
s. _ * 1989 * , _ 71 _ , 733775 allamandola , l. ; hudgins , d. ; sandford , s. modeling the unidentified infrared emission with combinations of polycyclic aromatic hydrocarbons .
_ astrophys .
j. lett . _ * 1999 * , _ 511 _ , l115 malloci , g. ; joblin , c. ; mulas , g. on - line database of the spectral properties of polycyclic aromatic hydrocarbons .
phys . _ * 2007 * , _ 332 _ , 353359 bauschlicher jr , c. ; boersma , c. ; ricca , a. ; mattioda , a. ; cami , j. ; peeters , e. ; de armas , f. s. ; saborido , g. p. ; hudgins , d. ; allamandola , l. the nasa ames polycyclic aromatic hydrocarbon infrared spectroscopic database : the computed spectra . _ astrophys . j. suppl
. s. _ * 2010 * , _ 189 _ , 341 malloci , g. ; mulas , g. ; joblin , c. electronic absorption spectra of pahs up to vacuum uv .
astrophys . _ * 2004 * , _ 426 _ , 105117 rieger , r. ; mllen , k. forever young : polycyclic aromatic hydrocarbons as model cases for structural and optical studies .
_ j. phys .
_ * 2010 * , _ 23 _ , 315325 crawford , m. ; tielens , a. ; allamandola , l. ionized polycyclic aromatic hydrocarbons and the diffuse interstellar bands . _
astrophys . j. _ * 1985 * , _ 293 _ , l45l48 canuto , s. ; zerner , m. c. ; diercksen , g. h. theoretical studies of the absorption spectra of polycyclic aromatic hydrocarbons .
_ astrophys . j. _ * 1991 * , _ 377 _ , 150157 rastogi , s. ; pathak , a. ; maurya , a. polycyclic aromatic hydrocarbon molecules in astrophysics .
_ aip conf .
proc . _ * 2013 * , _ 1543 _ , 49 carelli , f. ; grassi , t. ; gianturco , f. electron attachment rates for pah anions in the ism and dark molecular clouds : dependence on their chemical properties .
. astrophys . _ * 2013 * , _ 549 _ , a103 malloci , g. ; mulas , g. ; cecchi - pestellini , c. ; joblin , c. dehydrogenated polycyclic aromatic hydrocarbons and uv bump .
_ astron .
astrophys . _ * 2008 * , _ 489 _ , 11831187 carelli , f. ; sebastianelli , f. ; baccarelli , i. ; gianturco , f. electron - driven reactions in proto - planetary atmospheres : metastable anions of gaseous o - benzyne .
. j. _ * 2010 * , _ 712 _ , 445 carelli , f. ; sebastianelli , f. ; satta , m. ; gianturco , f. gas - phase route to polycyclic aromatic hydrocarbon formation in protoplanetary atmospheres : role of stabilized benzyne anions . _
mon . not .
r. astron .
soc . _ * 2011 * , _ 415 _ , 425430 charles w. bauschlicher , j. ; ricca , a. the infrared spectra of polycyclic aromatic hydrocarbons with some or all hydrogen atoms removed .
_ astrophys .
j. _ * 2013 * , _ 776 _ , 102 watson , m. d. ; fechtenktter , a. ; llen , k. big is beautiful - aromaticity revisited from the viewpoint of macromolecular and supramolecular benzene chemistry . _ chem .
rev . _ * 2001 * , _ 101 _ , 12671300 doetz , f. ; brand , j. d. ; ito , s. ; gherghel , l. ; muellen , k. synthesis of large polycyclic aromatic hydrocarbons : variation of size and periphery . _ j. am .
_ * 2000 * , _ 122 _ , 77077717 liu , r. ; wu , d. ; feng , x. ; mullen , k. bottom - up fabrication of photoluminescent graphene quantum dots with uniform morphology . _
j. am .
_ * 2011 * , _ 133 _ , 1522115223 zhi , l. ; mllen , k. a bottom - up approach from molecular nanographenes to unconventional carbon materials .
_ j. mater .
* 2008 * , _ 18 _ , 14721484 feng , x. ; pisula , w. ; mllen , k. large polycyclic aromatic hydrocarbons : synthesis and discotic organization .
_ pure appl .
chem . _ * 2009 * , _ 81 _ , 22032224 pisula , w. ; feng , x. ; mullen , k. charge - carrier transporting graphene - type molecules . _ chem .
mater . _ * 2010 * , _ 23 _ , 554567 grimsdale , a. c. ; mllen , k. the chemistry of organic nanomaterials . _
* 2005 * , _ 44 _ , 55925629 pisula , w. ; zorn , m. ; chang , j. y. ; mllen , k. ; zentel , r. liquid crystalline ordering and charge transport in semiconducting materials .
_ macromol .
commun . _ * 2009 * , _ 30 _ , 11791202 am1 and zindo / s calculations were performed using vamp package included in accelrys materials studio software , version 5.0 ( http://accelrys.com/products/materials-studio ) .
wetmore , s. d. ; boyd , r. j. ; eriksson , l. a. electron affinities and ionization potentials of nucleotide bases .
_ * 2000 * , _ 322 _ , 129 135 caldas , m. j. ; pettenati , e. ; goldoni , g. ; molinari , e. tailoring of light emission properties of functionalized oligothiophenes . _
lett . _ * 2001 * , _ 79 _ , 25052507 dvila , l. y. a. ; caldas , m. j. applicability of mndo techniques am1 and pm3 to ring - structured polymers . _ j. comput .
_ * 2002 * , _ 23 _ , 1135 kubatkin , s. ; danilov , a. ; hjort , m. ; cornil , j. ; brdas , j. ; stuhr - hansen , n. ; hedegrd , p. ; bjrnholm , t. single - electron transistor of a single organic molecule with access to several redox states . _
nature ( london ) _ * 2003 * , _ 425 _ , 698701 steglich , m. ; jger , c. ; rouill , g. ; huisken , f. ; mutschke , h. ; henning , t. electronic spectroscopy of medium - sized polycyclic aromatic hydrocarbons : implications for the carriers of the 2175 uv bump .
_ astrophys .
j. lett . _ * 2010 * , _ 712 _ , l16 dewar , m. j. s. ; zoebish , e. g. ; healy , e. f. ; stewart , j. j. p. a new general purpose quantum mechanical molecular model .
_ j. am .
soc . _ * 1985 * , _ 107 _ , 39023909 ridley , j. ; zerner , m. an intermediate neglect of differential overlap techinique for spectroscopy : pyrrole and the azines .
_ theor .
acta _ * 1973 * , _ 32 _ , 111134 fischer - hjalmars , i. transition densities of conjugated molecules and intensities of forbidden bands . _ j. mol .
spectrosc . _ * 1971 * , _ 39 _ , 321331 dreuw , a. ; head - gordon , m. single - reference ab initio methods for the calculation of excited states of large molecules . _ chem .
rev . _ * 2005 * , _ 105 _ , 40094037 jensen , f. _ introduction to computational chemistry _ , 2nd ed . ; wiley , 2007 beenken , w. j. ; pullerits , t. spectroscopic units in conjugated polymers : a quantum chemically founded concept ?
_ j. phys .
b _ * 2004 * , _ 108 _ , 61646169 krueger , b. p. ; scholes , g. d. ; fleming , g. r. calculation of couplings and energy - transfer pathways between the pigments of lh2 by the ab initio transition density cube method .
_ j. phys .
b _ * 1998 * , _ 102 _ , 53785386 sun , m. ; kjellberg , p. ; ma , f. ; pullerits , t. excited state properties of acceptor - substitute carotenoids : 2d and 3d real - space analysis .
lett . _ * 2005 * , _ 401 _ , 558564 stein , s. e. ; brown , r. @xmath2-electron properties of large condensed polyaromatic hydrocarbons . _
j. am .
soc . _ * 1987 * , _ 109 _ , 37213729 cocchi , c. ; prezzi , d. ; ruini , a. ; benassi , e. ; caldas , m. j. ; corni , s. ; molinari , e. optical excitations and field enhancement in short graphene nanoribbons .
_ j. phys .
lett . _ * 2012 * , _ 3 _ , 924929 nakada , k. ; fujita , m. ; dresselhaus , g. ; dresselhaus , m. s. edge state in graphene ribbons : nanometer size effect and edge shape dependence .
b _ * 1996 * , _ 54 _ , 1795417961 sun , z. ; wu , j. open - shell polycyclic aromatic hydrocarbons . _
j. mater .
chem . _ * 2012 * , _ 22 _ , 41514160 katraro , r. ; ron , a. ; speiser , s. photophysical studies of coronene and 1 , 12-benzperylene .
self - quenching , photoquenching , temperature dependent fluorescence decay and temperature dependent electronic energy transfer to dye acceptors . _
phys . _ * 1979 * , _ 42 _ , 121132 weisman , j. l. ; lee , t. j. ; salama , f. ; head - gordon , m. time - dependent density functional theory calculations of large compact polycyclic aromatic hydrocarbon cations : implications for the diffuse interstellar bands . _ the astrophysical journal _ * 2003 * , _ 587 _ , 256 maruyama , y. ; iwasaki , n. absorption spectra of amorphous organic films .
lett . _ * 1974 * , _ 24 _ , 2629 orlandi , g. ; zerbetto , f. quantum - chemical analysis of the absorption and emission spectra of coronene .
* 1988 * , _ 123 _ , 175185 we recently addressed this issue in the analysis of the optical spectrum of icosahedral .
a detailed discussion can be found in ref .
an exhaustive and systematic analysis of the size effects on the electronic and optical properties of a representative set of pahs has been recently presented by malloci and coworkers , see ref .
kokkin , d. l. ; troy , t. p. ; nakajima , m. ; nauta , k. ; varberg , t. d. ; metha , g. f. ; lucas , n. t. ; schmidt , t. w. the optical spectrum of a large isolated polycyclic aromatic hydrocarbon : hexa - peri - hexabenzocoronene , @xmath30 .
_ astrophys .
j. lett . _ * 2008 * , _ 681 _ , l49 rouill , g. ; steglich , m. ; huisken , f. ; henning , t. ; mllen , k. uv / visible spectroscopy of matrix - isolated hexa - peri - hexabenzocoronene : interacting electronic states and astrophysical context .
_ j. chem .
phys . _ * 2009 * , _ 131 _ , 204311 prezzi , d. ; varsano , d. ; ruini , a. ; marini , a. ; molinari , e. optical properties of graphene nanoribbons : the role of many - body effects .
b _ * 2008 * , _ 77 _ , 041404(r ) yang , l. ; cohen , m. ; louie , s. excitonic effects in the optical spectra of graphene nanoribbons .
_ nano lett . _ * 2007 * , _ 7 _ , 31123115 cocchi , c. ; prezzi , d. ; ruini , a. ; caldas , m. j. ; fasolino , a. ; molinari , e. concavity effects on the optical properties of aromatic hydrocarbons . _ j. phys . chem
. c _ * 2013 * , _ 117 _ , 1290912915 malloci , g. ; cappellini , g. ; mulas , g. ; mattoni , a. electronic and pptical properties of families of polycyclic aromatic hydrocarbons : a systematic ( time - dependent ) density functional theory study .
* 2011 * , _ 384 _ , 1927 | the electronic and optical properties of polycyclic aromatic hydrocarbons ( pahs ) present a strong dependence on their size and geometry .
we tackle this issue by analyzing the spectral features of two prototypical classes of pahs , belonging to @xmath0 and @xmath1 symmetry point groups and related to coronene as multifunctional seed .
while the size variation induces an overall red shift of the spectra and a redistribution of the oscillator strength between the main peaks , a lower molecular symmetry is responsible for the appearance of new optical features .
along with broken molecular orbital degeneracies , optical peaks split and dark states are activated in the low - energy part of the spectrum . supported by a systematic analysis of the composition and the character of the optical transitions , our results contribute in shedding light to the mechanisms responsible for spectral modifications in the visible and near uv absorption bands of medium - size pahs .
* keywords * : astrochemistry , coronene , configuration interaction , graphene nanoribbons , transition density , zindo polycyclic aromatic hydrocarbons ( pahs ) are the focus of research in significantly diverse fields , ranging from biochemistry @xcite to theoretical chemistry @xcite , from molecular electronics @xcite to astrochemistry @xcite , where pahs are assigned an important role @xcite due to their chemical composition and spectral features @xcite . in this framework , a theoretical identification of the absorption and emission features of these molecules assumes a significant relevance @xcite , since they might drastically vary both upon small structural modifications @xcite and as a consequence of ionization @xcite or dehydrogenation @xcite . in the field of molecular electronics ,
the interest in medium- and large - sized pahs is intrinsically connected to graphene , as they are viewed as highly stable graphene molecules @xcite .
their intriguing and versatile optical properties @xcite , combined with their facile bottom - up synthesis @xcite and their capability to assemble into @xmath2-@xmath2 stacked superstructures @xcite , make them excellent candidates for opto - electronic applications , @xcite where slight structural variations can trigger different properties and functionalities @xcite . in this paper
we investigate the effects of size and anisotropy on the optical spectra of pahs , with the aim of providing an useful reference for understanding and predicting their optical response . to tackle this issue , we focus on two prototypical classes of molecules with @xmath0 and @xmath1 point group symmetry , which are related to coronene , viewed as a multifunctional seed . indeed , with its high symmetry and specific electronic structure , coronene is one of the most representative medium - size pahs @xcite . by means of a well - established quantum chemistry approach
, we perform a theoretical analysis to identify the size- and anisotropy - induced modifications to the spectral features in the visible and near - uv bands of pahs .
the present study aims at showing that the main effect of size increase , moving from coronene to larger hexagonal pahs of fixed symmetry , is an oscillator strength variation of the main excitations and a red shift of the whole spectrum , which brings the strong absorption peaks from the uv to the visible energy region , maintaining overall the character of the excitons .
moreover , the additional features produced in the spectra by the anisotropical elongation of coronene in one direction will be investigated ; specific focus will be devoted to the effects of symmetry breaking .
finally , it will be clarified that bright excitations in the visible region appear already for the smaller anisotropic molecules , with low mass compared to highly symmetric pahs . |
phase transition in quantum chromodynamics ( qcd ) is one of the central subjects in high energy nuclear physics both theoretically and experimentally .
first principle lattice qcd ( lqcd ) calculations have shown that the transition from quark - gluon plasma ( qgp ) to hadronic matter is of crossover type at physical quark masses @xcite , in which order parameters and thermodynamic quantities change smoothly as functions of temperature . at finite baryon density
, one expects that the nature of the transition can change .
unfortunately , little is known about the state of matter at high baryon density from lqcd calculations because of the difficulty in numerical simulation at finite baryon chemical potential @xmath3 @xcite .
various approximation methods applied so far seem to work only @xmath4 or a small volume or heavy quark mass region .
nevertheless , effective models which implement relevant symmetries in qcd and large @xmath5 studies have shown that rich phase structure exists in high density @xcite . in particular , if there is a first order phase transition at @xmath6 and large @xmath3 , there must be a critical point ( cp ) at which the first order phase transition line terminates and the transition becomes second order .
existence of cp is supported by many chiral effective models @xcite , but its location depends on the detail of the models @xcite . stimulated by these theoretical results , the first beam energy scan program at relativistic heavy ion collider ( rhic ) has been carried out in search for the cp @xcite . since lower colliding energies leaves the incident nucleons in the central region , one expects to explore higher baryon density region at lower energies .
there are a number of observables which might have potential to indicate the transition from qgp to hadronic matter . among them
, event - by - event fluctuations of conserved charges are intimately connected to critical behavior associated with the phase transition @xcite .
measurements of the net - proton number fluctuations as a proxy of the net - baryon one @xcite and net electric charges have been presented for au+au collisions at various energies from @xmath7 gev to 200 gev @xcite .
given the fact that multiplicity of different particle species are well described by statistical models @xcite , one may regard event - by - event fluctuations of conserved charges as those of the grand canonical ensembles at chemical freeze - out temperature @xmath8 and baryonic chemical potential @xmath3 . through systematic analyses of the location of @xmath9 corresponding to each colliding systems @xcite
, one can map experimental measurements for property of the matter on @xmath10 plane .
furthermore , recent lqcd results at physical quark masses indicate that the crossover region coincides with the chemical freeze - out at least for @xmath4 @xcite
. one may look for remnant of the chiral criticality in the crossover region , originating from second order phase transition in the vanishing quark mass @xcite .
property of the transition can be characterized by behavior of fluctuations of conserved charges as well as an order parameter and its fluctuations @xcite . in the case of the chiral phase transition in qcd , the chiral order parameter or quark - antiquark condensate @xmath11 couples to quarks carrying the baryon number and the electric charge .
thus , the second order chiral phase transition in the chiral limit at finite temperature is characterized by not only divergent fluctuation of the order parameter but also higher order cumulants of the net baryon number and the net electric charge @xcite .
the divergence of the conserved charge fluctuations is governed by the critical exponent of the specific heat @xmath12 which depends on the universality class qcd belongs to .
although it is not completely determined yet @xcite , recent simulations @xcite indicate @xmath13 in the three dimensions , as conjectured by pisarski and wilczek @xcite . in this case
, the first divergent cumulant appears at the sixth order . at finite but small quark masses ,
the divergence is replaced by sign change , owing to the property of the universal @xmath13 scaling function @xcite . at nonzero
net baryon number density , the divergence at @xmath14-th order cumulants in the chiral limit appears at @xmath15-th order one .
the tricritical point in the chiral limit becomes the cp , where the second order cumulants diverge
. the chemical freeze - out line may locate at lower temperature than the chiral phase boundary @xcite such that measured fluctuations might not reflect those at the phase transition @xcite .
nevertheless , the existence of the cp is accompanied by anomalous behavior of the cumulants such as negative third and fourth order cumulants around the cp @xcite and may lead to non - monotonic behavior of the higher order cumulants as functions of @xmath16 .
indeed , the measured net - proton number cumulants in @xcite seems to follow this expectation , although still inconclusive due to uncertainty .
the measurement of the cumulants is based on event - by - event multiplicity distribution .
once the fluctuations are regarded as those of the grand canonical ensemble , the multiplicity distribution can be identified with unnormalized probability distribution . while the cumulants are expressed by central moments of the probability distribution , it is convenient for theoretical studies to compute them by differentiating the thermodynamic pressure with respect to chemical potentials .
recently , one of the authors ( k.m . ) investigated the probability distribution of the net baryon number in models with phase transitions @xcite .
it turns out that sufficient information on the tail in the probability distribution is responsible for the critical behavior of the higher order cumulants and that the remnant of the @xmath13 criticality can be characterized by narrower tail than the corresponding reference distribution . in the probability distribution ,
such information on the phase transition is encoded in the @xmath17 dependence of the canonical partition function @xmath0
. since the grand partition function is more straightforward in relativistic quantum field theories where the number of particles are not definite , computations of the canonical partition function are not generally easy . in ref .
@xcite , hasenfratz and toussaint proposed that the canonical partition function , @xmath0 , is calculated through the fourier transformation of the grand canonical partition function,@xmath18 , evaluated at pure imaginary @xmath3 . the difficulty associated with the complex fermion determinant is replaced by the highly oscillating integral which requires extraordinary numerical precision @xcite .
nevertheless , the probability distribution gives further insights into property of the system including phase transitions . in ref .
@xcite , one of the authors ( a.n . ) pointed out that one can extract the fugacity parameter @xmath19 at the chemical freeze - out and construct @xmath0 for the net baryon number without any assumption on the property of equilibrium @xmath20 .
furthermore , once @xmath0 is known , one can obtain the grand partition function @xmath18 as a series of fugacity .
this enables us to apply yang - lee theory for the phase transition @xcite(for recent reviews , see , e.g. , @xcite ) , in which zeros of the partition function give information on the thermodynamic property of the system .
the zeros of the partition function are distributed on a line in the complex plane of an external parameter and its density grows up with the system volume , then finally coalesce into the line in the thermodynamic limit .
this property leads us , in principle , to obtain the location and order of the phase transition from the distribution of the zeros . even in the absence of the phase transition , the zeros accumulated on the edge of the distribution exhibit singular behavior
this singularity , known as yang - lee edge singularity @xcite , can be regarded as a cp in the complex plane and gives influence on the thermodynamics on the real axis @xcite . in both experiments and the canonical approach in lqcd , @xmath0 at large @xmath17 requires such high statistics that obtained information is limited to some finite @xmath17
, thus one has to truncate the fugacity polynomial there in reconstructing the grand partition function ( see @xcite for recent lqcd calculations ) .
it is not a priori clear whether one can obtain the correct information on the phase transition from such a truncated partition function .
the purpose of this paper is to clarify this point .
we employ a solvable model for the chiral phase transition in qcd .
we present the yang - lee zeros in a chiral random matrix model , both for the exact grand partition function and for the reconstructed one as a truncated fugacity series with the canonical partition function being the coefficients .
we discuss effects of the truncation on the distribution of the yang - lee zeros and compare it with the spurious zeros of the skellam partition function , originated from the truncation . in the next section
, we briefly summarize the general relation among the probability distribution , partition functions , and yang - lee zeros .
a chiral random matrix model and its yang - lee zeros are presented in sec .
[ sec : chrm ] .
we demonstrate differences of the truncation effects on the yang - lee zeros between the models with and without phase transition in sec . [
sec : truncation ] .
implications for heavy ion experiments are discussed in sec .
[ sec : discussion ] .
section [ sec : conclusion ] is devoted to concluding remarks .
detailed expressions for partition functions in the chiral random matrix model are given in the appendix .
we start from experimentally measured data of net - baryon number multiplicity distribution @xmath21 , where @xmath17 is the net - baryon number . in real experiments
one measures the net - proton number @xmath22 . in principle
one can reconstruct @xmath21 from @xmath23 , @xmath24 and @xmath25 @xcite . in this study
we entirely assume the isospin invariance and regard @xmath23 as a proxy of @xmath21 .
the shape of the distribution depends on the colliding energies , centrality etc .
the net - baryon number can take any value as long as it can be packed within the system volume .
owing to limited statistics , however , we do not observe such states that have too large @xmath17 far from its mean value @xmath26 .
thus , we define the possible minimum and maximum of @xmath27 and @xmath2 as @xmath28 in thermal equilibrium , probability distribution of the net - baryon number in the grand canonical ensemble reads , for the fugacity factor @xmath19 , @xmath29 where @xmath18 is the grand partition function @xmath30\ ] ] and @xmath0 is the canonical partition function @xmath31.\label{eq : canonical}\ ] ] assuming the measured multiplicity distribution is the equilibrium one , one finds @xmath17 dependence of @xmath21 comes from @xmath32 .
using the charge - conjugate symmetry @xmath33 , one can determine @xmath34 from @xmath21 and obtain the canonical partition function @xcite @xmath35 because of limited range of @xmath17 , the canonical partition function can be obtained for @xmath36 $ ] . for the energy scan range in rhic experiments , @xmath37 , _
i.e. _ , there are a few events in which more anti - protons are observed than protons , except for @xmath7 gev where @xmath38 @xcite .
note that we need only @xmath39 thanks to the charge conjugate symmetry .
thus , in most cases , one can extract the canonical partition function for @xmath40 .
this limitation in @xmath17 also applies to theoretical approaches such as model studies @xcite and lattice simulations @xcite . in the former
, the canonical partition functions have been calculated using a projection formula @xmath41 where integration contour @xmath42 in complex @xmath34 plane can be arbitrary , but it is convenient to take the unit circle @xmath43 . then the formula becomes @xmath44 where @xmath45 is related to an imaginary chemical potential @xmath46 . in refs .
@xcite , thermodynamic potential @xmath47 in landau theory @xcite and in chiral quark - meson model @xcite was used through @xmath48 .
owing to the rapid oscillation in large @xmath17 , it turned out that the numerical integration in double precision works up to @xmath49 . in lattice qcd simulations , two approaches can provide the canonical partition functions , i.e. , ( i ) the fugacity expansion of the fermion determinant @xcite and ( ii ) hasenfratz and toussaint method @xcite . in the fugacity expansion , we must diagonalize a matrix whose rank is proportional to the lattice spacial volume .
this requires large computational resource and currently one can not go to simulations on large lattices . in the method ( ii )
, as @xmath17 increases , more accuracy is needed , and consequently @xmath17 can not go to very large .
once the canonical partition function @xmath0 is obtained , one can also construct a _ truncated _ grand canonical partition function as a series in @xmath34 @xmath50 owing to the truncation of the series at @xmath51 and @xmath2 , this partition function is an approximation of the exact partition function which could be obtained if one can take @xmath52 with @xmath53 being the number of net - baryons fulfilling the system volume @xcite . for lattice qcd at finite temperature with @xmath54 lattice ,
@xmath55 @xcite was derived for the quark fugacity series as @xmath56 .
thus , one needs to establish relations of physical quantities obtained from the truncated partition function with those from the exact partition function .
as seen in the summation running from @xmath57 , relativistic partition functions contain negative powers of @xmath34 .
the suppression of high @xmath17 contribution to @xmath58 can not be realized by small @xmath34 .
one needs to know large @xmath17 behavior in @xmath0 .
similar studies on higher order cumulants of the net baryon number have been carried out in ref .
@xcite , in which sufficiently large @xmath2 depending on the order of the cumulants is shown to be necessary to obtain a correct value of the cumulants . in this paper , we focus on yang - lee zeros for the baryon chemical potential
. the zeros of the partition function in complex chemical potential plane can be obtained by solving an equation @xmath59 for complex @xmath3 .
owing to the negative powers of the fugacity , the equation is a polynomial one in @xmath34 with order @xmath60 . for the truncated partition function @xmath61
, one needs to solve @xmath62 of which the order of the polynomial is @xmath63 . in the exact case ,
the roots have both real and imaginary part and its distribution in the complex chemical potential plane is expected to form a line , which crosses the real axis at the transition point in the thermodynamic limit .
the behavior of the distribution depends on the nature of the phase transition . in ref .
@xcite , the behavior of the yang - lee zeros around the cp was studied by using a chiral random matrix model .
the singularity associated with the cp appears as a branch point in complex @xmath3 plane and its property is shown to be connected with the universality . in lattice qcd , the phase transition between different @xmath64 sector in the deconfined phase , roberge - weiss phase transition , has been recently analyzed from a view point of yang - lee zeros @xcite . in this work
, we use a chiral random matrix model similar to used in ref . @xcite but with an extension to periodic property in imaginary chemical potential as it is necessary to have integer net baryon number .
the partition function is written as a polynomial in @xmath65 . since a complex root @xmath66
is accompanied with its conjugate @xmath67 and the charge conjugate symmetry implies @xmath68 and @xmath69 are also roots , only the roots located in the first quadrant of the complex @xmath3 plane are independent . in practice , it is convenient to use joukowski transformation @xmath70 and reorganize the series in terms of @xmath71 to reduce the number of roots to search for . using a property of chebychev polynomial @xmath72 , one finds @xmath73 then eq . reduces to a series expression containing only positive powers . after expanding the chebychev polynomial by eq .
, one finds , @xmath74}(-1)^k \frac{(n - k-1)!}{k !
( n-2k)!}\omega^{n-2k}.\label{zg_truncated } \end{split}\ ] ] this formula could be also useful to compare a relativistic system with nonrelativistic ones .
the roots of @xmath71 space is easily converted into those in @xmath34 and @xmath3 plane as @xmath75 taking both signs , one can finds all the roots in the complex @xmath3 and @xmath34 plane .
in this section , we introduce a chiral random matrix model which is an effective model for the spontaneous chiral symmetry breaking in qcd . since this model is analytically solvable in the chiral and thermodynamic limit @xcite and analytic expression for the partition function in finite volume is known @xcite , we find this model as the most suitable one for the present purpose .
an apparent shortcoming of the model for applying to the net baryon number probability distribution is lack of periodicity in imaginary chemical potential , which is a consequence of @xmath76 symmetry .
thus , we first extend the model to exhibit the appropriate periodicity and the phase structure in the imaginary baryon chemical potential . in qcd , the partition function has a periodicity @xmath77 in the imaginary quark chemical potential @xmath78 , thus @xmath79 in the baryon number .
lqcd simulations have shown that there is no phase transition in imaginary baryon chemical potential at temperatures below chiral crossover temperature and thermodynamic quantities smoothly behave as @xmath80 @xcite . this fact combined with eq .
implies that the phase transition at large baryon number density is encoded in higher fourier coefficients of the smoothly oscillating function .
we start with a partition function of the chiral random matrix model with @xmath81 sites given in @xcite @xmath82 where @xmath83 denotes the variance of the random matrix @xmath84 which has @xmath85 dimension and @xmath86 is the @xmath87 matrix approximating the dirac operator . at @xmath88 , @xmath83 is the only dimensionful parameter .
we use it as a unit of mass in the model and put @xmath89 in expressions below .
the dirac operator takes the form @xmath90 the matrix @xmath42 describes the effect of temperature and chemical potential . in ref .
@xcite , it was chosen as @xmath91 for one half of eigenvalues and @xmath92 for the other half , is the chemical potential of the baryon number , thus @xmath93 stands for that of the quark number . ] with @xmath94 and @xmath95 being the dimensionless parameters .
the linear ansatz for the matrix @xmath42 - accounts for the fact that there are the two smallest matsubara frequencies @xmath96 .
this model does not have any thermal distribution which gives the fugacity factor @xmath97 thus nor the periodicity in imaginary chemical potential , since it appears as a result of summation over the matsubara frequencies . in order to make the partition function periodic ,
we perform a following replacement @xmath98 which gives a periodicity @xmath99 in @xmath100 to the partition function . compared to the original linear ansatz
, this replace does not change anything at @xmath101 but alters the phase structure at @xmath102 . the phase structure of the model
is easily evaluated by taking @xmath103 limit .
introducing an auxiliary @xmath104 complex matrix field @xmath105 and performing the gaussian integration with respect to @xmath84 , one obtains the partition function @xcite @xmath106\label{eq : z_saddlepoint}\ ] ] where @xmath107 stands for the effective potential .
then the partition function can be determined by the minimum of the potential , which is evaluated at the saddle point @xmath108 of the integrand : @xmath109 and @xmath110 the saddle point @xmath108 is related to the chiral condensate through @xmath111 where the four dimensional volume @xmath112 corresponds to @xmath81 such that @xmath81 represents the typical number of the instanton ( or anti - instanton ) in @xmath112 . for real @xmath113 ,
one expects @xmath108 is a real matrix proportional to the unit matrix .
therefore , the saddle point can be obtained by solving for the potential @xmath114\right.\nonumber\\ & \qquad \left.\left[(\phi+m)^2- \tilde{t}^2 \left(a\sinh\frac{\mu}{2t}-i\right)^2\right]\right\}\end{aligned}\ ] ] where @xmath115 and @xmath116 in the chiral limit @xmath117 , one finds that @xmath118 at @xmath119 and a second order phase transition occurs at @xmath120 and @xmath101 , where @xmath108 continuously approaches to zero .
thus , @xmath108 can be regarded as an order parameter of the chiral phase transition .
the parameters in the model , @xmath83 , @xmath94 and @xmath95 , are determined as follows .
the only dimensionful parameter @xmath83 is estimated to be @xmath121mev through eq . by putting @xmath122 at @xmath119 . since @xmath123 at @xmath101 , putting @xmath124 mev yields @xmath125 .
the remaining parameter @xmath95 connects the model to the density scale . with the linear ansatz for @xmath42 -
, one finds the first order phase transition at @xmath6 and @xmath126 .
we follow the choice of ref .
@xcite and put @xmath127 , corresponding to the first order transition point at @xmath128mev , though we do not have the same phase diagram as ref .
@xcite owing to the implementation of the periodicity - .
figure [ fig : phasediagram ] shows the phase diagram of the modified random matrix model in the chiral limit and in the presence of a small explicit symmetry breaking , @xmath129 , respectively . in the chiral limit , second order line continues with decreasing temperature down to @xmath130 and @xmath131 , where @xmath132 and @xmath133 is the location of the tricritical point ( tcp ) .
below @xmath134 , there is the first order phase transition line . at finite quark mass ,
the second order line is replaced by smooth crossover and tcp becomes cp with slightly decreased temperature and increased chemical potential , @xmath135 and @xmath136 , respectively . while these structures are the same as those in refs .
@xcite , the apparent singularity at @xmath6 in the periodic parametrization significantly modifies the phase boundary at low temperature .
we stress that our purpose in this paper is to explore the property of partition function zeros rather than determination of the phase structure . of a periodic chiral random matrix model in imaginary chemical potential @xmath137 at @xmath138 . ] with the parameter set for @xmath139 and @xmath83 , we find that this form also gives a reasonable thermodynamic quantities at imaginary chemical potential . and @xmath3 .
while the qualitative phase structure does not depend on the parameters in the linear ansatz , it does so when one employs the periodic parametrization . ]
figure [ fig : sigma - theta ] displays the behavior of the order parameter @xmath108 in the imaginary baryonic chemical potential @xmath137 .
one sees that our parameterization gives the correct periodicity @xmath79 and expected temperature dependence such as larger amplitude at higher temperature below @xmath140 @xcite .
owing to lack of a @xmath141 sector such as the polyakov loop background , this model does not exhibit the roberge - weiss phase transition @xcite at high @xmath8 . at finite @xmath81
, the partition function can be expressed as @xcite @xmath142 where an irrelevant constant factor is ignored and @xmath143 denotes the confluent hypergeometric function .
one may directly obtain zeros of this partition function , but one needs to expand @xmath58 in a series of the fugacity @xmath34 to examine effects of tails in the probability distribution function .
we put the details in the appendix [ app1 ] and write down only the result for the canonical partition function , for @xmath144 , @xmath145^{k_1+k_2-k_3 } \binom{k_3}{\frac{k_3-n}{2 } } & k_1 = k_2\label{eq : zc_expanded } \\
\displaystyle \delta \sum_{k_3=0}^{\delta } \left(-\frac{16}{a^2 } \right)^{k_3 } \frac{(\delta+k_3 - 1)!}{(\delta - k_3 ) ! ( 2k_3 ) ! } \sum_{k_4=0}^{k_1+k_2-k_3 } \binom{k_1+k_2-k_3}{k_4 } \left [ \frac{2(2-a^2)}{a^2 } \right]^{k_1+k_2-k_3-k_4 } \binom{k_4}{\frac{k_4-n}{2 } } & k_1 \neq k_2 \end{cases}\end{aligned}\ ] ] we compute the yang - lee zeros for the truncated partition function with the canonical partition function of the chiral random matrix model . taking @xmath146 in eq .
, one recovers the exact grand partition function .
the computation of the zeros requires a special care in numerical digits as cautioned in literature @xcite .
we perform the calculations in 50 - 300 digits utilizing ` fmlib ` package @xcite in fortran 90 .
plane for @xmath117 at @xmath147 .
open symbols stand for the zeros in different @xmath81 and solid line indicates the stokes boundary .
branch point is denoted by closed circles . ]
figure [ fig : lyz099_omega ] shows the distribution of the yang - lee zero of the periodic chiral random matrix model in the complex @xmath71 plane , for @xmath117 and at @xmath147 .
the distribution of the zeros is symmetric with respect to the horizontal axis because the partition function is an even order polynomial of @xmath71 and a root has its complex conjugate . the solid line in fig .
[ fig : lyz099_omega ] stands for stokes boundary , which can be regarded as an extension of the phase boundary to a complex chemical potential plane . in the thermodynamic limit @xmath148
, it satisfies @xmath149 where the first condition ensures the well - defined partition function at the saddle point of integrand in eq .
and the second condition denotes the continuity of the real part of pressure at the boundary @xcite .
@xmath150 and @xmath151 stand for the two out of five solutions of the gap equation @xmath152 and give the minimum of @xmath153 in both sides of the boundary , respectively .
the density of the zeros increases with @xmath81 and turns into the cut which constitutes the stokes boundary in the thermodynamic limit .
this is clearly seen in fig .
[ fig : lyz099_omega ] . there are two branch points located on the real axis .
since @xmath154 for real @xmath3 , the one at @xmath155 corresponds to the second order phase transition point in real @xmath3 , while the other one , @xmath156 is located on the line @xmath157 .
the stokes boundary exhibits a closed curve , reflecting the periodicity in imaginary @xmath3 and existence of the phase boundary at real @xmath3 axis and @xmath157 .
the phase structure can be more intuitively understood by going to complex @xmath3 plane .
figure [ fig : lyz099_mu]-left displays the distribution of the same yang - lee zeros as in fig . [
fig : lyz099_omega ] , but the zeros in @xmath158 are omitted since their locations are trivial according to the charge conjugate symmetry @xmath159 .
the branching point on the horizontal axis indicates the second order phase transition point .
the stokes boundary extends to both direction in imaginary @xmath3 and ends up at the other branch point .
note that the branch points at @xmath160 and @xmath161 are essentially the same because of the periodicity .
we refer to @xcite for behavior of the order parameter in complex @xmath3 plane and related topics .
the zeros distribute along the boundary and becomes more dense for large @xmath81 , but distance to the real axis is not so close for these values of @xmath81 .
the behavior of the density of the zeros is related to a property of the thermodynamic potential which can be described by an analogy to electrostatics @xcite . in this case
, @xmath153 can be regarded as the electrostatic potential on the @xmath162 plane and the normal component of the electric field @xmath163 to the stokes boundary has a discontinuity of which amount is proportional to the density of the zeros .
we confirmed that in this model these discontinuities at large @xmath164 , where the zeros are dense , are much larger than those at small @xmath3 , following the expectation .
although the density of the zeros far from the branching point is a model - dependent feature dependent on the shape of the stokes boundary , it is governed by the universality near the branch point on the real axis as pointed out in ref .
@xcite .
effects of the finite but small quark mass can be seen in the right panel of fig .
[ fig : lyz099_mu ] where the distribution of yang - lee zeros for @xmath129 at the same temperature is displayed .
owing the explicit chiral symmetry breaking , the phase transition becomes a crossover such that the branch point on the real axis moves to above . as a result
, there are two branch points of which are complex conjugate each other . the same thing occurs also to the branch point at @xmath165 . here
we emphasize that these complex singularities are , albeit unphysical , indicating existence of a chiral phase transition in the chiral limit .
these are also known as yang - lee edge singularities @xcite .
the critical point at finite density ( see fig . [
fig : phasediagram ] ) is realized by coalescence of the branch points close to real @xmath3 axis when temperature is decreased @xcite . as seen in fig .
[ fig : lyz099_mu]-right , the yang - lee zeros are fairly on the boundary line and exhibit expected behaviors .
as described in sec . [ sec : general ] , the connection of net baryon number multiplicity distribution with the reconstructed grand partition function could potentially enables us to extract the yang - lee zeros from experimental data .
since the results presented in the previous section correspond to @xmath52 , i.e. , no information on the exact partition function is lost , we need to evaluate whether one can obtain the correct distribution of the yang - lee zeros when the fugacity expansion is truncated . furthermore , even if one starts from a partition function which does not exhibit any phase transition such as an ideal boltzmann gas , the truncation produces the zeros of partition function because it is a polynomial of order @xmath2 . in this section
we investigate in detail the effects of the truncation on the distribution of the yang - lee zeros .
figures [ fig : lyz_tr1 ] and [ fig : lyz_tr_rm ] display the distribution of the yang - lee zeros from the truncated partition function of the periodic chiral random matrix model for various @xmath2 and @xmath129 . is defined for @xmath117 .
thus it is slightly lower than the chiral crossover temperature for @xmath129 . ]
hereafter we set @xmath166 .
we confirmed the following results does not depend on the specific choice of @xmath81 .
we plot only the first quadrant in complex @xmath3 plane according to the symmetry structure of the distribution . the left panel in fig .
[ fig : lyz_tr1 ] shows the case of @xmath167 , at which transition is of crossover type as seen in the branch point at @xmath168 . for @xmath169 ,
the zeros are located on the stokes boundary ( dashed line ) .
reducing @xmath2 by one , i.e. , removing @xmath170 from the series , one sees a drastic change in the distribution .
the distribution of the zeros at large @xmath164 and @xmath171 splits into the two lines , but the rest of the zeros remains unchanged .
further reduction of @xmath2 substantially modifies the distribution such that the splitting occurs closer to the edge closer to the real @xmath3 axis .
nevertheless , up to @xmath172 , the edge of the distribution which is the closest yang - lee zero to the real @xmath3 axis remains the same . beyond @xmath173 ,
the distribution no longer holds the information on the exact yang - lee zeros thus the apparent relation to the phase boundary is lost . , but for @xmath174 the behavior with respect to changing @xmath2 does not depend on temperature or corresponding phase transition . in the right panel of fig .
[ fig : lyz_tr1 ] , we plot the result of the same analysis for @xmath175 where the branch point appears on the real axis , indicating the critical point . reflecting the location of the branch point ,
the edge of the distribution also become closer to the real axis compared to the crossover case .
the edge is stable against decreasing @xmath2 down to @xmath176 , then it starts to deviate slowly when decreased further .
this is so also in the case of first order phase transition ( @xmath177 ) depicted in fig .
[ fig : lyz_tr_rm ] .
the branch point is hidden in unphysical riemann sheets @xcite and the edge is very close to the real axis .
we also note that there is always a zero at @xmath178 when @xmath2 is odd .
these zeros look special since it corresponds to negative real axis in both complex @xmath34 and @xmath71 plane .
however , this is a mathematical consequence because in this case the truncated partition function is an odd order polynomial , thus it has at least one real root . as seen in figs .
[ fig : lyz_tr1 ] and [ fig : lyz_tr_rm ] , it becomes the edge of one of the lines bifurcating from the exact yang - lee zeros .
these results indicate the stability of the edge does not depend on the detail of the phase structure , although the location of the edge seems to be connected with the shape of the stokes boundary which is model dependent through the @xmath3 dependence of the partition function .
in particular , the present results are obtained by employing the periodicity in the random matrix model which does not correctly take into account degrees of freedom with baryon charges @xcite .
we note that this modification causes unphysical behavior in thermodynamics , such as negative @xmath0 at some small @xmath17 at low @xmath8 , which presumably reflect the unusual curvature of the phase boundary in fig .
[ fig : phasediagram ] .
therefore , we note that the shape of the distribution itself might not be relevant for realistic situations .
nevertheless , below we shall see that the stability of the edge is specific to the case with a phase transition .
figure [ fig : lattice ] displays distribution of the yang - lee zeros above @xmath140 calculated in lattice qcd simulation via the canonical method @xcite . while calculations in the confinement phase is still numerically difficult thus we do not see clear indications of a phase transition at low @xmath8 , the roberge - weiss ( rw ) transition @xcite provides us a well - defined phase transition in high temperature quark - gluon plasma phase , though at imaginary chemical potential . in this figure ,
the data are calculated on @xmath179 lattices and @xmath180 which corresponds to @xmath181 . a more detailed analysis in lattice qcd with different lattice setups
can be found in ref .
@xcite . since quark mass is heavy , the calculation is not relevant for chiral phase transition .
the rw transition is regarded as a transition from one @xmath141 sector to another one when single quarks can be excited owing to deconfinement and is known to exhibit a first order phase transition at @xmath182 . in terms of baryon chemical potential , the transition lines reduce to @xmath165 which is shown as a dotted line in fig .
[ fig : lattice ] .
a brief explanation of the roberge - weiss phase boundary can be found in appendix [ sec : app_b ] .
since it is hard to compute @xmath183 near @xmath184 , the canonical approach in lattice qcd lacks large @xmath17 contribution when one constructs the truncated partition function .
one sees that in fig .
[ fig : lattice ] the behavior of the distribution of the yang - lee zeros against changing @xmath2 is similar to that of the random matrix model , despite the completely different origin of the phase transition .
therefore , we expect the similar splitting behavior of the distribution also appearing in refs .
@xcite is also due to the truncation effect . indeed ,
@xmath81 and @xmath2 dependence of the yang - lee zero shown in ref .
@xcite agrees with the truncation effects discussed here .
we expect that the bifurcation of the zero starts at large @xmath164 by improving the fugacity expansion , but one needs to take @xmath185 to completely produce the yang - lee zero along the transition line . in the rw transition
where the transition point at @xmath186 , the edge of the distribution is the closest zero to the imaginary axis .
one sees that this point is also stable against changing @xmath2 .
this fact suggests that the stability of the edge is not specific to the random matrix model or chiral phase transition but might be a general property of the distribution when partition function is truncated .
finally we examine a model without phase transition in order to check whether the stability of the edge is specific to phase transition or not .
we employ the skellam distribution @xcite of which probability distribution of the net baryon number @xmath17 is given by @xmath187 where @xmath188 and @xmath189 denote the thermal averages of the numbers of baryons and anti - baryons , respectively .
the mean @xmath26 and variance @xmath190 of the distribution are given by @xmath191 and @xmath192 , respectively .
for @xmath193 , the distribution becomes symmetric and the argument of the modified bessel function @xmath194 is reduced to @xmath195 .
this distribution can be derived from non - interacting boltzmann gas @xcite , thus the canonical and grand canonical partition functions read @xmath196\label{eq : skellam_zg}\end{aligned}\ ] ] where the temperature and volume dependence is encoded in @xmath190 .
obviously the grand partition function does not have any roots thus no phase transition exists .
when one constructs the truncated grand partition function from the canonical partition function , however , there exist complex roots .
consequently , one might see these spurious zeros even if the system does not have any phase transition , when one constructs the partition function through the fugacity expansion . here
we investigate such spurious zeros from the skellam partition function such that it has the same variance with the random matrix model at @xmath166 , @xmath167 and @xmath129 of which distribution of the yang - lee zeros is displayed in fig .
[ fig : lyz_tr1 ] .
since the information on the phase transition is encoded in the tail of the probability distribution @xmath20 , the skellam distribution with the same variance serves a useful reference distribution @xcite .
the probability distribution of the random matrix model and corresponding skellam distribution are shown in fig .
[ fig : skellam]-left .
both distributions almost agree for small @xmath17 , according to the same @xmath190 , but the deviation appears in the tail of the distribution with tiny probability .
figure [ fig : skellam]-right shows the distribution of zeros of the truncated partition function for the skellam distribution with @xmath197 . except for a splitting of the distribution for @xmath198 which is similar to those in the random matrix model , the distributions consist almost parallel lines moving to large real @xmath3 direction as @xmath199 increases .
this behavior reflects the fact that all the zeros go away to infinity as @xmath200 since the exact grand partition function does not have roots .
remarkably , the edges of the distributions also move together with the rest of zeros , in contrast to the random matrix model and lattice qcd . furthermore
, one notes that the distributions for @xmath201 in the random matrix model , shown in fig .
[ fig : lyz_tr1 ] , resemble those from the skellam distribution .
this observation indicates that the stability of the edge against @xmath2 is a consequence of the existence of phase transition and information on the phase transition is lost for too small @xmath2 .
in the previous section , we have shown that the edge of the distribution of the yang - lee zeros remains unchanged when the tail part of the canonical partition function is missing . in practice
, this property gives implications for necessary statistics in heavy ion studies of the net - baryon number fluctuations and in lattice qcd calculations .
since the sufficient @xmath2 to see the stable edge depends on the system volume , here we compare it with corresponding @xmath202 for @xmath203-th order cumulants .
here we consider only even order ones for net - baryon number at @xmath101 , since we are looking at @xmath0 rather than @xmath20 which becomes asymmetric with respect to @xmath17 at @xmath204 .
thus the first central moment @xmath205 .
the second , fourth and sixth order cumulants @xmath206 @xmath207 read @xmath208 the property of the higher order cumulants of net - baryon number probability distribution for changing @xmath2 was studied in ref .
@xcite by using a chiral quark - meson model . for sufficiently large volume
, it was shown that @xmath202 for the cumulants approximately scale with @xmath209 . .@xmath2
necessary for reconstructing @xmath203-th order cumulants and edge of the yang - lee zeros from @xmath183 in the random matrix model at @xmath167 and @xmath129 . [ cols="^,^,^,^,^",options="header " , ] we summarize the values of each @xmath2 in table [ tbl : nmax_vs_cumulants ] .
the calculations are done for @xmath167 and @xmath129 in the random matrix model .
owing to the narrow @xmath183 , even the sixth order cumulant for @xmath210 only requires @xmath211 , i.e , @xmath212 to reconstruct it from @xmath0 , while the edge of yang - lee zeros demands @xmath213 .
the small @xmath214 implies the system volume is not large enough to exhibit @xmath209 scaling regime of the cumulants .
this can be understood from the small value of @xmath190 in the random matrix model calculations .
the rapid decay of @xmath20 give a rather weak dependence of @xmath2 for the higher order cumulants . for a sufficiently large volume
, one expects that @xmath20 resembles gaussian near the peak , while the probability distribution ( fig .
[ fig : skellam ] ) has a sharp peak .
thus we can not assess the value of @xmath2 needed in a realistic situation relevant for heavy ion collisions .
moreover , the baryon number carried in this model is not a physical one , as mentioned above .
all we can say is that one may need much more statistic than higher order cumulants . .
] in the fluctuation measurements at rhic , observed @xmath20 can be well described by the skellam distribution and deviation from the skellam distribution exists in the tail , resulting in higher order cumulants different from the skellam case .
the obtained variance reaches @xmath215 @xcite at the most central bin .
thus it is instructive to give a reference for the distribution of the spurious yang - lee zeros based on the skellam distribution .
here we pick up the data for @xmath7 gev at the most central bin , which gives @xmath216 and @xmath217 with available bin from @xmath218 to @xmath219 . and
@xmath219 have only 1 event.]as mentioned in sec .
[ sec : general ] , one can construct @xmath183 from @xmath220 according to charge conjugation symmetry @xcite . in the skellam distribution for @xmath221 , we find that @xmath222 and 26 for second , fourth and sixth order cumulant , respectively .
note that these @xmath202 apply to the cumulants at @xmath101 .
the data does not have enough statistic for the sixth order cumulant at freeze - out @xmath3 .
we plot the distribution of the yang - lee zeros for the constructed skellam distribution with @xmath221 .
the basic feature is the same as the small @xmath190 case ( fig .
[ fig : skellam ] ) .
the line of zeros moves toward infinity as @xmath2 increases .
one sees that some zeros below @xmath223 appear on the imaginary @xmath3 axis , which corresponding to the unit circle in complex fugacity plane .
this means that , for a large volume case , the zeros can appear on the imaginary axis when the tail of the @xmath183 is not provided . in the skellam distribution
, these zeros can be obtained directly by looking at the truncated partition function on the imaginary @xmath3 axis , for @xmath43 , @xmath224 which converges into eq . with oscillations giving zeros on imaginary @xmath3 .
in this paper , we present analyses on partition function zeros which can be obtained from a truncated series of the fugacity expansion . by solving an extended chiral random matrix model which has a periodicity in the imaginary chemical potential ,
we compare the exact location of the yang - lee zeros and those obtained from the truncated series .
we found that the edge of the distribution of the zeros is insensitive to the truncation of higher order terms in the fugacity expansion to some degree .
we found the similar behavior in lattice qcd at high temperature in the context of the roberge - weiss phase transition .
this observation indicates that those higher order terms may have limited influences in search for the location of the phase boundary in lattice qcd calculations and heavy ion experiments .
although the distribution of zeros exist in systems without phase transition , due to the truncation , the zeros closest to the real @xmath3 axis are stable against truncation if the system has a phase transition or crossover .
the spurious zeros in the skellam distribution moves toward infinity against the truncation .
therefore , one can distinguish whether the distribution is related to the phase transition or not by looking at the stability of the edge of the distribution against the truncation .
although the information on the stokes boundary is lost in the case of too small @xmath2 , we expect that it does not mean that all the relevant information on the phase transition gets lost in the truncated partition function .
this expectation follows from the fact the sixth and higher order cumulants at @xmath101 should be influenced by the phase transition and the truncated series is still able to reproduce them
. it would be interesting to see how the distributions of the zeros in the small @xmath2 cases in figs .
[ fig : lyz_tr1]-[fig : lyz_tr_rm ] are related to the remnant of the phase transition .
the order of the truncation in the fugacity series to obtain the stable edge of the yang - lee zeros , @xmath2 , is nevertheless found to be much larger than those for higher order cumulants .
we can not make a quantitative assessment on realistic values for heavy ion experiments due to the lack of connection in the model to the real world .
we hope that such an estimate becomes feasible in the near future .
the authors would like to thank x. luo and n. xu for providing numerical data of the star collaboration .
they would like to gratefully thank b. friman and k. redlich for fruitful discussion and continuous encouragement .
they acknowledge stimulating discussions with ph .
de forcrand , f. karsch , j. knoll , v. koch , k. nagata and j. wambach .
this work was supported by the grants - in - aid for scientific research on innovative areas from mext ( no . 24105008 ) , by the grants - in - aid for scientific research from jsps ( no .
15h03663 , no.26610072 ) , hic for fair , and the polish science foundation ( ncn ) , under maestro grant 2013/10/a / st2/00106 .
in the following , we derive an analytic expression for @xmath225 from eq . .
first we rewrite the @xmath3 dependent part in terms of the fugacity @xmath65 . since @xmath226 \\ & = \left ( a^2 \sinh^2 \frac{\mu}{2 t } + 1 \right)^{k_1+k_2}\cos[2(k_1-k_2)\phi]\end{aligned}\ ] ] where @xmath227 and imaginary part vanishes after summation over @xmath228 and @xmath229 , using the chebychev polynomial @xmath230\label{eq : chebychev}\ ] ] and @xmath231 we have the partition function @xmath58 as @xmath232 expanding the chebychev polynomial by the following expression @xmath233 and using binomial expansion in the third line of , we can express @xmath58 in terms of @xmath234 .
we obtain eq . by applying the projection .
note that maximum power of @xmath34 is given by @xmath81 .
in this appendix , we give a brief explanation of the cut arising from the fermi distribution function and apply it to the roberge - weiss transition in qcd .
the thermodynamic potential of the free fermi gas is given by @xmath235 + ( \mu \rightarrow -\mu)\ ] ] where @xmath236 .
when the chemical potential @xmath3 has an imaginary part , @xmath237 , the imagary part gives a phase in front of the boltzmann factor : @xmath238 where @xmath239 .
therefore , for @xmath240 , the phase gives @xmath241 and the thermodynamic potential has a logarithmic cut at @xmath240 and @xmath242 .
the anti - particle term also gives the cut symmetric with respect to the imaginary axis . in ref .
@xcite , it is pointed out that the branch point singularity limits the convergence radius when one tries to analytically continue the results in the imaginary chemical potential to the real one .
since this cut originates from the fermi distribution , the same analytic structure appears in chiral models with fermions @xcite . in qcd at high temperature , quarks are deconfined and have a light mass owing to chiral restoration . since the deconfinement can be expressed as a breaking of @xmath64 symmetry , it is useful to resort to chiral effective models with the polyakov loop background @xcite which successfully describe the roberge - weiss transition @xcite
. then , the relevant leading single quark contribution to the thermodynamic potential reads @xmath243
\\ + ( \mu_q \rightarrow -\mu_q , \ , \phi \rightarrow \bar{\phi } ) \end{split}\ ] ] where @xmath244 is the quark chemical potential and @xmath245 is the the expectation value of the polyakov loop . for antiquark contribution , the conjugate @xmath246 couples to the thermal distribution .
at the imaginary chemical potential , polyakov loop @xmath245 acquires a complex phase @xmath247 .
one may express @xmath248 .
then the thermodynamic contribution becomes @xmath249 the phase of the polyakov loop @xmath247 varies as a function of the imaginary quark chemical potential @xmath250 .
the roberge - weiss transition at @xmath251 can be understood as a transition from one @xmath141 sector with @xmath252 to another one @xmath253 @xcite .
then , the coupling between @xmath247 and @xmath250 gives the prefactor @xmath241 in front of the boltzmann factor .
moreover , @xmath254 in the deconfined phase and the prefactor 3 allow this function to have the singularity at @xmath255 .
this feature gives the cut drawn as rw transition line in fig .
[ fig : lattice ] . a derivation based on the gaussian @xmath20 can be found in ref .
@xcite . in the confinement phase where @xmath256
, this term is suppressed and the thermal cut from the quark does not appear .
77ifxundefined [ 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 * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) in @noop _ _ , ( , ) p. @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) , @noop * * , ( ) @noop `` , '' @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop `` , '' ( ) , @noop `` , '' ( ) , @noop `` , '' ( ) , @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop `` , '' ( ) , @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop _ _ ( , ) @noop * * , ( ) @noop * * , ( ) , @noop * * , ( ) @noop * * , ( ) @noop @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) , | we investigate yang - lee zeros of grand partition functions as truncated fugacity polynomials of which coefficients are given by the canonical partition functions @xmath0 up to @xmath1 .
such a partition function can be inevitably obtained from the net - baryon number multiplicity distribution in relativistic heavy ion collisions , where the number of the event beyond @xmath2 has insufficient statistics , as well as canonical approaches in lattice qcd .
we use a chiral random matrix model as a solvable model for chiral phase transition in qcd and show that the closest edge of the distribution to real chemical potential axis is stable against cutting the tail of the multiplicity distribution . the similar behavior is also found in lattice qcd at finite temperature for roberge - weiss transition .
in contrast , such a stability is found to be absent in the skellam distribution which does not have phase transition .
we compare the number of @xmath2 to obtain the stable yang - lee zeros with those of critical higher order cumulants . |
pulsating da ( h - rich atmospheres ) white dwarfs , also called zz ceti or dav stars , are the most numerous class of degenerate pulsators , with over 148 members known today . they are characterized by multiperiodic brightness variations caused by spheroidal , non - radial @xmath1-modes of low degree with periods between 70 and 1500 s ( winget & kepler , althaus et al .
g117@xmath0b15a is the most well - studied star of this class of variables .
the rate of change of its 215.2 s period is very small : @xmath2 s / s ( kepler 2009 ) , with a stability comparable to that of the most stable millisecond pulsars .
the _ axion _ is a hypothetical particle that appears as a consequence of the symmetry postulated by peccei & quinn ( 1977 ) to solve the strong cp ( charge - parity ) problem in quantum chromodynamics .
axions are considered as candidates for dark matter of the universe , and their contribution depends on their mass ( raffelt 2007 ) .
interestingly enough , axion emission is supposed to take place in the cores of white dwarfs ( isern et al .
1992 , 2008 ) .
since axions can freely escape from such stars , their existence would increase the cooling rate and , consequently , the rate of change of the periods as compared with the standard one . in this work we present a new asteroseismological model for g117@xmath0b15a and
use the more recent measurement of the rate of change of the 215.2 s period to impose new constraints on the mass of the axion .
[ cols="<,^,^,^ " , ] [ table1 ] we performed a detailed asteroseismological study of the dav star g117@xmath0b15a using a grid of evolutionary models characterized by consistent chemical profiles for both the core and envelope , and covering a wide range of stellar masses , thicknesses of the hydrogen envelope and effective temperatures .
this constitutes the first asteroseismological application of the da white - dwarf models presented in althaus et al .
( 2010b ) .
these models were generated with the lpcode evolutionary code from the zams through the thermally - pulsing and mass - loss phases on the agb and finally to the domain of planetary nebulae and white dwarfs .
the effective temperature , the stellar mass and the mass of the h envelope of our da white dwarf models vary in the ranges : @xmath3 k , @xmath4 , @xmath5 , where the ranges of the values of @xmath6 are dependent on @xmath7 . for simplicity , the mass of he has been kept fixed at the value predicted by the evolutionary computations for each sequence .
we searched for a pulsation model that best matches the pulsation periods of g117@xmath0b15a . to this end , we sought the model that minimizes a quality function defined simply as the average of the absolute differences between theoretical and observed periods : @xmath8 , where @xmath9 is the number of the periods 215.20 s , 270.46 s and 304.05 s ( kepler et al . 1982 ) observed in g117@xmath0b15a .
the theoretical periods were assessed by means of the pulsation code described in crsico & althaus ( 2006 ) .
we found a best - fit model with the characteristics shown in table [ table1 ] .
the internal chemical stratification and a propagation diagram ( i.e. , the spatial run of the logarithm of the squared brunt - visl and lamb frequencies ) of this model are shown in fig .
[ figure1 ] .
each chemical transition region produces clear and distinctive features in @xmath10 , which are eventually responsible for the mode - trapping properties of the model . in the core region , there are several peaks at @xmath11 ( where @xmath12 ) resulting from steep variations in the inner c - o profile which are caused by the occurrence of extra mixing episodes beyond the fully convective core during central helium burning .
the extended bump in @xmath13 at @xmath14 is caused by the chemical transition of he , c and o resulting from nuclear processing in prior agb and thermally - pulsing agb stages .
finally , there is the he / h transition region at @xmath15 , which is smoothly shaped by the action of time - dependent element diffusion .
in order to compute the effects of axion emission , we adopted the axion emission rates of nakagawa et al .
axion emission produces a supplementary energy loss rate to those normally considered in the standard theory of white dwarf evolution . as a result
, axions accelerate the cooling process of white dwarfs .
such an acceleration of the evolution has a direct consequence on the pulsational properties of the star , because it produces a larger value of the period derivative of the oscillation modes , @xmath16 ( isern et al .
1992 ) . due to this extra cooling mechanism ,
the structure of the white dwarf itself is also affected but , fortunately , in such a way that for a fixed @xmath17 the pulsation periods @xmath18 are largely independent of the exact value of the axion energy losses ( crsico et al .
2001 ) .
we have computed the values of @xmath19 for the modes with periods @xmath20 s , @xmath21 s and @xmath22 s ( @xmath23 ) of our asteroseismological model by adopting different axion emissions .
as expected , the pulsation periods themselves do not vary appreciably , thus validating our asteroseismological model even with axion emission , but the rates of period change monotonically increase for increasing axion emission . in fig .
[ figure2 ] we display the theoretical value of @xmath19 corresponding to the period @xmath24 s. also shown is the most recent determination of the rate of period change for the period at @xmath25 of g117@xmath0b15a , according to kepler ( 2009 ) .
note that , if we consider a standard deviation from the observational value , we conclude that the observations are compatible with an axion mass lower than @xmath26 mev .
this value is about 5 times larger than that found by crsico et al .
( 2001 ) , but in good agreement with the range of values obtained by bischoff - kim et al .
( 2008 ) , which derived axion masses between 12 and 26.5 mev .
on the basis of a new asteroseismological model for g117@xmath0b15a , the archetype of dav stars , and making use of the most recent determination of the rate of period change for the dominant mode of this star , we have derived a new upper limit of the mass of the ( up to now ) elusive particle called _
we emphasize , however , that the derived upper limit for the axion mass ( @xmath26 mev ) is preliminary , since we have not quantified yet the uncertainties of white dwarf modeling ( core overshooting , rate of the reaction @xmath27c@xmath28o ) and the uncertainties related to the asteroseismological approach .
the main conclusion of this preliminary study is that , if we assume that our asteroseismological model is a good representation of g117@xmath0b15a , in order to explain the high observed value of @xmath29 it is necessary to invoke some extra cooling mechanism , being the axion emission a very plausible one .
althaus , l. g. , crsico , a. h. , isern , j. , & garc a berro , e. 2010a , a&ar , 18 , 471 althaus , l. g. , crsico , a. h. , bischoff - kim , a. , et al .
2010b , apj , 717 , 897 bergeron , p. , fontaine , g. , billres , m. , boudreault , s. , & green , e. m. 2004 , apj , 600 , 404 bischoff - kim , a. , montgomery .
m. h. , winget , d. e. 2008 , apj , 675 , 1512 crsico , a. h. , & althaus , l. g. 2006 , a&a , 454 , 863 crsico , a. h. , benvenuto , o. g. , althaus , et al .
2001 , new astr . , 6 , 197 isern , j. , garca berro , e. , torres , s. , & cataln , s. 2008 , apj , 682 , l109 isern , j. , hernanz , m. , garca berro , e. , 1992 , apj , 392 , l23 kepler , s. o. , 2009 , in _
`` new advances in helio an asteroseismology '' _ , iau joint discussion 11 , xxvii iau general assembly ( rio de janeiro ) kepler , s. o. , nather , r. e. , mcgraw , j. t. , robinson , e. l. 1982 , apj , 254 , 676 koester , d. , & holberg , j. b. 2001 , _
12th european workshop on white dwarfs _
, 226 , 299 peccei , r. d. , quinn , h. r. , 1977 , phys .
lett . , 38 , 1440 raffelt , g. g. 2007 , j. phys .
a , 40 , 6607 nakagawa , m. , kohyama , y. , itoh , n. , 1988 , apj , 326 , 241 winget , d. e. , & kepler , s. o. 2008 , ara&a , 46 , 157 | we perform an asteroseismological study on the dav star g117@xmath0b15a on the basis of a modern set of fully evolutionary da white dwarf models that have consistent chemical profiles at the core and the envelope .
we found an asteroseismological model for g117@xmath0b15a that closely reproduces its observed pulsation periods .
then , we use the most recently measured value of the rate of period change for the dominant mode of this pulsating star to impose a preliminary upper limit to the mass of the axion . |
in general relativity the presence of a rotation ( intrinsic angular momentum , or spin ) of the central body influences motion of a particle orbiting it . due to this reason the orbits of test particles differ in the schwarzschild and kerr backgrounds .
when , in turn , a test particle has spin as well , it will also influence the particle s orbit .
in particular , the motion of a spinning particle will differ from the non - spinning one even in the schwarzschild background .
let us consider a massive non - spinning test particle orbiting a central black hole ( bh ) .
there exists a minimal radius at which stable circular motion is still possible , it defines the so - called innermost stable circular orbit ( isco ) in given background . for the schwarzschild background
the radius of isco equals to @xmath2 , the schwarzschild radius @xmath3 , and other physical quantities which will be introduced further have the following dimensionalities : @xmath4=[m]$ ] , @xmath5=[m]$ ] , @xmath6=1 $ ] , @xmath7=[m]$ ] , @xmath8=[m]$ ] . ]
@xcite . in the kerr space - time circular motion
is possible only in the equatorial plane of bh and the radius of isco depends on the direction of motion of the particle in comparison with the direction of bh rotation , whether they co - rotate or counter - rotate .
co - rotation and counter - rotation cases correspond to parallel and antiparallel orientation of vectors of the orbital angular momentum of the particle and the bh angular momentum . for the case of the extreme kerr background
the difference between these two variants is quite considerable : we have @xmath9 for the antiparallel and @xmath10 for the parallel orientation @xcite .
values of the isco parameters ( radius , total angular momentum , energy , orbital angular frequency ) are determined by the kerr parameter @xmath0 .
the main subject of the present paper is to investigate how the isco parameters ( at a given value of @xmath0 ) are changed if a test particle has spin ( fig .
[ figure - im1 ] ) .
we consider motion of a spinning test particles in the schwarzschild and kerr metrics in the equatorial plane and analytically derive the corrections for the parameters of isco taking spin to be a small parameter .
we restrict our consideration by the case when vectors of black hole spin , particle spin and orbital angular momentum are collinear .
for non - spinning particles moving in the schwarzschild metric the isco parameters were found by kaplan @xcite .
the solution of einstein field equations around a rotating black hole was found by kerr @xcite . in the following works of carter @xcite and de felice @xcite the geodesic motion of general type
was studied for the kerr and kerr - newman ( charged bh ) geometry .
see also wilkins @xcite for bound orbits in kerr metric and dymnikova @xcite for review .
the parameters of isco in kerr space - time for a non - spinning particle were obtained in the works by ruffini & wheeler @xcite and bardeen @xcite .
the problem of the motion of a classical spinning test body in general relativity was considered in papers of mathisson @xcite , papapetrou @xcite and dixon @xcite , using different techniques .
the equations of motion of a spinning test particle in a given gravitational field were derived in different forms ; they are now referred to as mathisson - papapetrou - dixon equations . from these equations
it follows that the motion of the centre of mass and the particle rotation are connected to each other , and when the particle has spin the orbits will differ from geodesics of a spinless massive particle . in this paper
we use the equations of motion in the form derived in dixon s papers .
influence of spin on orbits in the schwarzschild metric was investigated in the paper of corinaldesi and papapetrou @xcite , and in the paper of micoulaut @xcite .
a motion of a spinning test particle in kerr metric was considered by rasband @xcite . using integrals of motion arising from the symmetries in the kerr space - time
rasband @xcite derived the equations of motion of the radial coordinate of a spinning particle for the motion in the equatorial plane of a bh .
also in that work the spin - induced corrections for the radii of last stable orbits for schwarzschild ( @xmath11 ) and extreme kerr ( @xmath12 ) metric were illustrated through numerical calculations . in the paper by tod , de felice & calvani @xcite
the influence of test body s spin on the radii of isco for different values of kerr parameter @xmath0 including naked singularities was numerically calculated , see also papers of abramowicz & calvani @xcite and calvani @xcite .
more general case of kerr - newman metric was analysed by hojman & hojman @xcite .
subsequently the number of works on this subject were published @xcite .
the detailed derivation of the equations of motion using the integrals of motion was presented in the work of suzuki & maeda @xcite for the schwarzschild case and in the work of saijo _ et al . _
@xcite for the kerr background .
_ @xcite suggested that the radius of the last stable orbit is independent of the particle s spin in the extreme kerr background ( for co - rotating orbits ) .
many authors used an effective radial potential for investigation of spinning particle orbits @xcite , @xcite , @xcite , @xcite , @xcite , @xcite .
method of calculation of isco parameters of spinning particle moving in kerr metric is presented in details in the paper @xcite .
the equations for the circular orbits in the equatorial plane resulting from the mathisson - papapetrou - dixon equations appear to be irresolvable analytically , and isco parameters for arbitrary value of @xmath0 are supposed to be found numerically .
linear corrections in spin for the isco parameters in schwarzschild metric were found by favata @xcite . in this work
we analytically obtain the small spin corrections for the isco parameters for the kerr metric at arbitrary value of @xmath0 . for the extreme @xmath12 and almost extreme @xmath13 kerr bh we succeed to find the exact analytical solution for the isco parameters for arbitrary spin , with only restrictions connected with applicability of mathisson - papapetrou - dixon equations .
it has been shown that the limiting values of isco radius and frequency for @xmath12 do not depend on the particle s spin while values of energy and total angular momentum do depend on it .
the present paper is organised as follows . in section [ section - spinlessep ]
we describe a motion of a test body without spin and introduce the notion of the effective potential ( ep ) . in section [ section - iscospinless ]
we describe how to find isco parameters with the help of eps , in the schwarzschild and kerr metric . in section [ section - mpd ]
we present the basic equations for motion of a spinning test body . in section [ section - epspin ]
we introduce eps for spinning particles and write explicitly equations for isco parameters for arbitrary kerr parameter and arbitrary spin . in section [ section - arbitrary ]
we derive formulae for calculation of the small - spin linear corrections to isco parameters for arbitrary value of @xmath0 . in sections
[ section - schw ] , [ section - slowly ] , [ section - extreme - kerr ] we consider in details the cases of schwarzschild metric , kerr metric with @xmath14 and extreme kerr metric . in section [ section - exact ]
we present an exact solution for extreme kerr bh in the case of co - rotation . in section [ section - circular ]
we consider circular orbits of arbitrary radius .
section [ section - conclusions ] is conclusions .
in the problems of motion of a test particle in a central field it is convenient to express the equations of motion for the radial coordinate in terms of a so - called `` effective potential '' energy . in the newtonian dynamics these equations for a particle moving in the gravitational field of a central body with mass @xmath10
can be written as ( see , for example , @xcite ) @xmath15 where @xmath16 is the total non - relativistic energy per unit mass , @xmath17 is the so - called effective potential ( ep ) @xmath18 here @xmath19 is the angular momentum per unit mass , @xmath20 is the gravitational constant . in the framework of general relativity ,
the schwarzschild space - time is given by the expression for the interval ( in spherical schwarzschild coordinates @xmath21 ) as : @xmath22 in this space - time the equation of motion for the radial coordinate is @xcite @xmath23 with the connection between @xmath24 and @xmath25 as @xmath26 here @xmath27 is the total energy at infinity per unit particle rest mass , @xmath19 is the angular momentum per unit particle rest mass , @xmath28 is the proper time . this equation can also be presented in terms of an effective potential @xmath29 here we identify with the effective potential the following expression : @xmath30 comparing equations ( [ e ] ) and ( [ b ] ) we should keep in mind that non - relativistic total energy @xmath16 is not an analogue of @xmath27 , since unlike the latter , it does not contain rest mass energy .
newtonian non - relativistic energy and effective potential are obtained as @xmath31 and @xmath32 , see @xcite .
let us consider the effective potential in the schwarzschild space - time ( [ c ] ) as a function of radial coordinate .
then , the angular momentum @xmath19 plays a role of a parameter defining the shape of @xmath33 curve . in the case
@xmath34 the effective potential has two extrema : maximum and minimum , at the radii of which unstable and stable circular motion are possible correspondingly . in the case
when @xmath19 equals to the boundary value @xmath35 , two extrema of ep merge into one inflection point .
this boundary value of @xmath19 defines parameters of the last stable orbit which is also called the innermost stable circular orbit ( isco ) , i.e. the boundary orbit on which the finite motion is still possible .
further , when the angular momentum @xmath19 of a test body is less than @xmath36 , the ep does not have an extremum .
for these values of angular momentum neither type of finite motion is possible and a test body will inevitably fall in the black hole whatever values of @xmath27 it may have . in boyer - lindquist coordinates the kerr metric is given by the expression @xcite @xmath37 where @xmath0 is the specific angular momentum of a black hole , @xmath38 , @xmath39 .
the equation of motion for the radial coordinate of a test particle in the equatorial plane ( @xmath40 ) of kerr metric is written as @xcite @xmath41 , \end{split } \label{f}\ ] ] with the connection between @xmath24 and @xmath25 @xmath42 here @xmath43 is a @xmath44-component of the angular momentum per unit rest mass , equal to the total angular momentum per unit mass , when @xmath44-axis is parallel to the axis of the rotation of bh .
so @xmath45 and @xmath46 corresponds to the co - rotation and the counter - rotation cases respectively , and @xmath0 is always positive .
we see at the right - hand side of the equation ( [ f ] ) that the expression in brackets is dependent on the energy @xmath27 . for this reason the effective potential in the kerr space - time
can not be defined in such a simple way as for the schwarzschild space - time .
let us write the equation for the radial motion as @xmath47 it is convenient to write this equation as @xcite @xmath48 where @xmath49 , @xmath50 and @xmath51 are defined as : @xmath52 l = 2 m r a l \ , , \\
& \gamma = a^2 l^2 - \delta ( r^2 + l^2 ) \ , . \end{split}\ ] ] let us rewrite ( [ f1 ] ) as @xmath53 and define ep equal to @xmath54 the positive square root must be taken here , see @xcite .
this definition keeps analogy with the ep in the schwarzschild space - time ( and , therefore , with the newtonian dynamics as well ) . in some other works , however , @xcite the entire right - hand side of the equation ( [ f ] )
is used instead of the effective potential @xmath55 , especially for the investigation of circular orbits and the isco parameters . despite
that it contains energy , some authors @xcite keep calling it an effective potential , because it has some features of the ep important for the investigation of circular orbits ( see next section ) .
let us describe how to find isco parameters and first consider the schwarzschild metric . for a circular motion
we need two conditions to be satisfied simultaneously : \(i ) the radial velocity should be equal to zero : @xmath56 what corresponds to the equality @xmath57
\(ii ) the acceleration of the radial coordinate should be absent : @xmath58 differentiating ( [ b1 ] ) with respect to @xmath28 and dividing by @xmath59 , we obtain @xmath60 \ , , \label{accel}\ ] ] so the condition ( [ circ-2 ] ) corresponds to the relation @xmath61 = 0 \ , .\ ] ] for a given @xmath27 we have @xmath62 solving the system of equations ( [ circ - schw-1 ] ) and ( [ circ - schw-2 ] ) , we obtain the expressions for the radius of the circular orbit @xmath63 and for the particle energy @xmath27 on it @xcite , as a function of @xmath19 in the form @xmath64 , \label{ff}\ ] ] @xmath65 in eq .
( [ ff ] ) the upper sign corresponds to the stable circular orbits in the minimum of ep and the lower one stands for the unstable circular orbits in its maximum . in eq .
( [ ff2 ] ) radius @xmath63 given by ( [ ff ] ) should be substituted . from the expression ( [ ff ] ) we see that for @xmath66 the radius becomes complex and we do not have a circular orbit at all .
mathematically this means that the ep with such values of @xmath19 does not have extrema . in order to find the last stable orbit ( isco ) we need to find the last value of @xmath19 at which the ep still has extremum .
the isco takes place when points of maximum and minimum of ep merge .
therefore we need the third condition : \(iii ) the inflection point of the ep satisfies relation : @xmath67 so , in order to find the parameters of isco ( @xmath63 , @xmath27 , @xmath19 ) , we need to solve three equations ( [ circ - schw-1 ] ) , ( [ circ - schw-2 ] ) , ( [ circ - schw-3 ] ) simultaneously . from ( [ ff ] ) it
is evident that the points of maximum and minimum of ep merge at the marginal angular momentum @xmath68 .
thus , the isco parameters in the schwarzschild space - time are @xcite : @xmath69 in the kerr space - time the finding of isco parameters is more complicated .
isco in the kerr space - time were considered in @xcite , @xcite , where , in particular , parameters of isco for the extreme kerr bh were found .
this problem is described at length , for example , in the textbook by hobson _
_ @xcite .
it is convenient to solve the problem by introducing a function @xmath70 the function @xmath71 is the right - hand side of eq .
( [ f1 ] ) and has qualities which are important in our research . the expression for acceleration is given by the first derivative @xmath72 ( compare with ( [ accel ] ) ) , and stability of the circular orbit is given by the sign of its second derivative @xmath73 .
it is convenient to use variable @xmath74 instead of @xmath63 and variable @xmath75 instead of @xmath19 . for the circular motion
we need the conditions for the velocity ( [ circ-1 ] ) and the acceleration ( [ circ-2 ] ) to be satisfied simultaneously .
therefore for the kerr metric we have from ( [ v - spinless ] ) the system of equations @xmath76 the usage of @xmath77 instead of @xmath63 does not change anything in the form of the system : since @xmath78 , the condition @xmath79 is equivalent to @xmath80 .
the solution of the system ( [ system - r ] ) defines the set of parameters @xmath75 and @xmath27 for stable and unstable circular orbits as functions of @xmath77 @xcite . for the stable circular orbit
we have @xmath81^{1/2}},\\ e(u ) & = \frac{1 - 2 m u \mp a \sqrt{m u^3}}{\left(1 - 3mu\mp 2a \sqrt{mu^3}\right)^{1/2}}.\\ \end{aligned } \label{xe}\ ] ] using @xmath82 , we have : @xmath83^{1/2 } } .
\label{l}\ ] ] the upper sign corresponds to the antiparallel orientation of particle s angular momentum @xmath19 and bh spin @xmath0 ( counter - rotation ) , the lower to the parallel one ( corotation ) . in order to find the parameters of isco ,
we need to add the condition @xmath84 written in terms of @xmath77 the third condition transforms to @xcite : @xmath85 since @xmath80 for a circular orbit , the third condition is @xmath86 .
the three equations @xmath87 , @xmath80 , @xmath88 form a closed system on three parameters of isco @xmath27 , @xmath89 and @xmath77 , which are then dependent only on @xmath10 and the kerr parameter @xmath0 .
solving this system @xcite , we obtain the equation for the inverse isco radii in the kerr metrics : @xmath90 or @xmath91 analytic solution for the isco radius can be found in the paper @xcite .
solutions of ( [ u-0 ] ) should be substituted into ( [ xe ] ) and ( [ l ] ) for finding @xmath27 and @xmath19 . in the limit @xmath92 , we have @xmath93 in the schwarzschild case . in the extreme kerr limit @xmath94
we obtain @xmath95 for the counter - rotating orbit and @xmath96 for the co - rotating case @xcite , @xcite .
for a description of the influence of body s spin on the parameters of its circular orbits we use the mathisson - papapetrou - dixon ( mpd ) @xcite equations of motion : @xmath97 here @xmath98 is a covariant derivative along the particle trajectory , @xmath28 is an affine parameter of the orbit @xcite , @xmath99 is the riemannian tensor , @xmath100 and @xmath101 are 4-momentum and 4-velocity of a test body , @xmath102 is its spin - tensor .
the equations were derived under the assumption that characteristic radius of the spinning particle is much smaller than the curvature scale of a background spacetime @xcite ( see also @xcite , @xcite ) and the mass of a spinning body is much less than that of bh .
these equations are , however , incomplete , because they do not define which point on the test body is used for spin and trajectory measurements , so we need some extra condition to do that @xcite .
we use the condition of tulczyjew @xcite that fixes the centre of mass of the test body @xmath103 let us consider the motion of a spinning particle in the equatorial plane of kerr bh . in this case the angular momentum of a spinning particle is always perpendicular to the equatorial plane @xcite .
therefore we can describe the test particle spin by only one constant @xmath1 which is the specific spin angular momentum of the particle .
value @xmath104 indicates the magnitude of the spin and @xmath1 itself is its projection on the @xmath44-axis .
it is more evident to think of the spin in terms of the particle s spin angular momentum @xmath105 which is parallel to the bh spin angular momentum @xmath106 , when @xmath107 , and antiparallel , when @xmath108 . here
@xmath109 is a unit vector in the direction of the @xmath44-axis and @xmath110 is a mass of the particle @xcite , @xcite .
saijo _ et al _ @xcite have derived the equations of motion of a spinning test particle for the equatorial plane of kerr bh .
the equations of motion for the variables @xmath63 , @xmath111 , @xmath112 in this case have the form @xcite @xmath113 + \frac{r^2 + a^2}{\delta}p_s,\\ & \sigma_s \lambda_s \dot \varphi = \left ( 1 + \frac{3ms^2}{r \sigma_s}\right)\left [ j -(a+s)e \right ] + \frac{a}{\delta}p_s.\\ \end{split } \label{spin - eqs}\ ] ] where @xmath114 ^ 2}{\sigma_s^3},\\ & r_s = p_s^2 - \delta \left\ { \frac{\sigma_s^2}{r^2 } + [ -(a + s ) e + j]^2 \right\ } , \\ & p_s = \left[r^2 + a^2 + \frac{a s ( r + m)}{r } \right ] e - \left(a + \frac{m s}{r } \right ) j. \end{split}\ ] ] here @xmath115 and the affine parameter @xmath28 is normalised as @xmath116 @xcite ; @xmath27 is the conserved energy per unit particle rest mass , and @xmath117 is the conserved total angular momentum per unit particle rest mass which is collinear to the spin of a bh
. equation for radial motion can be rewritten as @xmath118 where @xmath119 ^ 2 - \delta ( a + s)^2,\\ & \beta_s = \left [ \left(a + \frac{m s}{r } \right ) \left(r^2 + a^2 + \frac{a s ( r + m)}{r } \right ) - \delta ( a + s ) \right]j,\\ & \gamma_s = \left(a + \frac{m s}{r } \right)^2 j^2 - \delta \left[r^2 \left(1 - \frac{m s^2}{r^3}\right)^2 + j^2 \right ] .
\end{split } \label{abc - spin}\ ] ] this form is used by favata @xcite .
coefficients @xmath120 , @xmath121 , @xmath122 agree with rasband @xcite results .
the question of what we should consider as the ep in the case when a particle has spin does not differ from the `` spinless case '' in the kerr space - time .
if we want to keep analogy with the ep in the schwarzschild background , we should define it as the solution of equation @xmath123 with @xmath120 , @xmath121 , @xmath122 from ( [ abc - spin ] ) .
ep in this form can be found in the paper @xcite : @xmath124 with @xmath120 , @xmath121 , @xmath122 from ( [ abc - spin ] ) , see also favata @xcite , rasband @xcite , suzuki & maeda @xcite , @xcite , steinhoff & puetzfeld @xcite . for a review of misprints and comparison of effective potentials presented in different works see appendices of the paper @xcite .
using this ep for finding the isco parameters is presented , for example , in the paper of favata @xcite .
it is shown that conditions for a circular orbit ( the vanishing of @xmath59 and @xmath125 ) are equivalent to @xmath126 for the isco the third condition is added : @xmath127 a general solution of these three equations for the isco of a spinning particle is supposed to be found numerically . in the small - spin limit ,
favata @xcite analytically had found parameters of isco in the schwarzschild metric . since our purpose is to obtain the analytical expressions for circular orbits in the kerr metric
, it is more convenient to define the effective potential similar to ( [ v - spinless ] ) , by analogy with the work @xcite .
we rewrite equation ( [ l ] ) as : @xmath128 and define the effective potential as its right - hand side : @xmath129 it can be easily shown that solution of system of equations determining circular orbits is not changed if the effective potential is multiplied by some function which is not equal to zero ( with its first and second derivatives ) for physically relevant parameter values .
therefore it is more convenient for analytical calculations to define effective potential as @xmath130 with @xmath120 , @xmath121 , @xmath122 from ( [ abc - spin ] ) .
notice that in the case of a spinning particle we use the total angular momentum @xmath131 instead of the orbital angular momentum @xmath19 .
, whereas in the case of a spinning particle the conserved quantity is the total angular momentum @xmath132 , which includes spin terms @xcite . in this case
orbital angular momentum at infinity @xmath133 can also be introduced as @xmath134 , see @xcite .
further we can say that in both cases total angular momentum is conserved with a note that in spinless case it consists of orbital angular momentum part only .
] further , for the sake of convenience , we shall change variables and work not with @xmath63 and @xmath131 but with @xmath74 and @xmath135 , so the function @xmath136 will be used . in order to find the circular orbits we need to solve the system of equations @xmath137 the explicit form of these equations
is given in the first two equations in the system ( [ p ] ) .
the solution of this system defines the set of parameters @xmath89 and @xmath27 for stable and unstable circular orbits as functions of @xmath1 and @xmath77 . in figure [ figure - imc ]
we present the numerically calculated dependence of @xmath63 on total angular momentum @xmath131 and the influence of spin on it for the schwarzschild background , see also @xcite .
the point at which upper and lower branches on figure [ figure - imc ] meet defines the isco for the given value of spin . in order to find the dependence of the radius of the isco on spin
, we need to add the condition @xmath138 , which is equivalent to @xmath139 to the system ( [ n ] ) .
the resulting system for finding the isco parameters , thus , consists of three equations ( [ n ] ) and ( [ n1 ] ) and has the explicit form @xmath140 these three equations form a closed system on three parameters of isco @xmath27 , @xmath89 and @xmath77 , which are then dependent only on the kerr parameter @xmath0 and particle s spin @xmath1 .
this system can be used for numerical calculation of @xmath63 , @xmath27 , @xmath131 of isco at given @xmath0 and @xmath1 .
using ep in the form @xmath141 ( [ u - kerr - spin ] ) permits of splitting that system of three equations into two parts : equations @xmath142 and @xmath143 contain only @xmath63 and @xmath131 , and can be solved independently from equation @xmath144 .
so actually the problem is reduced to solving of the system of two equations .
it is clearly seen in the case of the schwarzschild metric for a spinless particle , see section 3 . in the case of using ep in the form @xmath136 ( [ v - spin ] ) the equations
@xmath145 and @xmath146 contain not only @xmath77 and @xmath89 , but also @xmath27 , so that in order to find these three parameters it is necessary to solve simultaneously three equations .
disadvantage of @xmath141 is that it contains radicals , and each differentiation increases complexity of expressions .
ep in the form @xmath136 , on the contrary , has polynomial form with respect to @xmath77 , and each differentation makes the expressions simpler .
all these properties are present in the spinless case also , and it is convenient to look for the isco parameters for the schwarzschild case using @xmath147 and the isco parameters for the kerr case using @xmath136 .
another important characteristics of the particle circular motion is its angular velocity .
we calculate the spin corrections for the orbital angular frequency of the particle at the isco , as seen from an observer at infinity .
this angular frequency is defined as @xmath148 the values @xmath149 and @xmath150 are found from the second and the third equations in ( [ spin - eqs ] ) , where we should substitute values of @xmath63 , @xmath27 and @xmath131 at a given orbit . to find the isco frequency @xmath151 we need to use the isco values of @xmath63 , @xmath27 and @xmath131 ,
see @xcite .
the system ( [ p ] ) does not admit of an analytical solution for arbitrary values of @xmath0 and @xmath1 .
using @xmath152 , we can expand the equations of the system ( [ p ] ) into series of powers of @xmath1 . in this section
we find analytical solution for isco parameters with small - spin linear corrections , for arbitrary value of @xmath0 . to perform analytical calculations and get analytical results for small - spin corrections , it is more convenient to use the following equivalent system of equations instead of the system ( [ n ] ) and ( [ n1 ] ) : @xmath153 similar approach for spinless case is used in @xcite .
we shall seek for solution of the system ( [ new - system ] ) in the form @xmath154 here @xmath155 , @xmath156 , @xmath157 are the values of @xmath89 , @xmath27 , @xmath77 in the spinless case , and @xmath158 , @xmath159 , @xmath160 are the linear corrections associated with spin .
the dimensionalities are @xmath161=[x_0]=[m]$ ] , @xmath162=1 $ ] , @xmath6=[e_0]=1 $ ] , @xmath163=[m^{-1}]$ ] , @xmath164=[u_0]=[m^{-1}]$ ] , @xmath165= [ m^{-2}]$ ] , @xmath8=[m]$ ] .
further we introduce these expressions into ( [ new - system ] ) and linearise it in spin .
since the equations of the system are to be satisfied for arbitrary values of spin , we obtain the system of six equations for six unknowns @xmath166 . first , there are three equations for three unknowns @xmath155 , @xmath156 , @xmath157 , at given @xmath0 : @xmath167 these equations describe the parameters of isco for the spinless case .
solution of the first two equations in this system gives us the values of @xmath155 and @xmath156 for an arbitrary circular orbit of given inverse radius , see ( [ xe ] ) .
solving three equations simultaneously , we can obtain the equation for @xmath157 , see ( [ u-0 ] ) .
the solution of equation ( [ u-0 ] ) for @xmath157 has a complicated form .
therefore it is more convenient to express the six unknowns not as explicit functions of @xmath0 but as the explicit functions of @xmath0 and @xmath157 , keeping in mind that @xmath157 can be found from eq.([u-0 ] ) at arbitrary @xmath0 .
before we proceed we need to notice that representation of all unknowns via @xmath157 given below could be rewritten in a different form using ( [ u-0 ] ) . to get @xmath155 and @xmath156 for isco we can substitute @xmath168 into the expressions for an arbitrary circular orbit ( [ xe ] ) .
but it is more convenient to obtain directly the connections between parameters @xmath155 , @xmath156 , @xmath157 for isco from three equations ( [ system - null ] ) .
it is easy to obtain that : @xmath169 whereas @xmath156 has positive value and we can write @xmath170 the value of @xmath155 can have both signs . from expression for @xmath155 in ( [ xe ] ) , we see that the second term in numerator is always bigger ( or equal ) than the first term , which means that the sign of @xmath155 is determined by sign in front of the second term in numerator .
therefore @xmath171 stands for co - rotating orbits and @xmath172 for counter - rotating orbits .
we can write : @xmath173 for @xmath156 we can also write : @xmath174 three equations for small - spin corrections @xmath158 , @xmath159 , @xmath160 , partially simplified with using the third equation in ( [ system - null ] ) and relation @xmath175 , are : @xmath176 after significant simplifications with using expressions ( [ x0e0viau0 ] ) , ( [ x0viau0 ] ) , ( [ e0viau02 ] ) we succeed to obtain the corrections in a simple form : @xmath177 to avoid singularity in the schwarzschild case ( @xmath11 , @xmath178 ) , we can also rewrite ( [ lineinye - popravki ] ) as @xmath179 let us summarise the results of this section . to obtain isco parameters for given @xmath0 in the form ( [ podstanovka ] ) one should find @xmath157 from equation @xmath180 and then calculate @xmath155 with using ( [ x0viau0 ] ) , @xmath156 with using ( [ e0viau01 ] ) , @xmath158 , @xmath159 , @xmath160 with using ( [ lineinye - popravki ] ) or ( [ lineinye - popravki - schw ] ) .
the value @xmath131 can be found by using @xmath181 : @xmath182 the isco radius @xmath63 can be found by using @xmath183 : @xmath184 we calculate also the isco frequency for arbitrary value of @xmath0 , as described at the end of section [ section - epspin ] .
using ( [ x0viau0 ] ) , ( [ e0viau02 ] ) and ( [ lineinye - popravki ] ) in ( [ omega - definition ] ) we obtain @xmath185 as @xmath186 @xmath187 or , in terms of @xmath188 as @xmath189 in all the formulae the upper sign corresponds to the antiparallel orientation of particle s angular momentum @xmath190 and bh spin @xmath191 ( counter - rotation , @xmath192 ) , the lower to the parallel one ( corotation , @xmath193 ) ; @xmath1 is the projection of spin on the @xmath44-axis and can be positive ( spins of particle and bh are parallel ) or negative ( antiparallel ) ; @xmath0 is positive or equal to zero ; the @xmath44-axis is chosen to be parallel to bh spin @xmath191 .
in order to illustrate the obtained results , let us consider some important limit cases : schwarzschild metric and kerr metric , when the bh rotation is slow ( @xmath14 ) and extremely fast ( @xmath194 ) .
the latter case is called `` extreme kerr bh '' , for at the values of @xmath0 exceeding @xmath10 the bh disappears and we have a naked singularity @xcite without an event horizon . in the schwarzschild case there is no prefential direction associated with bh spin ( @xmath11 ) , so @xmath44-direction can be chosen arbitrarily .
choosing @xmath44-axis along @xmath190 vector , and taking @xmath195 , we obtain : @xmath196 here instead of @xmath1 , which is the projection on @xmath44-axis that does not unabiguously correspond to any physical direction in schwarzschild case , we use @xmath197 which is the projection of particle s spin upon the direction of @xmath190 and is positive when the particle s spin is parallel to it and negative when it is antiparallel .
value @xmath131 is considered as positive in this case .
small - spin corrections for schwarzschild metric were derived by favata @xcite .
the case of slowly rotating kerr bh ( @xmath14 ) allows us to separate corrections that arise from different types of angular momentum coupling .
our general formulae , presented in section [ section - arbitrary ] , allow us to obtain correction terms proportional to @xmath0 , @xmath1 and @xmath198 . considering @xmath14 in equation ( [ eq - u0 ] ) , we get @xmath157 : @xmath199 using the first two terms from ( [ u0-slow - kerr ] ) in ( [ e0viau01 ] ) , ( [ x0viau0 ] ) , ( [ lineinye - popravki - schw ] ) , we obtain @xmath155 , @xmath156 , @xmath158 , @xmath159 and @xmath160 .
note that when using formulae ( [ lineinye - popravki ] ) for calculation of corrections @xmath159 and @xmath160 , the quadratic terms from ( [ u0-slow - kerr ] ) should be also taken into account .
corrections @xmath0 , @xmath1 , @xmath198 for isco parameters can be as well obtained directly by solving the system ( [ p ] ) , with keeping terms linear in @xmath0 and @xmath1 .
if we do nt assume any ratio between values @xmath0 and @xmath1 , a full description of the problem requires also to obtain quadratic terms @xmath200 and @xmath201 in isco parameters .
the scheme for obtaining the squared corrections is analogous to the linearisation procedure , only we need to express the parameters in the form @xmath202 and retain the squared in spin terms in the system ( [ p ] ) . for our purposes we can use more simple way : corrections @xmath200 can be found from solution of the system ( [ p ] ) for schwarzschild case , and @xmath201-corrections can be found from consideration of the spinless case .
finally , we obtain : @xmath203 different situations of mutual orientation of bh and particle spin are presented on figure [ figure - im2 ] , where the case @xmath14 is drawn on the basis of the first three terms in formula ( [ kerr - slow ] ) for @xmath204 .
for example , for particle @xmath205 we should use the upper sign and @xmath107 , for particle @xmath206 we should use the lower sign and @xmath108 .
this result allows to understand the character of the spin - orbital coupling in gr . from the terms @xmath207 and @xmath208 , we see that the the spin - orbital coupling depends on mutual orientation of the spin and angular momentum vectors : it is attractive and the radius decreases , when they are parallel , and repulsive , if they are antiparallel . the terms with @xmath198 , @xmath201 , @xmath200 in formulae ( [ kerr - slow ] ) are of second infinitesimal order and include information about interactions of particle spin , bh spin and orbital momentum . considering the formula for radius it is easy to show that the trinomial containing these three terms is negative for all possible values of @xmath0 and @xmath1 .
therefore , quadratic corrections always lead to decrease of the isco radius . substituting values of @xmath209 , @xmath210 and @xmath204 from ( [ kerr - slow ] ) into ( [ omega - definition ] ) ,
we obtain the isco frequency as @xmath211
in the extreme kerr background ( @xmath194 ) for a spinless particle the isco radius is @xmath9 in the antiparallel case ( counter - rotation ) , and @xmath10 in the parallel case ( co - rotation ) .
numerical solution of the system ( [ p ] ) for @xmath0 close to @xmath10 shows that isco radius has very different behaviour for co - rotating ( @xmath193 ) and counter - rotating ( @xmath192 ) particles , see fig .
[ figure - ekerr - co ] and [ figure - ekerr - rr ] .
we will therefore consider these cases separately . in the antiparallel case
we get , using ( [ e0viau01 ] ) , ( [ x0viau0 ] ) , ( [ lineinye - popravki ] ) , ( [ omega - arbitrary - a ] ) : @xmath212 for the parallel case we search for isco parameters at values of @xmath13 , @xmath213 , because divergences may appear at @xmath214 at some ways of calculations . using ( [ e0viau01 ] ) , ( [ x0viau0 ] ) , ( [ lineinye - popravki ] ) , ( [ omega - arbitrary - a ] )
, we get : @xmath215 we see that in the case of @xmath12 ( @xmath216 ) the corrections , linear in spin , are absent in formulae for isco radius and frequency .
this was also demonstrated in @xcite . in the next section
we discuss this feature in details and find the exact in spin expressions for the isco parameters .
the radius of horizon of the kerr bh , @xmath217 in the case @xmath218 will have the form @xcite @xmath219 as @xmath220 and @xmath221 for @xmath213 , it will always be that @xmath222 .
in the work @xcite on basis of the numerical calculation , it was noticed that in the extreme kerr background for the parallel case the magnitude of test - body s spin does not influence the radius of the last stable orbit and it always remains equal to @xmath10 . solving numerically eqs ( [ p ] ) , we plot fig . [ figure - ekerr - co ] where this conclusion is distinctly visible .
we also succeeded in proving this analytically . on basis of results of the numerical solution
we have substituted @xmath223 into the system ( [ p ] ) and have obtained the expressions for @xmath89 and @xmath27 , which is the analytical proof of existence of such solution of the system .
the exact solution is then @xmath224 using ( [ kerr - exact ] ) in ( [ omega - definition ] ) , we obtain the isco frequency as @xmath225 developing ( [ kerr - exact ] ) in series in @xmath1 , we get the linear in spin terms which are in agreement with ( [ kerr - co ] ) at @xmath216 .
this result allows us to obtain the full in spin correction to the nearly extreme black hole , with @xmath218 , considered in the section [ section - extreme - kerr ] . for this purpose after substituting this value for @xmath0 we develop the system ( [ p ] ) in series in @xmath226 and seek for the solution in the form @xmath227 , where @xmath228 stands for @xmath229 .
after some algebraic calculation we obtain for the nearly extreme kerr metric the expressions : @xmath230 developing this result in series in @xmath1 , one obtains all terms in expressions ( [ kerr - co ] ) .
note that derivation of mpd equations was performed under condition @xmath220 , so always we have @xmath231 , @xmath232 in ( [ kerr - exact ] ) and ( [ kerr - delta ] ) .
indeed , equations of motion of a spinning particle were derived under the assumption that its characteristic size is very small compared with the characteristic length of the background field ( e.g. the distance from the central body ) @xcite , @xcite .
mller @xcite had shown in the context of special relativity that the classical spinning body must have a certain minimum characteristic size connected with its specific spin , see also wald @xcite . for particle at isco , in our system of units
the above conditions reduce to @xmath220 , see also @xcite , @xcite , @xcite .
in previous parts of the present paper we have discussed isco , and investigated how the presence of spin influences parameters of isco for different given @xmath0 .
let us now discuss not innermost stable but arbitrary circular orbits .
we will investigate how the presence of spin influences orbit of given radius
namely , how @xmath27 , @xmath131 , @xmath185 of a circular orbit of given radius are changed if a test particle has spin .
let us suppose that inverse radius @xmath77 of circular orbit is known and consider the system ( [ n ] ) .
we shall seek for its solution in the form @xmath233 here @xmath155 and @xmath156 are the values of @xmath89 and @xmath27 in the `` spinless '' case , see ( [ xe ] ) , @xmath158 and @xmath159 are the linear corrections associated with spin .
we obtain for the linear corrections : @xmath234 these expressions are valid for both stable and unstable circular orbits with the zero order isco parameters given by ( [ xe ] ) .
value of total angular momentum @xmath235 can be obtained using @xmath181 .
the expressions ( [ circ - first ] ) can be simplified further . using ( [ xe ] )
we obtain : @xmath236 @xmath237 written as functions of @xmath63 the resulting formulae are @xmath238 @xmath239 substituting ( [ e1-long ] ) and ( [ j1-long ] ) into ( [ omega - definition ] ) , we obtain angular frequency @xmath240 at the circular orbit with a given inverse radius @xmath77 as @xmath241 or , as function of radial coordinate @xmath63 , as @xmath242 the linear corrections for the energy , angular momentum and frequency at a given radius for the circular orbits are given also in @xcite , @xcite in another presentation , see also appendices a and b in @xcite where comparison of spin orientation in different papers is discussed .
\(i ) the linear in spin corrections for the isco parameters : radius , total angular momentum , energy , orbital angular frequency for arbitrary kerr parameter @xmath0 , are found analytically , see section [ section - arbitrary ] .
\(ii ) small - spin linear corrections for schwarzschild metric ( firstly obtained by favata @xcite ) were derived , see formulae ( [ schw ] ) .
\(iii ) expressions for the isco parameters of a slowly - rotating kerr bh up to second order in spin and kerr parameter are obtained , see ( [ kerr - slow ] ) .
the case of a slowly rotating kerr bh gives the idea of spin - orbital coupling s influence on the parameters of isco .
it is found that spin - orbital coupling has an attractive character when the total angular momentum @xmath131 and spin of either bh or a test - body are parallel and repulsive when they are antiparallel .
\(iv ) linear in spin corrections to the isco parameters for extreme kerr bh both for counter- and co - rotating cases are obtained , see ( [ kerr - counter ] ) and ( [ kerr - co ] ) .
\(v ) the exact parameters of isco for a spinning test - body in the extreme kerr background in corotation case are found , see ( [ kerr - exact ] ) and ( [ kerr - delta ] ) .
it is proved analytically that radius and angular frequency of such orbits are independent of the particle s spin while the values of energy and total angular momentum depend on it .
\(vi ) figures [ figure - im1 ] and [ figure - im2 ] illustrate the number of possible configuration for rotating bh and a spinning test particle with parameters calculated in the present work .
we are greatful to an anonymous referee for useful suggestions leading to improvement of the paper .
the work of gsbk , oyut and pije was partially supported by the russian foundation for basic research grant no .
14 - 02 - 00728 and the russian federation president grant for support of leading scientific schools , grant no . nsh-261.2014.2 . | we consider the motion of classical spinning test particles in schwarzschild and kerr metrics and investigate innermost stable circular orbits ( isco ) .
the main goal of this work is to find analytically the small - spin corrections for the parameters of isco ( radius , total angular momentum , energy , orbital angular frequency ) of spinning test particles in the case of vectors of black hole spin , particle spin and orbital angular momentum being collinear to each other .
we analytically derive the small - spin linear corrections for arbitrary kerr parameter @xmath0 .
the cases of schwarzschild , slowly rotating and extreme kerr black hole are considered in details . for a slowly rotating black hole
the isco parameters are obtained up to quadratic in @xmath0 and particle s spin @xmath1 terms . from the formulae
obtained it is seen that the spin - orbital coupling has attractive character when spin and angular momentum are parallel and repulsive when they are antiparallel . for the case of the extreme kerr black hole with co - rotating particle we succeed to find the exact analytical solution for the limiting isco parameters for arbitrary spin .
it has been shown that the limiting values of isco radius and frequency do not depend on the particle s spin while values of energy and total angular momentum depend on it .
we have also considered circular orbits of arbitrary radius and have found small - spin linear corrections for the total angular momentum and energy at given radius .
system of equations for numerical calculation of isco parameters for arbitrary @xmath0 and @xmath1 is also explicitly written .
pacs numbers : : 04.20.-q |
cluster algebras were first introduced by s. fomin and a. zelevinsky in @xcite to design an algebraic framework for understanding total positivity and canonical bases for quantum groups .
a cluster algebra is a subring of a rational function field generated by a distinguished set of laurent polynomials called cluster variables .
the long - standing positivity conjecture , now proved in @xcite and @xcite , asserts that the coefficients in any cluster variable are positive integers . from the combinatorial point of view
, the positivity conjecture suggests that these coefficients should count some combinatorial objects .
lots of research focuses on building such combinatorial models .
we give a brief summary of the pros and cons of four such models .
* * @xmath1-paths : * in @xcite , schiffler ( independently carroll price and fomin zelevinsky in their unpublished work ) obtained a formula for the cluster variables of a cluster algebra of finite type @xmath0 ( see 2 for the definition ) in terms of @xmath1-paths . this formula has been modified and generalized to cluster algebras coming from surfaces @xcite . * * perfect matchings of a snake diagram : * a description that is similar to the @xmath1-path model but has a more graph - theoretic flavor @xcite .
interesting combinatorics , for example the snake graph calculus @xcite , arises in the study of this model . this formula is also restricted to cluster algebras coming from surfaces . *
* compatible pairs in a dyck path : * in @xcite , the cluster variables of rank 2 quivers , which do not necessarily come from surfaces , are described in terms of dyck paths . a more general construction of the so - called compatible pairs is used in the study of greedy bases in @xcite , and another generalization called gcc is used in @xcite .
* * broken lines and theta functions : * discovered in @xcite , they are the most general combinatorial models so far . but the model is mysterious that even the finiteness of the number of broken lines is not immediate from the definition .
our ultimate goal is to find a combinatorial model that is both general and effective in computation . even though this goal appears out of reach for
now , we feel that the model of maximal dyck paths and compatible pairs has the potential to be generalized .
this motivates the main goal of this paper : _ for a type @xmath0 quiver , give a new formula for the cluster monomials using a combinatorial model similar to compatible pairs , and find the bijections to other known models .
_ we reach this goal by proving three equivalent formulas . in theorem [ thm:01sequence ] , we give a formula using a sequence of 0 - 1 sequences called gcs ( globally compatible sequence ) , where each vertex of the quiver is assigned a 0 - 1 sequence satisfying a certain compatibility condition .
in theorem [ thm : gccxa ] , we give a formula using globally compatible collections ( gccs ) in dyck paths .
this formula has a similar flavor to the combinatorial formula for greedy bases in @xcite .
in [ dvecpipe ] , we first use a combinatorial gadget called pipelines to decompose the * d*-vector of a cluster monomial into the ones of cluster variables , then give a formula for cluster variables using gccs in theorem [ mainthm ] .
moreover , we construct a bijection between gcss ( which is equivalent to gccs ) and broken lines in theorem [ mainthm2 ] ( the even rank case ) and [ mainthm2 ] ( the general case ) , which relates our new formula with the theta functions constructed in @xcite .
the simplicity of this bijection came as a surprise for us : namely , under our setting , the @xmath2-th number in a gcs ( which is a 0 - 1 sequence ) is 0 if and only if the corresponding broken line bends at the @xmath2-th coordinate hyperplane @xmath3 .
this suggests that there could be further connections between our new combinatorial formulas and theta functions , and thus could provide a new approach to understanding broken lines ( which are difficult to describe explicitly in general ) .
the paper is organized as follows . in
2 we recall the definition of cluster algebra and some facts about type @xmath0 quivers . in
3 we define the * d*-vector of a cluster monomial and introduce its decomposition using pipelines .
4 consists of the statements of the main results of the paper . in
5 we prove the gcc formula for cluster variables ( theorem [ mainthm ] ) by establishing a bijection from gccs to perfect matchings . then in 6 we give the proof of the other main results of 4 . in
7 we give the bijection between gcss and broken lines .
then we give some examples in 8 . in the appendix
, we give another proof of theorem [ mainthm ] using @xmath1-paths .
_ acknowledgement .
_ we are grateful to ralf schiffler for valuable correspondences , and to man wai cheung , mark gross and greg muller for very helpful discussion on scattering diagrams and broken lines .
in this section , we recall some definitions and fix notations about quivers and skew - symmetric cluster algebras ( 2.1 ) and some special type @xmath0 quivers ( 2.2 ) .
recall that a finite oriented graph is a quadruple @xmath4 formed by a finite set of vertices @xmath5 , a finite set of arrows @xmath6 and two maps @xmath7 and @xmath8 from @xmath6 to @xmath5 which send an arrow @xmath9 respectively to its head @xmath10 and its tail @xmath11 .
an arrow @xmath9 whose head and tail coincide is a _ loop _ ; a _
@xmath12-cycle _ is a pair of distinct arrows @xmath13 and @xmath14 such that @xmath15 and @xmath16 .
similarly , it is clear how to define @xmath17-cycles for @xmath18 .
a vertex is a _ source _ ( respectively a _ sink _ ) if it is not the head ( resp .
the tail ) of any arrow . in this paper ,
a _ quiver _ is a finite oriented graph without loops or 2-cycles . given a quiver @xmath19 and a vertex @xmath20
, the mutation @xmath21 is the new quiver @xmath22 obtained as follows : 1 . for every path of the form @xmath23 , add a new arrow from @xmath24 to @xmath25 .
2 . reverse all arrows incident to @xmath26 .
3 . remove all 2-cycles .
let @xmath4 be a quiver .
let @xmath27 ( for simplicity , we denote @xmath28 by @xmath2 in this paper if no confusion arises ; and later we also use notation @xmath29 to denote the same set ) .
let @xmath30 be the field of rational functions in @xmath31 with rational coefficients .
seed _ is a pair @xmath32 where @xmath33 is a set of elements of @xmath34 which freely generate the field @xmath34 . for any vertex @xmath35 ,
we denote @xmath36 the mutation @xmath37 is the seed @xmath38 where @xmath39 and @xmath40 is obtained from @xmath41 by replacing @xmath42 by @xmath43 let @xmath44 be the initial seed .
a _ cluster _ is a set @xmath40 which appears in a seed @xmath38 obtained from the initial seed by iterated mutations .
an element in a cluster is called a _
cluster variable_. a _
cluster monomial _ is a product of cluster variables in the same cluster .
the ( coefficient - free ) _ cluster algebra _ @xmath45 associated with @xmath19 is the subring of @xmath34 generated by all cluster variables .
next we recall the definition of the cluster algebra @xmath46 with principal coefficient corresponding to the coefficient - free cluster algebra @xmath47 .
let @xmath48 , @xmath49 .
define a quiver @xmath50 with the vertex set @xmath51 and the edge set @xmath52 in other words , @xmath50 is obtained from @xmath19 by adding an arrow from @xmath53 to @xmath2 for each @xmath54 .
we call @xmath55 unfrozen vertices and @xmath56 frozen vertices . starting with the initial seed @xmath57
, we mutate it iteratively similar as above , with the restriction that we only use the mutations @xmath58 for @xmath59 ( that is , only mutate at the unfrozen vertices ) .
for each new seed @xmath60 , the first @xmath17 rational functions @xmath61 form a cluster .
the union of all clusters gives the set of cluster variables .
the cluster algebra @xmath46 is the subring of @xmath62 generated over @xmath63 $ ] by all cluster variables .
let @xmath64 be another quiver .
we say that @xmath19 is a _ subquiver _ of @xmath22 if @xmath65 and @xmath66 and @xmath67 for any arrow @xmath68 .
we say that @xmath19 is a _ full subquiver _ of @xmath22 if @xmath19 can be obtained from @xmath22 by removing vertices @xmath69 and their incident arrows . here
we define type @xmath0 , linear , completely extended linear , and extended linear quivers .
the relation among the four classes can be described as follows : @xmath70{-45}{\large $ \supset$ } } \ , \{\textrm{linear}\ } \endaligned\ ] ] by definition , type @xmath0 quivers are those that are mutation equivalent to quivers of the form @xmath71 . in @xcite ,
a. buan and d. vatne showed that a type @xmath0 quiver is a connected quiver such that * all nontrivial simple cycles in the underlying graph have length 3 , and the corresponding directed subgraphs are oriented ( 3-cycles ) ; * the vertex degrees of the underlying graph are at most 4 ; moreover , a degree-4 vertex belongs to two 3-cycles , a degree-3 vertex belongs to one 3-cycle . for two integers @xmath72 and @xmath73 ,
we denote @xmath74=\{a , a+1,\ldots , b\}$ ] if @xmath75 , and @xmath74=\emptyset$ ] if @xmath76 .
a _ linear quiver _ is a quiver with @xmath17 vertices @xmath77 and @xmath78 arrows in which any two consecutive vertices @xmath28 and @xmath79 @xmath80)$ ] are connected by a single arrow in either direction and there are no others arrows . in order to have a convenient description for a linear quiver @xmath19
, we construct a sequence @xmath81 such that @xmath82 if there is an arrow going from the vertex @xmath28 to the vertex @xmath79 , and @xmath83 otherwise .
for example , for the quiver @xmath84 , we have @xmath85 .
we define a _ completely extended linear quiver _ @xmath22 as obtained from a linear quiver @xmath19 by attaching a 3-cycle to every edge , a 3-cycle to @xmath86 , and a 3-cycle to @xmath87 .
( so @xmath22 has @xmath88 vertices . ) by abuse of terminology , we also call the pair @xmath89 a completely extended linear quiver , whenever we need to specify the linear quiver @xmath19 .
let @xmath89 be a completely extended linear quiver with @xmath90 and @xmath91 .
if @xmath92 is adjacent to both @xmath93 and @xmath94 then we also denote @xmath95 . for the 3-cycle attached to @xmath86 , the head ( resp .
tail ) of the outgoing ( resp .
incoming ) arrow is denoted @xmath96 ( resp .
@xmath97 ) .
we define @xmath98 and @xmath99 similarly . [ quiverw3cycles ] the quiver @xmath100 in figure [ celnquiver ] is a completely extended linear quiver , where @xmath19 is the linear part @xmath101 .
in there , @xmath102 , @xmath103 , @xmath104 , @xmath105 , @xmath106 , @xmath107 and @xmath108 .
( v1 ) at ( 0:1 ) [ red ] 1 ; ( v2 ) at ( 0:2.732 ) [ red ] 2 ; ( v3 ) at ( 0:4.464 ) [ red ] 3 ; ( v4 ) at ( 0:6.196 ) [ red ] 4 ; ( v5 ) at ( -120:1 ) 5 ; ( v6 ) at ( 120:1 ) 6 ; ( -60:1.732)++(1,0 ) node ( v7 ) 7 ; ( -60:1.732)++(2.732,0 ) node ( v8 ) 8 ; ( -60:1.732)++(4.464,0 ) node ( v9 ) 9 ; ( 30:1.732)++(6.196,0 ) node ( v10 ) 10 ; ( -30:1.732)++(6.196,0 ) node ( v11 ) 11 ; ( v1 ) to ( v5 ) ; ( v5 ) to ( v6 ) ; ( v6 ) to ( v1 ) ; ( v1 ) ( v2 ) node[midway , above , font= ] @xmath109 ; ( v2 ) to ( v7 ) ; ( v7 ) to ( v1 ) ; ( v2 ) ( v3 ) node[midway , above , font= ] @xmath110 ; ( v3 ) to ( v8 ) ; ( v8 ) to ( v2 ) ; ( v4 ) ( v3 ) node[midway , above , font= ] @xmath111 ; ( v3 ) to ( v9 ) ; ( v9 ) to ( v4 ) ; ( v4 ) to ( v10 ) ; ( v10 ) to ( v11 ) ; ( v11 ) to ( v4 ) ; for convenience , if @xmath112 , then we denote the variable @xmath113 .
an _ extended linear quiver _
@xmath114 is obtained from a completely extended linear quiver @xmath89 by removing some vertices ( or none ) in @xmath69 and the arrows incident with them .
equivalently , we can characterize @xmath115 as a quiver obtained from @xmath19 by adding some ( or none ) of the following : * a 3-cycle or an edge hung on @xmath86 , or * a 3-cycle or an edge hung on @xmath87 , or *
3-cycles attached to some edges of @xmath19 .
there is an obvious way to obtain a completely extended linear quiver @xmath89 from an extended linear quiver @xmath114 ( up to relabeling vertices in @xmath116 ) .
an example is shown in figure [ completeaquiver ] .
lll ( v1 ) at ( 0:1 ) [ red ] 1 ; ( v2 ) at ( 0:2.732 ) [ red ] 2 ; ( v3 ) at ( 0:4.464 ) [ red ] 3 ; ( v4 ) at ( -120:1 ) 4 ; ( -60:1.732)++(2.732,0 ) node ( v5 ) 5 ; ( v1 ) to ( v4 ) ; ( v2 ) to ( v1 ) ; ( v2 ) to ( v3 ) ; ( v3 ) to ( v5 ) ; ( v5 ) to ( v2 ) ; & ( v1 ) at ( 0:0 ) ; ( v2 ) at ( 0:2 ) ; ( v3 ) at ( -60:1.732 ) ; ( v1 ) to ( v2 ) ; & ( v1 ) at ( 0:1 ) [ red ] 1 ; ( v2 ) at ( 0:2.732 ) [ red ] 2 ; ( v3 ) at ( 0:4.464 ) [ red ] 3 ; ( v4 ) at ( -120:1 ) 4 ; ( -60:1.732)++(2.732,0 ) node ( v5 ) 5 ; ( v6 ) at ( 120:1 ) [ blue ] 6 ; ( -60:1.732)++(1,0 ) node ( v7 ) [ blue ] 7 ; ( 30:1.732)++(4.464,0 ) node ( v8 ) [ blue ] 8 ; ( -30:1.732)++(4.464,0 ) node ( v9 ) [ blue ] 9 ; ( v1 ) to ( v4 ) ; ( v4 ) to ( v6 ) ; ( v6 ) to ( v1 ) ; ( v2 ) to ( v1 ) ; ( v1 ) to ( v7 ) ; ( v7 ) to ( v2 ) ; ( v2 ) to ( v3 ) ; ( v3 ) to ( v5 ) ; ( v5 ) to ( v2 ) ; ( v3 ) to ( v8 ) ; ( v8 ) to ( v9 ) ; ( v9 ) to ( v3 ) ;
it is well known that any cluster algebra associated to a type @xmath0 quiver with @xmath17 vertices can be constructed from a triangulation on the @xmath117-gon .
a _ diagonal _ on the @xmath117-gon is a line segment connecting two non - adjacent vertices . a connected curve on the polygon
is called a _
pseudo - diagonal _ if it is isotopic to a diagonal ( and its endpoints are the same as those of the diagonal ) and if its interior is in the interior of the polygon .
two pseudo - diagonals are said to be _ crossing _ if they intersect in the interior of the polygon .
let @xmath19 be a type @xmath0 quiver with @xmath17 vertices , and let @xmath118 be the diagonals given by the corresponding triangulation on the @xmath117-gon and @xmath119 the boundary edges .
it is also known that the cluster variables of @xmath120 are in natural bijection with all the diagonals of the polygon . using this bijection ,
a cluster monomial of @xmath120 can be identified with a finite set of pairwise non - crossing pseudo - diagonals , or equivalently with @xmath121where @xmath122 is a non - negative integer , @xmath123 are pairwise non - crossing diagonals , and @xmath124 are positive integers .
the following definition uses a natural intersection number , which was already considered in @xcite . for two diagonals @xmath125 on the @xmath117-gon
, we define the intersection number @xmath126 of @xmath127 and @xmath128 as follows : @xmath129 then the * d*-vector of the cluster monomial @xmath130 is defined by @xmath131 it is easy to see that the * d*-vector @xmath132 of any cluster monomial satisfies the following : * property a. * for any 3-cycle @xmath133 in @xmath19 such that @xmath134 are positive and satisfy the triangle inequalities ( i.e. , the sum of any two numbers is strictly greater than the third ) , the sum @xmath135 is even .
let @xmath136 be the set of all integer vectors @xmath132 satisfying property a. in this subsection we prove that the cluster monomials of @xmath120 are in bijection with @xmath136 .
we will define a map from @xmath136 to the cluster monomials , which would then induce the immediate bijection .
let @xmath137_+=\max(x,0)$ ] for any real number @xmath138 .
let @xmath139 .
we define a function @xmath140 as follows : @xmath141_+-[x - y - z]_+ -[y - x - z]_+}{2}=\begin{cases } x , & \textrm { if } y > x+z;\\ y , & \textrm { if } x > y+z;\\ 0 , & \textrm { if } z > x+y;\\ \displaystyle\frac{x+y - z}{2 } , & \textrm { otherwise}. \end{cases}\ ] ] for convenience , we denote @xmath142 ( if no confusion shall arise , we denote @xmath143 . )
we are ready to construct the so - called pipelines associated to @xmath144 .
* step 1 : * if @xmath145 then draw @xmath146 marking points on the diagonal @xmath147 to separate it into @xmath148 segments . if @xmath149 then draw @xmath150 pipes so that these pipes are pairwise non - crossing pseudo - diagonals isotopic to @xmath147 and that they do not cross any other @xmath151 @xmath152 .
* step 2 : * if @xmath147 and @xmath151 are two sides of a triangle with the third side @xmath153 , then for @xmath154 , we join the two @xmath155-th marking points on @xmath147 and @xmath151 ( ordering in the increasing distance from the common endpoint of @xmath147 and @xmath151 ) by a pipe inside the triangle .
draw these pipes so that they are disjoint from each other and from the pipes constructed in step 1 .
* step 3 : * suppose that @xmath156 form a triangle .
then for each marking point on @xmath147 that is not connected by a pipe to any marking point on @xmath151 or @xmath153 , we draw a pipe from the marking point to the common endpoint of @xmath151 and @xmath153 .
draw these pipes inside the triangle in such a way that they are non - crossing with each other and with the pipes constructed in step 1,2 .
a pipeline is a union of pipes connected consecutively through the marking points ( but not through the vertices of the @xmath117-gon ) .
then the pipelines are pairwise non - crossing .
since the endpoints of each pipeline are non - adjacent vertices of the polygon , every pipeline is a pseudo - diagonal .
hence the union of these pipelines corresponds to a cluster monomial .
clearly the * d*-vector of this cluster monomial is equal to @xmath157 . using the above construction of pipelines
, it is straightforward to prove the following : [ d - vector criteria ] let @xmath19 be a type @xmath0 quiver
. then 1 .
@xmath158 is the * d*-vector of some cluster monomial of @xmath120 if and only if @xmath159 .
two distinct cluster monomials have different * d*-vectors .
this allows us to denote the ( unique ) cluster monomial with * d*-vector @xmath160 by @xmath161 $ ] or @xmath162 $ ] .
each pipeline @xmath163 corresponds to a 0 - 1 sequence @xmath164 such that @xmath165 note that @xmath163 corresponds to a linear full subquiver of @xmath19 .
let @xmath166 be the multiset of all such sequences . then @xmath167=\prod_{{\bf b } \in s } x[{\bf b}].\ ] ] [ ex3332431 ] the first two pictures in figure [ figure : pipelines ] are a type @xmath0 quiver with 7 vertices and its corresponding triangulation on the 10-gon . for clearer illustration , the 10-gon
is drawn as a concave polygon .
the bottom illustrates the construction of pipelines for @xmath168 .
( v1 ) at ( 150:1.732 ) 1 ; ( v2 ) at ( 0:0 ) 2 ; ( v3 ) at ( 0:1.732 ) 3 ; ( 30:1.732)++(1.732,0 ) node ( v4 ) 4 ; ( v5 ) at ( -60:1.732 ) 5 ; ( -120:1.732)++(0.866,-1.5 ) node ( v6 ) 6 ; ( -60:1.732)++(0.866,-1.5 ) node ( v7 ) 7 ; ( v5)++(0,-2 ) node [ below ] @xmath19 ; ( v2 ) to ( v1 ) ; ( v2 ) to ( v3 ) ; ( v3 ) to ( v5 ) ; ( v5 ) to ( v2 ) ; ( v5 ) to ( v6 ) ; ( v6 ) to ( v7 ) ; ( v7 ) to ( v5 ) ; ( v3 ) to ( v4 ) ; = [ font= ] ( v1 ) at ( 0:0 ) ; ( v2 ) at ( 60:1.732 ) ; ( v3 ) at ( 0:1.732 ) ; ( v4 ) at ( -60:1.732 ) ; ( v1 ) circle [ radius=2pt ] ; ( v2 ) circle [ radius=2pt ] ; ( v3 ) circle [ radius=2pt ] ; ( v4 ) circle [ radius=2pt ] ; ( v1.center ) to ( v2.center ) ; ( v2.center ) ( v3.center )
node[midway , left ] 1 ; ( v3.center ) to ( v1.center ) ; ( v3.center ) to ( v4.center ) ; ( v2)++(1.732,0 ) node ( v5 ) ; ( v3)++(1.732,0 ) node ( v6 ) ; ( v4)++(1.732,0 ) node ( v7 ) ; ( v5 ) circle [ radius=2pt ] ; ( v6 ) circle [ radius=2pt ] ; ( v7 ) circle [ radius=2pt ] ; ( v5.center ) to ( v2.center ) ; ( v7.center ) to ( v4.center ) ; ( v3.center ) ( v5.center ) node[midway , left ] 2 ; ( v5.center )
( v6.center ) node[midway , right ] 3 ; ( v6.center )
( v3.center ) node[midway , above ] 5 ; ( v6.center ) ( v7.center )
node[midway , right ] 6 ; ( v7.center )
( v3.center ) node[midway , left ] 7 ; ( v2)++(2.598,1.5 ) node ( v8 ) ; ( v3)++(2.598,1.5 ) node ( v9 ) ; ( v8 ) circle [ radius=2pt ] ; ( v9 ) circle [ radius=2pt ] ; ( v5.center ) to ( v8.center ) ; ( v8.center ) to ( v9.center ) ; ( v9.center )
( v5.center ) node[midway , above ] 4 ; ( v9.center ) to ( v6.center ) ; ( v4)++(3.464,0 ) node ( v10 ) ; ( v10 ) circle [ radius=2pt ] ; ( v6.center ) to ( v10.center ) ; ( v10.center ) to ( v7.center ) ; ( v7)++(0,-0.5 ) node [ below ] @xmath169 ; there are 5 pipelines , passing through edges sets @xmath170 , @xmath171 , @xmath172 , @xmath173 , @xmath174 , respectively .
so the cluster monomial @xmath162 $ ] is decomposed as @xmath175=x[1,1,1,1,0,0,0 ] \cdot x[1,1,0,0,1,1,0 ] \cdot x[1,1,0,0,1,0,1 ] \cdot x[0,0,1,1,1,1,0 ] \cdot x[0,0,1,0,1,1,0]$ ] .
[ a to a+ ] for @xmath158 , assume that @xmath176_+$ ] satisfies property a and @xmath177_+]=\prod x[{\bf b}]$ ] is a factorization into cluster variables in the same cluster .
then @xmath167=x[[{\bf a}]_+]\prod_i x_i^{[-a_i]_+}=\prod x[{\bf b}]\prod_i x_i^{[-a_i]_+},\ ] ] and that @xmath178 $ ] and @xmath179 ( @xmath149 ) are in the same cluster . if @xmath149 , then there is no @xmath180 satisfying @xmath181
. thus no pseudo - diagonals corresponding to pipelines for @xmath176_+$ ] will cross the diagonal @xmath147 .
therefore , after adding @xmath147 we still get a set of non - crossing pseudo - diagonals , which means that @xmath178 $ ] and @xmath179 ( @xmath149 ) are in the same cluster .
it then follows that @xmath162=x[[{\bf a}]_+]\prod_i x_i^{[-a_i]_+}$ ] .
now we recall the definition of * d*-vectors ( a.k.a .
denominator vectors ) , following @xcite .
every nonzero element @xmath182 ( more generally , for @xmath183 in the upper cluster algebra ) can be uniquely written as @xmath184 where @xmath185 is a polynomial with coefficients in @xmath186 ( which should be replaced by the integral group ring @xmath187 of the coefficient semifield for a cluster algebra with coefficients ; see @xcite for details ) which is not divisible by any cluster variable in @xmath188 .
then @xmath189 is called the _ * d*-vector _ of @xmath183 with respect to the cluster @xmath190 . [ rmk2.1 ] assume that @xmath19 is a full subquiver of @xmath22 .
we compare cluster variables in various cluster algebras . for simplicity denote the vertex sets @xmath191 and @xmath192 . by definition ,
the cluster algebra with coefficients , denoted @xmath193 , is generated by only those cluster variables in @xmath194 obtained from iteratively mutating the initial cluster variables @xmath188 only at vertices in @xmath195 .
( vertices in @xmath196 are called frozen vertices .
the coefficients are in @xmath197 $ ] . )
thus , there is a natural bijection sending a cluster variable in @xmath193 of the form @xmath198 $ ] to the cluster variable @xmath199\in \mathcal{a}(q')$ ] of the same expression .
there is also a natural bijection sending a cluster variable @xmath198\in\mathcal{a}(q , q')$ ] to the cluster variable @xmath198\in\mathcal{a}(q)$ ] , given by substituting @xmath179 by @xmath200 for @xmath201 $ ] .
more generally , if @xmath19 is a full subquiver of @xmath115 , and @xmath115 is a full subquiver of @xmath22 , then there is a natural bijection sending a cluster variable @xmath198\in\mathcal{a}(q , q')$ ] to @xmath198\in\mathcal{a}(q , p)$ ] given by substituting @xmath179 by @xmath200 for @xmath202 .
if @xmath19 is a full subquiver of @xmath203 , and @xmath22 is the vertex - induced subquiver of @xmath203 whose vertex set consists of vertices in @xmath5 and those adjacent to @xmath5 , then the cluster variables in @xmath193 have the same expressions as those in @xmath204 .
the `` only if '' part is easy , by checking the 3-cycle quiver .
the `` if '' part in the case @xmath205_+$ ] can be proved by the construction in [ s6.1 ] , using the fact that the diagonals in @xmath115 corresponding to the extended pipelines , therefore are non - crossing .
the `` if '' part for general @xmath157 follows from the following lemma .
in this section , we give three formulas for the cluster monomial @xmath162 $ ] for @xmath159 . all results here will be proved in 6 .
first we reduce to a special case .
we can replace @xmath157 by @xmath176_+$ ] , thus can assume @xmath206 , because of lemma [ a to a+ ] .
moreover , for any edge @xmath207 of @xmath19 that is not in a 3-cycle , we can add a vertex @xmath208 and two arrows @xmath209 and @xmath210 ( the vertex @xmath208 is a frozen vertex ) .
indeed , assume the modified quiver is @xmath22
. then by remark [ rmk2.1 ] , once we have a formula for cluster monomials for @xmath22 , we can set @xmath211 for all @xmath212 and obtain a formula for cluster monomials for @xmath19 . in the rest of the paper , we assume that @xmath205_+$ ] and @xmath213 for every 3-cycle @xmath133 , @xmath214 is a nonnegative integer by proposition [ d - vector criteria ] , and it is not hard to verify that @xmath215 . fix a deg-2 vertex @xmath216 of @xmath19 ( which exists because of ) . for @xmath35 , denote by @xmath217 be the distance ( i.e. , the length of the shortest directed path ) from @xmath216 to @xmath2 .
let @xmath218 be a 0 - 1 sequence , and define @xmath219 we say that a sequence of 0 - 1 sequences @xmath220 is a _ globally compatible sequence ( abbreviated gcs ) _ if the following holds for any 3-cycle @xmath133 : * if @xmath221 , then @xmath222 for @xmath223 ; * if @xmath224 , then @xmath225 for @xmath223 ; * if @xmath226 , then @xmath227 for @xmath223 . [ thm:01sequence ] for any * d*-vector @xmath228 ( i.e. , @xmath229 ) , we have the following formula for the corresponding cluster monomial : @xmath230=\left(\prod_{l=1}^n x_l^{-a_l}\right ) \sum_{\mathbf{s}}\left(\prod_{i } x_{i}^{e_i}\right),\;\textrm { where } e_i=\sum_{i\to j}|\,\bar{\mathbf{s}}_j\,|+\sum_{k\to i } |\mathbf{s}_k|-\sum_{i\to j\to k\to i}\sigma_{jki},\ ] ] here @xmath231 runs through all gcss .
this formula specializes to the formula given in @xcite for a linear quiver .
we recall the following definition from @xcite .
let @xmath232 be a pair of nonnegative integers .
let @xmath233 .
maximal dyck path _ of type @xmath234 , denoted by @xmath235 , is a lattice path from @xmath236 to @xmath237 that is as close as possible to the diagonal joining @xmath238 and @xmath237 , but never goes above it .
a _ corner _ is a subpath consisting of a horizontal edge followed by a vertical edge .
[ corner - first ] let @xmath239 ( resp .
@xmath240 ) be the set of horizontal ( resp .
vertical ) edges of a maximal dyck path @xmath235 .
we label @xmath241 with the _ corner - first index _ in the following sense : 1 .
edges in @xmath239 are indexed as @xmath242 such that @xmath42 is the horizontal edge of the @xmath2-th corner for @xmath243 $ ] and @xmath244 is the @xmath2-th of the remaining horizontal ones for @xmath245 $ ] , 2 .
edges in @xmath240 are indexed as @xmath246 such that @xmath28 is the vertical edge of the @xmath2-th corner for @xmath243 $ ] and @xmath247 is the @xmath2-th of the remaining vertical ones for @xmath248 $ ] .
( here we count corners from bottom left to top right , count vertical edges from bottom to top , and count horizontal edges from left to right . )
= [ font= ] ( 0,0 ) grid ( 6,4 ) ; ( 0,0 ) ( 6,4 ) ; ( p00 )
( p10 ) node[midway , above ] @xmath249 ( p10 ) ( p20 ) node[midway , above ] @xmath250 ( p20 ) ( p21 ) node[midway , right ] @xmath86 ( p21 ) ( p31 ) node[midway , above ] @xmath251 ( p31 )
( p32 ) node[midway , right ] @xmath252 ( p32 ) ( p42 ) node[midway , above ] @xmath253 ( p42 ) ( p52 ) node[midway , above ] @xmath254 ( p52 ) ( p53 ) node[midway , right ] @xmath255 ( p53 ) ( p63 ) node[midway , above ] @xmath256 ( p63 )
( p64 ) node[midway , right ] @xmath257 ; [ local_compatibility ] let @xmath258 , @xmath259 , @xmath260 .
we say that @xmath261 and @xmath262 are _
@xmath263-compatible _ if for every @xmath264 , either @xmath265 or @xmath266 .
in other words , neither of the first @xmath263 corners are in the subpath @xmath267 .
note that the following definition of global compatibility is different from @xcite .
[ definition : gcc ] let @xmath228 be a * d*-vector . for each pair @xmath207 in @xmath19 ,
let @xmath268 be a maximal dyck path @xmath269 .
we label @xmath268 with the corner - first index ( described in definition [ corner - first ] ) , whose horizontal edges are denoted @xmath270 and vertical edges are denoted by @xmath271 .
we say that the collection @xmath272 is a _ globally compatible collection ( abbreviated gcc ) _ if and only if for any @xmath273 in @xmath19 : * if @xmath209 is also an arrow in @xmath19 , then @xmath274 and @xmath275 are @xmath214-compatible , and @xmath276$};\ ] ] * otherwise , @xmath277$}.\ ] ] [ thm : gccxa ] assume @xmath278 . for any
* d*-vector @xmath228 , we have the following formula for the corresponding cluster monomial : @xmath279=\left(\prod_{l=1}^n x_l^{-a_l}\right ) \sum\left(\prod_{i\to j } x_{i}^{\big|s^{(i , j)}_2\big| } x_{j}^{\big|s^{(i , j)}_1\big|}\right)\cdot \prod_{i\to j\to k\to i } x_i^{-\sigma_{jki}},\ ] ] where the sum runs over all gccs .
( note that because of the rotational symmetry , each 3-cycle contributes three terms to the last product . )
the third method of computing the cluster monomial with given * d*-vector is to first decompose the cluster monomial into a product of cluster variables using pipelines and then use a formula for cluster variables using gccs .
fix a planar embedding of @xmath19 such that all 3-cycles are clockwise .
let @xmath169 be a graph whose edges corresponds to vertices of @xmath19 , and whose 3-cycles corresponds to 3-cycles of @xmath19 .
then @xmath169 is the subgraph of a triangularization of a polygon that consists of diagonals , and each edge @xmath2 of @xmath169 is attached a nonnegative integer @xmath146 .
next we make each triangle of @xmath169 a _ triangular pipe dream _ , in the following way : step 1 : on each edge @xmath2 , draw @xmath146 marking points to separate it into @xmath148 equal segments .
step 2 : if @xmath2 and @xmath280 are two sides of a triangle with the third side @xmath208 , then for @xmath154 , we join the two @xmath155-th marking points on edge @xmath2 and edge @xmath280 ( ordering in the increasing distance to the common endpoint of @xmath2 and @xmath280 ) by a pipe . drawing these pipes so that they are disjoint from each other .
a pipeline is a union of pipes connected consecutively .
( it may degenerate to a point . )
then the set of pipes is partitioned into disjoint pipelines .
assume that @xmath19 satisfies .
( note that a pipeline naturally corresponds to a completely linear subquiver of @xmath19 ) . for given * d*-vector
, we construct pipelines as in the subsection [ constpipelines ]
. then each pipeline corresponds to a cluster variable .
we give the gcc formula of the cluster variable @xmath162 $ ] in a completely extended linear quiver @xmath89 , where @xmath281 such that @xmath282 if @xmath283 , and @xmath284 otherwise . for any arrow @xmath285 of @xmath19
we attach a dyck path @xmath286 , which consists of one horizontal edge and one vertical edge .
( recall that @xmath287 is defined in
[ subsection : linear quivers ] . )
then definition [ definition : gcc ] specializes to the following : let @xmath288 for @xmath289 $ ] , @xmath290 $ ] .
we say that the collection @xmath291 is a _ gcc _
if @xmath292 , \textrm { and the following holds for $ i\in [ 2,n-1]$:}\ ] ] 1 . if @xmath293 , then @xmath294 ; 2 . if @xmath295 , then @xmath296 ; 3 . if @xmath297 , then @xmath298 ; 4 . if @xmath299 , then @xmath300 .
[ mainthm ] let @xmath89 be a completely extended linear quiver , and @xmath301 , @xmath302 .
define @xmath303 such that @xmath282 if @xmath283 , and @xmath284 otherwise .
then the cluster variable with * d*-vector @xmath304 is @xmath305=\left(\prod_{i=1}^n x_{i}^{-1}\right ) \sum \left(\prod_{i=0}^{\lqv } y_i \right),\ ] ] where the sum runs over all gccs @xmath291 , @xmath306 for @xmath307 $ ] , and @xmath308 [ rmk 4.7 ] note that the above theorem induces a formula for the cluster variables of any type @xmath0 quiver .
indeed , let @xmath50 be any type @xmath0 quiver and non - initial @xmath162 $ ] be a cluster variable with * d*-vector @xmath281 ( here @xmath309 ; the formulas for initial cluster variables are trivial ) .
then the subset of vertices @xmath310 is equal to the set of vertices @xmath5 of a linear full subquiver @xmath19 . by relabeling vertices if necessary ,
we assume @xmath191 ( thus @xmath311 and @xmath312 and for convenience we denote @xmath313 . by removing vertices in @xmath314 but not in or adjacent to @xmath5
, we get an extended linear quiver @xmath114 ; from this extended linear quiver we can obtain a completely extended linear quiver @xmath22 . define @xmath302 , @xmath315 .
thanks to remark [ rmk2.1 ] , a formula for cluster variable @xmath316\in \mathcal{a}(q , q')$ ] induces a formula for @xmath317\in \mathcal{a}(q , p)$ ] by substituting @xmath211 for @xmath202 , which is also a formula for @xmath162=x[1^n0^{n'-n}]\in \mathcal{a}(\tilde{q})$ ] .
( see example [ example62 ] . )
we also give a formula for cluster variables of any type @xmath0 quiver in terms of gcs as follows .
let @xmath50 be a type @xmath0 quiver and @xmath19 be a linear full subquiver of @xmath50 , as in remark [ rmk 4.7 ] .
define @xmath318 relabelling vertices if necessary , we assume that @xmath19 is @xmath319 where each arrow can go either direction .
we will give a formula of the cluster variable @xmath320 $ ] .
let @xmath321 , @xmath322 for every @xmath323 .
we say that @xmath231 is a gcs if @xmath324 for every arrow @xmath207 in @xmath6 . because @xmath325 for @xmath326 , by abuse of notation we also view @xmath231 as an element in @xmath327 .
let @xmath328 be the number of arrows in @xmath329 whose tails are in @xmath5 and heads are the vertex @xmath2 , let @xmath330 be the number of arrows in @xmath329 whose heads are in @xmath5 and tails are the vertex @xmath2 .
let @xmath331 be the set of vertices @xmath332 such that there exists a 3-cycle @xmath333 with @xmath334 ; or equivalently , @xmath335 denote @xmath336 .
we have the following formula , where arrows @xmath207 are in @xmath50 : @xmath337 = \left(\prod_{r\in q_0 } x_r^{-1}\right ) \frac{\sum_{\mathbf{s}}\left(\prod_{i\to j } x_i^{\bar{s}_j}x_j^{s_i}\right ) } { \prod_{k\in k}x_k } = \sum_{\mathbf{s } } z_{\bf s},\quad \textrm{where } z_{\bf s}=\frac{\prod_{i\to j } x_i^{\bar{s}_j}x_j^{s_i}}{\prod_{r\in q_0\cup k } x_r}\ ] ] and @xmath231 runs through all gcss .
in this section , we first recall the construction of snake diagram and the formula of cluster variables using perfect matching as in @xcite , then give a bijective proof of theorem [ mainthm ] via perfect matching .
associated to a completely extended linear quiver @xmath89 , we recursively construct the snake diagram by gluing @xmath17-tiles together as follows : we first put the @xmath338-tile to the right side of the @xmath339-tile ; suppose the @xmath340-tile is placed , we add the @xmath341-tile to the right side or on top of the @xmath340-tile such that the @xmath342 , @xmath343 and @xmath341-tiles are in the same row or column if and only if @xmath344 .
next , we label the edges as follows . * the common edge of the @xmath340-tile and the @xmath341-tile
is labeled @xmath345 . *
denote by @xmath346 the parallelogram bounded by the main diagonals of the @xmath340-tile and the @xmath341-tile and two boundary edges .
any edge forming an angle of @xmath347 with the main diagonal of the @xmath340-tile will be labeled @xmath348 ( where @xmath280 indicates the parallelogram to which the edge belongs ) . * for convenience ,
we let @xmath349 be the right triangle with legs @xmath350 and @xmath351 , and let @xmath352 the right triangle with legs @xmath353 and @xmath354 . * the edges of the first and the last tiles are labeled as in figure [ firstlasttiles ] .
= [ font= ] ( p00 )
( p10 ) node[midway , below ] @xmath355 ; ( p11 )
( p10 ) node[midway , above , sloped ] @xmath356 ; ( p11 ) ( p01 ) node[midway , above ] @xmath357 ; ( p01 ) ( p00 ) ; at ( 0.5,0.5 ) [ fill = lightgray ] @xmath358 ; ( p10 ) ( p20 ) node[midway , below ] @xmath359 ; ( p20 ) ( p21 ) node[midway , right ] @xmath360 ; ( p21 ) ( p11 ) node[midway , above ] @xmath345 ; ( p11 ) ( p20 ) ; ( p11 ) ( p20 ) ; at ( 1.5,0.5 ) [ fill = lightgray ] @xmath2 ; ( p21 ) ( p22 )
node[midway , right ] @xmath361 ; ( p22 ) ( p12 ) ; ( p12 ) ( p11 )
node[left , near start ] @xmath362 ; ( p12 ) ( p21 ) ; at ( 1.5,1.5 ) [ fill = lightgray ] @xmath363 ; at ( 1,-1 ) @xmath364 ; = [ font= ] ( p00 ) ( p10 ) ; ( p11 ) ( p10 ) node[midway , above , sloped ] @xmath356 ; ( p11 ) ( p01 ) node[midway , above ] @xmath357 ; ( p01 ) ( p00 ) ; at ( 0.5,0.5 ) [ fill = lightgray ] @xmath358 ; ( p10 ) ( p20 ) node[midway , below ] @xmath359 ; ( p21 ) ( p20 ) node[midway , above , sloped ] @xmath345 ; ( p21 ) ( p11 ) node[midway , above ] @xmath360 ; ( p11 ) ( p20 ) ; at ( 1.5,0.5 ) [ fill = lightgray ] @xmath2 ; ( p20 ) ( p30 ) node[midway , below ] @xmath362 ; ( p30 ) ( p31 ) ; ( p31 ) ( p21 ) ; ( p21 ) ( p30 ) ; at ( 2.5,0.5 ) [ fill = lightgray ] @xmath363 ; at ( 1.5,-1 ) @xmath365 ; = [ font= ] ( p00 ) ( p10 ) node[midway , below=2pt ] @xmath366 ; ( p10 ) ( p11 ) node[midway , right ] @xmath367 ; ( p11 ) ( p01 ) node[midway , above ] @xmath368 ; ( p01 ) ( p00 ) node[midway , left ] @xmath369 ; at ( 0.5,0.5 ) [ fill = lightgray ] @xmath200 ; at ( 0.5,-1 ) first tile ; at ( 6.5,-1 ) last tile ; at ( 6.5,0.5 ) or ; = [ font= ] ( p00 ) ( p10 ) node[midway , below=2pt ] @xmath370 ; ( p10 ) ( p11 ) node[midway , right ] @xmath371 ; ( p11 ) ( p01 ) node[midway , above ] @xmath372 ; ( p01 ) ( p00 ) node[midway , left ] @xmath373 ; at ( 0.5,0.5 ) [ fill = lightgray ] @xmath374 ; = [ font= ] ( p00 ) ( p10 ) node[midway , below=2pt ] @xmath373 ; ( p10 ) ( p11 ) node[midway , right ] @xmath372 ; ( p11 ) ( p01 ) node[midway , above ] @xmath371 ; ( p01 ) ( p00 )
node[midway , left ] @xmath370 ; at ( 0.5,0.5 ) [ fill = lightgray ] @xmath374 ; a perfect matching of the snake diagram is a set of edges such that each vertex is incident to exactly one edge in the set .
= [ font= ] ( p00 ) ( p10 ) ; ( p10 ) ( p11 ) ; ( p11 ) ( p01 ) ; ( p01 ) ( p00 ) ; ( p10 ) ( p20 ) ; ( p20 ) ( p21 ) ; ( p21 ) ( p11 ) ; ( p21 )
( p22 ) ; ( p22 ) ( p12 ) ; ( p12 ) ( p11 ) ; for the fixed completely extended linear quiver @xmath89 , let @xmath375 be the set of all perfect matchings in the associated snake diagram , and @xmath376 be the set of all gccs .
we shall construction a bijective map @xmath377 and its inverse @xmath378 .
first we prove a simple lemma . [ inthesameperfectmatching ] for any perfect matching @xmath14 and @xmath379 $ ] , there is exactly one edge of @xmath14 lies in @xmath346 .
the statement is obviously true for @xmath380 and @xmath381 .
suppose the statement is false for some @xmath382 $ ] .
since @xmath14 is a perfect matching , we are in one of the following two cases .
case 1 : both @xmath383 are in @xmath14 .
if we remove @xmath346 ( 4 vertices and 3 edges ) , then the rest graph has two components which have odd number of vertices and have perfect matchings .
this is a contradiction .
case 2 : none of the three edges @xmath384 lies in @xmath14 .
then we remove the three edges ( but do not remove the vertices ) and apply the same argument as in case 1 .
thanks to the above lemma , we can write a perfect matching @xmath14 as @xmath385 where @xmath386
. \(i ) we define a map @xmath387 by sending @xmath388 to @xmath389 such that for @xmath390 $ ] , @xmath391 ( by lemma [ inthesameperfectmatching ] , exactly one of the three cases occurs . )
\(ii ) we define a map @xmath392 by sending @xmath389 to the set of edges @xmath393 such that @xmath394 for @xmath390 $ ] , and @xmath395 @xmath396 we assign a weight @xmath397 for each edge @xmath24 of the snake diagram as follows for all @xmath398 : @xmath399 for a perfect matching @xmath400 , define its weight @xmath401 .
in @xcite it is proved that the cluster variable with * d*-vector @xmath157 is @xmath402=\left(\prod_{i=1}^n x_{i}^{-1}\right ) \sum_\gamma w(\gamma).\ ] ] [ pmsgccs ] the maps @xmath403 and @xmath404 are well - defined and are inverses of each other .
moreover , @xmath405 , thus @xmath403 induces a bijective proof of theorem [ mainthm ] using .
\(i ) we show that @xmath403 is well - defined , that is , @xmath406 satisfies the condition .
it s clear from the construction that for every @xmath382 $ ] , @xmath407 .
next , we prove ( a ) and ( c ) of , since ( b ) and ( d ) can be proved similarly . for ( a ) , we suppose @xmath293 and need to show that @xmath408 .
this is true because the two edges @xmath409 and @xmath410 are incident to the same deg-2 vertex , thus exactly one of them is in @xmath14 .
( see the left diagram in figure [ labels ] . ) for ( c ) , we suppose @xmath411 and need to show that @xmath412 .
these two edges are opposite edges of a tile which is the middle of three tiles in a row or a column .
deleting these two edges will separate the snake diagram into two graphs with even number of vertices each .
thus the two edges must be both in @xmath14 or not in @xmath14 .
( see the right diagram in figure [ labels ] . )
\(ii ) we show that @xmath404 is well - defined , that is , @xmath413 is a perfect matching .
since the snake diagram has @xmath414 vertices and @xmath14 has @xmath415 edges , it suffices to show that all edges in @xmath14 are disjoint .
we assume the contrary that @xmath416 shares a vertex with @xmath417 for some @xmath418 . since @xmath419 and @xmath420 , @xmath421 and @xmath422 much be consecutive , thus @xmath423 .
we first assume @xmath424 .
we shall discuss two cases @xmath425 and @xmath426 , and omit the other two cases @xmath427 and @xmath428 since the proof is similar .
case @xmath425 : since @xmath291 is a gcc , we must have @xmath429 , @xmath430 , @xmath431 , or @xmath432 . correspondingly , @xmath433 it is obvious from figure [ cc+1c+2 ] that @xmath416 and @xmath434 are disjoint , a contradiction as expected .
= [ font= ] ( p00 ) ( p10 ) ; ( p11 ) ( p10 ) node[midway , above , sloped ] @xmath435 ; ( p11 ) ( p01 )
node[midway , above ] @xmath436 ; ( p01 )
( p00 ) ; at ( 0.5,0.5 ) [ fill = lightgray ] @xmath437 ; ( p10 ) ( p20 ) node[midway , below ] @xmath438 ; ( p20 ) ( p21 )
node[midway , right ] @xmath439 ; ( p21 ) ( p11 ) node[midway , above ] @xmath440 ; at ( 1.5,0.5 ) [ fill = lightgray ] @xmath441 ; ( p21 )
( p22 ) ; ( p22 ) ( p12 ) ; ( p12 ) ( p11 ) node[near start , left ] @xmath442 ; at ( 1.5,1.5 ) [ fill = lightgray ] @xmath443 ; = [ font= ] ( p00 ) ( p10 ) ; ( p11 )
( p10 ) node[midway , above , sloped ] @xmath435 ; ( p11 ) ( p01 ) node[midway , above ] @xmath436 ; ( p01 ) ( p00 ) ; at ( 0.5,0.5 ) [ fill = lightgray ] @xmath437 ; ( p10 ) ( p20 ) node[midway , below ] @xmath438 ; ( p21 ) ( p20 ) node[midway , above , sloped ] @xmath440 ; ( p21 ) ( p11 ) node[midway , above ] @xmath439 ; at ( 1.5,0.5 ) [ fill = lightgray ] @xmath441 ; ( p20 ) ( p30 ) node[midway , below ] @xmath442 ; ( p30 ) ( p31 ) ; ( p31 ) ( p21 ) ; at ( 2.5,0.5 ) [ fill = lightgray ] @xmath443 ; case @xmath444 : similar as the above case , @xmath445 we get the expected contradiction by observing figure [ cc+1c+2 ] .
the cases of @xmath446 and @xmath447 are proved by a similar argument .
\(iii ) the fact that @xmath403 and @xmath404 are inverse of each other , and @xmath405 , follows easily from their definitions . if @xmath448 then @xmath449 as shown in figure [ exmapmg ] .
= [ font= ] ( p00 ) ( p10 ) node[midway , below ] @xmath450 ; ( p10 )
( p11 ) node[midway , right ] @xmath451 ; ( p11 )
( p01 ) node[midway , above ] @xmath368 ; ( p01 ) ( p00 )
node[midway , left ] @xmath452 ; ( p10 )
( p20 ) node[midway , below ] @xmath453 ; ( p20 )
( p21 ) node[midway , right ] @xmath454 ; ( p21 )
( p11 ) node[midway , above ] @xmath455 ; ( p21 ) ( p22 ) node[midway , right ] @xmath456 ; ( p22 ) ( p12 ) node[midway , above ] @xmath457 ; ( p12 ) ( p11 ) node[midway , left , near start ] @xmath458 ; ( p22 ) ( p23 )
node[midway , right ] @xmath459 ; ( p23 )
( p13 ) node[midway , above ] @xmath460 ; ( p13 ) ( p12 )
node[midway , left ] @xmath461 ; ( v1 ) at ( 0:0 ) ; ( v2 ) at ( 0:2 ) ; ( v1 )
( v2 ) node[midway , above , font= ] @xmath403 ; = [ font= ] ( p20 ) ( p30 ) node[midway , below ] @xmath462 ; ( p30 )
( p31 ) node[midway , right ] @xmath463 ; ( p40 )
( p50 ) node[midway , below ] @xmath464 ; ( p50 )
( p51 ) node[midway , right ] @xmath462 ; ( p60 )
( p61 ) node[midway , right ] @xmath465 ; ( p60 ) ( p70 ) node[midway , below ] @xmath464 ;
we have explained in [ subsection3.3 ] that , in order to compute cluster variables , it suffices to have the formula for a completely extended linear quiver , namely theorem [ mainthm ] .
this theorem follows from theorem [ pmsgccs ] . in this section ,
we show how to derive theorem [ thm:01sequence ] and theorem [ thm : gccxa ] from theorem [ mainthm ] .
let @xmath466 $ ] be the right hand side of the formula in theorem [ thm : gccxa ] .
we shall show that ( i ) the gccs for the * d*-vector @xmath157 are in one - to - one correspondence with the collections of gccs for the * d*-vectors @xmath180 s described in ; ( ii ) @xmath466=\prod_{\bf b } x'[{\bf b}]$ ] ; and ( iii ) @xmath467=x[{\bf b}]$ ] using theorem [ mainthm ] .
it then follows that @xmath466=x[{\bf a}]$ ] .
\(i ) let @xmath468 be any gcc for the * d*-vector @xmath157 . for each pipeline @xmath163 , let @xmath469 , we construct a gcc @xmath470 for the * d*-vector @xmath180 by requiring the following for each arrow @xmath207 : if @xmath163 intersects the edge @xmath2 at the @xmath155-th marking point , then @xmath471 if and only if @xmath472 , if @xmath163 intersects the edge @xmath280 at the @xmath155-th marking point , then @xmath473 if and only if @xmath474 , ( in both case the marking points are ordered in the increasing distance from the common endpoint of @xmath2 and @xmath280 ) . to verify that @xmath470 is a gcc for the * d*-vector @xmath180 , we need to check the conditions in definition [ definition : gcc ] .
the only nontrivial condition to check is that for a 3-cycle @xmath333 , @xmath475 and @xmath476 are @xmath477-compatible .
that reduces to showing that @xmath478 in the case @xmath479 . in this case
, @xmath163 intersects edges @xmath2 and @xmath280 at the @xmath155-th marking points for some @xmath480 . then either @xmath481 or @xmath482 . in the former case , @xmath483 ; in the latter case , @xmath484 .
therefore @xmath485 .
it is easy to see that a unique gcc @xmath468 is determined if we take any collection of gccs @xmath486 for all pipelines @xmath163 .
so we have the desired one - to - one correspondence .
\(ii ) we show that @xmath487=x'[{\bf a}]$ ] .
since @xmath488 , it suffices to show that , for each gcc @xmath468 , letting @xmath470 be defined as in ( i ) , the following holds ( recall that @xmath469 depends on @xmath163 ) : @xmath489 since the left hand side is equal to @xmath490 it suffices to show that @xmath491 , @xmath492 , and @xmath493 .
the first two are clear . to show the last equality : first note that if @xmath163 is disjoint from the edge @xmath280 , then @xmath494 and thus @xmath495 .
so we only need to consider those @xmath163 that intersect @xmath280 .
let @xmath496 ( @xmath497 ) be the pipeline that intersects @xmath280 at the @xmath155-th marking point ( ordered in the increasing distance to the common endpoint of @xmath280 and @xmath208 ) .
if @xmath498 , then @xmath499 , thus @xmath500 ; otherwise , either @xmath494 or @xmath501 , thus @xmath502 . therefore @xmath493 .
\(iii ) we show that @xmath467=x[{\bf b}]$ ] . by remark [ rmk2.1 ]
, it suffices to show that , in the setting of theorem [ mainthm ] , the right hand side of is equal to @xmath466 $ ] .
it breaks down to show that , for @xmath503 $ ] , the following equality holds for the @xmath504-th 3-cycle @xmath133 in @xmath22 ( for @xmath505 $ ] , the @xmath504-th 3-cycle is the one that contains vertices @xmath28 , @xmath79 and @xmath506 ; the @xmath507-th 3-cycle is @xmath508 ; the @xmath17-th 3-cycle is @xmath509 ) : @xmath510 we shall only prove the case when @xmath511 $ ] and @xmath512 , because other cases can be proved in a similar way . in this case , the @xmath504-th 3-cycle is @xmath513 ( where @xmath514 ) , and the left hand side of is equal to @xmath515 so holds .
we give a bijection between gcss and gccs .
let @xmath231 be a gcs . consider a 3-cycle @xmath133 , labeled in the way that @xmath226 .
then we define @xmath516 it is then easy to check that the conditions of gcss and gccs , as well as the two theorems , are equivalent under this bijection .
this section is devoted to the construction of a bijection between gcss and broken lines in the even rank case ( theorem [ mainthm2 ] ) and the general case ( theorem [ mainthm2 ] ) . in
[ subsection : g - vectors ] , we give a description of the g - vector of a cluster variable , which will determine the initial direction of the broken lines . in
[ subsection : scattering diagrams and broken lines ] , we recall the necessary facts on scattering diagrams , broken lines , and theta functions . the rest of the section is to state and prove theorem [ mainthm2 ] and [ mainthm2 ] . in this section , we let @xmath50 be a type @xmath0 quiver and @xmath19 be a linear full subquiver of @xmath50 .
relabelling vertices if necessary , we assume that @xmath19 is @xmath517 it is shown in ( * ? ? ? * theorem 7.5 ( 4 ) ) that , if @xmath122 is the g - vector of a cluster variable , then the cluster variable is exactly @xmath518 .
so we first study g - vectors .
[ lem : g ] let @xmath304 be defined as in .
then the g - vector @xmath519 of the cluster variable @xmath320 $ ] is given by @xmath520 we use the description of g - vectors in @xcite : @xmath521 it is easy to verify that @xmath522 corresponds to the gcs @xmath523 where @xmath524 if and only if @xmath283 ( this is the unique gcs satisfying @xmath525 ) .
thus @xmath526 meanwhile , @xmath527 .
so we can compute the g - vector case by case : if @xmath528 : the power of @xmath529 in is equal to the number of arrows @xmath530 with @xmath531 , which is @xmath532 . since @xmath533 , @xmath534 . if @xmath535 and @xmath536 : then @xmath537 , thus the power of @xmath529 in is @xmath538 . if @xmath535 and @xmath539 : then @xmath540 or @xmath238 .
if it is @xmath428 , then @xmath541 , thus the power of @xmath529 in is @xmath542 ; if it is @xmath427 or @xmath238 , then @xmath543 and @xmath537 , thus @xmath529 does not appear in @xmath544 . in all cases ,
we get @xmath545 .
the above is equivalent to the following description of the g - vector of @xmath320 $ ] : @xmath546 we only recall the necessary facts needed in our paper , specialized in our setting . for more reference ,
see @xcite .
recall that in
[ subsection : quiver def ] we defined @xmath547 for a coefficient free cluster algebra @xmath548 of rank @xmath17 , and @xmath49 and @xmath549 for a principal coefficient cluster algebra @xmath46 of rank @xmath17 .
let @xmath550 , @xmath551 with a basis @xmath552 , @xmath553 with dual basis @xmath554 .
@xmath555 is equipped with a skew - symmetric form @xmath556 .
let @xmath557 be a sublattice of @xmath555 with basis @xmath558 .
define @xmath559 we assume the fundamental assumption : the map @xmath560 given by @xmath561 is injective .
choose a strictly convex top - dimensional cone @xmath562 , with associated monoid @xmath563 , such that @xmath564 for all @xmath55 .
let @xmath565}$ ] be the completion of @xmath566 $ ] at the maximal monomial ideal @xmath567 generated by @xmath568 . a _ wall _ is a pair @xmath569 where @xmath570 is a @xmath571-dimensional convex rational polyhedral cone contained in @xmath572 for some primitive vector @xmath573 , and @xmath574 .
a _ scattering diagram _
@xmath575 is a collection of walls such that there are only finite many walls @xmath576 satisfying @xmath577 for each @xmath578 .
for a smooth path @xmath579\to
m_{\mathbb{r}}\setminus{\rm sing}(\mathfrak{d})$ ] whose endpoints are not in @xmath580 and that it is transversal to each wall that it crosses , we define an automorphism @xmath581 of @xmath565}$ ] as follows : for each @xmath578 , we can find numbers @xmath582 such that @xmath583 , with @xmath584 , @xmath585 , and @xmath586 if @xmath587 , and @xmath263 taken as large as possible .
for each @xmath2 , define an automorphism @xmath588 to be @xmath589 where @xmath590 is the primitive vector annihilating @xmath591 and satisfying @xmath592 .
define @xmath593 a _ consistent scattering diagram _
@xmath575 is a scattering diagram if @xmath581 only depends on the endpoints of @xmath14 .
two scattering diagrams @xmath575 , @xmath594 are _ equivalent _ if @xmath595 for all paths @xmath14 for which both sides are defined .
let @xmath596 be given by @xmath561 . a wall @xmath597 is _ incoming _ if @xmath598 , where @xmath599 and @xmath573 ; otherwise , the wall is _ outgoing_. define the initial scattering diagram @xmath600 note that all walls in @xmath601 are incoming .
[ def of d ] let @xmath602 be a consistent scattering diagram such that @xmath603 consists only of outgoing walls .
( it is proved in ( * ? ? ?
* theorem 1.7 ) that @xmath575 exists and is unique up to equivalence . )
let @xmath575 be as defined in definition [ def of d ] , @xmath604 and @xmath605 .
a _ broken line _ for @xmath606 with endpoint @xmath607 is a piecewise linear continuous proper path @xmath608
\to \mathbb{r}^n\setminus{\rm sing } ( \mathfrak{d})$ ] with a finite number of domains of linearity .
this path comes along with the data of , for each domain of linearity @xmath609 $ ] of @xmath14 , a monomial @xmath610 $ ] .
this data satisfies the following properties : 1 .
if @xmath612 is the first ( and therefore unbounded ) domain of linearity of @xmath14 , then @xmath613 .
3 . for @xmath8 in a domain of linearity @xmath612 , @xmath614 .
4 . let @xmath615 be a point at which @xmath14 is not linear , passing from domain of linearity @xmath612 to @xmath616 .
let @xmath617 .
then @xmath618 is a term in the formal power series @xmath619 we denote the monomial attached to the final domain of linearity of @xmath14 by @xmath620 . for given @xmath604 and @xmath605 , the _ theta function _
is defined to be @xmath621 where @xmath14 runs over all broken lines for @xmath606 with endpoint @xmath607 .
[ conditions not needed ] the broken lines that we shall construct in this paper will not bend on walls that are not in the initial scattering diagram @xmath601 defined in .
a priori , there may exist broken lines that do bend on walls in @xmath603 ( i.e. , outgoing walls ) ; but we shall argue that there are no such broken lines in our setting ( where we only consider those appear in the theta function corresponding to cluster variables ( see [ subsection : bijective ] ) . in general , broken lines can bend on outgoing walls if we consider those appear in the theta function corresponding to cluster monomials . in this subsection , we give a bijection between gcss and broken lines for the cluster variable @xmath320 $ ] when @xmath622 ( the rank of the cluster algebra @xmath548 ) is even .
this is exactly the case when the exchange matrix @xmath623 is of full rank , which guarantees that the fundamental assumption in @xcite is satisfied .
( indeed , since ( * ? ? ?
* lemma 3.2 ) asserts that the rank is invariant under mutation , it is suffices to consider the linear quiver @xmath624 , in which case the determinant of @xmath623 is 1 if @xmath622 is even , 0 if @xmath622 is odd . )
[ df : adjustableposition ] for a given gcs @xmath231 , we say that @xmath625 $ ] is a _
adjustable _ position if \(a ) the @xmath155-th coordinate of @xmath231 is @xmath507 , and \(b ) the sequence obtained from @xmath231 by replacing the @xmath155-th coordinate by @xmath200 is still a gcs .
[ rmk : adjustableposition ] a adjustable position always exists as long as some coordinate of @xmath626 is 0 .
this follows from the following equivalent definition : let @xmath627 be the full subquiver of @xmath19 with vertex set @xmath628 . then @xmath155 is a adjustable position if and only if @xmath155 is a sink of @xmath627 . since @xmath19 is acyclic , @xmath627 is also acyclic , thus @xmath627 has at least one sink .
[ df : wisi ] for a given gcs @xmath231 , let @xmath629 .
define @xmath630 , and define @xmath631 ( @xmath632 ) and @xmath633 ( @xmath634 ) backward recursively as follows .
assume @xmath635 is defined for some @xmath2 satisfying @xmath632 .
define @xmath631 to be the smallest adjustable position , and the corresponding new gcs by @xmath636 .
[ df : g ] ( i ) define a function @xmath637 sending @xmath231 to @xmath638 where @xmath639 where @xmath640 is the number of arrows @xmath641 in @xmath642 such that @xmath643 and @xmath644 , and @xmath645 is the number of arrows @xmath646 in @xmath642 such that @xmath643 and @xmath647 .
\(ii ) define @xmath648 for @xmath649 .
[ rmk : mi-1to1 ] ( 1 ) note that @xmath650\in\{0,1\}^n$ ] .
we claim that @xmath519 coincides with the definition given in lemma [ lem : g ] .
indeed , note that @xmath651 .
if @xmath652 , nothing needs to be proved .
we assume @xmath653 in the rest of the paragraph . if @xmath536 , then we are in the second case of , @xmath654 ; if @xmath655 or @xmath427 , then we are in the second case of , thus @xmath656 ; if @xmath657 , then we are in the first case of , thus @xmath658 .
\(2 ) it is easy to check that every coordinate of @xmath659 satisfies @xmath660 .
indeed , @xmath661 is obvious ; to check @xmath662 , note that every vertex @xmath155 is adjacent to at most two vertices in @xmath5 , so we only need to show that in the second case of , it is impossible to have @xmath663 . but
this equality implies that @xmath155 is adjacent to at least two vertices in @xmath5 ; it follows from the description of type a quivers that @xmath155 and two vertices in @xmath5 form a 3-cycle , which contradicts the condition of the second case .
the following main theorem gives a bijective construction of @xmath664 $ ] .
[ mainthm2 ] assume the rank @xmath622 of the cluster algebra is even , and @xmath665 such that @xmath666 let @xmath606 be defined as in lemma [ lem : g ] .
then there is a bijection @xmath667 between the set of gcs and the set of broken lines for @xmath606 with endpoint @xmath607 , such that each gcs @xmath231 is sent to a broken line @xmath668 satisfying @xmath669 ( defined in ): using notation in definition [ df : wisi ] , the broken line @xmath14 can be explicitly described as follows : \(i ) it has @xmath670 domains of linearity @xmath671 ( where @xmath672 is unbounded ) ; \(ii ) it bends from the domain of linearity @xmath673 to @xmath674 at a point on the wall @xmath675 ; \(iii ) @xmath676 for @xmath677 ; \(iv ) the monomial attached to @xmath678 is @xmath679 .
the setting of @xcite in the type a setting is specialized as follows : the lattice @xmath680 with a basis @xmath681 and with a skew - symmetric bilinear form @xmath682 satisfying @xmath683 .
( since type a is skew - symmetric , all the multipliers @xmath684 as noted in ( * ? ? ?
* p114 ) ) .
the dual lattice @xmath685 has a basis @xmath686 dual to @xmath681 .
the vector @xmath93 defined in ( * ? ? ?
* p29 ) is @xmath687 : check @xmath688 contains a term @xmath689 , for @xmath632 .
( here @xmath690 is the primitive vector annihilating the tangent space to @xmath675 and that @xmath691 is positive . ) since @xmath692 , and by remark [ rmk : mi-1to1](2 ) , all coordinates of @xmath693 take value in @xmath694 , we must have @xmath695 .
meanwhile , by the initial scattering diagram described ( * ? ? ?
* p31 ) , @xmath696 .
so @xmath697 and we are left to show that @xmath698 , or equivalently @xmath699 , or equivalently @xmath700 where @xmath701 , @xmath702 . since @xmath635 and @xmath636 only differ in the @xmath631-th coordinate ( where the former has coordinate @xmath507 and the latter has coordinate @xmath200 ) , we have @xmath703 then can be shown case by case : ( 1 ) if @xmath704 , then @xmath705 or @xmath155 is not adjacent to @xmath631 , in either situation we have @xmath706 , thus holds ; ( 2 ) if @xmath707 : then @xmath708 , @xmath709 , and holds ; ( 3 ) if @xmath710 , then @xmath711 , @xmath712 and still holds . : denoting by @xmath713 the point where @xmath14 bends from the domain of linearity @xmath673 to @xmath674 , check that @xmath714 for @xmath715 ( assume @xmath716 ) .
note that @xmath717 are determined by the following conditions : @xmath718 we see that the coordinates of @xmath719 can be expressed as linear combinations of @xmath720 ( by treating @xmath720 as variables ) . for @xmath721 in @xmath722 and @xmath723 ,
we denote @xmath724 the purpose of introducing this definition is : if @xmath725 , then @xmath726 under the condition .
so step 2 is done once we prove the following lemma .
[ qi - qi+1 ] ( i ) the @xmath631-th coordinate of @xmath727 is @xmath728 .
\(ii ) @xmath729 .
\(iii ) @xmath730 for @xmath731 . as a consequence , @xmath732 .
\(i ) by , @xmath733 so to show @xmath734 , it is equivalent to show that the above degrees are both 0 . since @xmath735 , we must have @xmath736 for any arrow @xmath737 , otherwise it contradicts the assumption that @xmath635 is a gcs .
thus @xmath738 .
similarly , since @xmath739 , we must have @xmath740 for any arrow @xmath741 , otherwise it contradicts the assumption that @xmath636 is a gcs .
thus @xmath742 .
since @xmath636 and @xmath635 only differ in the @xmath631-th coordinate , we have @xmath743 .
\(ii ) assume that @xmath744 .
to determine @xmath745 , it suffices to consider the @xmath631-th coordinate on both sides : @xmath746 then ( ii ) holds since @xmath747 .
\(iii ) we prove it by fixing @xmath2 and using downward induction on @xmath504 . for @xmath748 , @xmath749 ,
hence the statement holds . for @xmath750 , using ( ii ) we have @xmath751 we argue in two cases : case 1 : @xmath752 .
then @xmath753 , thus @xmath754 .
then @xmath730 follows from @xmath755 .
case 2 : @xmath756 .
we shall show that @xmath757 ( which implies @xmath758 ) .
it suffices to show that the two degrees in the following expression are @xmath507 and @xmath200 , respectively : @xmath759 since @xmath760 is gcs and @xmath761 by the construction of @xmath760 , there is no arrow @xmath737 in @xmath19 satisfying @xmath762 .
thus @xmath763 .
next , we prove @xmath764 by contradiction : if @xmath765 , then there is no arrow @xmath741 in @xmath19 satisfying @xmath766 .
but then @xmath631 is also a adjustable position in @xmath760 , which contradicts the property of being `` smallest '' in the definition of @xmath767 ( definition [ df : wisi ] ) .
if @xmath768 , then by our assumption on @xmath19 ( see ) , there must be two arrows @xmath769 and @xmath770 in @xmath19 satisfying @xmath771 .
consider the longest path starting from @xmath631 and going left : @xmath772 in @xmath773 ( defined in remark [ rmk : adjustableposition ] ) .
then @xmath280 is a sink in @xmath773 , hence adjustable and satisfying @xmath774 , again contradicts the property of being `` smallest '' in the definition of @xmath767 .
it is easy to see that @xmath667 injective , since different @xmath775 determine different sets of @xmath631 ( i.e. , the sets of walls where the broken line bend ) , thus determine different broken lines . thus to show the bijectivity of @xmath667 , it suffices to show that the number of gcc is not less than the number of broken lines . to show the latter , we use ( *
* theorem 7.5 ( 4 ) ) which asserts that the cluster variable @xmath320 $ ] is equal to the theta function @xmath776 .
since the number of gcs is equal to @xmath320|_{x_1=\cdots = x_{n'}=1}$ ] , on the other hand each broken line contributes at least 1 to @xmath320|_{x_1=\cdots = x_{n'}=1}$ ] , we conclude that the number of gcs is not less than the number of broken lines .
therefore @xmath667 is bijective .
this completes the proof of theorem [ mainthm2 ] .
note that the above argument implies that each broken line that contributes to @xmath776 will not bend on any outgoing walls . to extend theorem [ mainthm2 ] to odd ranks ,
we need to consider principal coefficients .
we denote by @xmath777 $ ] the cluster variable in @xmath46 that corresponds to the cluster variable @xmath162 $ ] in @xmath548 .
[ lemma : g - vector principal ] let @xmath548 be a ( coefficient - free ) cluster algebra of rank @xmath17 , and @xmath46 the corresponding cluster algebra with principal coefficients .
let @xmath162 $ ] be a cluster variable in @xmath548 with g - vector @xmath778 .
then the corresponding cluster variable @xmath777 $ ] in @xmath46 has g - vector @xmath779 recall that the g - vector of @xmath162\in\mathcal{a}$ ] is the multidegree of the corresponding cluster variable in principal coefficients .
more precisely , start with matrix @xmath780 , and for each mutation @xmath781 , @xmath782 and @xmath783 are related by the rule @xmath784_+ , & \textrm { otherwise . } \end{cases}\ ] ] the new cluster variable is determined by @xmath785_+}+\prod_{i=1}^{2n}x_{i;t}^{[-b_{ik}^t]_+}.\ ] ] each cluster variable is homogeneous with the assignment that , for @xmath59 , @xmath786 , @xmath787 , where @xmath788 is the @xmath2-th column of @xmath623 .
this multidegree is the g - vector of the cluster variable .
the g - vector of @xmath777\in\mathcal{a}_{\rm prin}$ ] is defined similarly as above , with @xmath623 and @xmath789 being replaced by @xmath790 using , it can be proved by a simple induction that , @xmath791 ( in fact , @xmath792 if the sign - coherence conjecture is true ( * ? ? ?
* conjecture 8.8 ) ; in particular , it is true if @xmath623 is skew - symmetric , but we do not need this fact in this paper . ) using , it can be proved by a simple induction that @xmath777 $ ] can be obtained from @xmath162 $ ] by the substitution @xmath793 and @xmath794 , for @xmath59 . since @xmath795 the multidegree ( i.e. , the g - vector ) of @xmath777 $ ] must be @xmath796 where @xmath797 is the multidegree ( i.e. , the g - vector ) of @xmath162 $ ] . in the rest
we show that theorem [ mainthm2 ] can be adapted to @xmath46 .
[ mainthm2 ] assume @xmath798 such that @xmath799 ( there is no condition on @xmath800 . )
let @xmath801 where @xmath606 is defined in lemma [ lem : g ] .
then there is a bijection @xmath667 between the set of gcs and the set of broken lines for @xmath802 with endpoints @xmath803 , such that each gcs @xmath231 is sent to a broken line @xmath668 satisfying @xmath804 ( defined in ): using notation in definition [ df : wisi ] , the broken line @xmath14 can be explicitly described as follows : \(i ) it has @xmath670 domains of linearity @xmath671 ( @xmath672 is unbounded ) ; \(ii ) it bends from the domain of linearity @xmath673 to @xmath674 at a point on the wall @xmath675 ; \(iii ) @xmath805 for @xmath677 , where @xmath806 .
\(iv ) the monomial attached to @xmath678 is @xmath807 .
since @xmath808 is full rank , the fundamental assumption in @xcite is satisfied , thus the proof in [ section : proof ] works with little change ; we point out the nontrivial revision change . because of lemma [ lemma : g - vector principal ] , the @xmath809-vector of @xmath777 $ ] is indeed @xmath802 .
thus @xmath777 $ ] is equal to the theta function @xmath810 by ( * ? ? ?
* theorem 7.5 ( 4 ) ) . in step 1 , @xmath690 is replaced by @xmath811 . since the first @xmath622 coordinates of @xmath812 form the vector @xmath693 , we have @xmath813 . the function attached to the wall
@xmath591 is @xmath814 where @xmath815 .
thus @xmath816 in step 2 , since @xmath817 , we have @xmath818 with the same @xmath819 as in lemma [ qi - qi+1 ] .
in this section , we give several examples to illustrate the computation of cluster variables and cluster monomials using methods introduced in previous sections .
we compute @xmath820 $ ] , the cluster monomial with * d*-vector @xmath821 for the quiver @xmath822 ( a 3-cycle ) . using formula : choose @xmath823 .
then @xmath824 , @xmath825 , @xmath826 , @xmath827 .
a gcs @xmath828 satisfies @xmath829 a tedious computation using gives @xmath820=x_1^{-2}x_2^{-2}x_3^{-2}(x_1+x_2+x_3)^3 $ ] .
( but at least it is easy to see that there are 27 terms ; indeed , since each pair has 3 choices @xmath830 , the total number of gcss is @xmath831 . ) using formula : there are three dyck paths of size @xmath832 corresponding to the three arrows @xmath833 , @xmath834 , @xmath835 : = [ font= ] ( 0,0 ) grid ( 2,2 ) ; ( 0,0 ) ( 1,0 ) node[midway , above ] @xmath250 ( 1,0 ) ( 1,1 ) node[midway , right ] @xmath86 ( 1,1 ) ( 2,1 ) node[midway , above ] @xmath251 ( 2,1 ) ( 2,2 ) node[midway , right ] @xmath252 ; ( 0,0 ) grid ( 2,2 ) ; ( 0,0 ) ( 1,0 ) node[midway , above ] ( 1,0 ) ( 1,1 ) node[midway , right ] @xmath463 ( 1,1 ) ( 2,1 ) node[midway , above ] @xmath462 ( 2,1 ) ( 2,2 ) node[midway , right ] @xmath463 ; ( 1,0.1 ) ( 1,.9 ) ( 1.1,1 ) ( 1.9,1 ) ( 2,1.1 ) ( 2,1.9 ) ; ( 0,0 ) grid ( 2,2 ) ; ( 0,0 ) ( 1,0 ) node[midway ] ( 1,0 ) ( 1,1 ) node[midway , right ] @xmath462 ( 1,1 ) ( 2,1 ) node[midway ] ( 2,1 )
( 2,2 ) node[midway ] ; ( 1,0 )
( 1,1 ) ; ( 0,0 ) grid ( 2,2 ) ; ( 0,0 ) ( 1,0 ) node[midway , above ] @xmath463 ( 1,0 )
( 1,1 ) node[midway ] ( 1,1 )
( 2,1 ) node[midway ] ( 2,1 )
( 2,2 ) node[midway , right ] @xmath464 ; ( 0,0 ) ( 1,0 ) ( 2,1 ) ( 2,2 ) ; such that ( i ) we do not choose both @xmath250 and @xmath86 in each dyck path , and ( ii ) we choose @xmath836 in the @xmath2-th dyck path if and only if we do not choose @xmath837 in the @xmath838-th dyck path for @xmath839 ( by convention , the 4th dyck path is the 1st one ) . in the example of gcc in figure [ fig 9 ] ( right ) , the corresponding product @xmath840 computing all possible gccs gives @xmath820=x_1^{-2}x_2^{-2}x_3^{-2}(x_1+x_2+x_3)^3 $ ] .
using formula : first observe that there are 3 pipelines as shown in figure [ pipelines ] . according to
[ subsection3.3 ] , @xmath820=x[1,1,0]x[0,1,1]x[1,0,1]$ ] .
now we compute @xmath841 $ ] using theorem [ mainthm ] .
the pair @xmath842 can be @xmath843 .
correspondingly , we have @xmath844 . then @xmath845&=(x_1^{-1}x_2^{-1})\sum y_0y_1y_2=(x_1^{-1}x_2^{-1})(x_1x_{1,0}x_{2,0 } + x_2x_{1,1}x_{2,1 } + x_{1,2}x_{1,1}x_{2,0})\\ & = ( x_1^{-1}x_2^{-1})(x_1 + x_2 + x_3 ) \endaligned\ ] ] where the last equality is obtained by substituting @xmath846 , and @xmath847
. similarly , @xmath848=(x_2^{-1}x_3^{-1})(x_1 + x_2 + x_3)$ ] , @xmath849=(x_1^{-1}x_3^{-1})(x_1 + x_2 + x_3)$ ] .
thus @xmath820=x_1^{-2}x_2^{-2}x_3^{-2}(x_1+x_2+x_3)^3 $ ] .
[ example62 ] we compute some cluster variables of @xmath850 where @xmath851 is the following type @xmath0 quiver : [ scale=0.8 ] ( v2 ) at ( 0:1 ) 2 ; ( v1 ) at ( 120:1 ) 1 ; ( v5 ) at ( 240:1 ) 5 ; ( 30:1.732)++(1,0 ) node ( v6 ) 6 ; ( -30:1.732)++(1,0 ) node ( v3 ) 3 ; ( 0:1.732)++(2.5,-0.866 ) node ( v4 ) 4 ; ( 30:1.732)++(2.732,0 ) node ( v7 ) 7 ; ( v1 ) to ( v2 ) ; ( v2 ) to ( v5 ) ; ( v5 ) to ( v1 ) ; ( v2 ) to ( v6 ) ; ( v6 ) to ( v3 ) ; ( v3 ) to ( v2 ) ; ( v3 ) to ( v4 ) ; ( v6 ) to ( v7 ) ; as observed in remark [ rmk 4.7 ] , the set of non - initial cluster variables is in one - to - one correspondence with the set of * d*-vectors @xmath852 , where @xmath853 and @xmath310 is the vertex set of a linear full subquiver of @xmath851 .
so we can compute all cluster variables using theorem [ mainthm ] .
for example , we consider the * d*-vector @xmath854 .
then the subset of vertices @xmath855 is equal to the set of vertices @xmath5 of the full linear subquiver @xmath856 .
it is a subquiver of an extended linear subquiver @xmath115 and after completing @xmath115 , we get a completely extended linear quiver @xmath22 as shown in figure [ qq ] . lll [ scale=0.8 ] ( v1 ) at ( 0:1 ) [ red ] 1 ; ( -60:1.732)++(1,0 ) node ( v5 ) 5 ; ( 0:1.732)++(1,0 ) node ( v2 ) [ red ] 2 ; ( 0:1.732)++(2.732,0 ) node ( v3 ) [ red ] 3 ; ( 60:1.732)++(2.732,0 ) node ( v6 ) 6 ; ( -30:1.732)++(4.464,0 ) node ( v4 ) 4 ; ( v1 ) to ( v2 ) ; ( v2 ) to ( v5 ) ; ( v5 ) to ( v1 ) ; ( v3 ) to ( v2 ) ; ( v2 ) to ( v6 ) ; ( v6 ) to ( v3 ) ; ( v3 ) to ( v4 ) ; & ( v1 ) at ( 0:0 ) ; ( v2 ) at ( 0:2 ) ; ( v3 ) at ( -60:1.732 ) ; ( v1 ) to ( v2 ) ; & [ scale=0.8 ] ( v1 ) at ( 0:1 ) [ red ] 1 ; ( v8 ) at ( 120:1 ) [ blue ] 8 ; ( v9 ) at ( 240:1 ) [ blue ] 9 ; ( -60:1.732)++(1,0 ) node ( v5 ) 5 ; ( 0:1.732)++(1,0 ) node ( v2 ) [ red ] 2 ; ( 0:1.732)++(2.732,0 ) node ( v3 ) [ red ] 3 ; ( 60:1.732)++(2.732,0 ) node ( v6 ) 6 ; ( -30:1.732)++(4.464,0 ) node ( v4 ) 4 ; ( 30:1.732)++(4.464,0 ) node ( v10 ) [ blue ] 10 ; ( v1 ) to ( v8 ) ; ( v8 ) to ( v9 ) ; ( v9 ) to ( v1 ) ; ( v1 ) to ( v2 ) ; ( v2 ) to ( v5 ) ; ( v5 ) to ( v1 ) ; ( v3 ) to ( v2 ) ; ( v2 ) to ( v6 ) ; ( v6 ) to ( v3 ) ; ( v3 ) to ( v4 ) ; ( v4 ) to ( v10 ) ; ( v10 ) to ( v3 ) ; in @xmath115 , we have @xmath857 , @xmath858 , @xmath859 .
two 3-cycles are added and the new vertices are @xmath860 , @xmath861 and @xmath862 .
all gccs are described as follows .
lllll @xmath863 & = & l ( p00 ) ( p10 )
node[midway , below ] @xmath462 ; ( p10 ) ( p11 ) node[midway , left ] @xmath463 ; ( p20 )
( p21 ) node[midway , right ] @xmath464 ; ( p20 )
( p30 ) node[midway , below ] @xmath462 ; & , & @xmath864 + lllll @xmath865 & = & l ( p00 ) ( p10 )
node[midway , below ] @xmath462 ; ( p10 )
( p11 ) node[midway , left ] @xmath463 ; ( p20 )
( p21 ) node[midway , right ] @xmath464 ; ( p20 )
( p30 ) node[midway , below ] @xmath462 ; & , & @xmath866 + lllll @xmath867 & = & l ( p00 )
( p10 ) node[midway , below ] @xmath462 ; ( p10 )
( p11 ) node[midway , left ] @xmath463 ; ( p20 ) ( p21 ) node[midway , right ] @xmath464 ; ( p20 )
( p30 ) node[midway , below ] @xmath462 ; & , & @xmath868 + lllll @xmath869 & = & l ( p00 ) ( p10 ) node[midway , below ] @xmath462 ; ( p10 ) ( p11 )
node[midway , left ] @xmath463 ; ( p20 )
( p21 ) node[midway , right ] @xmath464 ; ( p20 ) ( p30 ) node[midway , below ] @xmath462 ; & , & @xmath870 + lllll @xmath871 & = & l ( p00 ) ( p10 ) node[midway , below ] @xmath462 ; ( p10 )
( p11 ) node[midway , left ] @xmath463 ; ( p20 )
( p21 ) node[midway , right ] @xmath464 ; ( p20 )
( p30 ) node[midway , below ] @xmath462 ; & , & @xmath872 the cluster variable with @xmath873-vector @xmath874 is @xmath875 = \sum_{i=1}^5 x(\beta_i ) = \dfrac{x_5x_6x_9x_{10}+x_2x_4x_5x_9+x_2x_6x_8x_{10}+x_2 ^ 2x_4x_8+x_1x_3x_9x_{10}}{x_1x_2x_3}.\ ] ] setting @xmath876 , we get the following cluster variable of @xmath850 : @xmath875=\dfrac{x_5x_6+x_2x_4x_5+x_2x_6+x_2 ^ 2x_4+x_1x_3}{x_1x_2x_3}.\ ] ] the table below shows some @xmath873-vectors and their corresponding cluster variables of @xmath850 .
clccl ( 1,0,0,0,0,0,0 ) & @xmath877 & & ( 1,1,0,0,0,0,0 ) & @xmath878 + ( 0,1,0,0,0,0,0 ) & @xmath879 & & ( 0,1,1,0,0,0,0 ) & @xmath880 + ( 0,0,1,0,0,0,0 ) & @xmath881 & & ( 0,0,1,1,0,0,0 ) & @xmath882 + ( 0,0,0,1,0,0,0 ) & @xmath883 & & ( 1,0,0,0,1,0,0 ) & @xmath884 + & & & ( 0,0,1,0,0,1,0 ) & @xmath885 + + cl ( 1,1,1,0,0,0,0 ) & @xmath886 + ( 0,1,1,0,1,0,0 ) & @xmath887 + ( 0,1,1,1,0,0,0 ) & @xmath888 + ( 0,0,1,1,0,1,0 ) & @xmath889 + ( 1,1,1,1,0,0,0 ) & @xmath890 here we give an example to illustrate statements in [ s7 ] .
let @xmath22 be the following type a quiver : [ scale=0.8 ] ( v1 ) at ( 0:1 ) 1 ; ( 60:1.732)++(1,0 ) node ( v4 ) 4 ; ( 0:1.732)++(1,0 ) node ( v2 ) 2 ; ( 0:1.732)++(2.732,0 ) node ( v3 ) 3 ; ( v2 ) to ( v1 ) ; ( v1 ) to ( v4 ) ; ( v4 ) to ( v2 ) ; ( v2 ) to ( v3 ) ; let @xmath19 be the full linear subquiver @xmath891 , and correspondingly @xmath892 .
there are 5 gcs @xmath893 : @xmath894 by lemma [ lem : g ] , @xmath895 .
consider @xmath896 . by definition [ df : wisi ] , @xmath897 , @xmath898 by definition [ df : g ] ,
@xmath606 is as above , @xmath899 the broken line @xmath668 described in theorem [ mainthm2 ] has 4 domain of linearity @xmath900 , bends on walls @xmath901 in that order .
the direction vector of @xmath14 on @xmath678 is @xmath902 for @xmath903 .
the monomial attached to @xmath900 are @xmath904 , @xmath905 , @xmath906 , @xmath907 , respectively .
thus @xmath908 .
follow the proof in step 2 of [ section : proof ] , we compute the coordinates of @xmath719 as follows : @xmath909 @xmath910 since @xmath911 , it is clear in this example that @xmath912 for all @xmath913 .
in this section , we first recall the construction of @xmath1-paths and the formula of cluster variables using @xmath1-paths as in @xcite , then give a bijective proof of theorem [ mainthm ] via @xmath1-paths .
let @xmath115 be a convex polygon with @xmath914 vertices .
a diagonal of @xmath115 is a line segment connecting two non - adjacent vertices .
two diagonals are said to be crossing if they intersect in the interior of @xmath115 .
a triangulation @xmath1 of @xmath115 is a maximal set of non - crossing diagonals together with the boundary edges of @xmath115 .
any triangulation has @xmath17 diagonals and @xmath915 boundary edges .
= [ font= ] at ( 0,0 ) ( a ) ; ( a.corner 1 ) ( a.corner 2 ) node[midway , left=3pt ] @xmath917 ; ( a.corner 2 ) ( a.corner 3 ) node[midway , left ] @xmath918 ; ( a.corner 3 )
( a.corner 4 ) node[midway , left ] @xmath919 ; ( a.corner 4 ) ( a.corner 5 ) node[midway , below ] @xmath920 ; ( a.corner 5 )
( a.corner 6 ) node[midway , right=2pt ] @xmath921 ; ( a.corner 6 ) ( a.corner 7 ) node[midway , right ] @xmath922 ; ( a.corner 7 )
( a.corner 1 ) node[midway , right=3pt ] @xmath923 ; ( a.corner 7 ) ( a.corner 2 ) node[midway , above ] @xmath924 ; ( a.corner 7 ) ( a.corner 3 ) node[midway , above ] @xmath925 ; ( a.corner 7 ) ( a.corner 4 )
node[midway , left ] @xmath926 ; ( a.corner 4 )
( a.corner 6 ) node[midway , above ] @xmath927 ; the process of constructing the initial triangulation starts with placing the diagonal @xmath924 .
we obtain @xmath350 by rotating @xmath924 in the counterclockwise direction .
the edge @xmath351 is obtained by rotating @xmath924 in the clockwise direction .
let @xmath928 be the common vertex of @xmath350 and @xmath351 .
suppose that the diagonal @xmath147 ( @xmath390 $ ] ) is drawn .
the diagonal @xmath929 is obtained by rotating @xmath147 in the counterclockwise direction if @xmath82 , in the clockwise direction if @xmath83 . the boundary edge between @xmath147 and @xmath929
is labeled @xmath345 . when @xmath930 then @xmath931 is the boundary edge clockwise from @xmath932 and @xmath933 is the boundary edge counterclockwise from @xmath932 .
denote the common vertex of @xmath933 and @xmath931 by @xmath934 .
we can view both the snake diagram and the triangulation @xmath1 as graphs .
then there is a natural graph homomorphism @xmath936 between them that sends an edge of the snake diagram to an edge of the triangulation as follows : @xmath937 in @xcite , fomin and zelevinsky showed that the cluster variables of @xmath45 are in bijection with the diagonals of the polygon @xmath115 where the initial cluster variables @xmath938 corresponds to @xmath939 .
[ deftpath ] @xcite a @xmath1-path @xmath9 from @xmath928 to @xmath934 is the sequence @xmath944 such that 1 ) @xmath945 are vertices of @xmath115 .
2 ) @xmath946 such that @xmath947 connects the vertices @xmath948 and @xmath949 for each @xmath950 .
3 ) @xmath951 if @xmath952 .
4 ) @xmath953 is odd .
5 ) @xmath947 crosses @xmath940 if @xmath208 is even .
6 ) if @xmath954 and both @xmath955 and @xmath947 cross @xmath685 then @xmath956 is closer to @xmath928 than @xmath957 is to @xmath928 .
let @xmath958 be the set of all @xmath1-path from @xmath928 to @xmath934 .
for any @xmath959 , let @xmath960 let @xmath961 such that @xmath282 if and only if @xmath283 .
the following formula of the cluster variable @xmath162 $ ] is proved in @xcite : @xmath962=\sum_{\alpha\in\p } x(\alpha)\ ] ] we define a map @xmath963 by sending @xmath389 to the @xmath1-path @xmath9 obtained by first constructing a path @xmath964 from @xmath928 to @xmath934 where @xmath965 for @xmath966 $ ] , @xmath967 for @xmath390 $ ] , and \(i ) we first define @xmath976 , which is exactly the folding map in @xcite . as defined in @xcite ,
a complete @xmath1-path @xmath9 from @xmath928 to @xmath934 is similar to a @xmath1-path from @xmath928 to @xmath934 defined in definition [ deftpath ] , in the sense that we require 1 ) , 2 ) , and where we use the order @xmath981 note that we do not require edges in @xmath9 to be distinct .
it is easy to see that a complete @xmath1-path has length @xmath982 . for simplicity ,
we denote @xmath9 using its edge sequence . for @xmath388 , we define ( recall that @xmath936 is defined in ): @xmath983$,\ ; $ l_{2j+1}=p(\gamma_j)$ for $ j\in [ 0,n]$.}\ ] ] note that the starting point of each @xmath678 is determined by @xmath984 .
the union of a perfect matching @xmath14 with all the diagonals of tiles form a path @xmath985 in the snake diagram . if we consider the quotient map from the snake diagram to the triangulation of @xmath115 , by identifying the diagonal edge @xmath2 with @xmath986 and identifying diagonal edge @xmath363 with @xmath410 , then the image of @xmath985 is the complete @xmath1-path @xmath987 .
\(ii ) we show that @xmath976 has a well - defined inverse map @xmath988 ( which is the unfolding map in @xcite ) , thus @xmath976 is bijective .
indeed , @xmath988 sends a complete @xmath1-path @xmath989 to @xmath990 , where @xmath991 is the unique edge in @xmath992 , that is , @xmath993 , @xmath994 , and @xmath995 ( resp .
@xmath996 , @xmath997 ) if @xmath998 is @xmath151 ( resp .
@xmath999 , @xmath997 ) for @xmath1000 $ ] . for @xmath1002 $ ] with @xmath1003 , @xmath991 and @xmath1004
are disjoint if @xmath1005 because @xmath1006 and @xmath1007 are disjoint .
so we assume @xmath1008 .
we shall only discuss the case @xmath1009 since other cases can be proved similarly .
\(iii ) we show that @xmath1021 is bijective by giving its inverse @xmath1022 .
suppose that @xmath1023 is a @xmath1-path from @xmath928 to @xmath934 .
if @xmath1024 , then @xmath9 is already a complete @xmath1-path , so we define @xmath1025 .
now assume @xmath278 .
the sequence @xmath1026 is obtained as a result of the following algorithm .
1 . initialize @xmath1027 .
let @xmath280 run from @xmath200 to @xmath17 : if @xmath1028 , then insert @xmath1029 to @xmath612 so that the resulting @xmath612 is nondecreasing with the order given in .
3 . define @xmath1030 .
finally , we show that @xmath1031 . by the construction of @xmath1032 in ( iii ) , @xmath1033 remains unchanged if we replace @xmath9 by the complete @xmath1-path @xmath1034 ; this is because each time we insert the pair @xmath1029 , the extra contribution to the product is @xmath1035 .
so it suffices to show @xmath1036 by 5 ) , @xmath1037 , so it suffices to show @xmath1038 , or to show that @xmath1039 for @xmath1040 $ ] .
indeed , @xmath1041 by , thus @xmath1042 by the definition of the weight @xmath25 in .
moreover , theorem [ pmsgccs ] asserts that @xmath1043 .
thus @xmath1039 . at ( 0,0 ) ( a ) ; ( a.corner 1 ) ( a.corner 2 ) node[midway , above , inner sep=3pt ] @xmath450 ; ( a.corner 2 ) ( a.corner 3 )
node[midway , left ] @xmath451 ; ( a.corner 3 )
( a.corner 4 ) node[midway , left ] @xmath455 ; ( a.corner 4 )
( a.corner 5 ) node[midway , below ] @xmath460 ; ( a.corner 5 )
( a.corner 6 ) node[midway , right=2pt ] @xmath459 ; ( a.corner 6 )
( a.corner 7 ) node[midway , right ] @xmath457 ; ( a.corner 7 )
( a.corner 1 ) node[midway , above ] @xmath452 ; ( a.corner 7 ) ( a.corner 2 ) node[midway , above=0pt ] @xmath924 ; ( a.corner 7 ) ( a.corner 3 ) node[midway , above ] @xmath925 ; ( a.corner 7 ) ( a.corner 4 )
node[midway , left ] @xmath926 ; ( a.corner 4 )
( a.corner 6 ) node[midway , above , inner sep=3pt ] @xmath927 ; 99 d. e. bayleran , d.j .
finnigan , a. haj ali , k. lee , c.m .
locricchio , m.r .
mills , d.p - p .
stiefel , t.m .
tran and r. urbaniuk , a new combinatorial formula for cluster monomials of equioriented type @xmath0 quivers , preprint 2014 , available at http://www.math.unl.edu/~klee24/cluster_a_public.pdf .
e. gunawan and g. musiker , @xmath1-path formula and atomic bases for cluster algebras of type @xmath127 , _ symmetry , integrability and geometry : methods and applications ( sigma ) _ * 11 * ( 2015 ) , 060 , 46 pages . | lots of research focuses on the combinatorics behind various bases of cluster algebras .
this paper studies the natural basis of a type @xmath0 cluster algebra , which consists of all cluster monomials .
we introduce a new kind of combinatorial formulas for the cluster monomials in terms of the so - called globally compatible collections .
we give bijective proofs of these formulas by comparing with the well - known combinatorial models of the @xmath1-paths and of the perfect matchings in a snake diagram . for cluster variables of a type @xmath0 cluster algebra
, we give a bijection that relates our new formula with the theta functions constructed by gross , hacking , keel and kontsevich . |
where the lower part of the classical instability strip intersects the main sequence , three distinct classes of multiperiodically pulsating variables can be found . the @xmath9 doradus stars pulsate in gravity modes of high radial order and have periods of the order of one day ( kaye et al .
the @xmath5 scuti stars have periods of the order of a few hours ( breger 1979 ) and are thus pressure and gravity mode pulsators of low radial order .
the fastest pulsations in this domain in the hr diagram are however excited in the rapidly oscillating ap ( roap ) stars ( kurtz 1982 , kurtz & martinez 2000 ) , with typical periods around 10 minutes , indicating pressure modes of high radial overtones .
the photometric semi - amplitudes associated with these pulsations are in most cases only a few mmag , which makes them difficult to detect .
the roap stars are also remarkable due to their spectral peculiarities , since they are pulsating representatives of cool magnetic ap stars of the srcreu subtype .
the magnetic fields of ap stars cause elemental segregation on the stellar surface . in other words ,
the chemical elements are arranged in patches on the stellar surface ( see , e.g. , kochukhov et al .
2004 or lueftinger et al .
2003 for well - documented examples ) .
this causes modulation of the mean apparent brightness of the star with the rotation period .
as the chemical elements show some alignment with the magnetic poles , the rotational light curves show a single - wave structure if only one magnetic pole is seen and a double - wave variation if both magnetic poles come into view during a rotation cycle . the strong magnetic fields present in the roap stars are usually not aligned with the stellar rotation axis . since the pulsation axis does however coincide with the magnetic axis , the roap stars are oblique pulsators ( kurtz 1982 ) .
this means that the pulsation modes excited in the roap stars are seen at different aspects during the rotation cycle , which , for nonradial modes , causes amplitude modulation over the rotation period ( which can then be inferred ) and allows us to put constraints on the geometry of the pulsator . the oblique pulsator model ( opm ) , in its simplest form
, then predicts that ( except for special geometric orientations of the axes ) dipole ( @xmath2 ) pulsation modes will be split into equally spaced triplets .
quadrupole ( @xmath10 ) modes will give rise to equally spaced frequency quintuplets , where the spacing of consecutive multiplet components is exactly the stellar rotational frequency . the magnetic field of the roap stars affects the pulsations in two additional ways .
firstly , the pulsation frequencies are shifted with respect to their unperturbed value ( cunha & gough 2000 ) and secondly , the pulsation modes are distorted so that a single spherical harmonic can no longer describe them fully ( e.g. , kurtz , kanaan & martinez 1993 , takata & shibahashi 1995 ) .
this is observationally manifested by the presence of additional multiplet components surrounding the first - order singlets , triplets or quintuplets , which are spaced by integer multiples of the rotation frequency . for a long time , the predominant observing method for studying the pulsations of roap stars was time - resolved high - speed photometry , yielding interesting results on the pulsational behaviour , geometry and asteroseismology of these stars ( e.g. , see matthews , kurtz & martinez 1999 ) .
the most recent observational advances however came from time - resolved spectroscopy : since the vertical wavelengths of the pulsation modes are comparable to the size of the line - forming regions in the atmospheres of these stars , the vertical structure of the atmospheres can be resolved ( ryabchikova et al .
2002 , kurtz , elkin & mathys 2003 ) .
spectroscopy is also more sensitive to the detection of roap star pulsations than photometry . due to the vertical stratification of chemical elements in their atmospheres ( e.g. ryabchikova et al .
2002 ) the pulsational radial velocity amplitudes of some spectral lines ( most notably rare earth elements ) can reach several kilometres per second .
the most extreme example of such high radial velocity amplitudes is hd 99563 ( xy crt ) .
this star was photometrically discovered to pulsate by dorokhova & dorokhov ( 1998 ) and was confirmed by handler & paunzen ( 1999 ) .
elkin , kurtz & mathys ( 2005 ) discovered pulsational radial velocity variations with semi - amplitudes of up 5 for some euii and tmii lines in their time - resolved spectroscopy of this star .
such high amplitudes are capable of yielding information on the structure of the atmospheres of ap stars with the best possible signal to noise .
however , one important piece of information that spectroscopy can not supply at this point is detailed knowledge of the stellar pulsation spectrum .
the reason is that the largest telescopes are necessary for obtaining spectra of the required time resolution and signal to noise .
however , observing time on these telescopes is sparse .
therefore , lengthy photometric measurements of these stars on small telescopes are still required to decipher the pulsational spectra fully . to this end , we included hd 99563 as a secondary target in a multi - site campaign originally devoted to the roap star hd 122970 ( handler et al.2002 ) to be observed at times when the latter star was not yet accessible . however , hd 99563 turned out to be quite interesting , which is why we continued to observe it after the original campaign was finished .
we also acquired some mean - light observations of hd 99563 in an attempt to determine its rotation period . in this paper , we report the results of these measurements .
our multi - site time - series photometric observations were obtained with seven different telescopes at four observatories : the 0.75 m t6 automatic photometric telescope at fairborn observatory ( fapt ) in arizona , the 0.5 m , 0.75 m , 1.0 m and 1.9 m telescopes at the south african astronomical observatory ( saao ) , the 0.6 m reflector at siding spring observatory ( sso ) in australia and the 0.9 m telescope at observatorio de sierra nevada ( osn ) in spain . whereas the latter observations consisted of strmgren @xmath11 photometry , the other telescopes used the b filter only .
the measurements are summarised in table 1 .
.journal of the observations of hd 99563 ; @xmath12 is the number of 40-second data bins obtained in the respective night and @xmath13 is the rms scatter of the residual light curves after prewhitening and low - frequency filtering ; the remaining columns are self - explanatory . [ cols="<,^,^,^,^,^,^ " , ] from the values in table 5 it can be seen that the mode of hd 99563 is dominated by the dipole component , which is reasonable as the magnetic fields in ap stars are predominantly dipoles . the contribution of the @xmath4 and @xmath10 terms to the mode are fairly small , but the @xmath3 component does have some influence .
table 6 shows that the observations are quite well reproduced by our fit
. it would be interesting to see whether the theory by saio & gautschy ( 2004 ) is capable of reproducing our results .
we can now check how well our model reproduces the observed pulsational amplitudes and phases over the stellar rotation cycle . to this end
, we subdivided the time series into pieces some @xmath14 pulsation cycles long and determined the amplitudes and phases for the 1557.653 @xmath0hz variation within these subsets .
the spherical harmonic decomposition method by kurtz ( 1992 ) yields a fit to the amplitude / phase behaviour over the rotation cycle , and we show it together with the data in fig . 6 .
we can see that both magnetic ( pulsation ) poles are actually in view for approximately the same amount of time ; the relative fractions are 53 and 47 per cent .
this is also the reason why it was so difficult to determine the rotation period of the star ; the pulsations as well as the mean light variations are almost symmetrical along the rotation period .
the fitted curves reproduce the observations fairly well , with one exception : the pulsation amplitude is overestimated for the pole that is for a shorter time in the line of sight , whereas the amplitude is underestimated for the other pole .
the harmonic frequencies can not be made responsible for this , because they do not affect the first - order amplitudes analysed here .
consequently , we believe that the poor fits near the pulsation amplitude maxima could be due to presently undetected additional multiplet components of the pulsation mode ; the amount of over- and under - fitting can be fully explained with the noise level in our residual amplitude spectrum .
our photometric multisite observations of the roap star hd 99563 resulted in the detection of a frequency quintuplet that is due to a single distorted dipole pulsation mode .
the splitting within this frequency quintuplet together with our mean light observations and published magnetic measurements allowed us to determine the stellar rotation period as @xmath1 d. within the errors , the mean light extrema occur in phase with pulsation amplitude maximum , suggesting that the abundance spots on hd 99563 should be fairly concentric around its magnetic poles . to our knowledge
, hd 99563 is only the fourth roap star where both magnetic poles become visible throughout a rotation cycle ( the others are hr 3831 , e.g. , see kurtz et al .
( 1993 ) , hd 6532 , e.g. , kurtz et al.(1996 ) , and hd 80316 e.g. , kurtz et al .
( 1997 ) ) , and only the third whose pulsational mode distortion has been quantified . in this context
it is interesting to note that the observational features of hd 99563 are in many aspects similar to another roap star , the well - studied hr 3831 ( e.g. , kurtz et al .
1993 , kochukhov 2004 ) : their geometrical orientations , rotation period and effective temperatures are alike .
only the pulsation period of hd 99563 is some 10 per cent shorter , and the phases its first - order combination frequencies are more consistent with them being harmonics .
it would thus be very interesting to compare these two objects in detail .
this would however require more in - depth studies of hd 99563 .
its basic parameters are still poorly known ; the radius we inferred is based on the assumption that the visual companion is physical , which needs to be checked .
if hd 99563 a and b were a physical pair , then the separation of the components is too wide for motion around a common centre of mass to be determined in a reasonable period of time .
however , common proper motion may be detectable within a few years .
the most efficient way to pin down the radius of the roap star would perhaps come from a combination of magnetic and polarisation measurements ( landolfi et al .
1997 ) . with that , the inclination @xmath15 and the magnetic obliquity @xmath16 can be constrained , and given the accurate rotation period and @xmath17 already available , a fairly accurate radius could be obtained .
there are several other possibilities that make hd 99563 interesting for future observations . as already argued by elkin et al .
( 2005 ) , the star is a very attractive target for pulsational radial velocity measurements .
given the high amplitude of the mean light variations of the star and its favourable geometry , doppler imaging of its surface should also be within reach .
finally , we have evidence that we have not yet deciphered the full pulsational content of the star s light curves .
another multi - site campaign , aiming at obtaining @xmath18 h of observations on 1-metre - class telescopes would therefore also be justifiable .
as we pointed out , hd 99563b is located within the @xmath5 scuti instability strip and should therefore also be tested for pulsations , with more suitable means than ours .
the austrian fonds zur frderung der wissenschaftlichen forschung partially supported this work under grants s7303-ast and s7304-ast .
we are grateful to don kurtz for supplying results prior to publication , for helpful discussions and for comments on a draft version of this paper .
we also thank hiromoto shibahashi for his accurate referee s report . | we undertook a time - series photometric multi - site campaign for the rapidly oscillating ap star hd 99563 and also acquired mean light observations over two seasons .
the pulsations of the star , that show flatter light maxima than minima , can be described with a frequency quintuplet centred on 1557.653 @xmath0hz and some first harmonics of these .
the amplitude of the pulsation is modulated with the rotation period of the star that we determine with @xmath1 d from the analysis of the stellar pulsation spectrum and of the mean light data .
we break the distorted oscillation mode up into its pure spherical harmonic components and find it is dominated by the @xmath2 pulsation , and also has a notable @xmath3 contribution , with weak @xmath4 and 2 components .
the geometrical configuration of the star allows one to see both pulsation poles for about the same amount of time ; hd 99563 is only the fourth roap star for which both pulsation poles are seen and only the third where the distortion of the pulsation modes was modelled .
we point out that hd 99563 is very similar to the well - studied roap star hr 3831 .
finally , we note that the visual companion of hd 99563 is located in the @xmath5 scuti instability strip and may thus show pulsation .
we show that if the companion was physical , the roap star would be a 2.03 @xmath6 object , seen at a rotational inclination of @xmath7 , which then predicts a magnetic obliquity @xmath8 .
stars : variables : other stars : oscillations techniques : photometric stars : individual : hd 99563 stars : individual : xy crt |
let us consider the initial - boundary value problem for the korteweg - de vries ( kdv ) equation with a decaying initial datum @xmath0 existence and uniqueness of real - valued , classical solutions can be proved via the inverse scattering transform , introduced by green , gardner , kruskal and miura in their seminal work @xcite .
the long time behavior of these last ones has been extensively investigated in the literature ( @xcite@xcite@xcite@xcite@xcite@xcite@xcite@xcite ) .
the solutions are known to eventually decompose into a certain number of solitons , travelling to the right , plus a radiation part , propagating to the left . in this paper
we wish to consider the so - called _ soliton region _
, formed by those points of the @xmath1-plane satisfying @xmath2 , for some fixed constant @xmath3 . in order to detail more about existing results ,
let us recall that the solutions of ( [ cp ] ) are uniquely individuated by the scattering data of the operator @xmath4 associated with the initial datum .
these last ones consist of a finite number of eigenvalues , @xmath5 , with @xmath6 , of the corresponding norming constants @xmath7 , and of the reflection coefficient @xmath8 .
the long time asymptotics of solutions of ( [ cp ] ) in the soliton region reads as follows @xmath9 here the phase - shifts are given by @xmath10}.\end{aligned}\ ] ] the term @xmath11 in ( [ nuova_sol ] ) is know to be small for large @xmath12 , its magnitude depending on the smoothness and decay properties of @xmath13 .
this formula was established by hirota @xcite , tanaka @xcite and wadati and toda @xcite independently , for vanishing reflection coefficient @xmath14 .
the general case was first treated by tanaka @xcite and shabat ( @xcite ) .
more recently , grunert and teschl proved such asymptotic behavior for initial data with lower regularity @xcite .
their approach relies on the steepest descent analysis of riemann - hilbert problem [ cesare ] ( see next section ) via a decomposition of the nonanalytic reflection coefficient @xmath14 into an analytic approximant and a small rest . in this paper
we examine the same riemann - hilbert problem via the modern dbar method , introduced by miller and mclaughlin in @xcite and @xcite . in particular , we establish a better estimate of @xmath15 for a larger class of initial data . our main result is the following [ hauptsatz ] let the reflection coefficient @xmath14 associated to the initial datum @xmath16 belong to @xmath17 , for some integer @xmath18 .
assume that @xmath14 and its first @xmath19 derivatives tend zero at @xmath20 .
moreover , let @xmath21 belong to the wiener algebra on the real line .
that is , assume that this last one is the image , via fourier transform , of some function in @xmath22 .
fix @xmath3 .
then there exists a constant @xmath23 such that @xmath24 for all @xmath12 sufficiently large and @xmath25 .
notice that these hypotheses are satisfied by all initial data admitting @xmath26 moments ( @xcite ) : @xmath27 theorem [ hauptsatz ] is achieved via a careful treatment of the imaginary part of the phase @xmath28 defined in ( [ def_phi ] ) .
after the standard procedure , the jump ( [ salto ] ) of the riemann - hilbert problem [ cesare ] across the real axis is decomposed and displaced partly below and partly above it .
accordingly , @xmath28 develops a non - vanishing real part , providing a decay of the decomposed jumps towards the identity matrix .
the novelty here is that subsequently we reconsider the oscillations originating from the imaginary part of @xmath28 . from their analysis ,
we extract additional information about the decay of the error term @xmath29 , corresponding exactly to our improvement of the estimates . to our best knowledge ,
the idea of this last step is new in the literature .
+ further advantages of our approach are the following .
first of all , our analysis requires less sophisticated technical means , employing basically calculus at an undergraduate level and the van der corput lemma . using them we provide simpler and more explicit expressions for @xmath29 ( see formulas ( [ ciliegio ] ) , ( [ saldini ] ) and the following ones in section [ nuova_ossessione ] ) .
these ones can be easily employed for a more detailed , long time asymptotic expansion including higher order corrections .
moreover , they might turn out to be useful for the analysis of analogue riemann - hilbert problems beyond the framework of integrability .
this a current research interest of ours .
let us fix an integer @xmath18 and a constant @xmath3 .
we assume @xmath25 for the remaining part of the paper and prove theorem [ hauptsatz ] . the hypotheses on the reflection coefficient
are also understood to hold , without further recalling them .
analogously to @xcite , we produce the solution of kdv corresponding to the initial datum @xmath13 - or equivalently , to the associated scattering data @xmath30 , @xmath31 and @xmath32 - via the following [ cesare ] find a function @xmath33 meromorphic away from the real axis , with simple poles at @xmath34 , satisfying : i _ `` jump condition '' . for every @xmath35 one
has @xmath36 where @xmath37 `` residue condition '' @xmath38 for @xmath39 .
`` symmetry condition '' @xmath40 `` normalization condition '' @xmath41 here the phase is given by @xmath42 the solution of the system ( [ cp ] ) at an arbitrary time @xmath43 is then recovered by the formula @xmath44 the right - hand side being computed from the expansion @xmath45
our proof of theorem [ hauptsatz ] consists in subsequent reformulations of the riemann - hilbert problem , till obtaining a convenient , equivalent integral equation defined on the plane . in this section
we remove the jump of @xmath46 across the real axis , exchanging it for some non - analytic behaviour on a strip around this last one .
let us fix the parameter @xmath47 we define the following regions @xmath48 with @xmath49 and @xmath50 we will indicate the corresponding reflected strips w.r.t .
the real axis ( see figure [ figura_striscie ] ) .
let us also fix the notation @xmath51 for the remaining part of the paper .
we wish to consider the following non - analytic extension of the reflection coefficient @xmath14 to the whole upper - half plane : @xmath52}\cdot \chi{\left(}\frac{v}{\delta } { \right)}.\end{aligned}\ ] ] here @xmath53 is chosen as follows is by no means unique .
any other smooth function with analogue `` cut - off '' properties would do . ]
@xmath54 } } & 1\leq v < 2\\ 0 & v \geq 2 \end{array } \right.\end{aligned}\ ] ] notice that @xmath55 vanishes on @xmath56
. moreover , there exists a constant @xmath57 such that @xmath58 where @xmath59 these are the main properties motivating our choice of such extension . using ( [ def_r ] ) , one can decompose the jump matrix @xmath60 as follows @xmath61 here @xmath62 and @xmath63 mimicking the classical nonlinear steepest descent method ( @xcite,@xcite ) , we introduce @xmath64 riemann - hilbert problem [ cesare ] for @xmath46 is then equivalent to the following [ dbar_problem_per_m_tilde ] find a two dimensional , vector - valued function @xmath65 continuous on @xmath66 and differentiable with continuity as a function of two real variables away from the real axis , such that * i. * : : `` @xmath67-condition '' @xmath68 in particular , @xmath69 is holomorphic on @xmath70 * ii . * : : `` residue condition '' .
the vector - valued function @xmath69 has simple poles at @xmath71 , where it satisfies @xmath72 for @xmath73 .
* iii . * : : `` symmetry condition '' @xmath74 * iv . *
: : `` normalization condition '' @xmath75 [ remark_claudia ] the solution @xmath69 of the meromorphic @xmath67-problem needs actually to be differentiable also on the real axis , although possibly not with continuity .
this is easily deduced , using ( [ diseg_r ] ) , from the representation @xmath76 this is nothing else than the generalization of the cauchy integral formula for smooth functions @xcite . here
@xmath77 is understood to be a small , compact rectangle containing the point @xmath78 , which for our purposes is chosen on the real line .
our next goal is to remove the poles from vector @xmath69 . to this purpose ,
we introduce in this section a model riemann - hilbert problem .
an explicit expression of its solution wo nt be necessary for our analysis , but only some of its elementary properties concerning regularity and asymptotic behavior .
these last ones are provided by proposition [ roseto ] .
find a two times two matrix valued meromorphic function @xmath79 whose only poles are simple and lie in @xmath80 , satisfying : ii _ : : `` residue condition '' @xmath81 for @xmath39 .
iii _ : : `` symmetry condition '' @xmath82 iv _ : : `` normalization condition '' @xmath83 for some constant @xmath84 possibly depending on @xmath85 and @xmath12 .
let us remark that one can not normalize @xmath79 imposing it to approach the identity matrix at infinity .
no solution would then exist for a set of exceptional points @xmath1 which accumulate in the neighborhood of the peaks of the solitons as @xmath86 ( see @xcite , chap .
38 for more details about this kind of issues ) .
normalization ( [ norm_matricial ] ) will do for our purposes .
all the information we need about the model riemann - hilbert problem is contained in the following [ roseto ] there exists a unique solution to the model riemann - hilbert problem .
this last one has the form @xmath87}\end{aligned}\ ] ] where @xmath88 and @xmath89 the functions @xmath90 and @xmath91 and their derivative w.r.t .
@xmath85 are bounded in the whole @xmath1-plane .
the constant @xmath84 is determined by conditions @xmath92 and @xmath93 , and has the following asymptotic behaviour @xmath94 here @xmath23 is a positive constant and @xmath95},\quad\quad\quad j=1,2,\ldots , m.\end{aligned}\ ] ] ansatz ( [ aneins ] ) and ( [ anzwei ] ) follow from points @xmath93 and @xmath96 of the model riemann - hilbert problem .
the `` residue conditions '' translate , concerning the vector @xmath97 , into a system of linear equations @xmath98 here @xmath99 indicates the m - dimensional column vector whose entries are all one .
the matrices @xmath100 and @xmath101 , respectively symmetric and diagonal , are given by @xmath102 the constants @xmath103 s are defined as follows : @xmath104 they vary between zero and @xmath105 for @xmath85 and @xmath12 real .
the matrix @xmath100 is easily proved to be positive definite . consequently also @xmath79 is , for all @xmath85 and @xmath12 real , and the system ( [ sist_orig ] ) has a unique solution .
now , by elementary calculations the inverse of @xmath79 is shown to be entry - wise bounded as the entries of the diagonal matrix @xmath101 vary between zero and @xmath105 .
this proves that also @xmath97 is bounded .
differentiating ( [ sist_orig ] ) w.r.t .
@xmath85 , one obtains @xmath106 where @xmath107 a solution @xmath108 for this system exists and is unique . rewriting ( [ sist_orig ] ) as @xmath109 the right - hand side is evidently bounded .
so , in view of ( [ collegamento ] ) , also @xmath110 is .
this results then into boundedness for @xmath108 .
the function @xmath111 can also be treated similarly .
the residue conditions yield in this case the system @xmath112 here @xmath79 is defined as above and @xmath113 now , from the `` residue conditions '' and from ( [ aneins ] ) one has @xmath114 so that both @xmath115 and its derivative w.r.t .
@xmath85 are bounded on the whole @xmath1-plane . the same is then proved for the vector @xmath116 , via arguments analogue to the ones above .
finally , the asymptotic estimate ( [ clas]-[sic ] ) is a classical result , already available in @xcite . we now wish to estimate the discrepancy between @xmath69 and the solution of the ( matricial ) model riemann - hilbert problem @xmath79 . to this purpose ,
let us introduce the `` error vector '' @xmath118}^{-1}. \end{aligned}\ ] ] we start with a characterization of its following directly from the meromorphic @xmath67-problem for @xmath69 .
it consists in the following find a 2-dimensional vector - valued function @xmath117 continuous on the whole complex plane and differentiable with continuity ( as a function of two real variables ) on @xmath119 , such that * i _ * : : `` @xmath67-condition '' . for all @xmath120 one has @xmath121 where @xmath122e^{t\phi(w ) } & 0
\end{array } { \right)}}\ , [ m(w)]^{-1 } & \mbox { if } { \text{im}}(w)\geq 0,\\ m(w)\,\scriptsize{{\left(}\begin{array}{cc } 0 & [ { \overline{\partial}}r(-w)]e^{-t\phi(w ) } \\ 0 & 0 \end{array } { \right)}}\ , [ m(w)]^{-1 } & \mbox { if } { \text{im}}(w)\leq 0 . \end{array}\right.\normalsize\end{aligned}\ ] ] * iii * _ : : `` symmetry condition '' .
the vector - valued function @xmath123 is an even function @xmath124 * iv * _ : : `` normalization condition '' .
@xmath125 from conditions ( [ resuno]-[norm_matricial ] ) on deduces via simple arguments of complex analysis that @xmath126 from formula ( [ def_e ] ) , it is then not evident that @xmath117 is smooth in the origin .
this follows indeed from property ( [ simm_e ] ) together with the observation that @xmath69 is differentiable in zero ( see remark [ remark_claudia ] ) .
let us now define the operator @xmath127 as follows @xmath128}{\left(}w { \right)}:= -\frac{1}{\pi } \iint_{{\mathbb{r}}^2 } \frac{{\mathbf{e}}(s)b(s)}{w - s}{\textrm{d}}a(s).\end{aligned}\ ] ] the smooth @xmath67-problem above is easily shown to be equivalent to the following integral equation @xmath129 ( see @xcite and @xcite for further details ) .
this is the final reformulation of the riemann - hilbert problem [ cesare ] , on which we will perform our analysis starting from the next section . in this section
we study existence and uniqueness of a solution for equation ( [ mostar]-[sarajevo ] ) .
[ proposizione_principale ] there exists a constant @xmath23 , depending on @xmath130 , such that @xmath131 for all @xmath132 and all @xmath25 .
fix such @xmath12 and @xmath85 and let @xmath117 belong to @xmath133 . by elementary algebraic manipulations one obtains @xmath134 on @xmath56 the matrix @xmath135 vanishes , because @xmath136 does . on @xmath137 one
has @xmath138 in view of this last one , of ( [ diseg_r ] ) and of ( [ norm_matricial ] ) one also has @xmath139 on @xmath140 this last one further simplifies to @xmath141 it follows that @xmath142}^{-\frac{1}{2}}}_{{l^2{\left(}{\mathbb{r } } , { \textrm{d}}a { \right ) } } } { \textrm{d}}b\\ & \leq \frac{c_2}{2 } \cdot\sqrt[4]{\frac{\pi^3}{3\delta}}e^{-c_0\delta t}\int_{\delta}^{2\delta } \frac{{\textrm{d}}b}{\sqrt{\abs*{b - v}}}\\ & \leq c_3 \cdot\sqrt[4]{\delta}\ , e^{-c_0\delta t}.\end{aligned}\ ] ] let us now consider the region @xmath143 . in view of ( [ stima_insieme ] ) one has @xmath144}^{-\frac{1}{2}}}_{{l^2{\left(}{\mathbb{r } } , { \textrm{d}}a { \right)}}}\nonumber\\ & \leq c_5 \cdot \int_0^{\delta } \frac{b^{n-1 } e ^{-c_0 tb}}{\sqrt[4]{tb}\sqrt{\abs*{b - v } } } { \textrm{d}}b \nonumber \\ & \leq \frac{c_6}{t^{n-\frac{1}{2}}}.\end{aligned}\ ] ] the regions @xmath49 and @xmath50 are treated analogously .
this completes the proof .
a direct consequence of the analysis above is the following , fundamental [ corollario_sano ] the integral equation ( [ mostar]-[sarajevo ] ) has a unique solution in @xmath133 , whenever @xmath12 is sufficiently large and @xmath85 is greater or equal than @xmath145 . moreover , for such solution @xmath146 , one has @xmath147 uniformly with respect to @xmath148 . in view of ( [ stima_inf ] ) , one can invert the operator @xmath149 by means of neumann series .
this easily yields both parts of the thesis .
this last result guarantees that the integral equation ( [ mostar]-[sarajevo ] ) is an equivalent characterization of the error vector
@xmath117 introduced in ( [ def_e ] ) .
such an equivalence will be exploited in order to extract as much explicit information as possible about its asymptotic behavior . from the integral equation ( [ mostar ] - [ sarajevo ] )
it is easy to deduce that the error vector @xmath123 has the following asymptotic expansion @xmath151 where @xmath152 here @xmath78 is understood to approach infinity along the imaginary axis . in this regime ,
the nonalaytic vector @xmath153 coincides with @xmath154 .
so ( [ def_e ] ) yields @xmath155 plugging ( [ norm_matricial ] ) and ( [ drina ] ) into ( [ costarica ] ) gives @xmath156 for the first component of the vector @xmath157 defined in ( [ tisa ] ) . plugging this last identity into ( [ sava ] ) ,
one obtains @xmath158 comparing with ( [ nuova_sol ] ) and recalling ( [ clas ] ) yields @xmath159 for some constant @xmath57 . estimating the magnitude of this quantity will complete the proof of theorem [ hauptsatz ] . to that purpose
, we will need the following [ ionio ] the solution @xmath160 of the integral equation ( [ mostar]-[sarajevo ] ) is differentiable with respect to @xmath85 for all positive @xmath85 and all sufficiently large @xmath12 .
there exists a constant @xmath23 , dependent on @xmath130 , such that @xmath161 for all sufficiently large @xmath12 and all @xmath162 .
the @xmath163-norm above is understood to be computed w.r.t .
the complex variable @xmath78 . solving the integral equation ( [ mostar]-[sarajevo ] ) by neumann series , one obtains @xmath164}(w;x , t).\end{aligned}\ ] ] for each term of the series on the right hand side , the recursive formula @xmath165}(w;x , t ) = { \mathbb{j}}_x{\left\ { } \newcommand{\rb}{\right\}}{\mathbb{j}}^{n-1}{\left[}(1,0 ) { \right]}(w;x , t ) \rb + { \mathbb{j}}{\left\ { } \newcommand{\rb}{\right\}}{\frac{\partial}{\partial x}}{\mathbb{j}}^{n-1}{\left[}(1,0 ) { \right]}(w;x , t ) \rb , & & n\geq 1 , \end{aligned}\ ] ] holds , where @xmath166}(w ) : = -\frac{1}{\pi}\iint_{{\mathbb{r}}^2}\frac{{\left[}{\frac{\partial}{\partial x}}b(s;x , t){\right]}{\mathbf{e}}(s)}{w - s } { \textrm{d}}a(s ) & & { \mathbf{e}}\in { \textrm{l}^\infty ( \mathbb{r}^2 ) } .\end{aligned}\ ] ] by analogue calculations as in the proof of theorem [ proposizione_principale ] , ( [ fortuna ] ) yields the estimate @xmath167}(w;x , t)}_{\infty } \leq n\cdot { \left(}\frac{c}{t^{n-\frac{1}{2 } } } { \right)}^ n , & & x\geq c_0 t\end{aligned}\ ] ] valid for some constant @xmath23 , all positive integers and all @xmath12 sufficiently large .
as a consequence , @xmath168}(w;x , t)\end{aligned}\ ] ] converges uniformly in this @xmath1-region , and coincides with the derivative with respect to @xmath85 of the right - hand side of ( [ fleons ] ) .
this gives differentiability of the left - hand side and estimate ( [ gemona ] ) .
the main point here is to control the first term in the right - hand side of ( [ ciliegio ] ) . using ( [ sila ] )
, we get for this one the expression @xmath169}{\textrm{d}}a(s).\end{aligned}\ ] ] taking the derivative under the sign of integral and applying lemma [ ionio ] and corollary [ corollario_sano ] one obtains @xmath170 where @xmath171}\rb { \textrm{d}}a(s ) .
\end{aligned}\ ] ] now , in view of lemma [ roseto ] , one can determine a positive constant @xmath172 such that @xmath173 for @xmath174 . so that @xmath175}{\textrm{d}}b\\ & \leq c_2 \int_{0}^{2\delta } b^n{\left[}\frac{1}{\sqrt{tb } } + \frac{1}{{\left(}tb { \right)}^{\frac{3}{2 } } } { \right]}e^{-c_0 tb } { \textrm{d}}b\\ & = \frac{c_2}{t^{n+1}}\int_0^{+\infty } { \tilde{b}}^n { \left(}\frac{1}{\sqrt{{\tilde{b } } } } + \frac{1}{{\tilde{b}}^{\frac{3}{2 } } } { \right)}e^{-c_0 { \tilde{b } } } { \textrm{d}}{\tilde{b}}\,\ , \leq \,\
, c_3 t^{-n-1}\end{aligned}\ ] ] substiting in ( [ celoria ] ) one obtains @xmath176 where the asymptotic estimate is to be understood as uniform with respect to @xmath85 greater or equal than @xmath177 . again by means of ( [ banda_semplice ] ) one easily determines a positive constant @xmath178 such that @xmath179 for all @xmath12 suffciently large and @xmath85 greater or equal than @xmath177 .
we then turn to analyse , in this same @xmath1-regime , @xmath180 an explicit expression for the integrand above is provided by @xmath181 where @xmath182 and the function @xmath183 was defined in ( [ aneins ] ) .
one can then rewrite ( [ brasile ] ) as follows @xmath184 put @xmath185 from the original definition ( [ def_phi ] ) of @xmath28 , separating real and imaginary parts , @xmath186}.\end{aligned}\ ] ] in view of our assumptions on the reflection coefficient @xmath14 , there exists a function @xmath187 such that @xmath188 substituting in ( [ kolmogorov ] ) and using fubini s theorem , one obtains then @xmath189}e^ { 2ita{\left[}4a^2 + { \left(}{\frac{x}{t}}- 12 b^2 { \right)}- \frac{\rho}{2 t } { \right ] } } { \textrm{d}}a \rb { \textrm{d}}\rho\end{aligned}\ ] ] put @xmath190}e^
{ 2ita{\left[}4a^2 + { \left(}{\frac{x}{t}}- 12 b^2 { \right)}- \frac{\rho}{2 t } { \right ] } } { \textrm{d}}a . \end{aligned}\ ] ] to study this integral we use the van der corput lemma ( see @xcite , pp 334 ) .
we obtain in this way that there exists a ( universal ) constant @xmath191 such that @xmath192}}_{l^1({\mathbb{r}},{\textrm{d}}a ) } \rb\end{aligned}\ ] ] from explicit expression ( [ aneins ] ) one deduces the existence of a constant @xmath193 such that @xmath194 uniformly for real @xmath85 and @xmath12 .
it immediately follows that @xmath195 and that @xmath196}}_{l^1({\mathbb{r}},{\textrm{d}}a ) } & \leq c_{7}{\left[}\int_{-\infty}^{+\infty } e^{-24tba^2}{\textrm{d}}a + \int_{-\infty}^{+\infty } tba^2 e^{-24tba^2 } { \textrm{d}}a { \right]}\\ & = c_{7}{\left[}\frac{1}{\sqrt{tb}}\int_{-\infty}^{\infty}e^{-24{\tilde{a}}^2}{\textrm{d}}{\tilde{a}}+ \frac{1}{\sqrt{tb}}\int_{-\infty}^{+\infty } { \tilde{a}}^2 e ^{-24{\tilde{a}}^2 } { \textrm{d}}{\tilde{a}}{\right]}\,\leq \,\frac{c_{8}}{\sqrt{tb}}.\end{aligned}\ ] ] via ( [ bombieri ] ) this yields @xmath197 by elementary estimates in ( [ cicerone ] ) , then , @xmath198 here the constant @xmath199 is understood to depend also on the @xmath200-norm of @xmath201 .
this treatment of integral @xmath202 was inspired by @xcite , lemma 5.1 . finally , substituting according to this last one in ( [ tricomi ] ) , one obtains @xmath203 plugging this inequality and ( [ cile ] ) into ( [ saldini ] ) gives the thesis .
research partially supported by the austrian science fund ( fwf ) under grant no .
y330 and by the indam group for mathematical physics .
the author wishes to thank gerald teschl , ira egorova and spyros kamvissis for valuable discussions .
its , a.r . , asymptotic behavior of the solution of the nonlinear schr'odinger equation , and isomonodromic deformations of systems of linear differential equations , _ dokl .
sssr _ 261 ( 1981 ) 14 - 18 ( in russian ) ; _ soviet .
_ 24 ( 1982 ) , 452 - 456 ( in english ) .
mclaughlin , kt - r .
, miller p.d . ,
the dbar steepest descent method and the asymptotic behavior of polynomials orthogonal on the unit circle with fixed and exponentially varying nonanalytic weights , _ imrp int .
_ 2006 , art .
i d 48673 , 177 .. | we address the problem of long - time asymptotics for the solutions of the korteweg - de vries equation under low regularity assumptions .
we consider decreasing initial data admitting only a finite number of moments .
for the so - called `` soliton region '' , an improved asymptotic estimate is provided , in comparison with the one in @xcite .
our analysis is based on the dbar steepest descent method proposed by p. miller and k. t. d. -r .
mclaughlin . |
the grist project ( ) is enabling astronomers and the public to interact with the grid projects that are being constructed worldwide , and bring to flower the promise of easy , powerful , distributed computing .
our objectives are to understand the role of service - oriented architectures in astronomical research , to bring the astronomical community to the grid particularly teragrid , and to work with the national virtual observatory ( nvo ) to build a library of compute - based web services . the scientific motivation for grist derives from creation and mining of wide - area federated images , catalogs , and spectra
. an astronomical image collection may include multiple pixel layers covering the same region on the sky , with each layer representing a different waveband , time , instrument , observing condition , etc .
the data analysis should combine these multiple observations into a unified understanding of the physical processes in the universe .
the familiar way to do this is to cross - match source lists extracted from different images .
however , there is growing interest in another method of federating images that reprojects each image to a common set of pixel planes , then stacks images and detects sources therein .
while this has been done for years for small pointing fields , we are using the teragrid to perform this processing over wide areas of the sky in a systematic way , using ( pq ) survey data .
we expect this `` hyperatlas '' approach will enable us to identify much fainter sources than can be detected in any individual image ; to detect unusual objects such as transients ; and to deeply compare ( e.g. , using principal component analysis ) the large surveys such as sdss , 2mass , dposs , etc .
( williams et al .
2003 ) .
grist is helping to build an image - federation pipeline for the palomar - quest synoptic sky survey ( djorgovski et al .
2004 ) , with the objectives of mining pq data to find high redshift quasars , to study peculiar variable objects , and to search for transients in real - time ( mahabal et al .
our pq processing pipeline will use the teragrid for processing and will comply with widely - accepted data formats and protocols supported by the vo community .
the grist project is building web and grid services as well as the enabling workflow fabric to tie together these distributed services in the areas of data federation , mining , source extraction , image mosaicking , coordinate transformations , data subsetting , statistics
histograms , kernel density estimation , and r language utilities exposed by services ( graham et al .
2004 ) , and visualization .
composing multiple services into a distributed workflow architecture , as illustrated in figure [ fig : workflow ] , with domain experts in different areas deploying and exposing their own services , has a number of distinct advantages , including : * proprietary algorithms can be made available to end users without the need to distribute the underlying software . *
software updates done on the server are immediately available to all users .
* a particular service can be used in different ways as a component of multiple workflows . *
a service may be deployed close to the data source , for efficiency .
interactive deployment and control of these distributed services will be provided from a workflow manager . we expect to use nvo services for data access images , catalogs , and spectra as well as the nvo registry for service discovery .
as described in section [ sec : workflow ] , much of the pipeline and mining software for grist will be built in the form of web services .
one of the reasons for building services is to be able to use them from a thin client , such as a web browser .
however , for such services to be able to process private data or use high - end computing , there must be strong authentication of the user . the vo and grid communities are converging around the idea of x.509 certificates as a suitable credential for such authentication .
however , most astronomers do not have such a certificate , and we do nt want to make them go through the trouble of getting one unless it is truly necessary .
therefore , we are building services with `` graduated security '' , meaning not only that small requests on public data are available anonymously and simply , but also that large requests on private data can be serviced through the same interface . however in the latter case , a certificate is necessary .
thus the service `` proves its usefulness '' with a simple learning curve , but requires a credential to be used at full - strength ( see illustration in figure [ fig : grad_sec ] ) .
a key science - driven workflow we are constructing is illustrated in the schematic in figure [ fig : pq - pipeline ] .
the primary objectives are to search for high redshift quasars and optical transients in data from the palomar - quest sky survey .
the pipeline begins by federating multiwavelength datasets , and matching objects detected with the _ z _ filter with catalogs at other frequencies .
cluster analysis performed on the resulting color - color plots ( e.g. , _
i - z _ vs.
_ z - j _ ) yield new quasar candidates , and outliers may indicate the presence of other objects of interest .
single epoch transients are indicated by objects that are detected in one filter but not others .
an object that is detected in the reddest filter is of special interest since it could be a highly obscured object or a high redshift quasar . for multi - epoch transient search , illustrated in the lower part of figure [ fig : pq - pipeline ] ,
we compare new data with a database of past epochs to detect new transients or other variable objects .
as described above , a primary objective of the pq survey is the fast discovery of new types of transient sources by comparing data taken at different times .
such transients should be immediately re - observed to get maximum scientific impact , so we are experimenting with `` dawn processing '' on the teragrid , meaning that data is streamed from the telescope to the compute facility as it is taken ( rather than days later ) .
the pipeline itself is being built with streaming protocols so that unknown transients ( e.g. , newly identified variables or asteroids ) can be examined within hours of observation with a view to broadcasting an email alert to interested parties .
grist is developing a library of interoperable grid services for astronomical data mining on the teragrid , compliant with grid and vo data formats , standards , and protocols .
for ease of use , grist services are built with graduated security , requiring no more formal authentication than is appropriate for a given level of usage .
grist technology is part of a palomar - quest data processing pipeline , under construction , to search for high red - shift quasars and optical transients .
more information on grist can be found on our project web site at . | the grist project is developing a grid - technology based system as a research environment for astronomy with massive and complex datasets .
this knowledge extraction system will consist of a library of distributed grid services controlled by a workflow system , compliant with standards emerging from the grid computing , web services , and virtual observatory communities .
this new technology is being used to find high redshift quasars , study peculiar variable objects , search for transients in real time , and fit sdss qso spectra to measure black hole masses .
grist services are also a component of the `` hyperatlas '' project to serve high - resolution multi - wavelength imagery over the internet . in support of these science and outreach objectives
, the grist framework will provide the enabling fabric to tie together distributed grid services in the areas of data access , federation , mining , subsetting , source extraction , image mosaicking , statistics , and visualization . |
the study of the space - time evolution of hadronization during lepto - production on nuclear targets is relatively straightforward because in this case , only one string ( two jets of hadrons ) is produced .
the possible influence of the two string mechanism was studied in ref .
@xcite , in which its contribution was determined to be small .
another important issue is the selection of conditions under which cascading processes in a nucleus do not occur , i.e. , the hadronization occurs beyond the nucleus .
consequently , for the investigation of the hadronization process , it is very important to know the formation lengths of hadrons in lepto - production on the elementary nuclear target ( nucleon ) .
+ such investigations were performed in refs .
@xcite . in @xcite
, it was claimed that for hadrons as composite systems , the notion of formation length is ambiguous as the different constituents of the hadrons originate at different distances .
thus , the question of which of the following two length scales plays the more important role in the hadronization process is an open and model - dependent question : ( i ) the constituent formation length @xmath0 , which is the distance between the dis and the first constituent production points , or ( ii ) the yo - yo formation length @xmath1 , which is the distance between the dis and yo - yo formation points , where yo - yo is the object with the quantum numbers of the final hadron but without its `` sea '' . + in refs .
@xcite , the formation lengths of pions were presented in the form @xmath2 , where @xmath3 is the probability of the formation length being @xmath4 . when @xmath5 and @xmath4 were obtained in the framework of the lund model , comparison with experimental data gave @xmath6 .
this result has confirmed the conclusion of ref .
@xcite on the importance of the constituent formation length
. we will further consider @xmath5 as a formation length .
the parameter @xmath5 is a function of the variables @xmath7 and @xmath8 ( the energy of a virtual photon and the fraction of this energy carried away by the final hadron with energy @xmath9 ( @xmath10 ) , respectively ) .
+ the aim of this work to obtain the average formation lengths of hadrons in cc neutrino- and cc antineutrino - production on a proton target and to compare these values with those for electro - production , which were obtained earlier .
+ the paper is organized as follows . in section 2 ,
we briefly present a description of the model .
the results and discussion , as well as necessary information for the calculations , are presented in section 3 .
the conclusions are presented in section 4 .
we begin by considering the distribution of the constituent formation lengths @xmath11 of hadrons carrying away fractional energy @xmath8 : @xmath12 @xmath13 where @xmath14 is the full hadronization length and @xmath15 is the string tension . + @xmath16 is the scaling function .
it is defined by the condition that @xmath17 is the probability that the first hierarchy ( rank 1 ) primary hadron carries away the fraction of energy @xmath8 of the initial string in the small interval @xmath18 .
we use the symmetric lund scaling function @xcite @xmath19 where @xmath20 and @xmath21 are parameters of the model , @xmath22 is the transverse mass of the emitted hadron and @xmath23 is a normalization factor .
+ information about the functions @xmath24 and @xmath25 is presented in the next subsection .
+ the functions @xmath26 are distributions of the constituent formation lengths @xmath11 of the rank @xmath27 hadrons carrying fractional energy @xmath8 .
to calculate the distribution functions , we use the recursion equation of ref .
@xcite .
the functions @xmath24 and @xmath25 are the probabilities of obtaining in some process the compositions of valence quarks for leading ( rank 1 ) and subleading ( rank 2 ) hadrons on the proton target . for hadrons of rank greater than @xmath29 , the condition @xmath30 is fulfilled .
we consider the three types of processes : electro - production , cc neutrino- and cc antineutrino - production .
@xmath25 does not depend on the process type .
the baryons in our scheme are constructed of the valence quark and diquark .
the function @xmath24 is a composition of three factors : ( i ) the probability that as a result of dis , the first constituent ( quark , antiquark ) of the final meson is knocked out ; ( ii ) the probability that as a result of the first break of the string , the second constituent of the final meson arises ; and ( iii ) the probability that these partons transform into the desired meson .
the function @xmath25 is composed of two factors : ( i ) the probability that as a result of two consequent breaks of the string , the first and second constituents of the final meson arise and ( ii ) the probability that these partons transform into the desired meson . for final baryons ,
the antiquark should be replaced by a diquark .
these functions for electro - production were presented in refs .
@xcite , and for ( anti)neutrino - production , they can be obtained using refs . @xcite .
we present here a realistic approach for the calculation of the average formation lengths considering the type of process in which the hadron was produced , the types of hadron and the target .
unfortunately , for the symmetric lund scaling function , the analytic summation of the produced hadrons sequence over all ranks is impossible .
therefore , we restrict ourselves to @xmath31 in eq.(1 ) .
+ in the general case , the function @xmath32 , @xmath33 can be written in the form : @xmath34 @xmath35 @xmath36 when combined with direct production , a few resonances may also contribute . here , @xmath37
( @xmath38 ) is the probability that the composite parton system transforms into a hadron ( resonance ) .
we use the condition @xmath39 . + the distribution functions @xmath40 and @xmath41 were described in detail in refs .
all the calculations were performed at fixed values of @xmath42 , @xmath7 and @xmath43 equal to @xmath44 , @xmath45 and @xmath46 , respectively .
+ the scaling function @xmath16 in eq.(2 ) has two parameters @xcite @xmath47 , @xmath48 and depends on the type of produced hadron . in the calculations ,
the types of observed hadrons and their parent resonances were considered , while for other hadrons , a summation over their types was performed .
the string tension was fixed at a value of @xmath49 .
+ the leading order parton distribution functions from @xcite were used for the quarks ( antiquarks ) in the proton . + we assume that new @xmath50 pairs are @xmath51 with probability @xmath52 , @xmath53 with probability @xmath54 and @xmath55 with probability @xmath56 .
it follows from isospin symmetry that @xmath57 . for these calculations
, we use the set of values for @xmath58 @xcite : @xmath59 . + for baryon production , we use the set of probabilities for the production of diquark - antidiquark pairs . for simplicity , we express them using the probabilities of light quark - antiquark pairs production .
we use : @xmath60 for the @xmath61-diquark with @xmath62 ( @xmath63-spin , @xmath64-isospin ) ; @xmath65 for light diquarks with @xmath66 ; and @xmath67 for strange diquarks .
+ = 8.cm
the normalized average formation lengths of the pseudoscalar mesons on the proton target , calculated using the symmetric lund model as functions of @xmath8 , are presented in fig . 1 for electro- ( panels a , b ) , cc neutrino- ( panels c , d ) , and cc antineutrino- ( panels e , f ) production .
the formation lengths are represented for pions in panels a , c and e and for kaons in panels b , d and f. the values of the parameters of the symmetric lund model are also presented .
the following parameters are indicated : the formation lengths for the direct pions ( dashed - dotted lines ) ; the sum of the direct pions and the pions from the decay of @xmath68 mesons ( dotted lines ) ; the sum of the direct pions and the pions from the decays of the @xmath68 and @xmath69 mesons ( dashed lines ) ; and the sum of the direct pions and the pions from the decays of the @xmath68 , @xmath69 and @xmath70 mesons ( solid lines ) .
the other indicated parameters are as follows : the formation lengths for the direct kaons ( dotted lines ) ; the sum of the direct kaons and the kaons from the decay of @xmath70 mesons ( dashed lines ) ; and the sum of the direct kaons and the kaons from the decays of the @xmath70 and @xmath71 mesons ( solid lines ) . for electro- and cc neutrino - production , the upper and lower curves of each type represent the formation lengths of positively and negatively charged hadrons , respectively . for cc antineutrino - production ,
the upper and lower curves represent the formation lengths of negatively and positively charged hadrons , respectively .
2 , the normalized average formation lengths of baryons and antibaryons in the symmetric lund model as functions of @xmath8 are presented in panels a and b , c and d , and e and f for electro- , cc neutrino- and cc antineutrino - production , respectively . the results for protons and antiprotons are presented in panels a , c and e , and the results for @xmath72 and @xmath73 are presented in panels b , d and f. the contribution of @xmath74 ( @xmath75 ) resonance and @xmath76 ( @xmath77 ) resonance is considered for protons ( antiprotons ) and @xmath72 ( @xmath73 ) , respectively .
+ from figs . 1 and 2 , certain general features of the average formation lengths are apparent : ( i ) for electro- and cc neutrino - production , the positively charged hadrons ( @xmath78 , @xmath79 and proton ) have larger formation lengths than the negatively charged hadrons ( @xmath80 , @xmath81 and antiproton ) ; for cc antineutrino - production , the situation is reversed .
( ii ) all the curves have a characteristic form with values equal to zero on two boundary points along @xmath8 and one maximum in the vicinity of @xmath82 for pions , @xmath83 for kaons , @xmath84 for protons and @xmath85 for @xmath72 .
the magnitudes of the maxima of the distribution do not exceed @xmath86 .
( iii ) the contribution from the decay of resonances is maximal for pions .
this contribution reaches @xmath87 for @xmath78 ( @xmath80 ) meson for the electro- and cc neutrino- ( cc antineutrino- ) production processes .
as expected , the maximal contribution originates from the @xmath68 meson .
the contribution of resonances can be neglected for kaons and baryons .
( iv ) a large difference between the formation lengths of oppositely charged hadrons is observed for kaons and protons .
the maximal difference occurs for cc neutrino - production .
( v ) for electro - production , the difference between the average formation lengths of particles and antiparticles vanishes at @xmath88 ; for cc neutrino- and cc antineutrino - production , the difference is large enough for @xmath8 to equal unity . ( vi ) unlike charged hadrons ,
the formation length of a neutral baryon ( @xmath72 ) is larger than that of the corresponding antibaryon ( @xmath73 ) for all three types of the production processes ; however , the difference between the average formation lengths of @xmath72 and @xmath73 is very small .
+ let us briefly discuss why the average formation lengths of positively ( negatively ) charged hadrons are larger than those of negatively ( positively ) charged ones for electro- and cc neutrino- ( cc antineutrino- ) production . in electro - production , this phenomenon occurs because of the large probability of knocking out a @xmath89 quark as a result of dis . the knocked out quark enters a composition of the leading hadron , which has a maximal formation length .
the @xmath79 meson has an average formation length that is larger than the @xmath78 meson because , in the first case , the influence of resonances is smaller .
the @xmath81 meson has a smaller average formation length than the @xmath80 meson because it is constructed from the `` sea '' quarks of the proton and can not be a leading hadron , whereas the @xmath80 meson can be leading because of the @xmath90 quark entering its composition . in cc neutrino- ( cc antineutrino- )
production , the knocked out parton obtains a positive ( negative ) electric charge , which implies that the leading hadron can be preferably positively ( negatively ) charged .
it should be noted that in the string model , the formation length of the leading ( rank 1 ) hadron @xmath91 , does not depend on the type of process or on the hadron or target type .
in this work , for the first time , the average formation lengths of pseudoscalar mesons , baryons and antibaryons for cc neutrino- and cc antineutrino - production on a proton target in the framework of the symmetric lund model have been determined .
+ the results of the calculations were compared with the previously determined results of electro - production ( see refs .
it was shown that despite certain differences in shape and magnitude , the average formation lengths for electro- and cc neutrino - production have similar behaviors .
cc antineutrino - production differs from the others by the replacement of the particles and antiparticles on antiparticles and particles , respectively .
+ finally , we would like to discuss the possible application of the obtained results .
recently , we have used the simplified version of the average formation lengths calculation for pions obtained in the framework of the symmetric lund model to fit sidis data , in which the possibility of daughter pions production was neglected in eq.(3 ) and the condition @xmath92 was applied .
these simplifications were necessary to reduce the computation time .
the two - parametric fit yielded satisfactory agreement with the data @xcite .
99 n.akopov , l.grigoryan , z.akopov , eur.phys.j . * c52 * ( 2007 ) 893 t.chmaj , acta phys.polon .
* b18 * ( 1987 ) 1131 a.bialas , m.gyulassy , nucl.phys . *
b291 * ( 1987 ) 793 n.akopov , l.grigoryan , z.akopov , eur.phys.j . *
c44 * ( 2005 ) 219 n.akopov , l.grigoryan , z.akopov , eur.phys.j . * c70 * ( 2010 ) 5 l.grigoryan , arxiv:1208.2339 [ hep - ph ] b.andersson et al . , phys.rep.*97 * ( 1983 ) 31 t.sj@xmath93strand , l.lonnblad , s.mrenna , hep - ph/0108264 ( 2001);lu tp 01 - 21 m.gl@xmath94ck , e.reya , a.vogt z.phys.*c67 * ( 1995 ) 433 l.grigoryan , phys.rev .
* c81 * ( 2010 ) 045207 l.grigoryan , phys.rev . *
c83 * ( 2011 ) 014904 v.barone , c.pascaud and f.zommer , eur.phys.j.*c12 * ( 2000 ) 243 w .- m.yao et al . ,
journal of physics g * 33 * ( 2006 ) 1 | the average formation lengths of the hadrons produced during the deep inelastic scattering ( dis ) of leptons on protons are studied in the framework of the symmetric lund model .
it is shown that these formation lengths essentially depend on the electric charges of the hadron .
for electro - production and charged current ( cc ) neutrino - production , the average formation lengths of positively charged particles are larger than those of negatively charged antiparticles .
this situation is reversed for cc antineutrino - production . in all the mentioned cases , the main mechanism is the direct production of hadrons .
the additional mechanism of hadron production , through the decay of resonances , is essential only for pions and leads to a decrease in the average formation lengths . * formation lengths of hadrons in lepto - production
* l. grigoryan alikhanyan national science laboratory ( yerevan physics institute ) , br.alikhanian 2 , 375036 yerevan , armenia |
during migraine attacks , localized pathological excitation can spread through cortical tissue and invade large areas before it abates .
this activity causes migraine aura , that is , neurological symptoms preceding the headache phase @xcite .
the underlying process is a phenomenon called cortical spreading depression ( sd ) .
it is assumed to be a reaction - diffusion process in the cortex , although reactions and diffusion processes that provide the mechanism of propagtion are still under debate @xcite . however , the generic dynamics of reaction - diffusion systems are largely independent of the interaction details and shared among various biological systems @xcite . therefore , to describe the spatio - temporal patterns of sd ,
the cortex can be approximated as a continuous excitable media supporting reaction diffusion waves @xcite .
psychophysical studies on visual processing in migraine patients suggest that changes in their networks of cortical neurons lead to an interictal state of changed excitability , i.e. , an anomalous cortical state in the interval between migraine attacks @xcite .
this motivates efforts to understand how the spread of reaction diffusion waves is controlled by nonlocal network connectivity . to include this , we investigate in this work the hypothesis that the emergence of sd waves can be attributed to these network changes as well
previously , we have investigated how to change parameters of an excitable medium so as to efficiently protect cortical tissue surrounding a stimulus against recruitment @xcite .
our current results suggest that failures in synaptic transmission result in increased susceptibility of cortical tissue to sd .
such a modulation of excitability becomes of crucial importance when the cortical state is close to the bifurcation of the onset of wave propagation .
the clinically relevant conclusion to be drawn from this is that therapy might target network connectivity that modulates cortical tissue excitability , even though a specific network connectivity is not required for the initiation or propagation of sd in a regime far from the bifurcation .
( left ) visual migraine aura symptom in the shape of a jagged crescent pattern moving through the visual field ( modified from @xcite ) .
( right ) translation of the visual field disturbance by inverse retinotopic mapping .
the crescent patterns translates into a wave segment resembling a `` critical finger '' @xcite .
such patterns are unbounded wave segments observed close to the excitability boundary , i.e. , a bifurcation of the systems described by eqs .
( [ eq : fhnplusnlc-1]-[eq : fhnplusnlc-2 ] ) . ]
a variety of neural network models for sd have been proposed , though there is not yet consent on the mechanism .
roughly speaking , two classes of models exist .
one taking a bottom up approach based on biophysical laws including several ionic currents , ion pumps , membrane potentials , and osmotic forces @xcite .
the other approach is top down trying to incorporate the system properties without detailing any first - level subsystems .
hodgkin proposed the first simplified reaction - diffusion approach to sd .
the model was never published , but communicated to grafstein @xcite . according to this references ,
hodgkin suggested to consider a balance equation for potassium with a cubic source function and diffusion .
the three roots of the cubic function being the resting state , a threshold , and a potassium ceiling level , respectively . based on methods introduced by huxley , solving the equation led to the approximate speed of sd . however , the diffusion coefficient used was four times higher than that of potassium in aqueous solution .
this may either indicate the anomalous nature of potassium migration in the cortex @xcite , or the leading role of a faster transcelluar reaction - diffusion mechanism @xcite .
another model @xcite followed grafstein s potassium hypothesis with two extensions .
we shortly summarize this work , because we follow a similar approach but with a different interpretation and aim .
firstly , the model included a second dynamic variable describing a refractory phase of sd .
the source term in the balance equation of potassium is replaced by a quartic polynomial as the major nonlinearity of this activator - inhibitor system .
secondly , the reaction - diffusion model was connected to a neural network building together a hybrid model .
the neural network was originally used to study cortical dynamics and sensory map reorganization . in the hybrid model
it was used to explain visual field defects occurring during migraine with aura .
we also use a reaction - diffusion system combined with a nonlocal interaction coming from a neural network .
our goal is to investigate which neural network connectivity can prevent sd , as suggested and studied using cellular automatons in @xcite .
we use the spatially extended fitzhugh - nagumo ( fhn ) system @xcite , which has a cubic nonlinearity , as a generic model of sd waves @xmath3 the model approximates the cerebral cortex as a two - dimensional surface with the ability to support sustained sd wave propagation . as a generic model this system does not make an explicit distinction between the various species involved in sd . in effect
it lumps together sodium inward currents and extracellular potassium concentration @xmath4_o$ ] into a single activator variable @xmath5 and their combined kinetics into the cubic source term .
likewise , a single inhibitor variable @xmath6 is related to recovery processes , such as effective regulation of @xmath4_o$ ] by @xmath7-@xmath8 ion pumps and the glia - endothelial system @xcite . whether a transcellular or extracellular route is taken ,
is at this level not specified .
the main reason we use the fhn mechanism as the reaction - diffusion part of the model is that it has been shown to successfully reproduce the two - dimensional spatio - temporal pattern of sd @xcite .
our study is essentially based on these pattern formation properties of sd waves and less on its detailed biophysical mechanism .
we extend the fhn system to encompass cortical lateral interactions , i.e. , connections running parallel to the cortical surface .
they are accounted for in the form of nonlocal coupling terms @xmath9.\end{aligned}\ ] ] the signal @xmath10 can either be the activator @xmath5 or inhibitor @xmath6 .
a connection in the cortex can extend over several millimeters and it either mediates competitive or cooperative interactions .
the parameter @xmath1 describes the connection length and the coupling strength @xmath0 of the interaction .
lateral connections in the cortex can form clusters at regular intervals @xcite .
their structure and how they might interact with sd waves will be further considered in sec . [
sec : conpat ] until then , we consider only one nonlocal coupling term occurring either in the activator ( [ eq : fhnplusnlc-1 ] ) or inhibitor ( [ eq : fhnplusnlc-2 ] ) balance equation , with fixed values for @xmath1 and @xmath0 .
this leads to four different coupling schemes : two of cross coupling ( cc ) activator and inhibitor , and two in which each dynamic variable is coupled via eq .
( [ eq : nlc ] ) into its own balance equation ( ncc ) .
parameter space of the fhn system at the section @xmath11 .
@xmath12 is the propagation boundary .
below @xmath12 any confined perturbation of arbitrary profile decays eventually . above @xmath12 some wave profiles are stable and propagate with constant speed .
they correspond to homoclinic orbits in a co - moving frame .
the simulations have been done with a fhn system at @xmath13 and @xmath14 ( solid black circle ) . a successful suppression of reaction - diffusion waves by nonlocal coupling indicates a shift of @xmath12 of the combined system beyond the point at @xmath13 and @xmath14.,scaledwidth=50.0% ] the investigations were done in a one - dimensional spatial fhn system .
we chose the parameter of the fhn system such that it is without the nonlocal coupling term in eqs .
( [ eq : fhnplusnlc-1]-[eq : fhnplusnlc-2 ] ) above but near to the excitability boundary of a one - dimensional system .
this boundary was obtained by transforming the system into a co - moving frame and searching for homoclinic orbits .
these orbits correspond to pulse solutions in the original coordinates .
the excitability boundary ( @xmath12 ) is defined by a saddle - node bifurcation at which the stable and unstable homoclinic orbit disappear @xcite .
the parameter values of @xmath15 , @xmath16 , and @xmath17 at which homoclinic orbits disappear constitute a boundary of codimension one . in other words
, @xmath12 separates the parameter plane into a regime where local stimulations is transmitted without damping and a regime where such sustained 1d reaction - diffusion waves do not exist ( fig .
[ fig : pb ] ) .
we chose a fhn system near @xmath12 because the transient nature of the observed symptomatic and electrophysiological events during migraine suggest such a regime @xcite .
in the regime below but close to @xmath12 transient wave forms exist in a one - dimensional system @xcite . above
@xmath12 transient wave forms exist in two - dimensional systems untill excitability reaches a boundary called @xmath18 .
there , sustained wave segments , called `` critical fingers '' , propagate without reentering tissue ( c.f .
[ fig : ret ] ) .
the regime between @xmath12 and @xmath18 is therefore called sub - excitable .
the regime in which this transition takes place is also well investigated in chemical model systems in experiment and theory , for a review see @xcite . to investigate the influence of various nonlocal connectivity schemes on wave propagation in the regime of sub - excitability , we start by setting a super - threshold stimulation in the one - dimensional system , choosing a particular fhn system with parameter values @xmath13 , @xmath14 , @xmath11 , and @xmath19 .
once a stable one - dimensional wave profile is obtained , the nonlocal lateral network is switched on .
different networks for various parameter values @xmath0 and @xmath1 are classified by their effect on the wave .
we distinguish two cases . either the wave is suppressed .
this indicates that the excitability boundary @xmath12 of the combined system is shifted to higher excitability values ( upwards in fig .
[ fig : pb ] ) into a regime where without the nonlocal coupling pulse solutions would exit . or the wave continues to spread , though its profile and speed might change . from a clinical point of view
, the wave suppression is a desirable control goal for the network achieved within the solid black regions in the @xmath20-planes in fig .
[ fig : controplane ] . parameter plane ( @xmath21 ) of the nonlocal control term .
black areas indicate successful suppression of wave propagation .
( a ) non - cross coupling ( ncc ) in the activator equation ( [ eq : fhnplusnlc-1 ] ) .
( b ) ncc in the inhibitor equation ( [ eq : fhnplusnlc-2 ] ) .
( c ) cross coupling ( cc ) in the activator equation ( [ eq : fhnplusnlc-1 ] ) .
( d ) cc in the inhibitor equation ( [ eq : fhnplusnlc-2 ] ) .
, scaledwidth=75.0% ] we find that wave propagation can be suppressed with a ncc ( non - cross - coupled ) setup only with positive coupling strength @xmath0 .
when the ncc term appears in the activator balance equation , the desired control goal is achieved largely independent of the connection length @xmath1 ( fig .
[ fig : controplane ] a ) , as long as @xmath1 is in the range of the wave width , including its refractory tail .
when the nonlocal coupling term appears in the inhibitor balance equation , a similar picture arises , though waves are suppressed for connection lengths ranging into the refractory tail of the wave ( @xmath22 ) only for a narrow regime of @xmath0 .
suppression completely fails for @xmath23 ( fig .
[ fig : controplane ] b ) . cross coupling of inhibitor and activator achieves the desired control goal for both positive and negative coupling strengths @xmath0 , depending on the connection length @xmath1 ( fig .
[ fig : controplane ] c - d ) .
the area in the parameter plane @xmath2 where this control goal is achieved resembles a mexican - hat - type network connectivity .
this is readily seen in fig .
[ fig : mex ] .
when the nonlocal term appears in the inhibitor balance equation ( [ eq : fhnplusnlc-2 ] ) the regimes of successful control in the @xmath0 direction is much wider ( fig .
[ fig : controplane ] d ) than the regime for cross coupling in the activator balance equation ( [ eq : fhnplusnlc-1 ] , fig .
[ fig : controplane ] c ) .
the `` mexican hat '' connectivity profile of successful wave suppression for cross coupling ( cc ) is clearly visible when the control plane is rotated and the space coordinate @xmath1 is plotted as the distance ranging from negative to positive values .
shown is the successful control area ( black ) for the cc term appearing in the inhibitor equation ( [ eq : fhnplusnlc-2 ] ) .
when the cc term is in the activator equation ( [ eq : fhnplusnlc-1 ] ) the profile of the mexican - hat connectivity is inverted .
, scaledwidth=50.0% ]
in the previous section , we have shown which network failures lead to the emergence of reaction - diffusion waves . from this alone
it is not deducible whether some of these network failures can also explain the anomalous cortical state in the interval between migraine attacks .
it would be a plausible hypothesis , however , that the same network changes that cause the ictal migraine events , i.e. , sd waves , lead to the anomalous interictal state .
changes causing the latter have been attributed to abnormal aspects in early visual processing in the cortex @xcite .
various , seemingly contradictory explanations have been given , such as lack of both intra - cortical inhibition and excitation .
they are also referred to as cortical hypo- or hyperexcitability ( see @xcite and references therein ) .
there is substantial work on the functional role of lateral connectivity for cortical processing , but little is known how the mechanism of sd is coupled to it .
evidence supporting a coupling comes from two independent sources . on the one hand , there is the structure of the hallucinatory aura patterns , in particular the typical zigzags ( see fig .
[ fig : ret ] ) .
such patterns were suggested to reflect the cortical network organization @xcite .
the main idea is that the approaching wave initially affects cortical cells which possess the highest spontaneous activity and are clustered in patches . within these patches the neuronal response properties remain relatively constant .
their feature distribution corresponds to the organization of the receptive field structure in the cortex .
for example , the connection pattern in the visual cortex is unspecific in the immediate vicinity of each neuron , while long - range connections primarily run between so - called iso - orientation columns @xcite .
cells in an orientation column have the same oriented receptive field structure , thus they are responsive to edges with the same orientation .
these edges are literally seen during a migraine attack as the building blocks of the hallucinatory zigzag pattern .
therefore , it is reasonable to assume that the sd wave interacts with this neural network structure in form of a synchronization process that occurs at the front of the sd wave and extends over the typical spatial length scale of iso - orientation columns .
the other line of evidence comes from _ in vivo _ studies in animal research on sd . in @xcite
herreras et al . showed that a synchronization of the firing pattern is possible up to the order of millimeters ahead of the sd wave .
the peculiarity of this activity is that it is resistant to synaptic transmission blockade .
this led to the hypothesis of direct neuron - to - neuron communication by previously closed gap junctions .
they suggested that sd propagates through transcellular pathways using a reaction - diffusion mechanism .
computer simulations of shapiro support this scheme @xcite .
a complete description of sd , however , must additionally include the full network connectivity of synaptic transmission when sd occures close to a bifurcation .
such a description of sd is beyond the scope of the present study .
the time and space scales of these dynamics differ by several orders of magnitude such that a separate treatment is justified .
therefore we investigate the stability of the suggested neuron - to - neuron communication by gap junctions separately .
we assume that the way this transcellular pathway interacts with synaptic transmission is in principle described in the previous section .
in a previous study @xcite we used the fhn system modeling two individual neurons with a diffusive coupling in the activator variable .
we showed that two fhn - neurons , each oscillating under its own source of noise , can synchronize .
the application of time - delayed feedback to only one of two subsystems was shown to change coherence and time scales globally .
time delayed feedback is also able to induce stochastic synchronization under certain conditions .
this motivates the approach pursued here to examine a time - delayed coupling between two identical neurons .
since the time - delay can introduce rich dynamics we study the case without random fluctuations . to distinguish this system of two neurons from the spatially extended fhn reaction diffusion system describing the cortical tissue in eqs .
( [ eq : fhnplusnlc-1]-[eq : fhnplusnlc-2 ] ) , we use the variables @xmath24 and @xmath25 with a subscript 1 and 2 identifying the two neurons .
the variables represent the membrane potential and the gating , respectively .
the diffusive coupling occurs in the membrane potential .
this is a discrete model of a gap junction - mediated electrical coupling , because ionic currents through gap - junctions give rise to strong electrical coupling of the neurons .
gating mechanisms of neuronal gap junctions have not been described as yet .
therefore , we do not consider any gating .
but we include a time delay @xmath26 because if the spread in the transcellular pathway is diffusion limited , as the slow propagation speed of sd clearly suggests , the transmission time can be in the order of the excitation cycle .
@xmath27 individual neurons have only one stable fixed point ( for @xmath28 ) .
it is readily shown that when @xmath29 , the coupled system also has only one stable fixed point . with a non - vanishing delay time
the phase space is infinite dimensional .
then , the fixed point is given by the four coordinates above as well as their respective history functions of length @xmath26 , which need to be constant .
along the lines of @xcite , it can be shown that this fixed point is stable .
we find that for adequate parameter values @xmath30 , @xmath16 and @xmath26 , the system is multi - stable .
it can avoid the stable fixed point and instead exhibit a mutual resonance phenomenon . in the 4d phase space section at time @xmath31 , this results in a stable firing oscillation of period @xmath32 between the two sub - systems ( see fig.[fig : phaseportrait ] ) .
thus , for two fhn - neurons in the excitable regime , a non - vanishing delay enables a synchronous operation of the two subsystems .
we showed that certain control schemes of an inhibitor - activator type system shift the emergence of wave propagation towards higher values of excitability .
the control we investigated is of the form of a nonlocal coupling given in eq .
( [ eq : nlc ] ) .
this nonlocal transaction was added to the reaction - diffusion mechansim either in the inhibitor or the activator balance equation .
the sum of all individual cross coupling terms that achieve a clinically desirable control goal takes the shape of an upright or inverted mexican hat , respectively .
this supports our assumption that the nonlocal coupling results from intrinsic lateral cortical connections .
dichotomic lateral interaction is an architecture widely used in models of topographic feature maps .
the prototypical example of such maps is the orientation preference in primary visual cortex , which is activated by the sd wave .
the link between sd and the neuronal network architecture is still missing .
one possibility is that gap - junction - mediated oscillatory patterns trigger sd .
if so , these oscillatory patterns are likely to be modulated by lateral synaptic connections , although their existence is in general resistant to synaptic transmission blockade @xcite . however , when sd is close to the bifurcation of the onset of wave propagation , as suggested by the spatio - temporal patterns ( e.g. in fig . [
fig : ret ] ) , therapy might target network connectivity as to prevent spread .
although we are still far from modeling the full mechanism of migraine with aura , neural network models have become sophisticated enough to constrain and validate possible underlying cortical circuitry of involved subsystems .
to understand the origin of the gap - junction - mediated oscillatory patterns better , we performed simulations in a system of two gap - junction - coupled neurons .
we showed that a time - delay is sufficient to produce sustained oscillations in an otherwise merely excitable ensemble .
thus , opening gap junctions between neurons , which are closed in a healthy state , can explain a localized pathological synchrony in the cortex when there exits a time delay .
to summarize , in modeling migraine a major objective is to understanding cortical susceptibility to focal neurological symptoms in terms of neural circuitry @xcite .
this could open up to us new strategies for therapy using methods of controlling complex dynamics .
control of complex dynamics has evolved during the last decade as one of the central issues in applied nonlinear science @xcite .
progress toward clinical implementation of nonlinear methods has been done so far in neurology in particular in parkinson s disease , a neurological diseases also characterized by pathological brain synchrony .
there , techniques based on control of complex dynamics @xcite are now tested in clinical studies and fundamentally novel therapy methods are being evolved @xcite .
it is hoped that this success can be expanded .
this work has been partially supported by the dfg within the framework of sfb 555 and the sachbeihilfe da-602/1 - 1 .
b. grafstein : _ neural release of potassium during spreading depression .
_ , in _ brain function .
cortical excitability ans steady potentials .
_ , edited by m. a. b. brazier ( university of california press , berkely , 1963 ) , pp . 87124 .
o. herreras , c. largo , j. m. ibarz , g. g. somjen , and r. martin del rio : _ role of neuronal synchronizing mechanisms in the propagation of spreading depression in the in vivo hippocampus _
, j neurosci * 14 * , 7087 ( 1994 ) .
n. hadjikhani , m. sanchez del .
, o. wu , d. schwartz , d. bakker , b. fischl , k. k. kwong , f. m. cutrer , b. r. rosen , r. b. tootell , a. g. sorensen , and m. a. moskowitz : _ mechanisms of migraine aura revealed by functional mri in human visual cortex _
usa * 98 * , 4687 ( 2001 ) .
p. a. tass , j. klosterkotter , f. schneider , d. lenartz , a. koulousakis , and v. sturm : _ obsessive - compulsive disorder : development of demand - controlled deep brain stimulation with methods from stochastic phase resetting _ ,
neuropsychopharmacology * 28 suppl 1 * , 27 ( 2003 ) . | transient wave forms in neural networks with diffusive and nonlocal coupling have attracted particular interest because they may mediate recruitment of healthy cortical tissue into a pathological state during migraine . to investigate this process
, we use a reaction - diffusion system of inhibitor - activator type as a generic model of pathological wave propagation and set it close to bifurcation in the sub - excitable regime .
we report the influence of various nonlocal connectivity schemes on wave propagation .
wave propagation can be suppressed with cross coupling inhibitor and activator for both positive and negative coupling strength @xmath0 , depending on the connection length @xmath1 .
the area in the parameter plane @xmath2 where this control goal is achieved resembles a mexican - hat - type network connectivity .
our results suggest that nonlocal synaptic transmission can control wave propagation , which may be of therapeutic value .
nonlinear dynamical systems , excitability , control , nonlocal and time delay coupling m. a. dahlem , f. m. schneider , a. panchuk , g. hiller , e. schll [email protected],2 [email protected] [email protected],3 [email protected] [email protected] |
the energetic gamma - ray experiment telescope ( egret ) detected 66 active galactic nuclei ( agn ) with @xmath0-ray emission at energies greater than 100 mev @xcite .
almost all of the agn identified were blazars , triggering immense interest and prompting a series of studies focused on the multi - wavelength properties of these agn .
egret s successor , the large area telescope ( lat , onboard the _ fermi gamma - ray space telescope _ ) , has identified over 1000 @xmath0-ray sources with energies greater than 100 mev @xcite , revealing the @xmath0-ray sky in detail .
the lat instrument presents the opportunity to revisit questions first examined in the egret era with far superior data .
egret s discovery opened a new direction in investigating the nature and physical processes of blazars .
blazars are radio sources characterized by a compact core and relativistic jets that are aligned at a small angle to our observational line - of - sight , i.e. flat - spectrum radio quasars and bl lac objects @xcite .
it is supposed that @xmath0-ray emission from agn is produced from inverse - compton ( ic ) processes occuring within the jets and becomes apparently enhanced via relativistic beaming @xcite .
although there is good support for this scenario , several questions remain , particularly with respect to the origin of @xmath0-ray production within the jets @xcite , as well as how ic processes act to produce @xmath0-ray emission @xcite .
relativistic beaming ( or doppler boosting ) is evidenced by other associated properties that characterize blazars : high @xmath0-ray luminosity ( which comprises a large portion of the total radiative output ) @xcite ; apparent superluminal motion @xcite ; and rapid variability from radio to @xmath0-ray wavelengths @xcite . from the non - thermal nature of this emission
, it is understood that it is produced from within the agn jets and is relativistically beamed along the line - of - sight .
this could explain why @xmath0-ray - loud agn are also radio - loud but not all radio - loud agn are @xmath0-ray - loud . if @xmath0-ray emitting regions move at faster speeds than radio emitting regions , @xmath0-ray emission would be doppler boosted within a narrower cone @xcite .
outside the @xmath0-ray beaming cone and within the radio beaming cone , @xmath0-ray radiation would be doppler dimmed , giving the appearance of a @xmath0-ray - quiet yet radio - loud agn .
a second possibility suggests that the appearance of @xmath0-ray emission may be related to agn jet - bending @xcite .
if a bend in the jet occurs downstream of the @xmath0-emitting region and upstream of regions of extended radio emission , the emissions would be doppler - boosted in alignment to their respective sections of jet .
as such , it may be possible that @xmath0-ray emission is beamed away from us whilst radio emission is beamed towards us due to a better alignment of the radio - emitting jet section with our line - of - sight .
several studies have compared @xmath0-ray - loud and @xmath0-ray - quiet agn populations with respect to characteristics of the agn core and jet components .
statistically significant differences between these populations are found for core brightness temperature , jet opening angles and core polarization @xcite .
the distinction between populations is also established for radio flux density variability between @xmath0-ray - loud and @xmath0-ray - quiet agn @xcite , suggesting that radio and @xmath0-ray emission are produced within the same region within agn jets @xcite .
@xcite propose that radio - to-@xmath0-ray variability is caused by oblique shocks in the agn jets .
the role of oblique shocks has also been explored in jets on parsec @xcite and kiloparsec scales @xcite . @xcite
finds a correlation between @xmath0-ray and radio flux in agn after consideration of biases unaccounted for in previous studies @xcite .
studies have also correlated superluminal ejections and @xmath0-ray flares in agn jets , suggesting that @xmath0-ray emission is produced within the parsec - scale region of the jets @xcite .
furthermore , studies have related @xmath0-ray emission with jet kinematics @xcite , and there is evidence to suggest that fsrqs and bl lac objects may have intrinsically different mechanisms for @xmath0-ray emission @xcite . in examining the jet bend scenario of @xcite , (
* hereafter paper i ) used egret identifications to compare @xmath0-ray - loud and @xmath0-ray - quiet agn with respect to their jet - bending characteristics .
their results showed a tendency in @xmath0-ray - quiet sources to both have more jet bends and more pronounced jet - bend angles at parsec scales .
@xcite did not find evidence to support this scenario , but their study was limited in only examining parsec - to - kiloparsec jet misalignment angles .
the jet bend proposition was revisited in @xcite and @xcite with inconclusive results ( limited to 4 egret candidates and 30 lat sources respectively ) .
similarly , @xcite found that no significant difference in jet - bend angles between @xmath0-ray - loud and @xmath0-ray - quiet populations ( limited to 19 lat sources ) .
utilizing lat s superior @xmath0-ray detection capabilities and a larger sample size , we re - examine the suggestion that jet - bending may be a significant factor for the detection of @xmath0-ray emission in agn .
we follow the analysis of paper i , testing the statistical significance of jet - bend angles , as well as the number of bends between @xmath0-ray - loud and @xmath0-ray - quiet agn .
we use the ` clean sample ' of the second lat agn catalog ( 2lac , * ? ? ?
* ) and vlbi images accessible in the published literature or from the radio fundamental catalog ( rfc , version ` rfc2013d ' ) .
we examine the statistics of two samples , a large but inhomogeneous sample and a smaller but homogeneous subsample based on ( * ? ? ?
* hereafter mojave ) .
in order to investigate the significance of jet - bending between @xmath0-ray - loud and @xmath0-ray - quiet agn populations , we surveyed the literature for vlbi images and measured source properties .
the most recent version of the rfc contains 8310 compact radio sources studied in 5719 vlbi observing sessions , making it the most complete catalog of radio sources with available vlbi data .
we have compared the rfc with the crates catalog @xcite , a nearly uniform survey of of flat - spectrum radio sources ( @xmath1 ) containing flux densities and spectral indices for over 11,000 objects . by cross - referencing data from the crates catalog to sources in the rfc ,
we obtain a large pool of sources to apply selection criteria and draw upon in order to define a ` rfc - based sample ' for the purpose of our investigation .
there are a number of considerations in formulating our selection criteria for the rfc - based sample . a true cross - identification between radio / x - ray and @xmath0-ray counterparts
can only be determined when there is a correlation in variability in both bands . in practice ,
cross - identifications were made only for 28 sources in the 2-year fermi - lat sources catalog ( 2fgl , * ? ? ? * ) .
as such , most sources identified as @xmath0-ray - loud in 2lac are high - confidence statistical associations determined by three association methods using the detected positions and uncertainties of radio / x - ray and @xmath0-ray emissions within a source s 95% error ellipse @xcite , along with the physical properties of likely candidates .
the determination of a high - confidence association in 2lac is also subject to a number of considerations @xcite .
candidate gamma - ray sources are ` detected ' in 2fgl if a region of concentrated flux against its background obtains a test statistic ts @xmath2 25 ( corresponding to over 4 @xmath3 significance ) .
there is low availability of spectroscopic information for radio sources in southern declinations .
there are in some case multiple associations within a gamma - ray source s 95% error ellipse .
a candidate s gamma - ray spectral intensity may not lie within lat s energy - dependent sensitivity curve .
candidates with weak radio flux would also have reduced likelihoods of association as they become indistinguishable from other weak radio sources within the lat error ellipse .
these factors would cause bias to higher unassociated agn despite possibly having detectable @xmath0-ray flux . to address these issues , we employ the ` clean sample ' subset of 2lac where sources are excluded if they presented difficulties in the analysis of making an association .
we also apply a low threshold to flux densities to account for the possibility of false non - associations below the cutoff .
there is also possible bias introduced by @xmath0-ray attenuation due to the extragalactic background light ( ebl ) , resulting in a lower rate of association at higher redshifts .
predictions from ebl models show that @xmath0-rays with energies below @xmath4 10 gev from redshifts up to z @xmath4 3 do not undergo significant attenuation @xcite , and @xcite find no redshift dependence in the flux ratio of @xmath0-ray photons f(@xmath2 10 gev)/f(@xmath2 1 gev ) across blazar subclasses from z = 0 to above z = 2 . applying an upper limit to redshift
can eliminate the possibility of this bias in our sample .
furthermore , errors in modelling the diffuse @xmath0-ray background near the galactic ridge or in nearby interstellar cloud regions may introduce false associations between @xmath0-ray and radio emissions belonging to two distinct sources , identifying ` @xmath0-ray - loud ' agns that are actually @xmath0-ray - quiet .
however , 2lac selection criteria include only sources with high galactic latitudes ( @xmath5 ) , eliminating the possibility of false or non - associations occuring near the galactic ridge . in review of these factors in @xmath0-ray detection and source association ,
our source selection criteria are therefore as follows : 1
. identified in rfc ; 2 .
galactic latitude @xmath5 ; 3 .
total 4.85 ghz flux density @xmath6 jy ; 4 .
redshift z @xmath7 2 ; 5 .
spectral index @xmath8 ; 6 .
included in clean sample if included in 2lac ; 7 . associated with vlbi images in literature .
these criteria define our rfc - based sample , consisting of 351 agn in total .
the sample contains 151 @xmath0-ray - loud agn from the clean sample , a subset of sources from the second lat catalog that possess high - confidence associations between radio and @xmath0-ray counterparts @xcite ( see table [ tab : gl_list ] ) .
a @xmath0-ray - quiet subsample was defined by the 200 agn not detected by lat ( see table [ tab : gq_list ] ) .
we consider a subsample of the rfc - based sample defined by sources existing in mojave to investigate concerns of sample inhomogeneity @xcite .
the ` mojave subsample ' is an unbiased subset since the sources included conform to our selection criteria for the rfc - based sample .
the sample contains 65 @xmath0-ray - loud agn and 29 @xmath0-ray - quiet agn , giving a total sample of 94 agn .
the sources were observed using the vlba and processed uniformly .
the selection criteria used in mojave are also relatively homogeneous , only departing slightly from uniformity in the criterion for radio flux density between agn at northern ( 1.5 jy ) and southern ( 2.0 jy ) declinations to account for differences in instrument sensitivity .
the high fidelity imaging of this sample makes it ideal for jet - bend measurements .
the sample is also relatively large , which is advantageous from a statistical standpoint .
the data from mojave were not available when paper i was published .
all mojave agn were imaged at 15 ghz in the period august 1994 to september 2007 @xcite .
@xcite suggest two scenarios where jet bending could influence @xmath0-ray detection in radio - loud agn . if @xmath0-ray emission is produced within an agn jet , and if the jet bends significantly downstream of the @xmath0-ray emitting region , there is a possibility that the @xmath0-ray emission is beamed away from our line of sight ( doppler - dimmed ) whilst radio emission is beamed towards us . in this case
we would observe a @xmath0-ray - quiet but radio - loud agn .
the other possibility is that there may be agn with jets either with no bends or with bends upstream of the @xmath0-ray - emitting region , producing aligned radio and @xmath0-ray beaming cones . in this case
, we would observe a @xmath0-ray and radio - loud agn .
such possibilities may account for @xmath0-ray observations not otherwise explained by relativistic beaming .
two other simple scenarios may also be considered .
first , objects for which we lie within the @xmath0-ray beaming cone but not the radio beaming cone .
such sources would be observable to us as @xmath0-ray - loud , radio quiet agn .
however , it is reasonable to assume that substantial radio emission occurs in regions of @xmath0-ray emission , making this scenario unlikely .
second , agn jets for which both @xmath0-ray and radio beaming cones are misaligned with our line of sight . these sources will likely not be identified in @xmath0-rays and may appear as weak radio sources . in our approach ,
the number of parsec - scale jet bends and the maximum parsec - scale jet - bending angles are recorded for agn with vlbi images that present a discernible jet structure from which jet - bend data may be extracted .
the hypothesis of @xcite allows the possibility that jet - bending could be significant on kiloparsec scales .
however , neither @xcite nor paper i find a significant difference in the parsec - to - kiloparsec misalignment angle between @xmath0-ray - loud and @xmath0-ray - quiet populations .
in addition , paper i does not find a statistically significant difference in the number of kiloparsec - scale jet - bends .
more recent investigations assert that @xmath0-ray emission originates in the parsec - scale region , near the core , although the precise region continues to be discussed @xcite . in the scenario proposed in these studies ,
both @xmath0-ray and radio emission may be produced far upstream of any kiloparsec - scale bending , allowing the detection of both kinds of emission regardless of the existence of kiloparsec bending .
kiloparsec - scale bending does not appear to impact @xmath0-ray detection in agn , consistent across a number of studies , and is not explored here .
we applied the following method to our jet - bending measurements . for a given agn
, we compared vlbi images available from the rfc and in other major surveys .
considering the resolution and common features between epochs , we determined the agn s jet structure that could be discernable by eye . in this process , we assume that an abrupt change in the apparent direction of a jet , associated with a ` shock ' region of high flux density , corresponds to the bending between two jet sections .
we selected the best vlbi image to derive its jet properties , with preference always given to available mojave images due to their superior image quality .
the initial angle of the jet from the core was measured east of north in the equatorial coordinate system .
the angles of any existing bends in the jet are then measured ( in the same sense ) with respect to the previous jet trajectory closer to the core .
the number of bends and the maximum jet - bending angle were then recorded for each source , shown in tables [ tab : gl_list ] and [ tab : gq_list ] .
there are a number of factors affecting an agn s derived jet properties .
inhomogeneity is introduced from the variable resolution and sensitivity of vlbi instrumentation , the choice of radio frequency for imaging , and different image epochs for different sets of images associated with a particular agn . as such
, the low fidelity of some vlbi images may affect the consistency of jet measurements across the whole sample . despite this , using the best available vlbi images for each agn will provide the most accurate jet - bending properties to retest the jet - bending hypothesis .
the ability to resolve jet components also diminishes with decreasing radio flux density . in weak radio sources ,
this is particularly evident in the reduced ability to resolve low flux density jet components ( and therefore jet bending ) at large distances from the core .
deflections in jets may not be obvious at small jet - bending angles , and where there are jet - bends , they may not occur at a single , well - defined point along the jet .
the comparison between multiple vlbi images of agn across different imaging frequencies and epochs also allows us to obtain jet properties that reflect the jet structure from the observer s point of view as accurately as possible .
furthermore , bright individual features with different trajectories within or around a jet may also give the false appearance of jet bending in jet structure in certain imaging epochs @xcite .
comparing vlbi images at different epochs assists in identifying bright regions that may temporarily distort the appearance of the jet structure .
measurements are subject to image fidelity and morphological interpretation , however , these considerations should provide a robust result when testing the significance of jet bending in @xmath0-ray detection compared to the findings of previous , more - limited studies .
we employ a number of statistical tests to explore differences between @xmath0-ray - loud and @xmath0-ray - quiet populations in our sample . a two - sided kolmogorov - smirnov
( k - s ) two - sample test was used to determine the statistical significance of differences at 4.85 ghz flux density and maximum jet - bend angle distributions between @xmath0-ray - loud and @xmath0-ray - quiet agn populations .
since the values for the number of jet bends are discrete , a @xmath9 test was applied to characterize differences between the distributions in this case . 13 @xmath0-ray - loud and 32 @xmath0-ray - quiet sources did not show discernible jets and thus were omitted from the statistical testing of jet properties .
for all tests , the difference between sample populations is considered significant for a given property when , assuming the null hypothesis represents both samples being drawn from the same parent population , the probability of observing a test statistic at least as extreme as the one obtained ( @xmath10 ) is less than 0.05 ( 95% confidence that the null hypothesis can be rejected ) .
these results are summarized in table [ tab : agn_stat ] .
figures [ fig : lfx_agn_gl ] and [ fig : lfx_agn_gq ] show similar distributions for agn total flux densities at 4.85 ghz .
a k - s test determines that the probability that the two samples have been drawn from the same parent population is @xmath11 , finding no statistically significant difference between the two distributions .
the result gives confidence that there no biases in lat detections that may skew jet - bending statistics .
we now consider the jet - bending statistics of @xmath0-ray - loud and @xmath0-ray - quiet samples .
the distributions between samples for the number of jet bends ( figures [ fig : nb_agn_gl ] and [ fig : nb_agn_gq ] ) show no significant differences .
a @xmath9 test finds a statistically insignificant probability of @xmath12 ( @xmath13 with 2 degrees of freedom ) .
similarly , histograms for the parsec - scale maximum jet - bend angle ( figures [ fig : mb_agn_gl ] and [ fig : mb_agn_gq ] ) demonstrate no significant difference between the two populations .
the probability that the two samples belong to the same parent population is @xmath14 .
we find that there are no statistically significant differences between @xmath0-ray - loud and @xmath0-ray - quiet agn populations for any of the tested properties .
the results of the statistical tests on the mojave subsample are given in table [ tab : agn_stat ] .
the @xmath0-ray - loud and @xmath0-ray - quiet distributions for these properties are shown in figure [ fig : lmn_mjv ] .
we have investigated the suggestion that @xmath0-ray - quiet agn have a larger tendency for jet bending than @xmath0-ray - loud agn @xcite .
we present a statistical analysis of a large sample of agn that represents the population of radio - loud agn , using data from the fermi lat instrument .
we also conducted an analysis of a homogeneous subsample ( mojave ) to address the inhomogeneity of selection criteria and vlbi image quality in the rfc - based sample .
our analysis of the rfc - based sample shows that jet - bending does not play a significant role in predicting @xmath0-ray detectability in radio - loud agn . as such our results
do not support the suggestion of @xcite .
this conclusion is in contradiction to paper i , and supports the findings of @xcite .
the contrast of results obtained by new data with those of paper i reveals the strong dependence on the fidelity and homogeneity of vlbi images in jet - bend measurements .
the limited quality and homogeneity of vlbi images available at the time of study appears to be the reason for the reported conclusions in paper i. the results of the mojave subsample generally support the conclusions drawn from the rfc - based sample .
statistically insignificant results were found for both the number of parsec - scale jet bends and maximum parsec - scale jet bend , indicating that the rfc - based sample is not adversely affected by vlbi image inhomogeneity and giving additional confidence to the null result .
we thank the referee for providing helpful comments and suggestions .
this research has made use of data from the radio fundamental catalog ( rfc ) maintained by l. petrov , solution rfc2013d ( unpublished , available on the web at http://asrtogeo.org/vlbi/solutions/rfc2013d ) ; the mojave database , maintained by the mojave team ( lister et al .
2009 , aj , 137 , 3718 ) ; the crates catalog ( healey et al .
2007 , apjs , 171 , 61 ) ; and the nasa / ipac extragalactic database ( ned ) , operated by the jet propulsion laboratory , california institute of technology , under contract with the national aeronautics and space administration .
abdo , a. a. , ackermann , m. , ajello , m. , et al .
2010b , , 723 , 1082 ackermann , m. , ajello , m. , allafort , a. , et al .
2011a , 741 , 30 ackermann , m. , ajello , m. , allafort , a. , et al .
2011b , 743 , 171 aller , m. f. , aller , h. d. , & hughes , p. a. 1996 , in iau symposium 175 , extragallactic radio sources , ed .
r. d. ekers , c. fanti , & l. padrielli ( dordrecht : kluwer ) , 283 balsara , d. s. , & norman , m. l. 1992 , , 393 , 631 beasley , a. j. , gordon , d. , peck , a. b. , et al .
2002 , , 141 , 13 ( rfc ) bondi , m. , padrielli , l. , fanti , r , et al .
1996 , , 308 , 415 bower , g. c. , backer , d. c. , wright , m. , et al .
1997 , , 484 , 118 dodson , r. , fomalont , e. b. , wiik , k. , et al .
2008 , , 175 , 314 fey , a. l. , & charlot , p. 1997 , , 111 , 95 fey , a. l. , & charlot , p. 2000 , , 128 , 17 fey , a. l. , clegg , a. w. , fomalont , e. b. 1996 , , 105 , 299 fichtel , c. e. , bertsch , d. l. , chiang , j. , et al .
1994 , , 94 , 551 fomalont , e. b. , frey , s. , paragi , z. , et al .
2000 , , 131 , 95 fomalont , e. b. , petrov , l. , macmillan , d. s. , gordon , d. , & ma , c. 2003 , , 126 , 2562 ( rfc ) franceschini , a. , rodighiero , g. , & vaccari , m. 2008 , , 487 , 837 hartman , r. c. , bertsch , d. l. , bloom , s. d. , et al . 1999 , , 123 , 79 healey , s. e. , romani , r. w. , taylor , g. b. , et al .
2007 , , 171 , 61 ( crates ) henstock , d. r. , browne , w. a. , wilkinson , p. n. , et al .
1995 , , 100 , 1 homan , d. c. 2012 , int .
, 8 , 163 hughes , p. a. , aller , m. f. , & aller , h. d. 2011 , , 735 , 81 immer , k. , brunthaler , a. , reid , m. j. , et al .
2011 , , 194 , 25 ( rfc ) impey , c. 1996 , , 112 , 2667 jorstad , s. g. , marscher , a. p. , mattox , j. r. , et al .
2001 , , 556 , 738 kellermann , k. i. , lister , m. l. , homan , d. c. , et al .
2004 , , 609 , 539 kovalev , y. y. , aller , h. d. , aller , m. f. , et al .
2009 , , 696 , l17 kovalev , y. y. , petrov , l. , fomalont , e. b. , & gordon , d. 2007 , , 133 , 1236 ( rfc ) lhteenmki , a. , & valtaoja , e. 1999 , in asp conf .
159 , bl lac phenomenon , ed .
l. o. takalo & sillanp ( san francisco : asp ) , 213 lhteenmki , a. , & valtaoja , e. 2003 , , 590 , 95 lebedev , s. v. , ampleford , d. , ciardi , a. , et al .
2004 , , 616 , 988 lee , s .- s . , lobanov , a. p. , krichbaum , t. p. , et al .
2008 , , 136 , 159 len - tavares , j. , valtaoja , e. , tornikoski , m. , lahteenmki , a. & nieppola , e. 2011 , , 532 , a146 linford , j. d. , taylor , g. b. , romani , r. w. , et al .
2011 , , 726 , 16 linford , j. d. , taylor , g. b. , romani , r. w. , et al .
2012 , , 744 , 177 lister , m. l. , aller , h. d. , aller , m. f. , et al . 2009a , , 137 , 3718 ( mojave ) lister , m. l. , aller , m. f. , aller , h. , et al .
2011 , , 742 , 27 lister , m. l. , cohen , m. h. , homan , d. c. , et al . 2009b , , 138 , 1874 mattox , j. r. , schachter , j. , molnar , l. , hartman , r. c. , & patnaik , a. r. 1997a , , 481 , 95 mattox , j. r. , wagner , s. j. , malkan , m. , et al .
1997b , , 476 , 692 mcke , a. , pohl , m. , reich , p. , et al .
1997 , , 320 , 33 nieppola , e. , tornikoski , m. , valtaoja , e. , et al .
2011 , , 535 , a69 nolan , p. l. , abdo , a. a. , ackermann , m. , et al .
2012 , , 199 , 31 ojha , r. , fey , a. l. , charlot , p. , et al .
2005 , , 130 , 2529 ojha , r. , fey , a. l. , johnston , k. j. , et al .
2004 , , 127 , 3609 ojha , r. , kadler , m. , bck , m. , et al .
2010 , , 519 , a45 pearson , t. j. , & readhead , a. c. s. 1988 , , 328 , 114 petrov , l. , kovalev , y. y. , fomalont , e. , & gordon , d. 2005 , , 129 , 1163 ( rfc ) petrov , l. , kovalev , y. y. , fomalont , e. b. , & gordon , d. 2006 , , 131 , 1872 ( rfc ) petrov , l. , kovalev , y. y. , fomalont , e. b. , & gordon , d. 2008 , , 136 , 580 ( rfc ) petrov , l. 2011 , , 142 , 105 ( rfc ) petrov , l. 2012 , , 416 , 1097 ( rfc ) petrov , l. 2013 , , 146 , 5 ( rfc ) petrov , l. , kovalev , y. y. , fomalont , e. b. , & gordon , d. , 2011b , , 142 , 35 ( rfc ) petrov , l. , phillips , c. , bertarini , a. , murphy , t. , & sadler , e. m. 2011c , , 414 , 2528 ( rfc ) petrov , l. , & taylor , g. b. 2011d , , 142 , 89 ( rfc ) pushkarev , a. b. , kovalev , y. y. , & lister , m. l. 2010 , , 722 , l7 pushkarev , a. b. , kovalev , y. y. , lister , m. l. , & savolainen , t. 2009 , , 507 , l33 richards , j. l. , max - moerbeck , w. , pavlidou , v. , et al . 2011 , , 194 , 29 salamon , m. h. , & stecker , f. w. 1994 , , 430 , l21 salamon , m. h. , & stecker , f. w. 1998 , , 493 , 547 savolainen , t. , homan , d. c. , hovatta , t. , et al .
2010 , , 512 , a24 scott , w. k. , fomalont , e. b. , horiuchi , s. , et al .
2004 , , 155 , 33 shen , z .- q . ,
wan , t .- s . ,
moran , j. m. , et al .
1998 , , 115 , 1357 sikora , m. , stawarz , . , nalewajko , k. , & madejski , g. m. 2009 , , 704 , 38 stecker , f. w. , de jager , o. c. , & salamon , m. h. 1992 , , 390 , l49 stecker , f. w. , malkan , m. a. , & scully , s. t. 2006 , , 648 , 774 ( erratum , 658 , 1392 [ 2007 ] ) stickel , m. , meisenheimer , k. , & khr , h. 1994 , a&as , 105 , 211 taylor , g. b. , healey , s. e. , helmboldt , j. f. , et al . 2007 , , 671 , 1355 taylor , g. b. , vermeulen , r. c. , pearson , t. j. , et al .
1994 , , 95 , 345 taylor , g. b. , vermeulen , r. c. , readhead , a. c. s. , et al .
1996 , , 107 , 37 tingay , s. j. , jauncey , d. l. , king , e. a. , et al .
2003 , , 55 , 351 tingay , s. j. , jauncey , d. l. , reynolds , j. e. , et al .
1996 , , 111 , 718 tingay , s. j. , murphy , d. w. , & edwards , p. g. 1998 , , 500 , 673 ( paper i ) tingay , s. j. , reynolds , j. e. , tzioumis , a. k. , et al .
2002 , , 141 , 311 valtaoja , e. , & tersranta , h. 1995 , , 297 , l13 von montigny , bertsch , d. l. , chiang , j. , et al .
1995 , , 299 , 680 xu , w. , readhead , a. c. s. , pearson , t. j. , polatidis , a. g. , & wilkinson , p. n. 1995 , , 99 , 297
+ + j0049@xmath155738 & 1.338 & 302/@xmath1528 & 1 & 28 & 10 , 12@xmath16 , 14 + j0051@xmath150650 & 0.841 & 307 & 0 & 0 & 14 + j0116@xmath151136 & 1.488 & 325 & 0 & 0 & 2 , 4 , 6 , 14@xmath16 + j0120@xmath152701 & 1.000 & 156 & 0 & 0 & 6 , 11 , 14@xmath16 , 15 + j0136@xmath174751 & 2.016 & 330 & 0 & 0 & 2 , 4 , 6 , 8 , 9@xmath16 , 13 , 14 , 18 + j0137@xmath152430 & 0.956 & 75 & 0 & 0 & 2 , 6 , 11@xmath16 , 14 + j0141@xmath150928 & 0.940 & 224 & 0 & 0 & 2 , 3 , 6@xmath16 , 14 + j0145@xmath152733 & 0.833 & 53 & 0 & 0 & 14 + j0204@xmath151701 & 1.350 & 12 & 0 & 0 & 2 , 4@xmath16 , 6 , 14 + j0204@xmath171514 & 3.073 & 303 & 0 & 0 & 1 , 2 , 4 , 6 , 8 , 9@xmath16 , 14 , 18 + j0006@xmath150623 & 2.463 & 279 & 0 & 0 & 2 , 3 , 6 , 8 , 9@xmath16 , 14 + j0010@xmath171724 & 0.989 & 266 & 0 & 0 & 4@xmath16 , 14 + j0013@xmath174051 & 1.035 & 331 & 0 & 0 & 2 , 3 , 6 , 14@xmath16 , 20 + j0019@xmath177327 & 1.583 & 144 & 0 & 0 & 2 , 3 , 6 , 8 , 9@xmath16 , 13 , 14 , 18 + j0024@xmath154202 & 2.036 & 297 & 0 & 0 & 14 + j0029@xmath173456 & 1.182 & 56 & 0 & 0 & 3 , 6 , 14@xmath16 , 16 + j0038@xmath174137 & 1.144 & 113 & 0 & 0 & 6@xmath16 , 7 , 14 + j0042@xmath172320 & 1.604 & 118 & 0 & 0 & 2 , 4 , 6 , 14@xmath16 + j0051@xmath154226 & 0.926 & 40 & 0 & 0 & 14 + j0057@xmath173021 & 0.914 & 309 & 0 & 0 & 6@xmath16 , 14 + total 4.85 ghz flux density & k - s & 0.121 & no + maximum pc - scale jet bend & k - s & 0.998 & no + number of pc - scale jet bends & @xmath9 & 0.945 & no +
total 4.85 ghz flux density & k - s & 0.145 & no + maximum pc - scale jet bend & k - s & 0.948 & no + number of pc - scale jet bends & @xmath9 & 0.402 & no + | we investigate the hypothesis that @xmath0-ray - quiet agn have a larger tendency for jet bending than @xmath0-ray - loud agn , revisiting the analysis of @xcite . we perform a statistical analysis using a large sample of 351 radio - loud agn along with @xmath0-ray identifications from the fermi large area telescope ( lat ) .
our results show no statistically significant differences in jet - bending properties between @xmath0-ray - loud and @xmath0-ray - quiet populations , indicating that jet bending is not a significant factor for @xmath0-ray detection in agn . |
chiral effective field theory ( eft ) as proposed in the pioneering works of weinberg @xcite is a powerful tool for the derivation of nuclear forces . in this scheme
there is an underlying power counting which allows to improve calculations systematically by going to higher orders in a perturbative expansion .
in addition , it is possible to derive two- and corresponding three - nucleon forces as well as external current operators in a consistent way . over the last decade or
so it has been demonstrated that the nucleon - nucleon ( @xmath3 ) interaction can be described to a high precision within the chiral eft approach @xcite . following the original suggestion of weinberg , in these works the power counting
is applied to the @xmath3 potential rather than to the reaction amplitude .
the latter is then obtained from solving a regularized lippmann - schwinger equation for the derived interaction potential .
the @xmath3 potential contains pion - exchanges and a series of contact interactions with an increasing number of derivatives to parameterize the shorter ranged part of the @xmath3 force . for reviews we refer the reader to refs .
@xcite . in the present contribution
i focus on recent investigations by the groups in bonn - jlich and munich on the baryon - baryon interaction involving strange baryons , performed within chiral eft @xcite . in these works the same scheme as applied in ref .
@xcite to the @xmath3 interaction is adopted .
first i discuss the application to the strangeness @xmath0 sector ( @xmath4 , @xmath5 ) . here
the extension of our study @xcite to next - to - leading order ( nlo ) is in progress @xcite and a first glimpse on the ( still preliminary ) achieved results for the @xmath4 and @xmath5 interactions will be given .
then i report results of a study on the strangeness @xmath6 sector , i.e. for the @xmath7 , @xmath8 , and cascade - nucleon ( @xmath9 ) interactions .
predictions obtained at leading order ( lo ) @xcite are reviewed and implications for the @xmath2-dibaryon are discussed , based on our framework , in the light of recent lattice qcd calculations where evidence for the existence of such a state was found . at lo in the power counting , as considered in the aforementioned investigations @xcite , the baryon - baryon potentials involving strange baryons consist of four - baryon contact terms without derivatives and of one - pseudoscalar - meson exchanges , analogous to the @xmath3 potential of @xcite .
the potentials are derived using constraints from @xmath10 flavor symmetry . at nlo one
gets contributions from two - pseudoscalar - meson exchange diagrams and from four - baryon contact terms with two derivatives @xcite .
the paper is structured as follows : in sect .
2 a short overview of the chiral eft approach is provided . in sect .
3 results for the @xmath4- and @xmath5 interactions obtained to nlo are presented . in sect .
4 results for the the @xmath6 ( @xmath11 , @xmath9 @xmath12 ) systems are briefly reviewed and connection is made with lattice qcd results for the @xmath2-dibaryon case .
the paper ends with a short summary .
the derivation of the chiral baryon - baryon potentials for the strangeness sector at lo using the weinberg power counting is outlined in refs .
details for the nlo case will be presented in a forthcoming paper @xcite , see also @xcite .
the lo potential consists of four - baryon contact terms without derivatives and of one - pseudoscalar - meson exchanges while at nlo contact terms with two derivatives arise , together with contributions from ( irreducible ) two - pseudoscalar - meson exchanges .
the spin- and momentum structure of the potentials resulting from the contact terms to lo is given by @xmath13 in the notation of @xcite where the @xmath14 s are so - called low - energy coefficients ( lecs ) that need to be determined by a fit to data .
due to the imposed @xmath15 constraints there are only five independent lecs for the @xmath3 and the @xmath16 sectors together , as described in ref .
@xcite where also the relations between the various @xmath14 s are given .
a sixth lec is , however , present in the strangeness @xmath6 channels with isospin @xmath17 . in
next - to - leading order one gets the following spin- and momentum structure : @xmath18 the transferred and average momentum , @xmath19 and @xmath20 , are defined in terms of the final and initial center - of - mass ( c.m . )
momenta of the baryons , @xmath21 and @xmath22 , as @xmath23 and @xmath24 . the @xmath25 s ( actually @xmath14 s )
are additional lecs .
performing a partial wave projection and imposing again @xmath15 symmetry one finds that in case of the @xmath16 interaction there are eight new lecs entering the @xmath26-waves and @xmath26-@xmath27 transitions , respectively , and ten coefficents in the @xmath28-waves .
there are further ( four ) lecs that contribute only to the @xmath6 system .
the spin - space part of the one - pseudoscalar - meson - exchange potential is similar to the static one - pion - exchange potential ( recoil and relativistic corrections give higher order contributions ) and follows from the @xmath15 invariant pseudoscalar - meson baryon interaction lagrangian with the appropriate symmetries as discussed in @xcite : @xmath29 here , @xmath30 is the mass of the exchanged pseudoscalar meson .
the coupling constants @xmath31 at the various baryon - baryon - meson vertices are fixed by the imposed @xmath10 constraints and tabulated , e.g. , in @xcite . they can be expressed in terms of @xmath32 ( @xmath33 , @xmath34 mev ) and @xmath35 , the so - called @xmath36-ratio , for which we adopted the @xmath37 value ( @xmath38 ) . note that we use the physical masses of the exchanged pseudoscalar mesons .
thus , the explicit @xmath10 breaking reflected in the mass splitting between the pseudoscalar mesons is taken into account .
the @xmath39 meson was identified with the octet @xmath39 ( @xmath40 ) and its physical mass was used .
the two - pseudoscalar - meson - exchange potential can be found in refs .
@xcite .
the reaction amplitudes are obtained from the solution of a coupled - channels lippmann - schwinger ( ls ) equation for the interaction potentials : @xmath41 the label @xmath42 indicates the particle channels and the label @xmath43 the partial wave .
@xmath44 is the pertinent reduced mass .
the on - shell momentum in the intermediate state , @xmath45 , is defined by @xmath46 .
relativistic kinematics is used for relating the laboratory energy @xmath47 of the hyperons to the c.m . momentum .
we solve the ls equation in the particle basis , in order to incorporate the correct physical thresholds .
depending on the specific values of strangeness and charge up to six baryon - baryon channels can couple .
for the @xmath0 sector where a comparison with scattering data is possible the coulomb interaction is taken into account appropriately via the vincent - phatak method @xcite .
the potentials in the ls equation are cut off with a regulator function , @xmath48 $ ] , in order to remove high - energy components of the baryon and pseudoscalar meson fields @xcite .
we consider cut - off values in the range 500 , ... , 700 mev , similar to what was used for chiral @xmath3 potentials @xcite .
, @xmath49 , @xmath50 , @xmath51 and @xmath52 as a function of @xmath53 .
the green ( grey ) band shows the chiral eft results to lo for variations of the cut - off in the range @xmath54 550@xmath55700 mev , while the red ( black ) band are results to nlo for @xmath54 500@xmath55700 mev .
the dashed curve is the result of the jlich 04 @xcite meson - exchange potential .
, title="fig:",height=211 ] , @xmath49 , @xmath50 , @xmath51 and @xmath52 as a function of @xmath53 .
the green ( grey ) band shows the chiral eft results to lo for variations of the cut - off in the range @xmath54 550@xmath55700 mev , while the red ( black ) band are results to nlo for @xmath54 500@xmath55700 mev .
the dashed curve is the result of the jlich 04 @xcite meson - exchange potential . ,
title="fig:",height=211 ] , @xmath49 , @xmath50 , @xmath51 and @xmath52 as a function of @xmath53 .
the green ( grey ) band shows the chiral eft results to lo for variations of the cut - off in the range @xmath54 550@xmath55700 mev , while the red ( black ) band are results to nlo for @xmath54 500@xmath55700 mev .
the dashed curve is the result of the jlich 04 @xcite meson - exchange potential . , title="fig:",height=211 ] , @xmath49 , @xmath50 , @xmath51 and @xmath52 as a function of @xmath53 .
the green ( grey ) band shows the chiral eft results to lo for variations of the cut - off in the range @xmath54 550@xmath55700 mev , while the red ( black ) band are results to nlo for @xmath54 500@xmath55700 mev .
the dashed curve is the result of the jlich 04 @xcite meson - exchange potential .
, title="fig:",height=211 ] , @xmath49 , @xmath50 , @xmath51 and @xmath52 as a function of @xmath53 .
the green ( grey ) band shows the chiral eft results to lo for variations of the cut - off in the range
@xmath54 550@xmath55700 mev , while the red ( black ) band are results to nlo for @xmath54 500@xmath55700 mev .
the dashed curve is the result of the jlich 04 @xcite meson - exchange potential .
, title="fig:",height=211 ]
the imposed @xmath10 flavor symmetry implies that at lo five independent lecs contribute to the @xmath16 interaction @xcite .
these five contact terms were determined in @xcite by a fit to the @xmath16 scattering data . already in that scenario a fairly reasonable description of the 35 low - energy @xmath16 scattering data could be achieved for cutoff values @xmath56 mev and for natural values of the lecs . at nlo
there are eight new contact terms contributing to the @xmath26-waves and the @xmath57-@xmath58 transition , and ten in the @xmath28-waves .
once again the corresponding lecs were fixed by fitting to the data .
the results obtained at nlo are presented in fig .
[ fig:1 ] ( black ( red ) bands ) , together with those at lo ( grey ( green ) bands ) .
the bands represent the variation of the cross sections based on chiral eft within the considered cutoff region
. for comparison also results for the jlich 04 @xcite meson - exchange models are shown ( dashed line ) . obviously , and as expected , the energy dependence exhibited by the data can be significantly better reproduced within our nlo calculation .
this concerns in particular the @xmath59 channel .
but also for @xmath60 the nlo results are now well in line with the data even up to the @xmath5 threshold .
furthermore , one can see that the dependence on the cutoff mass is strongly reduced in the nlo case .
note that in case of lo as well as at nlo no @xmath15 constraints from the @xmath3 sector were imposed in the fitting procedure .
the leading order @xmath15 breaking in the one - boson exchange diagrams ( coupling constants ) is ignored .
besides an excellent description of the @xmath16 data the chiral eft interaction also yields a correctly bound hypertriton , see table [ tabb ] .
indeed this binding energy had to be included in the fitting procedure because otherwise it would have not been possible to fix the relative strength of the ( s - wave ) singlet- and triplet contributions to the @xmath60 interaction .
table [ tabb ] lists also results for two meson - exchange potentials , namely of the jlich 04 model @xcite and the nijmegen nsc97f potential @xcite , which both reproduce the hypertriton binding energy correctly . obviously , the scattering lengths predicted at nlo are larger than those obtained at lo and now similar to the values of the meson - exchange potentials .
the @xmath61 scattering length in the @xmath57 partial wave is positive , as it was already the case for our lo potential , indicating a repulsive interaction in this channel .
@xmath16 singlet ( s ) and triplet ( t ) scattering lengths ( in fm ) and the hypertriton binding energy , @xmath62 ( in mev ) . [ cols="^,>,>,>,>,>",options="header " , ] calculations for the four - body hypernuclei @xmath63 and @xmath64 based on those interactions are reported in ref .
in this section i review results obtained for the @xmath6 sector , specifically for the coupled @xmath65 system , within chiral eft at lo @xcite . as mentioned above , at lo one additional lec occurs in this specific channel with @xmath17 which , in principle , should be determined from experimental information available for this sector .
however , the scarce data ( @xmath66 and @xmath67 cross sections @xcite ) are afflicted with large uncertainties and , thus , do not allow to establish reliably its value as found by us @xcite .
some results for strangeness @xmath6 published in @xcite are reproduced here in fig .
[ fig:2 ] . as
before the band reflects the dependence of the results on variations of the cutoff @xmath68 .
the cutoff was varied between 550 and 700 mev ( like in case of the lo @xmath16 potential ) and under the constraint that the @xmath7 @xmath69 scattering length remains practically unchanged @xcite . as reference
we have taken the result for @xmath70 mev and with the value of the additional lec fixed in such a way that @xmath71 .
the scattering length turned out to be @xmath72 fm @xcite .
analyses of the measured binding energy of the double - strange hypernucleus @xmath73 @xcite suggest that the @xmath74 scattering length could be in the range of @xmath75 to @xmath76 fm @xcite . a first determination of the scattering length utilizing data on the @xmath7 invariant mass from the reaction @xmath77 @xcite led to the result @xmath78 fm @xcite .
one particular interesting aspect of the coupled @xmath65 system is the @xmath2-dibaryon , a deeply bound 6-quark state with @xmath79 predicted by jaffe from the bag model @xcite , that should occur in this channel .
so far none of the experimental searches for the @xmath2-dibaryon let to convincing signals @xcite .
however , recently evidence for a bound @xmath2-dibaryon was claimed based on lattice qcd calculations @xcite . extrapolations of those computations , performed for @xmath80 mev , to the physical pion mass suggest that the @xmath2-dibaryon could be either loosely bound or move into the continuum @xcite . unfortunately , because the mentioned additional contact term can not be reliably fixed from the data in the @xmath6 sector , no immediate predictions can be made for the @xmath2-dibaryon within chiral eft . however , the framework can be used as a tool to study the dependence of its conjectured binding energy on the masses of the involved hadrons and , thus , allows for an alternative extrapolation of the results , obtained in lattice qcd calculations at unphysical meson- and baryon masses , down to the physical point . in particular one can fine - tune the `` free '' lec to produce a bound @xmath2 with a given binding energy for meson- and baryon masses corresponding to the lattice simulations , and then study its properties @xcite for physical masses . note that at lo the lecs do not depend on the meson masses ( strictly speaking , on the quark masses ) .
variations of the meson masses enter only in the potential in eq .
( [ pot ] ) , those of the baryon masses ( via @xmath44 ) in the ls equation ( [ ls ] ) .
let me first consider individual variations of the masses of the involved particles . to begin with
i examine the dependence of the @xmath2 binding energy on the pion mass @xmath81 and keep all other ( meson and baryon ) masses at their physical value .
corresponding results are displayed in fig .
[ fig : mpi ] ( left ) . adjusting the lec so that a @xmath2 binding energy of 13.2 mev is predicted for @xmath82 mev , corresponding to the result published by nplqcd @xcite , yields the dashed curve .
the solid curve corresponds to a @xmath2-dibaryon that is bound by 1.87 mev at the physical point , i.e. with the same binding wave number ( 0.23161 @xmath83 ) as the deuteron in the @xmath3 case . enlarging the pion mass to around 400 mev for the latter scenario ( i.e. to values in an order that corresponds to the nplqcd calculation @xcite ) increases the binding energy to around 8 mev and a further change of @xmath81 to 700 mev ( corresponding roughly to the hal qcd calculation @xcite ) yields then 13 mev .
note that the dependence on @xmath81 obtained agrees at least on a qualitative level with that presented in ref .
specifically , our calculation exhibits the same trend ( a decrease of the binding energy with decreasing pion mass ) and our binding energy of 9 mev at the physical pion mass is within the error bars of the results given in @xcite . on the other hand
, we clearly observe a non - linear dependence of the binding energy on the pion mass . as a consequence , scaling our results to the binding energy reported by the hal qcd collaboration @xcite ( 30 - 40 mev for @xmath84 mev ) yields binding energies of more than 20 mev at the physical point , which is certainly outside of the range suggested in ref .
however , it has to be said that for such large pion masses the lo chiral eft can not be trusted quantitatively .
now let me look at the dependence of the @xmath2 binding energy on the masses of the involved baryons .
in case of the @xmath2-dibaryon one is dealing with three coupled channels , namely @xmath74 , @xmath9 , and @xmath12 .
since we know from our experience with coupled - channel problems @xcite that coupling effects are sizeable and the actual separation of the various thresholds plays a crucial role , we expect a considerable dependence of the @xmath2 binding energy on the thresholds ( i.e. on the @xmath85 , and on the @xmath86 and @xmath87 masses ) .
corresponding results are displayed in fig .
[ fig : mpi ] in the right panel .
note that the pion mass and the masses of the other pseudo - scalar mesons are kept at their physical value while varying the @xmath88 thresholds .
for the isospin - averaged masses used in the actual calculation the thresholds are at 2231.2 , 2257.7 , and 2385.0 mev , respectively .
thus , the physical difference between the @xmath74 and @xmath9 thresholds is around 26 mev while the @xmath89 threshold is separated from the one for @xmath74 by roughly 154 mev .
first i discuss the effect of the @xmath89 channel because its threshold is quite far from the one of @xmath74 so that there is a rather drastic breaking of the su(3 ) symmetry . indeed , when the @xmath85 mass is decreased so that the nominal @xmath89 threshold ( at 2385 mev ) moves downwards and finally coincides with the one of the @xmath74 channel ( 2231.2 mev ) , a concurrent fairly drastic increase in the @xmath2 binding energy is observed , cf . the solid curve in fig . [ fig : mpi ] for results based on the interaction with a binding energy of 1.87 mev for physical masses of the mesons and baryons .
in this context i want to point out that the direct interaction in the @xmath8 channel is actually repulsive for the low - energy coefficients fixed from the @xmath16 data plus the pseudoscalar meson exchange contributions with coupling constants determined from the su(3 ) relations @xcite , and it remains repulsive even for lec values that produce a bound @xmath2-dibaryon .
but the coupling between the channels generates a sizeable effective attraction which increases when the channel thresholds come closer .
the dashed curve is a calculation with the contact term fixed to simulate the binding energy ( @xmath90 mev ) of the nplqcd collaboration at @xmath91mev .
as one can see , the dependence of the binding energy on the @xmath85 mass is rather similar .
the curve is simply shifted downwards by around 4.5 mev , i.e. by the difference in the binding energy observed already at the physical masses .
the asterisks and crosses represent results where , besides the variation of the @xmath8 threshold , the @xmath9 threshold is shifted to coincide with that of the @xmath74 channel .
this produces an additional increase of the @xmath2 binding energy by 20 mev at the physical @xmath8 threshold and by 9 mev for that case where all three @xmath88 threshold coincide .
altogether there is an increase in the binding energy of roughly 60 mev when going from the physical point to the case of baryons with identical masses , i.e. to the su(3 ) symmetric situation .
this is significantly larger than the variations due to the pion mass considered before . after these exemplary studies let me now try to connect with the published @xmath2 binding energies from the lattice qcd calculations @xcite .
the results obtained by the hal qcd collaboration are obviously for the su(3 ) symmetric case and the corresponding masses are given in table i of ref .
thus , one can take those masses and then fix the additional lec so that their @xmath2 binding energy is reproduced with those masses . to be concrete :
the masses @xmath92mev and @xmath93mev are used , and the lec is fixed so that @xmath94mev . when now the masses of the baryons and mesons are changed towards their physical values the bound state moves up to the @xmath74 threshold , crosses the threshold , crosses also
the @xmath9 threshold and then disappears .
in fact , qualitatively this outcome can be already read off from the curves in fig .
[ fig : mpi ] by combining the effects from the variations in the pion and the baryon masses .
based on those results one expects a shift of the @xmath2 binding energy in the order of 60 to 70 mev for the mass parameters of the hal qcd calculation . in case of the nplqcd calculation the values provided in ref .
@xcite are taken .
those yield then 17 mev for the @xmath9-@xmath74 threshold separation ( to be compared with the physical value of roughly 26 mev ) and 77 mev for the @xmath12-@xmath74 separation ( physical value around 154 mev ) .
we also use the meson masses of ref .
@xcite , specifically @xmath95 mev . with those baryon and meson masses
again the lec is fixed so that the @xmath2 binding energy given by the nplqcd collaboration is reproduced , namely @xmath96 mev @xcite .
again the masses of the baryons and mesons are changed so that they approach their physical values .
also here the bound state moves up to and crosses the @xmath74 threshold .
however , in the nplqcd case the state survives and remains below the @xmath9 threshold at the physical point .
specifically , an unstable bound state @xcite is observed in the @xmath9 system around 5 mev below its threshold and a corresponding resonance at a kinetic energy of 21 mev in the @xmath74 system phase shifts for the @xmath74 and @xmath9 channels are presented in fig .
[ fig : phases ] , for the relevant partial wave ( @xmath69 ) .
the solid line is the result for the @xmath88 interaction that produces a loosely bound @xmath2-dibaryon with @xmath97 mev .
the phase shift for the @xmath9 channel , fig .
[ fig : phases ] ( b ) , is rather similar to the one for the @xmath57 @xmath3 partial wave where the deuteron is found , see e.g. @xcite .
specifically , it starts at @xmath98 , decreases smoothly and eventually approaches zero ( for large energies not shown in the figure ) .
the result for @xmath74 , fig .
[ fig : phases ] ( a ) , behaves rather differently .
the pertinent phase commences at zero degrees , is first negative but becomes positive within 20 mev and finally turns to zero again for large energies .
this behaviour of the phase shifts was interpreted in @xcite as a signature for that the bound @xmath2-dibaryon is actually predominantly a ( bound ) @xmath9 state .
indeed , in that work it was argued that it follows already from the assumed ( approximate ) su(3 ) symmetry of the interaction , that any @xmath2-dibaryon is very likely a bound @xmath9 state rather than a @xmath74 state .
the dashed curve corresponds to the interaction that was fitted to the result of the hal qcd collaboration and reproduces their bound @xmath2-dibaryon with their meson and baryon masses .
the results in fig .
[ fig : phases ] are those obtained with physical masses of the mesons and baryons .
the phase shift of the @xmath9 channel shows no trace of a bound state anymore .
still the phase shift rises up to around @xmath99 near threshold , a behavior quite similar to that of the @xmath69 @xmath3 partial wave where there is a virtual state ( also called antibound state @xcite ) .
indeed , such a virtual state also seems to be present in the @xmath9 channel as a remnant of the original bound state .
the effect of this virtual state can be seen in the @xmath74 phase shift where it leads to an impressive cusp at the opening of the @xmath9 channel , cf . the dashed line in fig .
[ fig : phases ] ( a ) . the @xmath9 phase shift for the nplqcd scenario ( i.e. for the interaction that reproduces their bound @xmath2-dibaryon with their meson and baryon masses )
, see the dash - dotted curve , starts at @xmath98 , a clear indication for the presence of a bound state .
however , in this case the bound state is not located below the @xmath74 threshold but above , as already mentioned before .
consequently , the corresponding @xmath74 phase shift exhibits a resonance - like behavior at the energy where the ( now quasi ) bound @xmath2-dibaryon is located .
phase shifts for the @xmath69 @xmath12 partial wave can be found in ref .
the predictions of the three considered cases for this channel are practically the same .
chiral effective field theory , successfully applied in ref .
@xcite to the @xmath3 interaction , also works well for the baryon - baryon interactions in the strangeness @xmath0 ( @xmath100 ) and @xmath6 ( @xmath101 ) sectors .
as shown in our earlier work , already at leading order the bulk properties of the @xmath4 and @xmath5 systems can be reasonably well accounted for .
the new results for the @xmath16 interaction presented here , obtained to next - to - leading order in the weinberg counting , look very promising .
first there is a visible improvement in the quantitative reproduction of the available data on @xmath102 and @xmath103 scattering and , secondly , the dependence on the regularization scheme is strongly reduced as compared to the lo result .
indeed the description of the @xmath16 system achieved at nlo is now on the same level of quality as the one by the most advanced meson - exchange @xmath16 interactions . the recently reported evidence for the so - called @xmath2-dibaryon from lattice qcd calculations stimulated us to investigate also the quark - mass dependence of binding energies for baryon - baryon systems in the strangeness @xmath6 sector within the chiral eft framework .
here i presented results of an analysis performed at leading order in the weinberg counting .
we found rather drastic effects caused by the su(3 ) breaking related to the values of the three thresholds @xmath7 , @xmath8 and @xmath9 .
for physical values the binding energy of the @xmath2 is reduced by as much as 60 mev as compared to a calculation based on degenerate ( i.e. su(3 ) symmetric ) @xmath88 thresholds . translating this observation to the lattice qcd results reported by the hal qcd collaboration @xcite , we see that the bound state has disappeared at the physical point . for the case of the nplqcd calculation @xcite , a resonance in the @xmath7 system might survive .
i would like to thank n. kaiser , u .- g .
meiner , a. nogga , s. petschauer , and w. weise for collaborating on the topic covered by my talk .
work supported in part by dfg and nsfc ( crc 110 ) .
s. r. beane _ et al .
_ , phys .
* 106 * ( 2011 ) 162001 .
t. inoue _ et al .
_ , , phys .
lett . * 106 * ( 2011 ) 162002 .
s. r. beane _ et al .
_ , phys . rev .
d * 85 * ( 2012 ) 054511 .
t. inoue _ et al .
_ , nucl .
phys . a * 881 * ( 2012 ) 28 . | results from an ongoing study of baryon - baryon systems with strangeness @xmath0 and @xmath1 within chiral effective field theory are reported .
the investigations are based on the scheme proposed by weinberg which has been applied rather successfully to the nucleon - nucleon interaction in the past .
results for the hyperon - nucleon and hyperon - hyperon interactions obtained to leading order are reviewed .
specifically , the issue of extrapolating the binding energy of the @xmath2-dibaryon , extracted from recent lattice qcd simulations , to the physical point is addressed .
furthermore , first results for the hyperon - nucleon interaction at next - to - leading order are presented and discussed .
hyperon - nucleon interaction , hyperon - hyperon interaction , lattice qcd , effective field theory |
planetary nebulae ( pne ) are known to possess a variety of small - scale structures that are usually in a lower ionization state than the main body of the nebulae .
the morphological and kinematic properties of these low - ionization structures ( liss , gonalves , corradi & mampaso , 2001 ) vary from type to type , in the sense that liss can appear in the form of pairs of knots , filaments , jets , and isolated features moving with velocities that either do not differ substantially from that of the ambient nebula , or instead move supersonically through the environment . the total number of pne that are known to possess liss is 55 , i.e. about 10% of all the 527 galactic pne imaged in filters of high- and low - ionization emission lines ( balick 1987 ; schwarz , corradi & melnick 1992 ; manchado et al .
the different types of liss may be easily seen in corradi et al .
( 1996 ) . at present
, the origin of jets and pairs of knots in pne is not completely clear . from the theoretical point of view
, the principal physical process behind the formation of collimated liss is the interplay between the stellar agb and post - agb winds ( for single stars ) or between stellar and disk winds ( if the central star is binary ) .
according to the various studies dedicated to this issue ( e.g. garca - segura 1997 ; garca - segura & lpez 2000 ; steffen , lpez & lim 2001 ; blackman , frank & welch 2001 ) jets and knots originated by this interplay are predicted to be supersonic , highly collimated and two - sided . in the case of single stars these
are expected to be produced at the same time as the main pn shell , but in the case of a binary star origin they may be younger than the host pn . however , important properties of these liss such as density contrasts , the peculiar nitrogen abundance and main excitation mechanisms , appear hard to explain ( see dwarkadas & balick 1998 ; balick & frank 2002 ; gonalves et al .
2003 ) .
k 4 - 47 ( pn g149.0 + 04.4 ) is a compact pn that contains liss .
it is composed by a small , high ionization nebular core and a pair of low - ionization , high - velocity knots , connected to the core by a much fainter low - ionization lane ( corradi et al .
2000 ; see figure 1 ) .
gonalves et al .
( 2001 ) have proposed that the low - ionization lanes and knots of k 4 - 47 are genuine jets .
their morphological and kinematical properties are explainable if the jets and knots were formed by accretion disks , attaining velocities of several hundred kilometers per second ( the main properties of these models are also summarized in gonalves et al .
2001 and balick & frank 2002 ) .
these highly supersonic velocities imply that the resulting liss are likely to be shock - excited . k 4 - 47 is a poorly studied pn , for which statistical methods provide an unrealistic wide range of distances , for instance , 8.5 kpc ( cahn , kaler & stanghellini 1992 ) or 26 kpc ( van de steene & zijlstra 1994 ) .
corradi et al .
( 2000 ) computed a distance between 3 kpc and 7 kpc assuming that the object participates to the ordered rotation of the disk of the galaxy , but note that the relatively large height of k 4 - 47 on the galactic plane ( 0.54 kpc for a distance of 7 kpc ) adds some further uncertainty to this determination .
tajitsu & tamura ( 1998 ) estimated a distance of 5.9 kpc using the integrated iras fluxes under the ( crude ) assumption of constant dust mass for all pne .
lacking of anything better , we shall adopt in the following this distance of 5.9 kpc .
lumsden et al .
( 2001 ) mapped the h@xmath0 emission from k 4 - 47 finding that it is excited by shocks .
the object also appears in the 6 cm vla radio survey of aaquist & kwok ( 1990 ) showing a very compact radio core , with a diameter of 0.25 arcsec , and one of the largest brightness temperatures ( t@xmath1=8700 k ) found in pne .
so far , the properties ( luminosity and temperature ) of its central star as well as its nebular ( physical and chemical ) properties are not known . in this work ,
we address the debated issue of the nature and origin of high - velocity liss , through the determination of the physical parameters , excitation and chemistry of k 4 - 47 .
spectra of the core and the pair of knots are analyzed using two different models , which consider the gas to be either fully photoionized by the pn central star , or fully ionized by shocks .
we will show that k 4 - 47 is particularly interesting for this study because , in contrast to most of the liss studied up to now ( dopita 1997 ; dwarkadas & balick 1998 ; miranda et al .
2000 ; gonalves 2003 ; gonalves et al . 2003 ) , its pair of liss is mainly shock excited .
spectra of k 4 - 47 were obtained on 2001 august 28 at the 2.5 m isaac newton telescope ( int ) at the observatorio del roque de los muchachos on la palma ( spain ) , using the intermediate dispersion spectrograph ( ids ) .
the ids was used with the 235 mm camera and the r300v grating , providing a spectral coverage from 3650 to 7000 with a spectral reciprocal dispersion of 3.3 pix@xmath2 and a resolution of 6.9 .
the spatial scale of the instrument was [email protected] pix@xmath2 with the tek5 ccd .
seeing varied from [email protected] to [email protected] .
the slit width and length were 1.5 and 4 , respectively .
the slit was positioned through the centre of the nebula at p.a .
= 41@xmath4 , passing through the knots , and the exposure times were 3@xmath51800 s. bias frames , twilight and tungsten flat - field exposures , arcs and exposures of standard stars bd+332642 , cyg ob2#9 , hd19445 , and bd+254655 , were obtained .
spectra were reduced and flux calibrated using the standard iraf package for long - slit spectra .
line fluxes were measured separately for the three nebular regions indicated in figure [ image ] , namely , and .
the observed line fluxes are given in table 1 .
errors in the fluxes were calculated taking into account the errors in the measurement of the fluxes , as well as systematic errors of the flux calibrations , background determination , and sky subtraction .
the bottom lines of table 1 give the estimated accuracy of the measured fluxes for a range of line fluxes ( relative to ) in each of the regions .
absolute fluxes integrated along the slit in each region are as follows : f()@xmath6=2.41 ; f()@xmath7=2.76 ; and f()@xmath8=1.64 ( in units of @xmath9 erg @xmath10 s@xmath2 ) .
figure [ ratio ] shows the spatial profiles of low- to high - excitation oxygen emission lines ( 6300 , ( 3726+3729 ) and 5007 ) ; we clearly see that the knots have a much lower excitation than the core . while the low - ionization profiles ( 6300 and ( 3726+3729 ) ) show local maxima at the positions of the knots , the 5007 profile presents its maxima in the core and does not peak at the knots .
the average values of 6583/ at the knots are 2.0 ( ) and 2.1 ( ) , much higher than those usually found for spherical and elliptical pne ( e.g. , aller & czyzak 1983 ) , although such values are frequently found in the peimbert type i objects ( peimbert 1978 ; peimbert & torres - peimbert 1983 ) . the , on the other hand , has a less extreme 6583/ line ratio , 1.2 , even though still higher than in most pne . peculiar line ratios like those found in k 4 - 47 have been interpreted as indicative of the presence of shock - excitation or anomalous abundances ( see , for instance , miranda & solf 1992 ) .
fluxes were extinction - corrected using ( the logarithmic ratio between observed and dereddened fluxes ) , determined from the observed / and /h@xmath11 ratios , for each of the selected features of the nebula , and the reddening law of cardelli , clayton , & mathis ( 1989 ) .
the derived are : [email protected] ( ) ; [email protected] ( ) ; [email protected] ( ) . the physical properties ( for all the nebular regions ) and chemical abundances of the
were then evaluated assuming pure photoionization by the central star .
we estimate the electron densities of k 4 - 47 , which are listed in table 1 , using the 6717/6731 line ratio .
both knots are denser than the core ( 1900 @xmath13 ) by factors of 2.4 and 1.2 for and , respectively . adopting these densities ,
electron temperatures are estimated using standard line ratios : @xmath14 is obtained from the 4959/4363 line ratio , which is appropriate to zones of medium to high excitation , and @xmath14 from 6583/5755 , representative of low - excitation regions . the @xmath14 for and the @xmath14 of
the , are lower limits .
we find that , within the errors and when both determinations are available , the two temperatures are similar .
errors on @xmath14 at the position of the knots are significantly higher than those of the temperatures simply because of the low - ionization nature of these regions .
note that is remarkably higher than the typical values for pne of around @xmath15 k ( kaler 1986 ) .
figure [ shockphot ] shows several diagnostic diagrams which are commonly used to distinguish photoionized nebulae from shock - excited objects , adapted from sabbadin , minello & bianchini ( 1977 ) and phillips & cuesta ( 1999 ) . in all of them ,
the knots of k 4 - 47 are located close to the shock - excited zones while the core is displaced toward the locus where pne are usually found , i.e. photoionized objects .
this suggests that shock - excitation plays an important role in the knots , but not in the core .
the fact that in the ( 4959+5007)/ _ vs. _ ( 6716+6731)/diagram do not fall exactly within the shock regime " area is not against this interpretation , as its boundaries were originally computed for young stellar objects , using parameters ( gas abundances and shock velocities ) which can be different from those of evolved objects like k 4 - 47 .
the large expansion velocities of the knots and their broadened line profiles ( corradi et al .
2000 ) are also typical of shocked flows like h - h objects . in addition , the shock - excited nature of the knots would be consistent with their measured @xmath14 ( table 1 ) , unusually large for low - ionization microstructures in pne , and with their extremely high 6583/ line ratio . on the contrary ,
the velocities observed in the core are substantially lower than in the knots , and so is its 6583/ line ratio .
it is therefore suggested that the high velocity knots and of k 4 - 47 are mainly shock heated , while the is mainly photoionized . with the aid of existing photoionization and shock modelling
we will explore in the following these alternatives .
ionic and total abundances for the were computed using the ionization correction factors ( _ icf _ ) scheme , following kingsburgh & barlow ( 1994 ) , as described in corradi et al .
( 1997 ) and gonalves et al .
results are shown in table 2 .
note that as the @xmath14 in table 1 is only a lower limit , the abundances of the low - ionization ions are upper limits .
errors on the total abundances are obtained by propagating the errors in the ionic abundances as well as on the _ icf _ , and are indicated within brackets in the table as percentage errors ( 14% for he , for which no _ icf _ is applied , and 3050% for o and
n ) . according to its large he and n / o abundances ,
k 4 - 47 is similar to most bipolar pne ( perinotto & corradi 1998 ) , and would be classified as a type i pn ( peimbert & torres - peimber 1983 , see also kingsburgh & barlow 1994 ) .
however , the o , ne and s abundances of k 4 - 47 are among the lowest measured for galactic pne , and are more typical of some halo ( or type iv ) pne ( see e.g. howard , henry & mccartney 1997 ) .
high he / h abundances ( 0.15 @xmath16 he / h @xmath16 0.20 ) , and some ( but much less extreme than in k 4 - 47 ) oxygen deficiency ( 1.35 @xmath16 o / h ( @xmath17 ) @xmath16 2.7 ) , were found in other bipolar pne , such as ngc 6302 ( pottasch & beintema 1999 ) , ngc 6537 and he 2 - 111 ( pottasch , beintema & feibelman 2000 ) . very much like in the case of k 4 - 47 , these pne have considerably high electron temperatures ( average @xmath14 , from optical , iso and iue spectra for the three of them are 17200 k up to 18300 k ) .
therefore , we have the somewhat contradictory result that while k 4 - 47 seems to be an extreme bipolar pn in terms of its morphology and he and n abundances , its o , ne and s abundances rather resemble pne in the galactic halo , as its relatively large height on the galactic plane might also suggest ( see section 1 ) , and contrary to bipolar pne which are highly concentrated toward the galactic plane ( corradi & schwarz 1995 ) . to further investigate these issues , in the next two sections we will model the spectrum using , first , a shock excitation code , and , then , a pure photoionization model .
shock models have successfully reproduced the emission line ratios , line profiles and , recently , velocity maps in hh objects ( beck et al .
2004 ) . to check the possible influence of shock excitation on the spectrum we have explored a variety of steady plane - parallel models , for which we use the photoionization - shock code mappings ic ( dopita , binette & tuohy 1984 ; binette , dopita & tuohy 1985 ) . one important characteristic of shock models is that the predicted spectra strongly depends on the preshock ionization .
we have compiled a set of fully pre - ionized shock models ( i.e. the incident gas has been ionized ) and a set of local equlibrium preionization for the gas entering the shock ( as described by shull & mckee 1979 , hartigan et al .
relative abundances typical of type i pne were assumed , and two values for the pre - shock density were considered ( 100 , 1000 @xmath13 ) . the input shock velocities were varied from 95 to 140 km s@xmath2 , i.e. the range producing large ( 4959,5007)/ratios ( hartigan et al .
1987 ) , as observed .
note , however , that the measured expansion velocity of the is smaller than 50(corradi et al .
2000 ) , pointing to lower shock velocities .
none of the predicted spectra could reproduce the large observed ratios of intermediate to high excitation emission lines with respect a h i balmer line ( i.e. 3727/ , 5007/ , [ ne iii ] 3868/ and he ii 4686/ ) . moreover ,
the shocked spectra predicted by our models imply 6300/ and ( 6716+6731)/ larger than the observed values .
a large discrepancy was also found with the observed 4959/4363emission line ratio , which is more than 2 times larger than the model values . and
, finally , 6583/5755 is very high , @xmath18 300 , because the intensity of 5755 is largely understimated by these models .
we conclude that the plane - parallel shock models are not able to reproduce the spectrum of the core of k 4 - 47 .
we have used the photoionization code cloudy 95.06 ( ferland et al . 1998 ) .
cloudy needs as input the information on the shape and intensity of the radiation from the ionizing source , the chemical composition and geometry of the nebula , as well as its density and size . as mentioned in section 1 , the distance of k 4 - 47 , and thus its size and the luminosity of the central star , are poorly known .
we adopt the distance of 5.9 kpc computed by tajitsu & tamura ( 1998 ) , although extending the calculations to the full range of distances from 3 kpc to 7 kpc proposed by corradi et al .
( 2000 ) . a spherical geometry , and a filling factor of 1.0 have been assumed . using the h@xmath19 and images from corradi et al .
( 2000 ) a core diameter of 1.9 arcsec was estimated ( 10% contour extension corrected for the finite resolution ; tylenda et al .
the density was initially kept to constant and equal to its empirical value ( 1900 @xmath13 , see table 1 ) .
however , as noted in the introduction , aaquist & kwok ( 1990 ) discovered a bright and very compact radio core at the centre of k 4 - 47 , implying a high density of 72000 @xmath13 ( assuming an optically thin nebula at 5 ghz , the distance of 5.9 kpc and following goudis 1982 ) .
for this reason , models with much higher density values were also explored .
dust grains have been included , since they have an important effect , particularly on the temperature structure of pne .
this effect depends on the type of grain ( graphite and/or silicate ) , the grain abundances , and the grain - size distribution , being more relevant in the inner regions of the nebula ( see dopita & sutherland 2000 ) .
both graphite and silicate grains were considered . as we do not know the grain size distribution and grain abundances for k 4 - 47 and more in general for pne , we adotped ism size distribution , with ism gas - phase depletions due to grains , following van hoof et al .
( 2004 ) .
we adopted a blackbody spectrum for the central star , with derived from the , and emission line ratios , using a modified zanstra method as done for instance by mikoajewska et al .
this gives = 130000 k ( from 4686/ ) and = 115000 k ( from 4686/5876 ) .
a lower limit to the central star luminosity comes from the iras spectral energy distribution ( tajitsu & tamura 1998 ) , which gives l@xmath20 16 d@xmath21 @xmath22 ( where d is the distance in kpc ) yielding 550 @xmath22 at 5.9 kpc . as with the elemental abundances , we first run models with the empirical abundances given in table 2 , and subsequently with `` average '' values for either normal ( type - ii ) , or type - i pne , following kingsburgh & barlow ( 1994 ) .
cloudy models with the empirical abundances do not match the observed line fluxes .
both helium and ( 6548,6583 ) emission lines are largely overestimated , while oxygen , sulphur and neon lines are underestimated , mainly because the very low o / h , ne / h and s / h input abundances are not compensated by a high electron temperature in the model as empirically determined .
tests with type - ii pne abundances were also not successful because the predicted 6548,6583intensities are very low , which is expected as the main difference between type - i and type - ii abundances is that the latter are depleted in nitrogen by a factor of @xmath234 ( kingsburgh & barlow 1994 ) .
assuming type - i abundances , we found a partial agreement with the observed spectrum for = 120000 k , l=550 @xmath22 and n@xmath24= 1900 @xmath13 .
important optical lines like heii 4686 , ( 4959,5007 ) , ( 6548,6584 ) and ( 6716,6731 ) are well reproduced by this model , with discrepancies of 10% or less .
hei 4471 , 6678 , heii 5412 , and [ s iii ] 6312 lines agree with the observed spectrum within 12% to 20% .
other relatively intense lines such as ( 3726+3729 ) , [ neiii ] 3868 , and hei 5876show discrepancies of 35% to 50% with respect to the model .
furthermore , the model underestimates the intensities of the 4363 and the 5755 lines by a factor of about 3 .
these lines are the keys for the determination of the electron temperature .
finally , the model underestimates [ ni ] 5200 and [ oi ] 6300 by factors
larger than 30 .
let us note , in addition , that this model requires a size for the core of 4.0 arcsec , twice the optical size observed .
it is then instructive to explore alternative scenarios .
let us first note that in some pne and related objects ( see e.g. corradi 1995 ) , high 4363 and 5755 line intensities have been interpreted as the signature of very high core densities , as at @xmath25 larger than about 10@xmath26 @xmath13 the auroral ( 4363 and 5755 ) to nebular ( 4959 and 6583 ) line ratios are indicators of density rather than of temperature ( gurzadyan 1970 ) .
for this reason , we have calculated other models assuming much higher ( albeit still constant ) densities .
as noted above , high core densities are implied by the radio size and flux ( aaquist & kwok 1990 ) .
cloudy models for the observed radio core size , 0.25 arcsec , and such high densities ( from 72000 k up to @xmath27@xmath13 ) show that both 4363 and 5755 intensities can be now reproduced , but other nebular lines , in particular ( 3726+3729 ) , ( 6548,6583 ) , and ( 6717,6731 ) , become now largely underestimated because of collisional quenching . a natural way to solve the problem might be to assume a strong density stratification in the core of k 4 - 47 , with a very dense inner zone where 4363 and 5755 are mostly formed , and a lower density outer region where other important nebular lines are produced .
this idea should be tested by means of an appropriate photoionization model ; 3-d codes ( like mocassin ; ercolano et al .
2003 ) are much better suited than cloudy to deal with such extreme density variations . in summary
, we find that none of the constant density models is able to account , simultaneouly , for all optical emission lines in the .
in particular , the 4363 and 5755 intensities are strongly understimated if the nebular density is the one derived empirically from the lines . a model with a strong density stratification could possibly offer a solution to the problem .
the fact that we are not able to provide an accurate representation of the spectrum , also prevents us to determine the radiation escaping from the core and reaching the knots , and thus to attempt a reliable photoionization modelling of the latter .
a consistent photoionization modelling of the whole nebula ( core+knots ) would also require a fully 3-d modelling , that goes beyond the scopes of the present work .
given the evidence that the knots might be excited by shocks ( section .
3.2 ) , we instead attempt to describe their spectrum using existing shock models . we have used the bow - shock models described in raga & bohm ( 1986 ) and hartigan et al . (
we consider a bow shock with a functional form , @xmath28 , where @xmath29 is measured along the symmetry axis , @xmath30 is the cylindrical radius , the parameter @xmath31 determines the shape of the bow shock and the constant @xmath32 determines its size ( beck et al .
the emission from the bow - shock is modeled as described by hartigan et al .
( 1987 ) , being the geometry of the bow shock the parameter that determines the shock velocity .
the plane - paralel shock models required to predict the bow - shock emission line ratios are obtained with the photoionization - shock code mappings ic .
we have adopted a pre - shock density of 400 @xmath13 , that is within the range generally assumed for stellar jets .
the pre - shock magnetic field was taken to be negligible .
we assumed elemental abundances for he , c , n , o , ne , s , and ar to be within the range of type - i pn values . as for the preionization of the gas , local equilibrium is assumed .
we consider the bow - shock velocity as a free parameter , adopting a set of values from 50 to 300 km s@xmath2 .
we have obtained the best fit to the observed spectra for @xmath31=3 .
a summary of the results is presented in figure [ bshock ] for a number of emission line ratios plotted as a function of 6583/. note that the later line ratio increases in strength as the shock velocity increases , and in the case of k 4 - 47 is matched by only the largest shock velocities considered , somewhat larger than the estimate of 150 by corradi et al .
( 2000 ) .
for these velocities , the bow - shock models fit pretty well the line ratios ( 4069+4076)/(6717+6731)@xmath33 4072/6725 , and 6583/5755 , sensitive to the temperature in the recombination zone where and are formed , but not equally well the 4959/4363 ratio , indicating that the predicted electron temperature is too large . the and electron temperatures predicted by these models , for both knots , are 17800 k and 27600 k , respectively . the ratios 6300/ and ( 3726+3729)/ are also well reproduced .
larger discrepancies appear for the helium lines hei 5876 and heii 4686 , and specially for [ ne iii ] 3868 and 5007 in , and ( 6716+6731 ) and [ n i ] 5200 in both knots .
note that a similar problem for the ( 6717+6731)/ ratio is also found in hh objects ( raga et al .
the large discrepancy for the [ n i ] 5200/ emission line ratio is intriguing , as strong line intensities for this ion can only be produced with low shock velocities , which could however not account for the observed flux of higher ionization stages like 6583 .
different shock geometries , producing a different distribution of shock velocities , might help to solve this problem .
also note that the bow - shock models are based on appraisal of what would be reasonable parameters for the knots of k 4 - 47 , and we did not intend to reproduce the line intensities separately for each of the knots .
but the knots are not identical : in particular , high excitation lines like heii 4686 , [ ne iii ] 3868 and 5007 are much brighter in than in , which indicates that the radiation field of the central star would affect the emission line spectrum much more than that of .
the observed spectra of would also fit a model with a somewhat larger n / h abundance and a lower bow shock velocity .
summarising , the analysis of the shock - excited knots suggests that they are high - velocity ( 250 to 300 ) condensations with high electron temperatures ( = 17800 k ) and chemical abundances similar to type - i pne .
the analysis presented in this paper is the first attempt to determine the physical parameters , excitation mechanism and chemical abundances of k 4 - 47 . among other
pne with liss , m 2 - 48 ( lpez - martn et al .
2002 ) and kj pn 8 ( lpez , vzquez & rodrguez 1995 ) have , similarly to k 4 - 47 , a high electron temperature and the evidence for shock - excited liss . but note that in these pne , except for one of the m 2 - 48 s knots ( for which is 20100 k , twice the core temperature ) , and are not estimated in a detailed zone - by - zone basis , because 4363 and 5755 were not measured .
the two distinct models that we used for the analysis of k 4 - 47 ( pure photoionization for the core and pure shock - excitation for the pair of knots ) , provide some clues for the interpretation of this nebula .
first , the analysis of the core would suggests he and n overabundances typical of type - i ( galactic disk ) pne , but also an extreme oxygen deficiency which is instead more typical of pne in the galactic halo .
these abundances calculation would however be invalidated if more detailed modelling proves the existence of a strong density stratification within the core , with an inner region with extremely high density ( @xmath34 @xmath13 ) , and an outer zone with a much lower density matching the empirical determination from the lines .
in fact , in this case the determined empirically using the 4959/4363 and 6583/5755 line ratios would be wrong , as the auroral and nebular lines involved would be formed in different regions .
note that , albeit unusual , such high density cores were actually found in several other pne ( corradi 1995 ) , which like k 4 - 47 show an extreme degree of collimation .
second , the knots seem to be mainly shock - excited
. this conclusion is strenghtened by diagnostic line ratio diagrams , and , interestingly , by the fact that the h@xmath0 emission of k 4 - 47 is also shock - excited , contrary to most pne where instead it is excited by fluorescence ( lumsden et al .
our shock modelling also indicates that knots have type - i abundances .
according to these models , knots would move with space velocities of 250 - 300 , about twice the value estimated by corradi et al .
( 2000 ) from the analysis of the lines profiles .
moreover , while the shock model is successful in matching the ( 6583/5755 ) line ratio at the knots it underestimate their ( 4959/4363 ) ratio by some 30% and 70% for and , respectively . as a final remark ,
the two physical processes explored separately in this paper , i.e. photoionization and shock heating , are likely to be simultaneously present in a real nebula like k 4 - 47 , as gas can be ionized by both energetic photons from the central star and from the shocks associated with the observed supersonic outflows .
taking this into account would require a sophisticated modelling which is not presently available , and which would also need to be supported by a better knowledge of the nebular and stellar parameters .
in particular , very little information is presently known about the distance and the properties of the central star of k 4 - 47 , a basic data that could be addressed by future observations .
we thank the anonymous referee for his / her comments , which helped us to improve the paper significantly .
we also thank joanna mikoajewska for fruitful discussions and corrado giammanco for helping with cloudy .
the work of drg , am , rlmc and llm is partially supported by a grant from the spanish ministry of science and technology ( aya 2001 - 1646 ) . the work of ar is supported by the grant aya 2002 - 00205 . | k 4 - 47 is an unusual planetary nebula composed of a compact high - ionization core and a pair of low - ionization knots .
long - slit medium - resolution spectra of the knots and core are analyzed in this paper .
assuming photoionization from the central star , we have derived physical parameters for all the nebular components , and the ( _ icf _ ) chemical abundances of the core , which appear similar to type - i pne for he and n / o but significantly deficient in oxygen .
the nebula has been further modelled using both photoionization ( cloudy ) and shock ( mappings ) codes . from the photoionization modelling of the core , we find that both the strong auroral 4363 and 5755 emission lines observed and the optical size of the core can not be accounted for if a homogeneus density is adopted .
we suggest that a strong density stratification , matching the high - density core detected at radio wavelengths and the much lower density of the optical core , might solve the problem . from the bow - shock modelling of the knots , on the other hand , we find that knots chemistry is also represented by type - i pn abundances , and that they would move with velocities of 250 - 300 .
[ firstpage ] planetary nebulae : individual ( k 4 - 47 ) - ism : kinematics and dynamics - ism : jets and outflows |
in a recent paper @xcite , spectral problem for the one - dimensional cauchy process ( that is , the symmetric @xmath3-stable process ) killed upon exiting a half - line or an interval was studied . for the half - line , an explicit formula for generalized eigenfunctions of transition operators
was obtained using methods developed in the theory of linear water waves .
the argument of @xcite relies on some properties specific to the cauchy process , and it does not easily generalize to other lvy processes .
the purpose of this article is to derive a similar formula for generalized eigenfunctions in a more general setting , using a modified method .
the class of processes considered here consists of symmetric ( one - dimensional ) lvy processes @xmath4 with lvy measure having completely monotone density function on @xmath2 .
alternatively , this class can be described as ( one - dimensional ) subordinate brownian motions , corresponding to subordinators with lvy measure having completely monotone density function .
yet another characterization is given by the condition that the lvy - khintchine exponent of @xmath4 has the form @xmath0 for a _ complete bernstein function _ @xmath5 .
the equivalence of the above is given in proposition [ prop : cbf ] .
this class of lvy processes have attracted much attention in the last decade ; see , for example , @xcite for some recent developments .
there is also an extensive literature focused specifically on symmetric @xmath6-stable processes ( @xcite ) and relativistic @xmath6-stable processes ( @xcite ) , which are included here as examples .
we consider the process @xmath4 killed upon leaving the half - line @xmath2 .
let @xmath7 , @xmath8 and @xmath9 denote the corresponding transition operators , the @xmath10 generator , and the domain of @xmath8 , respectively ( formal definitions are given in the preliminaries ) .
the following generalization of theorem 2 from @xcite is the main result of the article .
its main advantage is the explicit description of the eigenfunctions .
[ th : eigenfunctions ] suppose that the lvy - khintchine exponent of @xmath4 has the form @xmath0 for a complete bernstein function @xmath1 . for all @xmath11 ,
there is a bounded function @xmath12 on @xmath13 which is the eigenfunction of @xmath7 and @xmath8 : for all @xmath14 .
the function @xmath12 is characterized by its laplace transform : for @xmath15 such that @xmath16 .
furthermore , for @xmath14 we have where @xmath17 , and @xmath18 is a bounded , completely monotone function on @xmath2 .
more precisely , we have and @xmath19 is the laplace transform of a finite measure @xmath20 on @xmath2 .
when @xmath5 extends to a function @xmath21 holomorphic in the upper complex half - plane @xmath22 and continuous in @xmath23 , and furthermore @xmath24 for all @xmath25 , then the measure @xmath20 is absolutely continuous , and for @xmath25 . in some cases , the above formulae can be substantially simplified .
for example , when @xmath4 is the symmetric @xmath6-stable process , then @xmath26 , and @xmath27 ; see section [ sec : examples ] . any complete bernstein function @xmath5
extends holomorphically to @xmath28 $ ] .
hence , the assumption in the last part of theorem [ th : eigenfunctions ] concerns the existence of boundary values of @xmath1 ( approached from the upper complex half - plane ) along the branch cut @xmath29 $ ] . a formula similar to
can be given in the general case , as discussed in remark [ rem : gamma ] .
it should be emphasized that although theorem [ th : eigenfunctions ] extends the result of @xcite , its proof is essentially different .
[ rem : anal ] entirely analytical formulation of theorem [ th : eigenfunctions ] is available ; see @xcite for the analytical definition of @xmath8 .
also , our arguments are purely analytical , with the exception of probabilistic proofs of lemmas [ lem : d2 ] and [ lem : smooth : domain ] .
the eigenfunctions @xmath12 _ never _ belong to @xmath30 : this reflects the fact that the spectrum of transition operators @xmath7 , considered as operators on @xmath30 , is purely continuous .
nevertheless , the functions @xmath12 yield a _ generalized eigenfunction expansion _ of @xmath7 . in classical eigenfunction expansion ,
a function is decomposed with respect to a complete orthonormal set of eigenfunctions , and the operator acts at each component independently .
informally , in _ generalized _ eigenfunction expansions , the principle is the same , but eigenfunctions are no longer square - integrable and there are uncountably many of them . a formal statement is given in the following result , which generalizes theorem 3 in @xcite , and explicitly provides functional calculus for @xmath8 . here
@xmath8 and @xmath31 denote the @xmath30 generator of the semigroup @xmath7 , and the domain of @xmath8 , respectively ( again , see preliminaries for formal definitions ) .
[ th : spectral ] with the notation of theorem [ th : eigenfunctions ] , let then @xmath32 extends to a unitary operator on @xmath30 , and furthermore , @xmath33 if and only if @xmath34 is in @xmath30 , and below we provide a complete proof of theorem [ th : spectral ] under an additional condition that @xmath35 is injective , and we verify that @xmath35 is injective when @xmath4 is the symmetric @xmath6-stable process , @xmath36 $ ] ( that is , @xmath37 ) . in the preliminary version of this article , a relatively easy to check , but rather restrictive sufficient condition for injectivity of @xmath35 was given ( lemma 19 in @xcite ) , and it was conjectured that in fact @xmath35 is always injective .
after a few months , this conjecture was solved in affirmative by jacek maecki , and the proof of injectivity of @xmath35 is given in the recent preprint @xcite .
we announce the result of @xcite and state theorem [ th : spectral ] in full generality .
there is a striking similarity between formulae and on one hand , and some formulae in fluctuation theory of lvy processes on the other .
more precisely , the exponent in and resembles known formulae for the laplace exponent of ladder processes ( see corollary 9.7 in @xcite ) and the laplace transform of the supremum functional ( theorem 1 in @xcite ) .
this phenomenon can be easily explained at the analytical level : in both cases explicit formulae are obtained using the _ wiener - hopf method _
, a technique for solving some integral equations in the half - line via fourier transform . fluctuation theory of lvy processes relies upon the wiener - hopf factorization of the @xmath38-potential operator ( that is , the resolvent ) of the transition semigroup of a lvy process .
the fourier symbol of this operator is @xmath39 . in the proof of theorem
[ th : eigenfunctions ] , the operator with fourier symbol @xmath40 is factorized .
a probabilistic formulation of the fluctuation theory is available through local times , excursion theory and ladder processes ; see @xcite .
it is an interesting open problem whether there is a similar _ probabilistic _ derivation of the formula for @xmath12 .
the proof of theorem [ th : spectral ] ( under the assumption that @xmath35 is injective ) follows the line of @xcite .
proofs of theorems [ th : eigenfunctions ] and [ th : spectral ] are rather technical .
therefore , it may be helpful to keep in mind the following well - recognized example .
suppose that @xmath4 is the brownian motion , with variance @xmath41 ( so that the generator of @xmath4 is the one - dimensional laplace operator @xmath42 , and @xmath43 ) . in this case
@xmath8 is the laplace operator in @xmath2 with dirichlet boundary condition at @xmath44 , and @xmath7 is the classical heat semigroup on @xmath13 with dirichlet boundary condition at @xmath44 .
the eigenfunctions of @xmath8 and @xmath7 are simply @xmath45 , with corresponding eigenvalues @xmath46 and @xmath47 , respectively . the integral transform
@xmath35 is the fourier sine transform on @xmath2 , and theorem [ th : spectral ] states that the fourier sine transform is the unitary mapping ( up to a constant factor @xmath48 ) , which diagonalizes the action of the heat semigroup . to the author s knowledge ,
no results similar to theorems [ th : eigenfunctions ] and [ th : spectral ] were available for lvy processes other than the brownian motion ( possibly with drift ; this requires a minor modification to the above example ) and the cauchy process ( studied in @xcite ) .
although generalized eigenfunction expansions similar to theorem [ th : spectral ] have been studied for various classes of operators ( see @xcite for the case of general markov processes , and @xcite for diffusion processes with feynman - kac potential ) , the explicit description of eigenfunctions was not available .
consequently , their applications were limited .
the novelty of theorem [ th : spectral ] is in that the generalized eigenfunctions @xmath49 are given in a fairly explicit form , allowing various estimates , asymptotic analysis and numerical approximation .
for this reason , one can expect that theorems [ th : eigenfunctions ] and [ th : spectral ] will find a variety of applications ; some are already discussed below .
theorems [ th : eigenfunctions ] and [ th : spectral ] yield formulae for the transition density @xmath50 of @xmath4 killed upon leaving the half - line @xmath2 , and for the distribution of the first exit time @xmath51 from the positive half - line .
these results are contained in the following two theorems . by @xmath52
we denote the probability for the process @xmath4 starting at a fixed point @xmath14 .
[ th : pdt ] let @xmath53 .
if @xmath54 is integrable in @xmath25 , then [ th : fpt ] let @xmath53 . if and then if @xmath55 is integrable in @xmath56 for all @xmath53 , then furthermore for @xmath57 .
theorem [ th : fpt ] in full generality is proved in the recent preprint @xcite . in this article
we provide a much simpler proof , under a more restrictive condition this condition implies both and , as well as integrability of @xmath55 ( see the proof of the theorem ) . the proof given below relies on theorem [ th : spectral ]
noteworthy , many interesting examples , including symmetric @xmath6-stable processes and relativistic @xmath6-stable processes , satisfy . by corollary [ cor : power2 ] ,
condition is satisfied when @xmath1 is regularly varying of positive order at @xmath44 and at @xmath58 .
it should be emphasized that the integrand in has two oscillatory factors , and cancellations in the integral are essential .
hence , formula is problematic for numerical computations . for the cauchy process
, a significant simplification of is possible ( see theorem 4 in @xcite ) .
it is an open problem whether formula can be simplified in the more general case , even for symmetric @xmath6-stable processes for general @xmath36 $ ] .
cancellation is of less importance in theorem [ th : fpt ] ; see @xcite , where is used to obtain estimates for the density function of the distribution of @xmath51 .
since @xmath59 for all @xmath25 , the supremum in is always not greater than @xmath60 ( see preliminaries ) .
the assumptions and are rather mild regularity and growth conditions for @xmath5 ; examples are given in section [ sec : examples ] . by symmetry ,
the first exit time @xmath51 for @xmath4 starting at @xmath14 has the same distribution as the first passage time @xmath61 through a barrier at @xmath62 for @xmath4 starting at @xmath44 . on the other hand ,
first passage times are related to the supremum functional : hence , theorem [ th : fpt ] gives , in a rather general setting , an explicit expression for the distribution of first passage times and the supremum functional , which are fundamental objects in fluctuation theory of lvy processes .
noteworthy , the double laplace transform ( in @xmath63 and @xmath62 ) of the distribution of @xmath61 is known for general lvy processes since 1957 ( theorem 1 in @xcite ) .
nevertheless , explicit formulae for @xmath64 are known only in some special cases ; see @xcite for some recent developments in this area .
also , a formula for the single laplace transform ( in @xmath63 ) of @xmath65 for a large class of symmetric lvy processes was obtained only very recently in @xcite .
theorems [ th : eigenfunctions ] and [ th : spectral ] can be applied to certain systems of pdes , related to traces of two - dimensional diffusions , as discussed in section [ sec : applications ] . furthermore , there are at least four recent preprints exploiting the results of the present article .
an application of theorem [ th : eigenfunctions ] to the spectral theory of symmetric @xmath6-stable process in the interval can be found in @xcite .
the case of the relativistic @xmath3-stable process in half - line and interval is studied in detail in @xcite .
this application may be of particular interest in quantum physics ; see @xcite for related research . in @xcite ,
theorems [ th : eigenfunctions ] and [ th : spectral ] are used to obtain refined semi - classical asymptotics for eigenvalues of higher - dimensional isotropic @xmath6-stable processes in domains .
finally , the proof of theorems [ th : spectral ] and [ th : fpt ] in full generality , as well as their application to a detailed analysis of first passage times , can be found in @xcite .
we conclude the introduction with a brief description of the structure of the article .
the preliminaries section contains definitions and standard properties used in the article , and auxiliary lemmas .
it includes some background on distribution theory , lvy processes and their generators , subordinate brownian motions and complete bernstein functions . in section [ sec : cbf ] a certain transformation related to the wiener - hopf factorization is studied .
next , in section [ sec : eigenfunctions ] , we derive the formula for @xmath12 and prove theorem [ th : eigenfunctions ] .
theorems [ th : spectral ] , [ th : pdt ] and [ th : fpt ] are proved in section [ sec : spectral ] .
examples , including symmetric @xmath6-stable processes and relativistic @xmath6-stable processes , are studied in detail in section [ sec : examples ] . in section [ sec : applications ] we give an application of our results to systems of pdes ( theorem [ th : pde ] ) .
in the theory of partial differential equations , it is a common case that one first finds a weak solution to a problem , and then , by showing that the solution is sufficiently regular , one argues that it is in fact a strong solution .
this approach will be used to prove theorem [ th : eigenfunctions ] : first we find a distributional eigenfunction of @xmath8 , and then we show that in fact it belongs to the domain of @xmath8 .
our argument relies heavily on fourier methods , and the operators involved are typically non - local , so the required background on distribution theory is slightly different from the one used in partial differential equations .
it should be emphasized that the use of distributions could be avoided , at least partially , at the price of less clear exposition .
however , the language of distribution theory seems to be well - suited for problems involving generators of killed lvy processes .
let @xmath66 denote the class of schwartz functions in @xmath67 ( @xmath68 ) , and let @xmath69 be the space od tempered distributions in @xmath67 . if @xmath70 and @xmath71 , we write @xmath72 for the value of @xmath73 at @xmath74 .
below we recall some well - known properties of tempered distributions ; for a detailed exposition of the theory , see e.g. @xcite . if @xmath70 , the fourier transform of @xmath74 is @xmath75 . for @xmath71
, @xmath76 is the tempered distribution satisfying @xmath77 .
the convolution of @xmath78 is defined in the usual way , @xmath79 .
when @xmath70 and @xmath71 , then @xmath80 is an infinitely smooth function , defined by the convolution of two distributions is not well - defined in general .
suppose that @xmath81 .
we say that @xmath82 and @xmath83 are _
@xmath69-convolvable _ if for all @xmath78 , the functions @xmath84 and @xmath85 are convolvable in the usual sense , i.e. the integral @xmath86 exists for all @xmath62 . when this is the case , the _ @xmath69-convolution _
@xmath87 is the unique distribution @xmath73 satisfying @xmath88 for @xmath78 . note that there are other non - equivalent definitions of the convolution of distributions , and @xmath69-convolution is often denoted as @xmath89 ; for the discussion of various notions of convolvability , the reader is referred to @xcite .
recall that the _ support _ of a distribution @xmath73 is the smallest closed set @xmath90 with the property that @xmath91 for all @xmath70 such that @xmath92 for @xmath93 . if any of the tempered distributions @xmath82 , @xmath83 has compact support , or ( in the one - dimensional case ) if both @xmath82 and @xmath83 are supported in @xmath94 , then @xmath82 and @xmath83 are automatically @xmath69-convolvable .
it is well known that the distributions @xmath95 and @xmath96 need not be equal ; however , if the pairs @xmath97 and @xmath98 are @xmath69-convolvable , and furthermore the functions @xmath84 , @xmath85 , and @xmath99 are convolvable ( that is , @xmath100 is integrable in @xmath101 , @xmath102 for all @xmath103 ) for any @xmath104 , then the @xmath69-convolution of @xmath82 , @xmath83 and @xmath105 is associative ; see @xcite , section 4.2.8 .
any @xmath69-convolvable distributions @xmath82 , @xmath83 satisfy the _ exchange formula _
@xmath106 , where the multiplication of distributions @xmath107 and @xmath108 extends standard multiplication of functions in an appropriate manner , see @xcite .
since we only use the exchange formula when @xmath107 and @xmath108 are genuine functions , or when @xmath107 is a measure and @xmath108 is a function , we do not discuss the notion of multiplication for general distributions and refer the interested reader to @xcite . in the one - dimensional case @xmath109 , the laplace transform can be defined for tempered distributions . if @xmath70 is supported in @xmath94 , then the laplace transform of @xmath74 is denoted by @xmath110 ( @xmath111 ) .
this is a holomorphic function of @xmath112 in the right complex half - plane @xmath16 , continuous at the boundary .
clearly , @xmath113 .
if @xmath71 is supported in @xmath94 , the laplace transform of @xmath73 is defined for @xmath15 with @xmath16 by where @xmath114 and @xmath115 is any infinitely smooth function such that @xmath116 for @xmath117 and @xmath118 for @xmath119 .
this definition does not depend on the choice of @xmath115 ( see formula ( 9.1.4 ) in @xcite ) .
the above definition of @xmath120 extends the usual definition of the laplace transform of signed measures on @xmath94 .
note that while the fourier transform of a distribution is again a distribution , the laplace transform is always a ( holomorphic ) function . for a fixed @xmath53 ,
the function @xmath121 is the fourier transform of the distribution @xmath122 ( where again @xmath123 ) , and as @xmath124 , the distributions @xmath121 converge to @xmath76 in @xmath69 .
the exchange formula holds also for the laplace transform : if both @xmath81 are supported in @xmath94 , then they are @xmath69-convolvable and @xmath125 for all @xmath112 with @xmath16 .
we say that two distributions @xmath82 , @xmath83 restricted to an open set @xmath126 are equal if @xmath127 for any @xmath70 vanishing in @xmath128 .
equivalently : @xmath129 is supported in @xmath128 .
below we recall and extend some standard definitions and properties , which can be found , for example , in @xcite . in this article
we are only concerned with one - dimensional subordinate brownian motions . however , lemma [ lem : d2 ] below might be of interest for general lvy processes .
for this reason , in this and the next subsection , we work with the general case .
let @xmath4 be a lvy process in @xmath67 .
we write @xmath52 and @xmath130 for the probability and expectation for the process @xmath4 which starts at @xmath103 .
the process @xmath4 is completely determined by its lvy - khintchine exponent @xmath131 : for @xmath132 we have where here @xmath133 is a nonnegative definite @xmath134 matrix ( the _ diffusion coefficient _ ) , @xmath135 is a vector in @xmath67 ( the _ drift _ ) , and @xmath136 is the _ lvy measure _ of @xmath4 : a radon measure on @xmath137 such that @xmath138 . if @xmath136 is absolutely continuous , we denote its density function with the same symbol @xmath139 .
the _ transition operators _ of @xmath4 are defined by whenever the integral is absolutely convergent .
each @xmath140 is a convolution operator , with convolution kernel given by the distribution of @xmath141 under @xmath142 . by the lvy - khintchine formula , furthermore , for @xmath70 , see theorem 3.3.3 in @xcite ( note that in @xcite , fourier transform is defined with @xmath143 instead of @xmath144 ) .
the operators @xmath140 form a contraction semigroup on each of the lebesgue spaces @xmath145 ( @xmath146 $ ] ) , on the space @xmath147 of bounded continuous functions , and on the space @xmath148 of continuous functions vanishing at infinity ( all equipped with usual norms ) .
on @xmath145 ( @xmath149 ) and on @xmath148 , this semigroup is strongly continuous ( see section 3.4 in @xcite ) .
the generator of each of these semigroups is denoted by the same symbol @xmath150 , and we write @xmath151 for the corresponding domain , where @xmath152 is either @xmath145 ( @xmath146 $ ] ) , @xmath147 or @xmath148 . more precisely , @xmath153 if @xmath154 and the limit exists in the topology of @xmath152 .
note that the limit , if it exists , does not depend on the choice of @xmath152 ( apart from the fact that in @xmath148 and @xmath147 it is defined pointwise , while in @xmath145 only up to a set of zero measure ) .
therefore , using a single symbol @xmath150 for operators acting on different domains @xmath151 causes no confusion .
the schwartz class @xmath66 is a core of @xmath150 on each of the spaces @xmath145 ( @xmath149 ) and @xmath148 , and by , for @xmath70 , we abbreviate , for example , @xmath155 to @xmath156 .
[ rem : spectral : free ] the spectral theory of @xmath140 and @xmath150 is very simple , thanks to the lvy khintchine formula .
indeed , the function @xmath144 is the eigenfunction of @xmath140 , with eigenvalue @xmath157 .
hence , @xmath144 belongs to @xmath158 , and it is an eigenfunction of @xmath150 with eigenvalue @xmath159 . the generalized eigenfunction expansion of @xmath140 and @xmath150 is given by and ; the fourier transform plays the same role as the integral transform @xmath35 in theorem [ th : spectral ] . if @xmath4 is symmetric , then @xmath160 is real .
since @xmath161 is a linear combination of @xmath144 and @xmath162 , it is also the eigenfunction of @xmath140 and @xmath150 .
note that the eigenfunctions @xmath12 in theorem [ th : eigenfunctions ] behave asymptotically as @xmath163 as @xmath164 . when @xmath70 , then the generator of @xmath4 can be written in the form where @xmath133 , @xmath135 and @xmath136 are as in ( theorem 31.5 in @xcite ) .
furthermore , where @xmath165 depends only on the process @xmath4 , and @xmath166 is the maximum of @xmath167 norms of @xmath74 and its first and second partial derivatives ( see the proof of theorem 31.5 in @xcite ) .
we need the following extension of . for @xmath148 instead of @xmath147 ,
this is well - known , see theorem 31.5 in @xcite .
the statement given below is , however , difficult to find in the literature , so we provide a proof .
let @xmath168 be the class of functions @xmath73 such that @xmath73 and all partial derivatives of @xmath73 ( of all orders ) belong to @xmath147 .
[ lem : gen : cb ] if @xmath169 , then @xmath170 and [ rem : domain : counterexample ] note that in contrast to the @xmath148 case , @xmath73 may fail to belong to @xmath171 when @xmath73 and its first and second partial derivatives are in @xmath147 .
a simple counterexample can be easily constructed for one - dimensional brownian motion : if @xmath172 , but @xmath173 is not _
uniformly _ continuous , then the convergence of @xmath174 may fail to be uniform .
we omit the details .
let @xmath175 denote the maximum of @xmath167 norms of @xmath176 and its partial derivatives of order not greater than @xmath177 , and denote @xmath178 .
there is a sequence @xmath179 such that @xmath180 and its first and second partial derivatives converge locally uniformly to @xmath73 and the corresponding first and second partial derivatives of @xmath73 , and furthermore @xmath181 ( with @xmath165 depending only on the dimension @xmath182 ) . for @xmath183 and @xmath103 , we have ( see @xcite ) in particular , @xmath184 .
since @xmath185 , by we have by , since @xmath186 and @xmath150 commute with partial differential operators on @xmath66 , by , finally , by , as @xmath187 , we have @xmath188 and @xmath189 ( by dominated convergence ) . by ,
taylor s theorem and dominated convergence , @xmath190 converges to the right - hand side of , which we denote by @xmath191 , and the convergence is locally uniform in @xmath103 .
it follows that since @xmath103 was arbitrary , we have @xmath170 and @xmath192 , as desired . informally , the generator @xmath150 is also a convolution operator , but the convolution kernel is a tempered distribution .
we give this a precise meaning ; see also @xcite , and @xcite for the case of symmetric @xmath6-stable processes .
[ def : gen : distr ] the _ distributional generator _ of @xmath4 is the tempered distribution @xmath193 defined by the formula let @xmath194 for @xmath70 .
we define @xmath195 ( the _ reflection _ of @xmath196 ) by the formula @xmath197 . hence , @xmath198 and @xmath199 for @xmath70 ( see @xcite ) . by , @xmath200 and @xmath201 for @xmath132 .
[ prop : gen : distr ] when @xmath202 , then @xmath203 .
we have , by the definition of @xmath69-convolution , recall that @xmath204 .
hence , @xmath205 is the @xmath206 limit of @xmath207 . since @xmath208 , @xmath209 and the convolution kernel of @xmath140
are integrable , and @xmath73 is bounded , by fubini we have by dominated convergence , as desired .
[ cor : gen : cb ] if @xmath169 ( as in lemma [ lem : gen : cb ] ) , then @xmath210 , and @xmath150 commutes with partial differential operators on @xmath168 . led @xmath126 be a partial derivative operator ( of arbitrary order ) . by proposition
[ prop : gen : distr ] , we have @xmath211 .
when @xmath4 is one - dimensional and symmetric , then takes the form where @xmath212 and @xmath213 . for @xmath214 ( as in lemma [ lem : gen : cb ] )
, we have here @xmath215 denotes the cauchy principal value integral : of course , when @xmath4 is symmetric , then @xmath216 .
the main references for the notion of a killed process ( or part of a markov process ) are @xcite , where general strong markov processes are studied . here
we consider a lvy process @xmath4 in @xmath67 .
let @xmath217 be the time of first exit from @xmath126 , @xmath218 . the _ killed process _
@xmath219 is a strong markov process in @xmath126 with lifetime @xmath217 , such that @xmath220 for all @xmath221 .
the transition operators @xmath222 , the generator @xmath223 , and domains @xmath224 corresponding to the killed process @xmath219 are defined in the same way as for the free process @xmath4 , that is , and whenever the limit exists in the topology of the function space @xmath152 ; in this case , we write @xmath225 .
again we abbreviate , for example , @xmath226 to @xmath227 . when @xmath4 is the brownian motion , then @xmath150 is the laplace operator in @xmath67 , and @xmath223 is the laplace operator in @xmath126 with dirichlet boundary condition . even in this case
the relation between @xmath223 and @xmath150 is very delicate . on the other hand ,
this relation is crucial for the proof of theorem [ th : eigenfunctions ] .
hence , we now discuss this topic in more detail .
suppose first that @xmath4 is a compound poisson process : @xmath228 , @xmath229 and @xmath136 is a finite measure . in this case
, the situation is clear .
the following result seems to be well - known to specialists , but it is difficult to find in literature ( cf .
example 3.3.7 in @xcite for the case @xmath230 ) .
[ lem : cp ] let @xmath4 be a compound poisson process in @xmath67 , with lvy measure @xmath136 , and let @xmath146 $ ] .
then the semigroup @xmath222 is strongly continuous on @xmath231 , the generator @xmath223 is a bounded operator on each @xmath231 , and for all @xmath232 , with the usual extension @xmath233 for @xmath234 .
denote by @xmath235 the intensity of jumps of @xmath4 , and let @xmath236 be the sequence of ( random ) times when @xmath4 jumps .
let @xmath176 be in some @xmath231 ( @xmath146 $ ] ) , and let @xmath237 be the starting point of @xmath4 .
note that indeed , in the first summand @xmath238 implies that @xmath221 , and in the second one , when @xmath239 and @xmath240 , then @xmath241 .
we have @xmath242 in the first summand , and in the second one , @xmath243 is independent of @xmath244
. furthermore , @xmath245 . hence ,
if @xmath246 , we have it follows that , and so , finally , the proof will be complete once we show that each of the summands on the right - hand side converges to @xmath44 in @xmath231 .
since @xmath247 are increase times of a poisson process , we have where @xmath248 .
in particular , @xmath249 and @xmath250 , which shows that the first two summands in converge to @xmath44 in @xmath145 .
furthermore , since the total mass of @xmath251 is @xmath252 , we conclude that which is of order @xmath253 as @xmath124 .
hence , also the third summand in converges to @xmath44
. when @xmath4 is not a compound poisson process , then the generators of @xmath4 and @xmath219 are unbounded operators , and the relation between the domains of @xmath150 and @xmath223 is much less obvious .
a point @xmath254 is said to be a _ regular boundary point of @xmath126 _ if @xmath255 a.s . with respect to @xmath52
( note that here the inequality @xmath53 is strict , while @xmath256 is used in the definition of @xmath217 ) .
let @xmath257 denote the space of @xmath148 functions vanishing in @xmath128 .
if every @xmath254 is a regular boundary point and @xmath4 has the strong feller property ( that is , @xmath140 maps @xmath167 to @xmath147 ) , then the operators @xmath222 form a strongly continuous contraction semigroup on @xmath257 ( that is , @xmath219 has the feller property ; see @xcite ) . note that if @xmath4 is a symmetric lvy process in @xmath258 ( and not a compound poisson process ) and @xmath259 , then @xmath44 , the only boundary point of @xmath126 , is regular ( see theorem 47.5 in @xcite ) .
we have the following fundamental result due to dynkin .
let @xmath103 and denote by @xmath260 the first exit time from the ball centered at @xmath62 and with radius @xmath261 .
if the limit exists , we write @xmath262 . if @xmath262 for all @xmath237 , we write @xmath263 , and we define @xmath264 .
[ lem : d ] suppose that @xmath4 is a lvy process in @xmath67 which is a strong feller process .
suppose furthermore that @xmath265 is open , and every boundary point of @xmath126 is regular . let @xmath266 .
then @xmath267 if and only if @xmath263 and @xmath268 . in this case ,
@xmath269 for @xmath237 .
the above lemma states that , in a sense , the operators @xmath223 and @xmath150 have the same _ pointwise _ definition , but their domains are different .
this can be made even more explicit using the language of distribution theory , as we now describe .
recall that @xmath196 is the distributional generator , defined in definition [ def : gen : distr ] , and @xmath195 is the reflection of @xmath196 .
[ lem : d2 ] suppose that @xmath4 is a lvy process in @xmath67 which is a strong feller process .
suppose furthermore that @xmath265 is open , and every boundary point of @xmath126 is regular . let @xmath270 .
if the distribution @xmath271 restricted to @xmath126 is equal to a @xmath257 function , then @xmath272 and @xmath273 for @xmath237 . in the proof ,
we use the _ dynkin s formula _
( @xcite , formula ( 5.8 ) ) : when @xmath274 is a markov time , @xmath275 , and @xmath276 , then we denote by @xmath277 and @xmath278 the open and closed balls centered at @xmath62 and with radius @xmath261 . fix @xmath237 and let @xmath279 .
let @xmath280 be an approximation to identity : @xmath281 , @xmath282 and @xmath283 when @xmath284 .
we define @xmath285 .
then @xmath286 ( see lemma [ lem : gen : cb ] ) , and ( the convolution of @xmath195 , @xmath73 and @xmath287 is associative because @xmath73 is a bounded function , @xmath287 has compact support , and @xmath288 is integrable for all @xmath70 . ) in particular , @xmath289 , and therefore @xmath276 .
let @xmath290 be the time of first exit from @xmath277 .
by dynkin s formula , suppose that @xmath291 .
then using for the right - hand side , we obtain consider the limit as @xmath292 .
then @xmath287 approximates the dirac delta measure @xmath293 .
hence , @xmath285 converges uniformly to @xmath73 , and ( again by ) @xmath294 converges to @xmath295 ( @xmath237 is fixed ) .
it follows that as @xmath296 , the right - hand side tends to zero
. therefore , @xmath297 and @xmath298 .
since @xmath237 was arbitrary , we conclude that @xmath299 , and @xmath300 for @xmath237 .
the result follows from lemma [ lem : d ] .
lemma [ lem : d2 ] seems to be now .
however , a related distributional approach to generators of the free and killed processes was used for example in @xcite , in the study of harmonic functions with respect to symmetric @xmath6-stable processes .
we need the following auxiliary result .
its @xmath301 analogue is relatively easy to prove using dirichlet forms .
the uniform version seems to be new , especially in the case of unbounded @xmath126 and @xmath73 not vanishing at infinity .
[ lem : smooth : domain ] suppose that @xmath279 , @xmath265 is an open set , @xmath169 , and @xmath302 whenever @xmath303 .
suppose furthermore , that @xmath304 for all @xmath254
. then @xmath305 and @xmath306 for @xmath237 .
recall that @xmath277 is the open ball centered at @xmath62 and with radius @xmath261 . in the proof , the following well - known result is used . for completeness
, we give a simple proof .
[ prop : ball ] for all @xmath307 , @xmath308 is bounded in @xmath53 .
let @xmath309 satisfy @xmath310 , @xmath118 for @xmath311 , and @xmath312 for all @xmath62 .
define @xmath313 .
then , by dynkin s formula , since there are few tools to handle functions non - vanishing at infinity , we use the definition of @xmath223 . for @xmath314
$ ] , denote by @xmath315 the set of @xmath62 such that @xmath316 , and @xmath317 .
our strategy is to show that away from the boundary of @xmath126 ( in @xmath315 ) we have @xmath318 , while near the boundary ( in @xmath319 ) , @xmath320 .
let @xmath321 .
note that @xmath322 does not depend on @xmath103 , and that by proposition [ prop : ball ] , @xmath323 is a bounded function of @xmath53 .
first , we consider @xmath324 , where @xmath314 $ ] will be specified later . by lemma [ lem : gen : cb ] , @xmath170 .
since @xmath73 vanishes on @xmath325 , for @xmath324 we have observe that the condition @xmath326 and @xmath240 implies that before time @xmath63 , the process _ first _
exits @xmath126 , and _ after that
_ it exits @xmath325 .
furthermore , @xmath327 a.s .
hence , by the strong markov property , since @xmath328 is increasing , and @xmath329 , we have we conclude that for @xmath324 , now we consider @xmath62 close to the boundary .
fix ( small ) @xmath330 . by corollary [ cor : gen : cb ] , @xmath331 is lipschitz continuous . hence , we can choose @xmath332 $ ] small enough , so that @xmath333 when @xmath334 .
let @xmath179 be the sequence approximating @xmath73 as in the proof of lemma [ lem : gen : cb ] .
let @xmath335 and @xmath336 .
by dynkin s formula , when @xmath337 , we use the estimate when @xmath338 , we simply have it follows that recall that @xmath180 and @xmath339 converge to @xmath73 and @xmath331 , respectively , locally uniformly , and that @xmath340 for some @xmath341 . by taking the limit as @xmath187 and applying dominated convergence , we obtain that for @xmath335 , we have @xmath333 for @xmath342 ( because @xmath334 ) . hence , finally , @xmath343 a.s . , and therefore we conclude that recall that as @xmath124 , @xmath323 is bounded .
furthermore , @xmath344 converges uniformly to @xmath331 .
hence , and imply that for @xmath63 small enough , since @xmath330 was arbitrary , the proof is complete .
from now on , we only consider one - dimensional lvy processes . the process @xmath4 corresponding to @xmath345 , @xmath229 and @xmath136 vanishing , is the brownian motion , running at twice the usual speed ( that is , @xmath346 ) .
we denote the transition density of the brownian motion by @xmath347 , a lvy process @xmath4 is said to be a _ subordinate brownian motion _ if it can be written in the form @xmath348 , where @xmath349 is the brownian motion , @xmath350 is a _ subordinator _ ( a non - decreasing lvy process ) , and @xmath349 and @xmath350 are independent processes .
we assume that under @xmath52 , @xmath349 starts at @xmath62 and @xmath350 starts at @xmath44 a.s . to avoid trivialities , we assume that @xmath4 is non - constant ,
that is , @xmath351 or @xmath136 is non - zero .
the subordinator @xmath350 is completely described by its _
laplace exponent _ @xmath5 : we have @xmath352 ( @xmath15 ,
@xmath16 ) , where here @xmath212 is the same as in , and @xmath353 is the lvy measure of @xmath350 : a radon measure on @xmath2 satisfying @xmath354 .
the relation between the lvy measures @xmath353 and @xmath136 of @xmath350 and @xmath4 is with @xmath347 defined in .
the lvy - khintchine exponent of the subordinate brownian motion @xmath4 satisfies @xmath355 . the class of subordinate brownian motions is naturally divided into compound poisson processes ( with bounded laplace exponents @xmath1 ) and processes corresponding to unbounded @xmath1 .
the latter class consists of strong feller processes : if @xmath356 a.s .
, then @xmath357 has absolutely continuous distribution .
this property is important for applications of lemma [ lem : d2 ] .
see @xcite for more information on subordinate brownian motions .
one needs to distinguish the _ killed subordinate _ and the _ subordinate killed _ processes .
the first one is @xmath358 introduced above , while the other one is obtained by subordination of the killed brownian motion .
we emphasize that these two processes are essentially different .
in particular , the spectral theory for the subordinate killed process is trivial : its eigenfunctions are the same as the eigenfunctions of the ( insubordinate ) killed brownian motion , and only the corresponding eigenvalues are different . for a discussion of the relation between killed subordinate and subordinate killed processes , see e.g. @xcite .
we recall some standard definitions ; see @xcite for further information .
a function @xmath359 is said to be _ completely monotone _ on @xmath2 if @xmath360 for all @xmath361 and @xmath362 .
a function @xmath359 is a
_ bernstein function _
if @xmath176 is nonnegative and @xmath363 is completely monotone . by bernstein s theorem ,
every completely monotone function is the laplace transform of some radon measure on @xmath94 , and every bernstein function has the form for some @xmath364 and a radon measure @xmath365 on @xmath2 satisfying @xmath366 .
whenever we write @xmath367 when @xmath176 is a bernstein function , we mean @xmath368 .
note that if @xmath369 , then @xmath176 has the form ; hence , a bernstein function @xmath176 is the laplace exponent of a subordinator if and only if @xmath369 .
a bernstein function @xmath176 for which the measure @xmath365 in the representation has a completely monotone density function is said to be a _
complete bernstein function _
( cbf in short ; also known as _ operator monotone function _ ) . recall that @xmath4 is a subordinate brownian motion , @xmath353 is the lvy measure of the underlying subordinator , and @xmath1 is its laplace exponent ( so that the lvy - khintchine exponent of @xmath4 is @xmath0 ) .
[ prop : cbf ] the following conditions are equivalent : 1 .
@xmath1 is a complete bernstein function ; 2 .
@xmath370 has a completely monotone density function @xmath371 ; 3 . for a radon measure @xmath372 on @xmath2 such that the integral
@xmath373 is finite , 4 .
@xmath1 extends to a holomorphic function in @xmath28 $ ] , which leaves the upper and the lower complex half - planes invariant ; 5 .
@xmath139 is a completely monotone function on @xmath2 .
a symmetric lvy process satisfying ( e ) is automatically a subordinate brownian motion .
equivalence of ( a ) and ( b ) is just the definition of complete bernstein functions .
equivalence of ( b)(d ) is standard , see @xcite , chapters 6 and 11 .
equivalence of ( c ) and ( e ) seems to be well - known to specialists ( cf .
@xcite ) , but a direct reference to the literature is difficult to find .
the proof is a rather straightforward calculation involving fourier transform and an application of bernstein s representation theorem ; for completeness , we provide the details .
suppose that ( c ) holds .
then , for @xmath374 , by an explicit calculation , for @xmath375 , hence , by fubini , comparing this with , we obtain which proves that @xmath139 is completely monotone on @xmath2 .
conversely , if ( e ) holds , that is , if @xmath139 is completely monotone on @xmath2 , then , by bernstein s theorem , @xmath139 has the representation , and the above reasoning can be reversed to prove that @xmath4 is a subordinate brownian motion , and that ( c ) holds .
the assumptions of the main theorems can be put in the following form .
[ asmp ] the process @xmath4 is a subordinate brownian motion satisfying any of the equivalent conditions of proposition [ prop : cbf ] .
we note the following immediate consequence of proposition [ prop : cbf ] .
[ corr : cbf ] the following conditions are equivalent : 1 .
@xmath176 is a complete bernstein function ; that is , holds for some @xmath364 and a measure @xmath365 with completely monotone density ; 2 . for some @xmath364 and a radon measure @xmath376 on @xmath2 such that @xmath377 , 3 .
@xmath378 for @xmath361 and either @xmath176 is constant , or @xmath176 extends to a holomorphic function in @xmath28 $ ] , which leaves the upper and the lower complex half - planes invariant ; 4 .
@xmath379 is the lvy - khintchine exponent of a symmetric lvy process , whose lvy measure has a completely monotone density function on @xmath2 .
the right - hand side of formula defines a holomorphic function in @xmath28 $ ] .
we often denote this extension by the same symbol @xmath176 .
we will also use the related notion of a stieltjes function : @xmath176 is said to be a _ stieltjes function _ if there are @xmath380 and a radon measure @xmath381 on @xmath2 satisfying @xmath382 such that since @xmath383 , we see that @xmath176 has the representation if and only if it is the laplace transform of the measure on @xmath94 , where @xmath293 is the dirac delta measure . in other words , @xmath176 is the laplace transform of a completely monotone function , plus a nonnegative constant .
we list some standard properties of complete bernstein and stieltjes functions .
[ prop : cbf : prop ] suppose that @xmath176 and @xmath384 are not constantly equal to @xmath44 . 1 .
the following conditions are equivalent : @xmath385 is a cbf ; @xmath386 is a cbf ; @xmath387 is a stieltjes function ; @xmath388 is a stieltjes function . 2 .
if @xmath176 , @xmath384 are cbf , @xmath389 and @xmath390 , then @xmath391 , @xmath392 , @xmath393 , @xmath394 ( extended continuously at @xmath395 ) , @xmath396 and @xmath397 are cbf .
one of the fundamental properties of complete bernstein and stieltjes functions is that the characteristics in representations and can be easily recovered from ( the holomorphic extension of ) the function .
we illustrate this for stieltjes functions . by @xmath293
we denote the dirac delta measure at @xmath44 .
[ prop : cbf : repr ] let @xmath176 be ( the holomorphic extension of ) a stieltjes function with representation . then and with the limit understood in the sense of weak convergence of measures .
formula is well - known in complex analysis , it is a variant of the sokhotskyi - plemelj formula for the cauchy - stieltjes transform in the upper complex half - plane .
the proof of a version for complete bernstein functions is contained in theorem 6.2 in @xcite ( see corollary 6.3 and remark 6.4 therein ) .
we provide a simple argument based on the representation theorem for harmonic functions in half - plane .
since @xmath398 is a nonnegative harmonic function in the upper complex half - plane , it can be represented using the poisson kernel , for unique costant @xmath399 and unique measure @xmath400 . on the other hand , by
, formula follows simply by comparing the above two representations .
the expressions are direct consequences of and dominated convergence .
we need three more properties of complete bernstein functions .
[ prop : cbf : ratio ] if @xmath176 is a complete bernstein function and @xmath401 , then @xmath402 and @xmath403 are complete bernstein functions . here
we extend @xmath384 and @xmath115 continuously at @xmath395 , @xmath404 and @xmath405 . by proposition [ prop : cbf : prop](b ) , @xmath406 is a cbf . note that @xmath407 , and so also @xmath408 is a cbf .
furthermore , @xmath409 is a cbf .
again by proposition [ prop : cbf : prop](b ) , @xmath410 is a cbf .
[ lem : g ] let @xmath176 be ( the holomorphic extension of ) a complete bernstein function , and let @xmath401 .
define then @xmath411 is a stieltjes function , and it is the laplace transform of a completely monotone function @xmath412 on @xmath2 .
furthermore , @xmath412 is the laplace transform of a finite measure on @xmath2 .
let @xmath176 have the representation , and let @xmath413 .
fix @xmath414 $ ] .
since we have furthermore , we conclude that by the definition , @xmath384 is the stieltjes function , and by the remark following , @xmath384 is the laplace transform of a completely monotone function @xmath415 , where @xmath416 is a finite measure on @xmath2 .
[ prop : cbf : ests ] if @xmath176 is a complete bernstein function , then 1 .
@xmath417 for @xmath361 ; 2 .
@xmath418 for @xmath361 ; 3 .
@xmath419 for @xmath420 , @xmath421 , @xmath422 ; 4 .
@xmath423 for @xmath102 as in ( c ) ; 5 .
@xmath424 for @xmath102 as in ( c ) . in ( c)(e )
, the holomorphic extension of @xmath176 is denoted by the same symbol .
all statements follow easily from the representation and differentiation under the integral sign .
indeed , ( a ) reduces to which follows from the inequality @xmath425 for @xmath426 . in a similar manner ,
( b ) follows from @xmath427 . to prove ( c)(e ) , observe that for @xmath420 with @xmath428 $ ] , hence , by a simple calculation , therefore , ( c ) is a consequence of @xmath429 , and ( d ) is proved using @xmath430 . finally , to prove ( e ) , use ( c ) and the inequality @xmath431 , verified easily by a direct calculation . for further properties of complete bernstein functions and related notions ,
see chapters 6 and 7 in @xcite , or chapter 3 in @xcite .
recall that @xmath348 is a subordinate brownian motion , @xmath5 is the laplace exponent of the subordinator @xmath350 , and @xmath355 is the lvy - khintchine exponent of @xmath4 .
assumption [ asmp ] is in force ; that is , @xmath1 is assumed to be a _ complete
_ bernstein function . in this section
we introduce the operation @xmath432 .
although it will be used for complete bernstein functions only , the definition is quite general .
[ def : dagger ] when @xmath1 is a positive function on @xmath2 , and the integral @xmath433 is finite , then we define for @xmath1 as definition [ def : dagger ] , substitute @xmath434 in .
note that defines a holomorphic extension of @xmath435 to the half - plane @xmath16 .
we denote this extension by the same symbol @xmath436 .
the definition of @xmath435 lies at the very heart of the wiener - hopf method . roughly speaking ,
if @xmath436 can be extended continuously to the boundary of the region @xmath16 , that is , to the imaginary axis , then we have @xmath437 , a wiener - hopf factorization of @xmath0 . for complete bernstein functions ,
this is formally proved in lemma [ lem : cbfextension ] below .
the wiener - hopf method is described in section [ sec : eigenfunctions ] .
the function @xmath435 plays an important role in fluctuation theory of lvy processes .
when @xmath4 is a symmetric lvy process , then its lvy - khintchine exponent @xmath131 is symmetric , @xmath355 for some nonnegative @xmath1 on @xmath94 .
suppose that @xmath4 is not a compound poisson process . by corollary 9.7 in @xcite
, the laplace exponent @xmath438 of the bivariate ascending ladder process satisfies this relation was used extensively in @xcite .
the fundamental lemma [ lem : cbfextension ] states that if @xmath1 is a cbf , then @xmath435 is a cbf .
before we prove it , we establish some simple properties of @xmath435 . [ prop : dagger : prop ] whenever both sides of the following identities make sense , we have : 1 .
@xmath439 , @xmath440 ( @xmath441 ) and @xmath442 ; 2 .
@xmath443 for @xmath341 ; 3 .
when @xmath43 , then @xmath444 ; 4 .
if appropriate limits of @xmath1 exist , then the corresponding limits of @xmath435 exist , and @xmath445 , @xmath446 .
properties ( a ) follow immediately from formula . in
( b ) , one uses also @xmath447 . for @xmath43 , by we have here we used the identity @xmath448 , proved easily by a substitution @xmath449 . finally , to show ( d ) , assume that a finite , positive limit @xmath450 exists .
it can be proved that the functions @xmath451 of @xmath452 are uniformly integrable as @xmath453 ( this is clear when @xmath1 is a cbf ; we omit the details in the general case ) .
it follows that and hence @xmath454 .
when the limit of @xmath1 is @xmath44 or @xmath58 , one uses in a similar way fatou s lemma ; we omit the details
. limits at @xmath58 are dealt with in a similar way .
[ prop : explogbound ] let @xmath1 be as in definition [ def : dagger ] .
if @xmath455 , @xmath441 and @xmath456 for all @xmath25 , then , with the same constants , @xmath457 , @xmath16 . in a similar manner ,
if @xmath458 for @xmath25 for some @xmath455 , @xmath441 , then @xmath459 , @xmath16 . for @xmath460 ( @xmath53 , @xmath461 ) , we have suppose that @xmath462 for @xmath375 .
then and the right - hand side extends to a bounded below harmonic function in the upper half - plane @xmath463 ( with vanishing linear term in the representation similar to ) . by the poisson formula , hence , this proves the first statement .
the other one follows by reversing the inequalities in the above argument . in proposition 2.1 in @xcite it
is proved that if @xmath5 and @xmath464 are increasing , then @xmath465 . for complete bernstein functions ,
this was independently proved in proposition 3.7 in @xcite .
recall that under assumption [ asmp ] , @xmath1 is a complete bernstein function .
hence , @xmath1 extends to a holomorphic function in @xmath28 $ ] ; we use the same symbol @xmath1 for this extension .
first , we define an auxiliary function so that @xmath466 .
note that @xmath467 is an ( integral - type ) weighted geometric mean of the family of complete bernstein functions @xmath468 ( @xmath375 ) , so that @xmath467 is a complete bernstein function .
this is formally proved in the following simple result .
[ lem : cbfmean ] if @xmath1 is a complete bernstein function , then @xmath469 and @xmath467 are complete bernstein functions . clearly , @xmath470 for @xmath25
furthermore , when @xmath471 , we have @xmath472 .
it follows that @xmath473 , that is , @xmath474 and @xmath475 . finally ,
@xmath476 , so @xmath477 and @xmath478 when @xmath479 .
the result follows by corollary [ corr : cbf](c ) .
below we prove that @xmath466 is a complete bernstein function ; this is stronger than the assertion of lemma [ lem : cbfmean ] , see proposition [ prop : cbf : prop](b ) .
[ lem : cbfextension ] if @xmath1 is a complete bernstein function , then @xmath435 is a complete bernstein function .
the holomorphic extension of @xmath435 ( denoted by the same symbol ) satisfies [ rem : wh : fact ] the first statement of the above lemma was proved independently in proposition 2.4 in @xcite using theorem 6.10 in @xcite .
after the preliminary version of the article was made available , jacek maecki pointed out to the author that the other statement of the lemma , formula , at least for purely imaginary @xmath112 , can be deduced from the wiener - hopf factorization of the lvy - khintchine exponent in fluctuation theory , see formula ( vi.4 ) in @xcite .
the novelty of lemma [ lem : cbfextension ] lies in the combination of the two parts , which is one of the key steps in the derivation of the explicit formula for the eigenfunctions @xmath12 in theorem [ th : eigenfunctions ] .
ccc ( 1,1.0675136 ) ( 0,0 ) .
dashed line depicts the boundary of the region in which the integrand is meromorphic . in ( a ) , @xmath16 and @xmath471 , and
the integrand @xmath451 is meromorphic in the region @xmath480 with pole at @xmath481 , outside the contour of integration . in ( b ) , @xmath482 and @xmath471 , and the integrand @xmath451 is meromorphic in @xmath483 with a simple pole at @xmath484 . ,
title="fig : " ] ( 0.29742268,0.42240919)(0,0)[lb ] ( 0.6542081,0.06303393)(0,0)[lb ] ( 0.73789728,0.46891429)(0,0)[lb ] ( 0.29836725,0.1298281)(0,0)[lb ] & ( 1,0.90680997 ) ( 0,0 ) .
dashed line depicts the boundary of the region in which the integrand is meromorphic . in ( a ) , @xmath16 and @xmath471 , and the integrand @xmath451 is meromorphic in the region @xmath480 with pole at @xmath481 , outside the contour of integration . in ( b ) , @xmath482 and @xmath471 , and the integrand @xmath451 is meromorphic in @xmath483 with a simple pole at @xmath484 .
, title="fig : " ] ( 0.26634191,0.33373005)(0,0)[lb ] ( 0.61959603,0.30444088)(0,0)[lb ] ( 0.07921712,0.78757119)(0,0)[lb ] ( 0.32160686,0.51651332)(0,0)[lb ] + ( a)&(b )
let @xmath176 be ( the holomorphic extension of ) the complete bernstein function defined in ( see lemma [ lem : cbfmean ] ) ; hence , @xmath466 for @xmath25 , and this identity defines a holomorphic extension of @xmath435 to the half - plane @xmath16 .
when @xmath16 , by we have for a fixed @xmath112 , the integrand on the right - hand side is a meromorphic function of @xmath452 in the region @xmath480 with a simple pole at @xmath481 .
furthermore , by proposition [ prop : cbf : ests](e ) , for any @xmath307 the integrand decays at least as fast as @xmath485 when @xmath486 in the region @xmath487 . hence , by a standard contour integration ( using contours shown in figure [ fig : contour ] ( a ) with @xmath488 ; the pole at @xmath481 is outside the contour ) and then parametrization @xmath489 of @xmath490 , we obtain that the formula on the right - hand side clearly defines a holomorphic function in @xmath28 $ ] ( integrability follows by proposition [ prop : cbf : ests](e ) ) , which is the holomorphic extension of @xmath436 .
suppose that @xmath482 and @xmath471 . using the parametrization @xmath491 of @xmath492
, we obtain that for a fixed @xmath112 , the integrand on the right - hand side is a meromorphic function of @xmath452 in the region @xmath483 with a simple pole at @xmath484 . furthermore , again by proposition [ prop : cbf : ests](e ) , for any @xmath307 , the integrand decays at least as fast as @xmath493 when @xmath486 in the region @xmath494 .
hence , again using standard contour integration ( along contours shown in figure [ fig : contour ] ( b ) with @xmath488 ; this time the pole at @xmath484 is inside the contour ) and the residue theorem , we obtain that recall that @xmath482 and @xmath471 .
hence @xmath495 , so that @xmath496 .
formula follows when @xmath482 and @xmath471 .
since @xmath1 and @xmath176 are complete bernstein functions , by corollary [ corr : cbf](c ) , @xmath497 and @xmath498 are bounded harmonic functions in the region @xmath482 , @xmath471 , taking values in @xmath499 $ ] and @xmath500 $ ] , respectively .
hence , @xmath501 and @xmath502 are poisson integrals of their ( bounded , measurable ) boundary values @xmath503 , @xmath504 ( defined on @xmath29 \cup [ 0 , i \infty)$ ] ) .
furthermore , we have @xmath505 and @xmath506 for @xmath507 , and @xmath508 $ ] , @xmath509 $ ] for almost all @xmath507 .
observe that when @xmath482 , @xmath471 , we have hence , @xmath510 is a bounded harmonic function in the region @xmath482 , @xmath471 , with boundary value @xmath384 given by @xmath511 on @xmath29 $ ] , and @xmath512 on @xmath513 .
it follows that @xmath514 $ ] for almost every boundary point @xmath112 , and so @xmath515 $ ] for all @xmath112 such that @xmath482 , @xmath471 .
it follows that @xmath516 in this region .
we have already seen that @xmath517 when @xmath16 , @xmath471 .
finally , @xmath518 , and so @xmath519 when @xmath479 . by corollary
[ corr : cbf](c ) , @xmath435 is a cbf , and extends to all @xmath520 .
we note a third formula for ( the holomorphic extension of ) @xmath436 , other than and : when @xmath1 is a complete bernstein function , then see the above proof of lemma [ lem : cbfextension ] . in section [ sec :
eigenfunctions ] , the transformation @xmath432 is applied to the function this definition is continuously extended at @xmath521 by letting @xmath522 .
we denote @xmath523 . by proposition [ prop : cbf :
ratio ] , @xmath524 and @xmath525 are complete bernstein functions , and @xmath526 .
it is easy to see that @xmath527 ( note that @xmath528 ) .
since @xmath1 is concave , its graph lies below the tangent line at @xmath521 , that is , formula ( plus @xmath529 ) combined with proposition [ prop : explogbound ] yields that we remark that more detailed estimates of @xmath524 and @xmath530 can be found in @xcite .
below we prove theorem [ th : eigenfunctions ] . our strategy is as follows .
first we state a distributional version of the spectral problem for @xmath8 , and in lemma [ lem : eigenfunctions ] we prove that a distributional solution is automatically a strong solution . the ( simpler ) case of compound poisson processes
requires a different approach and is studied separately in proposition [ prop : poisson ] .
next , we rephrase the distributional problem as a wiener - hopf equation , and we solve it using the wiener - hopf method . finally , we combine the two parts and prove theorem [ th : eigenfunctions ] .
in addition , we establish some basic properties of the eigenfunctions . in this part , @xmath4 can be an arbitrary one - dimensional _ symmetric _ lvy process with lvy - khintchine exponent @xmath131 .
we define @xmath1 by the formula @xmath355 .
note that @xmath5 need not be a bernstein function . recall that @xmath150 is the generator of @xmath4 , and @xmath8 is the generator of the process @xmath4 killed upon leaving @xmath13 , defined on an appropriate domain .
here we only consider the @xmath531 , @xmath532 and @xmath533 generators , formal definitions are given in preliminaries .
recall also that @xmath193 is the distributional generator of @xmath4 , see definition [ def : gen : distr ] .
the fourier transform of the distribution @xmath196 is @xmath534 , and since @xmath196 is symmetric , it is equal to its reflection @xmath195
. let @xmath535 . as usual ,
@xmath73 is extended to entire real line by setting @xmath302 for @xmath536 .
given some regularity of @xmath4 , if @xmath537 then , by lemma [ lem : d2 ] , the condition @xmath538 is equivalent to the apparently weaker condition @xmath539 for @xmath540 .
this motivates the following definition .
[ def : weak ] let @xmath11 .
a tempered distribution @xmath73 is said to be a _
distributional eigenfunction _ of @xmath8 , corresponding to the eigenvalue @xmath541 , if @xmath73 is supported in @xmath94 , @xmath69-convolvable with @xmath196 , and @xmath542 is supported in @xmath29 $ ] .
informally , [ lem : eigenfunctions ] suppose that @xmath4 is a symmetric lvy process , which has the strong feller property , and that @xmath44 is a regular boundary point of @xmath13 .
let @xmath11 , and suppose that @xmath73 is a distributional eigenfunction of @xmath8 , corresponding to the eigenvalue @xmath541 .
if @xmath73 is bounded and continuous on @xmath258 , and if for some @xmath543 , then @xmath544 and @xmath538 . since @xmath73 does not vanish at infinity , we can not apply lemma [ lem : d2 ] directly to @xmath73 to prove that @xmath544 .
thus , we decompose @xmath73 into the sum of a @xmath545 function and a smooth bounded function .
denote @xmath546 . by remark [ rem : spectral : free ] , @xmath547 and @xmath548 . by proposition [ prop : gen : distr ] , @xmath549 .
we choose an infinitely smooth function @xmath115 such that @xmath118 for @xmath550 and @xmath116 for @xmath62 large enough .
furthermore , we require that ( note that such a choice is always possible , as we do not assume that @xmath551 . )
we write @xmath552 , where @xmath553 and @xmath554 .
the function @xmath555 is infinitely smooth , with all derivatives bounded , so by lemma [ lem : gen : cb ] , @xmath556 .
we have @xmath557 for @xmath550 , and by and , @xmath558 .
hence , by lemma [ lem : smooth : domain ] and proposition [ prop : gen : distr ] , we have @xmath559 and @xmath560 for @xmath540 .
we claim that also @xmath561 and @xmath562 for @xmath540 .
once this is proved , we have @xmath544 and @xmath563 for @xmath14 , as desired .
hence , it remains to prove the claim .
we will apply lemma [ lem : d2 ] to @xmath554 .
since @xmath73 and @xmath115 vanish on @xmath29 $ ] , we have @xmath564 for @xmath536 . also , @xmath565 , and both @xmath566 , @xmath567 vanish at infinity , so that @xmath568 . in other words , @xmath569 .
furthermore , by assumption , @xmath570 restricted to @xmath13 is equal to @xmath571 . we have already seen that @xmath572 is a continuous function on @xmath258 which vanishes at @xmath44 .
hence , @xmath573 restricted to @xmath13 is equal to a continuous function , vanishing continuously at @xmath44 .
furthermore , for @xmath14 , by assumption , the first term on the right - hand side converges to @xmath44 as @xmath164 . since @xmath574 for @xmath62 large enough
, also @xmath575 vanishes as @xmath164 ( by lemma [ lem : gen : cb ] and proposition [ prop : gen : distr ] ) .
we conclude that @xmath573 restricted to @xmath13 is equal to a @xmath545 function . by lemma [ lem : d2 ] , @xmath576 ( and therefore also @xmath561 ) , and @xmath562 for @xmath14 .
continuity of @xmath73 at @xmath44 is essential . indeed ,
when @xmath4 is the brownian motion ( i.e. @xmath43 ) , then @xmath577 is the distributional eigenfunction of @xmath578 in @xmath2 , but @xmath73 is not in the @xmath167 domain of @xmath8 .
in particular @xmath579 is not equal to @xmath580 , as can be verified by a direct calculation .
when @xmath4 is a compound poisson process , the proof is much easier .
[ prop : poisson ] suppose that @xmath4 is a symmetric compound poisson process .
let @xmath11 , and suppose that @xmath535 is a distributional eigenfunction of @xmath8 , corresponding to the eigenvalue @xmath541 .
then @xmath538 .
the result is a straightforward application of lemma [ lem : cp ] and the fact that @xmath196 is the signed measure @xmath581 .
a method of solving problems of the form was found in @xcite in the case when @xmath196 is an integrable function satisfying some growth conditions .
for this reason , problems of this kind are called _ wiener - hopf equations _ , and the algorithm for finding their solutions is the _ wiener - hopf method _ ; see @xcite for a detailed exposition .
we begin with a brief description of the method .
a version of paley - wiener theorem states that a function is supported in @xmath94 or @xmath29 $ ] if and only if its fourier transform extends holomorphically to the upper or lower complex half - plane , respectively .
therefore , can be re - written as where @xmath582 .
the main ingredient of the _ wiener - hopf method _ is factorization of the symbol @xmath583 into the product of two functions , one extending holomorphically to the upper half - plane , the other one to the lower half - plane ( the _ wiener - hopf factors _ ) .
then @xmath76 is taken to be the multiplicative inverse of the first factor .
the fundamental condition for the wiener - hopf method is that the symbol does not vanish . in our case
, we have @xmath584 for @xmath585 .
for this reason , we apply the wiener - hopf method to the _ regularized _ symbol here we use the notation of . from now on ,
assumption [ asmp ] is in force .
it is required in order to establish some properties of @xmath586 and its wiener - hopf factors .
in particular , since @xmath524 is a complete bernstein function , the factorization is given by lemma [ lem : cbfextension ] : the wiener - hopf factors are @xmath587 and @xmath588 , and the solution is expected to satisfy @xmath589 , up to a multiplicative constant .
due to the singularity in the denominator , we prefer to work with the laplace transform of @xmath73 .
below we provide details for the above idea .
fix @xmath11 , and define a tempered distribution where @xmath196 is the distributional generator of @xmath4 and @xmath293 is the dirac delta measure . hence , @xmath590 .
we decompose @xmath591 into the convolution of the distribution ( the _ singular part _ ) and two infinitely divisible measures @xmath592 and @xmath593 ( _ regular parts _ ) supported in @xmath94 and @xmath29 $ ] respectively . clearly , we define the tempered distribution @xmath594 by means of fourier transform , with @xmath524 defined in .
since @xmath595 is compactly supported , @xmath595 and @xmath594 are @xmath69-convolvable , and since @xmath596 , by the exchange formula we have [ lem : laplacerp ] there is a probability measure @xmath592 supported in @xmath94 , such that ( see ) furthermore , @xmath592 has a completely monotone density on @xmath2 ( but it may have an atom at @xmath44 ) . if @xmath597 is the reflection of @xmath592 , then @xmath598 and @xmath599 ( @xmath374 ) . by lemma [ lem : cbfextension ]
, @xmath530 is a complete bernstein function . by proposition [ prop : cbf : prop](a ) , @xmath600 is a stieltjes function , and therefore it is the laplace transform of a measure @xmath592 , which has completely monotone density function on @xmath2 . by proposition [ prop :
dagger : prop](d ) , @xmath601 .
hence , @xmath592 is a probability measure .
finally , @xmath602 ( where the holomorphic extension of @xmath530 is denoted by the same symbol ) .
let @xmath593 be the reflection of @xmath592 .
by , for @xmath374 .
the lemma follows by the exchange formula .
[ lem : reduction ] suppose that @xmath71 is a distribution supported in @xmath94 , for which @xmath80 is bounded for any @xmath70 .
if @xmath603 is supported in @xmath604 , then @xmath73 is a distributional eigenfunction of @xmath8 .
note that both @xmath73 and @xmath605 are supported in @xmath94 , and so they are @xmath69-convolvable . furthermore , for any @xmath104 , @xmath606 is bounded , and @xmath607 , @xmath608 are integrable .
therefore , the @xmath69-convolution of @xmath73 , @xmath605 and @xmath593 is associative .
in other words , if we let @xmath609 , then @xmath610 .
clearly , if the support of @xmath611 is @xmath604 , then @xmath612 is supported in @xmath29 $ ] .
lemma [ lem : reduction ] enables us to describe a family of solutions to . if @xmath603 is supported in @xmath604 , then @xmath613 for some polynomial @xmath614 .
therefore , the fourier transform of a distributional eigenfunction @xmath73 is expected to have the form @xmath615 ; however , extra care is needed because of the zeros of the denominator @xmath616 at @xmath617 .
let ( see ) where is a normalization constant .
since @xmath530 is a complete bernstein function , @xmath176 extends to a meromorphic function in @xmath28 $ ] , with poles at @xmath618 .
we denote this extension again by @xmath176 . by and
proposition [ prop : dagger : prop](b ) , this holds true for complex @xmath112 with @xmath16 .
it follows that @xmath469 is equal to the right - hand side of . in order to prove theorem [ th : eigenfunctions ] , we need to show that @xmath176 is the laplace transform of a function @xmath73 of the form , and that @xmath73 is a ( distributional , and hence @xmath531 ) eigenfunction of @xmath8 .
we define and note that @xmath619 is the laplace transform of @xmath620 .
hence , @xmath621 is the laplace transform of by lemma [ lem : g ] ( applied to a cbf @xmath622 ) , @xmath384 is a stieltjes function , it is the laplace transform of a completely monotone function @xmath191 , and @xmath412 is the laplace transform of a finite measure .
hence , @xmath176 is the laplace transform of @xmath623 .
[ prop : residue ] we have @xmath624 , where @xmath625 is given by . by , for @xmath626
we have we claim that the integrand on the right - hand side is dominated by an integrable function .
then , by dominated convergence , as desired ( see and ) . hence , it remains to prove the claim .
observe that since @xmath524 is nonnegative and concave , we have @xmath627 for @xmath628 , and @xmath629 for @xmath630 .
this and monotonicity of @xmath524 gives that it follows that which is integrable in @xmath375 .
the above proposition and formulae and give that @xmath631 .
we summarize our development in the following result .
[ cor : f ] the function @xmath176 is the laplace transform of a function @xmath623 , where @xmath412 is completely monotone on @xmath2 and it is the laplace transform of a finite measure on @xmath2 .
[ lem : f ] the function @xmath73 constructed above is a distributional eigenfunction of @xmath8 . by
, we have @xmath632 .
since @xmath633 and both @xmath634 and @xmath412 are bounded , also @xmath73 is a bounded function .
hence , @xmath73 is @xmath69-convolvable with @xmath605 , and @xmath635 when @xmath16 . by the exchange formula , @xmath636 , which is a distribution supported in @xmath604 .
an application of lemma [ lem : reduction ] completes the proof .
clearly , also the distributional derivatives ( of all orders ) of @xmath73 are distributional eigenfunctions of @xmath8 .
however , they are not in the @xmath10 domain of @xmath8 . in order to apply lemma [ lem : eigenfunctions ] , we need to show that @xmath73 vanishes continuously at @xmath44 . [
lem : f0 ] the function @xmath73 has a nonnegative right limit at @xmath44 .
if @xmath1 is unbounded , then @xmath73 vanishes continuously at @xmath44 . since @xmath412 is non - increasing ( even completely monotone ) , the limit of @xmath637 as @xmath638 exists .
since @xmath639 , where @xmath176 is given by , we have the above expression is nonnegative , proving the first part .
furthermore , by proposition [ prop : dagger : prop](a , c ) , @xmath640 .
if @xmath1 is unbounded on @xmath94 , then @xmath641 tends to @xmath44 as @xmath642 ( see ) .
hence , by proposition [ prop : dagger : prop](d ) , @xmath643 converges to @xmath44 as @xmath642 , and therefore @xmath644 . in the remark following
it was already observed that @xmath645 coincides with the right - hand side of the formula for @xmath646 , so that @xmath647 .
we will show that @xmath12 satisfies also the remaining conditions of theorem [ th : eigenfunctions ] . by corollary [ cor : f ]
, @xmath648 has the form with completely monotone @xmath649 , and @xmath19 is the laplace transform of a finite measure @xmath20 on @xmath2 . by lemma [ lem : f ]
, @xmath12 is the distributional eigenfunction of @xmath8 .
if @xmath1 is unbounded , then @xmath650 vanishes continuously at @xmath44 ( lemma [ lem : f0 ] ) , and @xmath651 ( because @xmath652 tends to @xmath44 at infinity ) .
furthermore , the subordinate brownian motion @xmath4 has the strong feller property , and @xmath44 is a regular boundary point of @xmath13 ( see preliminaries ) .
hence , lemma [ lem : eigenfunctions ] applies : @xmath653 and @xmath654 .
when @xmath1 is bounded , then @xmath8 is a bounded operator on @xmath10 , and @xmath654 by proposition [ prop : poisson ] . in both cases
@xmath12 is an @xmath531 eigenfunction of @xmath8 .
note that ( see @xcite ) for @xmath540 . for a fixed @xmath62 , this integral equation
is easily solved , and we obtain @xmath655 , that is , @xmath12 is the eigenfunction of @xmath7 . by proposition [ prop : residue ] , @xmath656 .
recall that @xmath657 ( by proposition [ prop : dagger : prop](d ) ) , and hence ( by representation for @xmath530 ) @xmath658 .
it follows that @xmath659 , as desired .
it remains to prove formula .
suppose that ( the holomorphic extension of ) @xmath5 extends to a continuous function @xmath21 ( not to be confused with @xmath436 ) in the region @xmath660 , and @xmath661 for all @xmath25 .
recall that @xmath662 ( with @xmath384 given by ) , and @xmath652 . hence , by the remark following , @xmath663 , where @xmath381 is the representation measure in for the stieltjes function @xmath384 . to prove formula
, it suffices to apply proposition [ prop : cbf : repr ] , as we now describe .
we have by lemma [ lem : cbfextension ] , when @xmath471 and @xmath664 , hence , @xmath384 extends to a continuous function @xmath665 in the region @xmath660 , and for @xmath25 , note that @xmath666 and @xmath667 are real for @xmath25 . by proposition [ prop : cbf : repr ] , @xmath668 is the measure @xmath381 in the representation of the stieltjes function @xmath384 . since @xmath663 , @xmath669 and @xmath522 , we obtain an application of completes the proof .
using , formulae of theorem [ th : eigenfunctions ] can be written in a more concise form .
we record them for easier reference : [ rem : gamma ] the formula for @xmath20 remains true for general cbf @xmath1 , as long as it is understood in the distributional sense .
more precisely , it can be proved that @xmath20 is the limit of as @xmath296 .
we omit the details . in order to extend theorem [ th : eigenfunctions ] to more general symmetric lvy processes ,
one needs two ingredients : formula must define a function @xmath12 , and the convolution @xmath670 needs to be associative ( that is , some regularity of @xmath73 and @xmath671 is needed ) .
this seems to be problematic , even for general subordinate brownian motions .
we begin with a technical result , useful for the computation and estimates of @xmath625 .
we have by series expansion and integration by parts , as desired .
the following formula for @xmath625 is often more convenient than . for @xmath11 , by substituting @xmath672 in
, we obtain a substitution @xmath673 in the integral over @xmath674 and @xmath675 in the integral over @xmath676 yields [ prop : diff : theta ] the function @xmath625 is differentiable in @xmath11 , and can be differentiated in @xmath38 under the integral sign .
let @xmath677 for @xmath678 and @xmath507
. then @xmath679 .
therefore , it suffices to prove that @xmath680 is integrable over rectangles @xmath678 , @xmath681 , where @xmath682 .
we have using twice cauchy s mean value theorem ( for two functions @xmath683 and @xmath684 ) , we obtain that for some @xmath685 $ ] and @xmath686 $ ] , since @xmath687 is bounded ( by proposition [ prop : cbf : ests](b ) ) , the above expression is uniformly bounded in @xmath681 and @xmath678 . furthermore ,
when @xmath681 and @xmath688 , by the usual mean value theorem , for some @xmath165 depending on @xmath689 and @xmath690 .
the proof is complete .
[ prop : theta ] we have note that @xmath691 . hence , by , it follows that finally , by , if @xmath692 for some @xmath11 , then @xmath1 is linear , @xmath4 is a brownian motion , and @xmath692 for all @xmath11 .
since @xmath524 is increasing , for @xmath678 we have @xmath693 , and it is easily checked that equality holds if and only if @xmath1 is linear .
hence , by , @xmath694 , and if @xmath1 is not linear , then @xmath695 .
now we turn to the properties of @xmath12 .
[ prop : lf:0 ] we have the result follows from @xmath657 ( see proposition [ prop : dagger : prop](d ) ) , @xmath696 , and .
[ lem : gest ] for all @xmath697 , @xmath698 . furthermore , clearly , @xmath699 , and by lemma [ lem : f0 ] , we have @xmath700 .
this proves the first statement .
furthermore , @xmath701 . by , , and
proposition [ prop : residue ] , [ prop : lfest ] we have since @xmath696 , by and we have by proposition [ prop : cbf : ests](b ) , @xmath702 , and follows .
[ lem : lfest2 ] suppose that there are @xmath703 such that @xmath704 for @xmath112 large enough .
then ( cf . )
whenever @xmath11 , and @xmath16 .
since @xmath1 is concave and increasing , for @xmath705 we have let @xmath706 be the supremum in ( finite by the assumption ) , so that @xmath707 for all @xmath708 .
then by proposition [ prop : explogbound ] , we have the lemma follows from @xmath709 and . [ lem : power ] suppose that there are @xmath703 such that @xmath704 for sufficiently large @xmath112 .
then for each @xmath11 , the function @xmath12 is hlder continuous with any exponent less than @xmath6 .
more precisely , there is an absolute constant @xmath710 such that for all @xmath711 , @xmath11 and @xmath712 .
we decompose @xmath713 into the difference of @xmath714 and @xmath715 in a similar way to the decomposition @xmath716 in the derivation of the formula for @xmath12 , but this time @xmath714 and @xmath715 have a better decay rate at infinity . when @xmath16 , define note that @xmath717 is the laplace transform of the function @xmath718 , and @xmath719 . therefore , @xmath714 is the laplace transform of note that @xmath720 is lipschitz continuous on @xmath258 .
in fact , for @xmath14 we have so that the lipschitz constant of @xmath720 is not greater than @xmath38 .
the function @xmath715 extends continuously to @xmath721 .
denote the summands in the parentheses in the definition of @xmath715 by @xmath722 and @xmath723 , so that @xmath724 .
we first consider @xmath725 for @xmath726 $ ] .
we have when @xmath727 , then we understand that @xmath728 $ ] denotes the interval @xmath729 $ ] . by proposition [ prop : cbf : ests](c , d ) , @xmath730 and @xmath731 for @xmath732 ; the latter inequality is used frequently in the remainder of the proof .
using also @xmath733 , we obtain by , @xmath734 , where @xmath706 is the supremum in .
hence , when @xmath735 $ ] , we simply have @xmath736 , @xmath733 , so that using , we obtain that @xmath737 . together with @xmath738 , this gives similar estimates hold for @xmath739 .
it follows that the estimate for @xmath740 $ ] is obtained using @xmath741 ( here @xmath742 denotes the continuous boundary limit ) . by the inequalities @xmath743 , @xmath744 , we have hence , using also and the inequality @xmath745 , for @xmath740 $ ] we obtain this way we have estimated @xmath746 for all @xmath374 .
the function @xmath715 is the laplace transform of @xmath747 , and @xmath747 is the inverse fourier transform of @xmath748 ( @xmath374 ) .
fix @xmath711 .
we have since @xmath749 , the supremum is at most @xmath750 . furthermore
, in particular , @xmath747 is hlder continuous of order @xmath261 , with hlder constant at most @xmath751 .
the lemma is proved .
[ cor : power ] suppose that for some @xmath307 , ( see ) .
then there is @xmath341 ( depending on @xmath261 ) such that and note that in general , @xmath752 , see proposition [ prop : cbf : ests](b )
. choose @xmath753 small enough , so that holds with @xmath261 replaced by @xmath6 .
let @xmath754 .
for @xmath755 we have the integrand on the right - hand side is positive for @xmath756 for some @xmath455 .
hence the above expression is bounded uniformly in @xmath38 and @xmath112 , @xmath757 .
the result follows by lemmas [ lem : lfest2 ] and [ lem : power ] .
recall that a function @xmath176 is said to be _ regularly varying _ of order @xmath758 at @xmath58 or at @xmath44 , if for all @xmath759 , @xmath760 or @xmath761 , respectively .
[ cor : power2 ] suppose that @xmath1 is regularly varying of order @xmath762 at @xmath58 and regularly varying of order @xmath763 at @xmath44 .
then for all @xmath764 , the hypothesis of corollary [ cor : power ] holds true .
let @xmath765 .
since @xmath1 is a regularly varying bernstein function , also @xmath766 and @xmath767 are regularly varying at @xmath44 and @xmath58 ( see @xcite ) .
hence , if we define @xmath754 , then @xmath115 is regularly varying of order @xmath768 at zero , and regularly varying of order @xmath769 at @xmath58 , and a similar statement with orders decreased by @xmath3 is true for @xmath770 .
it follows that the function @xmath771 ( continuous on @xmath13 ) has positive limits at @xmath44 and @xmath58 , and we can now repeat the proof of corollary [ cor : power ] .
when @xmath1 is regularly varying at infinity , then the behavior of @xmath12 near @xmath44 for a fixed @xmath11 has a very simple description .
we write @xmath772 as @xmath638 or @xmath164 when @xmath773 converges to @xmath3 as @xmath638 or @xmath164 , respectively .
[ lem : regular ] if @xmath1 is unbounded in @xmath13 and @xmath5 is regularly varying at infinity of order @xmath774 $ ] , then consider an auxiliary function : since @xmath524 is a complete bernstein function , both @xmath775 and @xmath776 are increasing ( proposition [ prop : cbf : prop](a ) ) .
it follows that for @xmath25 , @xmath777 , we have @xmath778 . hence note that @xmath779 has a positive limit @xmath780 as @xmath642 .
hence , it is bounded above and below by positive constants when @xmath781 .
it follows that for some @xmath341 ( depending on @xmath38 ) , for @xmath782 $ ] , the inequalities in are reversed , and we obtain the same bound for @xmath783 . since @xmath1 is regularly varying of order @xmath6 and unbounded , for all @xmath375 . hence , by dominated convergence , since the integral of @xmath784 over @xmath2 is zero , we have these two formulae yield that ( see ) since @xmath785 , this ( and ) gives and follows . since @xmath12 is increasing on the interval @xmath786 ( see ) ,
formula follows by karamata s tauberian theorem and monotone density theorem ( see @xcite ) .
in this section we study the @xmath30 properties of the operators @xmath7 and prove theorems [ th : spectral ] , [ th : pdt ] and [ th : fpt ] .
let us define the operator @xmath787 by the formula for @xmath788 . here
@xmath789 is given by theorem [ th : eigenfunctions ] .
at least formally , @xmath787 is the adjoint of @xmath35 defined in . in this section ,
we write @xmath790 for @xmath791 , and @xmath792 for @xmath793 .
[ lem : pi : bounded ] the operator @xmath787 extends to a bounded operator on @xmath30 .
we follow the proof of theorem 3 in @xcite .
let @xmath788 .
define where @xmath794 .
clearly , @xmath795 .
let so that @xmath796 .
we have by lemma [ lem : gest ] , when @xmath797 , by symmetry , for all @xmath798 , hence , using hardy - hilbert s inequality ( see e.g. @xcite ) in the last step , the lemma follows .
the extension of @xmath787 to @xmath30 is denoted by the same symbol . by lemma [ lem : pi : bounded ]
, we see that @xmath787 is indeed the adjoint of @xmath35 defined in .
[ lem : pi : spec ] we have @xmath799 for @xmath800 .
furthermore , for @xmath801 such that @xmath802 is integrable in @xmath11 , we have again the argument follows the proof of theorem 3 in @xcite .
let @xmath803 be the kernel of @xmath7 ( i.e. the @xmath52 distribution of the killed process @xmath804 ) .
if @xmath788 , then @xmath805 is integrable in @xmath806 . theorem [ th : eigenfunctions ] and fubini yield in this case .
the general case @xmath801 follows by approximation .
let @xmath807 , @xmath808 , and define @xmath809 , @xmath810 . from it follows that @xmath811 and @xmath812 .
the operators @xmath7 are self - adjoint , so that by induction , we have @xmath813 for @xmath814 .
in particular , if @xmath815 and @xmath816 , then @xmath817 .
the right - hand side tends to zero as @xmath818 , so that @xmath819 and @xmath820 are orthogonal in @xmath30 . by approximation , this holds true for any @xmath800 such that @xmath821 for @xmath822 and @xmath823 for @xmath824 .
define @xmath825 for borel @xmath826 .
clearly , @xmath827 .
if @xmath828 and @xmath829 , then finally , suppose that @xmath830 , where @xmath831 and @xmath832 .
since @xmath833 converges to @xmath834 in @xmath30 , also @xmath835 converges to @xmath836 in @xmath30 , and so @xmath837 .
it follows that @xmath365 is an absolutely continuous measure on @xmath2 , and by approximation , for any @xmath800 .
the lemma will be proved if we show that @xmath838 .
fix @xmath839 and define @xmath840}$ ] for @xmath841 .
clearly , @xmath842 , and @xmath843 ) / \delta = \|\pi^ * f_\delta\|_2 ^ 2 $ ] .
furthermore , @xmath844 , where we study the behavior of @xmath503 , @xmath504 and @xmath845 as @xmath846 . by lemma [ lem : gest ]
, we have @xmath847 and @xmath848 .
hence , by fubini , so that @xmath849 tends to zero as @xmath846 .
the function @xmath504 is the imaginary part of the fourier transform of the function @xmath850}$ ] . by proposition [ prop :
diff : theta ] , we have @xmath851 for some @xmath341 ( depending on @xmath852 ) .
hence @xmath853 , and so @xmath854 .
it follows that @xmath855 also tends to zero as @xmath846 .
finally , for the function @xmath503 , by fubini , by taking the limit @xmath296 and using dominated convergence for the second term of the integrand , we obtain that hence , using @xmath856 ( @xmath857 ) , we conclude that for all @xmath839 , that is , @xmath858 . by lemma [ lem : pi : spec ] , @xmath859 is an @xmath30 isometry , and by the extra assumption , its adjoint @xmath32 is injective . hence , both operators are unitary on @xmath30 .
the theorem follows now directly from lemma [ lem : pi : spec ] and @xmath860 . in @xcite , theorem [ th : spectral ]
is proved in full generality .
more precisely , it is shown that @xmath35 is always injective .
here we are satisfied with the following observation : when @xmath4 is a symmetric @xmath6-stable process , then , by example [ ex : stable ] below , @xmath861 , and therefore @xmath862 . since
a self - adjoint isometry is necessarily unitary , we see ( by lemma [ lem : pi : spec ] ) that @xmath35 is injective . hence , we have proved the following result .
[ cor : stable ] when @xmath36 $ ] and @xmath4 is the symmetric @xmath6-stable process ( that is , @xmath37 ) , then the hypothesis of theorem [ th : spectral ] holds true . in the following proofs of theorems
[ th : pdt ] and [ th : fpt ] , we use the full statement of theorem [ th : spectral ] . in other words
, we assume that @xmath35 is injective ( which is proved to be true in general in @xcite ) .
let @xmath788 . by theorem
[ th : spectral ] , for all @xmath863 .
the function @xmath864 is jointly integrable in @xmath806 . by fubini ,
since @xmath788 is arbitrary , the kernel of @xmath7 has the form . by monotone convergence , where @xmath865 .
we have by theorem [ th : spectral ] , if we can use dominated convergence , then , by proposition [ prop : lf:0 ] , as desired .
hence , it remains to show that the integrand in is dominated by an integrable function when @xmath296 . by the assumption ,
there are @xmath703 such that @xmath866 for @xmath56 ( see the proof of corollary [ cor : power ] ) .
hence , @xmath867 is integrable in @xmath11 . by proposition [ prop : lfest ]
, we have @xmath868 , and by corollary [ cor : power ] ( for @xmath869 and @xmath870 ) , @xmath871 for some @xmath341 ( depending on @xmath62 ) .
hence , the integrand in is bounded by @xmath872 , which is integrable in @xmath11 .
formula is proved . by the same argument , the integrand in is integrable jointly in @xmath11 and @xmath873 .
formula follows by integrating both sides of it in @xmath63 over @xmath874 , using fubini and comparing the result with .
[ ex : stable ] let @xmath37 , where @xmath875 .
then @xmath350 is the @xmath876-stable subordinator , and @xmath348 is the symmetric @xmath6-stable lvy process . by and , ( note that the _ phase - shift _ of @xmath877 was conjectured , in somewhat different setting , in @xcite ; see @xcite for further discussion . ) by theorem [ th : eigenfunctions ] , the eigenfunctions of @xmath7 and @xmath8 are given by the formula where ( see and ) clearly @xmath878 does not depend on @xmath38 , and finally , where by lemma [ lem : regular ] , we have @xmath879 as @xmath638 . by theorem
[ th : spectral ] ( or corollary [ cor : stable ] ) , the functions @xmath12 yield a generalized eigenfunction expansion of @xmath8 and @xmath7 , and by theorems [ th : pdt ] and [ th : fpt ] , for @xmath880 , these results were obtained in @xcite , and the formulae for @xmath50 and @xmath64 can be significantly simplified in this case .
[ ex : relativistic ] if @xmath881 for some @xmath882 , then @xmath350 is the relativistic subordinator with mass @xmath365 , and the corresponding lvy process is the relativistic @xmath3-stable process with mass @xmath365 .
its infinitesimal operator @xmath150 is the ( quasi-)relativistic hamiltonian of a free particle with mass @xmath365 , and @xmath8 is the hamiltonian corresponding to the semi - infinite potential well . by , which increases with @xmath38 from @xmath44 as @xmath883 to @xmath884 as @xmath885 .
the eigenfunctions of @xmath7 have the form @xmath789 , where @xmath19 is the laplace transform of an explicit nonnegative function . by lemma [ lem : regular ] , @xmath886 as @xmath638 .
see @xcite for a detailed discussion of this example .
[ ex : cauchybrownian ] let @xmath887 , @xmath875 , @xmath351 .
then @xmath350 is the @xmath888-stable subordinator with drift , and the corresponding lvy process @xmath4 is a mixture of the brownian motion and the symmetric @xmath6-stable lvy process .
when @xmath880 , then , by , this decreases with @xmath38 from @xmath884 as @xmath883 to @xmath44 as @xmath885 , and can be written explicitly in terms of the dilogarithm function @xmath889 .
for general @xmath6 , the expression for @xmath625 is more complicated , and it can be proved that @xmath625 decreases from @xmath890 to @xmath44 . by lemma [ lem : regular ] , @xmath891 as @xmath638 .
[ ex : gamma ] let @xmath892 , so that @xmath350 is the gamma subordinator .
the subordinate process @xmath4 is called variance gamma process . by
, it can be proved that @xmath625 increases with @xmath38 from @xmath44 as @xmath883 to @xmath893 as @xmath885 . by lemma [ lem : regular ] , the eigenfunctions @xmath12
satisfy @xmath894 as @xmath638 , with @xmath895 .
in particular , @xmath12 is not hlder continuous .
note that the assumptions of theorem [ th : pdt ] are satisfied only for @xmath896 , but theorem [ th : fpt ] applies for all @xmath53 . on the other hand , the more restrictive condition is not met , so this case is not covered by the proof of theorem [ th : fpt ] given above . the general argument of @xcite needs to be used .
[ ex : loglog ] let @xmath897 .
it can be verified that in this case @xmath625 is greater than @xmath898 for some @xmath38 , e.g. @xmath899 .
this proves that it is not true in general that @xmath900 , even if @xmath1 is unbounded .
furthermore , @xmath867 is not integrable in @xmath11 for any @xmath53 , so theorem [ th : pdt ] is of no use here .
theorem [ th : fpt ] applies for @xmath896 , but the additional assumption is not satisfied .
[ ex : bounded ] let @xmath901 . in this case
the subordinator @xmath350 is the compound poisson process with exponential jump distribution , and the jumps of @xmath4 have laplace distribution with density @xmath902 .
note that @xmath903 is holomorphic in @xmath904 , and hence @xmath20 vanishes .
it follows that @xmath905 . by and a contour integration , we omit the details .
hence , @xmath906 satisfies @xmath907 for all @xmath540 . this can be alternatively proved by a direct calculation .
it can be easily proved that @xmath20 vanishes if and only if @xmath908 ( i.e. @xmath4 is a brownian motion ) or @xmath909 for some @xmath455 ( which corresponds to the process @xmath4 studied in this example , up to time and space scaling ) .
in general , @xmath20 may have non - zero singular part .
for example , consider @xmath910 and @xmath911 .
then @xmath912 , and hence @xmath20 has an atom at @xmath60 for @xmath911 .
as it was the case in @xcite , where the cauchy process was studied , the eigenfunctions @xmath12 are the boundary values of solutions of a certain system of pdes , a spectral problem with spectral parameter in the boundary . in @xcite ,
the formula for @xmath12 ( for the case of the cauchy process ) was found by solving the corresponding system of pdes .
here the argument goes in the opposite direction , and the result for pdes is an application of the formula for @xmath12 .
we need some background on kren correspondence . for a complete bernstein function @xmath1
, there is a unique _ string _
@xmath365 ( a non - decreasing , right - continuous unbounded function from @xmath94 to @xmath913 $ ] ) which is the _ kren representation _ of @xmath1 . for simplicity
, we only consider the case when @xmath914 , @xmath365 is continuously differentiable on @xmath2 , and @xmath915 for all @xmath916 .
denote @xmath917 .
then , for each @xmath918 there exists a ( unique ) nonnegative , non - increasing continuous function @xmath919 such that @xmath920 for @xmath916 , and @xmath921 .
furthermore , for the details , we refer to @xcite , or chapter 14 in @xcite , and the references therein .
the kren s correspondence between @xmath365 and @xmath1 is a rather complicated concept , and there are only few examples of explicit pairs , see @xcite .
let @xmath922 and define @xmath923 using the fourier transform in @xmath62 , it is easy to check that @xmath924 is the solution of the problem furthermore , so that @xmath925 , where @xmath150 is the operator with fourier symbol @xmath534
. theorems [ th : eigenfunctions ] implies the following result .
[ th : pde ] let @xmath926 be a positive continuous function on @xmath2 such that @xmath927 is integrable over @xmath674 , but not over @xmath676 .
for @xmath918 , let @xmath919 be the nonnegative , non - increasing continuous solution of @xmath920 ( @xmath916 ) , satisfying @xmath921 .
define @xmath928 , and let @xmath12 be the function given in theorem [ th : eigenfunctions ] .
finally , let then @xmath929 is continuous and bounded in @xmath930 , @xmath931 , twice differentiable in @xmath930 , @xmath916 , and it satisfies using the properties of @xmath932 , it is easy to check that @xmath929 is a weak solution of . since @xmath933 ( for a fixed @xmath931 ) is a sum of an @xmath934 function and @xmath935 ( in a similar way as in section [ sec : eigenfunctions ] ) , @xmath923 is a bounded function .
continuity of @xmath929 and its first and second partial derivatives in @xmath930 , @xmath916 follows by elliptic regularity theorem .
continuity of @xmath929 on the boundary is proved using the properties of @xmath932 and @xmath49 .
we provide two examples .
if @xmath936 for a constant @xmath401 , then we find that @xmath937 and @xmath938 .
hence @xmath1 is the characteristic exponent of the relativistic subordinator , described in example [ ex : relativistic ] . for a related example , see section 2.7 in @xcite .
consider now @xmath875 and @xmath939 , where @xmath940 .
we have @xmath941 , where @xmath942 is the modified bessel function of the second kind and @xmath943 .
moreover , @xmath944 , so that @xmath1 is the characteristic exponent of the @xmath888-stable subordinator , considered in example [ ex : stable ] . when @xmath880 , @xmath945 .
the problem with @xmath945 was studied in @xcite to find the eigenfunctions of @xmath7 .
earlier , a similar relation for more general domains ( also in higher dimensions ) was applied e.g. in @xcite , and for general symmetric stable processes e.g. in @xcite .
related problems appear frequently in hydrodynamics ( the _ sloshing problem _ ) , see the references in @xcite .
the operator @xmath946 is the generator of a two - dimensional diffusion @xmath947 on @xmath948 , with reflecting boundary . by repeating the construction given in example 3.1 in @xcite
, we can identify the subordinate brownian motion @xmath4 studied in the present article with the trace left on the horizontal axis by @xmath947 .
we remark that the connection between subordinate brownian motion in @xmath67 and traces of diffusions in @xmath949 was applied e.g. in @xcite to find formulae for the distribution of some lvy processes stopped at the time of first exit from the half - line or the interval .
a related problem for the trace of a two - dimensional jump - type stable process was studied in @xcite .
i would like to express my gratitude to tadeusz kulczycki for many enlightening discussions .
i also thank krzysztof bogdan , tomasz byczkowski , piotr graczyk , alexey kuznetsov , jacek maecki , micha ryznar , renming song and zoran vondraek for their comments to the preliminary version of the article .
i am particularly thankful to the anonymous referee for helpful remarks and many detailed suggestions .
chen , r. song , _ spectral properties of subordinate processes in domains_. stochastic analysis and partial differential equations , pp .
7784 ( eds .
g - q chen , e. hsu and m. pinsky ) , ams contemp .
( 2007 ) .
k. l. chung , _ doubly - feller process with multiplicative functional_. in : _
seminar on stochastic processes , 1985 _
( gainesville , fla . , 1985 ) , progr .
statist .
12 , birkhuser , boston , boston , ma . , 1986 , 6378 .
s. kotani , s. watanabe , _
krein s spectral theory of strings and generalized diffusion processes_. in _ functional analysis in markov processes ( katata / kyoto , 1981 ) _ , pp . 235259 , lecture notes in math .
923 , springer , berlin , 1982 .
m. g. kren , _ integral equations on a half - line with kernel depending upon the difference of the arguments_. transl .
( 2 ) 22 ( 1962 ) , pp .
( translated from uspekhi mat .
nauk 13(5 ) ( 1958 ) , pp . 3120 . )
n. wiener , e. hopf , _ ber eine klasse singulrer integralgleichungen_. sitzungber .
berlin ( 1931 ) , pp . 696706 .
( reprinted in _
norbert wiener : collected works _ , vol . 3 , mit press , cambridge , mass . , 1981 . | we study one - dimensional lvy processes with lvy - khintchine exponent @xmath0 , where @xmath1 is a complete bernstein function .
these processes are subordinate brownian motions corresponding to subordinators , whose lvy measure has completely monotone density ; or , equivalently , symmetric lvy processes whose lvy measure has completely monotone density on @xmath2 .
examples include symmetric stable processes and relativistic processes .
the main result is a formula for the generalized eigenfunctions of transition operators of the process killed after exiting the half - line .
a generalized eigenfunction expansion of the transition operators is derived . as an application , a formula for the distribution of the first passage time ( or the supremum functional )
is obtained . |
with the discovery of the new resonance of mass around 125 gev by the atlas and cms collaborations at the lhc in 2012 , a change of paradigm has taken @xcite .
the discovery has unambiguously established the role of the higgs mechanism in electroweak symmetry breaking ( ewsb ) .
all the properties of the new particle measured so far are consistent with that of the standard higgs boson .
thus , one may be tempted to conclude that for all practical purposes , the newly found particle is like that of the standard model ( sm ) higgs boson , and new physics effects are decoupled as far as the higgs sector is concerned . at the same time , it is well known that there are difficulties associated with the higgs sector of the sm that need to be addressed .
the main difficulty is the hierarchy problem associated with the quadratically diverging quantum corrections to the mass of the higgs boson when computed in the sm .
there is no remedy to this difficulty within the sm , and for a higgs boson of mass 125 gev , the new physics effects should show up within the tev range to cure this malady . assuming that the new physics effects are expected to appear only indirectly in the higgs sector , it is natural to consider these effects through effective couplings of the higgs bosons , with itself as well as with the gauge bosons and heavy fermions .
precise measurement of these couplings is very essential to establish the true nature of the ewsb mechanism .
while the lhc is capable of probing some of these couplings @xcite , especially the higgs couplings with the gauge bosons and top quark , one may need to rely on a cleaner machine like the international linear collider ( ilc ) @xcite for the required precision . within an effective lagrangian
, the effect of new physics could be studied in the various couplings through the quantum corrections they acquire .
such an effective lagrangian basically encodes the new physics effects in higher - dimensional operators with anomalous couplings .
+ the study of the higgs sector through an effective lagrangian goes back to refs.@xcite .
more recently , the lagrangian including a complete set of dimension - six operators was studied by refs .
@xcite . for some of the recent references discussing the constraints on the anomalous couplings within different approaches ,
please see refs .
reference @xcite studied the @xmath6 , where @xmath7 associated production at the lhc and tevatron to discuss the bounds obtainable from the global fit to the presently available data , whereas ref .
@xcite has discussed the constraint on the parameters coming from the lhc results as well as other precision data from lep , slc , and tevatron .
experimental studies on the higgs couplings at the lhc are presented in , for example , refs . @xcite .
+ the measurement of trilinear higgs couplings is best done through the process @xmath8 @xcite . at the same time , this process also depends on the higgs - gauge boson couplings , @xmath9 and @xmath10 , which will affect the determination of the @xmath11 coupling . another process that could probe the @xmath11 couplings is @xmath12 following the @xmath13 fusion @xcite , which is also affected by the @xmath14 and @xmath15 couplings . in a recent study @xcite , we investigated the effect of the @xmath16 coupling , where @xmath17 , in the extraction of the @xmath11 coupling , and found that a precise knowledge of the @xmath14 and @xmath9 couplings is necessary to derive information regarding the trilinear couplings . apart from this
, an investigation of higgs - gauge boson couplings has been performed in the recent study of wwh production at ilc @xcite .
the anomalous couplings like @xmath18 and @xmath19 has been studied @xcite in the process of @xmath20 .
the table [ table : constraints ] contains obtained limits on coupling parameters in previous studies .
+ the process @xmath21 is well suited to study the higgs to neutral gauge boson couplings @xcite .
this process is influenced by the trilinear couplings like @xmath22 and @xmath23 which can contaminate the effects of @xmath9 couplings . in this paper
we will focus our attention on this process in some detail within the framework of the effective lagrangian .
one goal of this study is to investigate cp - conserving higgs sector through higgs to gauge boson couplings and to understand the significance of other couplings in their measurement .
.current 95% cl bounds ( 2nd column ) and future higgs factory 2@xmath24 exclusion sensitivities ( 3rd and 4th columns ) on the coefficients of dim.-6 operators that contribute to @xmath25 .
[ cols="^,^,^,^ " , ]
the discovery of the higgs boson by the atlas and cms collaborations at the lhc has confirmed the higgs mechanism as the way to have ewsb providing masses to the fundamental particles .
the properties of the higgs boson measured by the lhc so far are consistent with the expectations of the sm .
it is expected that the lhc would measure the mass , spin , and parity of this particle along with the standard decay widths somewhat precisely . on the other hand , details of the couplings like the trilinear and quartic self - couplings as well as the couplings with the gauge bosons
are not expected to be measured precisely . at the same time , precise knowledge of these couplings is very important in reconstructing the ewsb mechanism . a precision machine like the ilc is expected to help in the precise measurement of these couplings . in this paper , the process @xmath0 , which is influenced by the higgs to gauge boson couplings , namely , @xmath26 and @xmath23 is considered .
the reach of an ilc at @xmath27 gev with an integrated luminosity of 300 fb@xmath1 in probing the different relevant parameters of the corresponding effective lagrangian is presented .
the influence of the presence of other couplings in the probe of each of the couplings are studied .
we observed that the study of the @xmath28 , @xmath29 and @xmath30 planes show that the allowed region can be narrowed to a very small band .
while this band is unaffected by the presence of @xmath31 , the effect is significant if @xmath32 .
consideration of the angular distributions of the higgs boson ( @xmath33 ) , the z - boson ( @xmath34 ) , @xmath35 , @xmath36 and @xmath37 and the transverse momentum as well as energy distributions of the higgs boson , z - boson and photon have proved to provide a handle in distinguishing the presence of different scenarios of @xmath38 , @xmath4 and @xmath5 .
all other parameters have an indistinguishable effect on these distributions .
the invariant mass distributions of the @xmath39 , @xmath40 and @xmath41 pairs are also sensitive to some parameters related to anomalous couplings . a quantitative estimate of the forward - backward asymmetry corresponding to the angle between @xmath42 and @xmath43 as well as z and@xmath42
show that large deviations of up to 77% are possible for moderate values of the couplings . in all cases ,
a suitably chosen beam polarization is found to be advantageous , as illustrated with an 80% left - polarized electron beam and a 30% right - polarized positron beam .
the statistics can be improved by a factor of 1.5 with the baseline polarization quoted above , which can be improved to an enhancement factor of 1.7 with the expected 60% positron beam polarization in the upgraded version of the ilc .
apart from the overall normalizing factor , some details are also affected , as is illustrated in the improvements in the forward - backward asymmetry , when beam polarization is used .
thus , the study shows that @xmath44 production at the ilc is useful in detecting the anomalous couplings in higgs - gauge boson interactions . a detailed analysis involving standard kinematic distributions
could be used to distinguish different scenarios involving the couplings .
while the numerical study needs to be improved with more realistic collider and detector information , as well as study of the background processes , we hope to have conveyed the importance of the process in determining and disentangling the effects of anomalous higgs - gauge boson couplings . + * acknowledgements * sk would like to acknowledge the financial support from the serb - dst , india , under the national post - doctoral fellowship programme , grant no .
pdf/2015/000167 and the iit - gandhinagar during the initial stages of the project .
we thank b. coleppa and p. poulose for useful discussions .
* a. * the most general formula for the matrix element square for arbitrary polarized @xmath45 beams is given in refs.@xcite .
if one considers massless electron and positron , the final formula is reduced to the following form @xcite , where @xmath47 is the helicity amplitude for the process under consideration , and @xmath48 and @xmath49 are the helicities of the electron and positron respectively .
@xmath50 = degree of longitudinal polarization and @xmath51 = degree of transverse polarization of electrons and positrons .
the @xmath52 refers to the angle of polarization of the electron and the positron , respectively . in our numerical analysis , @xmath53 is zero as the azimuthal angle of the initial beams is set to zero in madgraph . + * b. * throughout the paper , we calculated 3-@xmath24 limit from the cross section with an integrated luminosity ( @xmath54 ) of 300 @xmath55 using following formula .
+ @xmath56-@xmath24 = @xmath57 , where @xmath58 abc s. chatrchyan _ et al . _
( cms collaboration ) , phys . lett .
b * 716 * , 30 ( 2012 ) .
g. aad _ et al . _
( atlas collaboration ) , phys .
b * 716 * , 1 ( 2012 ) .
atlas collaboration , report no .
atlas - conf-2013 - 029 , 2013 [ http://cds.cern.ch/record/1527124/files/atlas-conf-2013-029.pdf ] .
g. aad _ et al . _
( atlas collaboration ) , phys .
b * 726 * , 88 ( 2013 ) .
s. chatrchyan _ et al . _
( cms collaboration ) , phys .
b * 716 * , 30 ( 2012 ) .
s. chatrchyan _ et al . _
( cms collaboration ) , j. high energy phys . * 06 * ( 2013 ) 081 .
g. aad _ et al .
_ ( atlas collaboration ) , phys .
b * 716 * , 1 ( 2012 ) .
v. khachatryan _ et al . _
( cms collaboration ) , eur .
j. c * 74 * , 3076 ( 2014 ) .
g. aad _ et al .
_ ( atlas collaboration ) , phys . rev .
d * 90 * , 112015 ( 2014 ) .
e. gabrielli , m. heikinheimo , l. marzola , b. mele , c. spethmann , and h. veermae , phys .
d * 89 * , 053012 ( 2014 ) .
j. brau _ et al . _
( ilc collaboration ) , arxiv:0712.1950 .
g. aarons _ et al . _
( ilc collaboration ) , arxiv:0709.1893 .
g. moortgat - pick , t. abe , g. alexander , b. ananthanarayan , a. a. babich , v. bharadwaj , d. barber , a. bartl _ et al .
. rep . * 460 * , 131 ( 2008 ) .
e. boos , v. bunichev , m. dubinin , and y. kurihara , phys .
b * 739 * , 410 ( 2014 ) .
s. weinberg , physica ( amsterdam ) * 96a * , 327 ( 1979 ) .
s. weinberg , phys .
lett . * 91b * , 51 ( 1980 ) .
h. georgi , annu .
nucl . part .
sci . * 43 * , 209 ( 1993 ) .
w. buchmuller and d. wyler , nucl .
b*268 * , 621 ( 1986 ) .
k. hagiwara , s. ishihara , r. szalapski , and d. zeppenfeld , phys .
d * 48 * , 2182 ( 1993 ) .
k. hagiwara , r. szalapski , and d. zeppenfeld , phys .
b * 318 * , 155 ( 1993 ) .
s. alam , s. dawson , and r. szalapski , phys .
d * 57 * , 1577 ( 1998 ) . v. barger , t. han , p. langacker , b. mcelrath , and p. zerwas , phys .
d * 67 * , 115001 ( 2003 ) .
g. f. giudice , c. grojean , a. pomarol , and r. rattazzi , j. high energy phys . *
* ( 2007 ) 045 .
r. contino , c. grojean , m. moretti , f. piccinini , and r. rattazzi , j. high energy phys .
05 ( 2010 ) 089 .
r. contino , arxiv:1005.4269 .
r. grober and m. muhlleitner , j. high energy phys . 06 ( 2011 ) 020 .
b. grzadkowski , m. iskrzynski , m. misiak , and j. rosiek , j. high energy phys .
* 10 * ( 2010 ) 085 .
m. baak , m. goebel , j. haller , a. hoecker , d. kennedy , r. kogler , k. moenig , m. schott _ et al .
_ , eur .
j. c * 72 * , 2205 ( 2012 ) .
m. b. einhorn and j. wudka , nucl .
b*876 * ( 2013 ) 556 .
r. contino , m. ghezzi , c. grojean , m. muhlleitner , and m. spira , j. high energy phys .
* 07 * ( 2013 ) 035
. s. willenbrock and c. zhang , annu .
nucl . part .
sci . * 64 * , 83 ( 2014 ) .
f. bonnet , m. b. gavela , t. ota and w. winter , phys .
d * 85 * , 035016 ( 2012 ) .
t. corbett , o. j. p. eboli , j. gonzalez - fraile , and m. c. gonzalez - garcia , phys .
d * 86 * , 075013 ( 2012 ) .
w. f. chang , w. p. pan , and f. xu , phys .
d * 88 * , 033004 ( 2013 ) .
j. elias - miro , j. r. espinosa , e. masso , and a. pomarol , j. high energy phys .
* 11 * ( 2013 ) 066 .
s. banerjee , s. mukhopadhyay , and b. mukhopadhyaya , phys .
d * 89 * , 053010 ( 2014 ) .
e. boos , v. bunichev , m. dubinin , and y. kurihara , phys . rev .
d * 89 * , 035001 ( 2014 ) .
e. masso and v. sanz , phys .
d * 87 * , 033001 ( 2013 ) . z. han and w. skiba , phys .
d * 71 * , 075009 ( 2005 ) .
t. corbett , o. j. p. eboli , j. gonzalez - fraile , and m. c. gonzalez - garcia , phys .
d * 87 * , 015022 ( 2013 ) .
b. dumont , s. fichet , and g. von gersdorff , j. high energy phys . * 07 * ( 2013 ) 065 .
a. pomarol and f. riva , j. high energy phys .
* 01 * ( 2014 ) 151 . j. ellis , v. sanz and t. you , j. high energy phys . * 07 * ( 2014 ) 036 .
h. belusca - maito , arxiv:1404.5343 .
r. s. gupta , a. pomarol , and f. riva , phys .
d * 91 * , 035001 ( 2015 ) .
atlas collaboration , phys .
b * 726 * , 88 ( 2013 ) .
d. teyssier ( atlas and cms collaborations ) , arxiv:1404.7311 .
a. de rujula , m. b. gavela , p. hernandez , and e. masso , nucl .
b*384 * , 3 ( 1992 ) .
a. gutierrez - rodriguez , m. a. hernandez - ruiz , o. a. sampayo , a. chubykalo , and a. espinoza - garrido , j. phys .
jpn . * 77 * , 094101 ( 2008 ) .
y. takubo , arxiv:0907.0524 .
j. tian , k. fujii and y. gao , arxiv:1008.0921 .
m. battaglia , e. boos , and w. m. yao , studying the higgs potential at the e+ e- linear collider , econf * 010630 * , e3016 ( 2001 ) .
v. barger , t. han , p. langacker , b. mcelrath , and p. zerwas , phys .
d * 67 * , 115001 ( 2003 ) .
r. killick , k. kumar , and h. e. logan , phys .
d * 88 * , 033015 ( 2013 ) .
a. djouadi , w. kilian , m. muhlleitner , and p. m. zerwas , eur .
j. c * 10 * , 27 ( 1999 ) .
s. kumar and p. poulose , arxiv:1408.3563 .
s. kumar , p. poulose and s. sahoo , phys .
d * 91 * , no . 7 , 073016 ( 2015 ) [ arxiv:1501.03283 [ hep - ph ] ] .
h. baer , t. barklow , k. fujii , y. gao , a. hoang , s. kanemura , j. list , h. e. logan _ et al .
_ , arxiv:1306.6352 .
c. castanier , p. gay , p. lutz , and j. orloff , arxiv:0101028 .
k. fujii , in proceedings of the higgs snowmass workshop , princeton , new jersey , 2013 [ slides available from http://physics.princeton.edu/ indico / conferencedisplay.py?confid=127 ] .
q. h. cao , h. r. wang and y. zhang , arxiv:1503.05060 [ hep - ph ] .
h. y. ren , chin .
c * 39 * , no .
11 , 113101 ( 2015 ) doi:10.1088/1674 - 1137/39/11/113101 [ arxiv:1503.08307 [ hep - ph ] ]
. v. m. abazov _ et al . _
[ d0 collaboration ] , phys .
lett . * 109 * , 121802 ( 2012 ) .
j. alwall , m. herquet , f. maltoni , o. mattelaer , and t. stelzer , j. high energy phys . * 06 * ( 2011 ) 128 .
j. alwall , r. frederix , s. frixione , v. hirschi , f. maltoni , o. mattelaer , h .- s .
shao , and t. stelzer _ et al .
_ , j. high energy phys . * 07 * ( 2014 ) 079 .
feynrules , http://feynrules.irmp.ucl.ac.be/wiki/hel .
a. alloul , b. fuks , and v. sanz , j. high energy phys .
* 04 * ( 2014 ) 110 .
r. kleiss , z. phys .
c * 33 * ( 1987 ) 433 .
doi:10.1007/bf01552550 k. i. hikasa , phys .
d * 33 * ( 1986 ) 3203 .
doi:10.1103/physrevd.33.3203 b. melic and m. patra , arxiv:1610.02983 [ hep - ph ] . | higgs boson couplings with gauge bosons are probed through @xmath0 in an effective lagrangian framework . for this study the beam polarization facility at the ilc along with the typical center - of - mass energy of 500 gev is considered .
the reach of the ilc with an integrated luminosity of 300 fb@xmath1 in the determination of @xmath2-conserving parameters is obtained .
sensitivity of the probe of each of these couplings in the presence of other couplings is investigated .
the most influential coupling parameters are @xmath3 .
other parameters of significant effect are @xmath4 and @xmath5 .
a detailed study of the various kinematic distributions represents possibilities to disentangle the effect of some of these couplings .
# 1phys .
rev .
* d#1 * |
the discovery of colossal magnetoresistance ( cmr ) in perovskite - type manganites has attracted considerable attention.@xcite double - exchange ( de ) mechanism , @xcite strong electron - phonon coupling , @xcite phase separation scenarios @xcite or a griffiths singularity @xcite were suggested to clarify the origin of the cmr effect , but a conclusive microscopic model has not yet been established .
ever since , the occurrence of cmr effects has been reported for various other classes of materials , such as pyrochlores,@xcite rare - earths based compounds like gdi@xmath8,@xcite and ternary chalcogenide spinels @xmath9cr@xmath8s@xmath10.@xcite these cmr materials have been classified in terms of spin - disorder scattering and a universal dependence of the magnetoresistence vs. carrier density has been suggested on theoretical grounds.@xcite ramirez _ et al .
_ drew attention to the spinel system in 1997.@xcite in polycrystalline with @xmath11 k , the cmr effect reaches values comparable to those observed in perovskite oxides . the substitution of fe by cu increases @xmath12 to temperatures above room temperature , and the cmr effect remains relatively strong ( @xmath137 % ) .@xcite in addition , solid solutions of the ferrimagnetic semiconductor fecr@xmath8s@xmath10 and the metallic ferromagnet cucr@xmath8s@xmath10 show a number of puzzling properties : from the very beginning , a controversial discussion has been arising whether the cu ions are mono- or divalent for @xmath14.@xcite for @xmath15 it was established that only monovalent and hence diamagnetic ( @xmath16 ) cu exists in the mixed crystals.@xcite moreover , fe@xmath0cu@xmath17cr@xmath8s@xmath10 shows two metal - to - insulator transitions as a function of @xmath18 , as the room - temperature resistivity reveals two minima at @xmath19 and @xmath20 and concomitantly the seebeck coefficient changes sign two times.@xcite additionally , band - structure calculations predicted that the fe@xmath0cu@xmath17cr@xmath8s@xmath10 system should exhibit a half - metallic nature.@xcite recent experimental investigations of single crystals indicated a strong dependence of their magnetic and magneto - transport properties on hydrostatic pressure suggesting a strong magneto - elastic coupling.@xcite measurements on the ac susceptibility in pure fecr@xmath8s@xmath10 exhibited a cusp in the low - field magnetization and the onset of magnetic irreversibilities at 60 k was explained by domain - reorientation processes.@xcite later on , ultrasonic studies indicated an anomaly in the temperature dependence of the shear modulus close to 60 k , and it was suggested that the onset of orbital order induces a structural distortion at this temperature.@xcite this result , however , is hardly compatible with the observation that orbital order is established in polycrystals close to 10 k , while an orbital glass state is found in single crystals.@xcite optical spectroscopy simultaneously probes the lattice and electronic degrees of freedom and is , therefore , ideally suited to investigate structural phase transitions and to clarify the importance of electron - phonon coupling for the cmr effect.@xcite earlier infrared ( ir ) studies in polycrystalline reported , in accordance with the crystal - lattice symmetry @xmath21 , the existence of four ir - active phonons , which strongly depend on temperature near and below @xmath12.@xcite we performed measurements of the optical properties of single crystals of fe@xmath0cu@xmath17cr@xmath8s@xmath10 ( @xmath22 and 0.5 ) to shed light on the interplay of structural and electronic properties in these compounds . since the optical properties of the samples with @xmath23 and @xmath24 were found to be very similar , we only show and discuss the corresponding data for @xmath24 in the following .
single crystals of were grown using a chemical transport - reaction method with chlorine as transport agent and the ternary polycrystals as starting material .
details of the sample preparation are described elsewhere.@xcite no indication for the existence of secondary phases was found by x - ray diffraction analysis of powdered single crystals .
x - ray single - crystal analysis confirmed the high structural homogeneity of the samples .
the composition and homogeneity of the samples were examined by electron - probe microanalysis .
the samples were optically polished platelets with dimensions of about @xmath25 mm@xmath26 .
structural , magnetic and electrical transport data are given in ref . .
two fourier - transform - infrared spectrometers with a full bandwidth of 10 to 8000 @xmath27 ( bruker ifs 113v ) and 500 to 42000 @xmath27 ( bruker ifs 66v / s ) together with a @xmath28he cryostat ( oxford optistat ) were used for measurements of the optical reflectivity in the energy range from 70 to 30000 @xmath27 due to small sample dimensions and for temperatures of . in order to investigate small fractions of the sample surface in the range of 0.1 mm@xmath29 we utilized an ir microscope ( bruker irscope ii ) , which works in the far- ( fir ) and mid - infrared ( mir ) range .
figure [ fig1 ] shows the temperature dependence of the fir reflectivity @xmath30 vs. wave number of pure . in the upper panel @xmath30
is plotted for 5 and 300 k. the four visible phonon peaks are attributed to the four ir - active @xmath31 modes ( symmetry group @xmath21 , # 227).@xcite to analyze the spectra , we used a 4-parameter fit assuming frequency - dependent damping constants to account for the asymmetry of the phonon peaks .
this fitting procedure infers a splitting of the longitudinal and transverse eigenfrequencies , @xmath32 and @xmath33 , and the corresponding damping constants , @xmath34 and @xmath35 .
@xcite the resulting curves describe the measured reflectivity down to 100 @xmath27 very well , without assuming an additional contribution of free charge carriers . a representative result of these fits is shown by the solid line in the upper panel of fig . [ fig1 ] for @xmath36 k. the detailed temperature dependence of the reflectivity is visualized in the two - dimensional ( 2d ) contour plot in the lower panel of fig . [ fig1 ] .
to enable a comparison of the phonon shift , the peak positions ( maxima in @xmath30 ) for @xmath36 k are indicated as vertical lines . around @xmath37
k a shift of the phonon frequencies can be observed , especially for the mode @xmath38 close to 100 @xmath27 .
the intensity of this mode strongly depends on temperature , too ( see upper frame of fig .
[ fig1 ] ) . of vs. wave number for @xmath36 k and 300 k. a fit of the reflectivity for @xmath36 k
is indicated by the solid line .
lower panel : 2d - contour plot of the reflectivity @xmath30 vs. @xmath39 and @xmath40 generated by interpolation of 17 spectra .
the vertical lines are highlighting the maxima of the ir - active - phonons in @xmath30 at 5 k.,width=321 ] the resonance frequencies @xmath32 and @xmath33 ( left frames ) and the corresponding damping rates @xmath34 and @xmath35 ( right frames ) are shown in fig . [ fig2 ] as a function of temperature .
above the curie temperature @xmath37 k , the resonance frequencies @xmath32 and @xmath33 of all modes reveal a similar quasi - linear increase with decreasing temperature , which can be fully ascribed to anharmonic contributions to the lattice potential.@xcite in contrast to the rather usual behavior in the paramagnetic regime , modes @xmath41 and @xmath42 soften for temperatures below @xmath12 , while @xmath32 and @xmath33 increase towards lower temperatures in the case of modes @xmath43 and @xmath38 .
these anomalous changes of the eigenfrequencies in the vicinity of @xmath12 suggest a correlation with the onset of magnetic order .
however , it has to be stated that the size of the effect is different for the observed modes : @xmath44k@xmath45/\omega(t=0~$]k@xmath46 is of the order of @xmath47 % for the internal mode @xmath38 , @xmath48 % for the bending mode @xmath42 , approximately @xmath49 % for the bending mode @xmath43 , and @xmath50 % for the stretching mode @xmath41 .
longitudinal and transverse eigenfrequencies behave rather similar .
( @xmath33 ) and damping constants @xmath34 ( @xmath35 ) obtained by a 4-parameter fit for the four ir - active phonons in as described in the text .
all solid lines are drawn to guide the eye.,width=321 ] the influence of magnetic order on phonons in magnetic semiconductors has been proposed by baltensperger and helman@xcite and baltensperger@xcite more than 30 years ago , and has recently been used by sushkov _
_ to describe the phonon spectra in zncr@xmath8o@xmath10.@xcite based on a model calculation , where superexchange interaction between the magnetic ions infers a spin - phonon coupling , relative frequency shifts up to 10@xmath51 have been predicted .
the order of magnitude of this effect corresponds nicely to the experimentally observed values in fecr@xmath8s@xmath10 and , therefore , wakamura@xcite considered this mechanism to dominate the phonons behavior for @xmath52 .
subsequently , wakamura and coworkers@xcite discussed the sign of the relative frequency shift in terms of nearest - neighbor fm exchange and next - nearest - neighbor afm exchange for cdcr@xmath8s@xmath10 , which exhibits phonon modes with a similar temperature dependence as fecr@xmath8s@xmath10 .
moreover , they could show that these anomalous changes in the phonon frequencies are absent in non - magnetic cdin@xmath8s@xmath10 , further corroborating their approach.@xcite thus , the positive shift of modes @xmath41 and @xmath42 would indicate that fm exchange ( cr - s - cr ) dominates in accordance with a strong influence of the ( cr - s ) force constants on these modes , and , correspondingly , the negative shift of modes @xmath43 and @xmath38 favors afm exchange ( cr - s - cd - s - cr ) with a strong influence of the ( cd - s ) force constants . note that a more rigorous theoretical treatment of anharmonic spin - phonon and phonon - phonon interactions in cubic spinels by wesselinova and apostolov@xcite confirms the above interpretation . in the interpretation of the effect of magnetic ordering on the ir active phonon modes becomes even more complicated , because there exist , besides fm nearest - neighbor cr - s - cr bonds , additional exchange paths via afm fe - s - fe and fe - s - cr - s - fe bonds .
nevertheless , the overall temperature behavior of the phonon frequencies in fecr@xmath8s@xmath10 is similar to cdcr@xmath8s@xmath10 and may be well interpreted , accordingly .
note , however , that a critical discussion of the above approach is given by bruesch and dambrogio.@xcite a straightforward interpretation of the temperature dependence of the damping constants ( right panel of fig .
[ fig2 ] ) is not obvious at all .
again , considering only the anharmonicity of ionic non - magnetic crystals , the damping is expected to show some residual low - temperature value and a quasi - linear increase in the high - temperature limit , just as observed for the longitudinal damping constants of modes @xmath41 and @xmath42 for @xmath53.@xcite however , the temperature dependence of @xmath34 and @xmath35 in general deviates from such a behavior : in the case of mode @xmath38 both damping constants show a broad maximum just above @xmath12 and a steep decrease towards lower temperature for @xmath54 .
mode @xmath43 follows a similar temperature dependence for @xmath54 , but the reduction of the damping constants is slightly smaller , and in the paramagnetic regime @xmath34 and @xmath35 remain almost constant in contrast to the results of wakamura.@xcite the behavior of modes @xmath41 and @xmath42 for @xmath52 appears even more complex , but one can identify the onset of enhancement damping close to @xmath11 k followed by broad cusp - like maxima close to 100 k , except for @xmath33 of mode @xmath41 that increases linearly with decreasing temperatures .
wakamura@xcite argues that the maxima of mode @xmath38 ( and @xmath43 ) are due to spin fluctuations of the fe spins , in agreement with the strong influence of the corresponding force constant on this mode according to bruesch and dambrogio .
@xcite furthermore , long range spin order assumingly leads to the anomalous changes of the damping constants for all modes below @xmath12 . in comparison to the temperature dependences of the damping constants in
, one finds that modes @xmath43 and @xmath38 behave similar to the case of .@xcite on the other hand , modes @xmath41 and @xmath42 in clearly reveal a more complex behavior than in , indicating a significant influence of the iron sublattice and the additional effective exchange coupling between fe - fe and fe - cr ions on these modes .
additionally , we want to mention the large increase in intensity ( about 20 % ) for mode @xmath38 ( close to 120 @xmath27 ) when cooling from room temperature to 5 k ( see fig . 1 ) .
the intensity remains almost constant above 200 k , while a linear increase with decreasing temperature is observed below 200 k. at this temperature , maxima appear in the temperature dependence of the damping constants , suggesting a correlation of the two phenomena with regard to the spin - fluctuation scenario discussed above .
when adopting the overall interpretation of the data in terms of spin - phonon coupling , one has to consider , however , that e.g. the appearance of the cusps in the damping constants may be connected to domain reorientation processes visible in the ac susceptibility@xcite and anomalies detected by ultrasonic investigations.@xcite although the absence of significant changes of the phonon frequencies contradicts the scenario of a structural phase transition at 60 k driven by orbital ordering as suggested in ref . , it becomes clear that the complex mechanisms dominating the damping effects demand further theoretical studies to single out the important contributions in detail . vs. wave number for with @xmath19 ( upper panel ) and @xmath24 ( lower panel ) at @xmath36
k ( open circles ) and @xmath55 k ( open triangles ) .
the dashed lines represent results of fits as described in the text.,width=321 ] having discussed the phonon properties of pure fecr@xmath8s@xmath10 we now turn to the temperature dependence of the phonon modes for fe@xmath0cu@xmath17cr@xmath8s@xmath10 .
figure [ fig3 ] shows the fir reflectivity for @xmath19 ( upper panel ) and @xmath24 ( lower panel ) for temperatures 5 k and 300 k each .
the results for @xmath23 are very similar to those obtained for @xmath24 and , hence , only the data for @xmath24 is shown and discussed .
the reflectivity of both samples , @xmath19 and 0.5 , shows a drude - like contribution due to the presence of free charge carriers , while fecr@xmath8s@xmath10 can be described as an insulator .
the highest drude - like conductivity is found for @xmath19 and the phonon modes are on the verge of being fully screened . for both compounds
the internal mode @xmath38 at @xmath56@xmath27 ( see fig .
[ fig1 ] for the pure compound ) can hardly be detected .
focusing on the group of external modes , a new mode @xmath57 appears close to @xmath58@xmath27 , while on increasing cu concentration @xmath18 mode @xmath41 at @xmath59@xmath27 becomes considerably reduced in intensity . without an accompanying lattice dynamical calculation one can not decide , if this new mode represents an impurity mode due to the doping with cu or a symmetry change .
there are reports in literature@xcite claiming the reduction of symmetry to @xmath60 because of the ordering of fe and cu ions on the _ a _ sublattice . of the ir - active phonon modes in fe@xmath0cu@xmath17cr@xmath8s@xmath10 for @xmath7 ( circles ) , 0.2 ( squares ) and 0.5 ( triangles ) .
the dashed ( @xmath7 ) , dash - dotted ( @xmath19 ) and dotted lines ( @xmath24 ) indicate the magnetic transition temperatures .
the solid lines are drawn to guide the eye.,width=321 ] in this case seven ir - active phonon modes are predicted .
assuming that the peak close to 275 @xmath27 ( mode @xmath43 ) is generated by two single phonon modes ( see inset in fig .
3 ) , five modes are visible , only , while the internal mode @xmath38 close to @xmath61@xmath27 remains screened . however , our results rather point toward a continuous evolution of the phonon modes on increasing @xmath18 and favor a statistical _ a_-site distribution of cu and fe ions throughout the lattice instead of a symmetry reduction resulting from a superstructure due to an ordered _ a _ sublattice . we tried to fit the complete spectra taking into account the reflectivity up to 10000 @xmath27 , using a 4-parameter fit for the phonon modes and a lorentz oscillator for the mid - infrared excitation at about @xmath62@xmath27 ( see next section ) . representative results of these fits at low wave numbers are shown in fig .
[ fig3 ] as dashed lines .
the temperature dependences of the longitudinal modes @xmath32 as derived from these fits are shown in fig .
[ fig4 ] together with corresponding data for @xmath7 .
the transverse eigenfrequencies behave rather similar ( not shown ) .
compared to the sample with @xmath7 , the temperature dependence of the damping constants in the doped compounds is weak and therefore not shown here either . regarding the sample with @xmath19 similar anomalies as in pure fecr@xmath8s@xmath10
can be seen in the vicinity of @xmath12 for the observable modes @xmath63 , and @xmath57 .
obviously , the temperature dependence of all phonon frequencies for @xmath24 is very weak and no clear anomalies around @xmath12 are visible . within the experimental uncertainties one can detect a slight decrease of @xmath32 towards lower temperatures except for mode @xmath43 , which behaves similarly to the case of fecr@xmath8s@xmath10 ( compare fig . [ fig2 ] ) . keeping in mind the influence of spin fluctuations and spin - phonon coupling on the phonon properties in fecr@xmath8s@xmath10
, cu - doping seems to reduce these features significantly .
this observation is in agreement with reduced spin - orbit coupling due to the substitution of jahn - teller active fe@xmath64 by non jahn - teller active fe@xmath65 .
therefore , for @xmath24 only fe@xmath65 with a half - filled @xmath38-shell is present in the system@xcite and the system becomes almost magnetically isotropic as it was confirmed by ferromagnetic resonance experiments.@xcite when the reflectivities of the doped compounds with cu concentrations @xmath19 and 0.5 ( fig .
[ fig3 ] ) are compared with that of pure fecr@xmath8s@xmath10 it becomes clear that contributions from free charge carriers have to be taken into consideration .
the metallic - like behavior is most significant for @xmath19 , but it becomes reduced again on further doping . for a consistent description of the drude - type behavior of the doped compounds , it is important to measure the reflectivity spectra to higher energies .
the room - temperature reflectivities of fe@xmath0cu@xmath17cr@xmath8s@xmath10 for @xmath19 and 0.5 are plotted in the upper panel of fig .
[ fig5 ] up to 3@xmath6610@xmath28 @xmath27 , corresponding to almost 4 ev , and are compared to the reflectivity of insulating fecr@xmath8s@xmath10 . for the kramers - kronig analysis of the smoothed reflectivity data we used a low - frequency hagen - rubens extrapolation and a high - frequency extrapolation with a @xmath67 power law up to 10@xmath68 @xmath27 and a subsequent @xmath69 high - frequency tail .
the resulting dynamic conductivities @xmath70 are shown in the lower panel of fig .
we carefully checked the high - frequency extrapolation , also trying smoother extrapolations , but found that the results are not influenced in the relevant energy range below 20000 @xmath27 .
the use of a hagen - rubens extrapolation is justified by the fact that we have the complete information on the absolute values of the dc conductivities and the corresponding temperature dependences for all compounds , although we are aware of the additional uncertainties originating from the hagens - rubens extrapolation , specifically for the sample with @xmath24 . however , the best fits of the reflectivity at room temperature , even in the limited spectral range , yielded dc conductivities of 150 ( @xmath71cm)@xmath72 for @xmath5 and 35 ( @xmath71cm)@xmath72 for @xmath73 , close to the dc values derived from the 4-probe measurements on single crystals by fritsch _ et al . _ @xcite cu@xmath17cr@xmath8s@xmath10 for cu concentrations @xmath7 , 0.2 and 0.5 .
lower panel : double - logarithmic plot of the real part of the dynamic conductivity @xmath74 as derived from the reflectivity spectra.,width=321 ] for @xmath7 a weak but well defined electronic transition is observed close to @xmath75@xmath27 and a further transition appears close to 20000 @xmath76 ev ) . on substituting iron by copper ,
metallic behavior shows up and for the dc conductivity is of the order 150 ( @xmath71cm)@xmath72 .
the transition at @xmath75@xmath27 , becomes almost fully suppressed for @xmath19 .
obviously , the @xmath38-electrons become strongly delocalized .
it is generally accepted that in an ionic picture monovalent cu is substituted inducing trivalent fe .
our results suggest that the system behaves as if holes are doped into an insulator driving the compound into a metallic regime .
unexpectedly , a broad peak appears again close to 2500 @xmath27 for @xmath24 .
the observed doping dependence of the conductivity spectra as documented in fig .
[ fig5 ] can be compared with band - structure calculations of these compounds .
@xcite local spin - density approximation ( lsda ) band - structure calculations predict a half - metallic ground state of , with a partly filled @xmath57 band at the fermi level .
correlation effects via lsda+__u _ _ yield a splitting of the fe @xmath57 band into a lower and upper hubbard band characterizing fecr@xmath8s@xmath10 as a mott - hubbard insulator.@xcite the splitting of the @xmath57 band is of the order of about 0.5 ev , and , hence , the peak close to 2000 @xmath27 may be interpreted as a transition between the lower and upper hubbard band .
accordingly , the high - energy excitation can be attributed to a cr(3__d _ _ ) to fe(3__d _ _ ) transition . using an ionic picture with localized fe @xmath38 states , alternatively
, the transition at 2000 @xmath27 may correspond to a transition between the lower @xmath57 doublet and the @xmath77 triplet of the fe @xmath38-states split in a tetrahedral crystal - field .
the expected crystal - field splitting for fe@xmath64 located in the tetrahedral site of the spinel structure is rather weak@xcite and a splitting of the order @xmath78
@xmath27 seems reasonable .
further support for this interpretation comes from the observation of crystal - field transitions as measured for diluted fe@xmath64 in cdin@xmath8s@xmath10 . here
a crystal - field splitting of approximately 2500 @xmath27 has been reported by wittekoek _
_ et al.__@xcite the appearance of the broad excitation for @xmath24 in the mid - infrared region at about 2500 @xmath27 , however , can not be explained easily . in an ionic picture only trivalent iron and monovalent copper are expected for @xmath24,@xcite and recent x - ray photoelectron spectroscopy @xcite strongly favors the existence of only monovalent cu for @xmath24 .
therefore , one can exclude the possibility that the broad excitation may be attributed to fe@xmath64 similarly to the well - defined electronic excitation for @xmath7 .
nevertheless , it has been concluded from mssbauer experiments in fe@xmath0cu@xmath17cr@xmath8s@xmath10 , that an ionic picture is not applicable at all .
@xcite for @xmath79 and @xmath54 the complicated mssbauer spectra indicate two different fe sites corresponding to fe@xmath64 and fe@xmath65 , while for @xmath53 a single line pointed towards a fast electron exchange between these two sites . for @xmath24
the line pattern for @xmath53 evidenced the existence of fe@xmath65 and a strong delocalization of the cu @xmath38-derived electrons .
hence , further studies beyond the scope of this paper are needed to clarify the nature of this mid - infrared excitation . in the following
we will discuss the optical conductivity results in the low frequency range in comparison with the dc conductivity data reported in ref . .
the room - temperature spectra for the concentrations @xmath5 and 0.5 , shown in fig .
5 have been used to estimate the drude - like conductivity . for all temperatures ,
the spectra could satisfactorily be described using a plasma frequency @xmath80 @xmath27 and a dielectric constant @xmath81 for @xmath5 , which is close to the value @xmath82 for @xmath7 .
for @xmath73 we used @xmath83@xmath27 and an enhanced dielectric constant @xmath84 .
the enhanced @xmath85 indicates strong changes in the electronic excitation spectrum at higher frequencies , but due to the complexity of the spectrum in this energy region there is also a larger uncertainty in @xmath85 for @xmath73 .
the decrease of the plasma frequency by a factor of 2.4 can be explained by a decrease of the charge carrier density , as fe@xmath0cu@xmath17cr@xmath8s@xmath10 approaches a metal - to - insulator transition close to @xmath24 . with these values ,
the conductivity below 500 @xmath27 could reasonably be fitted for all temperatures as indicated by the dashed lines in fig . 3 for the spectra at 5 k and 300 k. cu@xmath17cr@xmath8s@xmath10 as determined from fits to the reflectivity ( see text ) .
the dc conductivities as observed from 4-probe measurements@xcite are indicated by solid lines and were scaled to the room temperature optical values .
lower panel : temperature dependence of the drude - like relaxation rates .
ferrimagnetic ordering temperatures are indicated by arrows .
the dashed and dash - dotted lines are drawn to guide the eye.,width=321 ] the resulting temperature dependences of the dc conductivity ( upper panel ) and relaxation rates @xmath86 ( lower panel ) are shown in fig .
the dc conductivities as derived from 4-probe measurements@xcite are indicated by solid lines .
the dc conductivities were scaled at room temperature , utilizing a factor of 1.6 for @xmath19 and a factor of 1.05 for @xmath24 . above 100 k the 4-probe dc results and the dc values as derived from the optical measurements follow a similar temperature dependence . however , at low temperatures the dc measurements are dominated by localization effects , which appear much weaker in the high - frequency ( @xmath87@xmath27 ) derived optical data .
that localization effects are most significant in the low - frequency ( `` dc '' ) transport measurements becomes clear from the fact that in doped semiconductors the conductivity below the fir regime increases almost linearly with frequency .
@xcite in the sample with @xmath24 , which exhibits the lower conductivity , localization effects dominate already at higher temperatures .
this may be attributed to a significant decrease of the charge - carrier density and concomitant increase of disorder due to the statistical distribution of the cu ions in the lattice,@xcite further discarding the possibility of @xmath9-site order of fe and cu for fe@xmath0cu@xmath17cr@xmath8s@xmath10 .
finally , we want to draw attention to the temperature dependence of the relaxation rate @xmath88 ( lower panel of fig .
6 ) . in the magnetically ordered state below @xmath12 ,
the relaxation rates become significantly reduced , e.g. the reduction amounts to almost 50% for @xmath19 .
we recall that the plasma frequency has been kept constant for each compound as a function of temperature .
this indicates that the increase of the conductivity just below the magnetic ordering temperature results from the freezing - out of disorder scattering and not from a change of the carrier density via band - structure changes at the onset of ferrimagnetic order .
taking into account the classification of chalcogenide spinels @xmath9cr@xmath8s@xmath10 as systems where cmr originates from spin - disorder scattering,@xcite the observed reduction of the relaxation rate below @xmath12 has to be regarded as direct evidence of such a scenario : in external fields the onset of ferrimagnetic order shifts to higher temperatures .
concomitantly , a reduction of the scattering rate and the anomalous increase of the conductivity arise . as a consequence
, maximal cmr effects will show up just below @xmath12 as a function of an external magnetic field .
a similar scenario has been reported for gdi@xmath8 , where the magnetic and magneto - transport properties have been described successfully in terms of spin - fluctuations and their suppression by external magnetic fields in the vicinity of @xmath12.@xcite we would like to point out , that at low temperatures the relaxation rates for @xmath19 and @xmath24 are of the same order of magnitude @xmath89@xmath27 , indicating a similar level of disorder for the cu doped compounds .
in summary , we investigated the optical properties of fe@xmath0cu@xmath17cr@xmath8s@xmath10 single crystals for cu concentrations @xmath7 , 0.2 , 0.4 and 0.5 .
phonon excitations and dynamic conductivity for @xmath23 are very similar to the results for fe@xmath90cu@xmath90cr@xmath8s@xmath10 and were not discussed separately .
the phonon excitations were measured as a function of temperature between 5 k and room temperature .
pure fecr@xmath8s@xmath10 shows clear anomalies in the eigenfrequencies at the transition from the paramagnetic to the ferromagnetic state , which can be explained by spin - phonon coupling . concerning the complex behavior of the damping constants , spin fluctuations in the vicinity of @xmath12
may describe many of the anomalous changes , but further theoretical studies are necessary to corroborate this interpretation .
the influence of magnetic order on the eigenmodes is reduced with increasing @xmath18 , and the appearance of a new phonon mode close to 350 @xmath27 is attributed to an impurity mode rather than to a symmetry reduction due to a - site order .
morover , the charge dynamics of fe@xmath0cu@xmath17cr@xmath8s@xmath10 were investigated .
fecr@xmath8s@xmath10 is an insulator , but becomes metallic when slightly doped with cu .
the conductivity of the free charge carriers can be described by a normal drude - type behavior .
the dc conductivity for @xmath19 is enhanced by a factor of four in comparison to @xmath24 .
the temperature dependence of the optically derived dc conductivity for both doped compounds is is in good agreement with resistivity measurements , but localization effects at lowest temperatures appear weaker in the optical measurements .
the corresponding behavior of the scattering rate , which shows a strong decrease below the ferrimagnetic phase transition , evidences the freezing - out of disorder scattering below @xmath12 . in accordance with the proposed classification of the ternary chalcogenide spinels as spin - disorder magnetoresistive materials ,
the reduction of the relaxation rate corroborates such a scenario and makes clear that spin - disorder has to be considered a necessary ingredient towards a theoretical description of this fascinating class of materials .
it is a pleasure to thank h .- a .
krug von nidda , j. hemberger , and ch .
hartinger for fruitful discussions .
this work was partly supported by the dfg via the sonderforschungsbereich 484 ( augsburg ) , by the bmbf / vdi via the contract no .
, by the u.s .
civilian research & development foundation ( crdf ) and by the moldavian research & development association ( mrda ) via grant no .
mp2 - 3047 . e. z. kurmaev , a. v. postnikov , h. m. palmer , c. greaves , st .
bartkowski , v. tsurkan , m. demeter , d. hartmann , m. neumann , d. a. zatsepin , v. r. galakhov , s. n. shamin , and v. trofimova , j. phys . :
matter * 12 * , 5411 ( 2000 ) . | a detailed investigation of phonon excitations and charge carrier dynamics in single crystals of fe@xmath0cu@xmath1cr@xmath2s@xmath3 ( @xmath4 ) has been performed by using infrared spectroscopy . in fecr@xmath2s@xmath3
the phonon eigenmodes are strongly affected by the onset of magnetic order . despite enhanced screening effects , a continuous evolution of the phonon excitations can be observed in the doped compounds with @xmath5 ( metallic ) and @xmath6 ( bad metals ) , but the effect of magnetic ordering on the phonons is strongly reduced compared to @xmath7 . the drude - like charge - carrier contribution to the optical conductivity in the doped samples indicates that the colossal magneto - resistance effect results from the suppression of spin - disorder scattering . |
the authors would like to thank w. vogel for his valuable comments and discussions .
this work was support by the villum kann rasmussen foundation ( denmark ) , the eu projects q - essence and malicia , and danish national research foundation .
ngj is grateful for generous support from danmarks nationalbank ( denmark ) . | we provide a straightforward demonstration of a fundamental difference between classical and quantum mechanics for a single local system ; namely the absence of a joint probability distribution of the position @xmath0 and momentum @xmath1 . elaborating on a recently reported criterion by bednorz and belzig [ phys .
rev .
a * 83 * , 52113 ] we derive a simple criterion that must be fulfilled for any joint probability distribution in classical physics .
we demonstrate the violation of this criterion using homodyne measurement of a single photon state , thus proving a straightforward signature of the breakdown of a classical description of the underlying state .
most importantly , the criterion used does not rely on quantum mechanics and can thus be used to demonstrate non - classicality of systems not immediately apparent to exhibit quantum behavior .
the criterion is directly applicable any system described by the continuous canonical variables x and p , such as a mechanical or an electrical oscillator and a collective spin of a large ensemble .
the conceptual differences between classical and quantum physics have intrigued and sometimes bewildered the physics community since the early days of quantum mechanics .
this has led to a search for indisputable manifestations of the quantum world through observations of non - classical behavior in experiments .
a field of particular curiosity is that of identifying the quantum to classical cross - over for ever larger systems , thereby eventually identifying non - classical effects in macroscopic systems .
recently this has led to the observation of , e.g. , macroscopic entangled atomic ensembles @xcite , interference of large molecules @xcite and experiments pushing toward observing non - classical effects in mechanical oscillators @xcite . in parallel to this fundamental interest , non - classicality is of central importance to quantum information processing , the essence of which is to advance computation beyond what is classically possible @xcite .
however , in some instances quantum effects are claimed by demonstrating consistency with an appropriate quantum model . yet
any rigorous demonstration of genuine quantum behavior must exclude the possibility of classical explanations .
the importance of this is exemplified in ref .
@xcite , where a pair of coupled classical oscillators is shown to exhibit signatures easily mistaken for those of entanglement expected from a quantum model .
thus , a definite conclusion on the quantum nature of a system can only result from the breakdown of the classical description and not from verified agreement with quantum mechanics .
this approach is most rigorously demonstrated by the bell - inequalities , where the underlying model of the system is stripped of any physics and is reduced to the very basic assumptions of locality and realism , resulting in an indisputable non - classicality criterion .
the bell - inequalities can not however , by their very nature , be investigated by data obtained from a single system . in this letter
we provide a conceptually simple demonstration of one of the key discrepancies between classical and quantum mechanics , valid for systems of a single degree of freedom : classical systems can always be described by a joint probability distribution for @xmath0 and @xmath1 , the two canonically conjugated coordinates of a system , whereas such a description does not apply in quantum mechanics due to the heisenberg uncertainty principle .
this discrepancy is most evident when the phase space description of the state of a system is examined .
classically , the phase space distribution @xmath2 is the joint probability of finding the system in an infinitesimal area around @xmath3 , and hence it obeys all the requirements of a probability distribution including being a non - negative function . as mentioned , in the case of a quantum phase space formulation , introduced by wigner @xcite ,
the heisenberg uncertainty renders this definition meaningless , as a joint probability distribution for @xmath0 and @xmath1 does not exist .
the phase space distribution is only defined through the single coordinate ( marginal ) distributions , projected from the distribution function @xcite and this relaxation of constraints allows for negative values of the function in areas smaller than @xmath4 .
this negativity is not directly observable due to the vacuum fluctuations preventing simultaneous measurement of @xmath0 and @xmath1 .
however , one can still infer the phase space distribution from measurements of only a single observable at a time and detect such negativities , thereby illuminating the failure of classical theory .
the usage of these negativities as markers of non - classicality has been discussed and demonstrated in several quantum optics systems ( see , e.g. , @xcite ) , using tomographic techniques .
often such methods search for the quantum state most compatible with the experimental data using statistical inference or variational techniques@xcite and thus inherently rely on quantum mechanics .
these methods are therefore not applicable for demonstrating the absence of a classical description
. alternatively , given measurements of all the coordinate distributions , the underlying state can be uniquely determined , and the phase space distribution fully calculated using the inverse radon transformation @xcite without relying on quantum mechanics .
though such methods have been used in quantum optics for demonstrating various states , the mathematical transformation involved is highly complicated .
furthermore the numerical stability of the inverse transformation is problematic , leading to numerical uncertainty at high frequencies , and sometimes results in unphysical states @xcite .
these limitations are a drawback for using tomographic techniques for validating the breakdown of a classical description , and the application of these methods is usually cumbersome .
our simple , unambiguous demonstration of the absence of a classical probability distribution is based on recent theoretical work by bednorz and belzig @xcite that verifies the negativity of the wigner function based on moments . as discussed in detail below ,
their results lead to a hierarchy of inequalities , such that violation of any one inequality indicates negativity of the wigner function . full tomographic reconstruction with the associated numerical complexities is thereby avoided .
we extend this approach such that it can be applied to quadrature measurement of a single photon state , and use the experimental data from the heralded single photon generation to directly disprove the existence of a joint probability of the position and momentum for this system .
we start by re - iterating the key results of bednorz and belzig , through a reformulation that relies only on classical mechanics .
the phase space of a system with a single degree of freedom is fully characterized by a two - dimensional phase space distribution @xmath5 .
that is , given the phase space distribution , the ensemble averaged result of any measurable quantity @xmath6 can be obtained by @xmath7 where @xmath8 is the decomposition of the quantity @xmath6 in terms of the generalized coordinate @xmath0 and its canonically conjugated momentum @xmath1 . to disprove the existence of a classical probability distribution we examine the ensemble average of a non - negative test function @xmath9 over a classically explainable system , which must have a proper distribution function that results in the ensemble average of @xmath10 be non - negative :
@xmath11 violating this condition is a direct proof of the absence of a joint probability distribution .
the condition can , however , be violated in quantum mechanics , where @xmath5 is the wigner function that can contain negative values .
the objective therefore is to optimize a test function such that it will be dominant at the possible negative areas of the distribution function . for a rotationally invariant phase space both the phase space distribution and the test function
can be described solely by the phase space radius @xmath12 , defined by @xmath13 . for reasons to become clear later
, we choose a specific form for the test function @xmath10 , writing it as a square of an @xmath14th order , even polynomial @xmath15 with real coefficients @xmath16 ; @xmath17 minimizing the above expression for a given order @xmath14 is done by straight - forward linear optimization of the coefficients @xmath16 : @xmath18 for all @xmath19 .
notice that the linearity of the problem ensures that the obtained minimum of @xmath20 is global and therefore the most optimal indicator of a possible violation of eq .
( [ eq : require ] ) for a given polynomial order @xmath14 .
it is important to emphasize that this is only a sufficient criterion for non - classicality , and an optimized positive average for a chosen @xmath14 does not ensure a classical probability distribution , since the negativity may only be exhibited by the inclusion of higher order terms in @xmath15 .
however , it is clear that increasing the polynomial order @xmath14 can not increase the minimized value of @xmath20 , and we conjecture that the limit of @xmath21 will exhibit any negativity of the wigner function , as the polynomial can represent an arbitrarily ( analytical ) sharp peaked function @xmath10 focused at the negativity .
assuming the existence of all moments ( e.g. , due to an exponentially decaying tail of the phase space distribution at large @xmath12 ) , this then becomes a necessary criterion for the negativity of the distribution function .
we also note here , that similar polynomial expansion has been discussed @xcite in the context of the p - function distribution .
the p - function is , however , only defined within the framework of quantum mechanics , and hence can not be used to prove the absence of a classical description .
we assume that , as is the case for many systems , the system in question can only be experimentally accessed by measuring one of the canonically conjugated variables ( e.g. , @xmath0 or @xmath1 ) at a time .
since we are restricted to single coordinate measurement at a time , neither the intensity nor the phase space distribution function is directly accessible .
for this method to be applicable to such experimental data , the functional @xmath22 must be expressed in terms of the moments of the projected coordinates @xmath23 , where @xmath24 is a measureable rotated coordinate .
to do this we use the identity @xmath25 where @xmath26 this is where quantum and classical approaches diverge . while classically eq .
( [ eq : mathidentity ] ) represents a measurable physical quantity , it is missing the key vacuum uncertainty , allowing for the breakdown of the classical description .
it is interesting to note the implication of identity ( [ eq : mathidentity ] ) . for the @xmath27th moment of the radial distribution to be known ,
we need @xmath27 cuts in phase space ; i.e. , different coordinate measurements at equally distributed angles .
regardless of any assumption about the underlying state , the average of eq .
( [ eq : mathidentity ] ) directly gives @xmath28 in the special case of a symmetric distribution function these moments are all identical , and eq .
( [ eq : moments ] ) reduces to @xmath29 the radial moments can thus be indirectly calculated from the quadrature measurements .
substituting these radial moments into eq .
( [ eq : require ] ) using the functional form of @xmath9 given by eq .
( [ eq : functional ] ) , we get , for a given set of measured moments @xmath30 , a necessary condition for classicality of the underlying state . if eq .
( [ eq : require ] ) is violated by the solutions of eq .
( [ eq : optimazation ] ) , the underlying state can not be explained by a proper phase space probability distribution , and one can not assign a joint probability distribution to @xmath0 and @xmath1 . to demonstrate the absence of a joint probability distribution we are going to consider the phase space description of a single photon state . in phase space
this can be described by the first excited state of a harmonic oscillator , which is rotationally invariant and contain negative parts in the wigner functions .
fig .
[ fig : epsart2 ] shows the optimal functional forms obtained for the this state for low polynomial orders . as higher order terms
are included , the optimized test function is increasingly probing the negative part of @xmath31 , yielding a negative expectation value .
profiles of the test function @xmath10 minimizing the expectation value @xmath20 for the first excited state of a quantum harmonic oscillator , as a function of the phase space radius , for different orders @xmath14 ( see text ) plotted against the profile of the corresponding wigner function . as the order of the polynomial increases , the function becomes centered around the negativity , decreasing elsewhere . in this case ,
negative expectation values are obtained starting at @xmath32 .
inset shows the polynomial order required to observe negative expectation values , as a function of the single - photon fractional content in a mixture with vacuum . as the fraction of vacuum
is increased , the state approaches a classically describable state and higher moments are needed to observe the negativity . ]
we note that negative expectation values appear only from the fourth order onward .
this is because the peak of the test function at the position of the negativity must be narrower than heisenberg s uncertainty in order not to smear the negativity ; this is in full agreement with ref .
@xcite .
the experimental demonstration is achieved with single photons generated by an heralded cavity - enhanced non - degenerate parametric down - conversion .
the equivalence between a single mode electromagnetic field and an harmonic oscillator allows us to describe the em field by a phase space of a single degree of freedom .
the down - conversion process produces two photons , and as one is detected as a trigger , the result is a single photon state where the losses introduce a statistically mixed component of vacuum .
the projection measurements ( quadratures ) are obtained by measuring the statistics of the noise , using an optical homodyne detection scheme . in this scheme ,
the weak investigated optical field is overlapped with a strong laser pulse on a beam splitter , and the interference of the two fields is detected and subtracted .
the phase of the strong laser field determines the angle @xmath33 ( eq . ( [ eq : quadratures ] ) ) of the measured coordinate .
measurements were taken without fixing the phase of the local oscillator , thus smearing the resulting distribution .
this enables us to treat the results as rotationally invariant even if non - invariant features existed prior to smearing .
such measurements will generate a rotationally invariant reconstructed state for any underlying state , but this does not necessarily average out negativities in the wigner function ( see , e.g. , @xcite ) .
for details of the experimental setup and the charactarization of the resulting single photon see ref .
@xcite . the data set contained 180,000 measured quadratures .
we have here revised the optimization of the functional to also account for statistical uncertainties inherent to a limited data set .
this is done by optimizing @xmath34 where @xmath35 is the standard deviation of @xmath10 .
the results are shown in fig .
[ fig : epsart ] .
the fact that the expectation value for our test function is negative with certainty of almost twenty standard deviations , clearly demonstrates that the measured state in this experiment can not be explained by classical theory , unambiguously negating the possibility of existence of a joint probability distribution for @xmath0 and @xmath1 .
the appearance of negative values from the twelfth order polynomials and higher indicate the quantum mechanical description of this state in terms of a wigner function includes negative valued areas .
we note that the minimized function from eq .
( [ eq : functional ] ) is monotonically decreasing for increasing order @xmath14 , and the onset of negativity at a certain order therefore means that all higher orders will also be negative .
this suggests a sequential authentication procedure for an unknown state .
as mentioned above , for a pure single photon state , negative expectation values are observable from the @xmath36-th order polynomial onwards .
the twelfth order polynomial required here is due to the vacuum component of the field , requiring higher orders of the polynomial as shown in the inset of fig .
[ fig : epsart ] , and is in agreement with the results obtained in ref . @xcite reporting 62% fraction of single photon in the resulting mixed state .
expectation value for the square of a polynomial relative to its standard deviation , as a function of the polynomial s order for the experimental data .
negativity by almost @xmath37 standard deviations disproves the existence of a joint probability distribution for @xmath0 and @xmath1 .
the inset shows a histogram of the raw measured quadrature data ( arbitrary units ) . ] in conclusion , we have experimentally demonstrated the non - existence of a joint probability distribution of two canonical variables .
this is done by violation of an inequality derived without the assumptions of quantum mechanics , thus allowing for it as proof of the absence of a classical description in systems not immediately evident to display quantum behavior .
the procedure used here can thus provide a simple , practical tool for demonstrating the non - classicality of a state based on quadrature measurements , where the existence of a classical joint distribution of two conjugated variables can be negated . in this way ,
this procedure is closely linked to other criteria @xcite demonstrating contextuality of measurements , and thus disproving the classical local hidden variable view . unlike ref .
@xcite , which are applicable to discrete variables , the method demonstrated here applies for continuous variables such as position and momentum , collective spin operators @xcite and quadrature phase operators .
this makes it useful to systems containing many particles , where criteria based on counting particles are not easily implemented and interpreted .
this method complements the full tomographic reconstruction techniques in that it is simpler and avoids numerical complexities of inverse transformations .
these kinds of conceptual proofs , when extended to different detection schemes , can shed more light on the quantum to classical correspondence , especially where the control of claimed macroscopic quantum states is in question . |
the light - matter interaction has been subject of central interest since the early years of quantum mechanics . in @xmath0 , one of the first attempts to explain the results coming from experiments was the rabi model , that describes the simplest dipole semiclassical interaction between light and matter @xcite , and is reduced to a pseudospin-@xmath1 system driven by a monochromatic classical radiation field .
however , the advent of quantum technologies such as cavity qed @xcite , have allowed us to access the quantum regime of the radiation field , where the dynamical description is given by the celebrated jaynes - cummings model ( jcm ) @xcite .
this model predicts collapses and revivals of the population inversion , the appearance of jaynes - cummings doublets as a consequence of the excitation number conservation , and it has found a testbed in several hybrid setups such as trapped ions @xcite , quantum dots @xcite , and circuit quantum electrodynamics ( circuit qed ) @xcite .
circuit qed has been growing both theoretically and experimentally , and their complex proposals such as the generation of multipartite entanglement @xcite rely on the fundamentals of the jcm and the tavis - cummings model @xcite .
this can be justified because ratios between the coupling strength @xmath2 and the resonator frequency @xmath3 may grow from typical quantum optical values of @xmath4 to circuit qed values of @xmath5 .
note that , in the latter , the rotating - wave approximation ( rwa ) can still be applied .
nowadays , two key experiments with superconducting circuits @xcite have made a significant improvement in the coupling strength , reaching values @xmath6 in the so - called ultrastrong coupling ( usc ) regime @xcite . in this case , the rwa is not longer valid and all dynamical and statical properties have to be explained through the quantum rabi model ( qrm ) h_r=_z + a + g_x(a+ ) , [ hamr ] where @xmath7 and @xmath8 are pauli matrices , @xmath9 is the annihilation ( creation ) operator , @xmath10 is half of the qubit energy , @xmath3 is the resonator frequency , and @xmath2 stands for the coupling strength .
in addition , a recent proposal considers the case where the coupling strength @xmath2 becomes comparable or larger than the mode frequency @xmath3 , @xmath11 , which is called deep strong coupling ( dsc ) regime @xcite . in this case , the dynamics can be intuitively explained as photon number wavepackets propagating along a defined parity chain .
although the state - of - the - art in circuit qed does not provide this coupling strength yet , the main features of the dsc regime have been observed in an analog quantum simulation @xcite .
the qrm described by the hamiltonian ( [ hamr ] ) , has substantial differences as compared to the jcm , in fact only recently the properties of the qrm have been completely understood @xcite . in the qrm
, a discrete @xmath12-symmetry replaces the continuous @xmath13-symmetry of the jcm .
therefore the excitation number is no longer a conserved quantity and the hilbert space splits into two infinite - dimensional invariant subspaces , the parity chains @xcite .
each eigenstate can be labeled with a @xmath12-quantum number , the parity @xmath14 , taking values @xmath15 .
furthermore , it is possible to give analytical expressions for these eigenstates as elements of the bargmann space @xcite , that allows the well - defined computation of their norms and overlaps with eigenstates of the harmonic oscillator , without truncation of the hilbert space @xcite .
the aim of this work is to present a new insight of the quantum rabi model by stressing the use of dynamical correlation functions coming from the analytical solution obtained in ref .
@xcite . in this manner , we are able to explain relevant features such as the validity of the rwa in two well - defined regions , ranging from the jc regime to higher - coupling regimes of the quantum rabi model . in addition , as the coupling strength @xmath2 enters into the dsc regime , we find that the true eigenstates of the system can be well approximated by the shifted oscillator basis in each invariant parity chain without the need for a more complicated basis .
this fact is supported by the calculation of the wigner function of the eigenstates , whose unexpected fidelity can be understood in terms of the analytical form of the eigenfunctions @xcite , resembling closely to fock states . in this context
, we also find stationary schrdinger cat - like states in the dsc regime .
finally , we present our concluding remarks .
in this section , we shall use the exact dynamics of the quantum rabi model to characterize two coupling regions that we may call _ lower coupling region _ and _ higher coupling region _
, respectively .
the first region comprises the jc regime , where the rwa holds , together with the perturbative usc regime , where small deviations from the jc occur ( @xmath16 ) .
the second identified region comprises a jump towards the higher - coupling regime , @xmath17 , and forms a precursor of the dsc regime ( @xmath18 ) .
the intermediate region , that is for @xmath19 , determines a kind of _ dark zone _ where no intuitive physics has been identified up to now .
we begin by studying a suitable dynamical observable , the time average photon number |n_0 = _ t_0^tt
n_0(t ) , [ time - av ] which measures the break of the @xmath13-symmetry in the jcm .
as we will show , this quantity exhibits significant features to characterize the regions mentioned before . since @xmath20 commutes with the parity operator @xmath21 , it is useful to investigate the dynamics within a fixed parity chain
. let @xmath22 with @xmath23 correspond to the states @xmath24 and @xmath25 of the two - level system , respectively .
then the subspace @xmath26 with parity @xmath27 is spanned by states @xmath28 where @xmath29 denotes the ( anti-)symmetric part of @xmath30 which is an element of the hilbert space @xmath31 of the radiation mode .
@xmath31 is spanned by fock states @xmath32 , which are ( anti-)symmetric if @xmath33 is ( odd ) even .
there exists a transformation @xmath34 that maps the element @xmath35 of @xmath31 onto a parity eigenstate @xmath36 that belongs to a parity chain f_|= |,= _ s|+_a| .
[ isop ] the dynamical quantities in each chain depend only on the initial distribution of photons , which fixes @xmath37 at time @xmath38 .
the state at later times follows as solution of the schrdinger equation i_t|(t)=h_|(t ) , [ schp ] where @xmath39 acts on functions in @xmath31 @xcite .
the natural observable within the invariant subspaces is thus the photon number .
we consider first the time - dependent expectation value , @xmath40 for some initial @xmath41 and positive parity . for @xmath42 and an initial fock state @xmath43 ,
the average photon number at time @xmath44 reads ( @xmath45 ) n_m(t)=m+2 ^ 2(1-(t ) ) , [ ndelta0 ] which entails that the difference @xmath46 is independent of the initial photon number , always greater than zero and oscillates with the mode frequency @xmath47 .
because fock states have definite reflection symmetry , the initial state in the full hilbert space will be the product @xmath48 . the jc result of eq . and the exact result . panel ( b ) shows the difference @xmath49 .
the black line shows where the jcm matches exactly the analytical solution .
, scaledwidth=44.0% ] starting now from the photon vacuum @xmath50 we consider values @xmath51 .
then , the correlation function @xmath52 depends on the initial state and shows a complicated oscillatory behavior .
still the time evolution can be globally characterized by the time average quantity defined in eq .
( [ time - av ] ) .
its dependence on @xmath53 for fixed @xmath54 shows features that allow to discern a regime which corresponds to the validity of the rwa ( the jc - regime ) from a region where the non - conservation of the excitation number in the qrm becomes relevant . as shown in fig .
[ fig1](a ) , the jaynes - cummings model exhibits a sharp peak of @xmath55 at @xmath56 with vanishing width as @xmath57 , and @xmath58 .
the maximum value 1/2 is reached at resonance and corresponds to rabi oscillations within the first jc - doublet , @xmath59 .
one photon is periodically exchanged between qubit and radiation field which contains 1/2 photon on average . for general values of @xmath2 and @xmath60
, the value of @xmath61 in the jcm reads @xmath62 on the other hand , the exact result exhibits an unbounded @xmath55 due to the broken @xmath13-symmetry , together with a pronounced broadening of the resonance .
figure [ fig1](b ) shows the difference of the time - averaged photon number @xmath55 calculated from the exact analytical solution and from the jcm .
notice that for @xmath63 , there is good agreement between both quantities until a value @xmath64 , establishing what we called before the lower - coupling region .
this means that with regard to photon production , the rwa result is valid for couplings associated with the usc , at least exactly at @xmath56 . for @xmath65 the deviations set in much earlier .
interestingly , for @xmath66 , _ less _ photons are generated than expected from the rwa calculation .
this corresponds to a shift of the resonant @xmath60 to values below @xmath67 . finally , for couplings
@xmath68 , what we called before the higher - coupling region , the resonance is completely gone and the photon production exceeds the rwa prediction for all @xmath60 , growing rapidly with the coupling .
this marks the crossover to the deep strong coupling regime . within the dsc , @xmath69 , the qualitative behavior of the system changes again , accompanied by the stabilization of schrdinger cat - like states ( see section [ dsc - regime ] )
we study the time evolution of the autocorrelation function @xmath70 where we start with an initial fock state with @xmath71 . from this
the revival probability is obtained as @xmath72 . in fig .
[ fig2 ] , we observe the characteristic collapse and revivals of @xmath73 in the deep strong coupling limit as in @xcite .
this feature is not present in the usc but appears only for sufficiently strong g. starting in the vacuum as in fig .
[ fig2](a ) , the revivals are explained by the fact that the initial state @xmath74 is a coherent state expressed in terms of the ground state of the quantum rabi model @xmath75 for @xmath76 . for large @xmath77
the physical behavior of the @xmath76 case is already present . that the physics of this limit is present already in a considerable distance from it is discussed in sec .
[ secapproxbasis ] .
collapse and revivals are also observed for higher initial fock states as seen in fig .
[ fig2](b ) . then the revivals become sharper but have additional contributions from higher harmonics . of state @xmath78 and in ( b ) the revival probability @xmath79 of state @xmath80 for @xmath81 and three values of @xmath54 .
the result depicted is obtained by solving eq . for positive parity.,scaledwidth=44.0% ] in fig .
[ fig3 ] , we show the time evolution of the distribution obtained by projection on the complete fock state basis , i.e. the photon number distribution . again the collapse and revival oscillations are visible in form of the red nodes of the interference patterns particularly pronounced in fig .
[ fig3](d ) .
the second feature that is observed is the bouncing of photon number wave packets " @xcite . again
the phenomenon can be understood in terms of the limit @xmath76 where the eigen basis of the quantum rabi model becomes the basis of a shifted oscillator .
this is discussed in detail in the next section , sec .
[ secapproxbasis ] . for @xmath82 ( a , b ) and @xmath83 ( c , d ) and different initial fock states @xmath78 ( a , c ) and @xmath84 ( b , d ) .
the boundaries for the `` bouncing of photon number wave packets '' can be obtained in a simple way using fig .
[ fig4](b ) . , scaledwidth=46.0% ] the natural starting point for an approximation of the full quantum rabi model at large @xmath2 is the well known adiabatic approximation @xcite .
the hamiltonian to zeroth order for fixed parity reads h_0=a^a + g(a^+a ) .
[ hnull ] the characteristic rabi physics in the dsc can already be extracted by use of the eigenbasis @xmath85 of this simple hamiltonian .
any more sophisticated approximation to the eigenbasis improves the quality only marginally and does not reveal new physical behavior . for convenience
we restrict ourselves to the positive parity subspace .
the spectrum in @xmath86 is given by the set of zeros @xmath87 of the function @xcite , g_+(x)=_m=0^k_m(x ) ^m , [ sol ] as @xmath88 .
the functional form of @xmath89 reads g_+(x ) = g_+^0(x ) + _ m=1^ [ gg0 ] with @xmath90 and the @xmath91 vanish for @xmath42 .
@xmath92 is of slow variation on the scale given by @xmath47 , therefore the location of the zeros of @xmath89 is determined by the poles at @xmath93 .
this leads to an almost equidistant distribution of the @xmath87 for small @xmath94 : the @xmath33-th root @xmath87 lies on the left or on the right of @xmath95 according to the sign of @xmath96 . because the @xmath96 grow with @xmath97 , we conclude that @xmath98 for small @xmath94 .
the dynamics of the model _ for fixed parity _ is encoded in this smooth distribution of eigenvalues together with the fidelity of the adiabatic basis .
there is a natural association of the shifted oscillator basis @xmath99 to the true eigenbasis @xmath100 that can be inferred from the representation @xcite , [ eqrepbarg ] _
m=0^k_m(x_n ) m;g .
as the quantum rabi model approaches the adiabatic limit , the spectrum converges to the spectrum of the shifted oscillator , i.e. x_n n g / this limit is encoded in the representation of the rabi eigenstates by the divergence of ( x_n - n)^-1 g/. this factor selects the @xmath33th shifted oscillator in the sum in eq . for sufficiently large @xmath101 .
we conclude that the rabi eigenbasis in a fixed parity subspace is close to diagonal in the basis of shifted harmonic oscillators .
this behavior is illustrated in fig .
[ fig4](a ) where we plot the contributions from the projections of the eigenstates on the shifted oscillators @xmath102 on a logarithmic scale .
the resulting matrix shows a rapid decrease of the off - diagonal elements .
this reduction is already present for small values of @xmath33 and @xmath103 and becomes more pronounced for higher values .
( a ) and fock states @xmath104 ( b ) for @xmath82 and @xmath81 .
this is for positive parity , i.e. @xmath105 is an eigenstate of @xmath106 .
the arrows in panel ( b ) show how to determine the boundaries for the `` bouncing photon number wave packets '' that are visible in fig .
, scaledwidth=46.0% ] in fig .
[ fig4](b ) , by contrast , we plot the representation of the eigenstates in terms of fock states .
the figure displays a clear parabolic shape .
this shape can be completely understood within the adiabatic limit .
considering the representation of the shifted oscillator states in terms of fock states , a perfectly symmetric parabola is seen .
this is due to the fact that the fock basis is transformed into the shifted oscillator basis by the operator @xmath107 and the group property of @xmath108 , @xmath109 .
because the matrix element @xmath110 reads @xmath111 , where @xmath112 denotes a laguerre polynomial , it follows @xmath113 , furthermore , @xmath114 and therefore @xmath115 .
the finite value of @xmath81 then leads to the slight asymmetry visible in the parabola of fig .
[ fig4](b ) .
surprisingly , also fig .
[ fig4](a ) shows an almost symmetric shape which can not be explained by the preceding argument . for @xmath82 ( a ) and @xmath83 ( b ) .
@xmath52 is evaluated as sum over the photon distribution of fig .
[ fig3](a ) and ( c ) : @xmath116 where @xmath117 .
( i ) solid lines depict numerically exact results calculated with this expression , ( ii ) dashed lines are calculated with @xmath118 replaced by @xmath119 and ( iii ) dotted lines are calculated with the additional replacement of the spectrum by its first order corrected perturbation theory value used in eq . . for @xmath82 ,
the adiabatic approximation allows a good description of the time evolution only if the exact spectrum is employed as seen in panel ( a ) . for @xmath83 ,
the adiabatic approximation is highly precise also if the approximated spectrum is used .
, scaledwidth=44.0% ] the effect of bouncing photon number wave packets mentioned earlier and shown in fig . [ fig3 ] can be described solely by the presence of the parabola in fig .
[ fig4](b ) . to illustrate the determination of the boundaries for the bouncing phenomenon in fig .
[ fig3](b ) , we have added the arrows in fig . [
fig4](b ) .
the upwards pointing arrow indicates the initial state @xmath120 .
the horizontal arrows then lead to the minimum and maximum photon numbers contained in the eigenstates . in fig .
[ fig5 ] , we present results for the time evolution of the quantum rabi model calculated using the adiabatic approximation . fig .
[ fig5](b ) shows that for @xmath121 the adiabatic approximation gives a detailed picture of the full time dependence .
[ fig5](a ) by contrast shows that for moderate values of the coupling , here @xmath122 , the full adiabatic approximation , i.e. approximated spectrum and approximated basis , completely misses a physical description of the dynamics .
this insufficiency of the full adiabatic approximation is mainly due to deviations in the spectrum which yield wrong phases .
by contrast , the approximated basis combined with the exact spectrum leads to a qualitatively correct description of the dynamics . to further illustrate the point that the adiabatic basis contains already all of the physics even far away from the limit @xmath123 , while the adiabatic spectrum leads to unphysical results , we define a distance of the state @xmath119 of the approximate basis from the true eigenstate @xmath118 by [ distance ] d_n(g,)= 1- ^2 .
a distance of the first order corrected eigen energy from the true eigen energy is given by the difference [ distance2 ] d^e_n(g,)= n - g+ ( -1)^n n , g n ,- g - e_n .
these functions are shown in fig .
[ fig6 ] for @xmath124 .
while their magnitude can not be compared directly , their dependence on @xmath2 and @xmath60 allows to conclude the following .
the slope of the contour lines is much steeper in case of the basis ( fig .
[ fig6](a ) ) compared to the spectrum ( fig .
[ fig6](b ) ) .
this means that the quality of the approximate basis increases much faster with @xmath101 than that of the first order spectrum .
this is surprising , as the basis is only a zeroth order approximation while the spectrum is approximated to first order .
it is possible to improve the adiabatic basis in a systematic fashion @xcite or use the rwa on top of the adiabatic approximation @xcite .
there are other approaches which reduce the approximative diagonalization of @xmath20 in @xmath26 to diagonalization of finite matrices @xcite .
all these techniques are equivalent to a break - up of @xmath26 into a set of invariant subspaces with finite dimension .
this means that a continuous symmetry is superimposed on the quantum rabi hamiltonian , leading to an effective model of jaynes - cummings type @xcite . in the quantum rabi model ,
such an additional ( hidden ) symmetry is not even present in an approximate sense , because otherwise fig .
[ fig4](a ) would display a block - diagonal pattern .
instead , the off - diagonal components of the true eigenstates in terms of shifted oscillator states are distributed rather smoothly on both sides of the central diagonal being at the same time smaller by at least two orders of magnitude .
it can be concluded that no self - consistent reduction to finite - dimensional invariant subspaces @xcite is possible which improves the adiabatic basis without creating a strong ( and unphysical ) symmetry .
most of the physics in the dsc regime ( in fact , already for @xmath125 ) is captured by the simple adiabatic basis .
of the shifted oscillator ground state from the ground state of the quantum rabi model in the positive parity subspace is shown in panel ( a ) .
the corresponding difference of the eigen energies @xmath126 in panel ( b ) .
the contour lines in the plots are drawn at values of @xmath127 and @xmath128 . comparing
the slope of the contour lines between panel ( a ) and ( b ) shows that the quality of the basis increases faster than the quality of the spectrum .
, scaledwidth=46.0% ] besides these physical considerations , we point out the technical fact that the spectrum is much easier to compute than the basis .
combining the exact spectrum with the simple basis allows us to predict correctly dynamical quantities of the qrm in the blue sketched parameter region of fig .
[ fig6](a ) , as demonstrated in fig .
[ fig5 ] .
the above discussion in the dsc regime led us to find the shifted basis as a good approximation to the real one .
the feature can be further confirmed by analyzing the wigner functions associated to each eigenstate in both parity chains , @xmath129 .
[ fig7 ] shows the wigner functions corresponding to the first eigenstates in the parity chain @xmath130 .
we see that these functions resemble the fock states @xmath50 , @xmath131 , @xmath132 , and @xmath133 , though shifted from the center , as expected from the shifted oscillator basis . notice the appearance of some additional interference pattern coming from other fock states which contribute to the whole solution , see fig .
[ fig4](a ) .
it is noteworthy to mention that exactly the same behavior can be observed in the other parity chain @xmath134 .
both chains behave very similar regarding the form of the eigenvectors , the ground state in @xmath135 looks almost the same as the first exited state in @xmath86 . , and
considering the dsc regime .
( a ) grouns state , ( b ) first excited , ( c ) second excited , and ( d ) third excited states . we can see that they resemble the wigner functions of fock states @xmath50 , @xmath131 , @xmath132 , @xmath133 , respectively , but displaced from the center as expected from the shifted oscillator basis.,scaledwidth=52.0% ] fig .
[ fig6](a ) shows that the ground state of @xmath106 at strong coupling is well approximated by @xmath136 .
if the system is prepared in this state , it will only weakly depend on time . using the transformation @xmath34 from eq .
, the state @xmath137 is mapped to the following states in the original basis with fixed parity @xmath27 : cc |c_+= & e^-((a^)0 |e-(a^)0 |g ) + [ cat - states ] these schrdinger cat - like states contain qubit - field entanglement @xcite , where the terms @xmath138 and @xmath139 are ( anti-)symmetric superpositions of the semiclassical states @xmath140 and @xmath141 @xcite . indeed , as the average photon number @xmath142 in @xmath143 is @xmath144 , an experimentally realizable value of @xmath145 corresponds to a dsc value @xmath146 .
these states behave almost like eigenstates regarding expectation values of observables , e.g. the revival probability @xmath147 . as seen in fig .
[ fig8 ] , @xmath148 remains almost constant for @xmath149 and exhibits only small high - frequency variations , but not the collapse - and - revival behavior with frequency @xmath47 as the fock states in fig .
[ fig1 ] .
[ fig8](b ) shows that , for @xmath149 , the oscillations with @xmath47 are completely gone and all fluctuations occur on time scales much shorter than the oscillator period @xmath150 .
averaged over such a short time , @xmath147 remains constant and behaves effectively a conserved quantity although the corresponding state is not an eigenstate .
the phase variable of the protected coherent states depends directly on the coupling @xmath54 and confines the effect to the dsc regime if the coherent states are required to contain more than two photons on average . of coherent states
@xmath151 for different values of @xmath152 and @xmath81 . if @xmath153 , the time evolution of the coherent state remains almost constant , @xmath154 for all times .
this is to be compared with the time evolution depicted in fig .
[ fig1](b ) of the initial fock state @xmath155 that yields the same photon number @xmath156 as @xmath157 .
, scaledwidth=46.0% ]
we have studied the dynamical properties of the quantum rabi model . by using the recently developed analytical solutions @xcite
, we obtained qualitative features of the spectrum and the eigenstates in two parts of the qrm : a lower and a higher coupling region , respectively .
it turns out that it is mandatory to pay attention to the sole and important @xmath12-symmetry of the qrm , which leads to the separation of the hilbert space into two invariant subspaces , the parity chains @xcite . within each chain , the dynamics appears to be rather simple . especially in the deep strong coupling regime , @xmath158 , we find a quite regular periodic behavior of the photon number distribution which can be traced back to the almost equidistant separation of energy levels .
this is no longer true if the initial state has no definite parity . under such circumstances , e.g.
if the initial state is a product of a coherent state and a qubit eigenstate , the time development may be more complicated , as both chains interfere @xcite .
we have identified a simple observable , the average photon generation from the vacuum at positive parity , to discern the convenient lower - coupling and higher - coupling regions .
this quantity shows a remarkable sharp peak just for @xmath159 , which demonstrates in a direct way the resonant enhancement of coupling between qubit and cavity mode for small @xmath2 , which was the original motivation to introduce the rwa in the first place @xcite .
this determines the lower - coupling region . for stronger couplings ,
the resonance gets broader until the peak vanishes around @xmath160 , establishing the higher - coupling region . here , the average generated photon number becomes larger than the jc limit of 1/2 due to the counter - rotating terms in ( [ hamr ] ) , which break the conservation law of the jcm . as early as for @xmath161 ,
the characteristic features of the deep strong coupling regime begin to manifest .
the dynamics for fixed parity are dominated by the adiabatic basis and the almost equidistant spectrum .
interestingly , the nontrivial effects , which separate the qrm in this regime from the simple adiabatic limit , can be incorporated by using the exact spectrum together with the adiabatic basis .
in fact , the exact eigenstates are very close to their adiabatic approximants whereas deviations in the eigenvalues lead to phase differences which become apparent after longer times ( fig . [ fig5 ] ) .
the fidelity of the adiabatic basis is even more visible in the wigner representation , which demonstrates the similarity of parity eigenstates with fock states in the deep strong coupling regime ( fig .
[ fig7 ] ) .
finally , the dsc allows for a special class of states which are not exact eigenstates but lead to expectation values fluctuating with small amplitudes on very short time scales . in this sense
, they form stationary schrdinger cat - like states , unaffected by the system interaction .
we acknowledge funding from the deutsche forschungsgemeinschaft through trr 80 , spanish micinn juan de la cierva , fis2009 - 12773-c02 - 01 , basque government it472 - 10 , upv / ehu ufi 11/55 , solid , ccqed , and promisce european projects .
h. paik , d. i. schuster , l. s. bishop , g. kirchmair , g. catelani , a. p. sears , b. r. johnson , m. j. reagor , l. frunzio , l. i. glazman , s. m. girvin , m. h. devoret , and r. j. schoelkopf , phys .
lett . * 107 * , 240501 ( 2011 ) .
m. mariantoni , h. wang , r. c. bialczak , m. lenander , e. lucero , m. neely , a. d. oconnell , d. sank , m. weides , j. wenner , t. yamamoto , y. yin , j. zhao , j. m. martinis , and a. n. cleland , nature phys . * 7 * , 287 ( 2011 ) . | we study the quantum rabi model within the framework of the analytical solution developed in phys .
rev . lett . * 107 * , 100401 ( 2011 ) .
in particular , through time - dependent correlation functions , we give a quantitative criterion for classifying two regions of the quantum rabi model , involving the jaynes - cummings , the ultrastrong , and deep strong coupling regimes .
in addition , we find a stationary qubit - field entangled basis that governs the whole dynamics as the coupling strength overcomes the mode frequency . |
we start with the equations of motion for a particle of unit mass on a three - dimensional sphere @xmath0 or in a lobachevsky space @xmath1 ( pseudosphere ) .
the sphere @xmath0 ( pseudosphere @xmath1 ) can be parameterized using the cartesian ( redundant ) coordinates of the four - dimensional euclidean space @xmath4 ( the minkovsky space @xmath5 ) with the constraint @xmath6 where @xmath7 ( @xmath8 ) is the corresponding metrics .
hereinafter , an upper sign in @xmath9 is used for the sphere and a lower sign is used for the pseudosphere .
the metrics in @xmath4 ( @xmath5 ) generates a metrics in the sphere @xmath0 ( the lobachevsky metrics in the pseudosphere @xmath1 ) . in terms of the cartesian
coordinates , the lagrangian for the particle s motion in the field of the potential @xmath10 is @xmath11 with the constraint [ e1 ] . using the hamiltonian formalism for systems with constraints ( arnold et al .
1993 ) , we get @xmath12 then the canonical equations of motion are @xmath13 , @xmath14 . * two - body problem on @xmath0 ( @xmath1 ) .
* consider the two - body problem on the curved spaces @xmath0 ( @xmath1 ) , where bodies are assumed to be point masses .
let these masses move in the field of some potential @xmath15 ( @xmath16 are the coordinates of bodies on @xmath0 ( @xmath1 ) ) .
in this particular case the potential energy @xmath17 depends on the distance between two points ( this distance is measured along a geodesic ) . in our case
, there is a center of mass frame of reference such that the two - body problem can be reduced to the problem of the particle s motion in the field of a fixed attracting center ( i.e. to the kepler problem in the case of the newtonian interaction ) .
the analogue of the kepler problem is superintegrable on @xmath0 ( @xmath1 ) ( see kozlov 1994 , borisov et al .
1999 and killing 1885 ) .
the generalization of all the kepler laws to spaces of constant curvature is given by kozlov ( 1994 ) .
but in curved spaces the center - of - mass frame of reference does not exist and therefore if the interaction between two bodies is newtonian - like , the two - body problem is not integrable on @xmath0 and @xmath1 . in terms of the cartesian canonical variables
@xmath18 and @xmath19 , the hamiltonian of the two - body system is ( the index @xmath20 denotes the number of the mass @xmath21 ) @xmath22
@xmath23 * invariant manifolds . *
the three dimensional two - body problem is rather complicated .
therefore by analogy with the planar case , we will examine in detail the motion on the invariant submanifolds of the system .
the behavior of the system on invariant submanifolds allows us to make conclusions about some properties of the system ( nonintegrability , stochasticity ) in the whole phase space .
nevertheless the three - dimensional problem has not been investigated yet .
the invariant manifolds of the n - body problem in @xmath24 are planes and , similarly , if a space is curved , they are spheres @xmath2 ( pseudospheres @xmath3 ) .
there is a three - parameter family of such manifolds at any point of @xmath0 ( @xmath1 ) ( see borisov et al .
1999 ) . * restricted two - body problem . * in the euclidean space @xmath24 a passage to the limit is possible in the two - body problem as the mass of attracting center goes to infinity while the interaction energy remains finite .
there is an inertial frame of reference with origin at the heavier particle , therefore if the potential is newtonian , the restricted problem is the kepler problem . consider a similar passage to the limit on @xmath0 ( @xmath1 ) . in this case
the attracting center moves along a large circle of the sphere ( along a geodesic ) .
the second particle ( point mass ) moves in the field of attracting center and does not affect the motion of the attracting center . if the origin is at the first particle , the second particle moves in the field of fixed center and gyroscopic forces .
the lagrangian of the system is @xmath25 where @xmath26 is the angular velocity matrix of the frame of reference . here , @xmath27
( i.e. it is a skew - symmetric matrix ) in the case of @xmath0 and @xmath28 in the case of @xmath1 .
let a particle move in the field of newtonian - like potential on a sphere @xmath0 ( pseudosphere @xmath1 ) .
using spherical coordinates @xmath29 ,
@xmath30 , @xmath31 and @xmath32 , we can write the hamiltonian as @xmath33 separating variables in ( [ e5 ] ) gives @xmath34 @xmath35 where @xmath36 is the projection of the three - dimensional angular momentum vector @xmath37 ( here @xmath38 ) onto the axis @xmath39 , @xmath40 is the squared momentum @xmath41 , @xmath42 is the energy constant .
it is easy to see that the vector @xmath43 is an integral of motion .
( in the case of the lobachevsky plane , all the trigonometric functions of @xmath44 should be replaced with hyperbolic functions . )
let us see how the domain of possible motions on @xmath0 ( @xmath1 ) ( hereinafter dpm ) depends on the energy constant @xmath42 and the moment constant @xmath45 .
we put @xmath46 ( @xmath46 ) in ( [ e7 ] ) .
if @xmath42 and @xmath45 are fixed then the dpm are defined as follows @xmath47 thus to construct a bifurcation diagram we should consider the quadratic equation @xmath48 where @xmath49 . the bifurcation set ( i.e. the locus of @xmath50 at which the domain of possible motion changes topologically ) consists of the curves ( see fig .
1 ) @xmath51 if both roots @xmath52 and @xmath53 of ( [ e9 ] ) are complex ( domain i in fig .
1 ) the motion is impossible .
if both roots are real and positive ( domain ii ) , the possible values of @xmath54 are given by @xmath55 .
this implies that a particle moves in the ring @xmath56 , with @xmath57 for @xmath2 . if the lower root ( @xmath52 ) is negative ( domain iii ) ,
then @xmath58 for the real motion on the lobachevsky plane and @xmath59 for the motion on a sphere ( since @xmath54 is negative if @xmath60 ) .
it means that on @xmath3 a body moves exterior to the circle @xmath61 and if the space is @xmath2 , a particle moves in the ring @xmath62 , where @xmath63 , @xmath64 .
note that motions on a sphere are bounded because of compactness @xmath0 .
orbiting time is always finite .
note also that the curved kepler problems are trajectory isomorphic to their plane analogues , as was shown by serre ( see appell 1891 ) .
define the action variables in terms of spherical variables @xmath65 where the integral is taken over the whole cycle of the period of motion .
since @xmath66 we have from ( [ e10 ] ) @xmath67 .
the kinetic energy in terms of spherical coordinates on @xmath0 is @xmath68 , and on the invariant sphere @xmath2 where the particle moves , we have @xmath69 ; here @xmath70 is a true anomaly ( i.e. usual polar angle ) . equating these two expressions
we get @xmath71 .
the coordinates @xmath70 and @xmath72 change by @xmath73 per one revolution of the orbit . therefore , after integrating we have @xmath74 to compute the third integral of ( [ e10 ] ) , put @xmath46 ( @xmath46 ) and use the equation for the orbit @xmath75 ( see kozlov 1994 ; killing 1885 ) .
@xmath76 where @xmath77 is a parameter of the orbit , @xmath78 is the eccentricity .
this implies @xmath79 where @xmath80 , @xmath81 .
we get after integration @xmath82 for @xmath0 , and @xmath83 for @xmath1 . since @xmath84 , @xmath85 , we get with ( [ e11 ] ) the explicit expression of the hamiltonian @xmath86 similar to the euclidean space @xmath24 , the hamiltonian depends only on the sum @xmath87 , i.e. the frequencies @xmath88 , @xmath89 , corresponding to the variables @xmath90 , @xmath91 , @xmath92 , coincide .
this is the case of the complete degeneracy , because all the three - dimensional liouville arnold tori foliate into one - dimensional tori i.e. circles .
note that unlike the hamiltonian in the space @xmath24 ( see markeev 1990 ) , expression [ e18 - 3.6 ] has additional terms , which are proportional to @xmath93 .
define new variables @xmath94 , @xmath95 , @xmath96 , @xmath97 , @xmath98 , @xmath99 ( analogues of the poincare variables ) @xmath100 in terms of these variables , the hamiltonian is @xmath101 with [ e18 - 3.8 ] and [ e18 - 3.7 ] , we have @xmath102 equation [ e18 - 3.8 ] implies all the delaunay variables except @xmath97 are integrals of motion .
the angle @xmath97 is an analogue of the mean anomaly @xmath103 and changes uniformly with the time @xmath104 . here
@xmath105 is the time , when the particle passes the pericentre , @xmath106 is the period of orbit revolution , which depends only on the energy constant @xmath42 ( see killing 1885 ; kozlov 1994 ) : @xmath107 in terms of the angular length of the orbit s major axis @xmath20 , the energy constant @xmath108 @xmath109 .
the delaunay variables can be expressed in terms of orbit parameters like in the planar case as it shown by markeev ( 1990 ) and demin et al .
choose the angular constants so that if we make gnomonic projection @xmath98 , @xmath99 are the images of the pericentre parameter and the longitude of the ascending node .
denote them by @xmath110 and @xmath111 .
let @xmath112 be the analogue of orbit inclination .
this value is equal to the angle between the axis @xmath39 and the vector @xmath113 .
express the variables @xmath94 , @xmath95 , @xmath96 , @xmath97 , @xmath98 , @xmath99 in terms of the elements of the orbit @xmath114 , @xmath115 , @xmath112 , @xmath105 , @xmath110 , @xmath111 : @xmath116 in the case of the lobachevsky space @xmath117 .
the observation of mercury s perihelion shift is one of the experiments that proves the general relativity theory ( grt ) ( see eddington 1963 ) .
this shift arises as a result of curving of a space near a gravitating body .
let us prove that in newtonian mechanics , a keplerian orbit also precesses in a curved space .
although the laws of precession in these theories are different .
we will take the restricted two - body problem as a model problem .
this problem is not integrable but if the velocity of the heavier particle is low we can analyze the problem using the perturbation theory . here
we do not mean to give a new physical justification of the perihelion shift , already given in grt and accepted as classical .
we just point out that some phenomena of the practical celestial mechanics admit another interpretations ( together with the planet nonsphericity , atmosphere refraction and so on ) .
the addition of curvature to the classical newtonian mechanics is an example of such interpretations .
consider the restricted two - body problem on @xmath2 @xmath118 . as usual , the sphere ( pseudosphere ) is assumed to be embedded in @xmath24 @xmath119 : @xmath120 .
let an attracting center move along the geodesic on @xmath121 plane , and we choose the ( moving ) frame of reference that the attracting center is at the north pole of the sphere ( pseudosphere ) @xmath122 .
the lagrangian of point mass ( particle of unit mass ) is @xmath123 here @xmath124 is the angular velocity matrix of the frame of reference , @xmath125 let us assume that the typical size of the domain of motion of the point mass is small in comparison with the radius of curvature @xmath126
. then we can analyze the problem using perturbation theory .
suppose also that the angular velocity of the attracting center s motion is small in comparison with the rotation frequency of the point mass moving along the corresponding keplerian orbit .
take the length @xmath127 @xmath128 and azimuth angle @xmath129 as the coordinates on a sphere ( pseudosphere ) and transform [ e18 - 4.1 ] as @xmath130 here @xmath131 , and @xmath132 is the linear velocity of the motion of the noninertial frame of reference .
if @xmath133 , the problem is reduced to the planar kepler problem .
the terms in [ e18 - 4.2 ] , which are linear with respect to the velocity , have the order @xmath93 and can not be omitted . to study the evolution of the orbit s shape in the unperturbed kepler problem we express the equations [ e18 - 4.2 ] in terms of @xmath114 , @xmath110 , @xmath115 , @xmath129 . here ,
@xmath115 is the eccentricity , @xmath110 is the longitude of the orbit s pericenter , @xmath129 is the azimuth angle , @xmath114 is the orbit s parameter , associated with the energy @xmath42 of the unperturbed kepler problem by following @xmath134 the new variables are expressed in terms of coordinates and velocities ( hereinafter @xmath135 ) @xmath136 hereinafter , for the sake of simplicity , we do nt substitute the expression for @xmath54 in terms of @xmath114 , @xmath115 , @xmath110 , @xmath129 .
the poisson brackets for @xmath114 , @xmath115 , @xmath110 , @xmath129 are @xmath137 and the hamiltonian is @xmath138 expressions [ e18 - 4.5 ] and [ e18 - 4.6 ] imply that when @xmath139 , the variables @xmath114 , @xmath115 , @xmath110 are slow , @xmath129 is fast . to define the secular change of the orbit s parameters , when @xmath140 , we neglect the terms with order higher than @xmath93 and average the equations of motion over the period of unperturbed motion . averaging over the period is equal to averaging over @xmath129 with a weight function @xmath141 . here
the weight function @xmath142 is defined by the derivative @xmath143 from [ e18 - 4.4 ] as @xmath144 .
so , we have the system : @xmath145 the equations [ e18 - 4.7 ] has the integral @xmath146 this integral implies that there is no secular change of the energy of the unperturbed system [ e18 - 4.3 ] to this approximation ( laplace s theorem ) .
the phase portrait of [ e18 - 4.7 ] on the surface of the integral [ stars0 ] depends on the parameter @xmath147 .
figure 2 shows the projection of the trajectories onto @xmath148 plane for different values of @xmath149 .
the parameter @xmath149 describes the ratio of the velocity of the attracting center to the characteristic velocity of the particle along the keplerian orbit .
the equation [ e18 - 4.7 ] implies that the curvature sign determines the direction of the motion along the trajectory but not the shape of the trajectory .
the value of curvature defines the velocity of motion along the trajectory .
it is clear from the figures that the velocity of perihelion shift depends not only on the eccentricity of the orbit but also on the orientation of the orbit with respect to the direction of motion of the attracting center . if @xmath149 is small , there exist two stable periodic orbits with the non - zero eccentricity .
the main axis of the orbits is perpendicular to the direction of the attracting center s motion . at pericenter ,
the direction of the point mass motion along one of the orbit coincides with the direction of the attracting center s motion @xmath150 .
when the particle moves along another orbit @xmath151 at the pericenter , its direction is opposite to the direction of the attracting center s motion .
when @xmath149 increases the orbit with @xmath152 becomes unstable , and if @xmath149 is sufficiently large , the stable orbit with @xmath153 disappears.=1 the projections of the trajectories of the non - averaged system onto @xmath148 plane are also shown in the figures .
here , we can see small oscillations ( for the variables @xmath110 , @xmath115 ) near the trajectories of the averaged system ( see fig .
remind that standard explanation of the perihelion shift is based on the schwarzschield solution ( eddington 1963 ) and implies that the shift velocity does not depend on the orientation of the orbit @xmath110 .
this is not the case in our problem statement .
there always exist orbits such that only their eccentricities change their values but their pericenters have almost no shift .
moreover , there are fixed points of the system ( [ e18 - 4.7 ] ) , corresponding to the periodic orbits which do not change their form and orientation .
note that the schwarzschield - like metrics can be constructed if boundary conditions ( at infinity ) correspond to the space of constant curvature as it shown by chernikov ( 1992 ) . and also the restricted two - body problem can be generalized for such metrics .
it is clear that the velocity of the perihelion shift depends on both @xmath110 and @xmath115 .
consider the restricted two - body problem on the sphere @xmath2 in the general case without assumption that the curvature is small .
we define the new variables @xmath43 , @xmath154 using the map @xmath155 , by @xmath156 here , the canonical poisson brackets @xmath157 are transformed to the lie poisson brackets corresponding to @xmath158 algebra .
the equations of motion are @xmath159 where the hamiltonian is @xmath160 these equations ( see borisov et al .
2001 ) have two integrals of motion : the area integral @xmath161 and the geometric integral @xmath162 . in our case
note that this system describes the motion of spherical top in the potential @xmath164 and in the field of gyroscopic forces.=-1 for @xmath165 this system is ( super)integrable ( the kepler problem on @xmath2 ) , and in this case simple geometric interpretation of the motion exists : _ the variable @xmath166 and the projection of the trajectory onto the plane @xmath167 is a circle shifted from the origin _
( incidentally , hamilton noticed a similar thing , in the two - dimensional kepler problem ) .
indeed in the consequence of @xmath168 we have @xmath169 and using the equation ( [ e12 ] ) we obtain @xmath170 @xmath171 where @xmath70 is a longitude on the sphere @xmath172 . eliminating @xmath173 from the last equation we can get @xmath174 for @xmath175 , according to the liouville arnold theorem , an additional integral must exist the system to be completely integrable .
we will show soon that in the general case , the additional integral does not exist ( see also borisov et al .
1999 , chernovan et al . 1999 ) .
we construct the poincare map to study the problem numerically . to construct it we use the analogy with the motion of rigid body and choose the andoyer canonical variables @xmath176 according to borisov et al .
( 2001 ) as @xmath177 g=(\boldsymbol{m},\boldsymbol{m}),\quad g=\arccos\left(\frac{-\gamma_3 } { \sqrt{1-m^2_3/(\boldsymbol{m},\boldsymbol{m})}}\right ) .
\end{gathered}\ ] ] then the hamiltonian as a function of these coordinates is -0.5mm@xmath178 -1.5mmthe equations of motion are canonical : 0.5mm@xmath179 -0.5 mm let us fix the energy level @xmath180 and define the poincare section by the relation @xmath181 . on this two - dimensional surface
we choose the variables @xmath182 and @xmath97 as the coordinates of the poincare map ( similarly to borisov et al .
the domain of definition of the variables is compact : @xmath183 , @xmath184 and the flow ( [ e18 - 5.2 ] ) defines corresponding poincare map . by direct substitution into [ e18 - 5.2 ]
, it is easily proved that @xmath185 so , each trajectory @xmath186 with the initial conditions @xmath187 @xmath188 corresponds to a similar trajectory @xmath189 with initial conditions @xmath190 , and each point @xmath191 of @xmath186 corresponds to a point @xmath192 of @xmath189 .
this means the poincare map ( for the chosen section @xmath181 is central symmetric .
the phase portraits for the different values of the energy @xmath42 and parameter @xmath193 are shown in fig .
it is easy to see in the figures that the stochastic layer increases as the energy @xmath42 .
this proves that the two - body problem in general case is not integrable .
the fixed points in fig . 4 correspond to periodic trajectories of the particle , which play an impotent role in the qualitative analysis of the system .
after the publication of the book by borisov et al .
( 1999 ) and paper by chernovan et al .
( 1999 ) , at our suggestion , s.l.ziglin could prove that the additional meromorphic integral does not exist for the potentials that are the analogous to the newtonian and hooke interaction for any value of the parameters ( see ziglin 2001 and ziglin 2003 ) .
with the equation of motion [ e18 - 5.1_12 ] , the integral of energy of the restricted problem [ e18 - 5.1_34 ] can be written as @xmath194 where @xmath44 , @xmath129 are the spherical coordinates on @xmath2 ( it means that @xmath44 is a latitude and @xmath129 is a longitude ) . if the energy @xmath42 is fixed ( therefore , @xmath195 is also constant ) the domain of motion on @xmath2 is defined by @xmath196 by analogy with the classical restricted three - body problem ( arnold et al .
1993 ) , we will call this domain _ the hill domain of the restricted two - body problem on a sphere_. when the parameters of the system are fixed , the shape of the hill domains is defined by the singularities and critical points of the effective potential @xmath197 .
the singularities of @xmath198 are at the poles of a sphere . and , since @xmath199 and @xmath200{}\infty$ ] , hill domain is always not empty , because near @xmath201 the inequality ( [ stars1 ] ) holds .
each critical point of the function @xmath197 corresponds to the equilibrium position of the particle ( here , the frame of reference rotates with the attracting center ) .
this equilibrium position is usually called _ a relative equilibrium_. note that ( we will prove this below ) in the fixed frame of reference the attracting center moves along the large circle and the particle moves along the another circle parallel to the large circle , with it being at the same meridian with the attracting center .
find the location of critical points by solving the system @xmath202 we obtain the following results : 1@xmath203 _ if @xmath204 , there are four critical points on the meridians @xmath205 and @xmath206 ( two points on each meridian ) .
their latitudes @xmath207 are defined by the equation _
@xmath208 2@xmath203 _ if @xmath209 , the function @xmath197 has no critical points at all_. it is easy to see , that in the case @xmath210 the critical points @xmath211 , @xmath212 are the saddle points of the function @xmath198 , and the points @xmath213 , @xmath214 are the strict maxima . hill domains for both cases ( 1@xmath203 and 2@xmath203 ) are shown in fig . [ x1 ] , [ x2 ] .
it is clear from fig .
[ x1 ] that fixed points are in the semisphere opposite to the attracting center .
we will use linear approximation to investigate the stability of the obtained fixed points ( relative equilibria ) .
let the point @xmath215 , @xmath216 , be the corresponding fixed point , where @xmath217 , @xmath218 . according to [ stars4 ] , it is very convenient to parameterize @xmath219 by the latitude of the fixed point : @xmath220 since @xmath199 for attracting center , @xmath221 , moreover for @xmath222 @xmath223 and these points correspond to the maximum of the effective potential @xmath198 and @xmath224 to the saddle point . here
, @xmath225 denotes the value of @xmath44 , for which @xmath219 [ stars2 ] reaches the maximum @xmath226 .
let us introduce canonical impulses @xmath227
, @xmath228 corresponding to the spherical angles . in the fixed points their values are @xmath229 , @xmath230 .
expand the hamiltonian [ e18 - 5.1_34 ] in the vicinity of the fixed point up to the second power , using the following canonical variables : @xmath231 we obtain @xmath232 the eigenvalues of the corresponding linearized system are @xmath233 the study of the radical expressions in [ stars3 ] gives us the following results : for @xmath234 and @xmath235 , @xmath236 are always real and @xmath237 are real for @xmath234 and purely imaginary for @xmath235 .
this means that _ relative equilibria in the restricted two - body problem on a sphere are always unstable_. note that , according to the theorem of central manifold , the existence of two purely imaginary eigenvalues for the points @xmath238 and @xmath239 results in the existence of an unstable ( hyperbolic ) periodic solution near these points .
[ x3 ] shows these solutions for different values of energy .
this work was supported by rfbr ( 04 - 05 - 64367 and 05 - 01 - 01058 ) , crdf ( ru - m1 - 2583-mo-04 ) , intas ( 04 - 80 - 7297 ) and the program `` state support for leading scientific schools '' ( 136.2003.1 ) .
appell , p. : 1891
, ` sur les lois de forces centrales faisant dcrire leur point dapplication une conique quelles que soient les conditions initiales ' , _ american journal of mathematics _ , * 13 * , 153158 . ,
@xmath244 ) ( a part of the sphere is shown in the spherical coordinates ) .
the boundary of the hill domain ( see fig .
[ x1 ] for @xmath245 ) is shown for the critical value of the energy @xmath246 when other parameters @xmath247 , @xmath135 .
the values of the energy corresponding to shown orbits are @xmath42=0.0009 , 0.01 , 0.027 , 0.355 , 1.0 , 1.024 ] | we perform the bifurcation analysis of the kepler problem on @xmath0 and @xmath1 .
an analogue of the delaunay variables is introduced .
we investigate the motion of a point mass in the field of the newtonian center moving along a geodesic on @xmath2 and @xmath3 ( the restricted two - body problem ) .
when the curvature is small , the pericenter shift is computed using the perturbation theory .
we also present the results of the numerical analysis based on the analogy with the motion of rigid body .
keywords and phrases : kepler problem , bifurcation analysis , perihelion shift , delaunay variables , restricted problem . |
support by european community ( fp7 eu - india grant athena and erc starting grant no.203523 bismuth ) is gratefully acknowledged .
x.h . thanks claude ederer and chung - yuan ren for their helpful advices and comments .
thanks j.m .
perez - mato for useful discussions .
the calculations have been performed on the vienna scientific cluster ( vsc ) and , partially , in caspur supercomputing center in rome . | we present a study based on several advanced first - principles methods , of the recently synthesized pbnio@xmath0 [ j. am .
chem .
soc * 133 * , 16920 ( 2011 ) ] , a rhombohedral antiferromagnetic insulator which crystallizes in the highly distorted @xmath1 crystal structure .
we find this compound electrically polarized , with a very large electric polarization of @xmath2 100 @xmath3c/@xmath4 , thus even exceeding the polarization of well - known bifeo@xmath0 .
pbnio@xmath0 is a proper ferroelectric , with polarization driven by large pb - o polar displacements along the [ 111 ] direction .
contrarily to naive expectations , a definite ionic charge of 4 + for pb ion can not be assigned , and in fact the large pb 6@xmath5-o 2@xmath6 hybridization drives the ferroelectric distortion through a lone - pair mechanism similar to that of other pb- and bi - based multiferroics . _
introduction _ multiferroics are materials in which different ferroic orders such as ferromagnetism , ferroelectricity and/or ferroelasticity may coexist in one single phase @xcite . in the last few years , there has been a tremendous boom of interest in these materials , due to the potential applications in memory devices or in novel type of magnetic switching in magnetoelectric multiferroics , based on the cross - coupling between ferroelectric and magnetic channels @xcite .
furthermore , these materials offer a rich and fascinating playground for the complex physical mechanisms underlying the processes involved in the observed properties.@xcite it is obvious that the search and the prediction by material design of new multiferroics is of great importance for both fundamental physics and technological applications @xcite .
relatively few multiferroics have been identified so far .
@xcite without doubt , the most studied and well characterized multiferroic is bifeo@xmath0 @xcite .
it crystallizes in the polar space group _ r3c _ ( no . 161 ,
point group _ c_@xmath7@xmath8 ) , and is predicted to have a g - type antiferromagnetic ( afm ) alignment of the fe spins @xcite .
the _
r3c _ symmetry corresponds to the so - called linbo@xmath7-type structure , which can be viewed as a highly distorted double perovskite with rhombohedral symmetry .
the primitive unit cell contains two formula units ( 10-atoms ) , arising from counterrotations of neighboring o octahedra about the [ 111 ] axis ( see fig.[fig:1 ] ) .
the crystal symmetry allows the presence of a spontanous polarization along the [ 111 ] direction , which arise from the relative displacement of the bi sublattices with respect to the feo@xmath9 octahedra cages along [ 111 ] .
the origin of the large spontaneous polarization of bi@xmath10fe@xmath10o@xmath11@xmath7 , @xmath2 90 @xmath3c/@xmath4 @xcite , has been explained by first principles calculations within the density functional theory plus hubbard - u approach ( dft+u)@xcite and the `` modern theory of polarization '' ( mtp ) @xcite , in terms of the lone - pair electrons present at the bi sites which are ultimately responsible for the large displacement along the [ 111 ] direction @xcite . inspired by the recent reports of inaguma _ et al .
_ on the synthesis of a new antiferromagnetically ordered compound with linbo@xmath7-type structure , such as pbnio@xmath7@xcite , we explore here the possibility of multiferroic behavior in pbnio@xmath7 .
we first summarize the experimental findings .
inaguma and coworkers synthesized two high - pressure polymorphs of pbnio@xmath7 with a ( a ) perovskite - type structure and ( b ) the linbo@xmath7-type structure @xcite ; the latter ( hereafter called l - pbnio@xmath0 ) is thermodynamically more stable than the perovskite - type one at ambient pressure . with respect to the orthorhombic structure , in the acentric rhombohedral linbo@xmath7-type ( _ r3c _ ) structure ( fig .
[ fig:1 ] ) pb and o atoms are displaced against each other along the threefold [ 111 ] axis leading to a large distortion of the pbo@xmath12 and nio@xmath12 octahedra ( see fig.[fig:1 ] ) .
the pb atom is coordinated by six oxygens at 2.10 and 2.25 , while the ni o bond distance splits into two subgroups ( 2.07 and 2.11 ) . magnetic susceptibility and
resistivity measurements indicate that l - pbnio@xmath7 undergoes an afm transition at @xmath13=205@xmath14k , and exhibits semiconducting behavior .
the afm ordering in the acentric crystal structure suggests possible multiferroic behavior . in the following
we show that l - pbnio@xmath0 is a _ new _ room - temperature _ multiferroic _ , with an _ exceptionally large _ polarization ( @xmath15 ) of about 100 @xmath3c/@xmath16 , which is the highest polarization ever predicted for any bulk material so far , and @xmath17 10% larger than that of bifeo@xmath0 @xcite .
( color online ) schematic view of the linbo@xmath7-type pbnio@xmath7 within space group _ r3c _ structure built up from two cubic perovskite unit cells , the black ( large ) , gray ( medium ) and red ( small ) spheres denotes pb , ni and o atoms , respectively .
with respect to the centrosymmetric ( _ r3c _ ) paraelectric phase ( see text ) the cations are displaced along the [ 111 ] axis relative to the anions ( the arrows represent the displacement vectors ) , and the oxygen octahedra slightly rotate with alternating sense around the [ 111 ] direction @xcite . the crystal structure was drawn using the program vesta@xcite . ]
_ computational details _ we performed dft based calculations using the vienna _ ab initio _ simulation package ( vasp ) @xcite based on the projector - augmented - wave method@xcite , withing the perdew , burke and ernzerhof ( pbe ) parameterization scheme@xcite for the generalized gradient approximation ( gga ) .
to overcome the deficiencies of standard exchange - correlation approximations for localized ni _ d _ states , we made use of three beyond - dft approaches : ( i ) dudarev s gga+u @xcite , using u=4.6 ev in accordance with constrained dft calculation of ref . ;
( ii ) the renowned heyd - scuseria - ernzerhof ( hse ) screened hybrid functional @xcite , which has been shown to give an excellent account of materials properties for magnetic multiferroics@xcite . ( iii ) the recently introduced variational pseudo self - interaction - correction method vpsic@xcite ( adopting the hse optimized structure ) , implemented within plane - wave basis set and ultrasoft pseudopotential scheme in the pwsic code .
the cutoff energy was set as 600 ev and a 8@xmath188@xmath188 monkhorst - pack grid of _ k _ points was used . in hse
, the fraction ( 1/4 ) of fock exchange was sampled using the twofold reduced * k*-point grid , to reduce the computational load .
the lattice parameters and atomic positions were relaxed ( at gga+u and hse level ) until the total energy changed by less than 10@xmath19@xmath20 ev per unit cell and the residual force was smaller than 0.01 ev / .
.structural parameters , magnetic moment and electronic band gap for afm - g configuration of l - pbnio@xmath7 within gga , gga+u and hse method , compared with the available experimental data .
_ a _ , _ c _ are the lattice parameters in the hexagonal setting , whereas _
x _ , _ y _ , and _ z _ are the internal atomic positions of pb and o. ni ions sit in the ( @xmath21 ) ( 0,0,0 ) positions . bond lengths of pb o and ni o are also reported .
@xmath22 ( @xmath14 ) is the ni o
ni angle .
m _ indicates the magnetic moment of ni . [ cols="<,^,^,^,^,^ " , ] [ tab : born ] |
since the quantum cascade laser ( qcl ) was introduced more than 20 years ago @xcite it has continuously been improved and redesigned to operate from the mid - infrared all the way down to the terahertz ( thz ) range .
compact laser sources at these wavelengths are valuable for spectroscopic applications @xcite but a major problem is that thz qcls do not operate at room temperature .
this can be overcome via difference frequency generation which recently has been demonstrated at powers in the milli - watt range @xcite .
direct thz - qcls still generate a lot of interest in the research community due to the promise of higher wall - plug efficiency and the prospects of miniaturization of cryo - coolers .
the main temperature degrading mechanism of the thz qcls is currently not fully understood and is still debated by the community @xcite .
this is a challenge to theory and there is a great need of realistic modeling tools that are able to treat all important quantum effects on the same footing .
a summary of different methods for simulating these structures is found in @xcite .
monte carlo simulations have shown that electron - electron interaction , beyond the meanfield or hartree approximation can influence the dynamics of thz qcls @xcite . in this work
we include and study the effects of a simple electron - electron scattering via the single plasmon pole approximation ( sppa ) @xcite . in this approximation
we capture both the static limit as well as dynamic effects .
this gives an energy dependent ( non - local in time ) interaction beyond the hartree - fock approximation .
this has been studied in a similar model with promising results @xcite , and with this work we want to adapt the idea into our model described in ref . @xcite .
in other methods based on non - equilibrium green s functions ( negf ) and applied to qcls , electron - electron scattering has previously been implemented in a low order @xmath0 approximation employing a purely static screening model @xcite .
to simulate the periodic structure of the qcl , we use a product space of plane waves and confined states .
the green s function evaluation provides us with the energy resolution of the states , labeled by state and @xmath1 index , at the cost of inverting the self - energy matrices at each @xmath1 and energy grid point .
a number of elastic scattering processes as well as longitudinal optical phonon scattering is treated in this formalism at a high computational cost . adding a many - body term fully dependent on the polarization would increase the complexity by one order , a formidable task in this model , leading us to consider simpler schemes as done in this work .
this paper is organized as follows : first the theory of the sppa is discussed and we show how it is implemented in the context of our formalism
. then the dispersion of the plasmons is discussed as well as the exchange shift . in the last part a test structure is used to evaluate the effect of the inclusion of the new scattering mechanism
. finally , we present the conclusions of our study .
in thz quantum cascade simulations , the relevant sources of scattering , beside electrons , are interface roughness , impurities from donor atoms , acoustic and optical phonons as well as alloy composition .
all these effects have already been implemented and details can be found in ref .
@xcite .
the model discussed here employs a non - equilibrium green s function formalism where the system observables are extracted from the retarded and lesser parts of the green s function in the formalism of @xcite .
these are solved using self - energies for the part of the hamiltonian that is non - diagonal in @xmath2 and solved through the dyson equation and the keldysh relation . in this section
we will describe how we formulate the self - energy expressions for the screened interaction in the single plasmon pole approximation ( sppa ) .
first we will discuss how the conduction band is treated as an effective band in order to apply the named approximation .
then we will describe how we formulate and also evaluate the coulomb matrix elements .
after this we acquire the lesser and retarded part of the screened interaction before moving on to the actual construction of the self - energies .
this is done in close correspondence to the treatment of phonons in ref .
@xcite in order to keep the length of this section reasonable . to this treatment
we will here add the necessary exchange part .
before we move on to the next section where we test the model , we will also describe how the plasmon dispersion is treated . in frequency space the retarded screened potential is given by its own dyson equation , given here as a matrix equation : @xmath3^{-1}}_{\epsilon(\mathbf{q},e)^{-1 } } \mathbf{v(q)}.\end{aligned}\ ] ] where the @xmath4 is the bare interaction as a function of the momentum wave vector , described in more detail below , and @xmath5 is the polarization as a function of momentum and energy @xmath6 .
if we now assume an isotropic screening medium and that the dielectric function @xmath7 is not level dependent , we can write the approximated retarded screened potential as @xmath8 where @xmath9 now holds the matrix structure .
we consider the dielectric function @xmath7 given by the lindhard formula @xcite which is retarded due to poles in the lower half of the complex plane . while we assume an effective band for the screening , all other physical mechanisms are treated with full state dependence . in this work
we will assume a plasmon distribution in equilibrium with respect to the electron temperature .
this approximation will ensure that energy is not dissipated from the distribution of non - equilibrium electrons .
in addition , this assumption will allow us to link the lesser part of the screened interaction to the retarded through the fluctuation dissipation theorem @xcite as @xmath10 \ ] ] where @xmath11 is the ( bosonic ) plasmonic distribution function and @xmath12 is the spectral function , which can be written in terms of the retarded green s function leading to the rightmost expression
. the electron temperature will be represented as the expectation value @xmath13 for the effective conduction band .
this is with respect to the discrete energy for each state @xmath14 as the single particle energies are @xmath15 .
we calculate @xmath13 self - consistently in the model .
the lesser and retarded part are all that is needed to formulate the self - energy which will be described in detail below .
the coulomb matrix elements are given on the quasi-2d form following @xcite @xmath16 with @xmath17 being the dielectric constant and @xmath18 is the system area while the other constants have their usual meaning .
the relative constant @xmath17 is the equilibrium value for the main material system with no dopant atoms present . to this
we will add the non - equilibrium contribution as shown in eq .
, making up the total dielectric constant .
form factors _ are expressed as @xmath19 where the integral is taken along the growth direction , which defines the quantization direction of the quasi-2d system and @xmath20 are the quantized states in the @xmath21-direction used as a basis for the simulation .
the numerical integration is greatly simplified by expressing the wave functions as a truncated fourier series @xcite .
all form factors are then easily calculated and are thus available at a reasonable computational cost for all @xmath22 .
these form factors will also be close to exact as the number of fourier components needed for the well behaved wave functions of the heterostructures is limited .
we define the sppa following @xcite as @xmath23 where the effective plasmon pole is chosen to ensure that the approximation is correct both in the long wavelength @xmath24 the static limit , as thoroughly discussed in @xcite , and given in the two cases of screening dimensions as @xmath25 and @xmath26 where the plasma frequency is defined as @xmath27 where @xmath28 and @xmath29 are , respectively , the effective 2d and 3d carrier densities .
@xmath30 is the effective mass of the electron and @xmath31 is a term proportional to @xmath32 which should take into account the _ pair continuum
_ @xcite . in this work
we set @xmath33 .
furthermore , @xmath34 is the inverse of the static screening length given in the two cases as @xmath35 in 3d we use the debye limit of screening and for the 2d case the screening becomes a function of the occupation probability @xmath36 at @xmath37 for the effective conduction band . using the sppa and the dirac identity we find for the retarded part of the screened interaction @xmath38.\end{aligned}\ ] ] this
can be linked to the lesser propagator using the fluctuation - dissipation theorem as discussed above and it is found to be @xmath39 \\ & = -{{\rm i}}\pi \mathbf{v}(\mathbf{q } ) \frac{e_{pl}^2 } { e_q } \left [ f(e_q ) \delta(e - e_q ) + ( f(e_q)+1)\delta(e+e_q ) \right].\end{aligned}\ ] ] these are the standard non - interacting bosonic propagators , which are the same that we have used previously for the optical phonons in our model . following
the same treatment applied to phonons one can formulate expressions for the retarded and lesser self - energies . however the static term in the retarded interaction has to be taken into account .
this is the exchange shift and it is crucial to preserve the limits of the sppa , static and long wavelength cases .
according to standard feynmann rules we express the diagram for the screened interaction self - energy in the gw approximation @xcite as @xmath40 where the times are now on the keldysh contour . using the langreth rules
@xcite we can transform them to the real time axis .
schematically we have for products structured as above the following rules : @xmath41 where the time arguments are implicit and are ordered as in the product in eq . .
at this point we make a rather drastic approximation and express the screened interaction in the sppa .
the static term in the retarded screened interaction is present in the first and third term . in the first it gives rise to the exchange described below ,
however in the third , it does not contribute as the retarded green s function , considered as a function of energy , only has poles in the lower imaginary half plane .
closing the contour in the upper plane prevents contributions from this term .
we can now proceed and formulate the self - energies for the screened interaction , and the treatment is analogous to that presented in ref .
@xcite for optical phonons , with the addition of the exchange term treated below . here
we make the crucial assumption that the self - energies are functions of energy only , in contrast with the more computationally intensive approach of ref .
@xcite that uses full energy and @xmath1-dependent terms . here
the full @xmath1-dependence is replaced by an average contribution , by evaluating the scattering matrix elements for a set of typical values for @xmath42 and @xmath43 , as explained in detail in @xcite .
this brings the scattering matrix elements out of the integral over @xmath44 and we write the lesser and retarded self - energy as @xmath45 and @xmath46 \label{eqsigmaspparet } \end{split}\ ] ] where the last term contains an principal value integral . as this term involves yet another integral in energy space , it is of higher numerical complexity than the other terms and currently neglected to expedite the numerical calculations .
the factor @xmath47 is again the bose distribution where the electron temperature enters . in each term
there is an average over all possible scattering angles .
the scattering matrix elements are given as @xmath48 and the plus and minus variants used above comes from the choice of typical values for the evaluation . in our @xmath1-independent approximation we put @xmath49 with @xmath50 takes the level difference properly into account for all combinations of indices .
@xmath51 is a representative value chosen to give a similar scattering rate compared to the case of thermalised subbands at the given lattice temperature .
as the hartree - fock self - energies are local in time they depend only on the difference in coordinates , and the exchange term is normally evaluated directly from the diagrammatic rules to be @xmath52 which simplifies if only the difference in time matters and we find @xmath53 where we moved the coulomb matrix elements out of the sum over @xmath2 in order to do the sum and reach the final 2d sheet densities @xmath54 .
the factor of 1/2 appears as we do not sum over spin for the exchange self - energy .
the @xmath55 is now an angle averaged quantity over the possible angles between @xmath2 and @xmath56 , where we again use typical values for @xmath42 and @xmath44 .
our model , without electron - electron scattering has been previously tested successfully on superlattices and qcl heterostructures @xcite .
the case considered here is a qcl design from 2009 @xcite .
the periodic structure is displayed in fig .
[ figbands ] together with the probability density for the subbands relevant for transport .
these type of four - well thz qcl structures have previously been troublesome to simulate , see for example @xcite , due to a pre - peak feature which is not recovered experimentally , at least not to the same extent as in the simulations . in theory , a lot more weight is put on the pre - peak at the cost of a lower main peak , compared to the opposite situation in experiments .
this design was also tested experimentally for the different doping concentrations , which makes it ideal as we want to study the impact of electron - electron interaction .
the experimental findings are summarized in table [ exptable ] .
sample a is doped to acquire a sheet density normal for thz qcls while sample b has about a factor three higher doping concentration . for the sppa , there is clearly an issue with vanishing carrier sheet density .
in such a situation the physical picture of a continuous spectra of modes of collective oscillations is questionable as a valid representation of the actual physics . while the wells are pure gaas .
in the widest well dopant atoms are placed in order to n - dope the structure , giving rise to impurity scattering , whereas interface roughness scattering originates from imperfect material interfaces .
probability densities are shown for the most relevant states .
this design employs resonant tunneling injection from 4 to 5 , building up inversion between 5 and 2 .
the lower laser state 2 is efficiently depopulated via resonant tunneling out to 3 via 1 , combined with phonon extraction to the starting point ; state 4 , shifted by one period . in ( a ) the alignment at design bias is shown , at a bias of 54 mv per period . in
( b ) the alignment at the parasitic resonance at 34 mv is shown , giving rise to the pre - peak in fig .
[ figcurr].,title="fig:",scaledwidth=40.0% ] while the wells are pure gaas . in the widest well dopant atoms
are placed in order to n - dope the structure , giving rise to impurity scattering , whereas interface roughness scattering originates from imperfect material interfaces .
probability densities are shown for the most relevant states .
this design employs resonant tunneling injection from 4 to 5 , building up inversion between 5 and 2 .
the lower laser state 2 is efficiently depopulated via resonant tunneling out to 3 via 1 , combined with phonon extraction to the starting point ; state 4 , shifted by one period . in ( a ) the alignment at design bias is shown , at a bias of 54 mv per period . in
( b ) the alignment at the parasitic resonance at 34 mv is shown , giving rise to the pre - peak in fig .
[ figcurr].,title="fig:",scaledwidth=34.8% ] the results of adding the new self - energy to the model is displayed for both sample a and b in fig .
[ figcurr ] . as a reference
we show currents calculated without the sppa self - energy for both 100 and 200 k lattice temperature . for high temperature
the current increases almost linearly up to design bias , around 54 mv as depicted in fig .
[ figbands](a ) , however for low temperature the current enters a region of negative differential resistance ( ndr ) after @xmath57 mv for both samples . after entering this region
it is no longer possible to converge the simulations ; that is why the black line abruptly stops .
the lack of convergence is attributed to artificially low lifetimes of the states due to the lack of important scattering mechanisms . in order to reach convergence in the self - consistent process
, the energy resolution would have to be increased significantly .
this is currently being investigated but the higher numerical complexity prevents us from showing any results here .
the pre - peak feature occurs at a low bias , before there is sufficient gain to start laser operation and it indicates that some important scattering is missing in the model as this is not the behavior seen in experiments . for sample
a , a threshold current of 175 a/@xmath58 is observed for 10 k heatsink temperature and this gradually increases to 430 a/@xmath58 for the maximum operating temperature .
the structure should thus be able to reach design bias also for low temperature .
the ndr in both samples occurs close to the bias matching a potential drop of the energy of one optical phonon , which is about 36 mev for algaas / gaas systems .
this strongly favors the parasitic current channel shown in fig .
[ figbands](b ) . when the sppa self - energy is included in the low temperature simulations the current changes drastically , and employing a 2d or 3d screening model makes a big difference .
the orange lines in fig .
[ figcurr ] shows simulations with 2d screening at a lattice temperature of 100k , and the current continues off the scale and is much too large compared to the experimental data .
the reason for this is that the plasmon dispersion has a finite value for the @xmath59 limit for 3d but tends to zero in the 2d case , as shown in eqs .
( [ eeff2d]-[eeff3d ] ) .
this effectively gives additional screening for the 3d case as the plasmon quanta , limited downwards by the plasma energy , always have to be exchanged with the collective modes that are assumed to be in thermal equilibrium . for the 2d case , there is nothing limiting the system from exchanging plasmons with vanishing energy , making both the occupation factor in eqs .
( [ eqsigmasppaless]-[eqsigmaspparet ] ) and the coulomb matrix element tend to large values . in the long wavelength ( @xmath59 )
limit the system should tend to a 3d electron gas , which is respected by the 3d but not the 2d screening model .
this , together with the observation that the 2d screening model gives an unrealistic current contribution , is the reason for focusing the rest of our discussion on the 3d screening model .
a few different cases are displayed in fig .
[ figcurr ] for the simulations with the sppa self - energy with 3d screening at @xmath60 k. qualitatively we observe better agreement as the simulations are similar to the reference case at 200 k , which is expected .
the quantative agreement is shown in table [ exptable ] . here
we use a lattice temperature 30 k higher than the _ heatsink _ temperature reported in the experiment . no laser field
is included , only off - state simulations are shown .
the agreement is reasonable except for the @xmath61 for sample a. for sample b the agreement appears to be better but as we consider off - state current and on - state current probably is higher , this might be too large as well .
the experimental decrease in current with lattice temperature is thus not seen in the simulations . the way we interpret
this is that the possibility of exchanging plasmons with the environment opens up new transport channels for the electrons , and that this effect actually is larger than the hypothesized effect of increased dephasing which would decrease the tunneling currents , such as the parasitic resonance at the ndr peak .
this effect plays a larger role at low temperature where phonon scattering is weak . in fig .
[ figcurr ] different limits of plasmon - phonon coupling are shown , indicated by @xmath62 and @xmath63 . in case of a strong coupling of the plasmon and phonon baths the plasmons would be quickly cooled by the phonons .
this is shown as blue lines where the electron temperature has been set equal to the lattice temperature @xmath62 . the opposite would be a weak coupling , and here the plasmons are only coupled indirectly through the mean electron temperature calculated as @xmath64 with respect to the bottom of each subband .
this is shown as magenta lines in fig .
[ figcurr ] . due to the occupation factors in eqs .
( [ eqsigmasppaless]-[eqsigmaspparet ] ) a higher electron temperature increases the strength of the sppa self - energy .
the impact of switching the exchange self - energy off is shown by the dashed line of the respective color . in fig .
[ figsigret ] the imaginary part of the retarded self - energy is plotted as a function of energy .
this is directly related to the linewidths of the single particle states .
one finds that the effect of this self - energy on the linewidth is below one mev , for all the states .
.experimental results from ref .
@xcite compared to simulations . here
we list , from left to right , sample name , sheet density , current at the ndr feature where lasing stops at low temperature and lastly the maximum current at the temperature , observed in experiments , where the laser ceases to operate . [ cols="<,<,^,^,^,^",options="header " , ] . compared to other scattering mechanisms
, the sppa gives a small contribution to the linewidth : about 7% of the total at this bias .
the lattice temperature is 100 k.,scaledwidth=50.0% ]
the screened electron - electron interaction has been included in a crude gw approximation via the sppa , including dynamical screening .
this leads to increased scattering in our model and a behavior that better represents the data from experimental studies , showing that these scattering mechanisms are indeed an important source of inelastic scattering when the other mechanisms of this type , such as acoustic phonon scattering , are weak .
exchange effects are shown to play a minor role only at higher doping concentrations , while the choice of coupling of the plasmon and phonon baths has a larger impact .
we observe that a 3d screening model gives reasonable results due to the finite value of the plasmon energy in the low @xmath22 limit , and we conclude that it also preserves the long wavelength limit . in contrast , the 2d screening model largely overestimates current and does not respect the long wavelength limit .
financial support from the cost action mp1204 in order to execute short term scientific missions is thankfully acknowledged , as is support from the swedish research council ( vr ) .
10 url # 1#1urlprefix[2][]#2 faist j , capasso f , sivco d l , sirtori c , hutchinson a l and cho a y 1994 _ science _ * 264 * 553556 | in this work we include electron - electron interaction beyond hartree - fock level in our non - equilibrium green s function approach by a crude form of gw through the single plasmon pole approximation .
this is achieved by treating all conduction band electrons as a single effective band screening the coulomb potential .
we describe the corresponding self - energies in this scheme for a multi - subband system . in order to apply the formalism to heterostructures we discuss the screening and plasmon dispersion in both 2d and 3d systems .
results are shown for a four well quantum cascade laser with different doping concentration where comparisons to experimental findings can be made . + _ accepted for publication in * journal of physics : conference series * by iop publishing ltd .
, as a proceeding of the pngf6 workshop in lund , sweden , 2015 . |
for the vast majority of protoplanetary disks , modeling of the ir spectral energy distribution has been the principal tool for characterizing the properties of the dust disk .
models for disks derived from such data are non - unique since the viewing geometry ( inclination ) , assumed grain opacity model , and the presence or absence of structure in the disk can combine to produce similar seds .
moreover , disks are not static : iso and spitzer studies of t tauri stars have shown that variability in the 3 - 4.5 micron range is ubiquitous @xcite .
at least some higher mass analogs to t tauri stars , the herbig ae stars , also exhibit striking variability in the ir sed which has been interpreted as reflecting changes in the scale height of the dust disk at or near the sublimation radius @xcite . for protoplanetary disks where
there has been significant dust grain growth and settling , the dust disk may have flattened sufficiently to be partially or largely shadowed @xcite by material at or near the dust sublimation radius . under these circumstances ,
the amount of light reaching the outer disk is expected to be anti - correlated with the ir excess in the 1 - 10@xmath4 m range , originating from material at or near the dust sublimation radius @xcite .
hst and ground - based nir single - epoch studies of herbig ae disks associated with ir seds interpreted as arising from shadowed disks ( meeus group ii ) resulted in a high non - detection rate , even for mm - bright , large protoplanetary disks @xcite , but also in some surprising detections @xcite . multi - epoch imaging of one of these stars , hd 163296 , demonstrated that the illuminated portion of the disk varied with time @xcite .
these observations typically lacked simultaneous measurements of the ir sed , which would be needed to test whether the disk illumination is indeed anti - correlated with the near and mid - ir light .
another herbig ae star with documented near and mid - ir excess variability is mwc 480 ( hd 31648 ) , located at d=137@xmath5 pc @xcite .
this star is one of the brightest herbig ae stars at millimeter wavelengths @xcite , and has been the subject of detailed studies from the x - ray through the radio regime @xcite , as well as repeated high - contrast imaging with hst @xcite and nir imaging with subaru @xcite .
hst imaging at the epoch of iso sws observations resulted in a marginal scattered - light detection of this disk @xcite at a time when the ir sed suggested that the ir excess was weaker than average .
however , the disk detection rate in scattered light is known to increase as further steps to increase the disk - to - star contrast , such as use of polarimetric differential imaging , are taken , and as the inner working angle of the observations decreases .
we present the first firm scattered light detection of the disk of mwc 480 with an inner working angle of 02 ( 27 au ) .
we carried out h - band ( @xmath6 m ) linear polarimetric observations of mwc 480 using the high - contrast imaging instrument hiciao @xcite with a dual - beam polarimeter at the subaru 8.2 m telescope on 2010 january 24 .
these observations were part of the strategic explorations of exoplanets and disks with subaru ( seeds ; * ? ? ? * ) .
the polarimetric observation mode acquires _ o_-ray and _ e_-ray simultaneously , and images a field of view of @xmath7 with a pixel scale of 9.53 mas / pixel .
mwc 480 was observed using a circular occulting mask with @xmath8 diameter .
the exposures were performed at four position angles ( p.a.s ) of the half - wave plate , with a sequence of p.a .
= 0@xmath2 , 45@xmath2 , 22.5@xmath2 , and 67.5@xmath2 to measure the stokes parameters .
the integration time per wave plate position was 19.5 sec and we obtained twenty - four sets by repeating the cycle of the wave plate . the adaptive optics system ( ao188 ; * ? ? ? * ) provided a diffraction limited and mostly stable stellar psf with fwhm of 0.07@xmath9 in the _ h _ band .
low quality images were removed prior to production of the final images , resulting in 18 good data sets with a total integration time for the polarization intensity ( hereafter _ pi _ ) image of 351.0 sec .
note that the data sets are taken with adi ( angular differential imaging ) mode , and the field rotation is 32 degrees .
the polarimetric data reduction is described in @xcite and @xcite , using the standard approach for polarimetric differential imaging @xcite .
instrumental polarization of hiciao at the nasmyth instrument was corrected by following @xcite .
we calculated @xmath10 from @xmath11 .
this approach mitigates two of the limitations of high - contrast imagery : the psf observations are obtained simultaneously with the disk data , eliminating the effects of variable seeing , and also provides a psf measurement which is a perfect color match to the star .
this is particularly important for systems where the j - h or h - k colors are redder than any main sequence candidates .
while a psf reference target was observed immediately after mwc 480 , the seeing was gradually changing and the reference target data did not provide a good match to the psf of the mwc 480 data . as a consequence , we focus on the @xmath10 image rather than psf subtraction intensity ( @xmath12 ) image .
the @xmath10 image is shown in figure [ mwc480pi ] .
non - saturated images of mwc 480 in h band were obtained immediately after the coronagraphic sequence , using an nd filter which transmits 10% of the light .
similar observations of the photometric calibration star hd 18881 ( a0v , j - h=0.0@xmath13=0.055 , h - k=0@xmath13 ) were obtained on 2010 january 23 .
hd 18881 was selected from the ukirt bright standard catalog @xcite for the mauna kea filter set described by @xcite , @xcite , and @xcite .
we derived the zero point magnitude as 24.8 @xmath14 0.01 from the photometry of the calibration star .
we observed mwc 480 during six epochs between 1996 and 2010 , using the aerospace corporation s broad - band array spectrograph system ( bass ) on nasa s infrared telescope facility ( irtf ) .
bass uses a cold beamsplitter to separate the light into two separate wavelength regimes .
the short - wavelength beam includes light from 2.9 - 6 @xmath4 m , while the long - wavelength beam covers 6 - 13.5 @xmath4 m .
each beam is dispersed onto a 58-element blocked impurity band ( bib ) linear array , thus allowing for simultaneous coverage of the spectrum from 2.9 - 13.5 @xmath4 m .
the spectral resolution @xmath15/@xmath16 is wavelength - dependent , ranging from about 30 to 125 over each of the two wavelength regions @xcite .
the entrance aperture of bass is a 1-mm circular hole , whose effective projected diameter on the sky was 33 - 34 , while a re - design of the dewar optics increased this value to 40 in 2010 .
the observations are calibrated against spectral standard stars located close to the same airmass . due to its proximity in the sky
, @xmath17 tau usually serves as the flux calibration star .
we also observed mwc 480 with the spex spectrograph on irtf on 11 nights between 2006 and 2011 , extending the study of @xcite .
the spex observations were made using the cross - dispersed ( hereafter xd ) echelle gratings in both short - wavelength mode ( sxd ) covering 0.8 - 2.4 and long - wavelength mode ( lxd ) covering 2.3 - 5.4 @xcite .
these observations were obtained using a 08 wide slit , corrected for telluric extinction , and flux calibrated against a variety of a0v calibration stars observed at airmasses close to that of mwc 480 .
the data were reduced using the spextool software @xcite running under idl . due to the light loss introduced by the 08 slit used to obtain the sxd and lxd spectra , changes in telescope tracking and seeing between the observations of mwc 480 and a calibration star may result in merged sxd or lxd spectra with a net zero - point shift compared to their true absolute flux values .
we used a variety of techniques to check for any systematic zero - point shift in the absolute flux scale , as discussed below in greater detail .
these included using the low dispersion prism in spex with a 30 wide slit , jhk photometry with the spex guider , and the 3 - 13.5 @xmath4 m spectrophotometry from bass . on many nights we observed mwc 480 and its a0v calibration star using the low - dispersion prism and a 30 wide slit .
to avoid saturation of the detector and minimize the wavelength calibration arc line blends , the flat - field and wavelength calibration exposures required a narrower slit , and the 08 slit was used . on nights of good transparency and seeing 10 or better
, this generally produces the same flux calibration as the other methods . on such nights ,
very little light hit the slit jaws , and individual exposures showed minimal scatter .
however , during nights of exceptionally good seeing ( 07 or better ) motion of the star on the detector can introduce small wavelength shifts , which will result in under - corrected or over - corrected telluric bands with a characteristic `` p cygni '' shape .
these have minimal affect on the overall absolute flux calibration , however . on the same night that some of the spectra of mwc 480 were obtained with spex , images in the j ,
h , and k filters of the spex guider camera ( `` guidedog '' ) were obtained . this uses the mauna kea filter set .
three calibration stars were also observed .
two of these , hd289907 and sa-97 - 249 , are standard stars for the mauna kea filter set , and the magnitudes of @xcite were used .
the third was hd 31069 , the a0v star used for the some of the spectroscopy , where 2mass @xcite values were used .
even with short ( 0.12 sec ) integration times per image , slight saturation effects were evident in some of the k - band data , and these were not used for the absolute flux calibration .
the disk of mwc 480 is detected out to a radius of @xmath31 ( 137 au ) ( figure [ mwc480pi ] ) .
the central black circle indicates the 015 radius of the occulting mask .
we measure an integrated @xmath10 from 02 - 20 of 1.2 @xmath14 0.1 mjy , with [email protected] mjy exterior to 06 in radius . on the same night , the star was at @xmath18 magnitude , resulting in a polarized fractional intensity of 0.08% .
in contrast to the nicmos marginal detection @xcite , the disk is detected at all position angles , except for a dark lane along pa@xmath360@xmath1410@xmath2 .
this pa is in reasonable agreement with the average disk semi - minor axis position angle of [email protected]@xmath2 derived from millimeter interferometry for an average of 5 co and hco@xmath19 transitions @xcite and the pa of the jet @xcite .
we therefore interpret the dark band as a feature due to depolarization due to the grain scattering phase - function , with no indication of material deficits in the disk .
the vector map of the @xmath10 disk is shown in figure [ mwc480pivec ] , superposed on the bright @xmath10 disk .
the vectors show only polarization angle information , since we have no contemporary @xmath12 data .
the bright inner disk ( @xmath20 ) polarization vectors show centro - symmetric features ( e.g. a butterfly pattern ) , indicating that the @xmath10 disk is really scattered light from the central star . in common with other seeds studies of young stellar object disks
, we measure the disk radial polarized surface brightness profile along the disk major axis , or the axis of maximum elongation in the hiciao data .
we extracted a swath 013 wide ( 2x the psf fwhm ) along the major axis from the center of the coronagraphic spot out to 3 on both sides of the star .
@xmath10 brightness was measured in 14 pixel @xmath21 1 pixel , 013 @xmath21 001 ( figure [ mwc480piswath ] ) .
both radial profiles are shown in figure [ mwc480piradpl ] . in the profile , r
= 015 is the occulting mask edge . from 02 - 06 ( 27 - 84 au )
we find that the @xmath10 radial surface brightness of pa=150 degrees and 30 degrees drops as @xmath22 and @xmath23 , respectively . from 06 - 10 ( 84 - 137 au ) , the radial surface brightness ( sb )
profile is steeper , @xmath24 .
the outer disk radial sb profile is similar to that of hd 169142 from 123 - 200 au , @xcite , hd 100546 from 10 - 140 au @xcite , and sao 206462 @xcite .
the shallow radial sb profile from 02 - 06 is similar to that seen in sao 206462 in the region of the spiral arms where the polarization fractional intensity is 0.6 % @xcite , although no resolved spiral features are detected in the disk of mwc 480 at the resolution of our data .
mwc 480 has a history of correlated near- and mid - ir variability @xcite with changes in the amplitude of the ir excess , rather than changes in stellar temperature , which has continued into the period covered by this paper .
individual measurements obtained from 2006 to 2011 are shown , together with the envelope of nir variations from the literature ( shaded ) in figure [ mwc480photospheric ] .
the hiciao data were obtained at @xmath18 , a historic minimum of nir brightness for mwc 480 .
the constant temperature of the ir excess permits us to estimate the dust disk scale from the nir excess . at the time of the hiciao observation ,
the disk scale height at the sublimation radius was only @xmath360% of that seen in the 1998 data , which yielded a marginal detection of the disk .
by contrast , _ spitzer _ irs and bass observations made in 2004 corresponded to a maximum excess state .
the hiciao data support an anti - correlation between disk detectability and the nir excess , consistent with predictions for dust disks where grains have grown and settled to the disk midplane @xcite .
despite being the brightest herbig ae star at millimeter wavelengths @xcite , the disk of mwc 480 has proven elusive to imaging in scattered light . despite multiple attempts with hst @xcite and subaru / ciao @xcite
, the disk has previously been imaged in scattered light only once , and then with only a marginal detection and incomplete recovery of the disk at all position angles @xcite .
the hiciao @xmath10 image of this disk represents the first detection of the disk in polarized light , and a detection which moreover probes the disk at all position angles , except along the system semi - minor axis .
the successful detection of the disk relies on two factors .
first , the hiciao observation was made at a time when the nir excess was at a historic minimum , providing the most favorable illumination conditions for the outer disk .
the second factor is polarimetric differential imaging with simultaneous measurement of the stellar psf from the unpolarized stellar signal .
the @xmath10 image was made by combining the stokes @xmath25 and @xmath26 images .
the stokes @xmath25 and @xmath26 parameters were created by subtracting two split images on the hiciao image , which can subtract unpolarized light ( psf ) of mwc 480 itself .
this approach provides not only a perfect seeing match , but also an exact color match to the star .
nevertheless , the disk is faint , with a fractional polarized intensity of only @xmath270.1% . despite its relatively old age ( 7 myr ; * ? ? ?
* ) , the disk has yet to develop distinct structural features , including divots , rings , or a cavity .
our data do reveal what appears to be a break in the radial polarized intensity power law at 06 ( 84 au ) , with a shallower power law interior to that point .
a similarly shallow power law is seen in the region of the spiral arms for sao 206462 @xcite .
however , the absence of distinct structure in the disk of mwc 480 constrains any spiral features in the the disk to have relative scale height ( h / r ) @xmath280.05 , in order for them not to be detectable at the spatial resolution of the hiciao data . in tandem with the presence of on - going accretion , which manifests itself through a soft x - ray spectrum , fuv excess light and mass loss via a jet @xcite , the disk of mwc 480 has a distinctly primordial character .
as befits a dynamically cold disk , mwc 480 s dust disk is also geometrically quite thin .
our detection of the disk , in tandem with the previous nicmos marginal detection , allows us to better constrain where the optically thick surface of the disk lies in h band relative to the gas disk .
@xcite found from modeling of co interferometric data a 10 @xmath14 1.1 au gas disk scale height at 100 au , corresponding to a disk opening half angle of 6@xmath2 and a disk inclination of 37.5@xmath2 .
@xcite estimated that the angle subtended by the dust disk rim at the sublimation radius was 2.2@xmath2 for the 1998 nicmos marginal detection of the disk in scattered light .
for the hiciao data , the star was 0.5 magnitudes fainter at h. if we assume that variation in the nir excess light scales with changes in the scale height of the dust disk inner rim , the rim subtended only @xmath360% of the scale height seen in 1998 data .
this constrains the optically thick dust disk surface to lie above 1.3@xmath2 . in tandem with the disk scattered light detection , this constrains the optically thick dust disk surface to lie above of 1.3@xmath2 .
the available data therefore suggest that the surface of the dust disk lies between [email protected]@xmath2 . adopting the average of this range ,
the dust disk has a scale height at 100 au only @xmath330% as high as the gas disk ( h / [email protected] ) , demonstrating that this is a stratified disk .
the synoptic data for mwc 480 further suggest that the outer dust disk is only rarely illuminated , in contrast to hd 163296 @xcite , although quantifying the duty cycle for disk illumination will require a denser grid of photometric and spectroscopic data than are currently available .
a further advantage of the availability of synoptic data in tandem with more specialized observations is that we can start to extract the vertical structure of the disk . by combining the imagery ( nicmos 1.1@xmath4 m non - detection data ; * ? ? ?
* ) with the synoptic data , we find that the 2004 spitzer irs detection of pah emission in mwc 480 @xcite corresponded to a high excess state ( bass data ) , a factor of 3 above the iso sws data ( figure [ mwc480photospheric ] ) . in this configuration , the entire dust disk surface would have been shadowed , indicating that pah emission either originates from the inner rim of the dust disk , or that it must be associated with material in the gas disk at altitude , rather than in close proximity to the outer dust disk ( figure [ mwc480_cartoon ] ) . if extended to other datasets , this technique can allow us to understand which disk constituents are routinely exposed to uv radiation , and which have less irradiation .
we can compare the outer disk properties of mwc 480 with the roughly coeval sao 206462 , and the younger , solar analog lkca 15 , both of which have similar @xmath10 measurements using hiciao ( wisniewski , priv .
* ; * ? ? ?
* ) , and prior hst imagery .
both mwc 480 and sao 206462 have radial polarized intensity profiles beyond 06 proportional to r@xmath29 , if the polarization efficiency is constant , suggesting that the outer disks are not highly flared .
model fits to the sed and sub - millimeter visibility for sao 206462 indicate a disk scale height at the inner edge of the outer disk of 9.2 au at 46 au @xcite . to be illuminated at 100 au , the dust disk in that system must lie above 20 au , a factor of 6.7 times higher than that for mwc 480 . as for lkca 15 ,
the nicmos data demonstrate that the disk is detected out to 12 ( schneider , priv .
comm . ) , with no evidence of shadowing of the outer disk along the system major axis . to meet this illumination condition , assuming a wall height of 7 au at 50 au in radius @xcite , the dust disk of lkca 15 must extend more than 14 au above the midplane at 100 au in radius . the combination of a planet candidate in the disk of lkca 15 ( kraus et al .
2011 ) , and spiral arms which may also be associated with jovian - mass planets in sao 206462 @xcite , suggests that the outer disks of these transitional disk systems are dynamically excited by massive planets . the extent to which this process is present
may therefore constrain the masses of planets , even for systems where the inner cavity is inaccessible to current imaging technology . however
, the primordial character of mwc 480 s disk compared to coeval or even younger systems demonstrates that the process of giant planet formation may not be ubiquitous , or proceed at the same rate in all systems .
addressing how the planet hosting disks differ from those with no evidence of planet formation will be a major task both for alma and for the next generation of high - contrast imagers .
we report the detection of the disk of the herbig ae star , mwc 480 , in scattered , polarized light from 02 - 10 ( 28 - 137 au ) in data taken with hiciao at the subaru 8.2 m telescope in early 2010 . at the epoch of the detection ,
the nir excess was at a historic low . given a history of excess amplitude , but not temperature changes , this implies that the vertical extent of the inner rim of the dust disk , and by inference , the accretion rate onto the star , was at a historic low .
an anti - correlation between the scale height of the dust disk near the sublimation radius and the degree of illumination of the outer disk is expected only when the dust disk has settled toward the disk midplane .
the data presented here , in tandem with earlier synoptic photometry and spectroscopy and a history of high - contrast imaging observations , provide direct confirmation that herbig ae stars with ir excesses characterized by power laws of their h / r are disks where dust grains have grown , settled , and where the outer disk is largely shadowed .
higher cadence synoptic observations are required to quantify the illumination / shadowing duty cycle for mwc 480 , and to obtain sufficient data for other meeus group ii disks to establish whether mwc 480 is typical of the shadowed disks . the first steps in comparing the disk geometry of systems like mwc 480 to older , transitional disks have demonstrated that the transitional disks considered have not only wide gaps , but outer disks which are much more vertically extended than seen for mwc 480 .
as the seeds survey continues , we will have the opportunity to obtain data not only for other transitional disks , but also for additional primordial disks , and should be able to test the hypothesis that transitional disks have structural changes that extend over the entire disk reflecting the presence of giant planets within the gapped region .
this work is partially supported by kakenhi 22000005 ( mt ) , kakenhi 23103004 , 23740151 ( mf ) , nsf ast 1008440 ( cag ) , nsf ast 1009314 ( jpw ) , nasa nnx09ac73 g
( cag and mls ) and the ir&d program at the aerospace corporation ( rwr ) . | one of the key predictions of modeling from the ir excess of herbig ae stars is that for protoplanetary disks , where significant grain growth and settling has occurred , the dust disk has flattened to the point that it can be partially or largely shadowed by the innermost material at or near the dust sublimation radius .
when the self - shadowing has already started , the outer disk is expected to be detected in scattered light only in the exceptional cases that the scale height of the dust disk at the sublimation radius is smaller than usual .
high - contrast imaging combined with the ir spectral energy distribution allow us to measure the degree of flattening of the disk , as well as to determine the properties of the outer disk .
we present polarimetric differential imaging in @xmath0 band obtained with subaru / hiciao of one such system , mwc 480 .
the hiciao data were obtained at a historic minimum of the nir excess .
the disk is detected in scattered light from 02 - 10 ( 27.4 - 137au ) .
together with the marginal detection of the disk from 1998 february 24 by hst / nicmos , our data constrain the opening half angle for the disk to lie between [email protected]@xmath2 . when compared with similar measures in co for the gas disk from the literature , the dust disk subtends only @xmath330% of the gas disk scale height ( h / [email protected] ) .
such a dust disk is a factor of 5 - 7 flatter than transitional disks , which have structural signatures that giant planets have formed . |
at the heart of an active galactic nucleus lies the putative black - hole `` monster '' that supplies its power .
much of this power emerges in the form of x - rays , which are in turn efficient probes of the immediate regions surrounding the nucleus .
the fuel supply for the massive central engine is thought to arise from a disk of dense , accreting matter with column densities @xmath9 well in excess of @xmath10 @xmath11 . in the presence of such matter , the reflected x - ray spectrum has two distinct features .
one is the fe k@xmath2 fluorescence emission line at 6.4 kev ( fabian et al .
1989 , george & fabian 1991 ) .
the other is compton reflection , a broad hump centered on a few tens of kev with a shape governed by photoelectric absorption at low energies and by compton recoil and the declining scattering cross - section at high energies ( lightman & white 1988 ; guilbert & rees 1988 ) .
the ubiquity of fe k@xmath2 lines in seyfert galaxies is well established .
recent results from the advanced satellite for cosmology and astrophysics ( @xmath5 ) suggest that many of these lines in seyfert 1 galaxies exhibit relativistic profiles that are caused by photons emerging near a black hole , and thus they most likely originate in an accretion disk ( nandra et al .
whether compton reflection is as common is not known .
although hard x - ray observations with @xmath12 show that compton reflection is clearly present in the _ composite _ spectrum of eight bright seyfert 1 galaxies ( pounds et al .
1990 ) , the level of significance of this feature and constraints for theoretical reflection models are poor for most _ individual _ objects in the @xmath12 sample ( smith & done 1996 , nandra & pounds 1994 ) .
the rossi x - ray timing explorer ( @xmath1 ) provides the large bandpass coverage and collecting area necessary to allow the unambiguous detection of compton reflection in many individual seyfert galaxies for the first time .
mcg@xmath05 - 23 - 16 is an x - ray bright active galaxy with an optical classification of seyfert 1.9 ( veron et al . 1980 ) .
in the 20 years since its discovery , its 2 to 10 kev flux has varied by a factor of 4 from a high state of @xmath13 ergs @xmath11 s@xmath4 in 1978 ( tennant 1983 ) , to a low state of @xmath14 ergs @xmath11 s@xmath4 in 1989 ( nandra & pounds 1994 ) , to a high state of @xmath15 ergs @xmath11 s@xmath4 in 1996 ( this work ) . in its high - flux state ,
mcg@xmath05 - 23 - 16 is among the five brightest seyfert galaxies in hard x - rays .
@xmath5 observations show the fe k@xmath2 line to be significantly broadened with a fwzi of @xmath3120,000 km s@xmath4 and an unusually complex profile ( weaver et al .
1997 , hereafter w97 ) .
the signature of compton reflection has not before been detected in this galaxy .
here we present the first results for mcg@xmath05 - 23 - 16 from @xmath1 .
this work showcases the data quality , and more importantly , the power of using @xmath1 for purposes of x - ray spectroscopy studies of active galaxies .
this paper describes early results from a comprehensive spectral and variability study of mcg@xmath05 - 23 - 16 that involves simultaneous @xmath5 observations .
the @xmath5 results and a joint analysis of the @xmath5 and @xmath1 data will be presented in a future paper .
@xmath1 was launched on 30 december 1995 and carries three scientific instruments : the proportional counter array ( pca ) , the high - energy x - ray timing experiment ( hexte ) , and the all - sky monitor ( asm ) . here
we present data from the pca , which consists of five collimated ( 1@xmath16 fwhm ) proportional counter units ( pcus ) that contain three multi - anode detector layers with a mixture of xenon and methane gas .
the pcus are numbered 0 through 4 .
each has a bandpass of 2 to 60 kev , a geometric collecting area of @xmath31,400 @xmath17 , and a fwhm energy resolution of @xmath3 8% at 6.6 kev .
see glasser , odell , and seufert ( 1994 ) for a detailed description of the pca .
mcg@xmath05 - 23 - 16 was observed with @xmath1 from november 27 to 30 1996 , for about @xmath3100 ks .
there was an overlapping observation with @xmath5 that began about half way through the @xmath1 observation on 29 november 1996 and lasted for approximately 35 ks .
we do not discuss the most recent @xmath5 data here except to use the observed photon index and absorbing column density ( weaver et al .
, in prep ) to constrain the @xmath1 fits . to maximize the signal - to - noise ratio between 2 and 10 kev , we accumulate photon events from the top xenon / methane layer of the pca .
we also use only pcus 0 through 3 ; data from pcu 4 is discarded because this detector is turned on for a smaller fraction of the on - source time due to breakdown .
data is discarded for times of earth occult ( when the earth elevation angle is less than 10@xmath16 ) , passage of the satellite through the south atlantic anomaly ( saa ) , and when there is severe electron contamination .
this yields a total integration time of @xmath380 ksec . for the total spectrum ,
the small statistical errors on the data points combined with systematic uncertainties in the detector response matrix cause large values of @xmath18 . also , the background subtraction is not yet accurate for the complete observation due to small variations in the internal background that are not accounted for in the current background model .
we therefore use an 11 ksec exposure that coincides with the @xmath5 observation on 29 november 1996 , and is accumulated from @xmath19 s to @xmath20 s @xmath1 mission elapsed time . for this snapshot , poisson statistics are an acceptable description of the point to point variations in the data , the systematic errors have a minimal impact on our results because they are comparable to the statistical errors , and the background model works appropriately .
the pca detector response matrix was generated by adding together matrices for pcus 0 , 1 , 2 , and 3 that were supplied by the @xmath1 guest observer facility ( gof ) at the goddard space flight center .
the individual matrices represent the most up - to - date pca calibration and were created 26 august 1997 using _ pcarmf _ v2.2.1 . at the time of this writing the in - flight energy resolution of the pca detectors has not been precisely determined , but based on spectral fits to line emission from cas - a , the most likely value has been measured to within @xmath31% ( keith jahoda , 1997 private communication ) .
we therefore use two response matrices that bracket the possible values of the energy resolution .
the v2.2.1 matrix has a fwhm resolution of 7.5% at 6.6 kev ( resolution factor of 0.75 ) as the default value .
the two response matrices that bracket the most appropriate value for cas - a have multiplicative factors of 1.07 and 1.13 on the default resolution yielding effective resolution factors of 0.8 and 0.85 , or 8.0% and 8.5% fwhm resolution at 6.6 kev .
these response matrices are labeled m8 and m85 in the spectral fits discussed below . in the work presented here ,
the uncertainty in the resolution only affects the significance of detection of the broad fe k@xmath2 line .
power - law fits to the spectrum of the crab with the v2.2.1 response matrix yield residuals less than 2% at all energies included in our fits ( keith jahoda , 1997 private communication ) .
the amount of background internal to the pcus varies according to the orbital environment of the satellite .
there are no detectors offset from the science axis to allow a simultaneous measurement of the background , and so the standard procedure that has been developed by the @xmath1 gof is to model the internal and cosmic background .
the background for mcg@xmath05 - 23 - 16 was modeled in february 1997 with the @xmath21 program .
the internal background was predicted from the particle and activation models that have been derived from earth - occult data .
the particle model is based on the value of the housekeeping parameter q6 , which tracks the instantaneous particle flux by measuring the rate of events that trigger exactly 6 of the lower level discriminators in each pcu .
the activation model estimates the additional background that is present in the detectors after the satellite passes through an saa .
after background subtraction , the pca count rate for mcg@xmath05 - 23 - 16 is @xmath340 counts s@xmath4 .
the 11 ksec spectrum contains @xmath22 photons .
we estimate that the current background model is reliable up to energies of 30 kev and ignore pulse - height channels above this energy .
channels below 2.8 kev are ignored because of remaining calibration uncertainties involving partial charge collection ( keith jahoda , 1997 private communication ) .
data are modeled with the @xmath23 spectral fitting package and @xmath18 statistics are used .
mcg@xmath05 - 23 - 16 has a large absorbing column of @xmath9 @xmath32 @xmath24 @xmath11 .
this column is measurable with the pca , but because of ongoing calibration issues at the lowest energies , a systematic error may exist in this measurement that would affect our spectral results .
therefore , we use the information that the simultaneous @xmath5 data provide near energies of 2 kev to place an additional constraint on our spectral modeling . in all cases ,
we fix the absorbing column density at @xmath25 @xmath11 ( weaver et al . , in prep ) , which was obtained by fitting the most recent ( simultaneous with @xmath1 ) @xmath5 data from @xmath26 kev with an absorbed power - law model .
this column is identical to that measured during the 1994 observation ( @xmath27 @xmath11 ; w97 ) .
the results of spectral fitting are listed in table 1 .
we use three models to describe the spectrum : a power law ( n@xmath28 ) , a power law plus a gaussian , and a compton reflection model plus a gaussian .
the reflection model is the @xmath29 model in @xmath23 , which is an exponentially cut - off power - law spectrum reflected from a disk of neutral material ( magdziarz & zdziarski 1995 ) .
the free parameters are the photon index of the primary power - law spectrum ( @xmath30 ) , the cutoff energy of the primary power - law spectrum ( @xmath31 ) , the abundance of heavy elements ( @xmath32 ) , the relative amount of reflection compared to the directly - viewed primary spectrum ( @xmath33 ) , the inclination angle of the disk normal to our line of sight ( @xmath6 ) , and the power - law normalization ( @xmath34 ) .
the quantity @xmath33 is equal to 1 for the case of an isotropic x - ray source above a disk with a covering factor @xmath35 = 0.5 ( @xmath36 = 2@xmath37 ) as viewed from the x - ray source .
the amount of reflected flux depends on @xmath6 and @xmath32 , with larger inclinations and/or larger abundances producing less reflection ( george & fabian 1991 ) .
in addition , because of the contribution to the reflection hump by down - scattering of high energy photons , the amount of reflection depends on the cutoff energy of the primary power - law spectrum . if @xmath31 is larger than a few hundred kev , there is no significant reduction in the amount of reflection at @xmath3830 kev in comparison to a pure power - law spectrum ; however , if @xmath31 is less than a few hundred kev , there are fewer primary photons that can scatter to energies below 30 kev .
to test the effect of varying @xmath31 , we examine models with @xmath31 = 60 kev , 200 kev , and 500 kev .
cutoff energies much less than 60 kev are unlikely because they provide poor fits unless @xmath30 @xmath38 1.75 , which is outside the range of indices allowed for the intrinsic spectrum by the simultaneous @xmath5 data ( weaver et al .
, in prep ) .
a value of @xmath31 = 200 kev approximates the cutoff implied by fits to non - simultaneous osse and @xmath12 data for mcg@xmath05 - 23 - 16 ( fig . 7 in zdziarski ,
johnson , & magdziarz 1996 ) and is similar to the mean cutoff energy for radio - quiet seyfert 1 galaxies ( gondek et al .
the upper limit of 500 kev is chosen simply to have no appreciable effect in the pca bandpass .
we also assume solar abundances and allow only one of the parameters @xmath33 and @xmath6 to be free for a given fit . for all models ,
the spectral parameters and statistical error bars that are derived using response matrices m8 and m85 ( 2 ) are similar , if not identical , and so we list only the numbers derived using m8 in the table ( columns 2 through 8) . the difference between m8 and m85 does matter when assessing the significance of detection of the broad fe k@xmath2 line .
we therefore list the values of @xmath39 , the f - statistic , and the probability of exceeding f for both m8 and m85 in columns 10 , 11 , and @xmath40 . for fits that include reflection , we also list three best - fit values for the parameters that are most dependent on @xmath31 .
these are the iron line equivalent width ( w@xmath41 , column 5 ) and @xmath6 or @xmath33 ( columns 7 and 8) .
the top entry corresponds to @xmath31 = 500 kev , the middle entry corresponds to @xmath31 = 200 kev , and the bottom entry corresponds to @xmath31 = 60 kev .
models that do not include reflection provide formally unacceptable fits .
an absorbed power law with @xmath31 placed above the pca bandpass provides an exremly poor fit ( table 1 , fit 1 ) with @xmath42 for 57 d.o.f .
the poorness of this fit is illustrated in figure 1 , which shows the pca data and the power - law model folded through the instrumental response .
the residuals for the fit are plotted in the bottom panel . adding a narrow gaussian to the model provides a large improvement ( table 1 , fit 2 , fig .
2a ) , but the fit is still statistically unacceptable with @xmath39 @xmath3 1.5 for 55 d.o.f .
the probability of obtaining such a large value of @xmath18 by chance is only @xmath30.01 . if the line is allowed to be broad ( table 1 , fit 3 , fig .
2b ) , the fit is improved significantly for response m8 , but not for m85 , as indicated by the small f - statistic and large probability of exceeding f in columns 14 and 16 .
this fit , however , remains formally unacceptable for both matrices ( columns 10 and 11 ) with a probability of @xmath30.02 for obtaining such large values of @xmath18 by chance .
models that include reflection provide formally acceptable fits ( table 1 , fits @xmath43 ) . to examine
how significant this improvement is compared to models that include only the iron line , we compare the fit with a power law and a broad gaussian ( fit 3 ) , to a fit with a power law plus reflection and a narrow gaussian ( fit 4 , fig .
2c ) . for the fit with reflection ,
the decrease in @xmath18 is large compared to the fit without reflection ; @xmath44 = 14 for m8 and @xmath44 = 21 for m85 .
f - statistic values are not listed for fit 4 because the f - statistic can not be used in this case to accept or reject the hypothesis that reflection provides a better fit to the data than a broad line .
instead , we calculate the ratio of likelihoods ( edwards 1972 ) for the reflection plus narrow gaussian model versus the power law plus broad gaussian model . this ratio is defined as l@xmath45/l@xmath46 = exp[@xmath47 for two models .
we find that reflection plus a narrow gaussian is 1,100 to 36,000 times more likely to be the correct description of the data than a power law and broad gaussian , assuming that the errors are normally distributed .
we next examine whether a broad line is detected when reflection is included in the model ( table 1 , fits 5 to 7 ) . to test the range of allowed parameter space ,
we assume different values for @xmath6 and @xmath33 .
for fits 5 and 6 , we fix @xmath6 at values of @xmath48 and @xmath49 , which cover the range of most likely inclinations for a neutral accretion disk measured from @xmath5 studies of the fe k@xmath2 line profile ( w97 ) .
angles less than @xmath48 are the least preferred by the @xmath1 data because they yield poor fits unless @xmath31 is as low as @xmath50 kev , which we have already ruled out since this requires the intrinsic spectrum to be too flat . for @xmath51 , @xmath33 is small and ranges from 0.36 to 0.80 implying covering factors of @xmath52 ; however , for @xmath53 , @xmath33 is large and ranges from 2.9 to 5.9 , implying unphysical covering factors of @xmath54 .
if we instead fix @xmath33 at 1.0 ( fit 7 ) , then @xmath6 ranges from 63@xmath16 to 81@xmath16 ( best - fit values ) with a 90% confidence range of 50@xmath16 to 87@xmath16 .
our results clearly show that @xmath33 , @xmath6 , and @xmath31 are strongly coupled . for a given @xmath33 , smaller inclinations and
w@xmath41 are allowed for a low @xmath31 , while for a given @xmath6 , a low @xmath31 allows more reflection and hence a larger @xmath35 .
the confidence contours for fit 7 for the width of the fe k@xmath2 line vs. @xmath6 for @xmath31 = 200 kev are shown in figure 3 .
for all of the above cases , a broad line significantly improves the fit over a narrow line , as indicated by the f - statistic @xmath55 .
the probability of exceeding f by chance ranges from 0.05 for the worst case ( and the poorest overall fit ) to @xmath380.001 . for fit 7 , the @xmath56 kev flux is @xmath57 ergs @xmath11 s@xmath4 . finally , we examine whether the gas that reprocesses x - rays is ionized by replacing the @xmath29 model , which calculates reflection from a neutral disk , with the @xmath58 model , which calculates reflection from an ionized disk .
the @xmath58 model adds the free parameters of disk temperature and ionization parameter , but yields no improvement to the fit compared to the neutral case .
also , the best - fitting ionization parameter is zero .
this result is consistent with the fact that the peak energy of the fe k@xmath2 line is approximately equal to that expected for fluorescence from neutral iron ( 6.4 kev in the galaxy rest frame ) , and rules out significant ionization of the disk .
we conclude that compton reflection @xmath59 a broad fe k@xmath2 line from a dense , neutral reflector are detected in mcg@xmath05 - 23 - 16 with @xmath1 . in an 11 ksec exposure ,
the line width is well constrained .
there is also little difficulty in statistically separating the reflection component from the broad iron line , a problem that has plagued the analysis of data from missions such as @xmath5 and @xmath12 .
we have shown that both compton reflection and a broad fe k@xmath2 line are detected in the @xmath1 spectrum of mcg@xmath05 - 23 - 16 .
this is the first clear detection of reflection in this galaxy , and the broad fe k@xmath2 line confirms the reality of the broad line in the @xmath5 data ( w97 ) .
detecting a broad fe k@xmath2 line with @xmath1 in a seyfert 1.9 galaxy provides the first independent confirmation of @xmath5 detections of broad lines in these objects .
this also puts to rest the idea that the broad lines seen by @xmath5 are spurious features caused by a `` continuum conspiracy '' , where a complex continuum mimics a broad line .
the significance of detection of the broad line ( i.e. , having @xmath60 kev ) depends somewhat on the response matrix .
however , even for the worst - fitting reflection model , a broad line is detected at @xmath6195% confidence .
for the best - fitting reflection models , a broad line is detected at greater than 99% confidence .
examining the parameter space for reflection models , we tested disk inclinations that range from @xmath6 = 50@xmath16 to 87@xmath16 , as suggested by a detailed study of the fe k@xmath2 profile ( w97 ) .
the parameters of disk inclination ( @xmath6 ) , the amount of reflection ( @xmath33 ) , and the high - energy cutoff ( @xmath31 ) are strongly correlated .
if we allow the amount of reflection to be a free parameter , then an angle of 50@xmath16 provides the poorest fit of the models we tested . for @xmath31 = 500 kev
, @xmath33 is small but for @xmath31 = 60 kev , @xmath33 becomes consistent with 1 , which is the value expected if the reprocessing gas is in the form of a disk .
an angle of 87@xmath16 provides the best fit , but for this case we find the opposite result : for @xmath31 = 500 kev , @xmath33 is consistent with 1 , but for @xmath31 = 60 kev , @xmath33 is quite large and implies an unphysical covering factor of neutral reflecting material of @xmath54 . assuming instead a normal disk geometry with @xmath62 ( @xmath63 ) , the fit is statistically comparable to the best - fit , and we confirm the above result , namely that @xmath6 can range from 50@xmath16 to 87@xmath16 for high - energy cutoffs in the primary spectrum ranging from 60 to 500 kev .
the model parameters derived from @xmath1 are similar to those derived from @xmath5 ( w97 ) . for a reflection model with @xmath62 , @xmath30(@xmath5 )
ranges from 1.84 to 2.05 ( 90% confidence errors ) depending on how the fe k line is modeled , while the mean value for @xmath1 from reflection - model fits is @xmath30(@xmath1 ) = [email protected] .
the indices are consistent although not necessarily identical , but this is not a problem since the photon index varies in this galaxy ( w97 ) . comparing the fe k - line parameters is not straightforward because the line is clearly non - gaussian ( w97 ) and the energy resolutions of the experiments differ by a factor of four .
however , the line parameters are consistent for a single - gaussian approximation . in @xmath5 ,
the line has e@xmath65 = [email protected] kev , w@xmath41 @xmath3 260 ev , @xmath66 @xmath3 0.5 kev , and fwzi @xmath32.4 kev ( w97 ) . in @xmath1 ,
the line has e@xmath65 = 6.26@xmath67 kev , w@xmath41 = 230@xmath68 ev , @xmath66 = 0.43 @xmath640.15 kev , and fwzi @xmath32.5 kev . the inclination of the accretion disk is determined by different methods for the two experiments . for @xmath1 , it is measured from the amount of compton reflection , and for @xmath5 , it is measured from the shape of the fe k@xmath2 line . w97 and turner et al .
( 1997 ; hereafter t97 ) derive different inclinations from @xmath5 using different techniques .
both authors use a model that consists of a theoretical line profile from an accretion disk ( a ` disk line ' ; fabian et al . 1989 , laor 1991 ) and
assume fluorescence from a neutral disk with @xmath62 ; however , w97 use a two - component model that consists of a disk line and a narrow gaussian at 6.4 kev ( which represents emission from gas further out ) , while t97 fit the entire line with a disk line .
additionally , w97 assume an outer disk radius of @xmath69 ( @xmath70 = gm/@xmath71 ) and derive the following three parameters : the disk emissivity index , @xmath72 ( r@xmath73 ) , the inner disk radius , @xmath74 , and the inclination ( @xmath6 ) .
using a different approach ,
t97 assume @xmath75 , @xmath76 , and @xmath77 and derive only @xmath6 .
depending on the choice of schwarzschild or kerr geometries , w97 find @xmath72 = 5 to 7 , @xmath74 = 13 to 22@xmath70 , @xmath6 = 52@xmath16 to 89@xmath16 , w@xmath41(gaussian ) = 54 to 69 ev , and w@xmath41(disk line ) = 198 to 223 ev .
using a schwarzschild geometry , t97 find @xmath78 @xmath16 and w@xmath41(total ) = 362@xmath79 .
it is not possible to prove which assumptions are correct because the geometry of the disk , the emissivity profile as a function of radius , and even whether or not all of the fluorescence arises from an accretion disk , are not known _ a priori_. we conclude that the inclination derived from the line profile does not necessarily represent of the true inclination of the accretion disk .
the range of inclinations derived from @xmath1 from the amount of compton reflection is 50@xmath80 ( 90% confidence ) , in agreement with w97 .
one problem for all of these analyses , including the current one , is the large equivalent width of the disk line . for @xmath31 = 500 kev
, we find w@xmath41 @xmath3 230 ev with @xmath1 , which is much larger than the maximum equivalent width of 160 ev predicted from a neutral disk ( george & fabian 1991 ) .
the w@xmath41 of the disk line can be reduced if we follow w97 and assume that @xmath325% of the iron line flux originates from somewhere other than the disk .
in this case , we find w@xmath41(disk line ) @xmath3170 ev .
this is still a problem , however , because the equivalent width for an edge - on disk is expected to be much smaller than 160 ev .
in fact , w@xmath41 predicted for a neutral disk viewed at our best - fitting inclination of 81@xmath16 for an incident spectrum with @xmath30 = 1.9 is only @xmath350 ev ( george & fabian 1991 ) .
clearly , there is too much flux in the iron line to be produced by such a disk .
large equivalent widths can be produced if the accretion disk is ionized , but the @xmath1 data rule out this possibility .
besides the dependence of w@xmath41 on disk inclination and @xmath30 , w@xmath41 is proportional to the iron abundance and the covering factor of reflecting material .
if we assume almost complete coverage of the x - ray source ( @xmath35 = 0.99 ) , an equivalent width as large as 170 ev restricts the disk to be inclined by no more than 70@xmath16 ( george & fabian 1991 ) . for the case of @xmath31 = 500 kev ,
this inclination is inconsistent with our statistical lower limit of 73@xmath16 ( table 1 ) .
thus it is impossible to have a self - consistent interpretation of the data for @xmath31 = 500 kev without having a factor of @xmath33 overabundance of iron . for smaller cutoff energies ,
smaller inclinations are allowed . for @xmath31
= 60 and 200 kev , the 90% confidence range of inclinations is 50@xmath16 to 81@xmath16 , which allows equivalent widths as large as @xmath3130 ev for a neutral disk with @xmath35 = 0.5 . to produce a value as large as w@xmath81 ev
, it is thus possible to have @xmath35 as small as 0.65 and still have solar abundances .
if iron is overabundant , @xmath35 can be less .
this technique of comparing the relative fluxes in the fe k line and reflection hump thus allows us to infer the presence of a high - energy cutoff in the intrinsic spectrum and to place approximate limits on @xmath31 of 60 to 200 kev .
our analysis clearly demonstrates the model - dependence of disk inclination angle and covering factor on the shape of the incident spectrum and provides a self - consistent explanation of the data without invoking unusually large abundances .
the unified model describes seyfert 1 galaxies as having their nuclear regions viewed preferentially at small inclination angles so that the continuum source and broad line region are not blocked .
conversely , this model describes seyfert 2s and intermediate types ( i.e. , seyfert 1.8s and 1.9s ) as having their inner regions viewed preferentially edge - on so that the continuum source and broad line region are blocked by thick clouds . within the context of this model , the mean inclinations for the two types
are predicted to be @xmath330@xmath16 and @xmath360@xmath16 for 1s and 2s , respectively .
seyfert 1 galaxies appear to follow the expected trend , with a mean inclination angle of 30@xmath16 ( nandra et al .
on the other hand , seyfert 2 galaxies may not follow the expected trend .
t97 derive small angles of 13 to 33@xmath16 for four seyfert 1.9 and 2 galaxies , including mcg@xmath05 - 23 - 16 , and conclude that these objects have their inner regions preferentially seen face on , similar to seyfert 1s .
in contrast , the @xmath1 data imply that the disk in mcg@xmath05 - 23 - 16 is viewed edge - on .
the 90% confidence range of inclinations inferred from fits with compton reflection models , 50@xmath16 to 81@xmath16 , is entirely consistent with the geometry of the nuclear region that is predicted by the unified model based on its seyfert 1.9 classification .
clearly , @xmath1 provides valuable information that can be used to constrain the geometry and distribution of gas near the centers of active galaxies . comparing results in the literature ,
we find that using @xmath5 data alone to derive disk inclinations from iron line profiles is ambiguous because there are many unknown parameters and the results depend crucially on the assumptions that go into the modeling .
although an approximately face - on disk can be derived for certain techniques of modeling the @xmath5 data , this interpretation is not supported by the @xmath1 data , which instead favor large inclinations .
using @xmath1 , we have detected the signature of compton reflection in the seyfert 1.9 galaxy mcg@xmath05 - 23 - 16 .
we also confirm the broad fe k@xmath2 line seen with @xmath5 with fwhm @xmath3 48,000 km s@xmath4 .
this measurement provides an independent confirmation of the broad line that is thought to be the signature of emission from the inner regions of an accretion disk orbiting a black hole . from spectral modeling , we derive a 90% confidence range for the disk inclination of 50@xmath7 , which is consistent with the inclination predicted from the unified model for mcg@xmath05 - 23 - 16 based on its seyfert 1.9 classification .
the equivalent width of the fe k@xmath2 line from the disk is at least @xmath3170 ev .
this is much larger than that predicted for a highly inclined , neutral accretion disk unless we infer supersolar abundances , an unusually large covering factor of reflecting material , or the presence of a high - energy cutoff in the incident spectrum between @xmath360 and @xmath3200 kev . we wish to thank keith jahoda for invaluable help with implementing and understanding the pca calibration .
we also thank the members of the @xmath1 guest observer facility for their help with the data extraction .
this research was supported by a nasa long term space astrophysics grant . | we report the detection with the rossi x - ray timing explorer of a compton reflection signature in the seyfert galaxy mcg@xmath05 - 23 - 16 .
@xmath1 also resolves the fe k@xmath2 fluorescence line with fwhm @xmath3 48,000 km s@xmath4 .
this measurement provides the first _ independent _ confirmation of @xmath5 detections in seyfert galaxies of broad fe k@xmath2 lines that are thought to be the signature of emission from the inner regions of an accretion disk orbiting a black hole . under the assumption that reflection arises from an isotropic source located above a neutral accretion disk , and using a theoretical model that accounts for the dependence of the reflected spectrum on inclination angle ,
we derive a 90% confidence range for the disk inclination of @xmath6 = 50@xmath7 .
the large inclination is consistent with that expected from the unified model for mcg@xmath05 - 23 - 16 based on its seyfert 1.9 classification .
if we assume that the high - energy cutoff in the incident spectrum lies at energies larger than a few hundred kev , then the equivalent width of the fe k@xmath2 line is much larger than predicted for the amount of reflection .
this implies either an enhanced iron abundance , a covering factor of reflecting material @xmath8 , or a cutoff in the incident spectrum at energies between @xmath360 and @xmath3200 kev . |
complex network tools have been used in different fields from social , technological and biological networks @xcite , to analyze the structure and dynamics of the network .
different centrality measures have been used to find the important vertices in a network e.g. , degree centrality , eigenvector centrality , katz centrality , vertex betweenness centrality , edge betweenness centrality , closeness centrality@xcite , etc . among them ,
the simplest measure is the degree centrality .
degree of a vertex is the number of edges connected to it .
pagerank is another important measure and is used in network tasks like , information retrieval , link prediction and community detection@xcite . in 1998 , brin and page @xcite developed the pagerank algorithm to rank websites in their search engine .
pagerank score , @xmath0 , of a vertex @xmath1 in an undirected and unweighted network , is defined as @xmath2 , where @xmath3 is the damping factor , also known as teleportation factor , @xmath4 is the adjacency matrix corresponding to the network , @xmath5 is the degree of the vertex @xmath6 and @xmath7 is the intrinsic , non - network contribution .
many researchers have studied the significance of @xmath8 in pagerank @xcite . however , till now , no study has been taken place for the same for @xmath7 . in many cases
the value of @xmath7 has been taken as @xmath9 @xcite .
localization of eigenvector is a common phenomenon in most of the networks and it is frequently suggested that the main cause of localization is due to the presence of high degree vertices @xcite .
localization means that most of the weight of the centrality accumulates in a few number of vertices .
recently , empirical studies in pagerank suggest that pagerank does not show localization @xcite . in this paper ,
we show that depending upon intrinsic and non - network contribution , pagerank value changes . this article has been organized in the following way : in chapter [ aproof ]
, we show that if the intrinsic and non - network contribution of the vertices are the same or proportional to their degrees , then the pageranks remain same or proportional to the degree of the vertices .
we also show that if the intrinsic and non - network contribution of the vertices is zero then pagerank values become proportional to the degrees . in chapter
[ numerical ] , our study on simulated and empirical networks support our findings . on the contrary to the common belief
we show that , for some networks , pagerank centrality can be localized and the value of @xmath7 can effect in the localization of the pagerank centrality . chapter [ conclusion ] concludes with some discussion .
consider a simple undirected connected network @xmath10 of size @xmath11 , with adjacency matrix @xmath4 .
the pagerank centrality @xmath0 of a vertex @xmath1 in the network @xmath10 is defined as : @xmath12 where @xmath13 is the damping factor ( google search engine uses @xmath14 @xcite ) , @xmath5 denotes the degree of vertex @xmath6 , and @xmath7 is the intrinsic and non - network contribution to the vertex @xmath1 . in matrix form ,
the equation ( [ equ1 ] ) can be written as : @xmath15 where @xmath16 is the diagonal matrix with elements @xmath17 and @xmath18 .
equation ( [ equ2 ] ) can be written as : @xmath19 since @xmath20 equation ( [ equ3 ] ) can be written as the sum of an infinite series : @xmath21 now @xmath22 , is a column stochastic matrix . since @xmath10 is undirected and connected @xmath23 is diagonalizable .
one can write @xmath24 as a linear combination of the eigenvectors @xmath25 of @xmath23 , as @xmath26 then the equation ( [ equ5 ] ) becomes @xmath27 let @xmath28 be the eigenvalues of @xmath23 corresponding to the eigenvectors @xmath29 , respectively .
then @xmath30 after some simple calculation , we get @xmath31 since , @xmath32 , we get @xmath33 therefore , the pagerank vector @xmath34 of a network is the linear sum of eigenvectors of @xmath23
. if the value of @xmath24 is proportional to degrees , then equation ( [ equ6 ] ) becomes : @xmath35 where @xmath36 is the leading eigenvector of @xmath23 .
note that @xmath36 is proportional to the degree of the network .
therefore , equation ( [ equ10 ] ) reduces to @xmath37 hence , pagerank centrality is proportional to the degree of the network .
pagerank can also be quantified successively with an initial estimation of @xmath34 from equation ( [ equ1 ] ) or ( equivalently equation ( [ equ2 ] ) ) .
equation ( [ equ2 ] ) will be , @xmath38 where @xmath39 is any initial value of @xmath34 usually taken @xmath40 @xcite .
if the intrinsic and non - network contribution is zero , then from equation ( [ remark2 ] ) , we get @xmath41 in this case , we take @xmath39 any non - zero vector as @xmath40 will not contribute anything .
after @xmath42 iterative steps we have @xmath43 now , writing @xmath39 as a linear combination of the eigenvectors @xmath44 of @xmath23 , we get @xmath45 for some appropriate choice of @xmath46 .
then @xmath47 since @xmath48 for all @xmath49 , we get @xmath50 in the limit @xmath51 . as @xmath52 and @xmath8 are constants and @xmath36 is proportional to degree , it follows that pagerank centrality is proportional to the degree centrality .
we explore four different values of intrinsic and non - network contribution , @xmath53 . 1 .
@xmath7 is inversely proportional to the degree of vertex @xmath1 .
2 . all @xmath7 are equal to one .
@xmath7 is proportional to the degree of vertex @xmath1 .
@xmath7 is proportional to the square of the degree of vertex @xmath1 . in all the above cases ,
we have taken @xmath14 . our numerical simulation on erds - rnyi random network @xcite ,
we have observed that if the value of @xmath7 is proportional to the inverse of the degree of vertex @xmath1 then pagerank score increases for a few number of low degree vertices and for other , pagerank is stable ( see figure [ fig1](a ) ) . here
` stable ' means , whether the pagerank scores are positively correlated with the degree . if @xmath53 is equal or proportional to the degrees of the vertices then pagerank centrality is proportional to the degree of the vertices ( see figure [ fig1](c ) ) .
pagerank values are stable when @xmath53 is either equal to one or proportional to the square of the degree of vertex ( see figure [ fig1](b ) and figure [ fig1](d ) ) . in comparison with random network , small - world and scale - free network are stable with any value of @xmath53 that we have considered ( see figure [ fig2 ] and figure [ fig3 ] ) .
however , in real - world networks , the results are not consistent for the above mentioned cases . in blog network
, we have observed that pagerank score fluctuates when @xmath53 is inversely proportional to the degree of vertex ( see figure [ figblog](a ) ) .
when all @xmath7 are equal to one , pagerank value fluctuates but less than that in the previous case ( see figure [ figblog](b ) ) .
when @xmath7 is proportional to the degree of vertex @xmath1 , @xmath53 and pagerank score correlate positively ( see figure [ figblog](c ) ) .
pagerank values are stable when @xmath7 is proportional to the square of the degree of vertex @xmath1 ( see figure [ figblog](d ) ) . in p2p ( peer - to - peer ) network ,
we have seen high fluctuation of pagerank value for some high degree vertices when @xmath53 is inversely proportional to the degree of a vertex ( see figure [ figp2p](a ) ) .
similar observation is also found when all @xmath7 are equal to one ( see figure [ figp2p](b ) ) .
when @xmath7 is proportional to the degree of vertex @xmath1 , pagerank score is also proportional to the degree of vertex @xmath1 ( see figure [ figp2p](c ) ) .
pagerank score is stable when @xmath7 is proportional to the square of the degree of vertex @xmath1 ( see figure [ figp2p](d ) ) . for email network ,
we have observed strong fluctuation of pagerank value when @xmath7 is inversely proportional to the degree of vertex @xmath1 ( see figure [ figemail](a ) ) . when all @xmath7 are equal to one , we have seen less fluctuation of pagerank value than that in the first case ( see figure [ figemail](b ) ) .
pagerank value also becomes proportional to the vertex degree when @xmath7 is proportional to the same ( see figure [ figemail](c ) ) .
when @xmath7 is proportional to the square of the degree of vertex @xmath1 , higher degree vertices hold higher pagerank values ( see figure [ figemail](d ) ) . in pgp ( pretty good privacy )
network , some high degree vertices fluctuate in pagerank values when @xmath7 is inversely proportional to the degree of a vertex @xmath1 ( see figure [ figpgp](a ) ) .
similar result has been found when all @xmath7 are equal to one ( see figure [ figpgp](b ) ) .
when @xmath7 is equal or proportional to the degree of vertex @xmath1 , pagerank value is proportional to the same ( see figure [ figpgp](c ) ) .
pagerank value increases with the increment of degree of vertices when @xmath7 is proportional to the square of the degree of vertex @xmath1 ( see figure [ figpgp ] ) .
internet network is comparatively stable than other networks when @xmath7 is inversely proportional to the degree of vertex @xmath1 and when all @xmath7 are equal to one ( see figure [ figas](a ) and figure [ figas](b ) ) . when @xmath7 is equal or proportional to the degree of vertex @xmath1 , pagerank value is also proportional to the same ( see figure [ figas](c ) ) .
when @xmath7 is proportional to the square of the degree of vertex @xmath1 , high degree vertices take the high pagerank score ( see figure [ figas](d ) ) .
* localization of pagerank centrality : * the extent of localization can be measured by calculating inverse participation ratio ( ipr ) @xcite .
the ipr of a pagerank vector @xmath34 is calculated as : @xmath54 when ipr is of order @xmath55 as @xmath56 , the pagerank vector @xmath34 is localized .
if ipr @xmath57 , @xmath34 is delocalized @xcite .
we observe that random and small - world network do not show localization of pagerank centrality .
in these two networks , the lowest iprs observe when @xmath53 is inversely proportional to the degree of vertex and the highest iprs observe when @xmath53 is the square of the degree of the vertices ( see figure [ ipr ] ( a ) and [ ipr ] ( b ) ) .
however , in scale - free network the pagerank is localized and the highest ipr is observed when @xmath53 is proportional to the square of the degree of a vertex and the lowest ipr is observed when all @xmath7 are equal to one ( see figure [ ipr ] ( c ) ) . in blogs network , the pagerank is localized and the highest ipr is observed when the value of @xmath53 is inversely proportional to the degree of the vertices .
the lowest value of ipr is seen when @xmath53 is proportional to the degree of the vertex ( see figure [ ipr ] ( d ) ) .
the pagerank is not localized in p2p and email networks . in both of these networks ,
the highest value of iprs are observed when @xmath53 is proportional to the square of the degree of vertex and the lowest ipr is observed when all @xmath7 are equal to one ( see figure [ ipr ] ( e ) and [ ipr ] ( f ) ) .
the pagerank is localized in pgp network and the highest value ipr is observed when @xmath53 is proportional to the square of the degree and the lowest ipr is seen when @xmath53 is proportional to the degree of the vertex ( see figure [ ipr ] ( g ) ) . in internet network , pagerank centrality is localized .
the highest value of ipr is observed when @xmath53 is the square of the degree of the network and the lowest ipr is observed when @xmath53 is proportional to the degree of the vertex ( see figure [ ipr ] ( h ) ) .
in this work , we have shown that pagerank centrality depends upon intrinsic , non - network contribution of the vertices .
if the intrinsic and non - network contribution of the vertices is proportional to the degree then pagerank score becomes proportional to the degree of the vertices .
we have also shown that if the intrinsic and non - network contribution of pagerank centrality is zero , then it also becomes proportional to the degree of the network .
our numerical simulation of three network models ( erds - rnyi random network , watts - strogatz small - world network and barabsi - albert scale - free network ) shows that the pagerank scores are more resilient in the intrinsic and non - network contribution in small - world and scale - free networks than in the random network . among all real - world networks studied here ,
blog , p2p , e - mail and pgp , pagerank score fluctuates heavily when intrinsic and non - network contribution are either inversely proportional to the degree or equal to one .
when the value of @xmath53 is proportional to degree , pagerank keeps the same proportional value .
high degree vertices possess high pagerank value when the @xmath53 is proportional to the square of the degree .
internet network shows very small fluctuation of pagerank score when @xmath53 is either inversely proportional or equal to one and for other two values of @xmath53 show similar result as other real networks . on the contrary to the common belief
we show that pagerank centrality can show localization for some networks and the extent of localization depends upon the intrinsic and non - network contribution of the pagerank centrality . in synthetic networks
, we have seen that the value of ipr increases with the increase in the value of @xmath53 .
however , in the real - world networks studied here , no such correlation is found . in blog network
, ipr attains the highest value when @xmath53 is inversely proportional to the degree and for other networks the highest value of ipr was observed when @xmath53 is proportion to the square of the degree . in a nutshell , we have observed that the intrinsic and non - network contribution plays an important role in pagerank centrality calculation . here , we have considered the underlying undirected structure of all the networks .
the giant component ( i.e. , the largest connected sub - network ) has been considered when the network is not connected .
* three simulated networks : * erds - rnyi random network is constructed with 1000 vertices and two vertices are connected with probability @xmath58 @xcite . in watts - strogatz small - world network ,
the number of vertices is @xmath59 , average degree of initial regular graph is @xmath60 and the rewiring probability is @xmath61 @xcite .
barabsi - albert scale - free network is generated with @xmath59 vertices and size of the seed network @xmath62 .
a new vertex is added with the existing @xmath63 vertices @xcite .
* blogs network : * here vertices are the blogs and an edge represents a hyperlink between two blogs .
the number of vertices and edges are 1222 and 16714 , respectively .
the data were downloaded from konect @xcite and used in @xcite * gnutella peer - to - peer ( p2p ) network : * gnutella is a system where individual can exchange files over the internet directly without going through a web site . here , peer - to - peer means persons to persons .
the vertices are hosts in the gnutella network topology and the edges represent connection between them .
the number of vertices are 22663 and number of edges are 54693 .
p2p network data were downloaded from snap@xcite and used in @xcite . *
e - mail network : * e - mail network was constructed based on the exchange of mails between the members of the university of rovira i virgili ( tarragone ) . here
the vertices are the users , and two users are connected by an edge if one has sent or received email from other .
the number of vertices and edges are 1133 and 5451 , respectively .
the data were downloaded from @xcite and used in @xcite * pretty good privacy network : * here the vertices are the users of pretty good privacy ( pgp ) algorithm and edges are the interactions between them .
the number of vertices and edges are 10680 and 24316 , respectively .
the data were downloaded from @xcite and used in @xcite .
* internet network : * the vertices in this network are autonomous systems ( collection of computers and routers@xcite ) and the edges are the routes taken by the data traveling between them .
the number of vertices and edges are 22963 and 48436 , respectively .
internet data were downloaded from @xcite and used in @xcite .
the author thanks anirban banerjee and abhijit chakraborty for critical reading and suggestion .
author also thanks prof .
asok kumar nanda for helping to prepare the manuscript .
author gratefully acknowledges the financial support from csir , india through a doctoral fellowship . | pagerank centrality is used by google for ranking web - pages to present search result for a user query . here , we have shown that pagerank value of a vertex also depends on its intrinsic , non - network contribution . if the intrinsic , non - network contributions of the vertices are proportional to their degrees or zeros , then their pagerank centralities become proportion to their degrees . some simulations and empirical data
are used to support our study .
in addition , we have shown that localization of pagerank centrality depends upon the same intrinsic , non - network contribution . |
some of us have recently argued that rotating black holes surrounded by relic cold dark matter ( dm ) density spikes may act as particle accelerators @xcite . in the limit of maximally rotating ,
extremal kerr black holes , we showed that collisions between particles , e.g. weakly interacting massive dm particles , may reach arbitrarily high center - of - mass energies .
for some related works see @xcite . in this sequel
, we compute the escape fraction and flux at infinity of the highly blue - shifted particles .
we demonstrate that the escape fraction is finite and of interest for any value of @xmath0 although the price one pays for sub - extremality is that the achievable centre - of - mass frame ( cmf ) collision energy is limited . our earlier work attracted considerable comment with regard to the fact that extremality for a kerr black hole might not be achievable in nature @xcite , although not all authors concur on this point @xcite .
accretion spins up the hole but radiative back - reaction limits the degree of spin - up .
the limit @xmath1 was derived @xcite for a thin accretion disk , and criticised as being nonrealistic for more general situations @xcite .
more recently , the role of binary black hole mergers has been studied and superradiant extraction of angular momentum from the larger black hole has been shown to set the limit @xmath2 @xcite . however we note that these limits apply only to astrophysical constraints , and that string theory may provide alternative options for generating extremal black holes .
given these motivations , it is imperative that we obtain an estimate for the flux expected from the ultra - energetic collisions that take place in the near - horizon region of kerr black holes @xcite .
concretely , we will focus on the flux emitted by dm spikes that form around intermediate - mass black holes ( imbhs ) , as detailed in @xcite .
such dm spikes are expected to extend down to distances that are of order the schwarzschild horizon radius , and to have a density that is essentially set by the annihilation rate of the dm .
we take into account the fact that massless products of ultra - energetic collisions may be captured by the black hole by including a fully - relativistic escape function into the computation .
the paper is organised as follows : we first set the scene in the next section , and present a relativistic formula for the total flux expected at infinity from collisions around an imbh .
this formula requires knowledge of the escape function , which encodes the proportion of massless products that arrive at infinity , as well as the dm density profile around the black hole .
we discuss the escape function in section [ escapefraction ] , and the density profile in section [ density ] .
we present our results in section [ results ] , and end by expanding on some open issues that must be tackled to improve and extend the results of this paper .
our aim is to obtain the flux emanating from particle annihilations in a dm spike collapsing around a rotating black hole . we first explain how we intend to compute the flux , for steady - state ( time - independent ) conditions .
the first ingredient is the dm number density in the vicinity of the black hole .
we model this using a distribution function in phase space @xmath3 , where @xmath4 and @xmath5 are the coordinates of real space and velocity space respectively .
this distribution function @xmath3 , which is a lorentz scalar , enters the annihilation rate quadratically .
defining @xmath6 as the annihilation cross section for a pair of dm particles and @xmath7 as the relative velocity of two dm particles with velocities @xmath8 and @xmath9 , we propose the following formula for the _ invariant cross section _ @xmath10 where @xmath11 denotes the velocity space of the colliding particles .
the preceding equation tells us the number of annihilations taking place locally .
our objective is of course to find the flux of massless particles escaping the gravitational field of the black hole .
only massless particles produced with momenta satisfying certain conditions will propagate out to infinity . to obtain the number of escaping massless particles from the annihilation rate
, we must therefore dress the rate of eq .
( [ annihilation ] ) with an `` escape function '' which encodes the probability that the massless annihilation products escape to infinity .
we expect this function to go to zero at the horizon ( such that all of the annihilation products get captured there ) and approach one at infinity , so that all of the annihilation products very far from the black hole escape to infinity and contribute to the flux .
call the escape function @xmath12 , where we make explicit its dependence on the velocities of the colliding dm particles . folding this in with the annihilation rate above
, we have that the number of massless particles escaping the black hole per unit time @xmath13 is given by @xmath14 note that in principle we can use a similar formulation to obtain the flux spectrum , but our sole purpose is to show that the flux is potentially observable . if the observer is a distance @xmath15 away from the black hole , then to find the flux received per unit area per unit time by the observer we must divide the above by @xmath16 as usual .
note that in this work we limit the phase space integration domains @xmath17 to in - falling dm particles moving on geodesics that are captured by the black hole , the so - called plunge " orbits , as those are the ones which lead to collisions with high cmf energies .
for our purposes , the spatial integral will be computed in a spherical volume @xmath18 around the black hole , extending to a radius @xmath19 , measured in units where the schwarzschild radius @xmath20 is set to 1 .
the upper limit of the volume integral is a free variable ( in principle ) , and we may choose to restrict it in order to focus on the size of the flux emerging from the near horizon region , where the cmf energies of the collisions are high .
a crucial ingredient required to calculate the flux is the density profile of dm around the black hole . in this paper
we use a density profile corresponding to the spike " proposed in @xcite .
this density profile includes the effects of dm annihilation on the dm density ( as well as other effects ) and we shall make the assumption that it provides a good description in all regions down to the black hole horizon .
we stress that this density profile is also non - relativistic , and so its convolution with a fully relativistic ( and exact ) escape function is a source of potential inconsistency but this method provides a relatively straight forward way to estimate the density of dm particles around a black hole .
finally , the types of density spikes described in @xcite correspond to density profiles around non - rotating black holes .
the more energetic collisions occur for rotating black holes , and therefore in a more detailed analysis aimed at finding exact numbers a relativistic density profile is needed .
again we stress that we are only interested at this stage in an order of magnitude estimate for the size of the emergent flux so we leave such refinements to future investigations . regardless of how one estimates the density profile , in order to calculate the emergent flux we need to know what fraction of collision products escape the gravitational field of the black hole . to make the estimate as straightforward and as clear as possible
, we only consider the simplified case of the annihilation of two dm particles into two massless particles ( e.g. two neutrinos or two photons ) .
the escaping fraction of collision products is an @xmath21-dependent quantity . for the case of a schwarzschild black hole the escape fraction as a function of @xmath21
is a known result and a useful treatment can be found in @xcite .
when considering the case of plunge orbits , the escape fraction becomes more complicated due to the fact that the collision products are boosted towards the black hole . in the next section ,
we describe the calculation of the escape fraction as a function of @xmath21 for both schwarzschild and kerr black holes , including the relativistic effects of the boosts .
in this section , we present an analytic treatment of the escape fraction for massless particles .
these particles are produced in the collisions of the dm particles and are consequently highly boosted towards the centre of the black hole .
the details of this boost are dependent on the momentum of the dm particles .
this of course is determined by the angular momentum of the two dm particles .
it is a relatively straightforward exercise to construct the escape function for a massless particle produced in the gravitational field of a schwarzschild black hole @xcite . here
we want to do a similar calculation but with the massless particles boosted according to the kinematics of the initial colliding dm particles .
let us first examine the motion of massless particles in the locally - nonrotating - frame ( lnrf ) , which is the frame in which an observer who rotates with the local geometry does their measurements .
working in the equatorial plane , the lnrf momentum of a massless particle is given by p^lnrf=(p^lnrf _
, p^lnrf _ , p^lnrf _ ) where the components take the forms @xcite p^lhrf_(r , a)&= & , + p^lhrf_(r , a , , b)&= & , + p^lhrf_(r , a , b)&= & , where we are working in units where the mass of the black hole and the schwarzschild radius @xmath22 are set equal to 1 . in this frame
we can establish the conditions on the massless particle s 3-momentum as a function of @xmath21 such that the it escapes the gravitational field of the black hole .
this is most easily performed by analysing the turning points ( i.e. where @xmath23 ) of the massless particle as a function of @xmath21 and @xmath24 .
the equation of motion for @xmath25 on the equatorial plane is = 1-+ .
[ rdot ] setting this equal to zero and solving for @xmath24 as a function of @xmath21 we find the two solutions b(r)=. [ solb ] we can now use these solutions to tell us whether a massless particle produced at some distance @xmath21 , with particular values of @xmath24 and @xmath26 ( @xmath27 and determines whether the massless particle is moving to larger @xmath28 or smaller @xmath29 radii ) will escape the black hole . to do this for general @xmath30
analytically is tricky , and we choose to investigate two values of @xmath30 , namely @xmath31 , the schwarzschild limit , and the extreme kerr case of @xmath32 .
other values of @xmath30 give results that are numerically not too different from the analytic results presented below and we leave the details of a numerical treatment for arbitrary @xmath30 to future work @xcite .
our goal here is to find an estimate for the emergent flux rather than to give a precise prediction .
starting with the schwarzschild case , @xmath33 the two solutions of eq .
[ solb ] take the forms b^a=0_(r)= .
we will for the remainder of the schwarzschild discussion omit the superscript @xmath31 , to avoid clutter . by plotting these solutions in fig .
[ bvsr ] we see that for the positive solution there is a minimum and for the negative solution there is a maximum .
both these extrema occur at @xmath34 and have values @xmath35 and @xmath36 .
we can now read off the conditions on @xmath24 as a function of @xmath21 that determine whether the massless particle will escape or be captured by the black hole . as a function of @xmath21 ( in units where @xmath37 ) .
the upper ( red ) curve is @xmath38 whereas the lower ( blue ) curve is @xmath39.[bvsr ] ] the conditions for the massless particle to escape are for r<3 : = + 1 - 3 > b > 3 , [ con1 ] which simply reflects the fact that a particle produced in the region @xmath40 , with a value of @xmath24 in between the minimum and maximum , will move towards infinity unhindered by the potential barrier . if the particle were produced with @xmath41 , then it will hit the potential barrier and bounce back into the black hole .
similar reasoning then yields the following additional conditions for escape , which apply in the region @xmath42 : [ con2 ] r>3 : & = & + 1 b_-(r ) > b > b_+(r ) ; + [ con3 ] & = & -1 3>b > b_+(r ) ; + [ con4 ] & = & -1 b_-(r)>b > -3 . for the @xmath43 conditions
, the massless particle will initially be moving to smaller @xmath21 but will reach a turning point ( at a distance @xmath21 determined by the value of @xmath44 ) and will be reflected by the effective potential leading to the escape of the massless particle . to calculate the escape function , we apply the following logic .
we assume that the annihilation of two dm particles proceeds as @xmath45 where we assert that the @xmath46s are massless particles . as an estimate designed to provide a reasonable order of magnitude measure , we only concern ourselves with the question of whether @xmath47 escapes . by doing this , we avoid the complications introduced by trying to consistently analyse whether either , both or neither @xmath47 or @xmath48 escape .
we construct the measure by assuming that in the cmf of the collision , @xmath47 is produced isotropically , that is , there is no preferred direction . in the cmf , there is an @xmath21 dependent range of angles that correspond to particular values of @xmath24 and @xmath26 ( evaluated in the lnrf ) that satisfy the conditions for @xmath47 to escape .
in other words , the conditions on the values of @xmath24 and @xmath26 for the massless particle to escape in the lnrf ( eqs .
( [ con1 ] ) - ( [ con4 ] ) ) can be mapped onto a range of angles in the cmf of the collision in which @xmath47 is produced .
because of the assumption that the production in the cmf is isotropic , the escape fraction can be computed in terms of the ratio of this range of angles over the full @xmath49 in the cmf on the equatorial plane .
the situation can therefore be summarised as follows : two dm particles collide at a given @xmath21 with two respective angular momenta .
we have two frames : the lnrf frame , which we can think of conventionally as the lab frame , and the cmf of the collision .
the two frames are related by a boost , which is defined by the radial position and the angular momenta of the colliding dm particles .
it is this boost that provides the mapping between the escape conditions on @xmath24 and @xmath26 and the range of angles in the cmf .
one immediate important point is that by assuming the massless particles are only produced and escaping in the equatorial plane , we are potentially overestimating the escape fraction .
however , this is not entirely clear as some particles could still escape even if they have trajectories that take them off the equatorial plane . for simplicity
we restrict our analysis to the equatorial plane . in a paper to appear , a numerical analysis is performed , @xcite , where it is shown that this approximation is indeed a good one to the accuracy to which we wish to work in this analysis . to find the @xmath21-dependent range of angles in the cmf that corresponds to the full range of escape conditions for @xmath24 and @xmath26 in the lnrf
, we can construct dot products between the 2 spatial components of the massless particle momentum vectors evaluated in the cmf ( i.e. the @xmath21 and @xmath50 components ) corresponding to the extreme values of @xmath24 for the appropriate values of @xmath26 and solve for the angle between them . this angle is then the escape range in the cmf , and all massless particles produced inside it escape . first we need to find the expressions for the momenta in the cmf . to do this
we simply apply a boost to the lnrf momenta for the massless particles such that p^cmf=.p^lnrf , is the cmf momenta and where @xmath51 is a general lorentz boost in the equatorial plane that can be written as @xmath52 where @xmath53 , @xmath54 and @xmath55 are the usual parameters of the lorentz boost .
the exact form of this boost is determined by the momenta of the dm particles that are colliding together .
parameterising the momenta of the massive colliding particles ( where we have set the mass of the dm particle @xmath56 ) as q_i(r , _
i , a)&=&(q_i , q_i , q_i ) where the momentum vectors are parameterised by the angular momentum @xmath57 and we use @xmath58 instead of @xmath59 to make it clear that this momentum vector is for a massive particle as opposed to a massless particle . the explicit forms after taking @xmath31 are @xcite q_i(r , a=0)&= & , + q_i(r , _
i , a=0)&= & - , + q_i(r , _
i , a=0)&= & . the boost parameters in terms of these momentum components
are then found to be & = & , + & = & , + & = & , + & = & .
we are now in a position to construct the dot products . using the boosted momenta of the massless final state particles we can construct the dot products
remember that this is in the cmf now .
explicitly we construct the vectors as ( r , b , ) = . with reference to the conditions set out in eq .
( [ con1 ] ) for @xmath40 the range of escape angles , @xmath60 , in the cmf reads ^r<3(r)= , with the escape fraction for @xmath40 , @xmath61 , given by ef^r<3(r)=. we again state that we are only considering motion in the equatorial plane .
the corresponding expressions for @xmath42 are ( following the conditions stated in eqs .
( [ con2 ] , [ con3 ] , [ con4 ] ) ) @xmath62}\\ + \abs{\arccos\;\left[\frac{\tilde{p}(r , \;b_+(r ) , \;-1).\tilde{p}(r , \;3\sqrt{3 } , \;-1)}{\abs{\tilde{p}(r , \;b_+(r ) , \;-1)}\abs{\tilde{p}(r , \;3\sqrt{3 } , \;-1)}}\right ] } \\
+ \abs{\arccos\;\left[\frac{\tilde{p}(r , \;-3\sqrt{3 } , \;-1).\tilde{p}(r , \;b_-(r ) , \;-1)}{\abs{\tilde{p}(r , \;-3\sqrt{3 } , \;-1)}\abs{\tilde{p}(r , \;b_-(r ) , \;-1)}}\right]}\end{gathered}\ ] ] with the escape fraction for @xmath42 , @xmath63 , given by ef^r>3(r)=. putting this all together , the total escape fraction is ef_a=0(r)&= & , where @xmath64 for @xmath40 and zero otherwise .
in addition to the dependence of @xmath65 on @xmath21 there is an implicit dependence on the initial angular momenta of the two colliding dm particles @xmath66 and @xmath67 .
we can plot this escape fraction for any combinations of @xmath66 and @xmath67 and in fig .
[ 2min2 ] we show @xmath65 for @xmath68 and @xmath69 as an example . with different combinations of angular momenta the size of the escape fraction does not change significantly . for dm particles colliding with angular momentum @xmath68 and @xmath69 .
[ 2min2],width=302 ] as a consistency check , we can remove the boost from the calculation and consider massless particles produced at some position @xmath21 and allowed to have initial momenta in any direction in the lnrf .
this is obviously a much simpler problem . comparing with the calculation of the escape cone outlined in @xcite we find complete agreement .
we now move to the extremal kerr case .
the calculation for kerr follows the same steps outlined for the schwarzschild case .
let us repeat the reasoning for clarity : two dm particles collide at a given @xmath21 with two respective angular momenta .
we have two frames : the lnrf frame , which we can think of conventionally as the lab frame , and the cmf of the collision .
the two frames are related by a boost , which is defined by the radial position and the angular momenta of the colliding dm particles .
isotropic production of the massless particle in the cmf is again assumed .
the conditions for escape , given by a range in @xmath24 and @xmath26 , as observed in the lnrf , are once again computed by considering the turning points of the massless particle .
the boost is then applied to convert the momentum of the escaping particle to the cmf .
this allows us to map the escape conditions in the lnrf , which are conditions on @xmath24 and @xmath26 , to a range of angles in the cmf .
this is then divided by @xmath70 to obtain the escaping proportion of massless particles . although the formula for the turning points is now more complicated , the reasoning is identical to the schwarzschild case above .
the first step is to analyse the turning points of @xmath25 once again . using eq .
( [ solb ] ) and setting @xmath32 we have the two solutions b^a=1_+(r)&=&r+1 , + b^a=1_-(r)&=&- .
we have plotted these two solutions in fig .
[ ae1turn ] .
note that the conditions for escape are now more complicated , but the logic is the same . if the particle has @xmath71 and it does not encounter a potential barrier , then it escapes .
if it has @xmath72 and it encounters a potential barrier , then it bounces off and escapes .
the solution , @xmath73 , has a maximum at @xmath74 with value @xmath75 .
the solution @xmath76 has a lowest value of 2 which occurs at the horizon , @xmath77 .
again , we will drop the superscript @xmath32 from what follows , as for the remainder of this section we will be concerned only with extremal kerr . as a function of @xmath21 .
the dashed ( red ) line is @xmath78 whereas the solid ( blue ) curve is @xmath79 .
[ ae1turn],width=302 ] from this plot , we can read off the conditions needed for the massless particle to escape ( note that here @xmath73 refers to the lower ( negative ) branch of the curve , as displayed in fig .
[ ae1turn ] ) .
they are for [ con5 ] r<4 : & = & + 1 - 7 < b < b_+(r ) ; + [ con6 ] & = & -1 2 < b < b_+(r ) , and for [ con7 ] r>4 : & = & + 1 b_-(r ) < b < b_+(r ) ; + [ con8 ] & = & -12 < b < b_+(r ) ; + [ con9 ] & = & -1b_-(r ) < b < -7 .
let us give the physical meaning of the escape conditions for @xmath80 . in this case
, a particle which is moving away from the horizon ( @xmath81 ) , will escape unless it hits the lower solid curve ( @xmath82 ) of fig .
[ ae1turn ] . to avoid that
, it must have @xmath83 , which is the maximum of that curve .
on the other hand , if it has a value of @xmath24 that is larger than the dashed line ( @xmath84 ) , then it will hit that line and bounce back to the horizon : this explains the condition with @xmath85 . for @xmath86 and @xmath43 , the particle which is moving towards the horizon ,
then escape will occur if it hits the red line , thus having @xmath87 .
similar reasoning yields the remaining conditions .
we now need our final state massless particle momenta in the cmf .
we follow the same procedure as before but now the initial dm momenta have the form q_i(r , a=1 , _ i)&= & + q_i(r , a=1 , _ i)&=&- + q_i(r , a=1 , _ i)&= & _ i , [ ae1comps ] where @xmath88 .
the boost parameters are now set by the choices of @xmath89 of the initial dm particles . using the components in eq .
( [ ae1comps ] ) and the boost , we can calculate the cmf momenta of the massless final state particles .
these cmf momenta will be functions of @xmath24 , @xmath21 and @xmath26 and we can apply the conditions necessary for escape listed in eqs .
( [ con5 ] , [ con6 ] , [ con7 ] , [ con8 ] , [ con9 ] ) .
following again the procedure outline above we now want to construct dot products of the vectors p^_(r , b , ) = .
evaluated with parameters corresponding to the extreme escape values quoted in the escape conditions .
the prime notation is there to remind the reader that we are evaluating this with @xmath32 ( not to be confused with the tilde notation for the @xmath31 schwarzschild case ) .
thus , we are able to obtain the angular range in the cmf where produced massless particles escape .
this is in complete analogy to the schwarzschild case above . the range of escape angles that corresponds to the range in the values of @xmath24 and @xmath26 for @xmath80 , as listed in eqs .
( [ con5 ] , [ con6 ] , ) @xmath90 , in the cmf reads @xmath91}\\ + \abs{\arccos\;\left[\frac{p^{\prime}(r , \;b_{+}(r ) , \;-1).p^{\prime}(r , \;2 , \;-1)}{\abs{p^{\prime}(r , \;b_{+}(r ) , \;-1)}\abs{p^{\prime}(r , \;2 , \;-1)}}\right]}.\end{gathered}\ ] ] for @xmath92 , applying the conditions in eqs .
( [ con7 ] , [ con8 ] , [ con9 ] ) @xmath93 takes the form for dm particles colliding with angular momentum @xmath68 and @xmath94 .
[ escapekerr2 - 2],width=302 ] @xmath95}\\ + \abs{\arccos\;\left[\frac{p^{\prime}(r , \;b_{+}(r ) , \;-1).p^{\prime}(r , \;2 , \;-1)}{\abs{p^{\prime}(r , \;b_{+}(r ) , \;-1)}\abs{p^{\prime}(r , \;2 , \;-1)}}\right ] } \\\abs{\arccos\;\left[\frac{p^{\prime}(r , \;-7 , \;-1).p^{\prime}(r , \;b_{-}(r ) , \;-1)}{\abs{p^{\prime}(r , \;-7 , \;-1)}\abs{p^{\prime}(r , \;b_{-}(r ) , \;-1)}}\right]}.\end{gathered}\ ] ] the total @xmath21 dependent escape function is given by ef_a=1(r)&= & , where @xmath96 for @xmath80 and zero otherwise .
again @xmath97 has an implicit dependence on the initial dm angular momentum @xmath98 and @xmath99 . in fig .
[ escapekerr2 - 2 ] we show a plot of the escape function as a function of @xmath21 for initial angular momentum @xmath100 and @xmath101 . using this escape fraction
, we can not only calculate the emergent flux of massless particles but we can plot the escape fraction vs the cmf energy of the collision .
doing so gives a clear measure of how likely it is that a massless particle that is produced in a very high energy collision will escape the black hole .
we are able to do this because the cmf energy of the collision is purely a function of the radius and the angular momenta of the colliding dm particles , @xcite , @xmath102 given this relationship , we can construct the plot shown in fig .
[ evsesc ] , which displays the escape fraction versus cmf energy for two dm particles colliding with angular momentum @xmath103 and @xmath104 and is evaluated in the extreme kerr limit , @xmath32 .
it is noteworthy that even for cmf energies that are close to @xmath105 times the dm mass , the escape function is almost @xmath106 , which is non - negligible , especially if these collisions are taking place frequently which of course depends on the dm density in this region as well as the size of the cross section for dm collisions . to get an estimate for the size of the escape fraction for higher energies , a good approximation for the parametric behaviour plotted in fig .
[ evsesc ] is @xmath107 .
this approximation is good for @xmath108 , any lower and this approximate parametric dependence breaks down . and @xmath109 in the gravitational field of a kerr black hole with @xmath32 .
the energy is displayed in units of the dm mass @xmath110 .
[ evsesc],width=302 ]
now that we have the escape fraction for a rotating black hole , we must turn to the dm density profile in the area immediately around the black hole , a region where reliable modelling of the dm dynamics is difficult . in
what follows , we shall make some simplifying assumptions that enable us to make a reasonable estimate for that density .
we consider an imbh , taken to be of reference mass @xmath111 as predicted to exist in the inner galaxy as failed building blocks for the central supermassive black hole , and possibly existing throughout the halo as failed satellites @xcite . here
@xmath112 is the mass of the sun and @xmath113 is a dimensionless number parameterising the absolute mass of the black hole .
we caution that this is a hypothesis , and there may be other routes to building supermassive black holes .
in addition , we assume that the dm consists of massive weakly interacting massive particles that froze out in the early universe .
doing so gives us rough ball park figure for the size of the annihilation cross section .
such particles are cold and form a density spike via adiabatic contraction during the formation phase of the imbh by baryonic dissipation . for our purposes
the details of the spike or any other part of the dm profile are irrelevant if they concern distances bigger than @xmath114 .
all we need take from the profile is that the rising dm density as one approaches the horizon is tempered by its annihilation , so that in the immediate vicinity of the black hole an `` annihilation plateau '' is reached
. the ( larger ) radius at which the plateau is reached is unimportant for our calculation because it is much larger than @xmath115 . on the other hand , the radius up to which the plateau continues ( the inner radius of the plateau )
, we take to be the black hole horizon .
therefore , we assume that the dm density is essentially flat from the black hole horizon all the way up to values of the radius which are @xmath116 , where @xmath117 .
we emphasize that the details of the dm density profile at such radii are irrelevant for us , as we are interested in the most energetic particles resulting from annihilations taking place at radii that are @xmath118 .
the density in the annihilation plateau is given by @xmath119 , where @xmath120 is the thermal - averaged annihilation rate per unit volume , @xmath121 is the dm mass , and @xmath122 is the time - scale for spike formation . for details of the annihilation plateau ,
we refer the reader to @xcite .
although we now have an estimate for the _ total _ dm density in the near - horizon region , we are still short of information as to the composition of the dm particles with regard to their angular momenta .
this is of crucial importance because the energies of the emitted photons are set by the angular momenta of the colliding dm particles , and the very energetic collisions can occur for dm particles in plunge orbits .
so what we must do is obtain an estimate of the proportion of particles in the plateau which are in fact within the plunge range of the black hole .
the complexity of the problem means that this is beyond the scope of the paper , and , given that we are only interested in an order - of - magnitude estimate of the flux emitted in the immediate vicinity of the horizon , we simply assume a flat distribution in angular momentum .
in addition , we assume that for the particles colliding near the horizon , the proportions of plunge and bound geodesics are comparable and so we take the absolute value of the density of both to be of order the height of the plateau .
we remind the reader that the correct angular momentum decomposition of the dm particles in the plateau is likely to be very complicated , especially in the near - horizon region .
our aim in this analysis is to estimate the size of the total emergent flux originating from near horizon collisions .
it turns out that we are almost insensitive to the details of the angular momentum distribution as we are effectively integrating over the full range of angular momentum states .
if we wanted to calculate the energy spectrum of the emerging massless particles the choice of the angular momentum distribution becomes more important with the potential to change the shape of the spectrum .
in this section we put together the ingredients computed so far : the density in the near - horizon region and the escape fraction to calculate the flux .
going back to eqs .
( [ annihilation ] ) and ( [ flux ] ) , for plunge orbits we can exchange the dependence on the velocities for a dependence on the angular momentum of the colliding particles .
the phase space distributions , @xmath123 , can in principle be recast as functions of @xmath124 and @xmath89 .
the @xmath89 dependence encodes the distribution of angular momentum states which in general changes with @xmath124 .
this @xmath124 dependent distribution is unknown ( as discussed above ) and we make the approximation that we can factorize the @xmath89 dependence from @xmath123 ( such that this distribution is constant in @xmath124 ) .
we now assume that the remaining distribution function in @xmath89 is flat across the allowed range in @xmath89 .
in addition , the escape function @xmath125 changes slowly with @xmath57 and so we can treat it as a constant with respect to the integral over the angular momenta of the colliding particles .
further to this assumption we remind the reader that the escape fraction has been calculated for the equatorial plane only .
to do the full volume integral we need to know how the escape function changes as we go off the equatorial plane . instead of changing the form of the escape function to account for this variation ( which is a task most
easily completed numerically ) we assume a spherical symmetry such that we can use the same escape fraction for the full volume .
we can convert the number density @xmath126 to a mass density @xmath127 by dividing by the rest mass @xmath128 . for our purposes we set this density equal to the plateau density @xmath129 .
the final form for the flux arriving at some distance @xmath15 away from the black hole simplifies to a spherically symmetric volume integral ( a consequence of the assumed spherical symmetry of the density and escape function in the near - horizon region ) and can be written as @xmath130 where @xmath131 is to be taken as some typical annihilation cross section , @xmath132 is the mass of the dm particle , and @xmath15 is the distance to the black hole .
a further complication which we do not address here is that the cross section will in general be energy and therefore @xmath21 dependent . as we wish to make a simple estimate of the flux we take the cross section to be constant in energy .
a further simplification is that we assume that @xmath133 is dominated by an s - wave term , which is velocity independent . in doing
so @xmath133 is assumed to be just a constant . in general
there will be velocity dependent terms and consequently we would replace @xmath134 with the equivalent thermally averaged cross section .
note also that the form of the flux in eq .
[ flux ] is similar to the flux spectrum constructed in ref .
@xcite with the addition of the escape function .
ref.@xcite deals with the flux spectrum @xmath135 which requires an additional ingredient @xmath136 , which is the number of secondary particles produced per annihilation .
we are interested in the total flux of particles , rather than the flux spectrum ( the latter is obviously very important , but its calculation is much harder in a relativistic setting ) .
we can trivially convert the flux spectrum to total flux by assuming that photons reaching us will have energies bunched around the mass of the dm particle , so that it is safe for our purposes to write @xmath137 and integrate trivially over energy to obtain the total flux in eq .
( [ flux ] ) .
the escape fraction @xmath125 was computed in section [ escapefraction ] .
it is relativistic , and so its convolution with a non - relativistic density is one of the major approximations in this work .
as the escape function is slowly varying with @xmath57 we may choose @xmath138 as we wish ( if we knew the angular - momentum distribution of the plateau then we would have integrated over these two variables weighted by the distribution function in angular - momentum space ) .
we choose one of @xmath139 and @xmath140 to be the critical angular - momenta corresponding to the highest - energy collisions .
this choice will not alter the order of magnitude of the results below .
the density @xmath141 is simply the _ mass _ density of dm particles that are on plunge orbits .
as outlined above , we will use @xmath142 , again implicitly assuming that the proportion of plunge particles is of the same size as non - plunge particles .
note that we are integrating from the horizon @xmath143 up to an upper radius @xmath144 which is a few horizon lengths .
the factor of @xmath145 is necessary for converting the integral from @xmath37 units to si units . plugging everything in
we are left with @xmath146 where we have isolated the dimensionless integral @xmath147 inside which is found _ all _ of the dependence on the rotation of the black hole .
this dependence on the rotation parameter @xmath30 , which enters into the limits of the integral as well as the integrand , is obviously rather complicated . computing this integral for extremal kerr and for schwarzschild , with @xmath148 for both ( and with @xmath149 for kerr and @xmath150 for schwarzschild )
, we obtain @xmath151 and @xmath152 . for comparison
, we note that if we put 1 in place of the escape fraction we obtain the values @xmath153 and @xmath154 .
we can now give an estimate of the expected emergent flux .
we use the following fiducial values for our various parameters : we set the mass of the black hole using @xmath113 and set it equal to @xmath155 in units of @xmath156 solar masses .
we use @xmath157 , set the distance between us and the black hole to @xmath158 and assume that the growth time - scale of the black hole is @xmath159 years .
the total flux is then given by : @xmath160 where @xmath161 .
note that by writing the flux in this fashion we have placed all of the dependence on the geometry into the integral @xmath162 .
we see that imbhs can be bright dm annihilation sources , and this has been explored and constrained in the literature @xcite . for the case of extremal kerr , integrating up to @xmath163 ,
we obtain the flux to be @xmath164 .
this is well within the reach of planned neutrino detectors .
if we integrate only up to @xmath165 for extremal kerr , thereby including the emergent flux of massless particles coming from the most energetic annihilations , we obtain an expected number of @xmath166 futuristic large area high energy neutrino / particle detectors in space such as owl ( @xmath167 ) would have an effective detection area solid angle efficiency product in excess of @xmath168 and could detect such signals over a few days ( admittedly only above @xmath169 ) .
we note below that penrose boosting may allow such energies to be achievable at infinity . while the actual flux from an imbh depends on its distance from us and on the other reference parameters , which are uncertain by several orders of magnitude , we note that our calculation of a finite escape fraction which is weakly dependent on the black hole angular momentum means that even detection of a single event might open a window on new physics , attainable only via particle collisions at extreme cmf energies .
our final result should be viewed as an order of magnitude estimate of the flux received on earth , which we find to be the same with or without black hole rotation . on the other hand ,
given that very high ( even arbitrarily high ) energies can be achieved in the annihilations around a rotating black hole , while the schwarzschild geometry only allows _
maximal _ cmf energies of @xmath170 , our results imply that the spectrum received from a rotating black hole should typically contain signatures of highly energetic products .
a next step is to compute the spectrum as a function of @xmath30 to confirm this expectation .
we leave this for future work .
we obtained above an order - of - magnitude estimate for the blue - shifted flux emitted by dm annihilations in the near - horizon region of an extremal kerr black hole .
we made several approximations and assumptions along the way that made the calculation tractable .
we assumed that the escape function calculated purely on the equatorial plane was sufficient , restricting the colliding dm particles and the escaping particles to motion only on the equatorial plane .
we also assumed that the dm spike " around the black hole was non - relativistic and spherically symmetric , and that it was flat right down to the black hole horizon .
in our view , the former of these two concerns is unlikely to be a source of large errors , because the escape fraction goes from zero to one regardless of the dimensionality or the geometry .
it is very easy to check using our results above that the change from schwarzschild to kerr makes very little difference to the escape function .
moreover , some work we have done , and upcoming work @xcite , shows that going to three - dimensions does not change the results significantly , thus corroborating our intuition .
the concern about the dm density profile is rather more serious , because the density enters the flux quadratically .
several effects need to be studied more rigorously , chief among them being the effect of the rotating geometry on the dm spike , as well as the feasibility of a more accurate modelling of the spike close to the black hole horizon in a fully relativistic fashion .
the advantage of such a construction is that it should naturally yield the composition of angular momentum in the dm spike , which is crucial for the flux and its spectrum .
we should mention here that a reliable computation of the flux spectrum requires another ingredient : an explicit particle physics model of the dm and its interactions .
this is necessary because it determines the energy dependence of the differential cross section of the annihilation diagrams , which , coupled with the geometry - dependent cmf energy of the collisions , feeds directly into the differential flux spectrum .
the flux spectrum is therefore expected to be a far more revealing signature than the total flux , as it should contain clues as to the nature and composition of the dm spike .
for example , we naively expect the spectrum to bunch around the dm mass , which in itself would be an exciting observation .
there are further necessary refinements of our work that we would like to comment on .
the computation of the cmf energy in the kerr field reference @xcite neglected back - reaction effects , which were studied in @xcite .
this issue deserves further attention . finally , supermassive black holes , the nearest one of which is at our galactic centre ,
should not be ruled out as possible sources of annihilating dm . even after partial destruction by a complex merging history , something which is not necessarily an issue very close to @xmath22 , cusps , albeit with softer profiles , are regenerated by dynamical processes @xcite .
finally , one of the most intriguing aspects , that we reserve for future work , is the exploration of the penrose process @xcite , which may allow the escaping annihilation products to tap the rotational energy of the kerr black hole , provided that the collisions occur within the black hole ergosphere .
this effect was briefly considered in @xcite , but the potential remains to be fully explored .
the work of mb was partially supported by the js guggenheim memorial foundation and fondecyt ( chile ) grant # 1100282 and # 1090753 .
bh thanks christ church college , oxford , for financial support , and cern , where some of this work was carried out .
smw thanks the oxford physics department for hospitality and the higher education funding council for england and the science and technology facilities council for financial support under the sepnet initiative .
t. jacobson and t. p. sotiriou , phys .
* 104 * ( 2010 ) 021101 [ arxiv:0911.3363 [ unknown ] ] .
e. berti , v. cardoso , l. gualtieri , f. pretorius and u. sperhake , phys .
* 103 * ( 2009 ) 239001 [ arxiv:0911.2243 [ unknown ] ] .
a. a. grib and yu .
v. pavlov , arxiv:1001.0756 [ unknown ] .
s. w. wei , y. x. liu , h. guo and c. e. fu , arxiv:1006.1056 [ unknown ] .
k. lake , phys .
* 104 * ( 2010 ) 211102 [ arxiv:1001.5463 [ unknown ] ] . | the escape fraction at infinity is evaluated for massless particles produced in collisions of weakly interacting particles accreted into a density spike near the particle horizon of an extremal kerr black hole , for the case of equatorial orbits .
we compare with the schwarzschild case , and argue that in the case of extremal black holes , redshifted signatures can be produced that could potentially explore the physics of particle collisions at centre of mass energies that extend beyond those of any feasible terrestrial accelerator . |
the non - keplerian rotation curves of spiral galaxies provided the first observational evidence that galaxies are embedded in extensive dark matter haloes ( @xcite ) .
historically , the 21-cm emission from atomic neutral hydrogen gas ( hi ) has been the main tool to derive galaxy rotation curves , because of its capability to trace the gravitational field beyond the optical stellar disc .
however , hi is less useful in constraining the rotation curve in the central parts of discs usually due to insufficient spatial resolution and the lack of hi gas in the inner parts of galaxies ( e.g. , @xcite ) . the interstellar medium in the centre is instead dominated by gas in the molecular and ionised phases ( e.g. , @xcite ) .
unfortunately , the rotation curves derived through co emission , the common tracer for the molecular gas distribution , often show non - axisymmetric signatures such as wiggles ( @xcite ) .
hot ionised gas has the additional disadvantage that the observed rotation alone is often insufficient to trace the total mass distribution and that its velocity dispersion needs to be included .
this velocity dispersion is generally influenced by a typically unknown contribution from non - gravitational effects such as stellar winds and shocks ( e.g. , @xcite ) .
some of these shortcomings of gas tracers can be overcome through careful correction and analysis .
however , a broader shortcoming of gas tracers becomes apparent because of their dissipative nature .
the gas is easily disturbed by perturbations in the plane from , for example , a bar or spiral arm ( @xcite ) .
gas also settles in the galaxy disc plane ( or polar plane ) and is thus less sensitive to the mass distribution perpendicular to it .
the typical velocity dispersions of the neutral and ionized gas , respectively , are @xmath410 kms@xmath5(@xcite ) and @xmath420 kms@xmath5(e.g .
@xcite ) . instead of gas
, stars could be used as a tracer of the underlying gravitational potential .
stars are present in all galaxy types , are distributed in all three dimensions , and are collisionless making them less susceptible to perturbations .
the random motion of the collisionless stars tipycally ranges from 20 kms@xmath5to above 300 kms@xmath5(@xcite ) .
this implies a scale height difference of over an order of magnitude between the gaseous and stellar components ( e.g. , @xcite )
. however , we need to measure both the ordered and random motions of stars before they can be used as a dynamical tracer . their random motion can be different in all three directions , an effect known as the velocity anisotropy .
this anisotropy requires more challenging observational and modelling techniques to uncover the total mass distribution .
nonetheless , these techniques are becoming available .
integral - field spectrographs like ` sauron ` ( @xcite ) , which is used in this study , allow us to extract high - quality stellar kinematic maps and enable us to perform dynamical models .
a common approach for spiral galaxies is to apply the asymmetric drift correction ( adc ; @xcite ) to infer the circular velocity curve from the measured stellar mean velocity and velocity dispersion profiles .
this approach is straightforward since the velocity and dispersion profiles can be obtained from long - slit spectroscopy , and no line - of - sight integration is required due to an underlying thin - disc assumption .
using this method requires assuming both the disc inclination and the magnitude of the velocity anisotropy . as such , the adc approach is widely adopted in studies that use the stellar kinematics to infer the circular velocity curve .
for example , when investigating the tully - fisher relation for earlier - type spirals ( e.g. , @xcite , @xcite ) , the speed of bars ( e.g. , @xcite ) , as well as the inner distribution of dark matter ( e.g. , @xcite ) . in this paper , we adopt an alternative approach , namely fitting a solution of the axisymmetric jeans equations to stellar mean velocity and velocity dispersion fields to infer mass distributions .
we apply this approach to the data acquired with the integral - field spectrograph sauron from inner parts of 18 late - type sb sd spiral galaxies .
these jeans models are less general than orbit - based models ( e.g. , schwarzschild s method ; @xcite ) but are far less computationally expensive while still providing a good description of galaxies dominated by stars on disc - like orbits ( e.g. , @xcite ) .
the applicability of these fitted jeans solutions even applies to dynamically hot systems such as lenticular galaxies .
the jeans models take into account the two - dimensional information in the stellar kinematic maps as well as integration along the line - of - sight .
we also compare the resulting circular velocity curves with those obtained through adc to investigate the validity of the assumptions underlying the simpler adc approach . in section [ s : observations ] , we summarise the sample including the ` sauron ` observations and near - infrared imaging .
the jeans and adc modelling approaches are described in section [ s : method ] and applied via markov chain monte carlo technique in section [ s : analysis ] .
we then compare the circular velocity curves from both modelling approaches in section [ s : vcirc ] and discuss the possible reasons for the significant differences we find in section [ s : discussion ] .
we draw our conclusions in section [ s : concl ] .
= 2.0 mm * 12|l| ngc & type & d & pa & @xmath6 & _ i _ & @xmath7 & @xmath8 & @xmath9 & @xmath10 & @xmath11 & @xmath12 + ( 1 ) & ( 2 ) & ( 3 ) & ( 4 ) & ( 5 ) & ( 6 ) & ( 7 ) & ( 8) & ( 9 ) & ( 10 ) & ( 11 ) & ( 12 ) + 0488 & sa(r)b & 32.1 & 5 & 0.230 & 42 & 2299 & 53.6 & 19.2 & 42.7 & 9.9 & 0.4 + 0772 & sa(s)b & 35.6 & 126 & 0.340 & 51 & 2506 & 54.0 & 20.6 & 47.1 & 19.0 & 0.4 + 4102 & sab(s)b & 15.5 & 42 & 0.445 & 59 & 838 & 15.0 & 22.1 & 19.5 & 1.5 & 1.1 + 5678 & sab(rs)b & 31.3 & 5 & 0.475 & 61 & 1896 & 24.4 & 24.0 & 20.9 & 3.1 & 1.1 + 3949 & sa(s)bc & 14.6 & 122 & 0.360 & 53 & 808 & 20.4 & 22.1 & 16.8 & 4.6 & 1.3 + 4030 & sa(s)bc & 21.1 & 37 & 0.240 & 43 & 1443 & 30.2 & 19.2 & 26.5 & 15.3 & 0.7 + 2964 & sab(r)bc & 20.6 & 96 & 0.450 & 59 & 1324 & 20.8 & 21.0 & 16.9 & 1.0 & 1.2 + 0628 & sa(s)c & 9.8 & 25 & 0.190 & 38 & 703 & 90.7 & 22.1 & 70.7 & 12.3 & 0.3 + 0864 & sab(rs)c & 21.8 & 26 & 0.320 & 50 & 1606 & 38.3 & 15.1 & 28.1 & 1.6 & 0.5 + 4254 & sa(s)c & 19.4 & 50 & 0.270 & 46 & 2384 & 52.1 & 19.2 & 40.7 & 15.8 & 0.5 + 1042 & sab(rs)cd & 18.1 & 174 & 0.290 & 47 & 1404 & 73.4 & 15.1 & 52.5 & 4.4 & 0.3 + 3346 & sb(rs)cd & 18.9 & 100 & 0.160 & 42@xmath13 & 1257 & 55.3 & 17.8 & 35.4 & 3.2 & 0.5 + 3423 & sa(s)cd & 14.7 & 41 & 0.230 & 42 & 1001 & 52.6 & 20.1 & 40.1 & 9.7 & 0.5 + 4487 & sab(rs)cd & 14.7 & 77 & 0.370 & 53 & 1016 & 49.8 & 19.2 & 36.8 & 10.6 & 0.5 + 2805 & sab(rs)d & 28.2 & 125 & 0.240 & 43 & 1742 & 65.8 & 12.6 & 50.8 & 13.0 & 0.3 + 4775 & sa(s)d & 22.5 & 96 & 0.135 & 34@xmath13 & 1547 & 29.0 & 17.8 & 19.3 & 8.3 & 0.9 + 5585 & sab(s)d & 8.2 & 38 & 0.360 & 53 & 312 & 69.7 & 17.8 & 53.2 & 15.7 & 0.3 + 5668 & sa(s)d & 23.9 & 120 & 0.155 & 37@xmath13 & 1569 & 38.3 & 16.5 & 30.3 & 13.8 & 0.5 + [ tab : prop ] our sample consists of 18 nearby , late - type spiral galaxies with hubble types ranging from sb to sd .
the sample selection , observations , and data reduction are presented in detail in ( * ? ? ?
* ; * ? ? ?
* hereafter g06 and g09 ) .
the galaxies all have imaging data from the hubble space telescope ( hst ) including either wfpc2 and/or nicmos data @xcite .
all targets were selected to be brighter than @xmath14 according to the values listed in the @xcite catalogue , where interacting and seyfert galaxies were discarded .
the sample also excludes targets that are inaccessible to the 4.2-m william herschel telescope ( [ ss : sauronifs ] ) , so only galaxies with 0 @xmath15 ra @xmath16 and @xmath17@xmath18 have been selected . in table
[ tab : prop ] we list the main properties of the 18 galaxies , combined from the literature and our own measurements . their morphological type ranging between sb and sd , together with the galactocentric distance ( d ) , are taken from g06 using nasa / ipac extragalactic database ( ned ) .
the photometric position angle ( pa ) is measured by g09 using digital sky survey ( dss ) images , but the typical uncertainty of 5@xmath18 - 10@xmath18 comes from comparing the measured values of the pa in g06 to literature values compiled from the third reference catalog of bright galaxies ( rc3 ) . in the same way , we estimated the typical uncertainty of 15% for the ellipticity @xmath6 , which we also take from g06 to compare to the rc3 catalogue . from the axis ratio of the galaxies ( @xmath19 ) ,
we calculate the photometric inclination ( _ i _ ) of the galaxies using @xmath6 from table [ tab : prop ] in the following way ( @xcite ) : @xmath20 here @xmath21 is the intrinsic axis ratio of the galaxy ( e.g. @xcite ) and we adopt @xmath21 = 0.2 as commonly used ( e.g. , @xcite ) .
the change in the value of @xmath21 slightly affects our estimation of the inclination ( 1@xmath183@xmath18 ) . the total uncertainty of the inclination , considering the errors of @xmath22 and @xmath21 can reach 5@xmath18 . in column 6 of table [ tab
: prop ] , we present the inclinations we adopted throughout our entire analysis .
note here that the listed inclinations of the galaxies ngc3346 , ngc4775 and ngc5668 are slightly larger than the corresponding photometric values calculated in eq .
( [ eq : incl ] ) by 7@xmath18 , 1@xmath18and 2@xmath18 , respectively . for these three galaxies
, the estimated photometric inclination is below the minimum allowed mge inclination ( @xmath23 ) of the jam approach ( see sec .
[ ss : jam ] ; eq.[eq : ilim ] ) , and thus we adopt @xmath24 . nevertheless , the difference between the adopted and photometric inclination for these three galaxies is generally within the uncertainty of the photometric inclination , and does not affect our results .
we wish to keep a constant inclination through this analysis to focus on the differences between the jeans and adc approaches .
further , we measure the systemic velocity ( @xmath7 ) using the enforced point - symmetry method with typical uncertainty of 1 km s@xmath5 ( see sec .
[ ss : stellarkinematics ] ) and determine the galaxy effective radius @xmath25 ( half - light radius ) after applying a multi - gaussian expansion method ( sec .
[ ss : surfacebrightness ] ) to the surface brightness profiles of the galaxies . using chi - squared values of the fit
, we estimate that the typical @xmath26 uncertainty of the effective radius does not exceed 10% .
the radial extent @xmath9 of the stellar kinematic data is estimated via @xmath27 package ( @xcite ) , yeilding a typical uncertainty for our galaxies of 1% .
the disk scale length ( @xmath10 ) and effective radius of the bulge ( @xmath11 ) are taken from g09 including their typical uncertainties of 5% and 15% , respectively .
we then calculate the radial extent ( @xmath28 ) of the stellar kinematic data in terms of disk scale lengths with an associated typical uncertainty of 5 % ( see table [ tab : prop ] ) .
the sample galaxies were observed with the integral - field unit ( ifu ) spectrograph ` sauron ` at the 4.2-m william herschel telescope of the observatorio del roque de los muchachos on la palma , spain .
the ` sauron ` ifu @xcite has a @xmath29 field - of - view ( fov ) , sampled by an array of @xmath30 square lenses .
the fov corresponds to a typical radial extent of 1/5 to 1/3 of the galaxy s half - light radius ( @xmath25 ) .
the spectral resolution is @xmath4 4.2 @xmath31 ( fwhm ) , corresponding to an instrumental velocity dispersion of @xmath32 kms@xmath5 in the observed spectral range 4800 - 5380 @xmath31 ( at 1.1 @xmath31 per pixel ) .
this range includes a number of absorption features fe , mgb and h@xmath33 , which we use to measure the stellar kinematics .
emission lines in this range , such as [ oiii ] , [ ni ] and h@xmath33 , can be used to probe the ionised gas properties ( g06 ) .
the observations were reduced by g06 using the dedicated software @xmath34 @xcite . to obtain a sufficient signal - to - noise ratio ( s / n ) , we spatially binned the data cubes using the voronoi 2d binning algorithm of @xcite .
we created compact bins with a minimum s / n @xmath35 per spectral resolution element . in the central regions many individual spectra
have s / n @xmath36 60 and thus remained un - binned .
we parametrize the light distribution of the galaxies with the surface brightness profiles obtained by g09 .
they used archival ground - based @xmath37-band images from two - micron all sky survey ( hereafter , 2mass ) complemented with near - infrared hst nicmos / f160w images for 11 cases and optical hst wfpc2/f814w for the remaining 7 galaxies ( ngc1042 , ngc2805 , ngc3346 , ngc3423 , ngc4487 , ngc4775 , ngc5668 ) .
dss data were used for the outer parts of the same galaxies to obtain an accurate determination of the sky level and the galaxy disc geometry .
g09 derived the light distribution parameters using the ` ellipse ` task in imaging reduction and analysis facility ( iraf ) .
they first fitted elliptical isophotes to galaxy images with the centre , position angle and ellipticity left as free parameters in order to obtain the centre coordinates .
g09 then fit again ellipses to the images with the centre fixed , where the position angle and ellipticity were set as free parameters .
in the end , there were three photometric profiles and their combination gave a single near - infrared @xmath37-band profile with the maximum extent and inner spatial resolution allowed by the data .
the error introduced by this combination of optical and infrared images was negligible ( g09 , sec.3.3 ) and does not affect our analysis .
for most of the 18 late - type spiral galaxies , the ` sauron ` stellar mean velocity and velocity dispersion maps are , within the observational uncertainties , consistent with axisymmetry .
the non - axisymmetric features appear to be mainly due to dust obscuration . in half of the galaxies
, the effect of bars seems to be weak or more dominant in the outer parts .
these trends support the assumption of a stationary axisymmetric stellar system for the inner parts of the galaxies that we study here .
the methodology of @xcite has proven to be powerful in building detailed models for spherical , axisymmetric ( e.g. * ? ? ?
* ; * ? ? ?
* ) , and triaxial nearby galaxies @xcite , as well as globular clusters ( e.g. * ? ? ?
* ; * ? ? ?
the method finds the set of weights of orbits computed in an arbitrary gravitational potential that best reproduces all available photometric and kinematic data at the same time .
since higher - order stellar kinematic measurements are necessary to constrain the large freedom in this general modelling method , a less computationally intensive approach has been to construct simpler but still realistic
dynamical models based on the solution of the axisymmetric jeans equations ( @xcite ) .
we adopt the latter approach here . in the case of steady state axisymmetry both the potential
@xmath38 and distribution function ( df ) are independent of azimuth @xmath39 and time . by jeans theorem @xcite
the df depends only on the isolating integrals of motion : @xmath40 , with energy @xmath41 , angular momentum @xmath42 parallel to the symmetry @xmath43-axis , and a third integral @xmath44 for which in general no explicit expression is known .
however , @xmath44 usually is invariant under the change @xmath45 , though @xmath44 may loose this symmetry if resonances are present .
such symmetry implies that the mean velocity is in the azimuthal direction ( @xmath46 ) and that the velocity ellipsoid is aligned with the rotation direction ( @xmath47)20 degrees ( @xcite ) , and these vertex deviations may also correlate with morphological features ( @xcite ) ] .
when we multiply the collisionless boltzmann equation in cylindrical coordinates by @xmath48 and @xmath49 respectively and integrate over all velocities , we obtain two of the jeans equations @xcite , @xmath50 where @xmath51 is the intrinsic luminosity density has the same meaning as the luminosity density @xmath52 in @xcite ] . due to the assumed axisymmetry ,
all terms in the third jeans equation , that follows from multiplying by @xmath53 , vanish . the _ jeans axisymmetric model _ ( jam )
is based on solving the above two jeans equations and .
however , since there are four unknown second - order velocity moments @xmath54 , @xmath55 , @xmath56 and @xmath57 this is an underconstrained system , and we have to make two assumptions about the velocity anisotropy , or in other words about the shape and alignment of the velocity ellipsoid .
first , we assume the velocity ellipsoid is aligned with the cylindrical @xmath58 coordinate system , which gives @xmath59 .
we can then readily solve equation for @xmath55 .
second , we also assume a constant flattening of the velocity ellipsoid in the meridional plane , so we can write @xmath60 and solve equation for @xmath56 .
@xcite argue that this second assumption provides a good , general description for the kinematics of real disc galaxies .
when @xmath61 , the velocity distribution is isotropic in the meridional plane , corresponding to the well - known case of a two - integral distribution function @xmath62 ( e.g. * ? ?
* ; * ? ? ?
. given the intrinsic second - order velocity moments , we can then calculate the observed second - order velocity moment by integrating along the line - of - sight through the stellar system viewed at an inclination @xmath63 away from the @xmath43-axis : @xmath64\ , { \mathrm{d}}z',\end{aligned}\ ] ] where @xmath65 is the ( observed ) surface brightness with the @xmath66-axis along the projected major axis . for each position
@xmath67 on the sky - plane , @xmath68 yields a prediction of the ( luminosity weighted ) combination @xmath69 of the ( observed ) mean line - of - sight velocity @xmath70 and dispersion @xmath71 . under the above assumptions , the only unknown parameters are the anisotropy parameter @xmath72 , the inclination @xmath73 , and the gravitational potential @xmath38 , which is related to the total mass density @xmath74 through poisson s equation . to estimate @xmath74
, we derive the intrinsic luminosity density @xmath51 by deprojecting the observed surface brightness @xmath65 .
we then assume a total mass - to - light ratio @xmath75 to derive @xmath76 .
dynamical studies of the inner parts of galaxies typically consider @xmath75 as an additional parameter and further assume its value to be constant , i.e. , mass follows light ( e.g. * ? ? ?
since @xmath75 may be larger than the stellar mass - to - light ratio @xmath77 , this assumption still allows for possible dark matter contribution , albeit with a constant fraction .
a convenient way to arrive at @xmath38 is through the multi - gaussian expansion method ( mge ; * ? ? ?
* ; * ? ? ?
* ) , which models the observed surface brightness as a sum of @xmath78 gaussian components , @xmath79 \right\},\ ] ] each with three parameters : the central surface brightness @xmath80 , the dispersion @xmath81 along the major @xmath66-axis and the flattening @xmath82 .
the mge approach has several advantages . even though gaussians do not form an orthogonal set of functions , surface density distributions
are accurately reproduced . when the point spread function ( psf ) of the instrument is also represented as a sum of gaussians ,
the convolution with the psf becomes straightforward .
given the viewing direction , the mge parametrization can be deprojected analytically into an intrinsic luminosity density @xmath51 .
furthermore , the calculation of @xmath68 in equation reduces from the ( numerical ) evaluation of a triple integral to a straightforward single integral ( * ? ? ? * eq .
similarly , the gravitational potential @xmath38 can be calculated by means of one - dimensional integral ( * ? ? ?
* eq . 39 ) .
given the latter , the circular velocity from the jam model in the equatorial plane then follows upon ( numerical ) evaluation of @xmath83 where @xmath84 and @xmath85 are the total luminosity and the mass - to - light ratio of the @xmath86th gaussian .
the intrinsic dispersion and flattening , @xmath87 and @xmath88 , are related to their observed quantities , as @xmath89 with inclination @xmath73 , ranging from @xmath90 for face - on viewing to @xmath91 for edge - on viewing .
note here that the oblate mge deprojection is valid if @xmath92 for all gaussians ( @xcite ) .
therefore , the flattest gaussian in the mge fit defines the minimum possible inclination ( @xmath23 ) for which the mge model can be applied within the jam approach . in what follows , we assume mass follows light , so that the mass - to - light ratio is the same for each gaussian : @xmath93 .
we fix the galaxies inclinations to the values shown in table [ tab : prop ] , and we are left with two free parameters in jam : the total or dynamical mass - to - light ratio @xmath75 and the velocity anisotropy @xmath94 . for each galaxy , we obtain the values of these two parameters by fitting the observed second - order velocity moment @xmath69 , after which the jam circular speed follows from equation . in this work , we compare the jam - derived circular speed curves to those from the commonly adopted adc method . here
we relate the adc approach back to the jeans equations . instead of solving both jeans equations and
, we can instead evaluate the first equation in the equatorial plane ( @xmath95 ) and use @xmath96 and @xmath97 ( by symmetry ) to rewrite eq . as @xmath98 .\end{gathered}\ ] ] here , @xmath99 is the intrinsic mean velocity , and @xmath100 , @xmath101 are ( the square of ) the intrinsic mean velocity dispersions . in case of a dynamically cold tracer such as neutral gas , observed through i and
co emission at radio wavelengths , the mean velocity is typically much larger than the velocity dispersion ( @xmath102 ) , so that the circular velocity can be inferred from the deprojected observed rotation , @xmath103 ( e.g. , @xcite )
. however , the velocity dispersion can be non - negligible or even dominant for dynamically hot tracers like stars where it is possible that @xmath104 . in this case , the asymmetric drift must be taken into account since the observed rotation only captures part of the circular velocity ( e.g. , @xcite ) . from equation , this asymmetric drift correction depends on the two intrinsic velocity dispersions @xmath105 and @xmath106 , plus the cross term @xmath57 , which together define the velocity ellipsoid of the galaxy . to allow for a direct comparison with the jam model
, we adopt the same assumption of alignment of the velocity ellipsoid with the cylindrical coordinate systems , so that @xmath107 .
assuming that the velocity ellipsoid is furthermore symmetric around @xmath108 , it follows from appendix a from @xcite is efectively the epicycle approximation , which is applied only in adc and not in jam .
] of @xcite that @xmath109 with @xmath110 the radial logarithmic gradient of the intrinsic mean velocity : @xmath111 . under the assumption of a thin disc with @xmath112 ,
the circular velocity from the adc approach in the equatorial plane then reduces to @xmath113.\end{gathered}\ ] ] this final equation provides an estimate of @xmath114 that uses : ( 1 ) the surface brightness profile @xmath115 from the mge parametrization , ( 2 ) @xmath99 and @xmath110 from the observed mean line - of - sight velocity @xmath70 , and ( 3 ) @xmath105 from the observed line - of - sight velocity dispersion @xmath71 .
the latter two inferences also involve the velocity anisotropy @xmath116 and inclination @xmath73 as described in sec .
[ ss : radialprofiles ] below .
here , we describe the application of our model analyses to the observational data . in sec .
[ ss : surfacebrightness ] , we discuss the multi - gaussian expansion model adopted to the photometry of the ` sauron ` galaxies in order to derive a smooth , analytic representation of their surface brightness .
we explain the extraction of the stellar kinematic maps in sec .
[ ss : stellarkinematics ] .
the performance of the markov chain monte carlo analysis to the jam and adc models is shown in sec .
[ ss : mcmc ] .
adc holds a thin disc assumption , whereas jam does a full line - of - sight integration through the luminosity distribution to model the observed velocity moments . to infer the mass distributions of the 18 ` sauron ` galaxies
, we use the photometric bulge - disc decompositions of the total h - band surface brightness profile , as presented in @xcite and g09 ( see also sec .
[ ss : nirimaging ] ) .
these models represent each galaxy s profile as the superposition of an exponential disc and a srsic bulge .
the galaxies in our sample can contain a significant amount of interstellar dust , which is mostly transparent at near - infrared wavelengths .
the extinction @xmath117 is roughly a factor of 8 lower than the extinction @xmath118 @xcite , so we do not apply an internal extinction correction .
additionally , the near - infrared is considered as a good tracer of the stellar mass in galaxies as the light is dominated by old stars @xcite , although there is some concern that this might be not true due to uncertain influence of intermediate age populations ( @xcite ) .
@xcite and g09 use a one - dimensional mge decomposition @xcite of the bulge and disk profiles , using the implementation of @xcite .
they adopt 10 gaussians to represent the discs and between 13 and 19 gaussians to represent the bulges .
the fits to the one - dimensional profiles are excellent for all galaxies , but the mge is not applied to the two - dimensional images .
most of the galaxies ( 16 out of 18 ) display a clear light excess above the srsic fit to the bulge , which can be attributed to a bright nuclear star cluster @xcite , where a luminous mass contributes to the kinematics in the inner regions .
for these galaxies , we amend the mge decomposition of ( * ? ? ?
* chapter 5 ; sec.5.4 . ) by including one additional , circular gaussian to accurately represent the light excess in the centre .
figure [ fig : mge ] shows the mge light models of two representative galaxies in our sample .
the green asterisks indicate their observed light profiles .
the black thick curves represent the sum of the individual gaussians ( dotted lines ) of the three components : nucleus ( yellow ) , srsic bulge ( red ) , and exponential disc ( blue ) , which are marked in appendix [ a : mgetab ] with index of 0 , 1 and 2 , respectively .
ngc628 ( left panels ) is an example of a very good fit to the data , which is typical for the majority of our galaxies .
however , there are a few exceptions with mismatches due to bars or prominent spiral arms
. these are ngc864 ( figure [ fig : mge ] , right panel ) , ngc772 , and ngc1042 . nevertheless , we considered the mge fits to these three galaxies to be satisfactory for our needs , because fitting non - symmetric features could lead to uneven representation of the galaxies surface brightness profiles , and hence , gravitational potentials .
we converted the resulting peak surface brightnesses of the gaussians into physical units of @xmath119 , taking the absolute magnitude of the sun @xmath120 @xcite .
we measured the stellar kinematics using the penalised pixel - fitting ( ppxf ) method of @xcite . for spectral templates , we used a sub - sample of the miles stellar library @xcite , containing @xmath4115 stars that span a large range in atmospheric parameters such as surface gravity , effective temperature and metallicity .
we convolve the miles models of stars from their original spectral resolution to that of the data .
this is done by convolving with a gaussian with dispersion equal to the difference in quadrature between the final and starting resolution .
the ppxf method fits a non - negative linear combination of these stellar template spectra , convolved with the gaussian velocity distribution , to each galaxy spectrum by chi - square minimisation .
the higher - order gauss - hermite moments h3 and h4 are not included in this fit as free parameters .
the spectral regions affected by emission lines were masked out during this process .
a low - order polynomial ( typically sixth degree ) is included in the fit to account for small differences in the flux calibration between the galaxy and the template spectra .
this analysis yields the stellar mean velocity ( @xmath70 ) and stellar velocity dispersion ( @xmath71 ) for each bin .
their errors are estimated through monte - carlo simulations with noise added to the galaxy s best fitting model spectrum .
the ` sauron ` instrumental resolution is 105 kms@xmath5while the measured velocity dispersions of our galaxies can be significantly below that level .
@xcite used monte - carlo simulations to show that the intrinsic velocity dispersion is still well recovered . for a spectrum with signal - to - noise of @xmath121 and @xmath122 kms@xmath5 , the ppxf method will give velocity dispersions differing from the intrinsic ones by @xmath123 kms@xmath5 , consistent with the measured errors .
we might even expect larger uncertainties for some of our galaxies ( e.g. , ngc3346 , ngc4487 and ngc5585 ) , which own central velocity dispersions below this limit . on the other hand , recent papers put forward evidence that our adopted approach could accurately recover the intrinsic velocity dispersion to better than 10% precision given the simulations of @xcite and @xcite and our high s / n ( @xmath124 ) in each voronoi bin . in fig .
[ fig : maps1 ] we show the stellar kinematic maps of our sample of 18 late - type spiral galaxies resulting from the ppxf fits .
the first column shows the stellar flux in arbitrary units on a logarithmic scale . the next two columns display @xmath70 and @xmath71 maps in kms@xmath5 respectively .
we over - plotted the surface brightness contours of the galaxies as derived from their intensity maps .
for some of our galaxies , the centre position was not accurately determined during the data reduction due to foreground stars and dust lanes , which in some cases caused a significant offset in the measurement of the systemic velocity v@xmath125 .
therefore , we use the velocity field symmetrisation method described in appendix a of @xcite to estimate a robust v@xmath125 .
this method assumes that the velocity field ( 1 ) is symmetric with respect to the galaxy centre , ( 2 ) is uncorrelated and ( 3 ) varies linearly along the spatial coordinates .
then , for each position that has a counterpart , it computes their weighted mean velocities and combined errors ( rejecting data with errors @xmath126 km s@xmath5 ) . in this way
we obtain a robust estimate of v@xmath125 for each galaxy ( see table [ tab : prop ] ) .
galaxies with more concentrated light distributions generally have larger central peaks in their velocity dispersion fields .
elliptical galaxies usually have radially decreasing @xmath71 @xcite , and this is also the case for many early - type spirals , as a result of a centrally concentrated bulge .
but for very late - type spirals , we expect the @xmath71 field to be either flat ( due to their lower bulge - to - disc ratios ) or with a central decrease , because of cold components or counter - rotating discs ( @xcite ) . in comparison with early - type galaxies ,
the bulges of the late - type spirals are smaller and have lower surface brightness ( * ? ? ?
@xcite also find that @xmath71 generally decreases with radius with an @xmath127-folding length that is twice the photometric disc scale length .
thus , the variety seen in @xmath71 profiles of ` sauron ` galaxies might be also due to their limited radial coverage , ranging between 0.4 and 1.2 scale lengths . to obtain reliable circular velocity curves and associated uncertainties , we assume these properties are random variables viewed in a bayesian framework .
we then applied markov chain monte carlo ( mcmc ) to estimate these variables under both the adc and jam model approaches .
we used the @xmath128 code of @xcite , an implementation of an affine invariant ensemble sampler for the mcmc method of parameter estimation , together with jam code ( @xcite ) and our own adc routines all of which are implemented in @xmath129 .
this approach allows a robust understanding of the uncertainties in the modelling combined with their dependance on assumptions , which is particularly important since we ultimately compare adc and jam results . )
= 1.5 mm & + ngc & type & @xmath130 & @xmath131 & @xmath132 & @xmath133 & @xmath134 & @xmath135 & good fit + ( 1 ) & ( 2 ) & ( 3 ) & ( 4 ) & ( 5 ) & ( 6 ) & ( 7 ) & ( 8) & ( 9 ) + ngc0488 & sa(r)b & @xmath136 & @xmath137 & @xmath138 & @xmath139 & @xmath140 & @xmath141 & yes + ngc0772 & sa(s)b & @xmath142 & @xmath143 & @xmath138 & @xmath144 & @xmath145 & @xmath146 & yes + ngc4102 & sab(s)b & @xmath147 & @xmath148 & @xmath138 & @xmath149 & @xmath150 & @xmath151 & yes + ngc5678 & sab(rs)b & @xmath152 & @xmath153 & @xmath138 & @xmath154 & @xmath155 & @xmath156 & yes + ngc3949 & sa(s)bc & @xmath157 & @xmath158 & @xmath138 & @xmath159 & @xmath160 & @xmath161 & yes + ngc4030 & sa(s)bc & @xmath162 & @xmath163 & @xmath138 & @xmath164 & @xmath165 & @xmath166 & yes + ngc2964 & sab(r)bc & @xmath167 & @xmath168 & @xmath138 & @xmath169 & @xmath170 & @xmath171 & yes + ngc0628 & sa(s)c & @xmath138 & @xmath138 & @xmath172 & @xmath173 & @xmath174 & @xmath175 & no + ngc0864 & sab(rs)c & @xmath176 & @xmath177 & @xmath138 & @xmath178 & @xmath179 & @xmath180 & yes + ngc4254 & sa(s)c & @xmath181 & @xmath182 & @xmath138 & @xmath183 & @xmath184 & @xmath185 & yes + ngc1042 & sab(rs)cd & @xmath138 & @xmath138 & @xmath186 & @xmath187 & @xmath188 & @xmath189 & yes + ngc3346 & sb(rs)cd & @xmath138 & @xmath138 & @xmath190 & @xmath191 & @xmath192 & @xmath193 & yes + ngc3423 & sa(s)cd & @xmath194 & @xmath195 & @xmath138 & @xmath196 & @xmath197 & @xmath198 & yes + ngc4487 & sab(rs)cd & @xmath138 & @xmath138 & @xmath199 & @xmath200 & @xmath201 & @xmath202 & yes + ngc2805 & sab(rs)d & @xmath203 & @xmath204 & @xmath138 & @xmath205 & @xmath206 & @xmath207 & yes + ngc4775 & sa(s)d & @xmath138 & @xmath138 & @xmath208 & @xmath209 & @xmath210 & @xmath211 & no + ngc5585 & sab(s)d & @xmath138 & @xmath138 & @xmath212 & @xmath213 & @xmath214 & @xmath215 & no + ngc5668 & sa(s)d & @xmath138 & @xmath138 & @xmath216 & @xmath217 & @xmath218 & @xmath219 & no + [ tab : adcmc ] to infer circular velocity curves for our sample of galaxies using the adc method ( sect .
[ ss : adc ] ) , we first need their kinematic profiles along the projected major axis . to this end , we use the ` kinemetry ` package of @xcite that is based on harmonic expansion of two - dimensional maps along ellipses .
we extract the observed mean velocity @xmath220 and velocity dispersion @xmath221 profiles along ellipses with fixed axis ratio of the galaxies ( @xmath19 ) and photometric position angle , where @xmath6 and @xmath222 are shown in table [ tab : prop ] . however , for galaxies ngc3346 , ngc4775 and ngc5668 , we calculate the axis ratio @xmath22 from the adopted inclination @xmath223 using eq .
[ eq : incl ] ( see sec . [
ss : sample ] ) . then under the thin - disc assumption
we obtain the intrinsic mean velocity @xmath99 and radial velocity dispersion @xmath105 profiles from the observed profiles as : @xmath224 , \end{aligned}\ ] ] with systematic velocity @xmath225 and adopted inclination @xmath73 from table [ tab : prop ] . to get to equation
, we have used @xmath109 under the assumptions that the velocity ellipsoid is aligned with the cylindrical coordinate system and symmetric around @xmath108 ( see section [ ss : adc ] ) . as before , @xmath226 is the velocity anisotropy in the meridional plane , and @xmath227 is the radial logarithmic gradient of the intrinsic mean velocity .
+ given the assumptions of the adc approach , we relate the model parameters to the observed data as follows : @xmath228 where @xmath229 and @xmath230 .
to avoid numerical noise in the derivatives , we express @xmath231 and @xmath105 using a smooth , analytic functional forms and then compute the derivatives explicitly . as an analytical representation for @xmath232
, we use the power - law prescription of ( * ? ? ?
* eq.2.11 ) , which in the equatorial plane ( @xmath233 ) and assuming flat rotation curve , becomes @xmath234 where @xmath235 is the asymptotic velocity and @xmath131 is the core radius .
this model describes a rotation curve that increases linearly with radius @xmath236 when @xmath237 , and flattens to the value @xmath235 when @xmath238 .
however , in several cases the velocity profile shows little indication of flattening the observed region . in such cases ,
the core radius @xmath131 is not well constrained and fitting eq .
would lead to unphysical values for both @xmath131 and @xmath235 .
such galaxies are in the regime of @xmath237 and we adopt a linear solid - body rotation instead : @xmath239 where @xmath132 is the linear coefficient . + for @xmath105 , we use a linear expression of the form @xmath240 where @xmath133 and @xmath134 are free parameters .
we find the linear fits to the @xmath105 profiles are broadly representative , although the model does not reproduce the central part of the @xmath105 profiles for some of the galaxies ( e.g. , ngc772 , ngc5678 and ngc864 ) due to the marked dips in @xmath71 in the central parts of the galaxies ( @xcite ) .
such a descrease in the observed stellar velocity dispersion might be the result of a small dynamically cold component or counter - rotating disc of high surface brightness that obscures the bulge , dominating the luminosity - derived kinematic measurements ( @xcite ) or due to @xmath127-folding length of our galaxies ( see @xcite ) . to apply a bayesian framework to the problem
, we consider the free parameters of the model to be random variables and then calculate their posterior distributions using mcmc .
we model 5 free parameters for the galaxies in the power - law model for @xmath232 ( @xmath130 , @xmath131 , @xmath133 , @xmath134 , @xmath135 ) and 4 free parameters ( @xmath132 , @xmath133 , @xmath134 , @xmath135 ) for the galaxies with linear models for @xmath232 . for these vectors of parameters @xmath241
, we compute model profiles for @xmath242 and @xmath243 using eq.([eq : obsvmc ] ) and ( [ eq : obssigmc ] ) .
these profiles are compared to observed data using a chi - squared statistic , summing over radius positions with normalization set by the uncertainties in the observed data ( @xmath244 and @xmath245 ) : @xmath246 ^ 2/(\delta v_{\mathrm{obs},k}^2)\nonumber\\ & + & \sum_k [ \sigma_{\mathrm{mod},k}(\boldsymbol\theta ) - \sigma_{\mathrm{obs},k}]^2/(\delta \sigma_{\mathrm{obs},k}^2 ) .
\label{eq : chisq}\end{aligned}\ ] ] we adopt a log - probability function for the parameters given the data as @xmath247 , where @xmath248 is a set of priors .
we adopt uniform prior distributions for our parameters on fixed ranges : @xmath249 $ ] km s@xmath5 , @xmath250 $ ] , @xmath251 $ ] km s@xmath5 arcsec@xmath5 , @xmath252 $ ] km s@xmath5 , @xmath253 $ ] km s@xmath5 arcsec@xmath5 and @xmath254 $ ] . only for one galaxy ( ngc2964 ) we had to expand the range of the walkers to @xmath255 $ ] , since the distribution peaked at @xmath256 .
we run the adc - mcmc code with 60 walkers ( i.e. , members of the ensemble sampler ) , which have initial position set by the prior information and physical information .
we use a small random distribution around the parameters : @xmath130 , @xmath131 , @xmath132 , @xmath133 , @xmath134 , which are adopted from the least - squares minimisation routine ` mpfit ` in idl to find the values that minimize eq . [ eq : chisq ] .
we then start the walkers around the optimized values in a normal distribution with scale determined by the uncertainties in the fit .
the walkers of the parameter @xmath135 were allowed to move freely in the parameter space . on the other hand ,
the inclination of the galaxies is fixed to the value presented in column 6 of table [ tab : prop ] .
the mcmc code samples the posterior distribution , using 400 steps for burn - in and 1000 steps for sampling .
we find that all walkers converge after @xmath257 steps of burn - in to sample a similar distribution indicating a well - sampled unimodal posterior distribution .
we summarize the posterior distributions for the parameters in table [ tab : adcmc ] of this paper . additionally , we assign a quality flag to each adc - mcmc model in table [ tab : adcmc ] depending on the residual of the v profile ( @xmath258 ) and @xmath71 profile ( @xmath259 ) , where the label `` yes / no '' correspond to `` good / bad '' fit , respectively .
bad fit label is associated to galaxies ( i.e. , ngc0628 , ngc4775 , ngc5585 and ngc5668 ) if the median value of one of their residuals ( @xmath260 or @xmath261 ) is larger than 0.3 . the burn - in chains , posterior chains and corner plots of adc - mcmc method are shown in appendix [ c : chains ] .
we present the results of the analysis in fig .
[ fig : vc_c ] . in the first column
, we show the observed mean line - of - sight velocity @xmath70 ( filled green circles ) and its fiting function ( eq . [ eq : obsvmc ] , filled magenta triangles ) .
the galaxies are ordered by their morphological type from sb to sd .
we observe that sb sc galaxies ( at @xmath262 ) have higher observed velocity in contrast to scd sd .
we adopt a power - law model function to most of the @xmath232 radial profiles , except for the galaxies ngc0628 , ngc1042 , ngc3346 , ngc4487 , ngc4775 , ngc5585 and ngc5668 , which require a linear function . in the second column of fig . [ fig : vc_c ] , we present the observed line - of - sight velocity dispersion @xmath71 profiles ( filled green circles ) and its fiting function ( eq . [ eq : obssigmc ] , filled magenta triangles ) .
the @xmath71 profiles are more varied than the @xmath70 profiles .
some velocity dispersion profiles show an almost - linear decrease or increase with radius on the whole radial range available , some are flat , and others have a more complex behaviour that seems difficult to reproduce with a linear function .
sbc galaxies ( at @xmath262 ) are characterised by high observed velocity dispersion @xmath71 with respect to sc
most of our sb - sc galaxies have regular velocity fields with well - defined axisymmetric rotation and high amplitude , while sd velocity fields show more complex structure and lower amplitude of rotation . in our sample
we have four types of @xmath71 behaviour : decreasing outward ( ngc4102 ) , increasing outward ( ngc3949 ) , a central @xmath71-dip ( ngc0772 ) or flat ( ngc4775 ) .
we estimate the uncertainty of the velocity and velocity dispersion profiles at each radius ( see fig .
[ fig : vc_c ] ) from the corresponding error maps since the provided uncertainties from ` kinemetry ` routine are unrealistically small ( @xmath4 1 km@xmath5 ) .
thus , we took the absolute value of the median of the velocity bins and the median of the velocity dispersion bins within each annulus defined by ` kinemetry ` routine .
as described in sect .
[ s : method ] , we use a solution of the axisymmetric jeans equations based on a multi - gaussian expansion ( mge ) of the intrinsic luminosity density to predict the observed second velocity moment @xmath263 ( @xcite ) .
the solution has a constant meridional plane velocity anisotropy @xmath264 and constant mass - to - light ratio @xmath265 within the galaxy .
our analysis does not account for the variation of @xmath266 within each voronoi bin .
such variations will cause the averaged @xmath267 in a bin to be larger than the intrinsic @xmath266 along a ray , as evaluated by the jam code .
we do not anticipate this effect to bias our results significantly : since the surface brightness decreases with @xmath268 , the larger voronoi bins are found in the outer regions of the galaxies where the variation in the @xmath266 surface is low . at small @xmath268 , where the variation with position can be significant
, the higher surface brightness means the bins are smaller .
we symmetrize our observed @xmath266 fields in order to avoid outliers in the data ( see @xcite ) . in this way ,
the results will be not influenced by a few deviant values .
this is especially important in the case of the mcmc method application , where the best fit @xmath266 model rely on @xmath269-statistics .
moreover , the symmetrization renders the @xmath266 field axisymmetric which is the approximation consider by jam . at the same time we symmetrize the variance of the @xmath266 field : @xmath270 where the term @xmath271 indicates the symmetrized variance of @xmath266 field . + for this purpose we use the code presented in @xcite .
the procedure interpolates the bin values over a new grid generated by permuting the @xmath86th - bin coordinates as follow : @xmath272 this operation generates three additional velocity fields ( the first permutation obtained by @xmath273 corresponds to the original velocity field ) .
finally , the symmetrized velocity field is the mean of the four coordinate - permuted velocity fields .
given this , equation ( [ eq : dvrms ] ) reduces to the common addition of the error in quadrature : @xmath274 where @xmath275 , and + @xmath276 of a certain permutation .
+ although the interpolation is precise , the code does not consider the systematic effect that a given bin at @xmath277 position not always have a corresponding bin atat the other positions , e.g. , ( -x , y ) . to account for this
, the final @xmath266 field error is set by the maximum of either the values from equation ( [ eq : dvrms ] ) or half the deviation between neighbors @xmath266 field bin values ( m. cappellari private communication , see also @xcite ) .
+ we formulate a parallel mcmc approach for determining the posterior parameter distributions of the jam model . in this case
, there are only 2 free parameters for all galaxies ( @xmath264 and @xmath75 ) .
because of the large computational expense of evaluating a jam model , we run the jam - mcmc code with only 30 walkers , though the model still converges using 100 steps for a burn - in and 200 steps for sampling .
the prior distributions are taken to be uniform with , e.g. , @xmath278 $ ] and @xmath279~m_{\odot}/l_{\odot}$ ] . in jam - mcmc method
, we fix the inclination to the values in the column 6 of table [ tab : prop ] . for galaxies
ngc3346 , ngc4775 , and ngc5668 , we adopt @xmath280 regarding to the minimum possible inclination condition of jam model ( see sec .
[ ss : jam ] ; eq .
[ eq : ilim ] ) .
we use a similar log - likelihood statistic as we did for the adc , except in this case , only the model and observed @xmath266 are compared .
summaries of the posterior distribution of parameters for jam - mcmc model are presented in table [ tab : jammc ] .
the burn - in chains , posterior chains and corner plots are shown in appendix [ b : chains ] . in the fourth to fifth columns of fig .
[ fig : maps1 ] , we show the observed and the best fitting @xmath266 maps with over - plotted mge contours . for all galaxies , the mge contours are aligned with the surface brightness contours in position angle and ellipticity . in the sixth colum
, we calculate the residual maps @xmath281 between the observations ( @xmath282 ) and the models ( @xmath283 ) , where @xmath284=@xmath285 .
the median values of these residual maps @xmath286 indicate that the uncertainties of jam fit ranges between 3 % and 10 % .
note here that we do not count for systematics errors in these uncertainties . from the posterior distributions of jam - mcmc method
, we define the goodness of the fit with the label `` yes / no '' for `` good / bad '' fit , respectively .
bad fit labels are associated to galaxies ( i.e. , ngc4030 , ngc1042 , ngc3346 ) , which do not have well defined posterior distributions of the velocity anisotropy @xmath287 ( see appendix [ b : chains ] and table [ tab : jammc ] ) .
further , we find that the jam produces the best fits to the second velocity moment @xmath267 for sb and sc spirals ( e.g. , ngc488 , ngc5678 , and ngc2964 , characterised by fast - rotating discs ) and the model gets worse for scd and sd galaxies . given the observed values of @xmath70 and @xmath71 show evidence for non - monotonic behaviour ( figure [ fig : maps1 ] ) , the model provides an average representation of the data in the late - type galaxies .
we stress that the values derived for the parameters are the posterior distributions given the data , subject to the assumptions of the model .
= 1.5 mm & + ngc & type & @xmath75 & @xmath287 & good fit + ( 1 ) & ( 2 ) & ( 3 ) & ( 4 ) & ( 5 ) + 0488 & sa(r)b & @xmath288 & @xmath289 & yes + 0772 & sa(s)b & @xmath290 & @xmath291 & yes + 4102 & sab(s)b & @xmath292 & @xmath293 & yes + 5678 & sab(rs)b & @xmath294 & @xmath295 & yes + 3949 & sa(s)bc & @xmath296 & @xmath297 & yes + 4030 & sa(s)bc & @xmath298 & @xmath299 & no + 2964 & sab(r)bc & @xmath300 & @xmath301 & yes + 0628 & sa(s)c & @xmath302 & @xmath303 & yes + 0864 & sab(rs)c & @xmath304 & @xmath305 & yes + 4254 & sa(s)c & @xmath306 & @xmath307 & yes + 1042 & sab(rs)cd & @xmath308 & @xmath309 & no + 3346 & sb(rs)cd & @xmath310 & @xmath311 & no + 3423 & sa(s)cd & @xmath312 & @xmath313 & yes + 4487 & sab(rs)cd & @xmath314 & @xmath315 & yes + 2805 & sab(rs)d & @xmath316 & @xmath317 & yes + 4775 & sa(s)d & @xmath318 & @xmath319 & yes + 5585 & sab(s)d & @xmath320 & @xmath321 & yes + 5668 & sa(s)d & @xmath322 & @xmath323 & yes + [ tab : jammc ]
in order to compare the two modelling approaches adc and jam , we first combine the measured deprojected stellar velocity profile @xmath231 and radial dispersion profile @xmath105 .
next , we derive the circular velocity curve @xmath324 using the asymmetric drift correction ( adc ) formula in equation ( [ eq : adcvcirc ] ) .
these data are indicated by the blue - filled - pentagon curves with uncertainties derived from the transformed posterior distributions of adc - mcmc results .
second , we construct axisymmetric jeans models ( jam ) that fit the combined observed stellar mean velocity and velocity dispersion fields , we use equation ( [ eq : mgevcirc ] ) to obtain the circular velocity curve for each galaxy . the resulting @xmath325 curves from the best - fit jam model are plotted as red - filled - squares curves in fig .
[ fig : vc_a ] with their uncertainty from jam - mcmc code . for all galaxies ,
adc approach gives lower measurements of the circular velocity curve in comparison with the values obtained from jam models for at least part of the radial range . to quantify these differences , we plot in fig .
[ fig : resratio ] the velocity ratio between @xmath326 and @xmath327 .
overall , the velocity discrepancy for sb - sbc galaxies is larger in the inner parts ( @xmath328 ) and decreases in the outer parts ( @xmath329 ) , while for scd - sd galaxies the velocity discrepancy is constant or even increases in the outer parts .
the differences we found between the circular velocity curves from the two modelling approaches can have a strong impact on the inferred total mass distribution and thus also on any follow - up inference like the amount of dark matter in the inner parts of these late - type spiral galaxies .
we list the different assumptions adopted in both the axisymmetric jeans models ( jam ) and the asymmetric drift correction ( adc ) , and discuss the possible reasons for their discrepancy .
_ mass - follows - light ( jam ) : _ within the small radial range covered by the stellar kinematics , typically at @xmath330 , we expect the variations in the mass - to - light ratio to be small .
@xcite provides evidence that in a sample of 37 sb - sc galaxies , most of their rotation curves computes from the luminosity profiles assuming constant mass - to - light ratio provide a good match to the observed curve out to the radius where the predicted curve turns over .
additionally , @xcite investigate the effects of realistic radially varying mass - to - light ratios and find the overall effect to only be @xmath4 10% variations on the derived kinematic properties within @xmath331 .
_ constant anisotropy in the meridional plane ( jam ) : _ whereas the velocity anisotropy @xmath332 in the equatorial plane is inherent in the solution of the axisymmetric jeans equations ( section [ ss : axijeans ] ) , the velocity anisotropy @xmath333 in the meridional plane is a free parameter , given as @xmath334 .
there is little evidence for strong variation of @xmath72 .
for example , @xcite already argued that for spirals the @xmath333 is constant at approximately 0.6 , close to the measured value of @xmath335 in the solar neighbourhood @xcite .
this ratio of the anisotropy is measured in a few spirals of type sa to sbc @xcite , yielding slightly larger constant values between 0.6 and 0.8 .
based on long - slit spectra for a sample of 17 edge - on sb scd spirals , @xcite also adopt constant values , although slightly lower : 0.5 to 0.7 . while radial variation of the anisotropy is not excluded , a constant value of @xmath72 should be sufficient for our analysis .
_ shape of the velocity ellipsoid ( jam , adc ) : _ the adc and jam models in our study assume that the velocity ellipsoid in the meridional plane is aligned with the cylindrical coordinate system so that @xmath107 . additionally , the adc approach may allow for a tilt of the velocity ellipsoid through the parameter @xmath336 having intermediate values between @xmath337 and @xmath338 , corresponding to the cylindrical and spherical coordinate system , respectively .
however , we expect that the resulting circular velocity curve is only weakly dependent on this tilt , because most of the stellar mass is concentrated toward the equatorial plane , particularly for late - type spiral galaxies . in that case
, the assumption of axisymmetry gives @xmath107 .
_ dust ( jam , adc ) : _ the surface brightness distribution of the spiral galaxies that is used in both jam and adc modelling approaches can be strongly affected by extinction due to dust .
we have tried to minimise the effects of dust in various ways ( g09 ) : ( i ) selecting galaxies with intermediate inclinations so that they are not edge - on ( where dust extinction is strongest ) or face - on ( where the stellar velocity dispersion would be significantly below the spectral resolution ) , ( ii ) inferring the surface brightness distribution from images in the near - infrared where the extinction is significantly smaller than in the optical , and ( iii ) fitting smooth , analytical mge profiles to the radial surface brightness profile after azimuthally averaging over annuli to suppress deviations caused by bars , spiral arms , and regions obscured by dust .
the stellar kinematics are obtained from integral - field spectroscopy in the optical and thus could also be affected if the ( giant ) stars that contribute along the line - of - sight with different motions are affected by the dust in different ways . for example , if dynamically colder stars closer to the disc plane are relatively more obscured than dynamically hotter stars above the disc plane , the resulting combined ordered - over - random motion could be biased to lower values .
the effects of dust are strongly reduced for lines of sight only a few degrees away from edge - on ( @xcite ) and the effects of extinction on the line - of - sight velocity are negligible : the change in @xmath339 is only 1.3% higher from the dustless case ( @xcite ) . given the selection in our sample , the effects of dust on the inferred circular velocity curves from both modelling approaches are expected to be minimal . _
thin - disc ( adc ) : _ circular velocity curves of spirals nearly always come from ( cold ) gas , which is naturally in a thin disc .
the stellar discs of these late - type spiral galaxies are also believed to be thin , with inferred intrinsic flattening @xmath340 ( e.g. , * ? ? ?
even the bulges in late - type spiral galaxies are different from the classical bulges in lenticular galaxies .
srsic profile fits to their surface brightness ( @xmath341 ) show that towards later - types , the bulges are smaller in size , have profiles closer to exponential ( @xmath342 ) than de vaucouleurs ( @xmath343 ) , and are flatter ( e.g. , g09 ) .
hence , the thin - disc assumption adopted in the adc approach also seems to be reasonable .
independent of the nature of the dynamically hot stellar ( sub)system , it seems that the local value of the ordered - over - random motion is the key for understanding of the adc and jam discrepancy . to explore that , in fig .
[ fig : resratio ] , we plot the radial velocity ratio @xmath344 of the galaxies , as well as the local luminosity fraction of the fitted exponential disc compared to the total luminosity as a thin solid curve . for the sb sbc galaxies , the discrepancy in the estimated velocity ratio between the jam and adc models seems to be larger in the inner parts ( @xmath328 ) , where the luminosity is dominated by the presence of the bulge , and decreases outwards . for scd
sd galaxies the velocity ratio stays roughly constant and in a few cases increases towards larger radii , which could be due to the presence of a thick(er ) disc component . in the left panel of fig .
[ fig : vsrel ] , we plot the velocity ratios @xmath344 obtained from fig.[fig : resratio ] for all galaxies as measured at different radii versus the ordered - over - random motion @xmath1 ( the deprojected rotation vs. radial velocity dispersion ) at the same radii . in the right panel of fig .
[ fig : vsrel ] , we also show the ratio @xmath345 ( from the radial velocity dispersion profiles and deprojected rotation ) as measured at different radii versus the mass ratio @xmath346 at the same radii .
the discrepancy between the adc and jam inferred enclosed masses becomes gradually larger for increasing @xmath345 .
the impact of the models discrepancy is larger when is converted into mass ratio .
it also has the same order of magnitude as the expected uncertainty of the epicycle approximation ( @xmath347 , @xcite ) , which is actually applied in adc ( but not in jam ) approach .
we see that the circular velocity measurement of both modelling approaches are consistent for @xmath1 @xmath348 , corresponding to those radii in the sb sbc galaxies where the disc luminosity is dominating over the bulge luminosity , i.e. , where the thin solid curve in fig .
[ fig : resratio ] approaches unity .
since some of our galaxies do not have well fitting models , we mark their data with a different symbol in order to check for possible biases .
however , even after their removal , the general trends of the velocity and mass ratios are preserved .
the grey filled circles correspond to high - quality adc and jam model fits data , while the open pentagon and the open star correspond to low - quality adc and jam data , respectively .
the black filled squares show the median of the distribution of the data per bin .
the error bar corresponds to the uncertainties of the median value at the 25th and 75th percentiles of the distribution at each bin ( see also table [ tab : calc5 ] ) .
= 1.2 mm @xmath1 & @xmath349 & @xmath350 & @xmath351 + ( 1 ) & ( 2 ) & ( 3 ) & ( 4 ) + 0.1 & @xmath352 & 0.4 & @xmath353 + 0.3 & @xmath354 & 0.8 & @xmath355 + 0.5 & @xmath356 & 1.3 & @xmath357 + 0.7 & @xmath358 & 2.4 & @xmath359 + 0.9 & @xmath360 & 4.2 & @xmath361 + 1.1 & @xmath362 & 8.3 & @xmath363 + 1.3 & @xmath364 & 14.1 & @xmath365 + 1.5 & @xmath366 & 24.8 & @xmath367 + 1.7 & @xmath368 & 49.6 & @xmath369 + 1.9 & @xmath370 & 85.5 & @xmath371 + 2.0 & @xmath372 & 150.1 & @xmath373 + [ tab : calc5 ] since we are probing the inner parts of these spiral galaxies it might well be the presence of bulges and/or thick stellar discs that causes a break - down of the thin - disc / epicycle approximation in adc . as can be seen from fig .
[ fig : vc_a ] , the stellar rotation ( first column ) is of the same order and often even lower than the stellar dispersion ( second column ) .
the small ordered - over - random motion values implies that the stars are far from dynamically cold .
detailed photometric studies indicate that most disc galaxies contain a thick disc ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) , and in low - mass galaxies with circular velocities @xmath15120kms@xmath5 , like the scd sd galaxies in this sample , thick disc stars can contribute nearly half the luminosity and dominate the stellar mass ( @xcite ) .
moreover , recent hydrodynamical simulations that reproduce thick discs show that their typical scale lengths are around 35 kpc ( e.g. * ? ? ?
* ) , i.e. , twice range typically covered by our analysis .
additionally , various earlier studies have qualitatively indicated that the adc approach might not be suitable in the case of stellar systems that are ( locally ) not dynamically cold ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
we also find that the discrepancy between the two model is proportional to the error of the epicycle approximation in adc ( see @xcite ) , where @xmath374 .
@xcite show that rotation curves of early - type galaxies based on the dynamically cold molecular gas are well traced by the @xmath325 , and that the @xmath375 rotation curves from their ` sauron ` data exhibit noticeable asymmetric drift .
they also find the correction of the stellar rotation curve using adc approach is also consistent with @xmath325 .
however , they note that there is some indication that this becomes less true as velocity dispersion approaches the rotation speed , as we find in this study . it is not excluded that there are strong streaming motions in the centers of the ` sauron ` galaxies due to presence of non - axisymmetric features , which prevent an accurate asymmetric - drft correction . nevertheless , some of the discrepancy might be arise from jam overestimation of the m / l ratio .
@xcite showed that the overestimation can be severe in face - on barred system .
@xcite also discussed the possibility of an overestimation of the @xmath376 ratio due to disk averaging over various stellar populations .
we do not exclude the latter since the region probed by our data is in the central part of the galaxies , where multiple stellar populations can be present .
the former case should not influence some of our measurements , since all of our galaxies have intermediate inclinations . in future studies
, we will explore further this problem using large statistical sample of galaxies and morphologies .
the rotation curves of spiral galaxies , traced by atomic gas , provide the most direct path to estimate the total galaxy masses outside the stellar disc .
however , it is not straightforward to measure the rotation curve in the central parts of the galaxies , where the ionized and molecular gas dominate .
gas settles in the equatorial or polar plane and is thus less sensitive to the mass distribution perpendicular to it .
gas is also dissipative and easily disturbed by perturbations in the plane from features like bars or spiral arms .
thus , stars appear to be a better tracer as they are distributed in all three dimensions and , being collisionless , they are less sensitive to perturbations .
however , stars are not cold tracers as they move in orbits that are neither circular nor confined to a single plane .
we therefore need to measure both velocity dispersions and line - of - sight velocities to recover the total mass distribution of galaxies . in this paper
we compare two different approaches for inferring dynamical masses of spiral galaxies : the commonly used asymmetric drift correction ( adc ) and the axisymmetric jeans equations .
we used the stellar kinematics derived by integral field spectroscopy of a sample of 18 sb sd galaxies , observed with the ` sauron ` spectrograph .
we obtained stellar mean velocity and velocity dispersion maps and derived the galaxies circular velocity curves by fitting solutions to the jeans equations . using the same data we also derived the circular velocity curves via the adc technique .
we use markov chain monte carlo methods to determine credible values for model parameters and their associated uncertainties .
the adc approach gives systematically lower values than jam for the circular velocity curves , and hence the enclosed mass , when @xmath1 @xmath377 .
the velocity ratio @xmath378 for sb sbc galaxies is larger in the inner parts ( @xmath328 ) and decreases outwards .
however , for scd sd galaxies the mass ratio stays roughly constant and in a few cases increases towards larger radii .
the complete python codes , implementing mcmc analysis on both jam and adc models , described in this paper , can be downloaded from ` https://github.com/kalinova/dyn_models ` .
these codes provide a robust way to estimate the uncertainties in the derived mass distribution of galaxies through their circular velocity curves .
we are extremely grateful to the anonymous referee for their valuable contribution towards the completeness of this paper .
his / her reports were constructive and thorough and dramatically improved the quality of the result .
thanks ronald lsker for the fruitful discussions .
we thank michelle cappellari for the helpful suggestions about the calculation of the uncertainties of the symmetrized @xmath267 field .
we acknowledge financial support to the dagal network from the people programme ( marie curie actions ) of the european union s seventh framework programme fp7/2007- 2013/ under rea grant agreement number pitn - ga-2011 - 289313 .
j. f .- b . acknowledges support from grant aya2013 - 48226-c3 - 1-p from the spanish ministry of economy and competitiveness ( mineco ) .
v.k , d.c . and e.r .
are supported in part by a discovery grant from nserc of canada .
[ lastpage ] [ s : appendix ]
here we provide a table with the parameters of our mge models for each galaxy , where @xmath80 is the central surface brightness , @xmath379 is the the dispersion along the major @xmath66-axis , and @xmath82 is the flattening .
additionally , @xmath380 indicates the gaussians of the central , bulge or disc component of the galaxies with respective values of 0,1 and 2 . [ cols="^,^,^,^,^,^,^,^,^,^,^,^,^ " , ] | we infer the central mass distributions within 0.41.2 disc scale lengths of 18 late - type spiral galaxies using two different dynamical modelling approaches the asymmetric drift correction ( adc ) and axisymmetric jeans anisotropic multi - gaussian expansion ( jam ) model .
adc adopts a thin disc assumption , whereas jam does a full line - of - sight velocity integration .
we use stellar kinematics maps obtained with the integral - field spectrograph @xmath0 to derive the corresponding circular velocity curves from the two models . to find their best - fit values , we apply markov chain monte carlo ( mcmc ) method .
adc and jam modelling approaches are consistent within 5% uncertainty when the ordered motions are significant comparable to the random motions , i.e , @xmath1 is locally greater than 1.5 . below this value ,
the ratio @xmath2 gradually increases with decreasing @xmath1 , reaching @xmath3 .
such conditions indicate that the stellar masses of the galaxies in our sample are not confined to their disk planes and likely have a non - negligible contribution from their bulges and thick disks .
[ firstpage ] galaxies : bulge galaxies : disc galaxies : kinematics and dynamics galaxies : structure |
during last decade the integrable structures associated with conformal field theory ( cft ) and its @xmath0 extension were studied @xcite-@xcite . the main problem to which these papers were addressed is the simultaneous diagonalization of the corresponding infinite family of the integrals of motion ( i m ) .
the method used for this purpose is the continuous version of the quantum inverse scattering method ( qism ) @xcite .
the most successful results were obtained by means of the generalization of the baxter s @xmath1-operator relation @xcite . in this paper
we will follow this way constructing the @xmath1-operator for analysis of the integrable structure arising in the superconformal field theory ( scft ) .
this cft includes both usual left and right components of the energy - momentum tensor @xmath2 , @xmath3 and spin-@xmath4 local fields @xmath5 , @xmath6 where variable @xmath7 lies on a cylinder of circumference @xmath8 .
the full symmetry algebra of the theory is @xmath9 ( @xmath10 means superconformal algebra).we will concentrate here on the left - chiral component- sca .
the boundary conditions are : @xmath11 where `` + '' corresponds to ramond ( r ) sector and `` '' to neveu - schwarz ( ns ) one .
the @xmath10 gives possibility to construct two infinite abelian families of conserved charges @xcite,@xcite,@xcite where one is susy invariant and the other is not .
here we will consider the integrable structure invariant under the supersymmetry transformation . in the classical limit
the corresponding i m give the hamiltonians of the susy @xmath12=1 kdv hierarchy @xcite based on twisted affine lie superalgebra @xmath13 @xcite .
the supertrace of the monodromy matrix of the associated l - operator in the 3-dimensional representation of @xmath14 ( see appendix 1 ) is a generating function for the i m .
thus it s quantum generalization , the transfer matrix produces the i m of the scft .
we will review some results from part i @xcite concerning the construction of the quantum counterpart of the monodromy matrix and it s basic properties in sec .
2 . using the super q - oscillator representations of the upper borel subalgebra of quantum affine superalgebra @xmath15 we define the @xmath16 operators .
the transfer matrices in different evaluation representations can be expressed in such a way : @xmath17 where @xmath18 runs over integer and half - integer nonnegative numbers .
@xmath16 operators satisfy quantum super - wronskian relation : @xmath19 one should note , that we use only @xmath20-dimensional `` @xmath21-induced '' representations ( sometimes called atypical ) of @xmath14 .
it allows , however , to construct the fusion relations , see below . to construct the baxter - like relation we introduce additional `` quarter''-operators , constructed `` by hands '' from the @xmath1-operators : @xmath22 for odd integer @xmath23 .
the baxter s relations are : @xmath24 @xmath25 the fusion relations have the following form very similar to the @xmath26 case @xcite : @xmath27 but these relations are only `` fusion - like '' because the `` quarter''-operators do not seem to correspond to any representation of @xmath14 .
this will be the subject of sec .
3 . finally in sec .
4 we describe the truncation of these relations for different values of @xmath28 , being root of unity .
in the part i @xcite we have explicitly constructed the quantum counterpart of the monodromy matrix of the linear problem for the l - operator of the susy n=1 kdv equation .
this l - operator has the following explicit form : @xmath29 where @xmath30 , @xmath31 , @xmath32 are the chevalley generators of @xmath14 ( see appendix 1 ) , @xmath33 is a superderivative , the variable @xmath7 lies on a cylinder of circumference @xmath8 , @xmath34 is a grassmann variable , @xmath35 is a bosonic superfield .
the poisson brackets for the field @xmath36 , obtained by means of the drinfeld - sokolov reduction are : @xmath37 and the following boundary conditions are imposed on the components of @xmath36 : @xmath38 , @xmath39 . making a gauge transformation of the auxiliary linear problem we obtain a new superfield @xmath40 , where @xmath41 and @xmath42 generate the superconformal algebra under the poisson brackets : @xmath43 the corresponding infinite series of the integrals of motion @xcite : @xmath44 allows us to construct integrable evolution equations . for example
@xmath45 yields the susy n=1 kdv equation @xcite : @xmath46 which has the following form in components : @xmath47 these integrals of motion ( [ i m ] ) can be extracted from the monodromy matrix of the l - operator ( [ lop])@xcite : @xmath48 \big).\nonumber\end{aligned}\ ] ] it s quantum generalization can be represented in the quantum p - exponent form ( for the explanation of this notion see below ) : @xmath49 vertex operators @xmath50 have the following form @xmath51 , where : @xmath52=\frac{i}{2}\beta^2 , \quad [ a_n , a_m]=\frac{\beta^2}{2}n\delta_{n+m,0},\qquad \{\xi_n,\xi_m\}=\beta^2\delta_{n+m,0}.\nonumber\\ & & : e^{\pm\phi(u)}:= \exp\big(\pm\sum_{n=1}^{\infty}\frac{a_{-n}}{n}e^{inu}\big ) \exp\big(\pm i(q+pu)\big)\exp\big(\mp\sum_{n=1}^{\infty}\frac{a_{n}}{n}e^{-inu } \big)\nonumber.\end{aligned}\ ] ] @xmath50 satisfy the following properties : @xmath53 these relations gave us possibility to show that the reduced universal r - matrix of the quantum affine superalgebra @xmath54 ( see appendix 1 ) with lower borel subalgebra represented by operators @xmath55 have the p - exponent multiplication property @xcite .
thus we have called it the quantum p - exponent : @xmath56 the universal r - matrix represented by @xmath55 coincide with @xmath57 .
note , that the monodromy matrix is : @xmath58 unlike the cases of quantum ( super)algebra @xmath59 , @xmath60 , @xmath61 , @xmath62 @xcite-@xcite , @xcite this object ca nt be represented in the usual p - ordered form due to the singularities arising from the products of fermion free fields .
however as in all these cases @xmath63 which we will denote in the following as @xmath64 to simplify the notation satisfy the rtt - relation @xcite , @xcite : @xmath65 where @xmath18 , @xmath66 means that the corresponding object is considered in some representation of @xmath54 .
thus the supertraces of monodromy matrix ( `` transfer matrices '' ) @xmath67 commute .
it is very useful to consider the evaluation representations of @xmath54 , @xmath68 , where now symbol @xmath18 means integer and half - integer numbers ( see appendix 1 ) . denoting @xmath69 as @xmath70
we find that @xmath71 commute : @xmath72=0\end{aligned}\ ] ] the expansion of @xmath73 in @xmath74 ( the transfer matrix in the fundamental 3-dimensional representation ) is believed to give us as coefficients the local i m , the quantum counterparts of ( [ i m ] ) : @xmath75 where the densities @xmath76 are differential polynomials of @xmath2 and @xmath5 such that @xmath77 terms with derivatives .
@xmath2 and @xmath5 have the following representation in terms of free fields ( [ freefields ] ) ( see @xcite ) : @xmath78 one of the basic features of the i m is that their commutators with certain vertex operators reduce to total derivatives @xcite : @xmath79=\partial_u(:o^{(k)}_{\pm}(u ) w_{\pm}(u):)=\partial_u\theta^{(k)}_{\pm}(u)\end{aligned}\ ] ] ( where @xmath80 is the polynomial of @xmath81 , @xmath82 and their derivatives ) . by means of this important property one can show ( see @xcite ) that such integrals commute with the transfer matrices : @xmath83=0.\end{aligned}\ ] ] in our case it is also important to show that the generator of the supersymmetry transformation @xmath84 commutes with @xmath85 .
really , @xmath86=\pm i^{1/2}\partial_u : e^{\pm\phi(u)}:=\partial_u\xi^{\pm}(u ) .
\end{aligned}\ ] ] thus , @xmath87 can be included in the family of i m and using again the results of @xcite it is easy to show that @xmath88=0 . \end{aligned}\ ] ]
let s consider evaluation representations @xmath68 of the quantum algebra @xmath54 which we have already mentioned in the introduction and in previous section .
they are @xmath20 dimensional where @xmath18 runs over integer and half - integer numbers .
we call these representations `` @xmath89-induced '' because each triple @xmath90 and @xmath91 form @xmath89 superalgebra @xcite , and @xmath20-dimensional representation space is a representation space for each @xmath89 subalgebra irreducible representation ( see appendix 1 ) .
+ we will also use in the following the corresponding reducible infinite - dimensional representations @xmath92 , which look like @xmath89 verma representations for the @xmath90 triple . to simplify the notation
we denote @xmath93 as @xmath94 and @xmath95 as @xmath96 .
+ in order to construct the @xmath1-operator we introduce the auxiliary object , the super q - oscillator algebra @xcite with the following commutation relations : @xmath97=\pm\varepsilon_{\pm } , \qquad q^{\frac{1}{2}}\varepsilon_{+ } \varepsilon_{-}+q^{-\frac{1}{2}}\varepsilon_{-}\varepsilon_{+}=\frac{1}{q - q^{-1}}.\ ] ] using these relations it is possible to build realization for the upper borel subalgebra @xmath98 generated by @xmath99 : @xmath100 and we denote @xmath101 as @xmath102 .
next , following @xcite we define two operators which do not depend on representation of oscillator algebra but depend on ( and completely determined by ) the commutation relations and cyclic property of the supertrace : @xmath103 where @xmath104 multiplying two @xmath105 operators we arrive to such a result ( see appendix 2 ) : @xmath106 where @xmath107 is the supertrace of the monodromy matrix in @xmath92 representation .
defining then the @xmath1-operators : @xmath108 we find that the expression ( [ aa ] ) can be written in such a form : @xmath109 now let s recall that @xmath107 corresponds to infinite - dimensional representation @xmath110 . in order to obtain the transfer matrix corresponding to the finite - dimensional representation @xmath111 one should consider factor - representation + @xmath110/@xmath112 ( see appendix 1 ) .
thus one obtains : @xmath113 sign + in this formula appears because of taking a supertrace .
really , @xmath114 subrepresentation in @xmath110 representation space has a highest weight vector which is odd ( though
our convention is that highest weight vector in @xmath115 representation is even ( see appendix 1 ) ) .
that s why we have sign + instead of the usual . + rewriting ( [ ttt ] ) in terms of the @xmath16 operators we find : @xmath116 considering the case of 1-dimensional representation of @xmath54 ( @xmath117 ) we obtain a quantum super - wronskian ( qsw ) relation : @xmath118 in order to construct the baxter type relation @xcite we introduce the auxiliary `` quarter '' operators : @xmath119 for odd integer @xmath23 .
let s consider the @xmath120 operator and multiply it on @xmath121 operators .
using ( [ sw ] ) one can get the following relation of baxter type , which is similar to that one obtained in @xmath26 case @xcite : @xmath122
the main difference is that unlike the @xmath26 case we are not seeking for the eigenvalues of @xmath120 operator but we are interested in the operator corresponding to the 3-dimensional representation , @xmath123 , because we know that it is a generating function of the local i m @xcite .
thus we are looking for the relation of baxter type which includes this operator . + in order to do this we need to proceed through some additional steps .
first , let s unify the expressions ( [ tq ] ) , ( [ quarter ] ) : @xmath124 where @xmath125 , @xmath126 .
using the qsw relation it is not hard to obtain the expressions for @xmath85 operators only in terms of @xmath127 or @xmath128 : @xmath129 for example for @xmath123 we obtain the `` triple relation '' : @xmath130 due to the unifying relation ( [ utq ] ) and using again ( [ sw ] ) we find the fusion relations : @xmath131 however , as we already mentioned in the introduction these relations are different from fusion relations obtained for ( super)algebras @xmath26 , @xmath132 , @xmath133 , @xmath134 ( see @xcite,@xcite,@xcite,@xcite ) in the sense that in all these cases operators present in the fusion relation correspond to the representations of the associated algebras . in our case
we see that artificially constructed auxiliary quarter operators do not correspond to any representation of @xmath54
. + let s write the fusion relations for @xmath135 with @xmath74 multiplied on @xmath136 : @xmath137 multiplying both sides on @xmath121 we obtain second relation of baxter type : @xmath138
though the baxter @xmath1-operator method is very powerful there exists another approach which is useful for rational values of the central charge @xcite , it is the scft counterpart of baxter s commuting transfer matrix method . in this section
we will show that when q is the root of unity the system of fusion relations becomes finite .
so , let s put @xmath139 , @xmath140 , @xmath141
. then @xmath142 here @xmath125 , @xmath143 .
note also that @xmath144 with the use of such a transform one obtains : @xmath145 recalling the formula ( [ tq ] ) for @xmath146 we arrive to the following expression : @xmath147 for example for @xmath117 we obtain : @xmath148 using the basic formula ( [ tq ] ) and property ( [ mon ] ) @xmath149 and for @xmath150 one arrives to the following relations : @xmath151 @xmath152 comparing ( [ n/2 ] ) and ( [ n/2 - 1/2 ] ) we obtain the following simple formula : @xmath153 in the case when @xmath154 , where @xmath155 , @xmath156 there exist additional number of truncations : @xmath157 the relation ( [ trunc ] ) allows us to to rewrite the fusion relation system in the thermodynamic bethe ansatz equations @xcite of @xmath158 type as in @xmath26 case @xcite : @xmath159 with the use of identification : @xmath160 where we have denoted @xmath161 the eigenvalue of @xmath85 .
in this paper we studied algebraic relations arising from integrable structure of cft provided by the susy @xmath12=1 kdv hierarchy . as it was already mentioned in @xcite , @xcite , @xcite the construction of the @xmath1-operator as a `` transfer''-matrix corresponding to the infinite - dimensional q - oscillator representation could be also applied to the lattice models .
the relations like baxter s and fusion ones will be also valid in the lattice case because they depend only on the decomposition properties of the representations .
+ the use of supersymmetry in the considered model is that the limit of @xmath162 when @xmath163 is very simple due to the expansion @xcite : @xmath164 where @xmath165 , @xmath166 for @xmath167 and @xmath168 are the integrals of motion of the susy @xmath12=1 kdv .
thus @xmath169 .
we hope that such a nice result remains valid in the quantum case .
maybe this conjecture will give us possibility to simplify the study of the analytic properties of the eigenvalues of @xmath170 , @xmath1 operators and therefore the corresponding integral equations of destri - de vega type @xcite , @xcite .
we will consider these questions in the third part of our manuscript @xcite .
it will be interesting to consider the obtained tba equations ( [ tba ] ) in the context of the perturbed minimal models as it was done in @xcite .
+ in the following we also plan to study the quantization of @xmath171 susy kdv hierarchies , related with super - w conformal / topological integrable field theories .
we are grateful to f.a .
smirnov for useful discussions .
the work was supported by the dynasty foundation ( a.m.z . ) and rfbr grant 03 - 01 - 00593 ( p.p.k . ) .
& & \lambda^{-1}e_{\alpha_1}|s , l\rangle = c[s - l]^{+}[s+l+1/2 ] & & \lambda e_{-\alpha_1}|s , l\rangle= & & \lambda e_{-\alpha_1}|s , l\rangle=- & & \lambda^{-1}e_{\alpha_0}|s , l\rangle= & & \lambda^{-1}e_{\alpha_0}|s , l\rangle= & & \lambda e_{-\alpha_0}|s , l\rangle = c[s - l][s+l+1/2]^{+ } & & \lambda e_{-\alpha_0}|s , l\rangle =- c[s - l]^{+}[s+l+1/2 ] where @xmath207=\frac{q^x - q^{-x}}{q - q^{-1}}$ ] , @xmath207^{+}=\frac{q^x+q^{-x}}{q+q^{-1}}$ ] , @xmath208^{+})^{-1}$ ] .
+ we also note ( this will be important in taking the supertrace ) that we choose the highest weight @xmath209 to be even . + the representations @xmath210 of @xmath54 , which play also a crucial role in the construction of the @xmath1-operators ( see sec .
3 ) are infinite - dimensional and their explicit form is similar to that of ( [ repr ] ) but with @xmath211 going from @xmath18 to @xmath212 . from a point of view of the @xmath89 subalgebra , generated by the triple @xmath90 these representation
look like verma module for @xmath89 with the highest weight @xmath213 one can easily obtain that @xmath111 representations arise from the factor of two infinite - dimensional representations of @xmath214-type : @xmath110/@xmath114 .
the product of two @xmath105-operators ( [ aa ] ) can be rewritten in the following form : @xmath215 where @xmath216 and @xmath217 . + by the fundamental property of the universal r - matrix : @xmath218 to rewrite the tensor product of two @xmath64-operators from the previous formula in such a way : @xmath219 where @xmath220 @xmath221 , @xmath222 and @xmath223 satisfy the following commutation relations : @xmath224=\mp\alpha_{\pm } , \quad[\mathcal{h},\beta_{\pm}]=\mp\beta_{\pm},\\ & & q^{1/2}\alpha_-\alpha_++q^{-1/2}\alpha_+\alpha_-=-\frac{\mu^2}{q - q^{-1 } } , \quad q^{1/2}\beta_+\beta_-+q^{-1/2}\beta_-\beta_+=\frac{\mu^{-2}}{q - q^{-1}}\nonumber\end{aligned}\ ] ] let s choose as @xmath225 ( [ chi ] ) the highest weight representations of the q - oscillator algebra ( @xmath226 ) with highest weight vectors @xmath227 such that : @xmath228 , where @xmath229 is the space spanned by the following vectors : @xmath230 takes such values when @xmath231 are linearly independent .
it is easy to see how @xmath223 and @xmath232 act on these vectors : @xmath233 the action of @xmath234 is more complicated : @xmath235 where @xmath236[s+l+1/2]^{+},\quad s - l\in\mathbb{z}\\ & & a(l , s)=c[s - l]^{+}[s+l+1/2 ] , \quad s - l\in\mathbb{z}+1/2\nonumber\end{aligned}\ ] ] and @xmath237 , the coefficients @xmath238 can be calculated but we do nt need them in the following .
the elements @xmath239 and @xmath223 acting in the space spanned by @xmath231 for any constant @xmath240 coincide up to a constant shift in @xmath223 and shift in the `` @xmath241-direction '' of @xmath239 action coincides with action of the corresponding chevalley generators in the infinite dimensional @xmath210 evaluation representations ( [ repr ] ) .
when we take the trace in ( [ aal ] ) there is no contribution of shift in `` @xmath241-direction '' and the constant shift in @xmath223 action can be easily calculated , that is : @xmath242 now recalling the definition of @xmath243 we find that in chosen highest weight representation @xmath244 therefore , @xmath245 20 v.v .
bazhanov , s.l .
lukyanov , a.b .
zamolodchikov , comm .
( 1996 ) 381 .
bazhanov , s.l .
lukyanov , a.b .
zamolodchikov , comm .
( 1997 ) 247 .
bazhanov , s.l .
lukyanov , a.b .
zamolodchikov , comm .
( 1999 ) 297 .
d. fioravanti , f. ravanini , m. stanishkov , phys .
b 367 ( 1996 ) 113 . v.v .
bazhanov , a.n .
hibberd , s.m .
khoroshkin , nucl .
b 622 ( 2002 ) 475 . l.d .
faddeev , in : quantum symmetries / symmetries quantiques , eds .
a.connes et al .
( north - holland 1998 ) p. 149 .
baxter , exactly solved models in statistical mechanics .
( london : academic press 1982 ) a. bilal , j .- l .
gervais , phys .
b 211 ( 1988 ) 95 . p.p .
kulish , a.m. zeitlin , phys .
b 581 ( 2004 ) 125 ; arxiv : hep - th/0312159 .
kulish , a.m. zeitlin , phys .
b 597 ( 2004 ) 229 ; arxiv : hep - th/0407154 .
t. inami , h. kanno , comm .
( 1991 ) 519 .
kac , adv .
( 1977 ) 8 .
p. mathieu , phys .
b 203 ( 1988 ) 287 .
kulish , e.k .
sklyanin , lect .
notes phys . 151 ( 1982 )
bershadsky , v.g .
knizhnik , m.g .
teitelman , phys .
b 151 ( 1985 ) 31 .
b. feigin , e. frenkel , lect .
notes math .
1620 ( 1996 ) 349 .
kulish , n.yu .
reshetikhin , lett .
phys . 18 ( 1989 ) 143 .
kulish , theor .
86 ( 1991 ) 157 .
zamolodchikov , phys .
lett b 253 ( 1991 ) 391 .
bazhanov , s.l .
lukyanov , a.b .
zamolodchikov , nucl .
b 489 ( 1997 ) 487 . s.m .
khoroshkin , j. lukierski , v.n .
tolstoy , math.qa/0005145 .
p. p. kulish , a.m. zeitlin , in preparation . | the algebraic structures related with integrable structure of superconformal field theory ( scft ) are introduced .
the scft counterparts of baxter s q - operator are constructed .
the fusion - like relations for the transfer - matrices in different representations and their truncations are obtained .
superconformal field theory , super - kdv , quantum superalgebras , q - operator 11.25.hf ; 11.30.pb ; 02.20.uw ; 02.20.tw |
in recent years , there has been increasing interest in the quantitative study of aspects of circle packings , in particular for apollonian packings .
for instance , the asymptotic count of the number of circles with bounded curvature has been determined @xcite , as well as several questions concerning the arithmetic of apollonian packings , see @xcite . in this paper
, we investigate local statistics for the distribution of tangencies to a fixed circle in a circle packing , in a class of packings generated by reflections .
the packings that we study are formed by starting with an initial finite configuration @xmath0 of circles in the plane with disjoint interiors , and such that the gaps between circles are curvilinear triangles , as in figure [ fig configurations ] .
we then form the group @xmath1 generated by reflections in the dual circles defined by the triangular gaps of the configuration , see [ sec : construct ] , and by applying the elements of @xmath1 to the initial configuration @xmath0 , we obtain an infinite circle packing @xmath2 , see figure [ fig configurations ] for an example .
we create such initial configurations by starting with a finite triangulation @xmath3 of the riemann sphere @xmath4 . by the circle packing theorem of koebe , andreev and thurston @xcite
, there is a circle packing @xmath0 of @xmath5 having @xmath3 as its tangency graph , that is a collection of disks with disjoint interiors on @xmath5 corresponding to the vertices of @xmath3 where two disks are tangent if and only if the corresponding vertices are connected in @xmath3 .
the gaps between the disks are the connected components of the complement @xmath6 of the disks in @xmath5 .
they are triangular because we assume that @xmath3 is a triangulation , that is each face of @xmath3 is a triangle .
after stereographic projection from the point @xmath7 , we obtain a circle packing @xmath8 in the finite plane @xmath9 .
we fix a base circle @xmath10 and consider the subset @xmath11 of circles tangent to @xmath12 .
let @xmath13 be an arc ( if @xmath12 is a line , take @xmath14 to be a bounded interval ) .
let @xmath15 be the set of tangencies in @xmath14 , and @xmath16 be the subset of @xmath15 whose corresponding circles in @xmath17 have curvatures bounded by @xmath18 , see figure [ fig tangencies ] .
[ b ] associated to the apollonian-9 packing of figure [ fig configurations ] , with @xmath12 being the bounding circle.,width=188 ] we will show that the cardinality of @xmath16 grows linearly with @xmath18 : @xmath19 where @xmath20 is the standard arclength of @xmath14 , and @xmath21 is a constant depending only on @xmath2 and @xmath12 .
in particular , the tangencies are uniformly distributed in @xmath12 . to compare , the total number of circles @xmath22 with @xmath23 is @xmath24 for some @xmath25 @xcite .
the goal of our paper is to study the distribution of nearest neighbour spacings ( gaps ) in @xmath26 .
let @xmath27 be the sequence of tangencies in @xmath16 ordered by some orientation ( say , the counter - clockwise orientation ) .
the nearest - neighbour gaps , or spacings , between the tangencies are @xmath28 where @xmath29 denotes the arc - length distance .
the mean spacing is @xmath30 we define the gap distribution function to be @xmath31 clearly the definition of @xmath32 does not depend on the orientation that we choose .
we will show there is a limiting gap distribution , which is conformally invariant and independent of @xmath14 : [ gapdistributionintro ] given @xmath33 as above , there exists a continuous piecewise smooth function @xmath34 , such that @xmath35 moreover , the limit distribution @xmath36 is conformally invariant : if @xmath37 , @xmath38 , @xmath39 then @xmath40 our method does not just give the existence , but actually explicitly calculates the limiting spacing distribution in terms of certain areas , showing that it is a continuous , piecewise smooth function . we also show that @xmath41 is supported away from the origin , which is a very strong form of `` level repulsion '' . in
[ sec : examples ] we explicitly compute the limiting distribution for some examples . as a warm - up
, we start with classical apollonian packings , where we start with three mutually tangent circles and then fill in each curvilinear triangle with the unique circle which is tangent to all three sides of that triangle , and then repeat this process with each newly created curvilinear triangle . in that case
we recover a theorem of hall @xcite on the gap distribution of farey points . for a history and further results in this direction see @xcite , and see @xcite for treatments using homogeneous dynamics .
we then compute the gap distribution for two other classes of packings : the guettler and mallows @xcite packing in figure [ fig configurations two ] , called also apollonian-3 , where the curvilinear triangle is filled by three new circles , each tangent to exactly two sides , and for which the associated tangency graph is the octahedron ; and for the configuration ( called apollonian-9 ) found by butler , graham , guettler and mallows ( * ? ? ?
* figure 11 ) , for which the tangency graph is the icosahedron , as in figure [ fig configurations ]
. the density functions of these three cases are displayed in figure [ fig : threeappdensity ] .
of the gap distribution for classical apollonian packings ( dashed curve ) , which is the same as that discovered by hall for farey sequence , for the apollonian-3 packings of guettler and mallows ( dotted curve ) , and for the apollonian-9 packing of butler , graham , guettler and mallows ( solid curve).,width=453 ] we work on models where @xmath8 consists of two horizontal lines , with @xmath42 and @xmath43 , and the remaining circles lies in the strip between them .
we call such models generalized ford configurations .
let @xmath44 be the subgroup of @xmath1 consisting of orientation preserving elements of @xmath1 that fix the base circle @xmath12 .
it is a non - uniform lattice in @xmath45 ( this group was used crucially in sarnak s work @xcite on apollonian packings ) .
using geometric considerations , we reduce the spacing problem to counting elements @xmath46 of the fuchsian group @xmath44 such that the lower row @xmath47 lies in a dilated region defined by the intersection of certain quadratic inequalities , and such that @xmath48 lies in a given interval . after that
, we use the spectral theory of automorphic forms , specifically the work of anton good @xcite , to show that the number of lattice points is asymptotically a multiple of the area of this dilated region . in [ sec :
construct ] we give details how to construct @xmath2 from the initial configuration @xmath49 .
we will also construct related groups and prove their geometric properties . in [ sec : uniform dist of tgts ] and [ fordgapdistribution ] we will prove the uniform distribution of tangencies and theorem [ gapdistributionintro ] in the special case that @xmath2 is a generalized ford packing , and @xmath12 is one of the two lines from @xmath2 . in
[ transfer ] we show how to deduce the general case of and theorem [ gapdistributionintro ] from the results of [ fordequidistribution ] and [ fordgapdistribution ] . in
[ sec : examples ] we compute some examples .
we reduce the computational effort involved by using conformal invariance , observing that the packings in the examples admit extra symmetries .
for instance , in the case of apollonian-9 packings , the associated symmetry group is the non - arithmetic hecke triangle group @xmath50 .
we start with a ( finite ) triangulation @xmath3 of the sphere . by the circle packing theorem of koebe , andreev and thurston @xcite
, there is a circle packing @xmath0 of @xmath51 having @xmath3 as its tangency graph , that is a collection of disks with disjoint interiors on @xmath51 corresponding to the vertices of @xmath3 ( there must be at least @xmath52 vertices ) , where two disks are tangent if and only if the corresponding vertices are connected in @xmath3 .
the interstices , or gaps , between the disks , are the connected components of the complement @xmath53 of the disks in @xmath51 . because we assume that @xmath3 is a triangulation , that is each face of @xmath3 is a triangle , this means that the gaps are triangular .
we may identify @xmath51 with the riemann sphere @xmath4 , and then by picking a point which will correspond to @xmath7 and performing a stereographic projection onto the plane we obtain a circle packing @xmath8 in the finite plane @xmath9 .
if @xmath7 is contained in the interior of one of the disks , then @xmath49 is realized as a configuration of finitely many circles , with one circle containing the other ones , see figure [ fig configurations two ] .
if @xmath7 is a point of tangency of two disks , then @xmath49 is a configuration consisting of two lines and several other circles in between ; we call this type of configuration a generalized ford configuration , see figure [ fig : apollonian-3fordbis ] .
we start with a circle packing @xmath8 as above , of @xmath54 tangent circles @xmath55 in the plane , so that the gaps between circles are curvilinear triangles .
we build up a circle packing @xmath56 from @xmath8 by refections : for each triangular gap , we draw a dual circle which passes through the vertices of the triangle , see figure [ fig : apo config ] . is the dashed circle.,width=188,height=188 ] we can generate new circles in the gap by reflecting @xmath8 through the dual circles .
each of the circles forming the gap is fixed under this reflection , and the remaining circles are reflected into the gap .
denote these reflections by @xmath57 and the corresponding triangles by @xmath58 , where @xmath59 is the number of gaps .
note that all these reflections are anti - holomorphic maps in @xmath9 ( throughout this paper , by triangle we mean solid triangle with vertices removed ) .
let @xmath60 be the group generated by the reflections .
the reflection group @xmath61 has only the relations @xmath62 .
this can be seen from the ping pong lemma : @xmath63 are disjoint sets , and @xmath64 maps every @xmath63 into @xmath65 when @xmath66 .
if we apply @xmath61 to the configuration @xmath8 , we obtain a circle packing @xmath2 of infinitely many circles . pick a circle @xmath12 from @xmath2 , and
let @xmath17 be the circles tangent to @xmath12 and let @xmath67 .
we are interested in studying the counting and spacing problems on @xmath17 .
the first step is to place @xmath2 in an ambient hyperbolic space with @xmath12 being the boundary , and show that there is a finite covolume fuchsian group @xmath44 acting on @xmath17 and having finitely many orbits .
a general element in @xmath61 is given by @xmath68 with @xmath69 .
therefore each circle @xmath70 can be represented by @xmath71 , where @xmath72 .
we denote the triangular gap corresponding to @xmath73 by @xmath74 .
we define the length of the above word to be @xmath75 .
we say the word above is _ minimal _ for @xmath76 if we can not find any shorter expression , and we say @xmath76 is in @xmath75-th generation . we first show that there is a canonical way to express a circle with minimal length : [ g1]the minimal word for a given circle @xmath70 is unique .
moreover , @xmath77 is minimal if and only if we have @xmath78 we first prove the only if " direction by induction on the generation of @xmath76 . for the base case
first generation , if @xmath79 is minimal , then @xmath80 , otherwise @xmath81 will fix @xmath82 and then the word is not minimal . as a result , @xmath83 .
suppose the
only if " direction holds for generations up to @xmath84 and @xmath85 is a minimal word for @xmath76 , then @xmath86 has to be a minimal word for some circle . by our induction hypothesis , @xmath87 and @xmath88 .
since @xmath89 ( otherwise @xmath90 and the word is not minimal ) , we must have @xmath91 . from
we see that @xmath81 corresponds the unique triangle in @xmath92 where @xmath93 sits in , @xmath94 corresponds the unique triangle in @xmath92 where @xmath94 sits in , @xmath95 , @xmath96 corresponds he unique triangle in @xmath92 where @xmath97 sits in .
this canonical description shows the uniqueness of minimal expression for a circle .
we prove the if " direction by induction on the length of expression .
if @xmath80 , then @xmath79 is a circle lying @xmath98 .
the only circles with shorter words are the ones in @xmath99 , all of which do not lie in @xmath98 , so @xmath79 is minimal .
now suppose @xmath100 is the minimal word for @xmath76 .
then by our discussion above @xmath101 is the unique reflection that corresponds to the triangle in @xmath92 where @xmath76 sits in , so @xmath102 .
then @xmath103 satisfies the condition and the minimal word for @xmath93 is @xmath104 . by our induction hypothesis , @xmath105 , @xmath106 for each @xmath107 . if @xmath108 , we write @xmath109 , and define @xmath110 to be the _ minimal _ triangle to @xmath76 .
clearly @xmath76 is contained in @xmath111 .
now let @xmath112 be the subgroup of @xmath1 consisting of elements that fix a given circle @xmath12 . by relabeling
let @xmath113 be the set of circles that are tangent to @xmath12 , and their tangencies are @xmath114 .
let @xmath115 be the set of generators of @xmath1 that fix @xmath12 , where @xmath116 is the reflection corresponding to the triangle formed by @xmath117 . for our convenience , at this point we make the convention that @xmath118 and @xmath119 for @xmath120 , because @xmath121 s from @xmath122 are forming a loop around @xmath12 .
@xmath123 is generated by the reflections in @xmath124 . if @xmath125 where @xmath126 and @xmath127 , we first absorb the trivial identities @xmath128 if @xmath129 , then we cut the first and last few consecutive letters of the word @xmath130 that are from @xmath124 ( which may not exist ) .
we continue this procedure several times until it stabilizes , then we get
a word for @xmath12 , which has to satisfy condition so is minimal .
but the minimal word for @xmath12 is just @xmath12 . since each cutting corresponds to multiplying an element in @xmath131 , we have @xmath132 .
@xmath112 acts on @xmath17 , and has finitely many orbits on @xmath17 .
@xmath112 acts on @xmath17 because @xmath1 send pairs of tangent circles to pairs of tangent circles .
since @xmath112 fixes @xmath12 , it will send any circle tangent to @xmath12 to some circle tangent to @xmath12 .
now suppose @xmath133 and the minimal word for @xmath76 is @xmath134 . from the argument in lemma 2.1
, @xmath135 corresponds to a triangle @xmath136 , which has to be a triangle that contains part of @xmath12 where @xmath76 lies in , so @xmath81 has to be from @xmath124 . by the same argument
, one can show that @xmath137 , and @xmath138 .
therefore , the number of orbits is @xmath139 , the cardinality of @xmath122 .
now we put a hyperbolic structure associated to @xmath44 .
without loss of generality we assume that @xmath12 is the bounding circle of radius @xmath140 , whose interior is the unit disk @xmath141 , and let @xmath142 be the geodesic connecting @xmath143 and @xmath144 ( again we extend the definition for all @xmath120 by setting @xmath142=@xmath145 ) , so that the region bounded by @xmath142 and @xmath12 contains the triangle @xmath63 .
each @xmath146 preserves the metric in @xmath141 , so @xmath147 , the isometry group of @xmath141 .
let @xmath148 be the open region bounded by the loops formed @xmath142 s ( see figure [ fig : fundamentaldomain ] ) . for @xmath112 .
, width=151 ] the following theorem , due to poincar ( see ( * ? ? ? * theorem 7.1.3 ) ) , tells us that @xmath112 acts discontinuously on @xmath141 with fundamental domain @xmath148 .
let @xmath149 be a finite - sided , convex polyhedron in @xmath141 of finite volume all of whose dihedral angles are submultiples of @xmath150 .
then the group generated by the reflections in the sides of @xmath149 is a discrete reflection group with respect to the polyhedron @xmath149 . as corollaries ,
@xmath112 has @xmath151 as cusps , and the fundamental domain for @xmath112 has volume @xmath152 by gauss - bonnet . for our purpose ,
we want to take the orientation preserving subgroup of @xmath112 , which we denote by @xmath44 .
the group @xmath44 is a free , index-2 subgroup of @xmath112 generated by @xmath153 . to see @xmath44 is free , we apply the pingpong lemma to @xmath154 , which maps @xmath65 to @xmath63 when @xmath66 . from the properties of @xmath112
, we immediately see : [ propertyofgamma ] @xmath44 has cusps at @xmath143 , with stabilizer @xmath155 .
the area of a fundamental domain is @xmath156 .
we now study the distribution of tangencies on a fixed circle from a circle packing . recall that @xmath2 is generated from a finite circle packing @xmath49 , which is associated to a triangulated spherical graph .
we pick a circle @xmath157 and let @xmath17 be the collections of circles in @xmath56 which are tangent to @xmath12 . given a bounded arc @xmath13 , let @xmath16 be the collection of tangencies of circles from @xmath17 , whose curvatures are bounded by @xmath18 .
let @xmath158 be the standard arclength measure .
we will show that [ equidistribution ] as @xmath159 , @xmath160 with @xmath21 independent of @xmath14 given by @xmath161 where @xmath162 is the hyperbolic area of the region bounded by @xmath121 and the two geodesics linking @xmath143 to @xmath144 and @xmath143 to @xmath163 ( see figure [ fig : dregions ] ) .
moreover , @xmath21 is conformal invariant : if @xmath37 , @xmath164 , and @xmath165 , then @xmath166 .
we deduce theorem [ equidistribution ] from some old results of a. good @xcite , but it is in principle known ; for instance it is a special case of @xcite
. we will first prove theorem [ equidistribution ] for a generalized ford packing , when the initial configuration contains two parallel lines , one of them being the base circle @xmath12 .
the general case will follow by conformal invariance ( theorem [ transferprinciple ] ) whose proof is deferred to
[ transfer ] .
for an explicit formula for the areas @xmath162 in the case of a ford configuration , see .
we first show that any of our initial configurations may be brought into a standard shape , that of a ( generalized ) ford configuration , see figure [ fig : apollonian-3fordbis ] .
.,width=264 ] we can map the initial configuration @xmath49 by using a mbius transformation and possibly a reflection , to one which includes two parallels @xmath42 and @xmath167 , two circles tangent to both of @xmath168 and @xmath169 with their tangencies on @xmath168 is 0 and @xmath170 respectively and some circles in between .
we may use a mbius transformation of the riemann sphere to map the configuration to the interior of the strip @xmath171 as follows : pick a triangle in the tangency graph @xmath3 one of whose vertices corresponds to @xmath12 , and whose other vertices corresponds to two other circles @xmath172 and @xmath173 , so that @xmath12 , @xmath172 , @xmath173 form a triple of mutually tangent circles . setting @xmath174 , @xmath175 , @xmath176 to be the tangency points , corresponding to the edges of the triangle
, there is a ( unique ) mbius transformation @xmath37 which maps @xmath177 then @xmath178 will map @xmath12 and @xmath172 to a pair of tangent lines passing through @xmath179 and @xmath180 respectively , and @xmath173 will be mapped to a circle tangent to both at these points .
hence @xmath178 will map @xmath12 to the real axis @xmath181 , @xmath172 to the line @xmath182 and @xmath173 to the circle @xmath183 centered at @xmath184 and having radius @xmath185 , and the remaining circles all lie in the strip @xmath186 .
the triangular gap @xmath187 between @xmath12 , @xmath172 , @xmath173 is mapped into one of the two triangular gaps forming the complement in the strip @xmath186 of the circle @xmath183 . after applying
if necessary the reflection @xmath188 , we may assume that @xmath187 is mapped to the gap in the half plane @xmath189 , so that all the other circles in @xmath8 are mapped into the gap with @xmath190 .
since any triangulation of the sphere has at least @xmath52 vertices , there must be at least one such circle distinct from @xmath12 , @xmath172 , @xmath173 .
since in the resulting configuration the gaps are still triangular , that means that there is one other circle tangent to both @xmath168 and @xmath169 , say @xmath191 ( @xmath192 ) and that the images of all the other circles are mapped to circles lying between @xmath193 and @xmath194 , as in figure [ fig : apollonian-3fordbis ] .
we carry over all previous notations .
the initial configuration @xmath49 includes two parallel lines @xmath42 and @xmath167 , two circles @xmath195 and @xmath196 tangent to both of @xmath168 and @xmath169 , and some circles in between .
our base circle @xmath12 is @xmath168 , so @xmath197 and @xmath198 is the collection of all circles tangent to @xmath168 .
the reflection @xmath199 corresponds to the triangle formed by @xmath168,@xmath169 and @xmath200 . applying @xmath201 to @xmath49 ,
the fundamental domain for @xmath44 is formed by geodesics connecting tangencies of @xmath49 and @xmath202 on @xmath168 .
each tangency is a parabolic point for @xmath44 , and since @xmath203 , we have @xmath204 , from proposition [ propertyofgamma ] .
we record a computation which will be used at several places : [ curvatureformula ] let @xmath205 .
i ) if @xmath206 , then under the mbius transform @xmath178 , a circle @xmath207 will be mapped to the circle @xmath208 if @xmath209 , and to the line @xmath210 if @xmath211 .
when @xmath212 , the image circle is @xmath213 ii ) if @xmath206 then the lines ( degenerate circles ) @xmath214 are mapped to @xmath215 , and to the line @xmath216 if @xmath212 . for both i ) and ii ) ,
if @xmath206 , we use the bruhat decomposition to write @xmath178 as f @xmath217 for each factor above the transformation formula is simple , then the curvature of the image circle is obtained by composing these simple formulae together .
the case when @xmath212 is a simple check . to deduce theorem [ equidistribution ]
, we need to calculate the contribution from each @xmath218 . for @xmath219 ,
write @xmath220 . from lemma [ curvatureformula ]
it follows that @xmath221 sends @xmath121 , @xmath222 , to a circle of curvature @xmath223 and the line @xmath167 to a circle of curvature @xmath224 .
therefore , for @xmath219 , we need to calculate @xmath225 and for @xmath226 , we need to calculate @xmath227 this counting problem is a special case of a theorem of a. good , ( see the corollary on page 119 of @xcite ) , which we quote here : [ anton ] let @xmath44 be a lattice in @xmath228 .
suppose @xmath7 is a cusp of @xmath44 with stabilizer @xmath229 and @xmath230 is any cusp of @xmath44 with stabilizer @xmath231 .
choose @xmath232 so that @xmath233 .
any @xmath234 can be written uniquely in the form @xmath235 with @xmath236 , so this determines functions @xmath237 on @xmath228 .
let @xmath238 be two bounded intervals in @xmath168 .
then as @xmath159 , @xmath239 now apply theorem [ anton ] to deduce theorem [ equidistribution ] . for @xmath219 , we set @xmath240 , and @xmath241 so that @xmath242
. therefore , @xmath243 because @xmath244 contributes half of the fundamental period of the stabilizer at @xmath7 of @xmath245 ; the other half comes from @xmath246 .
the map @xmath247 is a bijection of the cosets @xmath248 and @xmath249 .
write @xmath250 , then @xmath251 ( when either @xmath163 or @xmath144 is @xmath7 , this expression is interpreted as a limit )
. then the condition @xmath252 from is the same as @xmath253 , or @xmath254 .
the condition @xmath255 is the same as @xmath256 , or @xmath257 .
therefore , counting is the same as counting @xmath258 recall the area of @xmath44 is @xmath259 .
then by setting @xmath260 , from theorem [ anton ] we obtain @xmath261 we interpret this in conformally invariant terms : let @xmath162 be the area of the region bounded by @xmath121 and the two geodesics @xmath262 , @xmath142 ( see figure [ fig : dregions ] ) .
( shaded ) bounded by the circles @xmath121 in the initial configurations , and by the two geodesics linking @xmath143 to @xmath163 and to @xmath144 .
displayed is an initial configuration , generating the classical ford circles , consisting of two parallel lines @xmath42 and @xmath167 , and two mutually tangent circles @xmath263 and @xmath264 . here
@xmath265 , @xmath266 , @xmath267 . , width=377 ] then @xmath268 will translate this region to a triangle with vertices @xmath7 , @xmath269 , and @xmath270 . note that @xmath271 is a point on the segment connecting @xmath270 and @xmath269 , and that @xmath272 .
we also have @xmath242 .
since @xmath268 is area preserving , we have @xmath273 hence we find @xmath274 the case of @xmath226 is simpler .
counting is the same as counting @xmath275 where @xmath276 is a fundamental period of @xmath229 , the length of which is twice the distance of @xmath277 and @xmath278 , which is @xmath279 .
we obtain @xmath280 noting that the area of the region bounded by @xmath172 and the two geodesics @xmath281 , @xmath282 is @xmath283 , we find @xmath284 adding up the contributions from all @xmath218 , we obtain theorem [ equidistribution ] , with the constant @xmath285 where @xmath162 is the hyperbolic area of the region bounded by @xmath121 and the two geodesics @xmath262 and @xmath142 .
explicitly , @xmath286 ( when one of @xmath287 is @xmath7 , we take the appropriate limiting value of @xmath162 ) .
we now study the gap distribution on @xmath12 . as in
[ sec : uniform dist of tgts ] , let @xmath288 be an arc ( or a bounded interval ) , and @xmath289 the tangencies in @xmath14 whose circles have curvature at most @xmath18 .
let @xmath27 be the sequence of tangencies in @xmath16 ordered by counter - clockwise direction .
the nearest - neighbour gaps , or spacings , between the tangencies are @xmath290 and the mean spacing is @xmath291 we define the gap distribution function on @xmath14 to be @xmath292 we will find the limiting gap distribution of @xmath293 : [ gapdistribution ] there exists a continuous piecewise smooth function @xmath41 , which is independent of @xmath294 such that @xmath295 the limiting distribution @xmath41 is conformal invariant : let @xmath296 be as in theorem [ equidistribution ] , and @xmath297 be the gap distribution function of @xmath298 from @xmath299 , then @xmath300 .
the explicit formula for @xmath301 is given in theorem [ thm : formula for f ] .
this section is devoted to proving theorem [ gapdistribution ] in the case of generalized ford packings .
recall that we assume the initial configuration @xmath49 includes two parallel lines @xmath168 and @xmath169 , two circles tangent to both of @xmath168 and @xmath169 with their tangencies on @xmath168 is 0 and @xmath170 respectively and possibly some circles in between .
our base circle @xmath12 is @xmath168 , so @xmath197 and @xmath198 is the collection of all circles tangent to @xmath168 .
[ comparecurv ] i ) let @xmath302 lie in a triangular gap @xmath187 bounded by the real axis @xmath168 and two mutually tangent circles @xmath303 .
then both @xmath304 and @xmath305 have radius greater than that of @xmath306 .
\ii ) if @xmath307 are such that @xmath308 lies in a triangular gap @xmath187 bounded by the real axis , which is disjoint from @xmath309 , then there is another circle @xmath310 which separates @xmath308 and @xmath309 and has smaller curvature ( i.e. bigger radius ) than that of @xmath308 . to see ( i )
, we may move the circle @xmath306 so that it is tangent to both the real line and the larger of the two initial circles @xmath304 and @xmath305 , say it is @xmath304 .
we get a configuration as in figure [ fig geom arg ] .
if we denote by @xmath311 the radius of the larger circle @xmath304 , and by @xmath312 the angle between the segment joining the centers of the two circles @xmath304 , @xmath305 , and the segment joining the point of tangency of the circle @xmath304 with real line and with the center of @xmath304 , then a computation shows that the radius of the third mutually tangent circle @xmath305 is @xmath313 .
this is an increasing function of @xmath314 .
since the angle @xmath315 is smaller than @xmath316 for any circle @xmath306 contained in the triangular gap and tangent to both @xmath304 and the real line , it follows that the circle @xmath306 has smaller radius than that of @xmath305 as claimed .
( shaded black ) are such that @xmath308 lies in a triangular gap @xmath187 bounded by the real axis , with @xmath187 disjoint from @xmath309 , then there is another circle @xmath310 ( shaded gray ) , which is one of the bounding circles of @xmath187 , which separates @xmath308 and @xmath309 and has smaller curvature than that of @xmath308.,width=302 ] for ( ii ) , note if @xmath309 is disjoint from the triangular gap @xmath187 containing @xmath308 , then one of the bounding circles @xmath76 of @xmath187 separates @xmath308 and @xmath309 .
by part ( i ) , @xmath76 has bigger radius than @xmath308 , see figure [ geom arg separate ] .
the following proposition is a crucial ingredient in understanding the gap distribution : [ 12 ] for any adjacent pair @xmath317 with corresponding circles @xmath318 and @xmath319 , there is an element @xmath320 and @xmath321 such that @xmath322 and @xmath323 .
the element @xmath324 is unique if @xmath318 , @xmath319 are disjoint , and if they are tangent then the only other element with this property is @xmath325 , where @xmath326 is the reflection in the dual circle corresponding to the triangle formed by @xmath327 and @xmath168 .
let @xmath328 be the minimal word for @xmath329 , so that the triangle associated to @xmath318 is @xmath330 .
let @xmath331 be the minimal word for @xmath332 and @xmath333 .
let s suppose @xmath334 .
suppose there exists an integer @xmath335 $ ] such that @xmath336 .
we take @xmath59 to be minimal , then @xmath337 and @xmath338 , and @xmath339 and @xmath340 are two disjoint sets .
since @xmath341 when @xmath342 , we see that @xmath343 maps @xmath339 and @xmath340 to two disjoint triangles lying in @xmath98 .
these two triangles contain @xmath329 and @xmath332 respectively .
we have a contradiction here because by lemma [ comparecurv](ii ) , there is another circle in @xmath17 separating @xmath329 and @xmath332 which has smaller curvature , so @xmath344 and @xmath345 are not neighbors .
therefore @xmath341 when @xmath346 $ ] , and @xmath347 now we claim that @xmath348 fixes @xmath121 when @xmath349 .
this will show that @xmath350 so that we may take @xmath351 , which satisfies @xmath352 and @xmath353 as required .
otherwise , let @xmath354 $ ] again be the smallest integer such that @xmath355 , then @xmath340 contains @xmath356 , and @xmath340 and @xmath121 are disjoint .
as a result @xmath357 maps @xmath121 and @xmath340 to two disjoint sets : @xmath344 and the triangle @xmath358 , the latter of which contains @xmath332 .
therefore , by lemma [ comparecurv](ii ) , between @xmath329 and @xmath332 there must be a circle whose curvature is smaller than that of @xmath332 , so that @xmath318 and @xmath319 can not be adjacent in @xmath359 , contradiction .
hence @xmath360 for @xmath361 $ ] and we have constructed @xmath351 . to address uniqueness of @xmath362 ,
assume @xmath363 also sends @xmath121 to @xmath329 , @xmath364 to @xmath332 , then @xmath365 fixes @xmath121 and @xmath364 simultaneously . therefore ,
if @xmath121 and @xmath364 are disjoint , @xmath366 ; if @xmath121 and @xmath364 are tangent , @xmath367 , where @xmath368 is the reflection corresponding to the triangle formed by @xmath121 , @xmath364 and @xmath168 .
we now use proposition [ 12 ] to relate the gap distribution problem to a question of counting certain cosets in @xmath44 . for each pair of circles @xmath370 in the initial configuration which are tangent to @xmath12 , we will define regions @xmath371 in the right @xmath372-plane @xmath373 as follows .
\a ) if @xmath374 , we require @xmath375 and in addition : \i ) if @xmath121 , @xmath364 are tangent , then @xmath376 satisfies @xmath377 and @xmath378 where @xmath379 , with @xmath380 is the reflection corresponding to the triangle formed by @xmath327 and @xmath168 .
\ii ) if @xmath121 and @xmath364 are disjoint ( non - tangent ) , then @xmath376 satisfies @xmath381 \b ) if one of @xmath382 is @xmath7 , say @xmath383 , we require @xmath384 and in addition : \i ) if @xmath121 , @xmath364 are tangent , then @xmath376 satisfies @xmath385 and @xmath386 \ii ) if @xmath121 and @xmath364 are disjoint ( non - tangent ) , then @xmath376 satisfies @xmath387 note that @xmath388 is a finite union of convex sets and @xmath389 the result is : [ omegaij ] for any @xmath390 and @xmath391 , the circles @xmath392 are neighbors in @xmath393 if and only if @xmath394 lies in some @xmath388 in the right half plane , defined by the conditions \i ) if @xmath374 then we require , with or in the case that @xmath121 , @xmath364 are tangent , and if they are disjoint then we require .
\ii ) if @xmath383 , then we require , with or in the case that @xmath121 , @xmath364 are tangent , and if they are disjoint then we require .
let @xmath395 be the reflection in the line @xmath396 , corresponding to the triangular gap bounded by the lines @xmath168 , @xmath169 and the circle with tangency @xmath170 . from proposition
[ 12 ] , passing to the orientation preserving subgroup @xmath44 , every ( unordered ) pair of neighbor circles can be expressed as @xmath397 or @xmath398 for some @xmath399 and @xmath390 with @xmath400 .
the expression is unique when @xmath121 and @xmath364 are disjoint because the simultaneous stabilizer for @xmath121 and @xmath364 is trivial .
however , we only need to calculate the contribution to @xmath301 from @xmath401 under @xmath44 : the contribution from @xmath402 is the same as that from @xmath403 because they are mirror symmetric under the reflection of the line @xmath404 . or more formally , @xmath405 and @xmath201 is metric preserving on @xmath168 .
when @xmath121 and @xmath364 are tangent , then @xmath402 is also in the orbit of @xmath403 under @xmath44 . in this case
, every pair of circles from @xmath406 can be uniquely expressed as @xmath407 for some @xmath399 .
if two neighbouring circles in @xmath408 can be expressed as @xmath409 , then we need their curvatures to satisfy @xmath410 by lemma [ curvatureformula ] we have @xmath411 and so we get .
all the circles in between @xmath412 and @xmath413 must have curvatures greater than @xmath18 .
we see from lemma [ comparecurv ] that it suffices to check finitely many circles .
a useful fact here is the following : [ inbetween ] if neither @xmath121 nor @xmath364 is the horizontal line from @xmath122 , which means @xmath374 , then for another circle @xmath138 , @xmath414 is in between @xmath412 and @xmath413 if and only if @xmath415 if @xmath364 is the horizontal line , so that @xmath383 , then @xmath414 is in between @xmath412 and @xmath413 if and only if @xmath416 the observation is that given real numbers @xmath417 and @xmath418 , a third real @xmath419 lies between them if and only if @xmath420 now take the numbers to be @xmath421 , @xmath422 and @xmath423 and compute @xmath424 from which the claim follows . in the case @xmath383 , we have @xmath425 which gives the claim .
there are two cases : + case 1 : @xmath121 and @xmath364 are tangent .
then @xmath426 does not lie between @xmath143 and @xmath427 , for all @xmath59 .
if @xmath428 , then the circles @xmath429 for @xmath430 are mapped by @xmath362 to some circles in between @xmath412 and @xmath413 , and in this case we need @xmath431 and using lemma [ curvatureformula ] which gives @xmath432 we obtain condition .
if @xmath433 , then from observation [ inbetween ] none of the circles @xmath434 lie between @xmath435 and @xmath436 , but the circles @xmath437 for @xmath430 are mapped by @xmath362 to some circles in between @xmath412 and @xmath413 ( recall that @xmath438 is the reflection corresponding to the triangle formed by @xmath327 and @xmath168 ) . in this situation
we need @xmath439 by lemma [ curvatureformula ] , @xmath440 if @xmath379 , which gives condition .
case 2 : if @xmath121 and @xmath364 are not tangent . we need to make sure that @xmath441 has curvature @xmath442 whenever @xmath441 is in between @xmath412 and @xmath413 .
again from observation [ inbetween ] , if @xmath433 , then the circles whose tangencies lie between @xmath143 and @xmath427 will be mapped by @xmath362 to some circles between @xmath412 and @xmath413 ; if @xmath428 , then the complement of will be mapped in between @xmath412 and @xmath413 .
thus we need @xmath443 if either @xmath444 or @xmath445 this gives condition once we use @xmath432 ( lemma [ curvatureformula ] ) . from lemma [ comparecurv ] , once or
hold when @xmath121 , @xmath364 are tangent , or is satisfied if @xmath121 and @xmath364 are disjoint , then all circles between @xmath412 and @xmath413 have curvatures greater than @xmath18 . putting the above together , we have established proposition [ omegaij ] . we need to calculate the contribution from each pair of circles @xmath403 to the proportion @xmath446 of gaps of size at most @xmath447 , and we denote this quantity by @xmath448 : , @xmath449 we will later show that these have a limit as @xmath159 : @xmath450 a direct computation shows that the distance between @xmath412 and @xmath413 is given by @xmath451 .
the average gap is asymptotically @xmath452 .
therefore , the relative gap condition @xmath453 in the definition of @xmath454 can also be written as @xmath455 we will work with a simpler condition @xmath456 which will finally lead to a limiting gap distribution function @xmath41 which is continuous , then in retrospect working with will lead to the same limiting function @xmath41 , by the continuity of @xmath301 .
if one of @xmath143 and @xmath427 is @xmath7 ( say @xmath383 ) , the above is changed to @xmath457 we define a region @xmath458 to be the elements @xmath459 satisfying ( or ) . then we have found that @xmath460 the region @xmath458 is compact because the condition already gives a compact region bounded by two sets of parallel lines with different slopes .
note also that @xmath458 is a finite union of convex sets . from the defining equation ,
it is clear that the region @xmath458 grows homogeneously with respect to @xmath18 : @xmath461 as a consequence of the analysis above , we find that the normalized gaps in our circle packings all bounded away from zero : there is some @xmath462 so all gaps satisfy @xmath463 thus the limiting distribution @xmath41 is supported away from the origin .
this is a very strong form of level repulsion , familiar from the theory of the farey sequence . to prove the assertion , note that we expressed the distribution function @xmath464 as a sum over all ( unordered ) pairs of distinct circles from the initial configuration @xmath370 : @xmath465 where @xmath466 here @xmath467 is the region in the plane of points lying in the compact sets @xmath468 of [ sec : determining ] , satisfying @xmath469 .
@xmath470 with @xmath471 the sets @xmath468 are compact , and therefore the functions @xmath472 are bounded on them .
thus if @xmath473 then @xmath474 and @xmath475 if @xmath476 . in the examples of section [ sec : examples ] , we have @xmath477 for the classical apollonian packing
[ sec : app1 ] ( which reduces to the farey sequence by conformal invariance ) , @xmath478 for the apollonian-3 packing [ sec : app3 ] , and @xmath479 for the apollonian-9 packing
[ sec : app9 ] .
we now pass to the limit @xmath159 , by relating the counting problem encoded in our formula for @xmath480 to the area of the region @xmath481 .
this is done via an equidistribution theorem of a. good . to formulate it ,
recall the iwasawa decomposition @xmath482 where @xmath483 and @xmath484 we can uniquely write any @xmath485 as @xmath486 we have the following joint equidistribution according to this decomposition , which is another special case of good s theorem ( see corollary on page 119 of @xcite ) : [ nak ] let @xmath14 be an bounded interval in @xmath168 , @xmath276 be an interval in @xmath487 , then as @xmath159 , @xmath488 theorem [ nak ] allows us to prove the following [ expanding ] for a nice " subset @xmath489 , @xmath490 as @xmath159 , where @xmath75 is the standard lebesgue measure in @xmath491 . here
`` nice '' means bounded , convex with piecewise smooth boundary .
we note that if @xmath492 then the iwasawa decomposition gives @xmath493 , @xmath494 so that @xmath495 are polar coordinates in the @xmath496 plane .
we prove proposition [ expanding ] in the special case that @xmath497 is bounded by two continuous and piecewise smooth curves @xmath498 with @xmath499 $ ] , and @xmath500 .
the sign @xmath501 is obtained only if when @xmath502 .
this special case suffices for what we need for @xmath503 .
this is essentially a riemann sum argument .
first we can express @xmath504 as @xmath505 divide the interval @xmath506 $ ] into @xmath507 equal subintervals @xmath508 .
for each subinterval @xmath509 , pick @xmath510 at which @xmath311 and @xmath511 achieve their maximum and minimum respectively .
let @xmath512 be the union of truncated sectors @xmath513 $ ] , and @xmath514 be the union of @xmath515 $ ] .
we have @xmath516 and @xmath517 we notice that the statement in proposition [ expanding ] satisfies finite additivity . from theorem
[ nak ] , we know proposition [ expanding ] holds for sectors , thus it holds for finite union of truncated sectors , which is the case for @xmath518 and @xmath514 . letting @xmath507 go to infinity , we prove proposition [ expanding ] for @xmath497 .
we can now state our main result , the formula for the limiting gap distribution @xmath41 , along the way proving theorem [ gapdistribution ] .
we keep our previous notation .
[ thm : formula for f ] for any interval @xmath288 , @xmath519 , where @xmath520 with @xmath521 given in .
we saw that @xmath522 and we need to estimate @xmath454 . from proposition [ 12 ] and proposition [ omegaij ] we can rewrite @xmath454 as @xmath523 in terms of the iwasawa coordinates , the condition @xmath524 is essentially equivalent to @xmath525 , when @xmath526 tends to @xmath527 , as we shall explain now .
writing @xmath528 , @xmath529 and @xmath530 , we have @xmath531 and since @xmath532 , this will essentially be @xmath533 provided @xmath534 is bounded away from zero .
we first excise small sectors @xmath535 of angle @xmath536 containing as bisectors those @xmath537 s such that @xmath538 , where @xmath539 or @xmath540 .
then for any @xmath541 , @xmath542 @xmath543 such that if @xmath544 and @xmath545 , then @xmath546 set @xmath547 so that the area of @xmath548 differs from that of @xmath549 by at most @xmath550 . then there are two intervals @xmath551 such that @xmath552 , and @xmath553 and so that @xmath554 applying proposition [ expanding ] , which approximates the above cardinalities by @xmath555 ( recall @xmath556 ) , and dividing by @xmath557 we obtain @xmath558 letting @xmath559 , we get @xmath560 with @xmath521 is as in , and @xmath561 a piecewise smooth , continuous function of @xmath447 .
summing over all pairs of circles in @xmath122 , we get @xmath562 the reason there is an extra factor of @xmath563 for @xmath327 disjoint is because @xmath564 only parametrizes half of the gaps formed by the @xmath112-orbits of @xmath327 , and the contribution from @xmath565 , @xmath566 is identical to that from @xmath121 , @xmath364 .
let @xmath567 be a mbius transformation . in this section we show
that if a circle packing @xmath2 and a circle @xmath12 from @xmath2 satisfy theorem [ equidistribution ] with some constant @xmath21 , and theorem [ gapdistribution ] with some piecewise smooth continuous function @xmath301 , then the packing @xmath568 and @xmath569 also satisfy theorem [ equidistribution ] and theorem [ gapdistribution ] with the same constant and distribution function .
in fact we will give a more refined statement : for any pair of distinct circles @xmath370 in the initial configuration , the densities @xmath570 defined in , are conformally invariant .
this is particularly useful in reducing the computations of the limiting densities in the examples of
[ sec : examples ] to manageable length .
the argument itself is routine : the claims are obvious if @xmath178 is a dilation , and are proved for general @xmath37 by localizing . [ transferprinciple ]
let @xmath288 be an arc ( if @xmath12 is a line then take a bounded interval ) , and @xmath37 .
assume @xmath571 such that @xmath572 for any @xmath573 .
i ) for the packing @xmath568 and the base circle @xmath569 , @xmath574 \ii ) for any pair of distinct circles @xmath370 in the initial configuration , the densities @xmath570 are conformally invariant : @xmath575 where on the rhs , @xmath576 refers to the gaps associated with the pair of circles @xmath577 , @xmath578 and the packing @xmath568 with base circle @xmath569 .
\iii ) as a consequence , for all @xmath178 and subarcs @xmath288 , @xmath579 we use the notation @xmath580 and @xmath581 throughout the proof related to the asymptotic growth as @xmath582 . without further explanation ,
all the implied constants depend at most on @xmath178 and @xmath14 .
\i ) we first show that for a circle @xmath76 with tangency @xmath583 on @xmath14 , the curvature of @xmath584 satisfies @xmath585 choose a parametrization @xmath586 for @xmath76 in a neighborhood of @xmath583 , with @xmath587
. then the curvature is given by @xmath588 therefore , a direct computation shows that @xmath589 which gives .
+ we divide the arc @xmath590 on @xmath569 into @xmath591 equal pieces @xmath592 with their preimages @xmath593 on @xmath12 .
we pick a point @xmath143 from the interval @xmath594 for each @xmath595 , and let @xmath596 .
we have @xmath597 for any circle @xmath76 on @xmath598 , we have @xmath599 therefore , @xmath600 such that @xmath601 dividing the above expression by @xmath18 , and letting @xmath159 , we obtain @xmath602 replace @xmath603 by , sum over all @xmath595 , and let @xmath604 , we obtain @xmath605 thus we prove the conformal invariance of the density of tangencies .
+ ii ) now we show that for each pair of circles @xmath606 , @xmath607 let @xmath608 be an ordered sequence of @xmath609 , and @xmath610 be the sequence of @xmath611 ordered in a way such that @xmath612 has the same orientation as @xmath608 .
let @xmath613 be the subset of @xmath614 consisting of those @xmath615 such that @xmath616 .
let @xmath617 be the subset of @xmath609 consisting of those @xmath618 such that @xmath619
. then @xmath620 from , the symmetric difference @xmath621 therefore , if we let @xmath622 and @xmath623 then @xmath624 and @xmath625 let @xmath626 and @xmath627 be the ordered sequence from @xmath628 and @xmath629 respectively .
then @xmath178 is a bijection on these two sequences .
let s say @xmath630 .
therefore , @xmath631 and @xmath632 now @xmath633
@xmath634 @xmath635 combining , , , we see that @xmath636 which only depend on @xmath178 and @xmath14 , and for any arbitrary small number @xmath637 , @xmath638 , such that when @xmath639 , @xmath640 we also have @xmath641 therefore , @xmath642 , @xmath643 as a result , as @xmath159 , we obtain @xmath644 since @xmath645 is a convex combination of @xmath646 , if we let @xmath647 , @xmath648 and use the continuity of @xmath649 ( as follows from ) , we obtain @xmath650 any bounded circle packing can be written as @xmath568 for some generalized ford packing @xmath2 .
the only issue we need to deal with is @xmath7 , if an arc @xmath276 in @xmath569 contain @xmath651 , the argument in the proof of theorem [ transferprinciple ] does not directly apply .
however , this is easily solved by precomposing @xmath178 with some @xmath652 , so that @xmath653 is a bounded arc , and @xmath654 for some @xmath655 and any @xmath656
. then we can apply theorem [ transferprinciple ] .
we compute the gap distribution function @xmath41 for three examples : the classical apollonian packing , where we start with a configuration of four mutually tangent circles , whose tangency graph is the tetrahedron ; the apollonian-3 packing in which one starts with three mutually tangent circles , and in either of the curvilinear triangles formed by three mutually tangent circles we pack three more circles , forming a sextuple whose tangency graph is the octahedron ; and the apollonian-9 packing introduced in @xcite where the initial configuration consists of 12 circles , with the icosahedron as its tangency graph .
we explain the procedure used in the computations : according to theorem [ thm : formula for f ] , the gap distribution @xmath41 is given in terms of a sum of functions @xmath657 over ( unordered ) pairs of distinct circles @xmath658 of areas of regions @xmath659 in the @xmath376 plane , explicitly described in [ sec : determining ] .
so for instance in the case of the apollonian-9 packing , there are @xmath660 such pairs .
we are able to cut down on the computational effort involved by using conformal invariance , once we note that the packings we study below enjoy more symmetries than a generic packing .
according to theorem [ transferprinciple ] , if there is some @xmath37 which preserves @xmath17 and @xmath12 ( note that @xmath661 only depends on @xmath17 , not on all of @xmath2 ) , and takes the pair of circles @xmath662 in @xmath122 to another pair @xmath663 in @xmath122 , then @xmath664 . in each of these packings
, there exists a group @xmath665 which is larger than @xmath44 that also acts on this given packing . in each of the three examples presented ,
the bigger group @xmath666 is one of the hecke triangle groups @xmath667 , which is the group of mbius transformations generated by the inversion @xmath668 and the translation @xmath669 , where @xmath670 : @xmath671 for @xmath672 we recover the modular group : @xmath673 .
for @xmath674 the groups @xmath667 are non - arithmetic . in all three cases ,
we show that @xmath675 is a normal subgroup of the appropriate @xmath667 . in the first example of the classical ford packing , @xmath676 which acts transitively on the ford circles @xmath17 . in the second example , of the apollonian-3 packing , @xmath677 . in the third example , the apollonian-9 packing , @xmath678 .
it is a simple check that in each of these cases , the generators @xmath679 send circles in @xmath99 to some circles in @xmath17 .
so if we show @xmath679 normalizes @xmath44 , then it follows that @xmath679 permutes the circles in @xmath680 .
we further show that @xmath667 acts transitively on pairs of tangent circles in @xmath99 for @xmath681 , and on pairs of disjoint circles for @xmath682 .
hence by conformal invariance of the components @xmath657 , this gives an expression for @xmath41 as a sum of areas of two regions for @xmath682 ( only one for @xmath672 ) .
these regions are in turn expressed as a finite union of certain subregions @xmath683 . for each subregion @xmath684 ,
the first two equations give the conditions @xmath685 , and the other equations except the last one gives the condition @xmath686 , where @xmath76 exhausts circles in @xmath122 such that @xmath687 is in between @xmath412 and @xmath413 .
the last quadratic equation corresponds to conditions or , capturing relative gap information .
the areas of these regions can be explicitly computed in elementary terms , but the resulting formulae are too long to record .
instead we display plots of the density of the normalized gaps ( the derivative of @xmath301 ) .
+ the resulting packing @xmath2 is a classical apollonian packing , generated by inscribing a unique circle into every curvilinear triangle produced by the four mutually tangent circles and repeating .
the circles @xmath17 tangent to the real line are precisely the classical ford circles @xmath689 , having as points of tangency the farey sequence @xmath690 .
the gap distribution of the farey sequence was found by hall @xcite , and we carry out the computation here as a warm - up to illustrate our method .
the group @xmath112 is generated by the three reflections @xmath693 so @xmath44 generated by @xmath694 and @xmath695 , which are the generators of the principal congruence subgroup @xmath696 which is a normal subgroup of @xmath697 .
any pair of circles from @xmath698 are _ conformally equivalent _ , in the sense that there exists some @xmath699 which preserves the packing and maps @xmath172 , @xmath173 to this given pair of circles : indeed , @xmath700 therefore we have @xmath701 .
and so by theorem [ thm : formula for f ] , @xmath702 and it remains to compute the area of @xmath703 .
we compute the region @xmath704 using , , .
a direct computation gives @xmath705 the distance between @xmath706 and @xmath707 is @xmath708 . when @xmath709 , @xmath710 will be mapped by @xmath362 to some circle in between @xmath711 and @xmath712 by observation [ inbetween ] , and the corresponding region is @xmath713 note that @xmath717 and @xmath718 are symmetric about the @xmath419-axis , so we only need to calculate one area .
therefore , by theorem [ thm : formula for f ] , @xmath719 the density function @xmath720 of the normalized gaps ( the derivative of @xmath41 ) is given in figure [ fig : classicapp3density ] .
our next example was discovered by guettler and mallows @xcite , in which one starts with three mutually tangent circles , and in either of the curvilinear triangles formed by three mutually tangent circles we pack three more circles , forming a sextuple @xmath0 , see figure [ ap3ford ] .
the base circle is @xmath42 and the circles of @xmath0 tangent to it are @xmath721 to see that @xmath44 is normal in @xmath726 , we check that the conjugates of the generators of @xmath44 by the generators @xmath201 , @xmath730 of @xmath726 still lie in @xmath44 : @xmath731 @xmath732 @xmath733 @xmath734 @xmath735 there are @xmath736 ( unordered ) pairs of circles from @xmath122 : four pairs of tangent circles @xmath737 , @xmath738 , @xmath739 , @xmath740 , and two pairs of disjoint circles @xmath741 .
each of these cases gives one equivalence class under the action of @xmath726 : @xmath742 and @xmath743 hence from formula @xmath744 first we consider @xmath745 : the distance between @xmath749 and @xmath750 is @xmath708 .
therefore , we need to consider the cases @xmath709 and @xmath714 separately . if @xmath709 , then the circles @xmath710 and @xmath751 will be mapped by @xmath362 to some circles in between @xmath711 and @xmath712 .
the corresponding region @xmath752 is @xmath753 if @xmath714 , then the circles @xmath754 will be mapped in between @xmath711 and @xmath712 .
the corresponding region @xmath755 is @xmath756 the region @xmath704 is the union of @xmath717 and @xmath718 which are symmetric , hence the area is @xmath757 next we consider @xmath758 . here @xmath759 and we need to split it into two cases . if @xmath760 , then @xmath761 will be in between @xmath762 and @xmath763 , and the corresponding region is @xmath764 if @xmath765 , then @xmath766 will be in between @xmath762 and @xmath763 , and , @xmath767 then @xmath758 is the union of @xmath768 and @xmath769 ( which have equal areas ) , and @xmath770 thus @xmath771 the density function @xmath720 of the normalized gaps ( the derivative of @xmath41 ) is given in figure [ fig : app3density ] .
this packing was introduced in @xcite .
the initial configuration @xmath0 , whose tangency graph is the icosahedron , is shown in figure [ ap9ford ] .
the base circle is @xmath42 and the circles @xmath772 tangent to it are @xmath773 the constant @xmath21 in this case is @xmath774 .
we claim that @xmath779 , the hecke-5 group .
which is generated by @xmath780 we check that each generator of @xmath44 can be written as word of @xmath201 and @xmath781 and hence lies in @xmath50 : clearly , @xmath782 moreover @xmath783 and @xmath784 as claimed .
since @xmath44 is generated by elements of @xmath50 , it is a subgroup of that group , and since both groups have finite volume , @xmath44 is a subgroup of finite index in @xmath50 , and in particular is non - arithmetic . to see that @xmath44 is normal in @xmath50
, we check that the conjugates of the generators of @xmath44 by the generators @xmath201 , @xmath781 of @xmath50 still lie in @xmath44 : @xmath785 @xmath786 @xmath787 @xmath788 @xmath789 and hence @xmath790 @xmath791 the pairs @xmath737 , @xmath792 are representatives of two conformally equivalent classes of pairs of circles from @xmath122 , characterized by the two circles being tangent or disjoint and @xmath50 acts transitively on each class : indeed , @xmath793 and @xmath794 each equivalence class contains 5 pairs of circles . therefore ,
@xmath795 we first compute @xmath704 .
if @xmath709 , then by observation [ inbetween ] , the circles @xmath800 will mapped by @xmath362 in between @xmath711 and @xmath712 .
the corresponding region is , using , @xmath801 the condition @xmath802 is redundant here . if @xmath714 , then @xmath803 will be mapped in between @xmath711 and @xmath712 .
the corresponding region is , by , @xmath804 the region @xmath704 is the union of @xmath752 and @xmath755 , which are symmetric . similarly , if @xmath808 , then @xmath766 will be in between @xmath762 and @xmath763 , and the region is @xmath809 the region @xmath758 is the union of @xmath810 and @xmath811 .
therefore , @xmath812 the density function of the normalized gaps is given in figure [ fig : app9density ] .
j. marklof .
_ fine - scale statistics for the multidimensional farey sequence_. limit theorems in probability , statistics and number theory , 4957 , springer proc .
42 , springer , heidelberg , 2013 . | we determine the distribution of nearest neighbour spacings between the tangencies to a fixed circle in a class of circle packings generated by reflections .
we use a combination of geometric tools and the theory of automorphic forms .
320755 ] |
hyper - kamiokande ( hk , @xcite ) is a proposed next - generation water cherenkov detector consisting of two cylindrical tanks with a fiducial volume of @xmath1kt each . the first tank is expected to start data - taking in 2026 , with the second tank following six years later .
hk has a broad physics programme covering many areas of particle and astroparticle physics , including oscillation of accelerator and atmospheric neutrinos , nucleon decay , solar neutrinos , supernova burst and relic neutrinos , indirect detection of dark matter through wimp annihilation and neutrino geophysics . for a galactic supernova at a fiducial distance of @xmath2kpc
, hk will detect @xmath0 neutrinos within 10s .
this high event rate enables hk to resolve fast time variations of the event rate , which could give us information on properties of the progenitor ( like its rotation ) or on details of the supernova explosion mechanism which are currently still unclear . in section [ sec :
sasi ] of this paper , we will describe a feature of supernova explosions that causes such fast time variations and we introduce a method of using this energy information . to determine whether this method can be successful in the presence of detector effects , we have started work on a detailed detector simulation , which is described in section [ sec : simulation ] .
to date , only two dozen supernova burst neutrinos have been detected , all from sn1987a @xcite , and progress in our understanding of the supernova explosion mechanism has come mostly from computer simulations . in 2003 ,
blondin and others @xcite reported that small perturbations of a spherical shock front can be self - sustaining and lead to rapidly growing turbulence behind the shock .
this feature , which has been called standing accretion shock instability ( sasi ) , has since played a crucial role in facilitating explosions in some two- and three - dimensional supernova simulations , but remained subdominant in others , leaving the role it plays in realistic supernova explosions the subject of debate @xcite . observing it would allow us to discriminate between different supernova models and improve our understanding of the explosion mechanism .
sasi is characterized by oscillations of both the number flux and the mean energy of supernova burst neutrinos with a frequency of @xmath3hz over a period of a few hundred ms .
previous work found that for an ideal scenario , these sasi oscillations of the number flux are clearly detectable in both icecube and hk @xcite .
however , the amplitude of these oscillations is direction - dependent and may be much smaller if the detectors on earth are at a large angle to the sasi plane . to improve our chances of detecting sasi in such a scenario
, we can make use of the additional event - by - event energy information that is available in hk but not in icecube .
we use data from a simulation by the garching group @xcite of a @xmath4 progenitor @xcite
. the preprocessed data set used here ( see appendix a of reference @xcite ) contains the mean energy , mean squared energy and luminosity of @xmath5 , @xmath6 and @xmath7 ( all other flavors ) during the first 550ms after the core - bounce for the observer direction with maximal sasi amplitude .
( green , dotted ) , @xmath6 ( blue , dashed ) and @xmath7 ( red , solid ) .
right : event rate in hk without oscillations ( blue , dashed ) and with full flavor swap between @xmath6 and @xmath7 ( red , solid ) .
sasi oscillations are visible at around 220ms and 500ms . ]
( green , dotted ) , @xmath6 ( blue , dashed ) and @xmath7 ( red , solid ) .
right : event rate in hk without oscillations ( blue , dashed ) and with full flavor swap between @xmath6 and @xmath7 ( red , solid ) .
sasi oscillations are visible at around 220ms and 500ms . ]
figure [ fig : sasi - oscillations ] displays the event rate expected in hk and the mean neutrino energy .
the sasi oscillations of both are in phase .
therefore , when the total event rate is high , the mean energy is also high , leading to a very high number of high - energy events ; while when the total event rate is low , the mean energy is also low , leading to a very low number of high - energy events .
accordingly , the amplitude of the sasi oscillations of the event rate will be larger in the highest - energy bin than in the total signal .
on the other hand , that high - energy bin contains fewer events , leading to a higher relative shot noise .
overall , considering the high - energy bin instead of the total signal is advantageous if the oscillations of the mean energy detected in hk are slightly higher than those of @xmath6 in this supernova simulation @xcite .
after showing that it is in principle possible to use the event - by - event energy information available in hk , we need to determine whether this can realistically be achieved in the presence of detector effects like backgrounds and finite energy uncertainty . to investigate this ,
we have started work on a detailed simulation of hk s response to a supernova neutrino burst .
we extract the neutrino flux from the data set described in section [ sec : sasi ] and multiply it with the energy - dependent inverse beta decay ( ibd ) cross - section @xcite to get the number of ibd events . since ibd is the dominant process for supernova neutrinos in hk , making up about 90% of the events , we scale this by a factor of @xmath8 to get an approximation for the total number of events .
signal of a 50mev positron from an inverse beta decay event in hk , simulated by wcsim .
the green dot at the centre of the detector marks the location of the event , while the orange and red dots signify pmt hits .
a faint cherenkov ring is visible . ]
we distribute the resulting positrons randomly within the detector volume and simulate the detector response with wcsim @xcite , a geant4-based application for simulating water cherenkov detectors ( see figure [ fig : wcsim ] ) .
we also simulate backgrounds , most importantly dark noise events from the pmts .
for reconstruction we use bonsai , a software fitter that was originally developed for super - kamiokande @xcite .
in a galactic supernova , sasi can lead to oscillations of the neutrino number flux that are detectable in hk .
the amplitude of these oscillations is larger for higher neutrino energies and under some conditions , it is theoretically advantageous to consider just a subset of the signal . to test
whether this holds up in the presence of detector effects , we are working on a detailed simulation of hk s response to a supernova neutrino burst and expect to present first results soon .
13 see talk by f. di lodovico in these proceedings .
slides are at http://neutrino2016.iopconfs.org/iop/media/uploaded/eviop/event_948/hyper-kamiokande_noamination.pdf hirata k _ et al .
_ 1987 _ phys . rev .
lett . _ * 58 * 14903 .
bionta r m _ et al .
_ 1987 _ phys .
lett . _ * 58 * 14946 .
alekseev e n , alekseeva l n , volchenko v i and krivosheina i v 1987 _ j. exp .
lett . _ * 45 * 58992 .
blondin j m , mezzacappa a and demarino c 2003 _ astrophys .
j. _ * 584 * 971980 hanke f , mller b , wongwathanarat a , marek a and janka h - t 2013 _ astrophys .
j. _ * 770 * 66 burrows a , dolence j c and murphy j w 2012 _ astrophys .
j. _ * 759 * 5 lund t , marek a , lunardini c , janka h - t and raffelt g 2010 _ phys .
_ d * 82 * 063007 tamborra i , hanke f , mller b , janka h - t and raffelt g 2013 _ phys .
lett . _ * 111 * 121104 for raw output of the simulations , see https://wwwmpa.mpa-garching.mpg.de/ccsnarchive/archive.html tamborra i , raffelt g , hanke f , janka h - t and mller b 2014 _ phys .
_ d * 90 * 045032 woosley s e , heger a and weaver t a 2002 _ rev .
* 74 * 101571 migenda j 2015 _ preprint _ 1609.04286 strumia a and vissani f 2003 _ phys .
_ b * 564 * 4254 https://github.com/wcsim/wcsim smy m 2007 _ proc .
. cosmic ray conf .
( mrida , mexico ) _ pp 127982 | hyper - kamiokande is a proposed next - generation water cherenkov detector .
if a galactic supernova happens , it will deliver a high event rate ( @xmath0 neutrino events in total ) as well as event - by - event energy information .
recent supernova simulations exhibit the standing accretion shock instability ( sasi ) which causes oscillations in the number flux and mean energy of neutrinos .
the amplitude of these oscillations is energy - dependent , so the energy information available in hyper - kamiokande could be used to improve the detection prospects of these sasi oscillations . to determine whether this can be achieved in the presence of detector effects like backgrounds and finite energy uncertainty , we have started work on a detailed simulation of hyper - kamiokande s response to a supernova neutrino burst |
the temporal nature of interference in wireless networks depends on the underlying traffic as well as the subcarrier allocations of neighbouring base stations ( which usually employ multicarrier systems like ofdm ) . in practice ,
due to the bursty nature of data traffic and uncoordinated subcarrier allocations across base stations , the resulting interference at the physical layer tends to be bursty .
in addition to the potential for harnessing such burstiness , feedback from the receivers is another resource available in wireless networks . with these motivations , in this paper
we study parallel ( multicarrier ) interference channels with bursty interference links and output feedback from the receivers . in @xcite and @xcite ,
the problem of harnessing bursty interference was studied for a single carrier setup without feedback .
a multicarrier version of @xcite was studied in @xcite . to study benefits of feedback
, @xcite considered a single carrier setup with bursty interference and output feedback from the receivers . in @xcite ,
bursty interference was modeled using a bernoulli random state ( instantiated i.i.d . over time ) and a complete capacity characterization was given for the linear deterministic setup . in this paper , we study the multicarrier version of @xcite _ i.e. , _ output feedback in multicarrier systems with bursty interference . since @xcite developed optimal single carrier schemes , a natural question arises in the multicarrier version : is it always optimal to treat each subcarrier _ separately _ and just copy the optimal scheme in @xcite on each subcarrier ? as the following example illustrates , such a separation may not be always optimal . [ [ toy - example ] ] toy example + + + + + + + + + + + consider two parallel symmetric @xmath0-user linear deterministic interference channels ( ldics ) @xcite as shown in figure [ fig : toy_example ] .
the first subcarrier has one direct link ( @xmath1 ) and one interfering link ( @xmath2 , hence @xmath3 ) and the second subcarrier has one direct link and three interfering links ( @xmath4 ) .
causal output feedback is available from the receivers to the respective transmitters .
bernoulli random states @xmath5 $ ] and @xmath6 $ ] indicate the presence of interference in the first and second subcarrier respectively and are instantiated i.i.d .
( over time ) from an arbitrary joint distribution @xmath7 . for this example
, we assume the expectation of both the states to be @xmath8 . subcarriers . ]
our goal here is to find the maximum achievable symmetric rate .
using the optimal single carrier schemes in @xcite , we can achieve symmetric rate @xmath9 from the first subcarrier and symmetric rate @xmath10 from the second subcarrier .
summing these rates , we can achieve a total symmetric rate @xmath11 .
now , we will show that rate @xmath12 is achievable by coding _ across _ the subcarriers rather than treating the subcarriers separately .
we use a block based pipelined scheme ( block length @xmath13 ) as follows .
the transmitters always send fresh symbols in the first subcarrier ( @xmath14-symbols for @xmath15 and @xmath16-symbols for @xmath17 as shown in figure [ fig : toy_example ] ) . in the first subcarrier , for sufficiently large @xmath13 , with high probability ( w.h.p . ) only @xmath18 @xmath14-symbols in a block get interfered at @xmath15 ( and @xmath18 @xmath16-symbols at @xmath17 ) . at the end of a block , due to feedback from @xmath15
, @xmath19 knows exactly which of its transmitted @xmath14-symbols caused interference at @xmath20 ( since the same state variable @xmath5 $ ] holds for both the receivers ) . for the next block ,
@xmath19 creates @xmath13 linear combinations of these @xmath18 @xmath14-symbols ( which caused interference at @xmath17 in the previous block ) and sends these @xmath13 linear combinations as @xmath21 $ ] ( in the second subcarrier ) over the next @xmath13 time slots .
due to bursty interference , w.h.p .
only @xmath18 of these linear combinations appear at @xmath17 ; but this is sufficient to decode @xmath18 @xmath14-symbols constituting the linear combinations . using these @xmath14-symbols @xmath17 can now recover all the interfered @xmath16-symbols in the previous block and hence achieve rate @xmath22 from the first subcarrier ( same for @xmath15 due to symmetry ) . for the remaining levels in the second subcarrier ,
the following is done : lowest levels are not used ( @xmath23 = d_3[t]=0 $ ] ) , and the transmitters send fresh symbols in the highest level which appear interference free at the receivers ( as the lowest levels are not used ) .
this leads to an additional rate @xmath22 from the second subcarrier .
adding rates from the two subcarriers , we achieve symmetric rate @xmath0 .
this is in fact the symmetric capacity ; an easy consequence of the outer bounds developed in this paper .
the above example demonstrates a _ helping _ mechanism ; the second subcarrier _ helped _ the first subcarrier in recovering interfered symbols in a pipelined fashion .
in this paper , we generalize this idea for an arbitrary collection of subcarriers with the following constraint : interference states across subcarriers are drawn from an arbitrary joint distribution ( instantiated i.i.d . over time ) and the marginal probability of interference is same for each subcarrier . the main idea behind the generalization is to use specific _ levels _ in very strongly interfered subcarriers to recover interfered signals for strongly and weakly interfered subcarriers in a pipelined fashion as shown in the toy example .
another aspect captured by the toy example is the importance of burstiness ; subcarriers in the above example are separable ( due to our results and @xcite ) when interference is always present .
hence , the proposed helping mechanism owes its relevance to bursty interference .
our main contributions are as follows : * in the linear deterministic setup , we have a complete capacity region characterization . in the setup with gaussian noise
, we have a tight generalized degrees of freedom ( gdof ) characterization and provide outer bounds on the capacity region . *
the inner and outer bounds are non - trivial extensions of single carrier results @xcite .
we identify regimes where treating subcarriers separately is optimal .
for the remaining regimes , we employ coding across subcarriers ( helping mechanism ) to achieve tight results .
the outer bounds involve a subset entropy inequality by madiman and tetali @xcite .
the remainder of this paper is organized as follows .
section [ sec : model ] deals with the notation and setup .
section [ sec : main_results ] summarizes the main results of this paper .
this is followed by section [ sec : outer_bounds ] on outer bounds and section [ sec : inner_bounds ] on inner bounds .
we conclude the paper with section [ sec : gdof ] on the gdof characterization .
we consider a system with two base stations ( transmitters ) @xmath19 and @xmath24 , and two users ( receivers ) @xmath15 and @xmath17 . for @xmath25
, @xmath26 has message @xmath27 for @xmath28 .
there are @xmath29 parallel channels from @xmath26 to @xmath28 ( subcarriers indexed by @xmath30 ) . in this paper , we consider two setups for the subcarrier channel : the first one is based on the linear deterministic model @xcite ( ld setup ) , and the second one is based on the gaussian interference channel ( gn setup ) . the subcarrier channel model for both the setups , followed by the statistics of bursty interference and rate requirements are described below .
[ [ subcarrier - channel - model ] ] subcarrier channel model + + + + + + + + + + + + + + + + + + + + + + + + in the ld setup , each subcarrier is modeled by a @xmath0-user ( symmetric ) ldic @xcite with a bursty interfering link ( explained below ) and feedback from respective receivers . at discrete time index @xmath31 ,
the transmitted signal in subcarrier @xmath32 of @xmath26 is @xmath33 \in \mathbb{f}^{q_j}$ ] where @xmath34 is a finite field .
the received signal in subcarrier @xmath32 at @xmath28 is given by : @xmath35 = \mathbf{g}_j^{q_j - n_j}\mathbf{x}^{(i)}_j[t ] + s_j[t ] \mathbf{g}_j^{q_j - k_j}\mathbf{x}^{(i')}_j[t ] \end{aligned}\ ] ] where @xmath36 is a @xmath37 shift matrix in the terminology of deterministic channel models @xcite , @xmath38 $ ] is a bernoulli random variable ( details in interference statistics below ) determining the presence of interference in subcarrier @xmath32 at time index @xmath39 , @xmath40 $ ] denotes the transmitted signal on subcarrier @xmath32 of user @xmath41 , and parameters @xmath42 and @xmath43 represent the direct and interfering link strengths @xcite in subcarrier @xmath32 .
figure [ fig : system_model ] shows the channel model for subcarrier @xmath32 . without loss of generality , we assume @xmath44 and let @xmath45 denote the normalized strength of the interfering signal in subcarrier @xmath32 .
for every time instant , it is convenient to consider a subcarrier as indexed levels of bit pipes @xcite ; each bit pipe carries a symbol from @xmath34 . in ld
setup : @xmath42 and @xmath43 represent direct and interfering link strengths .
presence of interference at time index @xmath39 is determined by bernoulli random variable @xmath38 $ ] . ] in the gn setup , at discrete time index @xmath31 , the transmitted signal in subcarrier @xmath32 of @xmath26 is @xmath46 \in \mathbb{c}$ ] , such that @xmath47|^2 \leq 1 $ ] . the received signal in subcarrier @xmath32 at @xmath28 is given by : @xmath48 = g_{d , j } x^{(i)}_j[t ] + s_j[t ] g_{i , j } x^{(i')}_j[t ] + z^{(i)}_j[t ] \end{aligned}\ ] ] where @xmath49 denote the direct and interfering channel gains , and @xmath50 \sim \mathcal{cn}(0,1)$ ] is gaussian noise . as in the ld setup ,
@xmath38 $ ] is the interference state . in both ld and gn setups
, @xmath26 receives causal feedback from @xmath28 ( feedback consists of the received signal and the interference state ) .
[ [ interference - statistics ] ] interference statistics + + + + + + + + + + + + + + + + + + + + + + + we consider the same interference statistics for both ld and gn setups .
as described above , the presence of interference in subcarrier @xmath32 at time index @xmath39 is given by a bernoulli random variable @xmath38 $ ] ( takes values in @xmath51 ) .
the @xmath29 bernoulli random variables @xmath52 , s_2[t],\ldots s_m[t]\}$ ] have a joint probability distribution @xmath53=s_1,s_2[t]=s_2 , \ldots s_m[t ] = s_m)=\mathbb{p}(s_1=s_1,s_2=s_2 , \ldots s_m = s_m ) $ ] instantiated i.i.d . over time . in this paper
, we restrict the analysis to joint distributions with the same marginal probabilities for every @xmath38 $ ] _
@xmath54 ) = p$ ] .
the transmitters are assumed to know the above statistics , but are limited to causal information on the interference realizations in the subcarriers ( through feedback ) .
[ [ rate - requirements ] ] rate requirements + + + + + + + + + + + + + + + + + we consider the same rate requirements for both ld and gn setups .
base station @xmath26 intends to send message @xmath27 to @xmath28 over @xmath55 time slots ( time index @xmath31 ) .
rate @xmath56 ( corresponding to @xmath27 ) is considered achievable if the probability of decoding error is vanishingly small as @xmath57 .
the capacity region for @xmath58 in the ld setup is given by the following rate inequalities , @xmath59 where @xmath60 and @xmath61 , and @xmath62 .
as shown in figure [ fig : crossover ] , the shape of the capacity region depends on the value of @xmath63 .
an intuitive interpretation of @xmath63 comes from our inner bounds ; @xmath64 implies there are enough levels in subcarriers with @xmath65 ( very strong interference ) to recover the interfered signals for subcarriers with @xmath66 ( weak interference ) and @xmath67 ( strong interference ) . and @xmath64 .
the dashed line representing inequality ( [ eq : ob_3 ] ) is active only when @xmath68 .
symmetric capacity ( @xmath69 ) for @xmath70 and @xmath71 is given by @xmath72 and @xmath73 respectively . ]
the details of the rate tuples @xmath74 marked in figure [ fig : crossover ] are listed below : + @xmath75 @xmath76 : @xmath77 + @xmath75 @xmath78 + @xmath75 @xmath79 + @xmath75 @xmath80 + @xmath75 @xmath81 + @xmath75 @xmath82 + @xmath75 @xmath83 + @xmath75 @xmath84 : @xmath85 .
the following rate inequalities are outer bounds on achievable @xmath58 in the gn setup . @xmath86 where @xmath60 and @xmath61 , and @xmath87 . in the gn setup , assuming @xmath88 , @xmath89 and @xmath90 ( rational @xmath91 ) , @xmath92 where @xmath69 denotes the symmetric capacity and @xmath93 .
[ cor : separate ] in the ld setup , for achieving symmetric capacity , treating subcarriers separately is optimal when all @xmath94 or all @xmath95 ( and for the degenerate case of @xmath96 ) . for the remaining cases , coding across subcarriers achieves symmetric capacity .
similarly , in the gn setup , treating subcarriers separately is gdof optimal when all @xmath97 or all @xmath98 .
in this section , we focus on proofs of outer bounds in the ld and gn setups . we refer to outer bounds ( [ eq : causal_ld ] ) and ( [ eq : causal_gn ] ) as causal outer bounds as they account for the causal knowledge of subcarrier interference states at the transmitter , for @xmath99 , the symmetric capacity @xmath100 stems from causal outer bound ( [ eq : causal_ld ] ) ; hence the subscript `` c '' for causal . ] . for proving these causal outer bounds ,
we use a subset entropy inequality by madiman and tetali which we describe in section [ sec : madiman_tetali ] , prior to the proofs .
then we introduce some additional notation in section [ sec : add_notation ] followed by outer bound proofs for the ld setup ( section [ sec : ld_ob ] ) and gn setup ( section [ sec : gn_ob ] ) .
we now describe a subset entropy inequality by madiman and tetali @xcite .
consider a hypergraph @xmath101 where @xmath102 is a finite ground set and @xmath103 is a collection of subsets of @xmath102 .
a function @xmath104 is called a fractional partition of @xmath101 if it satisfies the following condition @xmath105 .
@xmath106 with the above definition , the subset entropy inequality can now be stated as follows , @xmath107 where @xmath108 is a fractional partition and the above inequality holds for any collection of jointly distributed random variables @xmath109 .
the differential entropy version of the above inequality has the same form @xcite . to use these inequalities in our setups ,
we first choose a suitable fractional partition as explained below .
for @xmath110 , let @xmath111= \left ( s_1[t],\ ; s_2[t],\ldots s_m[t ] \right ) = \mathbf{s}$ ] denote the collection of interference states of all the @xmath29 subcarriers at time index @xmath39 . as specified in section [ sec : model ] , the occurrence of @xmath111=\mathbf{s}$ ] is governed by the joint probability distribution @xmath112=\mathbf{s})$ ] . to define a fractional partition
, we consider the ground set @xmath113 ( _ i.e. , _ the index set of subcarriers ) and view @xmath114 as a collection of @xmath29 indicator functions for representing any subset of @xmath102 .
the power set of @xmath102 ( excluding subsets @xmath115 such that @xmath112=\mathbf{s})=0 $ ] ) is chosen as set @xmath103 .
now , we define a fractional partition @xmath116 as follows .
@xmath117=\mathbf{s}_e ) } { p } \label{eq : fractional_cover_setup}\end{aligned}\ ] ] where @xmath118 and @xmath119 denotes the joint state where only the subcarriers whose index is in set @xmath120 face interference .
the fractional partition condition holds as follows .
@xmath121=\mathbf{s}_e ) } { p } = \frac{\mathbb{e}(s_j[t])}{p}=1\end{aligned}\ ] ] in section [ sec : ld_causal ] , we demonstrate the application of inequality ( [ eq : madiman_tetali ] ) , in conjunction with the fractional partition defined in ( [ eq : fractional_cover_setup ] ) , for proving outer bound ( [ eq : causal_ld ] ) .
similarly , in section [ sec : causal_gn ] , for proving outer bound ( [ eq : causal_gn ] ) we use the differential entropy version @xcite of inequality ( [ eq : madiman_tetali ] ) with the same fractional partition . for notational convenience ,
we use indicator functions @xmath122 and @xmath123 to denote the absence and presence of interference in subcarrier @xmath32 when the joint state realization across @xmath29 subcarriers is @xmath111=\mathbf{s } \in \{0,1\}^m$ ] .
also , in the proofs we use @xmath124 to denote @xmath125 . for convenience ,
we have listed all the additional notation used for outer bound proofs in ld and gn setups ( some notation is common to both setups ) .
notation used in ld setup proofs : * @xmath126,\;\mathbf{s}[2],\;\ldots \mathbf{s}[t ] ) $ ] .
* @xmath127 = ( \mathbf{y}^{(i)}_1[t],\;\mathbf{y}^{(i)}_2[t],\ldots \mathbf{y}^{(i)}_m[t])$ ] _
i.e. , _ received signal ( across @xmath29 subcarriers ) for @xmath28 at time index @xmath39 . *
@xmath128 $ ] : received signal ( across @xmath29 subcarriers ) for @xmath28 at time @xmath39 when @xmath111=\mathbf{s}$ ] .
the difference between @xmath127 $ ] and @xmath128 $ ] is that the state at time @xmath39 is assumed to be @xmath115 in the latter . *
@xmath129,\;\mathbf{y}^{(i)}[2],\;\ldots\mathbf{y}^{(i)}[t])$ ] . * @xmath130 $ ] : interfering signals ( across @xmath29 subcarriers ) for @xmath28 when @xmath111=\mathbf{s}$ ] .
* @xmath131}^{(i)}[1],\ ; \mathbf{v}_{\mathbf{s}[2]}^{(i)}[2 ] , \ldots \mathbf{v}_{\mathbf{s}[t]}^{(i)}[t])$ ] . * @xmath132 $ ] : interfering signals ( across @xmath29 subcarriers ) at @xmath28 when all its subcarriers face interference at time index @xmath39 .
this is equivalent to @xmath130 $ ] with @xmath133 . *
@xmath134 $ ] : received signal ( across @xmath29 subcarriers ) at @xmath28 when all its subcarriers are interference free at time index @xmath39 .
this is equivalent to @xmath128 $ ] with @xmath135 .
notation used in gn setup proofs : * @xmath126,\;\mathbf{s}[2],\;\ldots \mathbf{s}[t ] ) $ ] .
* @xmath127 = ( y^{(i)}_1[t],\ ; y^{(i)}_2[t],\ldots y^{(i)}_m[t])$ ] _
i.e. , _ received signal ( across @xmath29 subcarriers ) for @xmath28 at time index @xmath39 . *
@xmath128 $ ] : received signal ( across @xmath29 subcarriers ) for @xmath28 at time @xmath39 when @xmath111=\mathbf{s}$ ] .
the difference between @xmath127 $ ] and @xmath128 $ ] is that the state at time @xmath39 is assumed to be @xmath115 in the latter .
* @xmath129,\;\mathbf{y}^{(i)}[2],\;\ldots\mathbf{y}^{(i)}[t])$ ] .
* @xmath136=(z^{(i)}_1[t],\ ; z^{(i)}_2[t],\ldots z^{(i)}_m[t])$ ] _
i.e. , _ receiver noise ( across @xmath29 subcarriers ) for @xmath28 at time index @xmath39 .
* @xmath137 $ ] : receiver noise in interfered subcarriers for @xmath28 at time index @xmath39 when @xmath111=\mathbf{s}$ ] . *
@xmath138 $ ] : receiver noise in interference free subcarriers for @xmath28 at time index @xmath39 when @xmath111=\mathbf{s}$ ] .
* @xmath130 \oplus \mathbf{z}^{(i)}[t]= ( s_1[t ] g_{i,1 } x^{(i')}_1[t ] + z^{(i)}_1[t ] , \;s_2[t ] g_{i,2 } x^{(i')}_2[t ] + z^{(i)}_2[t ] , \ldots s_m[t ] g_{i , m } x^{(i')}_m[t ] + z^{(i)}_m[t ] ) $ ]
_ i.e. , _ interfering signal ( if present ) plus noise , across @xmath29 subcarriers , for @xmath28 at time index @xmath39 when @xmath111=\mathbf{s}$ ] .
* @xmath130 \oplus \mathbf{z}_{\mathbf{s}}^{(i)}[t]$ ] : interfering signal plus noise in interfered subcarriers for @xmath28 at time index @xmath39 when @xmath111=\mathbf{s}$ ] .
note that this does not include the noise terms for subcarriers which do not face interference at time @xmath39 ( unlike @xmath130 \oplus \mathbf{z}^{(i)}[t]$ ] ) . *
@xmath139}^{(i)}[1 ] \oplus \mathbf{z}^{(i)}[1 ] , \ ; \mathbf{v}_{\mathbf{s}[2]}^{(i)}[2 ] \oplus \mathbf{z}^{(i)}[2 ] , \ldots \mathbf{v}_{\mathbf{s}[t]}^{(i)}[t ] \oplus \mathbf{z}^{(i)}[t])$ ] .
* @xmath132 \oplus \mathbf{z}^{(i)}[t]$ ] : interfering signal plus noise ( across @xmath29 subcarriers ) at @xmath28 when all its subcarriers face interference at time index @xmath39 .
this is equivalent to @xmath130 \oplus \mathbf{z}^{(i)}[t]$ ] with @xmath133 .
* @xmath134\oplus \mathbf{z}^{(i)}[t]$ ] : received signal ( across @xmath29 subcarriers ) at @xmath28 when all its subcarriers are interference free at time index @xmath39 .
this is equivalent to @xmath128 $ ] with @xmath135 .
* @xmath140\oplus \mathbf{z}^{(i)}[1],\;\hat{\mathbf{x}}^{(i)}[2]\oplus \mathbf{z}^{(i)}[2],\ldots \hat{\mathbf{x}}^{(i)}[t]\oplus \mathbf{z}^{(i)}[t])$ ] .
see appendix [ sec : ld_ob_r1 ]
. using fano s inequality for @xmath15 , for any @xmath141 , there exists a large enough @xmath55 such that ; @xmath142 | \mathbf{y}^{(1)}_{1:t-1 } , \mathbf{s}_{1:t-1 } , \mathbf{s}[t ] ) \nonumber\\ & = \sum_{t=1}^{n } \sum_{\mathbf{s } } \mathbb{p}(\mathbf{s}[t]=\mathbf{s } ) h ( \mathbf{y}_{\mathbf{s}}^{(1)}[t ] | \mathbf{y}^{(1)}_{1:t-1 } , \mathbf{s}_{1:t-1 } , \mathbf{s}[t]=\mathbf{s } ) \nonumber \\ & \quad - \sum_{t=1}^{n } \sum_{\mathbf{s } } \mathbb{p}(\mathbf{s}[t]=\mathbf{s } ) h ( \mathbf{y}_{\mathbf{s}}^{(1)}[t ] | w^{(1 ) } , \mathbf{y}^{(1)}_{1:t-1 } , \mathbf{s}_{1:t-1 } , \mathbf{s}[t]= \mathbf{s } ) \nonumber\\ & \leq \sum_{t=1}^{n } \sum_{\mathbf{s } } \mathbb{p}(\mathbf{s}[t]=\mathbf{s } ) \sum_{j=1}^{m } n_j \mathbb{i}_{j \not \in \mathbf{s } } + \max(n_j , k_j ) \mathbb{i}_{j \in \mathbf{s } } \nonumber \\ & \quad - \sum_{t=1}^{n } \sum_{\mathbf{s } } \mathbb{p}(\mathbf{s}[t]=\mathbf{s } ) h ( \mathbf{v}_{\mathbf{s}}^{(1)}[t ] | w^{(1 ) } , \mathbf{y}^{(1)}_{1:t-1 } , \mathbf{s}_{1:t-1 } , \mathbf{s}[t]=\mathbf{s } ) \nonumber\\ & = \sum_{t=1}^{n } \sum_{j=1}^{m } \sum _ { \mathbf{s } } n_j \mathbb{p}(\mathbf{s}[t]=\mathbf{s } ) \mathbb{i}_{j \not \in \mathbf{s } } + \max(n_j , k_j ) \mathbb{p}(\mathbf{s}[t]=\mathbf{s } ) \mathbb{i}_{j \in \mathbf{s } } \nonumber\\ & \quad - \sum_{t=1}^{n } \sum_{\mathbf{s } } \mathbb{p}(\mathbf{s}[t]=\mathbf{s } ) h ( \mathbf{v}_{\mathbf{s}}^{(1)}[t ] | w^{(1 ) } , \mathbf{y}^{(1)}_{1:t-1 } , \mathbf{s}_{1:t-1 } , \mathbf{s}[t]=\mathbf{s } ) \nonumber\\ & = n \sum_{j=1}^{m } ( 1-p ) n_j + p\max(n_j , k_j ) \nonumber \\ & \quad- \sum_{t=1}^{n } \sum_{\mathbf{s } } \mathbb{p}(\mathbf{s}[t]=\mathbf{s } ) h ( \mathbf{v}_{\mathbf{s}}^{(1)}[t ] | w^{(1 ) } , \mathbf{y}^{(1)}_{1:t-1 } , \mathbf{s}_{1:t-1 } , \mathbf{s}[t]=\mathbf{s } ) \nonumber\end{aligned}\ ] ] @xmath143=\mathbf{s } ) h ( \mathbf{v}_{\mathbf{s}}^{(1)}[t ] | w^{(1 ) } , \mathbf{v}^{(1)}_{1:t-1 } , \mathbf{s}_{1:t-1 } ) \nonumber\\ & \stackrel{(a)}\leq n \sum_{j=1}^{m } ( 1-p ) n_j + p\max(n_j , k_j ) - p \sum_{t=1}^{n } h ( \tilde{\mathbf{v}}^{(1)}[t ] | w^{(1 ) } , \mathbf{v}^{(1)}_{1:t-1 } , \mathbf{s}_{1:t-1 } )
\nonumber\\ & = n \sum_{j=1}^{m } n_j + ( \max(n_j , k_j)-n_j ) p - p \sum_{t=1}^{n } h ( \tilde{\mathbf{v}}^{(1)}[t ] | w^{(1 ) } , \mathbf{v}^{(1)}_{1:t-1 } , \mathbf{s}_{1:t-1 } ) \label{eq : causal_1}\end{aligned}\ ] ] where ( a ) follows by using ( [ eq : madiman_tetali ] ) for the fractional partition defined in ( [ eq : fractional_cover_setup ] ) . using fano s inequality for @xmath17 , for any @xmath141
, there exists a large enough @xmath55 such that ; @xmath144 , \mathbf{y}^{(1)}[t ] | \mathbf{y}^{(2)}_{1:t-1 } , \mathbf{y}^{(1)}_{1:t-1 } , \mathbf{s}_{1:t-1 } , w^{(1 ) } , \mathbf{s}[t ] ) \nonumber\\ & = \sum_{t=1}^{n } \sum _ { \mathbf{s } } \mathbb{p}(\mathbf{s}[t]=\mathbf{s } ) ( h ( \mathbf{y}_{\mathbf{s}}^{(2)}[t],\mathbf{y}_{\mathbf{s}}^{(1)}[t ] | \mathbf{y}^{(2)}_{1:t-1 } , \mathbf { y}^{(1)}_{1:t-1 } , \mathbf{s}_{1:t-1 } , w^{(1 ) } , \mathbf{s}[t]=\mathbf{s } ) \nonumber\\ & \quad - h ( \mathbf{y}_{\mathbf{s}}^{(2)}[t ] , \mathbf{y}_{\mathbf{s}}^{(1)}[t ] | \mathbf{y}^{(2)}_{1:t-1 } , \mathbf{y}^{(1)}_{1:t-1 } , \mathbf{s}_{1:t-1 } , w^{(1 ) } , w^{(2 ) } , \mathbf{s}[t]= \mathbf{s } ) ) \nonumber\\ & = \sum_{t=1}^{n } \sum_{\mathbf{s } } \mathbb{p}(\mathbf{s}[t]=\mathbf{s } ) h ( \mathbf{y}_{\mathbf{s}}^{(2)}[t ] , \mathbf{y}_{\mathbf{s}}^{(1)}[t ] | \mathbf{y}^{(2)}_{1:t-1 } , \mathbf{y}^{(1)}_{1:t-1 } , \mathbf{s}_{1:t-1 } , w^{(1 ) } , \mathbf{s}[t]=\mathbf{s } ) \nonumber\\ & = \sum_{t=1}^{n } \sum_{\mathbf{s } } \mathbb{p}(\mathbf{s}[t]=\mathbf{s } ) h ( \hat{\mathbf{x}}^{(2)}[t ] , \mathbf{v}_{\mathbf{s}}^{(1)}[t ] | \mathbf{y}^{(2)}_{1:t-1 } , \mathbf{y}^{(1)}_{1:t-1 } , \mathbf{s}_{1:t-1 } , w^{(1 ) } , \mathbf{s}[t]=\mathbf{s } ) \nonumber\\ & \leq \sum_{t=1}^{n } \sum_{\mathbf{s } } \mathbb{p}(\mathbf{s}[t]=\mathbf{s } ) h ( \hat{\mathbf{x}}^{(2)}[t ] , \tilde{\mathbf{v}}^{(1)}[t ] | \mathbf{y}^{(2)}_{1:t-1 } , \mathbf{y}^{(1)}_{1:t-1 } , \mathbf{s}_{1:t-1 } , w^{(1 ) } , \mathbf{s}[t]=\mathbf{s})\nonumber\\ & = \sum_{t=1}^{n } h ( \hat{\mathbf{x}}^{(2)}[t ] , \tilde{\mathbf{v}}^{(1)}[t ] | \mathbf{y}^{(2)}_{1:t-1 } , \mathbf{y}^{(1)}_{1:t-1 } , \mathbf{s}_{1:t-1 } , w^{(1 ) } ) \sum_{\mathbf{s } } \mathbb{p}(\mathbf{s}[t]=\mathbf{s } ) \nonumber\\ & = \sum_{t=1}^{n } h ( \hat{\mathbf{x}}^{(2)}[t ] , \tilde{\mathbf{v}}^{(1)}[t ] | \mathbf{y}^{(2)}_{1:t-1 } , \mathbf{y}^{(1)}_{1:t-1 } , \mathbf{s}_{1:t-1 } , w^{(1 ) } ) \nonumber\\ & \leq \sum_{t=1}^{n } h ( \hat{\mathbf{x}}^{(2)}[t ] , \tilde{\mathbf{v}}^{(1)}[t ] | \mathbf{y}^{(1)}_{1:t-1 } , \mathbf{s}_{1:t-1 } , w^{(1 ) } ) \nonumber\\ & = \sum_{t=1}^{n } h ( \hat{\mathbf{x}}^{(2)}[t ] , \tilde{\mathbf{v}}^{(1)}[t ] | \mathbf{v}^{(1)}_{1:t-1 } , \mathbf{s}_{1:t-1 } , w^{(1 ) } ) \label{eq : causal_2}\end{aligned}\ ] ] using inequalities ( [ eq : causal_1 ] ) and ( [ eq : causal_2 ] ) , @xmath145 | \tilde{\mathbf{v}}^{(1)}[t ] , w^{(1 ) } , \mathbf{v}^{(1)}_{1:t-1 } , \mathbf{s}_{1:t-1 } ) \nonumber\\ & \leq n \sum_{j=1}^{m } n_j + ( \max(n_j , k_j)-n_j ) p + p \sum_{t=1}^{n } h ( \hat{\mathbf{x}}^{(2)}[t ] | \tilde{\mathbf{v}}^{(1)}[t ] ) \nonumber\\ & \leq n \sum_{j=1}^{m } n_j + ( \max(n_j , k_j)-n_j ) p + p \sum_{t=1}^{n } \sum_{j=1}^{m } ( n_j - k_j)^{+}\nonumber\\ & = n p \delta + n \sum_{j=1}^{m } ( 1+p ) n_j % \sum_{i=1}^{m } \max(n_i , k_i ) \nonumber \\ & \quad + \max(n_i - k_i,0 ) - 2n_i\nonumber\\\end{aligned}\ ] ] where @xmath146 .
the bound on @xmath147 follows by symmetry , and this completes the proof of outer bound ( [ eq : causal_ld ] ) .
the above proof demonstrates a connection between subset entropy inequalities and bursty interference in multicarrier systems . in @xcite
, we demonstrated a similar connection by using a sliding window subset entropy inequality @xcite to show tight outer bounds for the case without feedback ( in multicarrier systems with bursty interference ) .
+ see appendix [ sec : ld_ob_r1_r2_sum ] .
see appendix [ sec : gn_r1_ob ] . using fano s inequality for @xmath15 , for any @xmath141 , there exists a large enough @xmath55 such that ; @xmath148 | \mathbf{y}^{(1)}_{1:t-1 } , \mathbf{s}_{1:t-1 } , \mathbf{s}[t ] ) \nonumber\\ & = \sum_{t=1}^{n } \sum_{\mathbf{s } } \mathbb{p}(\mathbf{s}[t]=\mathbf{s } ) h ( \mathbf{y}^{(1)}[t ] | \mathbf{y}^{(1)}_{1:t-1 } , \mathbf{s}_{1:t-1 } , \mathbf{s}[t]=\mathbf{s } ) - \sum_{t=1}^{n } \sum_{\mathbf{s } } \mathbb{p}(\mathbf{s}[t]=\mathbf{s } ) h ( \mathbf{y}^{(1)}[t ] | w^{(1 ) } , \mathbf{y}^{(1)}_{1:t-1 } , \mathbf{s}_{1:t-1 } , \mathbf{s}[t]= \mathbf{s } ) \nonumber\\ & \leq \sum_{t=1}^{n } \sum_{\mathbf{s } } \mathbb{p}(\mathbf{s}[t]=\mathbf{s } ) h ( \mathbf{y}^{(1)}[t ] |\mathbf{s}[t]=\mathbf{s } ) - \sum_{t=1}^{n } \sum_{\mathbf{s } } \mathbb{p}(\mathbf{s}[t]=\mathbf{s } ) h ( \mathbf{y}^{(1)}[t ] | w^{(1 ) } , \mathbf{y}^{(1)}_{1:t-1 } , \mathbf{s}_{1:t-1 } , \mathbf{s}[t]= \mathbf{s } ) \nonumber\\ & \stackrel{(a)}\leq n \sum_{j=1}^{m } ( 1-p ) \log \left ( 1+|g_{d , j}|^2 \right ) + p \log \left ( 1 + \left ( |g_{d , j}| + |g_{i , j}| \right ) ^2 \right ) + nm \log ( \pi e ) \nonumber\\ & \quad - \sum_{t=1}^{n } \sum_{\mathbf{s } } \mathbb{p}(\mathbf{s}[t]=\mathbf{s } ) h ( \mathbf{y}^{(1)}[t ] | w^{(1 ) } , \mathbf{y}^{(1)}_{1:t-1 } , \mathbf{s}_{1:t-1 } , \mathbf{s}[t]= \mathbf{s } ) \nonumber\\ & = n \sum_{j=1}^{m } ( 1-p ) \log \left ( 1+|g_{d , j}|^2 \right ) + p \log \left ( 1 + \left ( |g_{d , j}| + |g_{i , j}| \right ) ^2 \right ) + nm \log ( \pi e ) \nonumber\\ & \quad - \sum_{t=1}^{n } \sum_{\mathbf{s } } \mathbb{p}(\mathbf{s}[t]=\mathbf{s } ) h ( \mathbf{v}_{\mathbf{s}}^{(1)}[t ] \oplus \mathbf{z}^{(1)}[t ] | w^{(1 ) } , \mathbf{y}^{(1)}_{1:t-1 } , \mathbf{s}_{1:t-1 } , \mathbf{s}[t]=\mathbf{s } ) \nonumber\\ & = n \sum_{j=1}^{m } ( 1-p ) \log \left ( 1+|g_{d , j}|^2 \right ) + p \log \left ( 1 + \left ( |g_{d , j}| + |g_{i , j}| \right ) ^2 \right ) + nm \log ( \pi e ) \nonumber\\ & \quad - \sum_{t=1}^{n } \sum_{\mathbf{s } } \mathbb{p}(\mathbf{s}[t]=\mathbf{s } ) h ( \mathbf{v}_{\mathbf{s}}^{(1)}[t ] \oplus \mathbf{z}_{\mathbf{s}}^{(1)}[t ] | w^{(1 ) } , \mathbf{y}^{(1)}_{1:t-1 } , \mathbf{s}_{1:t-1 } , \mathbf{s}[t]=\mathbf{s } ) \nonumber\\ & \quad - \sum_{t=1}^{n } \sum_{\mathbf{s } } \mathbb{p}(\mathbf{s}[t]=\mathbf{s } ) h ( \mathbf{z}_{\mathbf{s}^c}^{(1)}[t ] | \mathbf{v}_{\mathbf{s}}^{(1)}[t ] \oplus \mathbf{z}_{\mathbf{s}}^{(1)}[t ] , w^{(1 ) } , \mathbf{y}^{(1)}_{1:t-1 } , \mathbf{s}_{1:t-1 } , \mathbf{s}[t]=\mathbf{s } )
\nonumber\\ & = n \sum_{j=1}^{m } ( 1-p ) \log \left ( 1+|g_{d , j}|^2 \right ) + p \log \left ( 1 + \left ( |g_{d , j}| + |g_{i , j}| \right ) ^2 \right ) + nm \log ( \pi e ) \nonumber\\ & \quad - \sum_{t=1}^{n } \sum_{\mathbf{s } } \mathbb{p}(\mathbf{s}[t]=\mathbf{s } ) h ( \mathbf{v}_{\mathbf{s}}^{(1)}[t ] \oplus \mathbf{z}_{\mathbf{s}}^{(1)}[t ] | w^{(1 ) } , \mathbf{y}^{(1)}_{1:t-1 } , \mathbf{s}_{1:t-1 } , \mathbf{s}[t]=\mathbf{s } ) - \sum_{t=1}^{n } \sum_{\mathbf{s } } \mathbb{p}(\mathbf{s}[t]=\mathbf{s } ) \sum_{j=1}^m \mathbb{i}_{j \not \in \mathbf{s } } \log(\pi
e ) \nonumber \\ & = n \sum_{j=1}^{m } ( 1-p ) \log \left ( 1+|g_{d , j}|^2 \right ) + p \log \left ( 1 + \left ( |g_{d , j}| + |g_{i , j}| \right ) ^2 \right ) + nm \log ( \pi e ) \nonumber\\ & \quad - \sum_{t=1}^{n } \sum_{\mathbf{s } } \mathbb{p}(\mathbf{s}[t]=\mathbf{s } ) h ( \mathbf{v}_{\mathbf{s}}^{(1)}[t ] \oplus \mathbf{z}_{\mathbf{s}}^{(1)}[t ] | w^{(1 ) } , \mathbf{y}^{(1)}_{1:t-1 } , \mathbf{s}_{1:t-1 } )
- n m ( 1-p)\log(\pi e ) \nonumber \\ & \stackrel{(b)}\leq n \sum_{j=1}^{m } ( 1-p ) \log \left ( 1+|g_{d , j}|^2 \right ) + p \log \left ( 1 + \left ( |g_{d , j}| + |g_{i , j}| \right ) ^2 \right ) + nm \log ( \pi e ) \nonumber\\ & \quad - p \sum_{t=1}^{n } h ( \tilde{\mathbf{v}}^{(1)}[t ] \oplus \mathbf{z}^{(1)}[t ] | w^{(1 ) } , \mathbf{y}^{(1)}_{1:t-1 } , \mathbf{s}_{1:t-1 } ) - n m ( 1-p)\log(\pi e ) \nonumber\end{aligned}\ ] ] @xmath149 \oplus \mathbf{z}^{(1)}[t ] | w^{(1 ) } , \mathbf{v}^{(1)}_{1:t-1}\oplus \mathbf{z}^{(1)}_{1:t-1 } , \mathbf{s}_{1:t-1 } ) - n m ( 1-p)\log(\pi e ) \label{eq : gn_causal_a}\end{aligned}\ ] ] where ( a ) follows from the proof of ( [ eq : gn_looser_r1 ] ) ( see appendix [ sec : gn_r1_ob ] ) and ( b ) follows by using the differential entropy version of inequality ( [ eq : madiman_tetali ] ) for the fractional partition defined in ( [ eq : fractional_cover_setup ] ) . using fano s inequality for @xmath17 , for any @xmath141 , there exists a large enough @xmath55 such that ; @xmath144 , \mathbf{y}^{(1)}[t ] | \mathbf{y}^{(2)}_{1:t-1 } , \mathbf{y}^{(1)}_{1:t-1 } , \mathbf{s}_{1:t-1 } , w^{(1 ) } , \mathbf{s}[t ] ) \nonumber\\ & = \sum_{t=1}^{n } \sum _ { \mathbf{s } } \mathbb{p}(\mathbf{s}[t]=\mathbf{s } ) i(w^{(2 ) } ; \mathbf{y}^{(2)}[t ] , \mathbf{y}^{(1)}[t ] | \mathbf{y}^{(2)}_{1:t-1 } , \mathbf{y}^{(1)}_{1:t-1 } , \mathbf{s}_{1:t-1 } , w^{(1 ) } , \mathbf{s}[t]=\mathbf{s } ) \nonumber\\ & = \sum_{t=1}^{n } \sum _ { \mathbf{s } } \mathbb{p}(\mathbf{s}[t]=\mathbf{s } ) i(w^{(2 ) } ; \hat{\mathbf{x}}^{(2)}[t]\oplus \mathbf{z}^{(2)}[t ] , \mathbf{v}^{(1)}_{\mathbf{s}}[t ] \oplus \mathbf{z}^{(1)}[t ] | \mathbf{y}^{(2)}_{1:t-1 } , \mathbf{y}^{(1)}_{1:t-1 } , \mathbf{s}_{1:t-1 } , w^{(1 ) } , \mathbf{s}[t]=\mathbf{s } ) \nonumber\\ & = \sum_{t=1}^{n } \sum _ { \mathbf{s } } \mathbb{p}(\mathbf{s}[t]=\mathbf{s } ) i(w^{(2 ) } ; \hat{\mathbf{x}}^{(2)}[t]\oplus \mathbf{z}^{(2)}[t ] , \mathbf{v}_{\mathbf{s}}^{(1)}[t ] \oplus \mathbf{z}^{(1)}_{\mathbf{s}}[t ] | \mathbf{y}^{(2)}_{1:t-1 } , \mathbf{y}^{(1)}_{1:t-1 } , \mathbf{s}_{1:t-1 } , w^{(1 ) } , \mathbf{s}[t]=\mathbf{s } ) \nonumber\\ & \quad + \sum_{t=1}^{n } \sum _ { \mathbf{s } } \mathbb{p}(\mathbf{s}[t]=\mathbf{s } ) i(w^{(2 ) } ; \mathbf{z}^{(1)}_{\mathbf{s}^c}[t ] | \hat{\mathbf{x}}^{(2)}[t]\oplus \mathbf{z}^{(2)}[t ] , \mathbf{v}_{\mathbf{s}}^{(1)}[t ] \oplus \mathbf{z}_{\mathbf{s}}^{(1)}[t ] , \mathbf{y}^{(2)}_{1:t-1 } , \mathbf{y}^{(1)}_{1:t-1 } , \mathbf{s}_{1:t-1 } , w^{(1 ) } , \mathbf{s}[t]=\mathbf{s } ) \nonumber\\ & = \sum_{t=1}^{n } \sum _ { \mathbf{s } } \mathbb{p}(\mathbf{s}[t]=\mathbf{s } ) i(w^{(2 ) } ; \hat{\mathbf{x}}^{(2)}[t]\oplus \mathbf{z}^{(2)}[t ] , \mathbf{v}_{\mathbf{s}}^{(1)}[t ] \oplus \mathbf{z}^{(1)}_{\mathbf{s}}[t ] | \mathbf{y}^{(2)}_{1:t-1 } , \mathbf{y}^{(1)}_{1:t-1 } , \mathbf{s}_{1:t-1 } , w^{(1 ) } , \mathbf{s}[t]=\mathbf{s } ) \nonumber\\ & \leq \sum_{t=1}^{n } \sum _ { \mathbf{s } } \mathbb{p}(\mathbf{s}[t]=\mathbf{s } ) i(w^{(2 ) } ; \hat{\mathbf{x}}^{(2)}[t]\oplus \mathbf{z}^{(2)}[t ] , \tilde{\mathbf{v}}^{(1)}[t ] \oplus \mathbf{z}^{(1)}[t ] | \mathbf{y}^{(2)}_{1:t-1 } , \mathbf{y}^{(1)}_{1:t-1 } , \mathbf{s}_{1:t-1 } , w^{(1 ) } , \mathbf{s}[t]=\mathbf{s } ) \nonumber\\ & = \sum_{t=1}^{n } \sum _ { \mathbf{s } } \mathbb{p}(\mathbf{s}[t]=\mathbf{s } ) h(\hat{\mathbf{x}}^{(2)}[t]\oplus \mathbf{z}^{(2)}[t ] , \tilde{\mathbf{v}}^{(1)}[t ] \oplus \mathbf{z}^{(1)}[t ] | \mathbf{y}^{(2)}_{1:t-1 } , \mathbf{y}^{(1)}_{1:t-1 } , \mathbf{s}_{1:t-1 } , w^{(1 ) } , \mathbf{s}[t]=\mathbf{s } ) \nonumber\\ & \quad -\sum_{t=1}^{n } \sum _ { \mathbf{s } } \mathbb{p}(\mathbf{s}[t]=\mathbf{s } ) h(\hat{\mathbf{x}}^{(2)}[t]\oplus \mathbf{z}^{(2)}[t ] , \tilde{\mathbf{v}}^{(1)}[t ] \oplus \mathbf{z}^{(1)}[t ] | \mathbf{y}^{(2)}_{1:t-1 } , \mathbf{y}^{(1)}_{1:t-1 } , \mathbf{s}_{1:t-1 } , w^{(1 ) } , w^{(2 ) } , \mathbf{s}[t]=\mathbf{s } ) \nonumber\\ & = \sum_{t=1}^{n } \sum _ { \mathbf{s } } \mathbb{p}(\mathbf{s}[t]=\mathbf{s } ) h(\hat{\mathbf{x}}^{(2)}[t]\oplus \mathbf{z}^{(2)}[t ] , \tilde{\mathbf{v}}^{(1)}[t ] \oplus \mathbf{z}^{(1)}[t ] | \mathbf{y}^{(2)}_{1:t-1 } , \mathbf{y}^{(1)}_{1:t-1 } , \mathbf{s}_{1:t-1 } , w^{(1 ) } , \mathbf{s}[t]=\mathbf{s } ) - 2 nm \log(\pi e ) \nonumber\\ & \leq \sum_{t=1}^{n } \sum _ { \mathbf{s } } \mathbb{p}(\mathbf{s}[t]=\mathbf{s } ) h(\hat{\mathbf{x}}^{(2)}[t]\oplus \mathbf{z}^{(2)}[t ] , \tilde{\mathbf{v}}^{(1)}[t ] \oplus \mathbf{z}^{(1)}[t ] | \mathbf{y}^{(1)}_{1:t-1 } , \mathbf{s}_{1:t-1 } , w^{(1 ) } ) - 2 nm \log(\pi e ) \nonumber\\ & = \sum_{t=1}^{n } h(\hat{\mathbf{x}}^{(2)}[t]\oplus \mathbf{z}^{(2)}[t ] , \tilde{\mathbf{v}}^{(1)}[t ] \oplus \mathbf{z}^{(1)}[t ] | \mathbf{y}^{(1)}_{1:t-1 } , \mathbf{s}_{1:t-1 } , w^{(1 ) } ) - 2 nm \log(\pi e ) \nonumber\\ & = \sum_{t=1}^{n } h(\hat{\mathbf{x}}^{(2)}[t]\oplus \mathbf{z}^{(2)}[t ] , \tilde{\mathbf{v}}^{(1)}[t ] \oplus \mathbf{z}^{(1)}[t ] | \mathbf{v}^{(1)}_{1:t-1 } \oplus \mathbf{z}^{(1)}_{1:t-1 } , \mathbf{s}_{1:t-1 } , w^{(1 ) } ) - 2 nm \log(\pi e ) \label{eq : gn_causal_b}\end{aligned}\ ] ] using inequalities ( [ eq : gn_causal_a ] ) and ( [ eq : gn_causal_b ] ) , @xmath150 \oplus \mathbf{z}^{(1)}[t ] | w^{(1 ) } , \mathbf{v}^{(1)}_{1:t-1}\oplus \mathbf{z}^{(1)}_{1:t-1 } , \mathbf{s}_{1:t-1 } ) - n m ( 1-p)\log(\pi e ) \nonumber \\ & \quad + p\sum_{t=1}^{n } h(\hat{\mathbf{x}}^{(2)}[t]\oplus \mathbf{z}^{(2)}[t ] , \tilde{\mathbf{v}}^{(1)}[t ] \oplus \mathbf{z}^{(1)}[t ] | \mathbf{v}^{(1)}_{1:t-1 } \oplus \mathbf{z}^{(1)}_{1:t-1 } , \mathbf{s}_{1:t-1 } , w^{(1 ) } ) - 2pnm \log(\pi e)\nonumber \end{aligned}\ ] ] @xmath151\oplus \mathbf{z}^{(2)}[t ] | \tilde{\mathbf{v}}^{(1)}[t ] \oplus \mathbf{z}^{(1)}[t ] , \mathbf{v}^{(1)}_{1:t-1 } \oplus \mathbf{z}^{(1)}_{1:t-1 } , \mathbf{s}_{1:t-1 } , w^{(1 ) } ) - pnm \log(\pi e)\nonumber \\ &
\leq n \sum_{j=1}^{m } ( 1-p ) \log \left ( 1+|g_{d , j}|^2 \right ) + p \log \left ( 1 + \left
( |g_{d , j}| + |g_{i , j}| \right ) ^2 \right ) \nonumber\\ & \quad + p\sum_{t=1}^{n } \sum_{j=1}^m h ( g_{d , j}x_j^{(2)}[t ] + z_j^{(2)}[t ] | g_{i , j}x_j^{(2)}[t ] + z_j^{(1)}[t ] ) - pnm \log(\pi e)\nonumber \\ & \leq n \sum_{j=1}^{m } ( 1-p ) \log \left ( 1+|g_{d , j}|^2 \right ) + p \log \left ( 1 + \left ( |g_{d , j}| + |g_{i , j}| \right ) ^2 \right ) \nonumber\\ & \quad + p\sum_{t=1}^{n } \sum_{j=1}^m \log \left ( \pi e \left ( 1 + \frac{|g_{d , j}|^2}{1 + |g_{i , j}|^2 } \right ) \right ) - pnm \log(\pi e ) \nonumber\\ & = n \sum_{j=1}^{m } ( 1-p ) \log \left ( 1+|g_{d , j}|^2 \right ) + p
\log \left ( 1 + \left ( |g_{d , j}| + |g_{i , j}| \right ) ^2 \right ) + p \log \left ( 1 + \frac{|g_{d , j}|^2}{1 + |g_{i , j}|^2 } \right ) \nonumber\\ & = n \left ( ( 1+p ) \sum_{j=1}^{m } \log \left ( 1+|g_{d , j}|^2 \right ) + p \delta_{g } \right ) % --------------\end{aligned}\ ] ] where @xmath87 . the bound on @xmath147 follows by symmetry , and this completes the proof of outer bound ( [ eq : causal_gn ] ) .
+ see appendix [ sec : ob_sum_gn ] .
in this section , we focus on schemes for achieving the symmetric capacity in the ld setup ( see appendix [ sec : d1d2_ach ] and [ sec : q1q2_ach ] for achievability of remaining corner points in figure [ fig : crossover ] ) . in section [ sec : single_carrier ] , we briefly review the single carrier schemes in @xcite and describe a _ bursty relaying _ technique ( used in our multicarrier schemes ) . in section [ sec : separability ] , we mention the cases where treating subcarriers separately is optimal ( _ i.e. , _ simply copying the optimal single carrier scheme @xcite on each subcarrier leads to the symmetric capacity ) . for the remaining cases ,
we propose multicarrier schemes ( covered in sections [ sec : helping_mechanism ] and [ sec : insufficient_help ] ) , which employ a helping mechanism where some _ helper _ levels in subcarriers with @xmath65 are used to recover interfered signals in subcarriers with @xmath152 . for @xmath153 ( section [ sec : helping_mechanism ] ) ,
the helping mechanism is optimal ; whereas for @xmath70 ( section [ sec : insufficient_help ] ) the helping mechanism is run in parallel with the single carrier schemes @xcite to achieve symmetric capacity . after describing our multicarrier schemes , in section [ sec : ld_inner_examples ] we provide some illustrative examples . the single carrier version of our setup ( _ i.e. , _
@xmath154 ) was studied in @xcite . for notational consistency
, we use @xmath155 ( subcarrier index ) in stating the results from @xcite .
we simply restate below the schemes in @xcite for the regimes @xmath156 and @xmath157 ; but for the regime @xmath158 we mention a slightly different scheme that makes describing our multicarrier schemes in sections [ sec : helping_mechanism ] and [ sec : insufficient_help ] more convenient .
[ [ regime - alpha_1-leq-1 ] ] regime @xmath156 + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + for this regime , the symmetric capacity is @xmath159 . to achieve this , a two phase scheme ( same for @xmath19 and @xmath24 )
is used as briefly described below has slight variation from this scheme . for details ,
see @xcite . ]
( see @xcite for details ) : * phase @xmath160 : transmitters in phase @xmath160 at time index @xmath39 send fresh symbols on all @xmath161 levels .
if there is no interference at time index @xmath39 ( occurs w.p .
@xmath162 ) , all @xmath161 symbols can be decoded at the intended receiver and both transmitters stay in phase @xmath160 for time index @xmath163 . if there is interference ( occurs w.p .
@xmath164 ) , only the bottom @xmath165 symbols get interfered at a receiver and the transmitters transition to phase @xmath166 for time index @xmath163 . * phase @xmath166 : transmitters send the past interference ( obtained from receiver feedback ) on the top @xmath165 levels and fresh symbols on the remaining @xmath167 levels . both transmitters transition to phase @xmath160 for the next time index after phase @xmath166 .
figure [ fig : markov_chain ] shows the underlying markov chain for this scheme . .
] [ [ regime-1alpha_1-leq-2 ] ] regime @xmath168 + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + for this regime , the symmetric capacity is @xmath169 . to achieve this , a two phase scheme is used as briefly described below ( see @xcite for details ) : * phase @xmath160 : transmitters in phase @xmath160 at time index @xmath39 send fresh symbols on the top @xmath161 levels and the bottom @xmath170 levels are not used . if there is no interference at time index @xmath39 ( occurs w.p .
@xmath162 ) , all @xmath161 symbols can be decoded at the intended receiver and both transmitters stay in phase @xmath160 .
if there is interference at time index @xmath39 ( occurs w.p .
@xmath164 ) , only @xmath171 symbols get interfered at a receiver and the transmitters transition to phase @xmath166 for time index @xmath163 .
* phase @xmath166 : transmitters send fresh symbols in the top @xmath170 levels . in the next @xmath171 levels ( below the top @xmath170 levels ) , the @xmath171 interfering symbols ( obtained through receiver feedback ) from the previous time index are sent .
the remaining @xmath170 levels in the bottom are not used . both transmitters transition to phase @xmath160 for the next time index after phase @xmath166 .
the underlying markov chain in this scheme is same as the one in figure [ fig : markov_chain ] .
[ [ regime - alpha_1 - 2-bursty - relaying ] ] regime @xmath172 ( bursty relaying ) + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + for this regime , the symmetric capacity is @xmath173 . in @xcite ,
this is achieved using a markov chain based scheme similar to the ones described above . for convenience in describing our multicarrier schemes ,
we derive a block version of the scheme in @xcite as follows . in each block of duration @xmath13
, transmitters send fresh symbols on the top @xmath161 levels and never use the bottom @xmath161 levels . since the bottom @xmath161 levels are never used , the fresh symbols from the top @xmath161 levels are always received interference free .
this realizes rate @xmath161 . from the @xmath174 levels in the middle ( below the top @xmath161 levels ) , we realize an additional rate @xmath175 over two blocks as follows . for the first block , @xmath26 creates @xmath176 linear combinations from @xmath177 fresh symbols and sends these linear combinations in the middle @xmath174 levels . for large enough @xmath13 , w.h.p .
@xmath178 receives @xmath177 such linear combinations .
@xmath178 decodes the constituent fresh symbols from these linear combinations and sends them to @xmath179 ( through feedback ) .
@xmath179 now creates @xmath176 new linear combinations from these symbols and sends them in the @xmath180 middle levels during the next block . w.h.p .
@xmath181 receives @xmath177 such linear combinations and decodes all the constituent symbols .
this leads to an additive rate of @xmath182 at @xmath28 ( and similarly at @xmath178 ) . in the remainder of this paper
, we refer to this technique ( for middle levels in subcarriers with @xmath183 ) as _ bursty relaying _ since @xmath26-@xmath28 pair effectively acts as a relay for @xmath179-@xmath178 and vice versa .
figure [ fig : bursty_relaying ] illustrates this technique of bursty relaying .
adding the rate from bursty relaying in @xmath180 middle levels and rate @xmath161 from the top @xmath161 levels , we achieve rate @xmath173 .
middle levels ( below the top @xmath161 levels ) when @xmath158 . as shown , @xmath17 receives @xmath177 linear combinations in @xmath177 symbols during a block of duration @xmath13 .
it decodes and sends the constituent symbols to @xmath24 which again creates @xmath176 linear combinations from these symbols . in the next block
, @xmath15 receives @xmath177 linear combinations from @xmath24 and decodes the constituent symbols . ] using outer bounds ( [ eq : causal_ld ] ) and ( [ eq : ob_3 ] ) for ld setup and achievability rates for the single carrier schemes in @xcite , the following can be easily verified . * for @xmath184 _ i.e. , _ when interference is either never present or always present , the symmetric capacity can be achieved by treating the subcarriers separately . * for @xmath185 , when all subcarriers have @xmath94 , the symmetric capacity can be achieved by treating the subcarriers separately . * for @xmath185 , when all subcarriers have @xmath95 , the symmetric capacity can be achieved by treating the subcarriers separately . hence , the subcarriers are _ separable _ in the above cases . when we have subcarriers with @xmath94 as well as subcarriers with @xmath183 ( and @xmath185 ) , we employ coding across subcarriers ( through a helping mechanism described in the next subsection ) to achieve symmetric capacity ; we assume such a _ mixed _ collection of subcarriers in describing our multicarrier schemes in sections ( [ sec : helping_mechanism ] ) and ( [ sec : insufficient_help ] ) .
when @xmath71 , @xmath186 .
we will now describe the achievability of @xmath187 using a block based scheme . in each block of duration @xmath13
, fresh symbols are sent in the following levels ( same for both transmitters by symmetry ) : * all @xmath42 levels for subcarriers with @xmath66 .
* top @xmath42 levels for subcarriers with @xmath188 . and
the following levels are not used : * bottom @xmath189 levels of subcarriers with @xmath190 . *
bottom @xmath42 levels of subcarriers with @xmath65 . because of the above choices , in every block ( for large enough @xmath13 ) : * in subcarriers with @xmath66 , w.h.p .
@xmath191 fresh symbols get interfered . * in subcarriers with @xmath192 , w.h.p . @xmath193
fresh symbols get interfered . * in subcarriers with @xmath65 , the top @xmath42 fresh symbols are always received interference free . in total , each receiver needs to recover @xmath194 interfered symbols in each block .
this recovery is done in a pipelined fashion in the next block using a _ helping mechanism _ described below .
[ [ helping - mechanism ] ] helping mechanism + + + + + + + + + + + + + + + + + we will use the term _ helper levels _ for the middle @xmath195 levels ( below the top @xmath42 levels ) in subcarriers with @xmath183 ; hence @xmath196 helper levels in total .
after each block , due to feedback from @xmath28 , @xmath26 knows exactly which of its transmitted symbols caused interference at @xmath178 .
the number of such symbols , as described above , is w.h.p .
equal to @xmath194 .
@xmath26 now creates @xmath197 linear combinations of these symbols and sends the linear combinations on any @xmath198 of the helper levels in the subsequent block . w.h.p .
@xmath194 of such linear combinations are received at @xmath178 .
this is sufficient to recover all the interfered symbols at @xmath178 in the previous block .
as all the interfered symbols in a block are recovered using the above mechanism , we realize rate @xmath199 . if @xmath64 , some of the helper levels are still available ; precisely @xmath200 of them .
we realize an additional rate of @xmath201 from such leftover helper levels using the bursty relaying scheme described in section [ sec : single_carrier ] . adding the rate from the leftover helper levels to @xmath199
, we achieve the symmetric capacity @xmath202 .
when @xmath70 , @xmath203 .
before we proceed to the details , we give a high level idea of the scheme as follows .
simply copying the scheme for @xmath71 in section [ sec : helping_mechanism ] does not work for this case since there are not enough helper levels ( @xmath204 ) compared to the number of levels facing interference ( @xmath205 ) .
the trick in this case is to _ help as much as possible_. for each subcarrier with @xmath152 , we select @xmath206 _ helped _ levels ; these levels face interference and the interfered symbols are recovered using the helping mechanism described in section [ sec : helping_mechanism ] . the total number of helped levels @xmath207 equals the number of helper levels ( @xmath208 ) . for the remaining interfered levels in subcarriers with @xmath209 ,
we run the optimal single carrier scheme @xcite ( with a slight modification ) in parallel with the helping mechanism . adding the rates from the single carrier schemes and the helping mechanism , we achieve the symmetric capacity .
this high level idea can also be illustrated by rewriting @xmath210 as shown below .
@xmath211 where @xmath212 is the total number of helped levels , and for subcarriers ( with @xmath209 ) being helped the effective direct and interfering link strengths are @xmath213 and @xmath214 .
the last two terms in ( [ eq : rate_split ] ) come from the optimal single carrier schemes for @xmath209 ( that run in parallel with the helping mechanism ) .
we now describe the achievability of @xmath100 in detail . in subcarriers with @xmath95 ,
the transmitters always send fresh symbols in the top @xmath42 levels and never use the bottom @xmath42 levels .
this realizes rate @xmath215 . for each subcarrier with @xmath152
, we assign a non - negative integral value @xmath206 with the following constraints : ( @xmath14 ) @xmath216 for @xmath217 and @xmath218 for @xmath219 , ( @xmath16 ) @xmath220 .
simply put , @xmath206 denotes the number of helped levels in a subcarrier and the total number of such levels equals the number of helper levels available in subcarriers with @xmath183 .
having fixed @xmath206 for each subcarrier with @xmath209 , we now describe the modifications needed in the optimal single carrier scheme @xcite for parallel execution with the helping mechanism .
[ [ modification - for - alpha_j-1 ] ] modification for @xmath221 + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + the bottom @xmath206 levels ( of the direct link ) are selected as helped levels as shown in figure [ fig : parallel](a ) and interfered symbols in these levels are recovered using the helping mechanism described in section [ sec : helping_mechanism ] . for the modified single carrier scheme ,
phase @xmath160 remains the same as in @xcite and the modification is only in phase @xmath166 . for illustration purposes
consider that in phase @xmath160 for a subcarrier with @xmath222 , @xmath19 sends fresh symbols @xmath223 $ ] ( as shown in figure [ fig : parallel](a ) ) and @xmath24 sends fresh symbols @xmath224 $ ] . if there is no interference , all the fresh symbols are received and the transmitters stay in phase @xmath160 . if there is interference , the transmitters transition to phase @xmath166 . in the scheme in @xcite ,
all @xmath43 interfering symbols were sent on the top @xmath43 levels in phase @xmath166 ; in the modified scheme the transmitters just send the top @xmath225 interfering symbols in the top @xmath226 levels as shown in figure [ fig : parallel](a ) . in the remaining levels ,
fresh symbols are sent ( starred symbols in figure [ fig : parallel](a ) ) . ignoring the bottom @xmath206 levels , the resulting system of linear equations at the receivers is exactly the same as in @xcite with direct link strength @xmath227 and interfering link strength @xmath228 .
thus at end of phase @xmath166 , @xmath15 is able to decode @xmath229 ( interfered symbols in phase @xmath160 ) and @xmath230 ( fresh symbols in phase @xmath166 ) . to decode interfered symbols in the helped levels ,
the helping mechanism is used ( which collects all interfered symbols in helped levels during a block of duration @xmath13 and enables their recovery in the next block ) .
so effectively , the rate obtained from a subcarrier with @xmath66 is @xmath231 .
[ [ modification - for - alpha_j-1 - 1 ] ] modification for @xmath232 + +
+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + the case @xmath233 is just an aggregated version of the simple case @xmath234 . for this simple case , either @xmath235 or @xmath236 . if @xmath236 , we use the helping mechanism to recover the interfered symbols . if @xmath235 , there are no helped levels and we simply use the scheme for @xmath237 in @xcite .
[ [ modification - for-1alpha_j-2 ] ] modification for @xmath238 + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + the top @xmath206 levels ( of the direct link at the receiver ) are selected as helped levels as shown in figure [ fig : parallel](b ) .
again , phase @xmath160 remains the same as in @xcite and the modification is only for phase @xmath166 . for illustration purposes , consider that in phase @xmath160 for a subcarrier with @xmath238 , @xmath19 sends fresh symbols @xmath239 $ ] on the top @xmath42 levels . ]
( as shown in figure [ fig : parallel](b ) ) .
similarly , @xmath24 sends fresh symbols @xmath240 $ ] on the top @xmath42 levels .
the bottom @xmath189 levels are not used .
if there is no interference , all the fresh symbols are received and the transmitters stay in phase @xmath160 . if there is interference , the transmitters transition to phase @xmath166 . in phase @xmath166 of the scheme in @xcite ,
the bottom @xmath189 levels were not used and the @xmath241 interfering symbols in phase @xmath160 were sent on the @xmath241 levels above the unused levels . in the modified scheme , the transmitters send only @xmath242 interfering symbols ( from phase @xmath160 ) on the @xmath243 levels above the @xmath189 unused levels in the bottom .
these interfering symbols correspond to the @xmath243 levels below the top @xmath206 levels in the direct link at the receiver as shown in figure [ fig : parallel](b ) . in the remaining levels ,
fresh symbols are sent ( starred symbols in figure [ fig : parallel](b ) ) . ignoring the @xmath206 helped levels , the resulting system of linear equations at the receivers is exactly the same as in @xcite with direct link strength @xmath227 and interfering link strength @xmath228 .
thus at end of phase @xmath166 , @xmath15 is able to decode @xmath244 ( interfered symbols in phase @xmath160 ) and @xmath245 ( fresh symbols in phase @xmath166 ) . to decode interfered symbols in the helped levels ,
the helping mechanism is used ( which collects all interfered symbols in helped levels during a block of duration @xmath13 and enables their recovery in the next block ) .
so effectively , the rate obtained from a subcarrier with @xmath246 is @xmath247 .
taking into account the above modifications and adding the rates across subcarriers we achieve rate @xmath248 .
the toy example in section [ sec : introduction ] considered two subcarriers with @xmath249 , @xmath2 , @xmath250 and @xmath251 ( and @xmath8 ) .
as illustrated in the toy example , the middle level in the second subcarrier helped in recovering interfered symbols in the first subcarrier .
with reference to our achievability scheme for @xmath71 , the middle level in the second subcarrier is a helper level ( green level in figure [ fig : toy_example_revisited](a ) ) whereas the ( only ) level in the first subcarrier is a helped level ( red level in figure [ fig : toy_example_revisited](a ) ) .
since there is only one helped level and one helper level , @xmath252 and @xmath253 . to illustrate ideas behind our achievability schemes for @xmath68 and @xmath99
, we slightly modify the toy example as described below . and ( c ) example [ ex : delta_pos ] . ,
title="fig : " ] and ( c ) example [ ex : delta_pos ] . ,
title="fig : " ] and ( c ) example [ ex : delta_pos ] .
, title="fig : " ] [ ex : delta_neg ] compared to the original toy example , we have modified only the first subcarrier . for this case
, there are two levels in the first subcarrier which face may interference but there is only one helper level ( green level in figure [ fig : toy_example_revisited](b ) ) available in the second subcarrier . hence @xmath254 and @xmath255 .
we help the bottom level in the first subcarrier ( as we did in the original toy example ) and by simply copying the scheme in the original toy example we achieve rate @xmath0 . for the top level in the first subcarrier ( gray level in figure [ fig : toy_example_revisited](b ) ) ,
we use the optimal single carrier scheme for @xmath256 @xcite and achieve additional rate @xmath257 . in this example , it is easy to see that the helping mechanism and the single carrier scheme can be executed in parallel .
[ ex : delta_pos ] compared to the original toy example , we have modified only the second subcarrier such that it has one extra middle level ( blue level in figure [ fig : toy_example_revisited](c ) ) . for this case , @xmath258 and @xmath259 . the helping mechanism is used as in the original toy example to achieve rate @xmath0 .
additional rate @xmath260 is achieved using the bursty relaying technique for the extra middle level in the second subcarrier ( blue level in figure [ fig : toy_example_revisited](c ) ) .
in this section , we first describe tight outer bounds ( described below ) followed by tight inner bounds ( in sections [ sec : gdof_inner_pos ] and [ sec : gdof_inner_neg ] ) on the gdof for gn setup . as mentioned in section [ sec : main_results ] , for the gdof analysis we assume @xmath88 , @xmath89 and @xmath90 . we assume a rational @xmath91 to simplify the achievability schemes ( described in sections [ sec : gdof_inner_pos ] and [ sec : gdof_inner_neg ] ) . with the above assumptions ,
the gdof for gn setup is defined as follows , @xmath261 where @xmath262 is the symmetric capacity . from outer bounds ( [ eq : causal_gn ] ) and ( [ eq : ob_sum_gn ] ) for the gn setup , we have bounds on @xmath69 as follows , @xmath263 where @xmath264 . using ( [ eq : c_sym_bound ] ) , the following outer bound on gdof holds , @xmath265 where @xmath266 .
+ in the remainder of this section , we describe achievability schemes ( inner bounds ) which achieve outer bound ( [ eq : gdof_ob ] ) .
the schemes for the gdof setting mimic the achievability schemes for symmetric capacity in the ld setup by using techniques from @xcite .
hence , the scheme for @xmath267 ( section [ sec : gdof_inner_pos ] ) in the gdof setting mimics the scheme for @xmath71 in ld setup and the scheme for @xmath268 ( section [ sec : gdof_inner_neg ] ) mimics the scheme for @xmath70 in ld setup .
we use a block based scheme ( block size @xmath13 ) which mimics the scheme for @xmath71 in section [ sec : helping_mechanism ] for ld setup . for convenience in describing our scheme , we will work with the following _ real _ channel ( the achievable rate for the complex channel in gn setup is just twice the achievable rate for this channel ) .
@xmath48 & = \sqrt{snr}\ ; x^{(i)}_j[t ] + \left ( s_j[t ] \right ) \sqrt{inr_j}\ ; x^{(i')}_j[t ] + z^{(i)}_j[t ] \label{eq : gdof_real_channel}\end{aligned}\ ] ] where @xmath269,\ ; x^{(i')}_j[t ] \in \mathbb{r}$ ] , @xmath47|^2 \leq 1 $ ] and @xmath270 \sim \mathcal{n}(0,1)$ ] .
similar to the analysis in @xcite , we consider @xmath271 where @xmath272 and @xmath273 are positive integers .
furthermore , @xmath273 is such that @xmath274 is an integer ( always possible since all @xmath91 are rational ) . by letting @xmath273 grow to infinity
, we get a sequence of snrs that approach infinity . using ( [ eq : gdof_snr ] ) , the received signal in ( [ eq : gdof_real_channel ] ) can be rewritten as follows .
@xmath48 & = q^m x^{(i)}_j[t ] + \left ( s_j[t ] \right ) q^{m\beta_j } x^{(i')}_j[t ] + z^{(i)}_j[t ] \label{eq : gdof_q_channel}\end{aligned}\ ] ] following @xcite , we will express positive real signals in @xmath272-ary representation using @xmath272-ary digits @xmath275 ( which we will refer to as `` qits '' , similar to @xcite ) . to mimic the achievability scheme for @xmath71 in ld setup ( section [ sec : helping_mechanism ] ) , we use the following structure for the input signals ( we drop the time index for convenience ) . * for @xmath276 , @xmath277_q\end{aligned}\ ] ] where @xmath278 and for the remaining @xmath279 , @xmath280 .
[ eq : struct_2 ] * for @xmath281 , @xmath282_q \label{eq : struct_1}\end{aligned}\ ] ] where @xmath283 for @xmath284 . * for @xmath285 , @xmath286_q \label{eq : struct_1_2}\end{aligned}\ ] ] where @xmath287 and for the remaining @xmath288 , + @xmath289 .
the structure ( _ i.e. , _ non - zero qits ) used is same as in the scheme for ld setup ( section [ sec : helping_mechanism ] ) .
the restrictions on the values taken by non - zero qits arises from techniques in @xcite ( these simplify the analysis by preventing carry overs when signals interfere , see @xcite for details ) . in the absence of noise , it is easy to see the similarities between the ld setup and above setup ; qits in a signals are similar to _ levels _ in the ld setup .
the following example makes this similarity more precise for the case of subcarriers with @xmath290 . in a subcarrier with @xmath290 ,
the received signal at @xmath28 after interference ( in the absence of noise ) is as follows .
@xmath291_q + \left [ x^{(i')}_{j , m\beta_j } \ ; x^{(i')}_{j , m\beta_j-1 } \ ; \ldots \ ; x^{(i')}_{j , m+1 } \ ; 0\;0\;\ldots 0\;.\;0 \;0 \right]_q \end{aligned}\ ] ] clearly , the top @xmath273 qits of the direct signal ( _ i.e. , _
@xmath292 ) are interference free in the above scenario and by doing a modulo @xmath293 operation at the receiver , one can completely recover the direct signal . even in the presence of noise , due to bounded variance of the noise
, the higher qits can be decoded with negligible probability of error ( as @xmath294 ) .
having shown the similarity between ld setup and the above setup in the absence of noise , we now describe the rates that we can achieve from the subcarriers in the gdof setting .
[ [ beta_j - leq-1 ] ] @xmath295 + + + + + + + + + + + + + + + + + + + + + + + + + + + + in this case , over a block only @xmath296 qits in the direct signal are interfered .
assuming we are able to recover all ( except @xmath297 ) interfering qits ( using the helping mechanism described for @xmath290 below ) , we can achieve the following rate : @xmath298 the above rate follows directly from the analysis in @xcite .
[ [ beta_j - leq-2 ] ] @xmath299 + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + in this case , over a block only @xmath300 qits in the direct signal are interfered .
assuming we are able to recover all ( except @xmath297 ) interfering qits ( using the helping mechanism described for @xmath290 below ) , we can achieve the following rate : @xmath298 [ [ beta_j-2 ] ] @xmath301 + + + + + + + + + + + + + + + + + + + + + + + + + the top @xmath273 qits in the subcarriers with @xmath290 are always received interference free . so from them we can achieve rate : @xmath302 we now describe the helping mechanism for the gdof setting . for removing the interfering qits for subcarriers with @xmath303 in the previous block , we need to use @xmath304 _ helper _ qits in subcarriers with @xmath305 ; these are the middle @xmath306 qits below the top @xmath273 qits . since @xmath267
, we have sufficient number of such helper qits to recover all interfering qits in subcarriers with @xmath303 .
the helping mechanism is same as described for the ld setup ( with minor changes for the @xmath272-ary setup ) . from the leftover helper qits
, we can achieve an additional rate using the bursty relaying technique .
summing the rates for all subcarriers we have the following inner bound ( a factor of @xmath307 is included to account for the complex channel ) .
@xmath308 so , @xmath309 where ( a ) follows from large enough @xmath272 .
since the inner bound on gdof matches the outer bound , we have a tight result when @xmath310 . as in the case of @xmath267 in section [ sec : gdof_inner_pos ] , we focus on the real channel in ( [ eq : gdof_q_channel ] ) for our achievability scheme . the scheme for this case
mimics the achievability of symmetric capacity in ld setup for @xmath70 by using the techniques from @xcite .
since we have already illustrated the usage of techniques from @xcite ( for the case @xmath267 ) in mimicking the ld setup schemes for the gdof setting , we will briefly sketch the inner bound for @xmath268 . following the strategy of _ helping as much possible _ for the case @xmath70 in ld setup
, we use the middle @xmath306 qits ( below the top @xmath273 qits ) in subcarriers with @xmath290 as helper qits .
all the helper qits are used to recover interference in helped qits in subcarriers with @xmath311 ( each subcarrier with @xmath311 has @xmath206 helped qits and @xmath312 ) .
so we get the following rates from subcarriers : * for @xmath313 @xmath314 @xmath315 * for @xmath316 @xmath314 @xmath317 * for @xmath318 @xmath314 @xmath319 it should be noted that due to noise , some of the interfering qits in phase @xmath160 ( of the single carrier scheme executed in parallel with the helping mechanism ) may not be decoded correctly at @xmath26 ( after feedback ) and this may affect the recovery of qits in phase @xmath166 .
however , it can be shown that such an _ error propagation _
leads to @xmath297 reduction ( compared to the case without noise ) in the achievable rate for a subcarrier . combining the rates from all subcarriers
, we have the following bound ( factor of @xmath0 included for the complex channel ) .
@xmath320 where ( a ) follows from @xmath312 .
now , we have the following bound on the gdof ; @xmath321 where ( a ) follows from large enough @xmath272 .
the above inner bound matches outer bound ( [ eq : gdof_ob ] ) when @xmath268 and this completes the gdof characterization .
the work was supported in part by nsf awards 1136174 and 1314937 . additionally , we gratefully acknowledge support by intel and verizon . 99 n. khude , v. prabhakaran and p. viswanath , `` opportunistic interference management , '' in proc .
inf . theory ( isit ) , 2009 .
n. khude , v. prabhakaran and p. viswanath , `` harnessing bursty interference , '' in proc .
information theory workshop ( itw ) , 2009 .
s. avestimehr , s. diggavi and d. tse , `` wireless network information flow : a deterministic approach , '' ieee trans .
theory , vol .
1872 - 1905 , apr . 2011 .
m. madiman and p. tetali , `` information inequalities for joint distributions , with interpretations and applications , '' ieee trans .
theory , vol .
2699 - 2713 , june 2010 . j. jiang , n. marukala and t. liu , `` symmetrical multilevel diversity coding with an all - access encoder , '' in proc .
inf . theory ( isit ) , pp .
1662 - 1666 , 2012 .
wang , c. suh , s. diggavi and p. viswanath , `` bursty interference channel with feedback , '' in proc .
theory ( isit ) 2013 .
extended version available : https://sites.google.com/site/ihsiangw/isit13preprintburstyic s. mishra , i .- h . wang and s. diggavi , `` opportunistic interference management for multicarrier systems , '' in proc .
inf . theory ( isit ) 2013 .
extended version available : http://arxiv.org/abs/1305.2985 s. a. jafar and s. vishwanath , `` generalized degrees of freedom of the symmetric gaussian k user interference channel , '' ieee trans .
theory , vol .
56 , no . 7 , pp . 3297 - 3303 , jul .
2010 . using fano s inequality for @xmath15 , for any @xmath322
, there exists a large enough @xmath55 such that ; @xmath323| \mathbf{s}[t ] ) \nonumber\\ & = \sum_{t=1}^{n } \sum_{\mathbf{s } } \mathbb{p}(\mathbf{s}[t]=\mathbf{s } ) h(\mathbf{y}^{(1)}[t]| \mathbf{s}[t]= \mathbf{s } ) \nonumber\\ & \leq \sum_{t=1}^{n } \sum_{\mathbf{s } } \mathbb{p}(\mathbf{s}[t]=\mathbf{s } ) \sum_{j=1}^{m } n_j \mathbb{i}_{j \not \in \mathbf{s } } + \max(n_j , k_j ) \mathbb{i}_{j \in \mathbf{s } } \nonumber\\ & = n\sum_{j=1}^{m } n_j + p ( \max(n_j , k_j)-n_j ) \nonumber\\ & = n p\delta + n\sum_{j=1}^m n_j(1+p ) - ( n_j - k_j)^{+}p\end{aligned}\ ] ] where @xmath324 .
the outer bound on @xmath325 follows by symmetry and this completes the proof of outer bound ( [ eq : ob_4 ] ) . using fano s inequality for @xmath15 and @xmath17 , for any @xmath322
, there exists a large enough @xmath55 such that ; @xmath326 | \mathbf{s}[t ] ) + \sum_{t=1}^{n } h ( \hat{\mathbf{x}}^{(2)}[t ] |\mathbf{v}^{(1)}_{\mathbf{s}[t]}[t ] , \mathbf{s}[t ] ) \nonumber \\ & = \sum_{t=1}^{n } \sum_{\mathbf{s } } \mathbb{p}(\mathbf{s}[t]=\mathbf{s } ) h(\mathbf{y}^{(1)}[t ] | \mathbf{s}[t]=\mathbf{s } ) + \sum_{t=1}^{n } \sum_{\mathbf{s } } \mathbb{p}(\mathbf{s}[t]=\mathbf{s } ) h ( \hat{\mathbf{x}}^{(2)}[t ] | \mathbf{v}^{(1)}_{\mathbf{s}[t]}[t ] , \mathbf{s}[t]=\mathbf{s } ) \nonumber\\ & \leq \sum_{t=1}^{n } \sum_{\mathbf{s } } \mathbb{p}(\mathbf{s}[t]=\mathbf{s } ) \sum_{j=1}^{m } n_j \mathbb{i}_{j \not \in \mathbf{s } } + \max(n_j , k_j ) \mathbb{i}_{j \in \mathbf{s } } + \sum_{t=1}^{n } \sum_{\mathbf{s } } \mathbb{p}(\mathbf{s}[t]=\mathbf{s } ) \sum_{j=1}^{m } n_j \mathbb{i}_{j \not \in \mathbf{s } } + ( n_j - k_j)^{+ } \mathbb{i}_{j \in \mathbf{s } } \nonumber\\ & = \sum_{t=1}^{n } \sum_{j=1}^{m } n_j ( 1-p ) + \max(n_j , k_j ) p + \sum_{t=1}^{n } \sum_{j=1}^{m } n_j ( 1-p ) + ( n_j - k_j)^{+ } p \nonumber \\ & = np\delta + 2n \sum_{j=1}^{m } n_j \end{aligned}\ ] ] where @xmath324 .
this completes the proof of outer bound ( [ eq : ob_3 ] ) . using fano s inequality for @xmath15 , for any @xmath322
, there exists a large enough @xmath55 such that ; @xmath327 , \mathbf{y}^{(2 ) } [ t ] | \mathbf{y}^{(2)}_{1:t-1 } , \mathbf{y}^{(1)}_{1:t-1 } , w^{(1 ) } , w^{(2 ) } , \mathbf{s}_{1:n }
\right ) \nonumber\\ & = h \left ( \mathbf{y}^{(1)}_{1:n } , \mathbf{y}^{(2)}_{1:n } | w^{(2 ) } , \mathbf{s}_{1:n } \right ) - \sum_{t=1}^{n } h \left ( \mathbf{z}^{(1)}[t ] , \mathbf{z}^{(2 ) } [ t ] | \mathbf{y}^{(2)}_{1:t-1 } , \mathbf{y}^{(1)}_{1:t-1 } , w^{(1 ) } , w^{(2 ) } , \mathbf{s}_{1:n }
\right ) \nonumber\\ & = h \left ( \mathbf{y}^{(1)}_{1:n } , \mathbf{y}^{(2)}_{1:n } | w^{(2 ) } , \mathbf{s}_{1:n } \right ) - \sum_{t=1}^{n } h \left ( \mathbf{z}^{(1)}[t ] , \mathbf{z}^{(2 ) } [ t ] \right ) \nonumber\\ & = h \left ( \mathbf{y}^{(1)}_{1:n } , \mathbf{y}^{(2)}_{1:n } | w^{(2 ) } , \mathbf{s}_{1:n } \right ) - 2
n m \log \left ( \pi e \right ) \nonumber\\ & = \sum_{t=1}^{n } h \left ( \mathbf{y}^{(1)}[t ] , \mathbf{y}^{(2)}[t ] | \mathbf{y}^{(1)}_{1:t-1 } , \mathbf{y}^{(2)}_{1:t-1 } , w^{(2 ) } , \mathbf{s}_{1:n } \right ) - 2
n m \log \left ( \pi e \right ) \nonumber\\ & \leq \sum_{t=1}^{n } h \left ( \mathbf{y}^{(1)}[t ] , \mathbf{y}^{(2)}[t ] | \mathbf{y}^{(1)}_{1:t-1 } , \mathbf{y}^{(2)}_{1:t-1 } , w^{(2 ) } , \mathbf{s}[t ] \right ) - 2 n m \log \left ( \pi e \right ) \nonumber\\ & = \sum_{t=1}^{n } \sum_{\mathbf{s } } \mathbb{p}(\mathbf{s}[t]=\mathbf{s } ) h \left ( \mathbf{y}^{(1)}[t ] , \mathbf{y}^{(2)}[t ] | \mathbf{y}^{(1)}_{1:t-1 } , \mathbf{y}^{(2)}_{1:t-1 } , w^{(2 ) } , \mathbf{s}[t]=\mathbf{s } \right ) - 2 n m \log \left ( \pi e \right ) \nonumber\\ & \leq \sum_{t=1}^{n } \sum_{\mathbf{s } } \mathbb{p}(\mathbf{s}[t]=\mathbf{s } ) \sum_{j=1}^m \mathbb{i}_{j \not \in \mathbf{s } } \left ( \log \left ( \pi e \left ( 1 + |g_{d , j}|^2 \right ) \right ) + \log \left ( \pi e \right ) \right ) + \mathbb{i}_{j \in \mathbf{s } } \log \left ( ( \pi e)^2 \left ( 1 + |g_{d , j}|^2 + |g_{i , j}|^2 \right ) \right ) - 2 n m \log \left ( \pi e \right ) \nonumber\\ & = n \sum_{j=1}^m ( 1-p ) \log \left ( 1 + |g_{d , j}|^2
\right ) + p \log \left ( 1 + |g_{d , j}|^2 + |g_{i , j}|^2 \right ) \end{aligned}\ ] ] the bound on @xmath325 follows by symmetry and this completes the proof of outer bound ( [ eq : ob_r^i_gn ] ) .
we also prove a looser bound on @xmath56 as shown below ( the proof for this looser bound is used in the proof of outer bounds ( [ eq : causal_gn ] ) and ( [ eq : ob_sum_gn ] ) ) .
using fano s inequality for @xmath15 , for any @xmath322 , there exists a large enough @xmath55 such that ; @xmath328|\mathbf{y}^{(1)}_{1:t-1 } , w^{(2 ) } , w^{(1 ) } , \mathbf{s}_{1:n } ) \nonumber\\ & \leq h(\mathbf{y}^{(1)}_{1:n}| \mathbf{s}_{1:n } ) - \sum_{t=1}^{n } h(\mathbf{y}^{(1)}[t]|\mathbf{y}^{(2)}_{1:t-1 } , \mathbf{y}^{(1)}_{1:t-1 } , w^{(2 ) } , w^{(1 ) } , \mathbf{s}_{1:n } ) \nonumber\\ & = h(\mathbf{y}^{(1)}_{1:n}| \mathbf{s}_{1:n } ) - \sum_{t=1}^{n } h(\mathbf{z}^{(1)}[t]|\mathbf{y}^{(2)}_{1:t-1 } , \mathbf{y}^{(1)}_{1:t-1 } , w^{(2 ) } , w^{(1 ) } , \mathbf{s}_{1:n } ) \nonumber\\ & = h(\mathbf{y}^{(1)}_{1:n}| \mathbf{s}_{1:n } ) - \sum_{t=1}^{n } h(\mathbf{z}^{(1)}[t ] ) \nonumber\\ & = h(\mathbf{y}^{(1)}_{1:n}| \mathbf{s}_{1:n } ) -
n m \log(\pi e ) \nonumber\\ & \leq \sum_{t=1}^{n } h(\mathbf{y}^{(1)}[t]| \mathbf{s}[t ] ) - n m \log(\pi e ) \nonumber\end{aligned}\ ] ] @xmath329=\mathbf{s } ) h(\mathbf{y}^{(1)}[t]| \mathbf{s}[t]= \mathbf{s } ) - n m \log(\pi e ) \nonumber\\ & \leq \left ( \sum_{t=1}^{n } \sum_{\mathbf{s } } \mathbb{p}(\mathbf{s}[t]=\mathbf{s } ) \sum_{j=1}^{m } \log(\pi e ( 1+|g_{d , j}|^2 ) ) \mathbb{i}_{j \not \in \mathbf{s } } + \log \left ( \pi e \left ( 1 + \left ( |g_{d , j}| + |g_{i , j}| \right ) ^2 \right ) \right ) \mathbb{i}_{j \in \mathbf{s } } \right ) - n m \log(\pi
e ) \nonumber\\ & = \left ( \sum_{t=1}^{n } \sum_{j=1}^{m } ( 1-p ) \log \left ( \pi e \left ( 1+|g_{d , j}|^2 \right ) \right ) + p \log \left ( \pi e \left ( 1 + \left ( |g_{d , j}| + |g_{i , j}| \right ) ^2 \right )
\right ) \right ) - n m \log(\pi e ) \nonumber\\ & = n \sum_{j=1}^{m } ( 1-p ) \log \left ( 1+|g_{d , j}|^2 \right ) + p \log \left ( 1 + \left ( |g_{d , j}| + |g_{i ,
j}| \right ) ^2 \right ) \label{eq : gn_looser_r1}\end{aligned}\ ] ] as mentioned above , this is a looser bound compared to ( [ eq : ob_r^i_gn ] ) , but the above proof is used in proving outer bounds ( [ eq : causal_gn ] ) and ( [ eq : ob_sum_gn ] ) .
using fano s inequality for @xmath15 and @xmath17 , for any @xmath322 , there exists a large enough @xmath55 such that ; @xmath330 , \mathbf{y}^{(2)}[t ] | \mathbf{y}^{(1)}_{1:t-1 } , \mathbf{y}^{(2)}_{1:t-1 } , \mathbf{s}_{1:n } , w^{(1 ) } , w^{(2 ) } ) \nonumber\\ & = h(\mathbf{y}^{(1)}_{1:n } | \mathbf{s}_{1:n } ) + h ( \mathbf{y}^{(2)}_{1:n } | \mathbf{y}^{(1)}_{1:n } , \mathbf{s}_{1:n } , w^{(1 ) } ) - \sum_{t=1}^{n } h ( \mathbf{z}^{(1)}[t ] , \mathbf{z}^{(2 ) } [ t ] | \mathbf{y}^{(1)}_{1:t-1 } , \mathbf{y}^{(2)}_{1:t-1},\mathbf{s}_{1:n } , w^{(1 ) } , w^{(2 ) } ) \nonumber\\ & = h(\mathbf{y}^{(1)}_{1:n } | \mathbf{s}_{1:n } ) + h ( \mathbf{y}^{(2)}_{1:n } | \mathbf{y}^{(1)}_{1:n } , \mathbf{s}_{1:n},w^{(1 ) } ) - 2 n m \log(\pi e ) \nonumber\\ & = h(\mathbf{y}^{(1)}_{1:n } | \mathbf{s}_{1:n } ) + h ( \hat{\mathbf{x}}^{(2)}_{1:n } \oplus \mathbf{z}^{(2)}_{1:n } | \mathbf{y}^{(1)}_{1:n } , \mathbf{s}_{1:n } , w^{(1 ) } ) - 2 n m \log(\pi e ) \nonumber\\ & = h(\mathbf{y}^{(1)}_{1:n } | \mathbf{s}_{1:n } ) + h ( \hat{\mathbf{x}}^{(2)}_{1:n } \oplus \mathbf{z}^{(2)}_{1:n } | \mathbf{v}^{(1)}_{1:n}\oplus \mathbf{z}^{(1)}_{1:n } , \mathbf{s}_{1:n } , w^{(1 ) } ) - 2 n m \log(\pi e ) \nonumber\\ & \stackrel{(a ) } \leq n \sum_{j=1}^{m } ( 1-p ) \log \left ( 1+|g_{d , j}|^2 \right ) + p \log \left ( 1 + \left ( |g_{d , j}| + |g_{i , j}| \right ) ^2 \right ) \nonumber\\ & \quad + h ( \hat{\mathbf{x}}^{(2)}_{1:n } \oplus \mathbf{z}^{(2)}_{1:n } | \mathbf{v}^{(1)}_{1:n}\oplus \mathbf{z}^{(1)}_{1:n } , \mathbf{s}_{1:n } , w^{(1 ) } ) - n m \log(\pi e ) \nonumber\\ & \leq n \sum_{j=1}^{m } ( 1-p ) \log \left ( 1+|g_{d , j}|^2 \right ) + p \log \left ( 1 + \left ( |g_{d , j}| + |g_{i , j}| \right ) ^2 \right ) \nonumber\\ & \quad + \sum_{t=1}^{n } \sum_{\mathbf{s } } \mathbb{p}(\mathbf{s}[t]=\mathbf{s } ) h ( \hat{\mathbf{x}}^{(2)}[t ] \oplus \mathbf{z}^{(2)}[t ] | \mathbf{v}_{\mathbf{s}}^{(1 ) } [ t]\oplus \mathbf{z}^{(1 ) } [ t ] , w^{(1 ) } , \mathbf{s}[t]=\mathbf{s } ) - n m \log(\pi e ) \nonumber\\ & \leq n \sum_{j=1}^{m } ( 1-p ) \log \left ( 1+|g_{d , j}|^2 \right ) + p \log \left ( 1 + \left ( |g_{d , j}| + |g_{i , j}| \right ) ^2 \right ) \nonumber\\ & \quad + \sum_{t=1}^{n } \sum_{\mathbf{s } } \mathbb{p}(\mathbf{s}[t]=\mathbf{s } ) \sum_{j=1}^{m } \log(\pi e ( 1+|g_{d , j}|^2 ) ) \mathbb{i}_{j \not \in \mathbf{s } } + \log \left ( \pi e \left ( 1 + \frac{|g_{d , j}|^2}{1 + |g_{i , j}|^2 } \right ) \right ) \mathbb{i}_{j \in \mathbf{s } } - n m \log(\pi e ) \nonumber\\ & = n \sum_{j=1}^{m } ( 1-p ) \log \left ( 1+|g_{d , j}|^2 \right ) + p \log \left ( 1 + \left ( |g_{d , j}| + |g_{i , j}| \right ) ^2 \right ) \nonumber\\ & \quad + n \sum_{j=1}^{m } ( 1-p)\log \left ( 1+|g_{d , j}|^2 \right ) + p \log \left ( 1 + \frac{|g_{d , j}|^2}{1 + |g_{i , j}|^2 } \right ) \nonumber\\ % & = n \sum_{j=1}^{m } 2(1-p ) \log \left ( 1+|g_{d , j}|^2 \right ) + p \log \left ( 1 + \left ( |g_{d , j}| + |g_{i , j}| \right ) ^2 \right ) + p \log \left ( 1 + \frac{|g_{d , j}|^2}{1 + |g_{i , j}|^2 } \right ) \nonumber\\ & = n \left ( 2 \sum_{j=1}^{m } \log \left ( 1+|g_{d , j}|^2 \right ) + p \delta_g
\right ) \end{aligned}\ ] ] where ( a ) follows from the proof of ( [ eq : gn_looser_r1 ] ) ( see appendix [ sec : gn_r1_ob ] ) and @xmath87 .
as shown in figure [ fig : crossover ] , these corner points appear when @xmath64 .
we will describe the achievability of @xmath331 and achievability of @xmath332 follows by symmetry .
the achievability of @xmath331 is similar to achieving @xmath333 ( described in section [ sec : helping_mechanism ] ) ; with a slight modification for subcarriers with @xmath183 .
the additive term @xmath334 appears in @xmath187 because of bursty relaying in the leftover helper levels ( @xmath63 in number ) . for @xmath331 , to achieve @xmath335 , we use an asymmetric version of bursty relaying as follows : in every block @xmath19 sends @xmath336 linear combinations of @xmath337 fresh symbols in the leftover helper levels .
@xmath17 receives @xmath337 such linear combinations in every block ; it recovers the constituent symbols and forwards them to @xmath24 . in the next block
, @xmath24 creates @xmath336 linear combinations of the constituent symbols sent by @xmath15 and sends them on its leftover helper levels .
@xmath15 receives @xmath337 of these linear combinations and thus recovers the constituent symbols .
so compared to @xmath187 , @xmath15 now gains an additional rate @xmath201 but @xmath17 loses not using its leftover helper levels for its own messages ; it just uses them to relay messages for @xmath15 . ] rate @xmath201 .
this completes the achievability of @xmath331 .
both @xmath338 and @xmath339 are achieved using a separation based scheme ( _ i.e. , _ no coding across subcarriers ) .
we first describe the achievability of @xmath338 ; achievability of @xmath339 follows by symmetry . in @xmath340
we can rewrite rate @xmath341 as follows .
@xmath342 also , from the single carrier schemes in @xcite , the following rate tuples @xmath74 are achievable for a single carrier setup : | we study parallel symmetric @xmath0-user interference channels when the interference is bursty and feedback is available from the respective receivers .
presence of interference in each subcarrier is modeled as a memoryless bernoulli random state .
the states across subcarriers are drawn from an arbitrary joint distribution with the same marginal probability for each subcarrier and instantiated i.i.d . over time .
for the linear deterministic setup , we give a complete characterization of the capacity region . for the setup with gaussian noise , we give outer bounds and a tight generalized degrees of freedom characterization .
we propose a novel helping mechanism which enables subcarriers in very strong interference regime to help in recovering interfered signals for subcarriers in strong and weak interference regimes .
depending on the interference and burstiness regime , the inner bounds either employ the proposed helping mechanism to code across subcarriers or treat the subcarriers separately .
the outer bounds demonstrate a connection to a subset entropy inequality by madiman and tetali @xcite . |
recently fully microscopic calculations of nuclei with @xmath4 have become feasible @xcite .
the @xmath0be nucleus is such a system of special interest , as it allows tests of theories of interaction of composite particles @xcite .
the properties of low energy continuum of @xmath0be are of particular importance in this connection . however , the corresponding experimental data on the low energy photodisintegration of @xmath0be are not in mutual agreement ( see fig . 1 ) . in the present work we develop a semi microscopic model to describe the process , and we analyze the experimental data with its help .
the model accounts simultaneously for both resonant and non resonant contributions to the cross section .
an estimation of the reliability of various data is obtained and a theoretical photodisintegration cross section is derived .
we also calculate the reaction rates of the reaction @xmath0be@xmath5 and the reverse reaction for astrophysical conditions .
these reaction rates are of relevance in the high entropy bubble in type ii supernovae , an astrophysical site that has been suggested for the r process @xcite . the baryonic matter in this bubble is dominated in the beginning by @xmath6particles , neutrons , and protons .
the abundance distribution shifts then to higher masses through the recombination of the free @xmath6particles , neutrons , and protons .
this generates the so called @xmath6process leading to the formation of massive isotopes ( @xmath7 ) .
the reaction path in the @xmath6process is mainly determined by requirements of nuclear statistical equilibrium and depends also on the reaction rates of the various recombination paths bridging the mass 5 and 8 gaps .
it has been shown that there are three principal reaction paths from @xmath8he to @xmath9c : + ( i ) @xmath8he(2@xmath6,@xmath10)@xmath9c + ( ii ) @xmath8he(@xmath6 n,@xmath10)@xmath0be(@xmath6,n)@xmath9c + ( iii ) @xmath8he(2n,@xmath10)@xmath11he(@xmath6,n)@xmath0be(@xmath6,n)@xmath9c .
it was shown in refs .
@xcite that the triple alpha process ( i ) can be neglected compared to the reaction sequence ( ii ) via @xmath0be under r process conditions in the @xmath6process . also the reaction path ( iii ) via @xmath11he
can be neglected for this scenario @xcite .
this is true even if the reaction rate of @xmath8he(2n,@xmath10)@xmath11he is strongly enhanced @xcite , because then @xmath11he is also destroyed very effectively through photodissociation .
therefore , for the @xmath6 and r process the reaction @xmath8he(@xmath6 n,@xmath10)@xmath0be plays a key role in bridging the unstable mass gaps at @xmath12 and @xmath13 .
the reaction rates of @xmath8he(@xmath6 n,@xmath10)@xmath0be and the reverse photodisintegration of @xmath0be were estimated in ref .
@xcite from the experimental photodisintegration cross section
. however , ref .
@xcite did not include information on which experimental data their estimate was based .
in view of the astrophysical relevance of these reactions we recalculate in the present work the rates of the first step of the reaction ( ii ) above . the same problem is also addressed in ref .
these authors obtain the resonant contribution to the @xmath0be@xmath14be cross section from the breit wigner formula for the first excited state of @xmath0be with the parameters taken from ref .
@xcite . and
@xmath15 parameters of the resonance used in ref .
@xcite seem to be incorrect .
the resonant properties of the @xmath16 state of @xmath0be will be considered in our future work . ] in order to calculate the non resonant contribution they introduce a single particle potential with the depth chosen to reproduce the ground state , calculate both ground and final state continuum wave functions in this potential , and multiply the cross section obtained by the shell model spectroscopic factor .
they then add this cross section constructively or destructively to the resonant cross section to establish possible upper and lower bounds for the reaction rates .
this procedure has certain shortcomings : a resonant contribution to the cross section should not emerge as an addition to the dynamic model used , since a correct quantum mechanical model should necessarily contain such a contribution itself , along with the non resonant contribution and an interference term . besides
, the potential wells used for the ground state and continuum state should in fact be different : an additional spin orbit potential , for example , should be present in the ground p state as compared to the continuum s state .
in our model we use a three - body specification of the @xmath0be bound state , and a semimicroscopic continuum wave function which describes the essential scattering degrees of freedom at low relative energies . in sect . 2 this model
is formulated . in sect .
3 the results for the @xmath0be@xmath17 cross section are given . in sect .
4 the astrophysical rates for @xmath3be@xmath2 and the reverse reaction are calculated .
the relevant experimental data on the low energy @xmath0be@xmath17 cross section are presented in fig . 1 .
the available data are those in @xcite obtained with bremsstrahlung photons and those in @xcite obtained from @xmath10radiation from radioactive isotopes .
the peak at very low energy exhibited by the data of ref .
@xcite is not confirmed by other groups , and may arise from discrepancies caused by neutron energy loss in the targets @xcite .
the radioactive isotope techniques normally provide more reliable results due to the absence of difficulties with the energy resolution .
however , the cross section can be determined only for a few discrete photon energies with this method .
this drawback will be cured below by the use of an appropriate theoretical model
. our strategy will be thus to analyze the radioactive isotope data .
we shall consider the range of energies up to 0.5mev above threshold .
we need to obtain the ground state and continuum wave functions ( wf ) and calculate the transition matrix element .
we start with the three
body @xmath18 representation of the @xmath0be system . within this representation
the wf in the c.m .
system is the @xmath19 relative motion function times the intrinsic wfs of the two @xmath6particles . since a predominant contribution to the transition matrix element comes from distances large compared to the size of the @xmath6particle additional antisymmetrization
may be disregarded .
system @xcite , for example . ] the intrinsic @xmath6 particle wfs then will drop out from the calculation . in the following
we shall refer to the three particle relative motion function as to the wf of the system .
let us denote and @xmath20 the distance between the @xmath6 particles and that from their center of mass to the neutron , respectively .
the ground state is @xmath21 , and its wf is of the form @xmath22_l\chi_{1/2}]_j .
\label{eq : gswf}\ ] ] here @xmath23 is the neutron spin function , and the brackets @xmath24 $ ] stand for angular momentum coupling . because of the pauli principle and parity requirements @xmath25 is even , and @xmath26 is odd .
the wf in eq .
( [ eq : gswf ] ) was obtained from the three body schrdinger equation with @xmath27 and @xmath28 potentials reproducing the observed two body phase shifts .
these potentials , along with some details of the calculation , are listed in the appendix .
practically an exact solution to the @xmath18 bound state problem is obtained , but one can not get the experimental binding energy with these `` bare '' interparticle interactions .
a possible reason is that the two @xmath6particles may distort each other in the @xmath18 bound state compared to the pure @xmath29 case .
this may lead to a change in the @xmath29 interaction . to obtain a reasonable ground state
wf the strength of the attractive central component of the @xmath29 potential is reduced by 8% in our calculation .
this leads to values of 1.50mev and 2.48fm for the binding energy and charge radius of @xmath0be , sufficiently close to the experimental values of 1.5736mev and ( @xmath30)fm .
coming to the continuum wave function , we note first that the @xmath31be resonance produced in the reaction may be safely treated as a stable particle for the purposes of our calculation .
this is because its width of 7kev is extremely narrow on a nuclear scale .
the @xmath27 continuum wave function , taken at the resonance energy and normalized to unity in the interior region , decreases practically to zero in the coulomb barrier region as shown in fig . 2 ( solid line ) .
this function represents the wf of the resonance extremely well and is taken as the @xmath31be `` bound state '' wave function .
second , we argue that photodisintegration of @xmath0be proceeds entirely into the @xmath31be@xmath32 channel . indeed , one can estimate that , at small energies considered , the three
fragment @xmath18 disintegration channel is strongly suppressed due to the threshold regime .
the experimental data also strongly supports the absence of this channel @xcite . at the same time ,
fragment @xmath8he@xmath33he channel is still closed and ineffective due to the broad width of @xmath34he .
thus our cross section starts from the @xmath31be@xmath32 threshold of 1.6654mev .
the cross section in the region between this threshold and the @xmath19 threshold of 1.5736mev is known to be tiny @xcite and will be disregarded . as in the previous work ( e.g. @xcite )
we confine ourselves to an s wave relative @xmath35be motion , i.e. , with @xmath16 continuum states .
predictions of the above dynamic @xmath19 model for the photodisintegration of @xmath0be depend crucially on the position of the excited state of @xmath0be with respect to the threshold .
preliminary three body calculations gave us a peak lying too high in energy , and too broad .
this could not be used for a reasonable fit to the data , so in the following we shall formulate an alternative representation of the continuum wf .
we shall seek it in the form @xmath36 here @xmath37 is the intrinsic wave function of @xmath31be calculated with the same @xmath27 potential as for the ground state , and @xmath38 is the @xmath35be relative motion function , where for large @xmath39 the normalization @xmath40 is used .
generally speaking , the true continuum wave function differs from eq .
( [ eq : cwf ] ) not only in the @xmath35be interaction region but also in the outer region .
however , at energies in the vicinity of the long living excited state of @xmath0be the representation ( [ eq : cwf ] ) should approximately be valid in the outer region .
indeed , the decay of the long living state into the three
body @xmath19 channel is inhibited due to the threshold regime . due to the approximate validity of the wf in eq .
( [ eq : cwf ] ) in the outer region one obtains a correct energy dependence of the cross section when using this wf . for small energies the main energy dependence of the transition matrix element appears as a factor in the continuum wf and
is determined by an outer part of the wf , i.e. the phase shift .
besides , one can see below that just the outer region ( where @xmath38 takes the form of eq .
( [ eq : sin ] ) ) gives the biggest contribution to the transition matrix element .
one can therefore hope that a wf of the form of eq .
( [ eq : cwf ] ) suffices for the fitting purpose in the whole energy range considered .
we seek the relative motion function @xmath38 as a solution to the relative motion schrdinger equation with some potential whose parameters are chosen from a fit of the theoretical cross section to the data .
taking into account that the s
wave @xmath28 repulsion and the p wave @xmath28 attraction have comparable ranges one can assume a smooth attractive potential .
the woods
saxon family @xmath41 will be adopted below as a good representative .
consider now the representation of the photodisintegration cross section in our model .
the matrix element of the dipole transition operator @xmath42 between the wfs in eq .
( [ eq : gswf ] ) and eq .
( [ eq : cwf ] ) has to be calculated . after integrating over @xmath43 the matrix element
reduces to the overlap between the @xmath35be relative motion functions , namely the scattering function entering eq .
( [ eq : cwf ] ) and the `` effective bound state wf '' @xmath44_{j=3/2}\ ] ] obtained from eq .
( [ eq : gswf ] ) : @xmath45 using eqs .
( [ eq : cwf ] ) , ( [ eq : eff ] ) the cross section is of the same form as in the single particle case . the total ( s wave ) photodisintegration cross section is calculated in the simple form @xmath46 here @xmath47 is the @xmath35be reduced mass , and @xmath48 is the excitation energy @xmath49 that will be denoted as @xmath50 below . in case of a single particle description of the process , i.e. , the `` valence neutron '' model , eq .
( [ eq : crs ] ) is valid with the bound state wf normalized to unity , while in our case ( see below ) @xmath51 it is implied here and in eq .
( [ eq : crs ] ) that the @xmath31be and @xmath0be ground state wave functions are normalized to unity .
the function @xmath52 is shown in fig . 2 ( dashed line ) .
in ref . @xcite the data at the energies up to 185kev above threshold ( @xmath53 mev ) from refs .
@xcite were reproduced at a qualitative level within the following framework . the valence neutron model with woods
saxon type potentials was used .
the results obtained in this way were multiplied by some constant factors less than unity , so called `` reduction factors '' , to approach the experimental cross section . to obtain the continuum wave function the depth of the potential
was varied while the radius and diffuseness parameters were taken the same as for the central component of the potential in the bound state calculation .
two fits were found , one with a reduction factor of 0.53 leading to a weakly bound @xmath31be@xmath32 @xmath16 state , and the other one with a reduction factor of 0.31 leading to a virtual state . the first possibility was once preferred in view of the results of ref .
@xcite . in that work a two channel @xmath31be@xmath32 model of the ground state of @xmath0be
was introduced to cure the single particle description @xcite of photodisintegration .
the @xmath31be subsystem was allowed to be in the ground and first excited state , and that led to the reduction factor of 0.5 or 0.6 in the cross section depending on the assumptions . in ref .
@xcite the reduction factor of 0.56 was found for that model .
in contrast to ref . @xcite our results below definitely testify to a virtual @xmath31be@xmath32 @xmath16 state .
this is probably due to a more realistic treatment of the ground state of @xmath0be in our model .
in fact the reduction factors obtained in the two channel model of the ground state of @xmath0be @xcite should be used in conjunction with the channel coupling n@xmath54be dynamics , instead of using @xcite single channel dynamics . in ref .
@xcite the same data were fitted with the line shape @xmath55 .
this shape was derived incorrectly from the breit wigner cross section under the assumption that the @xmath16 level of @xmath0be is a bound or virtual @xmath31be@xmath32 state . in ref .
@xcite the data of ref .
@xcite were fitted with a one level r
matrix approximation .
the fit leads to a complex energy resonant state @xcite and the real part of the complex energy proves to be negative .
first we shall analyze the data of refs .
@xcite ( full circles , full squares , and full diamond in fig .
1 ) . a search of the parameters @xmath56 , @xmath39 , and @xmath57 of the potential ( [ eq : ws ] ) giving an acceptable fit to the data is performed .
several local minima of the quantity @xmath58 in the space of the parameters are found .
one of them is provided by @xmath59 these values seem to be very reasonable . for this
set the @xmath60(degrees of freedom ) value equals 0.62 .
another one is obtained with the parameters @xmath61 giving @xmath62 .
several other minima also exist with sizable higher but still acceptable @xmath63 values . in fig .
3 the photodisintegration cross sections obtained with the parameters ( [ eq:1 ] ) and ( [ eq:2 ] ) are shown as the solid and dashed curve , respectively .
the two cross sections prove to be quite close to each other .
to clarify partially the reason for this we note that the biggest contribution to the matrix element of eq .
( [ eq : int ] ) comes from the distances beyond the range of the potential . the distances larger than 5 fm in eq .
( [ eq : int ] ) provide 60 - 70% contribution to the cross section , and at such distances the wave functions deviate from the asymptotic ones , eq .
( [ eq : sin ] ) , by not more than 10% ( except for regions in the vicinity of zeros ) .
the asymptotic wave functions are determined by the phase shifts i.e. , predominantly by the scattering length @xmath57 and the effective range @xmath64 .
therefore the procedure is equivalent to some degree to fitting @xmath57 and @xmath64 values .
once this is done , the cross sections are not very dependent on the particular set of the potential parameters . the @xmath57 and @xmath64 values are @xmath65fm and @xmath66fm , respectively , for the set ( [ eq:1 ] ) , and @xmath67fm and @xmath68fm for the set ( [ eq:2 ] ) .
all the other above mentioned sets of potential parameters giving local minima to @xmath63 lead to very similar @xmath57 and @xmath64 values .
however considerable changes in the scattering wf inside the potential can influence the results , see the next paragraph .
the following way to interpolate between the data has also been tried .
let us denote by @xmath69 the cross section obtained in case when the right hand side of eq .
( [ eq : sin ] ) is used as a continuum wave function for all @xmath39 values .
this cross section has been calculated taking @xmath70 from the effective range expansion with the @xmath57 and @xmath64 values given by the potential ( [ eq:1 ] ) .
let us represent @xmath71 as @xmath72 and fit @xmath73 to experiment .
it is hoped that , in contrast to @xmath71 , the factor @xmath73 behaves in a smooth way and thus can be reliably obtained from an interpolation procedure .
indeed , the behavior of both @xmath71 and @xmath74 can be approximately described by the resonant factor @xmath75 times a slowly varying function .
even a fit with @xmath76 proves to provide a sufficiently low @xmath63 value .
the cross section obtained with this @xmath77 is shown in fig . 3 as the dotted curve .
presumably this procedure provides less accurate results than the previous one .
of course , the energy dependence @xmath75 is not accurate enough in the whole energy range , as a comparison with the exact solution for the potential ( [ eq:1 ] ) shows
. hence @xmath73 should include an energy dependence , but this could not be determined because of experimental uncertainties .
next we applied the procedure to the data of ref .
@xcite ( open circles in fig .
1 ) . three local minima with acceptable @xmath63 values are found .
however the parameters of the potential corresponding to all of them : @xmath78 prove to be rather unrealistic .
there exists one more difference between these potentials and those in eqs .
( [ eq:1 ] ) and ( [ eq:2 ] ) .
the latter potentials , as well as the other potentials ( [ eq : ws ] ) providing a good fit to the same data , support one deeply bound state and one state close to being bound . on the contrary , the potentials ( [ fu ] ) support only one very weakly bound state .
an existence of one deeply bound @xmath79 level in the neutron mean field in the @xmath0be nucleus , or , equivalently , one node in the low energy scattering wave function inside the potential , seems to be natural from the shell model point of view .
we think this point of view is sufficient to establish the correct number of nodes for the neutron motion inside the woods
saxon potential , even for such a clusterized system . in the @xmath6particle oscillator model of @xmath0be @xcite ,
for example , the first allowed neutron @xmath79state contains a substantial admixture of the nodeless @xmath80 function , but this leads not to a disappearance but only to a shift of the node . if one admits that the state considered is a mixture of @xmath80 and @xmath81 oscillator functions then there exists just one node located within the distance of 3 fm from the origin .
therefore we conclude that in the region where various data sets differ from each other the older radioactive isotope data are preferable .
we also note that the cross section we obtain with potentials ( [ eq:1 ] ) and ( [ eq:2 ] ) for the highest energies considered , being lower than the fitted datum of ref .
@xcite , agrees well with the bremsstrahlung jacobson data @xcite .
the @xmath0be@xmath82 reaction rate per nucleus per time unit is calculated via the usual averaging the elementary photodisintegration cross section @xmath83 with the approximate , or wien distribution for the photon density , @xmath84 where @xmath85 mev .
the rate of the reverse reaction ( the number of reactions per time unit per unit volume ) is @xmath86 where @xmath87 and @xmath88 are numbers of particles per unit volume .
the reaction constant @xmath89 is obtained from eq .
( [ lam ] ) using the reverse ratio @xmath90 @xcite : @xmath91 here @xmath92 is avogadro s number , @xmath93 is the temperature in @xmath94 k , @xmath95 with @xmath96 mev , and it is implied that the quantities ( [ lam ] ) and ( [ 3 ] ) are given in sec@xmath97 and cm@xmath11 sec@xmath97 mole@xmath99 , respectively .
use of the wien distribution instead of the exact , or planck , one , i.e. @xmath100^{-1}\rightarrow\exp(-e_\gamma / kt)$ ] , allows application of the above listed simple reverse ratio theory . for temperatures of @xmath101 and 10 , for example
, it gives the reaction constant ( [ lam ] ) with relative errors of 1% and 5.4% , respectively .
for @xmath102mev the cross section @xmath103 obtained in the preceding section with the potential ( [ eq:1 ] ) is used . for @xmath104 from 2.2mev up to 5mev the jacobson bremsstrahlung data @xcite are used .
the former energy region provides 96% and 62% contribution to the cross section for @xmath105 and 5 , respectively .
the contribution from energies @xmath104 higher than 5mev reaches 0.3% and 6.7% for @xmath101 and 8 , respectively .
the values of the rate ( [ 3 ] ) obtained can be represented by the fit @xmath106 \left(1+\sum_{n=1}^7a_nt_9^n\right)^{-1 } \label{fit}\ ] ] with @xmath107 the fit reproduces our @xmath108 values with the accuracy better than 1% at any @xmath93 in the range @xmath109 . in eq .
( [ fit ] ) @xmath110/kt$ ] .
the factor @xmath111 $ ] represents the asymptotic behavior of the @xmath31be formation contribution to the rate when @xmath93 tends to zero . in the table
our values for the three body reaction rate ( [ 3 ] ) are compared with those of ref .
@xcite and those of ref .
@xcite where constructive or destructive interference between the resonant and non resonant contributions at energies above the resonance energy was assumed .
summarizing , we have constructed a semi microscopic model for the low energy photodisintegration of the @xmath0be nucleus and have analyzed the experimental data with its help .
our analysis supports the older radioactive isotope data .
the theoretical cross section we derived may be compared with future microscopic calculations of the process .
we have calculated the astrophysical rates for the reaction @xmath3be@xmath2 and the reverse reaction .
our new reaction rates agree at @xmath112 with the ones given in ref .
they are somewhat smaller ( larger ) for lower ( higher ) temperatures than @xmath112 .
the reaction rates given in ref .
@xcite agree much better with our reaction rate at higher temperatures if one assumes in ref .
@xcite constructive ( destructive ) interference between the resonant and non resonant contributions at energies above ( below ) the resonance energy .
we are indebted to j.s .
vaagen and j.m .
bang for very fruitful comments .
this work was supported partially by the the fonds zur frderung wissenschaftlichen forschung in sterreich ( project p10361phy ) and the russian foundation for basic research ( grant no 97 - 02 - 17003 ) .
in our calculation of the ground state of @xmath0be the @xmath27 potential is taken in the form @xcite @xmath113-v_a\exp[-(\mu_a\rho)^2]\ ] ] with @xmath114mev , @xmath115fm@xmath97 , @xmath117fm@xmath97 , and @xmath118 , 320 , and 10mev for @xmath119=0 , 2 , and 4 , respectively . the coulomb @xmath27 interaction is also added .
the @xmath29 interaction in s , p , and d states is taken into account .
as in many previous studies @xcite the s wave repulsive potential @xmath120 $ ] with @xmath121mev and @xmath122fm is used .
the initial potential in p and d
states @xcite includes central and spin orbit components : @xmath123 with @xmath124mev , @xmath125fm , @xmath126fm , @xmath127mev@xmath128 , @xmath129fm , and @xmath1300.35fm .
the parameter @xmath131 is reduced to 39.6mev in the present three
body calculation , in order to reproduce the empirical g.s .
energy .
the three body dynamic equation is written in the form of the faddeev differential equations and each faddeev component is expanded over hyperspherical harmonics and hyperradial basis functions . using the raynal revai rotations of hyperspherical harmonics
the matrix elements are reduced analytically to two dimensional integrals .
the equations are projected onto subspaces of the basis functions retained that reduces the problem to the algebraic eigenvalue problem .
the number of basis functions retained is quite high and ensures the adequate convergence of the calculation .
pudliner , v.r .
pandharipande , j. carlson , s.c .
pieper , and r.b .
wiringa , phys .
c*56 * , 1720 ( 1997 ) .
k. arai , y. ogawa , y. suzuki , and k. varga , phys .
c*54 * , 132 ( 1996 ) .
woosely , j.r .
wilson , g.j .
mathews , r.d .
hoffmann , and b.s .
meyer , astrophys .
j. * 433 * , 229 ( 1994 ) .
k. takahashi , j. witti , and h.t .
janka , astron . and astrophys .
* 286 * , 857 ( 1994 ) .
s.e . woosely and r.d .
hoffman , astrophys .
j. * 395 * , 202 ( 1992 ) .
j. grres , h. herndl , i.j .
thompson , and m. wiescher , phys .
c * 52 * , 2231 ( 1995 ) .
efros , w. balogh , h. herndl , r. hofinger , and h. oberhummer , z. phys .
a * 355 * , 101 ( 1996 ) .
h. herndl , r. hofinger , and h. oberhummer , in editors : s. kubono and t. kajino , proceedings of the symposium on origin of matter and evolution of galaxies 97 ( omeg97 ) , nov . 5 - 7 , 1997 , atami , japan , world scientific , singapore , in press .
fowler , g.r .
caughlan , and b.a .
zimmerman , ann .
. astrophys .
* 13 * , 69 ( 1975 ) .
f. ajzenberg - selove , nucl . phys . *
a490 * , 1 ( 1988 ) .
jacobson , phys . rev . * 123 * , 229 ( 1961 ) . b.l .
berman , r.l .
van hemert , and c.d .
bowman , phys .
* 163 * , 958 ( 1967 ) .
gibbons , r.l .
maclin , j.b .
marion , and h.w .
schmitt , phys . rev . * 114 * , 1319 ( 1959 ) . w. john and j.m .
prosser , phys .
rev . * 127 * , 231 ( 1962 ) .
b. hammermesh and c. kimball , phys . rev . * 90 * , 1063 ( 1953 ) . m. fuishiro , t. tabata , k. okamoto , and t. tsuimoto , can . j. phys . * 60 * , 1672 ( 1982 ) .
f.c . barker and b.m .
fitzpatrick , aust . j. phys .
* 21 * 415 ( 1968 ) .
ryzhih , r.a .
eramzhyan , v.i .
kukulin , and yu.m .
tchuvilsky , nucl . phys .
* a563 * , 247 ( 1993 ) . m. fuishiro , k. okamoto , and t. tsuimoto , can . j. phys . * 61 * , 1579 ( 1983 ) .
francis , d.t .
goldman , and e. guth , phys . rev . * 120 * , 2175 ( 1960 ) .
e.g. corman , j.e .
sherwood , and w. john , phys .
lett . * 4 * , 146 ( 1963 ) .
blair , phys . rev . *
123 * , 2151 ( 1961 ) .
barker , nucl .
* 28 * , 96 ( 1961 ) . c. mahaux , nucl
71 * , 241 ( 1965 ) .
f.c . barker , can .
* 61 * , 1371 ( 1983 ) .
kunz , ann .
* 11 * , 275 ( 1960 ) .
fowler , g.r .
caughlan , and b.a .
zimmerman , ann .
. astrophys .
* 13 * , 69 ( 1975 ) .
s. ali and a.r .
bodmer , nucl . phys .
* 80 * , 99 ( 1966 ) .
j. bang and c. gignoux , nucl .
a313 * , 119 ( 1979 ) .
zhukov , b.v .
danilin , d.v .
fedorov , j.m .
bang , i.j .
thompson , and j.s .
vaagen , phys . rep . * 231 * , 151 ( 1993 ) . | a semi microscopic model for the low energy photodisintegration of the @xmath0be nucleus is constructed , and the experimental data are analyzed with its help .
the older radioactive isotope data are supported by this analysis .
the theoretical photodisintegration cross section is derived .
the astrophysical rates for the reaction @xmath1be@xmath2 and the reverse photodisintegration of @xmath0be are calculated .
the new reaction rate for @xmath3be@xmath2 is compared with previous estimations . |
the fermion dipole moments are some of the best measured quantities in physics .
while the limits for the cp - violating ones , the electric dipole moments , are impressive@xcite ( @xmath2 e - cm for the electron and the neutron ) , the agreement between the measurement of the anomalous magnetic dipole moments of the electron and of the muon and its theoretical prediction is such that provides the best determination of the fine structure constant .
since the first prediction for the electron anomalous magnetic moment computed by schwinger@xcite , the magnetic properties of the leptons played a central role in testing symmetries and the quantum prediction of the theory .
this could be done thanks to the relativity long lifetime of the muon and the stability of the electron in ordinary atoms .
for the heavy fermions the situation is very different . as short lived - particles , there is no possibility of such high precision experiments .
nevertheless , the magnetic dipole moments are , for on - shell particles , physical magnitudes that not only contain important information about the interactions , but may also give some insights on the mechanism of mass generation . due to the high mass of the tau lepton ( and other heavy fermions )
the measurement of the dipole moments is an interesting test for the standard model and possible new physics that may show up there . as expressed by m.perl@xcite in the section _ dreams and odd ideas in tau research _ : " ... it would be very nice to measure @xmath3 with enough precision to check this [ the schwinger term @xmath4 , as it was checked for the @xmath5 and the @xmath6 years ago._at present such precision is a dream . " _
previous bounds on the tau magnetic moment @xmath7 are indirect @xcite , or include a partial analysis of the amplitudes in the considered processes@xcite . in the first case they come from the observation that in general extensions of the standard model ( sm ) it is very difficult to generate a weak magnetic moment , @xmath8 , for a lepton without originating a coupling of the z boson to the lepton of the same order of magnitude .
this weak magnetic coupling is strongly bounded by lep1 , therefore , by assuming that the magnetic moment of the lepton ( @xmath7 ) has the same size , one obtains a rather strong bound on it . on the other hand , far from the resonance the @xmath7 magnetic moment couplings of the tau is more important and should be taken into account in the theoretical and experimental analysis . finally , the tau - w magnetic coupling was first considered in the w decays into tau plus leptons@xcite and also at low energies in the tau decays into leptons and pions@xcite .
while these arguments are plausible , a complete analysis of the data coming from tau - lepton production at lep1 , sld and lep2 , and data on w decays into tau leptons from lep2 and @xmath9 colliders , allows for a better understanding of the magnetic moment couplings of the tau to the photon , the z and the w bosons .
more technical details of what follows can be found in @xcite .
the anomalous magnetic moments of the tau lepton , as the rest of the leptons , receive the standard contribution from the schwinger term .
however , in the case of the tau lepton , larger contributions from new physics are still allowed . at the electroweak scale
those can be studied within the effective lagrangian approach .
magnetic dipole moments are usually defined with all particles in the magnetic moment vertex on - shell because _ these are the gauge invariant quantities of the theory , as shown in the classical paper _ @xcite . for the @xmath7 tau magnetic dipole moment
there is no available experiment satisfying the above requirement and to determine it one has to resort to general and consistent ( gauge invariant ) parameterizations of all physics in which the magnetic moment could play a significant role .
the most general of those parameterizations is based on the effective field theory language which will allow us to relate off - shell matrix elements with the physical on - shell anomalous magnetic moment @xmath7 . instead , the @xmath8 can be determined on - shell : we have tau pair production at the z - peak at various experiments and specific observables proportional to the weak magnetic moment have been calculated@xcite and measured@xcite .
the effective lagrangian approach @xcite we use is described in section 2 where we also fix the notation . magnetic moment are generated by non - renormalizable couplings , their effects grow with energy and , therefore , high energy experiments , though not as precise as experiments at lower energies , can still be relevant .
this is the case for lep1 and sld measurements as compared to lep2 ones .
besides , while the former are only sensitive to the z - magnetic couplings , lep2 is sensitive to both z and photon magnetic moments ( @xmath8 and @xmath7 ) .
far from the resonance the statistics rapidly decreases and the precision is not as good . in general , lower energy experiments do not provide stringent bounds : the suppression factor @xmath10 has to be compensated by higher precision in the experiment ( see for instance the bounds obtained from tau decays @xcite ) .
observables are studied in section 3 where tau production is considered in universality test and transverse polarization at lep1-sld , total production rates at lep2 , and lepton universality in @xmath11 decays in lep2 and hadron colliders .
what are the observables related to the magnetic moments ? there is no symmetry guide as , for example , in cp - violation by electric dipole moments , where specific cp - odd observables can be constructed .
however , magnetic couplings flip chirality , so in order to select the physical effects of the magnetic moments one should take these properties into account . in the standard model the only source of chirality flips are fermion masses , much lower than the electroweak scale .
this means that any contribution to inclusive observables are either suppressed by the fermion masses or need two operator insertions and then , they come as the electroweak magnetic moments squared .
furthermore , any new physics terms , not only that related to the magnetic moments will appear in these observables .
it is therefore convenient to also include observables that are exactly zero when chirality is conserved : these will be only sensitive to fermion masses and/or magnetic moments and will depend linearly on the couplings .
up to now , observables of this kind have been studied at lep1 but not at lep2 .
there are only two gauge invariant dimension six operators contributing to magnetic moments in the effective lagrangian .
therefore gauge invariance imposes a relationship among the three weak and electromagnetic magnetic moments.therefore one can achieve some insight on tau magnetic moments by studying w decays into tau leptons .
there exist good bounds on universality of w decays into leptons coming from lep2 , ua1 , ua2 , cdf and d0 that can be translated into limits on the magnetic moments . in section 4 we present the combined analysis of the observables , compare them to other bounds found in the literature and discuss the results .
the new physics phenomenology at low energy , _
i.e. _ at energies much lower than the scale of this new physics , can be parametrized by an effective lagrangian built with the standard model particle spectrum and gauge symmetry , having as zero order term just the standard model lagrangian , plus higher dimension gauge invariant operators suppressed by the scale of new physics @xmath12 .
the leading non - standard effects will come from the operators with the lowest dimension .
those are dimension six operators and among them there are only two operators that contribute to the tau magnetic moments : @xmath13 @xmath14 here @xmath15 is the tau leptonic doublet , @xmath16 is the higgs doublet , @xmath17 and @xmath18 are the u(1)@xmath19 and su(2)@xmath20 field strength tensors .
other dimension six operators like @xmath21 reduce to the operator of eq .
( [ eq : ob ] ) after the use of the standard model equations of motion .
the effective lagrangian is @xmath22 where we will take the couplings @xmath23 and @xmath24 real .
after spontaneous symmetry breaking , the higgs gets a vacuum expectation value and , as usual , the interactions in ( [ eq : interaccio ] ) can be written in terms of the gauge boson mass eigenstates @xmath25 , @xmath26 and @xmath27 .
our lagrangian , written in terms of the mass eigenstates , is @xmath28 where one can recognize the usual couplings related to the magnetic moments .
@xmath29 ( for @xmath30 ) is the abelian field strength tensor for the gauge bosons .
we have not written the non - abelian couplings involving more than one gauge boson because they do not contribute to magnetic moments at leading order .
the above new physics magnetic moments can be written as @xmath31 where @xmath32 is the higgs vacuum expectation value and @xmath33 , @xmath34 .
the lagrangian ( [ eq : leff_fin ] ) gives additional contributions to the anomalous standard model electromagnetic moment of the tau @xmath35 .
the same is true for the other magnetic moments that have been introduced for the weak magnetic moments : @xmath8 for the @xmath36-boson@xcite , and @xmath37 for the @xmath11-boson@xcite .
sometimes it will be convenient to consider the following dimensionless couplings , proportional to the magnetic moments @xmath38 where @xmath39 .
we would like to stress the following point . in the effective lagrangian approach ,
_ exactly _ the same couplings that contribute to processes at high energies also contribute to the ( magnetic moment ) form factors , @xmath40 , at @xmath41 .
the difference @xmath42 only comes from higher dimension operators whose effect is suppressed by powers of @xmath43 , as long as @xmath44 as needed for the consistence of the effective lagrangian approach .
this means that any measurement ( bound ) for the new physics contributions to magnetic moments , at an energy scale where some of the gauge bosons or other particles are off - shell , can be directly added to the on - shell standard model prediction in order to get a value ( limit ) on the magnetic dipole moments .
the effects of the operators in ( [ eq : leff_fin ] ) are suppressed at low energies .
this means that the most interesting bounds will come from the highest precision experiments at the highest available energies . at present
they are lep1 and sld ( @xmath36 decay rates and polarization asymmetries ) , lep2 ( cross sections and @xmath11 decays rates ) , cdf and d0 ( @xmath11 decay rates ) .
consequently in the following section we will study all those observables .
in this section we will present the observables best suited to set bounds on the magnetic moments .
we begin with tau pair production in @xmath45 from threshold to lep2 energies .
therefore we will include both photon and z - exchange with standard model vector and axial couplings to fermions , plus additional magnetic moment couplings given by eq .
( [ eq : leff_fin ] ) .
the expressions for the cross section can be found in @xcite .
it can be separated into three types of contributions : i ) the standard model tree level contribution , which is the dominant one , ii ) a contribution which is proportional to the tau mass .
this comes together with an insertion of the magnetic moment operators ( non - standard contribution ) or with an insertion of another fermion mass ( standard model contributions ) or both , two insertions of magnetic moment operators and two mass insertions ( non - standard contributions ) , and iii ) a contribution free of masses but with two insertions of the magnetic moment operators .
this can be easily understood , since standard model couplings of gauge bosons to fermions conserve chirality , while mass terms and magnetic moment couplings change it , therefore interference of magnetic moment contributions with standard ones should be proportional to the fermion masses and only the square of magnetic moments can be independent of fermion masses . in the limit of zero tau mass only the contribution iii ) is relevant , however there could be some range of the parameters in which contribution ii ) is higher than iii ) .
in fact for any finite value of the tau mass it is obvious that for large enough @xmath12 ii ) will always dominate over iii ) . in order to be as general as possible
we will include all three contributions .
bounds on the couplings @xmath7 and @xmath8 can be obtained from lep1-sld universality tests by assuming that only the tau lepton has anomalous magnetic moments ( muon and electron electromagnetic moments have been measured quite precisely @xcite ) . in order to compare with experimental data
it is convenient to define the universality ratio : @xmath46 here @xmath47 and @xmath48 are the quantities directly measured @xcite , @xmath49 is the standard model contribution ( including lepton - mass corrections ) , and @xmath50 and @xmath51 are the linear and quadratic terms , respectively , in the tensor couplings .
notice that this quotient eliminates the hadronic part that , although very well determined in experiments , has much higher theoretical uncertainties than @xmath52 .
the theoretical expression for @xmath52 can be computed from the cross sections , and it is @xmath53~ , \label{eq : ratio1}\ ] ] @xmath54 @xmath55 \,\frac{1}{v^2+a^2}\ , \ez^2\label{eq : rat2}\ ] ] with @xmath56 , @xmath57 .
notice that in the ratio eq .
( [ eq : ratiocross ] ) the electroweak radiative corrections cancel to a large extent and , therefore , we can use tree - level formulae .
however , if needed , the expressions in can be improved by using effective couplings @xcite . from the very precise experimental lep1 and sld measurements
@xcite we obtain @xmath58 comparing the equation ( [ eq : ratiocross1 ] ) with ( [ eq : ratiocross ] ) one gets @xmath59 that leads to the following two bands for @xmath60 : @xmath61 or @xmath62 at lep2 , the contributions coming from the photon - exchange are dominant over those coming from the @xmath36-exchange , and as a result both magnetic moments , @xmath7 and @xmath8 , will enter into the constraints .
present experimental errors from @xmath63 cross section are much milder at lep2 than at lep1 . a combination of the lep2 data @xcite on this cross section , for @xmath64 ( the invariant mass of the pair of tau leptons ) so that @xmath65 , is listed in table [ tabla1 ] .
this combination of data has been only made for the 183 gev and 189 gev data - sets as they have the highest luminosity and center - of - mass energies .
for comparison we also present the standard model prediction for the cross section . in both , experimental results and standard model predictions , initial - final state radiation photon interference is subtracted .
.combined experimental data for the @xmath66 cross section from aleph , delphi , l3 , opal at lep2 energies ( in gev ) .
@xmath67 is the cut in the invariant mass of the tau pair . [
cols="^,^,^,^",options="header " , ] [ tabla2 ] all data are used independently in the global fit discussed in the final section . just as an example of
how well the new couplings can be bound from lep2 let us find the limits on @xmath68 obtained by using only the data at @xmath69 gev .
the experimental value for the ratio @xmath70 is in this case : @xmath71 to be compared with the theoretical prediction ( assuming that @xmath60 is well bounded from lep1-sld , as it is ) @xmath72 from the two previous equations we find : @xmath73 this @xmath74 bound is comparable to the one obtained in the global fit given in the final section where all available data have been included .
chirality flipping observables vanish for massless taus and depend linearly on magnetic moments . on the other hand they will not get contributions from physics conserving chirality . in that sense they are truly magnetic moment observables . at lep1 , with the @xmath75 direction fully reconstructed in the semi - leptonic
decays it has been shown@xcite that one can measure the anomalous weak magnetic moment @xmath8 by measuring an azimuthal asymmetry on the tau and tau decay products angles .
this asymmetry selects the leading term in the weak magnetic moment in the tau pair production cross section , that appears in the p - odd , chirality flipping transverse tau polarization .
the expression one finds for the proposed asymmetry is : @xmath76~,\label{l3}\end{aligned}\ ] ] where @xmath77 , is the polarization analyzer for each hadron channel ( @xmath78 ) , @xmath79 , and @xmath80 . in addition , the sm contribution to this observable is doubly suppressed with respect to the non - standard one : by the fermion - boson vector coupling @xmath81 and by the @xmath82 factor . within @xmath74 , the lep1 measurement of this asymmetry and the sld determination of the transverse tau polarization@xcite , translate into the following values for the @xmath83 coupling @xmath84 combining these results one gets the bound : @xmath85
note that even though the transverse tau polarization has been measured at lep1-sld with a precision one order of magnitude worse than the universality test @xmath86 ( 2 - 4% typically for the asymmetry , and 0.5% for the tau - muon cross section ratio ) , the obtained bound eq .
( [ eq : ass1 ] ) is as good as the one coming from universality eq .
( [ eq : ratiocross1limit ] ) .
this is so because the asymmetry depends linearly on the couplings . at present there
does not exists a similar measurement at lep2 .
this would allow to disentangle the @xmath87 components from the @xmath36 components of the magnetic moments .
the lagrangian eq .
( [ eq : leff_fin ] ) shows that the same couplings that give rise to electromagnetic and @xmath36-boson magnetic moments , also contribute to the couplings of the @xmath11 gauge bosons to tau leptons .
the couplings appear in a different combination than that in the photon or @xmath36 couplings , so their study gives additional independent information on magnetic moment couplings . as was noticed in ref .
@xcite the best place to look for effects of the @xmath88 coupling is in the @xmath11 decay widths . using our effective lagrangian
we can easily compute the ratio of the decay width of the w - gauge boson in tau - leptons ( with magnetic moments ) to the decay width of the w to electrons ( without magnetic moments ) .
@xmath89 \label{eq : ratiow}\end{aligned}\ ] ] where @xmath90 , and @xmath88 can be rewritten in terms of @xmath91 and @xmath83 as in eq .
( [ eq : epsilonw ] ) .
note that @xmath92 , like the cross sections studied in subsection ( 1 ) , is a chirality even observable .
the decay of the @xmath11 into leptons has been measured to a rather good precision at lep2 , ua1 , ua2 , cdf and d0 .
there , results are presented in the form of universality tests on the couplings @xcite that we rewrite as a measurement on the ratio @xmath92 defined above @xmath93 then , from , we obtain the following limit on the @xmath11-boson magnetic moments . @xmath94
we have performed a global fit , as a function of the two independent couplings @xmath91 and @xmath83 , to the studied observables : * lepton universality @xmath95 at lep1 and sld ; * the ratio of cross sections @xmath96 , for the two highest energies measured at lep2 ; * the transverse tau polarization and polarization asymmetry @xmath97 measured at sld and lep1 ; * lepton universality of w decays @xmath98 measured at lep2 and @xmath9 colliders . regions coming from the different experiments : solid ( lep2 - 189 gev ) , dashed ( lep1-sld cross section ) , dot - dashed ( asymmetry ) .
we also have plotted the line @xmath100 ( dotted line ) .
this relationship appears when only the operator @xmath101 contributes .
[ fig : gfit ] ] for comparison we also present ( at 1@xmath99 ) the relevant limits set independently by the different observables , as discussed in the text . by projecting onto the axes one can read off the 1@xmath99 and 2@xmath99 limits on the different non - standard contributions to the anomalous electromagnetic and weak magnetic moments @xmath7
, @xmath8 @xmath102 using the relationship among @xmath91 , @xmath83 , @xmath23 and @xmath24 at a given value of the scale of new physics , one can easily obtain bounds on @xmath23 and @xmath24 .
alternatively , by assuming that @xmath103 or @xmath104 are order unity one finds a bound on the scale of new physics : @xmath105 tev .
bounds on the anomalous electromagnetic moment for the @xmath75 have been considered and measured from the radiative decay @xmath107 at lep1 @xcite . there , only the anomalous coupling @xmath7 is taken into account , while the contributions coming from the tau @xmath36-magnetic coupling @xmath8 or the effective 4-particle vertex are neglected .
the interpretation of off - shell form factors is problematic since they can hardly be isolated from the other contributions and gauge invariance can be a problem . in the effective lagrangian approach all those problems
are solved because form factors are directly related to couplings in the effective lagrangian , which is gauge invariant , and as discussed above , the difference @xmath42 only comes from higher dimension operators whose effect is suppressed by @xmath43 . using this approach , the analysis of the l3 and opal collaborations lead to the pdg @xmath108 cl limit@xcite : @xmath109 as can be seen from eq .
( [ eq : final2sigma ] ) our result , coming mainly from lep2 , is about one order of magnitude better than the ones obtained from the radiative @xmath36-decay .
the above bounds are completely model independent and no assumption has been made on the relative size of couplings @xmath23 and @xmath24 in the effective lagrangian ( [ eq : leff ] ) . for the sake of comparison with published data @xcite we present now the limits that can be found by considering separately only operator @xmath101 or only operator @xmath110 in the lagrangian ( [ eq : leff ] ) .
consider that only @xmath101 is present , as in ref .
@xcite , is equivalent to impose the relation @xmath111 or equivalently @xmath112 .
thus , from fig . [
fig : gfit ] , it is straightforward to obtain that the bounds on the anomalous magnetic moment ( at @xmath113 ) are reduced to @xmath114 , while little change is found on the weak - magnetic moment @xmath115 .
universality tests in @xmath11 decays do not provide any interesting constraint on @xmath8 and @xmath7 .
in fact , the straight lines coming from the direct 1@xmath99 bound from universality tests in @xmath11 decays lie well outside the figure .
however , because of the relationship ( [ eq : epsilonw ] ) , the lep1-sld and lep2 constraints on @xmath60 and @xmath68 can be translated into constraints on @xmath88 ( or @xmath37 defined in eq .
( [ eq : moments ] ) ) , the weak magnetic moment couplings of the @xmath11-gauge - boson to taus and neutrinos .
one obtains the 95% cl limits @xmath116 we have shown that the use of all available data at the highest available energies ( lep1 , sld , lep2 , d0 , cdf ) leads to strong constraints in all the magnetic moments ( @xmath117 ) of the tau lepton without making any assumption about naturalness or fine tuning .
the obtained bounds ( eq . ( [ eq : final2sigma ] ) and eq .
( [ eq : kweak ] ) ) are , to our knowledge , the best bounds that one can find in published data .
opal collaboration , g. abbiendi _
et al . _ , eur .
c6 ( 1999 ) 1 , eur .
c13 ( 2000 ) 553 ; aleph collaboration , r. barate _
et al . _ ,
c12 ( 2000 ) 183 ; l3 collaboration , m. acciarri _
et al . _ ,
b479 ( 2000)101 . | we obtain new bounds for the magnetic dipole moments of the tau lepton .
these limits on the magnetic couplings of the tau to the electroweak gauge bosons ( @xmath0 ) are set in a model independent way using the most general effective lagrangian with the @xmath1 symmetry .
comparison with data from the most precise experiments at high energies shows that the present limits are more stringent than the previous published ones . for the anomalous magnetic moment the bounds are , for the first time , within one order of magnitude of the standard model prediction . |
the hadronic b decays are important for testing the standard model ( sm ) , and also for uncovering the signal of new physics .
understanding non - leptonic @xmath1 meson decays is crucial for the ckm matrix elements measurements and cp violation detection .
especially the direct cp violation is expected to be measured in hadronic b decays .
recently , both belle and babar claimed to find direct cp asymmetry in @xmath2 and @xmath3 decays @xcite .
it is the start of direct cp measurement in b physics . in the theoretical side , there are much more complication in direct cp asymmetry predictions .
the direct cp violation in hadronic decays requires at least two decay amplitudes with different weak phase and strong phase . in the standard model ,
the weak phase comes from the so called ckm matrix , which is well defined .
although the weak phase from sm is clean , the strong phase requires hadronic matrix element calculation , which is usually model dependent .
the simplest case is two - body hadronic @xmath1 meson decays , for which bauer , stech and wirbel ( bsw ) proposed the naive factorization assumption ( fa ) in their pioneering work @xcite .
considerable progress , including generalized fa @xcite and qcd - improved fa ( qcdf ) @xcite , has been made .
on the other hand , technique to analyze hard exclusive hadronic scattering was developed by brodsky and lepage @xcite based on collinear factorization theorem in perturbative qcd ( pqcd ) . a modified framework based on @xmath0 factorization theorem
was then given in @xcite , and extended to exclusive @xmath1 meson decays in @xcite .
the predictions of branching ratios agree well with experiments in most cases , thus it is difficult to tell from experiments that which method is better than others . however , the strong phase , which is important for the cp violation prediction , is quite sensitive to various approaches .
the mechanism of this strong phase is quite different for various method , and give quite different results .
the recent experimental results @xcite can make a test for the validity of these approaches . in this paper
, we will first introduce the factorization approach in next section .
the qcd factorization and improved pqcd approach based on @xmath0 factorization are then introduced in section [ sec3 ] and section [ sec4 ] , respectively . in section [ sec5 ] ,
we upgrade the numerical results of these approaches using newest parameters .
we compare the three major approaches to show the difference of direct cp asymmetry predicted by them . finally the summary is presented .
the calculation of non - leptonic decays involves the short - distance lagrangian and the calculation of hadronic matrix elements which are model dependent .
the short - distance qcd corrected lagrangian is calculated to next - to - leading order @xcite . in 1987 ,
bauer , stech and wirbel first calculate the hadronic @xmath1 and @xmath4 decays using the naive factorization approach @xcite . in their factorization method ,
the hadronic matrix element is expressed as a product of two factors @xmath5 .
the first factor is proportional to the @xmath6 form factor , while the second one is proportional to the decay constant of @xmath7 meson .
all the perturbative channel dependent part is described by the wilson coefficients of four quark operators .
these effective four - quark operators are induced by the weak interaction in the quark level , which gives the short distance contribution to the non - leptonic decays of b mesons .
the effective hamiltonian for the charmless non - leptonic b decays is @xcite @xmath8 , \end{aligned}\ ] ] where @xmath9 and @xmath10 denotes the ckm factors .
the operators @xmath11 are tree level current operators .
the operators @xmath12 are qcd penguin operators . @xmath13 arise from electroweak penguin diagrams , which are suppressed by @xmath14 .
only @xmath15 has a sizable value whose major contribution arises from the @xmath16 penguin .
the @xmath17 s are the wilson coefficients of four quark operators with qcd corrections .
although this is a very simple method , later experiments show that many of the decay branching ratios explained well by the fa @xcite , especially for the color enhanced decays class i , iii and class iv @xcite .
when charmless @xmath1 meson decays are considered , efforts have been made to generalize the fa approach @xcite .
to explain the non - factorizable dominated class ii and class v decays , phenomenological parameter @xmath18 is introduced .
most of the branching ratios of hadronic @xmath1 decays agree well with experiments by @xmath19 @xcite .
an effort is also made to predict the cp violation parameters in different hadronic @xmath1 decays @xcite .
since direct cp violation requires a strong phase difference between decay amplitudes in addition to a weak phase difference , the precision of strong phase calculation is essential . in naive fa
, the hadronization is described by the form factor only , therefore , no strong phase is given .
( 130,160)(0,20 ) ( 60,72)(0,72 ) ( 60,72)(60,145 ) ( 90,72)(30,72,71 ) ( 120,72)(170,72)54 ( 60,72)5.5 ( 170,72)(230,72 ) ( 170,72)(170,145 ) ( 65,135)@xmath20 ( 96,105)@xmath21 ( 5,78 ) @xmath22 in generalized fa @xcite , the next - to - leading order wilson coefficients contain strong phases generated by the charm quark loop , where charm quark can be on shell . the diagram shown in fig.[fig1 ] is also called bss mechanism @xcite . the size of this kind of strong phase is sensitive to the momentum of the gluon connecting to the charm quark loop .
however , in fa , this momentum is not well defined , since all hadronic dynamics are defined only by form factors . in ref .
@xcite , the authors use @xmath23gev@xmath24 to give the cp asymmetry parameters of many channels .
in 1999 , beneke , buchalla , neubert , and sachrajda ( bbns ) proposed a formalism for two - body charmless @xmath1 meson decays @xcite . in this approach , they expand the hadronic matrix element by the heavy b quark mass and @xmath25 @xmath26.\label{bbnsf}\ ] ] here @xmath27 is a local operator in the effective hamiltonian and @xmath28 are bilinear quark currents . by neglecting the power corrections of @xmath29 , one need only calculate the order @xmath25 corrections including the vertex corrections for the four quark operators and the non - factorizable diagrams . these diagrams are shown in fig.[figbbns ] .
the first 6 diagrams have already been included in the generalized fa approach as next - to - leading order qcd corrections to local four quark operators .
what new are the last two non - factorizable diagrams , which has a hard gluon line connecting the four quark operator and the spectator quark .
they claimed that factorizable contributions , for example , the form factor @xmath30 in the @xmath31 decays , are dominated by long distance contributions .
hence , it is treated in the same way as fa , and expressed as products of wilson coefficients and form factor @xmath30 .
the non - factorizable contributions calculated perturbatively , are written as the convolutions of hard amplitudes with three @xmath32 meson wave functions .
annihilation diagrams are neglected as in fa .
values of form factors at maximal recoil @xmath33 and non - perturbative meson wave functions are all treated as input parameters .
it is easy to see from eq.([bbnsf ] ) , that this equation is only applicable for those color enhanced decay modes , where the factorizable contribution dominates the final results . while for the color suppressed modes , where the non - factorizable contributions are not small , the expansion of eq.([bbnsf ] ) is not right , since the large non - factorizable contribution is grouped into the next - to leading order term of eq.([bbnsf ] ) .
the numerical results show that the theory and experiments agree well for those class i and iv decays , which are color enhanced and dominated by the factorizable contribution .
this also agrees with the fa result , since the dominant part in eq.([bbnsf ] ) is the same as the fa .
the success of qcd factorization is that one can calculate the sub - leading @xmath34 non - factorizable contribution ( second term in eq.([bbnsf ] ) ) using perturbative qcd .
while in the fa , one has to input a free parameter @xmath18 to accommodate the non - factorizable contribution . in the qcd factorization calculations , people found that there exist endpoint divergence in the annihilation diagram calculations @xcite .
logarithm divergence occurred at twist 2 contribution , and linear divergence exists in twist 3 contribution . if not symmetric wave function , like @xmath35 meson , there is also soft divergence in the non - factorizable diagrams .
it is very difficult to treat these singularity in the bbns approach .
a cut - off is introduced to regulate the divergence , thus makes the qcd factorization approach prediction parameter dependent especially in the annihilation diagrams . as for the strong phase in bbns , like in fa
, it mainly comes from the bss mechanism . here
the momentum of the inner gluon is well defined , since wave functions are introduced .
however , it predicts too small strong phase , because of the small gluon momentum .
hence , small direct cp asymmetry is predicted .
there is also another source of strong phase from the annihilation diagrams , but strongly depends on the cut - off parameter .
the strong phase in qcd factorization can be large due to this cut - off . in a word , the qcd factorization approach is at least one step forward from naive factorization approach .
it gives systematic prediction of sub - leading non - factorizable contribution for the class i and class iv decays , which are dominated by the factorizable contribution .
problem remained is the endpoint singularity in higher order calculations , but may be solved with sudakov form factors like pqcd approach .
in this section , we will introduce the idea of pqcd approach .
the three scale pqcd factorization theorem has been developed for non - leptonic heavy meson decays @xcite , based on the formalism by brodsky and lepage @xcite , and botts and sterman @xcite . in the non - leptonic two body b decays , the @xmath1 meson is heavy , sitting at rest .
it decays into two light mesons with large momenta .
there must be a hard gluon to kick the light spectator quark ( with small momentum ) in the b meson to form a fast moving light meson .
so the dominant diagram in this theoretical picture is that one hard gluon from the spectator quark connecting with the other quarks in the four quark operator of the weak interaction . unlike the usual factorization approach ,
the hard part of the pqcd approach consists of six quarks rather than four .
we thus call it six - quark operators or six - quark effective theory
. there are also infrared ( soft and collinear ) gluon exchanges between quarks . summing over those leading soft contributions gives a sudakov form factor , which suppresses the soft contribution .
therefore , it makes the pqcd reliable in calculating the non - leptonic decays . with the sudakov resummation
, we can include the leading double logarithms for all loop diagrams , in association with the infrared contribution .
there are three different scales in the b meson non - leptonic decays @xmath36 , @xmath37 and @xmath38 .
the first scale describe the intrinsic electroweak decay of @xmath1 meson , through charged current .
the second scale @xmath37 denote the scale of energy release in the decay .
since the @xmath39 quark decay scale @xmath37 is much smaller than the electroweak scale @xmath40 , the qcd corrections to the four quark operators are non - negligible , which are usually summed by the renormalization group equation .
this has already been done to the leading logarithm and next - to - leading order for years @xcite .
the third scale @xmath38 involved in the @xmath1 meson exclusive decays is usually called the factorization scale , with @xmath39 the conjugate variable of parton transverse momenta . the dynamics below @xmath38 scale
is regarded as being completely non - perturbative , and can be parameterized into meson wave functions .
the meson wave functions are not calculable in pqcd .
but they are universal , channel independent
. we can determine them from experiments , and they are constrained by qcd sum rules and lattice qcd calculations . above the scale @xmath38 , the physics is channel dependent .
we can use perturbation theory to calculate channel by channel .
with all the large logarithms resummed , the remaining finite contributions are absorbed into a perturbative b quark decay sub - amplitude @xmath41 .
therefore the three scale factorization formula is given by the typical expression , @xmath42 , \label{eq : factorization_formula}\ ] ] where @xmath43 are the corresponding wilson coefficients , @xmath44 are the meson wave functions and the variable @xmath45 denotes the largest mass scale of hard process @xmath46 , that is , six - quark effective theory .
the quark anomalous dimension @xmath47 describes the evolution from scale @xmath45 to @xmath38 .
since logarithm corrections have been summed by renormalization group equations , the above factorization formula does not depend on the renormalization scale @xmath48 explicitly .
as shown above , in the pqcd approach , we keep the @xmath0 dependence of the wave function . in fact , the approximation of neglecting the transverse momentum can only be done at the non - endpoint region , since @xmath49 is qualified at that region . at the endpoint , @xmath50
, @xmath0 is not small any longer , neglecting @xmath0 is a very bad approximation . by , keeping the @xmath0 dependence , there is no endpoint divergence as occurred in the qcd factorization approach , while the numerical result does not change at other region .
furthermore , the sudakov form factors suppress the endpoint region of the wave functions .
the main input parameters in pqcd are the meson wave functions .
it is not a surprise that the final results are sensitive to the meson wave functions .
fortunately , there are many channels involve the same meson , and the meson wave functions should be process independent . in all the calculations of pqcd approach , we follow the rule , and we find that they can explain most of the measured branching ratios of b decays .
for example : @xmath51 decays , @xmath52 , @xmath53 decays @xcite , @xmath54 decays @xcite , @xmath55 decays @xcite , the form factor calculations of @xmath56 , @xmath57 @xcite , @xmath58 decays @xcite , @xmath59 decays @xcite etc .
we emphasize that non - factorizable ( fig.[fig2](c)(d ) ) and annihilation diagrams ( fig.[fig2](e - h ) ) are indeed sub - leading in the pqcd formalism as @xmath60 .
this can be easily observed from the hard functions in appendices of ref.@xcite .
when @xmath61 increases , the @xmath1 meson wave function enhances contributions to factorizable diagrams
. however , annihilation amplitudes , being independent of b meson wave function , are relatively insensitive to the variation of @xmath61 . hence ,
factorizable contributions become dominant and annihilation contributions are sub - leading in the @xmath62 limit @xcite . although the non - factorizable and annihilation diagrams are sub - leading for the branching ratio in color enhanced decays , they provide the main source of strong phase , by inner quark or gluon on mass shell .
the bss mechanism should also be present in the pqcd approach . however
this mechanism of strong phase is small , since it is at next - to - leading order @xmath63 corrections .
in numerical analysis , we use recently updated decay constants and form factors for the fa and qcdf approaches .
the branching ratios changes a little bit , since they are sensitive to these parameters .
but the direct cp asymmetry changes very little . in the pqcd approach ,
we use the full set of light meson wave functions , including two twist 3 distribution amplitudes , where only one used in the previous papers @xcite .
the threshold resummation for the endpoint of the hard part calculation is also newly included @xcite .
the numerical results for the branching ratios and cp asymmetries change only a little bit .
.direct cp asymmetries calculated in fa @xcite , qcdf @xcite and pqcd @xcite for @xmath64 and @xmath54 decays together with the averaged experimental results at percentage . [ cols="<,^,^,^,^",options="header " , ] as discussed in the previous section , the strong phase generated from pqcd approach is quite different from the fa and qcdf approaches .
the direct cp asymmetry in sm is proportional to the sine of the strong phase difference of two amplitudes .
therefore the direct cp asymmetry will be different if strong phase is different .
the predicted cp asymmetry by the three methods are shown in table [ cp ] .
it is easy to see that the fa @xcite and qcdf @xcite results are quite close to each other , since the mechanism of strong phase is the same for them .
recently the two b factories measure some channels with non - zero direct cp asymmetry @xcite , which are shown in table [ cp ] @xcite .
it is claimed that direct cp has been found in @xmath2 and @xmath65 decays with more than @xmath66 signal .
it is easy to see that our pqcd results of direct cp asymmetry @xcite agree with the experiments , especially for the experimentally well measured channels @xmath67 and @xmath68 decays .
although fa and qcdf are not yet ruled out by experiments , but the experiments at least tell us that the dominant strong phase should come from the mechanism of pqcd not qcdf .
charm quark loop mechanism , which gives the central value of strong phase in qcdf , is argued next - to leading order in pqcd .
this argument is now proved by b factories experiments . in summary , branching ratios of both fa and qcdf depend on the values of form factors .
the strong phase generated from these approaches is dominantly from perturbative bss mechanism . in the pqcd approach , dominant by short distance contribution ,
the form factors are sensitive to meson wave functions . by including the @xmath0 dependence and sudakov suppression ,
there is no endpoint divergence . in the pqcd formalism ,
annihilation amplitudes are of the same order as non - factorizable ones in powers of @xmath69 , which are both @xmath70 .
the strong phase comes mainly from the annihilation and non - factorizable diagrams in pqcd approach , which is quite different from the fa and qcdf approaches .
the experimentally measured direct cp asymmetry implies that pqcd gives at least the dominant strong phase than other approaches .
this will be further tested by experiments .
m. bauer , b. stech , m. wirbel , z. phys .
c * 29 * , 637 ( 1985 ) ; _ ibid . _ * 34 * , 103 ( 1987 ) .
a. ali , g. kramer and c.d .
l , phys .
d58 , 094009 ( 1998 ) ; c.d .
l , nucl .
74 , 227 - 230 ( 1999 )
cheng , phys .
b * 335 * , 428 ( 1994 ) ; h.y .
cheng , z. phys .
c * 69 * , 647 ( 1996 ) .
m. beneke , g. buchalla , m. neubert , and c.t .
sachrajda , phys .
* 83 * , 1914 ( 1999 ) ; nucl . phys . * b591 * , 313 ( 2000 ) .
lapage and s.j .
brodsky , phys .
b * 87 * , 359 ( 1979 ) ; phys . rev .
d * 22 * , 2157 ( 1980 ) ; c .- s .
huang , phys .
fortis et phys .
nucl . 4 ( 1980 ) 761 .
( chinese phys . 2 ( 1982 ) 179 .
cao , y .- b .
dai and c .- s .
huang , eur .
j. c11 ( 1999 ) 501 .
j. botts and g. sterman , nucl .
* b225 * , 62 ( 1989 ) .
li and g. sterman , nucl . phys .
* b381 * , 129 ( 1992 ) .
h - n . li and h.l .
yu , phys .
* 74 * , 4388 ( 1995 ) ; phys .
b * 353 * , 301 ( 1995 ) ; phys .
d * 53 * , 2480 ( 1996 ) .
chang , h.n .
li , phys .
d55 , 5577 ( 1997 ) ; t.w . yeh and h.n .
li , phys .
d56 , 1615 ( 1997 ) .
cheng , h - n .
li , and k.c .
yang , phys .
d * 60 * , 094005 ( 1999 ) . for a review , see g. buchalla , a.j .
buras , m.e .
lautenbacher , rev .
68 , 1125 ( 1996 ) .
m. neubert , v. rieckert , b. stech , q.p .
xu , heavy flavors , ed . by a.j .
buras , m. lindner , world scientific ; m. neubert , b. stech , adv .
high energy phys.15 , 294 - 344 ( 1998 ) .
a. ali , g. kramer and c.d .
l , phys .
d59 , 014005 ( 1999 ) ; c .- d .
l , p245 , proceedings of 3rd international conference on b physics and cp violation ( bconf99 ) , taipei , dec 1999 , world scientifc , hep - ph/0001321 .
m. bander , d. silverman and a. soni , phys .
43 , 242 ( 1979 ) ; j.m .
gerard and w.s .
hou , phys .
d43 , 2909 ( 1991 ) ; h. simma , g. eilam and d. wyler , nucl .
b352 , 367 ( 1991 ) .
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 . rev .
* d63 * , 074006 ( 2001 ) .
chen , h .-
li , phys .
d63 , 014003 ( 2001 ) .
t. kurimoto , h .-
sanda , phys .
d65 , 014007 ( 2002 ) ; c .- d .
l , m .- z .
yang , eur .
j. * c28 * 515 ( 2003 ) .
e. kou , a.i .
sanda , phys .
b525 , 240 - 248 ( 2002 ) .
chen , y .- y .
keum , h .-
li , phys .
d64 , 112002 ( 2001 ) ; s. mishima , phys .
b521 , 252 - 258 ( 2001 ) .
li , phys .
d66 , 094010 ( 2002 ) ; h.n .
li and k. ukai , phys .
b555 , 197 ( 2003 ) .
keum , a.i .
sanda , phys .
d67 , 054009 ( 2003 ) ; proceedings of 2nd workshop on the ckm unitarity triangle , durham , england , 2003 , hep - ph/0306004 ; y .- y .
keum , j. korean phys .
45 , s472 ( 2004 ) . | there are different approaches for the hadronic b decay calculations , recently . in this paper
, we upgrade three of them , namely factorization , qcd factorization and the perturbative qcd approach based on @xmath0 factorization , by using new parameters and full wave functions . although they get similar results for many of the branching ratios , the direct cp asymmetries predicted by them are different , which can be tested by recent experimental measurements of b factories . |
according to all hydro - dynamical simulation for structure formation run in the framework of a @xmath8cdm cosmology , large concentrations of galaxies are the best tracers of the filamentary web of dark - matter that our local universe is made of .
embedded into these filaments of shining ordinary matter ( stars in galaxies ) , should be hidden the , still to be found , largest reservoir of baryons in the local universe : the so called warm - hot intergalactic medium ( whim : e.g cen & ostriker , 2006 ) .
this , metal enriched ( through galaxy - igm feedback ) , otherwise primordial , medium should have temperatures in the range logt@xmath9 ( in k ) , and very low baryon densities @xmath10 @xmath11 . at such temperatures , h
and he are mostly fully ionized ( and so very difficult to detect ) , and the most abundant metal , oxygen , is mainly present in its stable he - like form : ovii .
the strongest bound - bound transition of the ovii ion is the k@xmath1 at @xmath12 , and should thus imprint absorption lines in the soft x - ray spectra of background objects whose lines of sight cross one or more whim filaments between us and the target .
however , these lines are expected to be extremely weak .
expected ovii column densities along a random line of sight crossing a whim filament , are n@xmath13_few_@xmath14 @xmath7 ( e.g. cen & fang , 2006 ) , giving rise to redshifted ovii k@xmath1 absorption lines with ew@xmath15 m . for these reasons detecting the whim
has proven to be very challenging .
the few detections so far , for the majority of the whim at logt@xmath16 , are either highly controversial ( e.g. nicastro et al .
, 2005a , b ) , or single - line and low statistical significance ( e.g. mathur , weinberg & chen , 2003 , nicastro , 2010 , zappacosta , 2012 ) .
perhaps the only exception that seemed to have gathered the largest consensus ( e.g. tananbaum et al . ,
2014 ) despite the large - compared to typical whim expectations - ew and associated ovii column density reported by the authors ( ew=@xmath17 m ) , is the proposed detection of a single - line ( ovii k@xmath1 ) whim filament at the redshift of the sculptor wall , reported by buote et al .
( 2009 : hereinafter b09 ) and fang et al .
( 2010 : hereinafter f10 ) along the line of sight to the blazar h 2356 - 309 .
recently , the same authors reported yet another evidence for a new ovii k@xmath1 whim absorber , again at @xmath18 , along the line of sight to the blazar mkn 501 ( ren , fang & buote , 2014 : hereinafter rfb14 ) .
indeed , the ovii k@xmath1 resonant line shifts to @xmath19 at @xmath20 , a redshift consistent with both the @xmath21 and @xmath22 intervals at which the sculptor wall and the hercules supercluster cross the lines of sight to h 2356 - 309 and mkn 501 , respectively . for this reason
b09 , f10 and rfb14 , identified the lines detected in the letg and rgs spectra of these two blazars at @xmath23 ( the average and maximum semi - dispersion of the two measurements in the letg and rgs spectra of h 2356 - 309 , respectively : f10 ) and @xmath24 ( rfb14 ) , as redshifted ovii k@xmath1 imprinted by two whim filaments permeating the two large scale structures of galaxies , the sculptor wall and the hercules superstructures . in a companion paper ( nicastro et al .
, 2015 : hereinafter n15 ) , we present a systematic study of the cold and mildly ionized interstellar ( ism ) and circum - galactic ( cgm ) medium of our galaxy through the modeling of the oi k@xmath1 ( @xmath25 ) , oii k@xmath1 ( @xmath26 ) and oii k@xmath3 ( @xmath27 ) absorption lines imprinted by our galaxy s ism / cgm in the spectra of two distinct samples of galactic and extragalactic sources . in particular , the oii k@xmath3 transition , firstly identified by gatuzz et al . , ( 2013a , b ) , is the weak ( oscillator strength @xmath28 , behar , private communication ) @xmath29 inner shell transition of the n - like ( 7 electrons ) ion of oxygen .
this line is hinted at in 9 of the 20 galactic x - ray binary ( xrb ) spectra and 8 out of the 29 agn spectra of the n15 sample , and has an average rest - frame wavelength of @xmath27 .
the 8 agns of the n15 sample , in whose spectra the oii k@xmath3 absorption line is hinted at , include the two blazars h 2356 - 309 ( @xmath30 , jones et al . , 2009 ) and mkn 501 ( @xmath31 , falco et al . , 2000 ) ,
for which the same line had been instead identified as intervening @xmath0 ovii k@xmath1 absorption tracing whim filaments ( b09 , f10 and rfb14 ) . here
we re - analyze all the available high resolution x - ray spectra of the two sightlines towards h 2356 - 309 and mkn 501 and demonstrate that , even for these two sightlines , the most likely identification of the @xmath32 line is indeed that of a @xmath2 oii k@xmath3 transition imprinted by a large amount of low - ionization metal medium ( limm ) that permeates the halo of our galaxy , at large distances from the galactic plane and perhaps up to the galaxy s cgm ( n15 ) .
this , at least for the line of sight to h 2356 - 309 ( the only one for which both the k@xmath1 and k@xmath3 transitions of oii are detectables ) does not completely rule out a possible contribution , at exactly the same wavelengths , by redshifted ovii k@xmath1 , but dramatically limits it to an ovii column density more than ten times lower than that claimed by b09 and f10 .
we reduced and analyzed archival hrc - letg and rgs data of the two blazars h 2356 - 309 and mkn 501 .
the data of h 2356 - 309 include 11 hrc - letg observations and 9 xmm - rgs observations , while mkn 501 has only 6 xmm - rgs observations .
all data were reduced with the latest versions of the xmm-_newton _ and _ chandra _ data reduction and analysis softwares ( `` science analysis system '' , gabriel et al . , 2004 - sas - v. 13.5.0 , and `` chandra interactive analysis of observation '' , fruscione & siemiginowska , 1999 - ciao - v. 4.6.1 ) and calibrations ( automatically set according to the given observation , for xmm-_newton _ data , and caldb v. 4.6.2 for _ chandra _ ) .
we followed the appropriate xmm-_newton _ and _ chandra _ data reduction / analysis threads and documentation to extract `` cleaned '' ( excluding periods of high background ) letg and rgs spectra and responses for all observations and co - added their spectra to maximize the snre ( * we also checked , a - posteriori , that all the results of our fitting procedures to the co - added spectra of h 2356 - 309 and mkn 501 where confirmed by simultaneous fitting of the individual spectra of the two sources * ) .
the final spectra of our two targets total 193 ks ( mkn 501 , xmm - rgs ) , 649 ks ( h 2356 - 309 , xmm - rgs ) and 466 ks ( h 2356 - 309 , hrc - letg )
. we used the fitting package _ sherpa _
, in ciao ( freeman , doe & siemignowska , 1999 ) to perform spectral fitting of the rgs and letg spectra of our targets , in the 6 - 30 ( rgs ) and 6 - 50 ( letg ) intervals , with the main aim to disentangle local ( i.e. @xmath2 ) absorption from our own galaxy s ism / cgm , from intervening absorption from putative whim filaments along the lines of sight to our two targets . following the procedure described in n15 , we first modeled the continuum of our targets with a power - law ( model _ xspowerlaw _ in _ sherpa _ ) attenuated by neutral absorption ( the xspec - native model _ xstbabs _ in _ sherpa _ : wilms , allen & mccray , 2000 ) . the best - fitting pure - continuum statistics for our three spectra are @xmath33 ( letg spectrum of h 2356 - 309 ) , @xmath34 ( rgs1+rgs2 spectra of h 2356 - 309 ) and @xmath35 ( rgs1+rgs2 spectra of mkn 501 ) .
figure [ res ] shows the 21.5 - 24 portions of the best - fitting pure - continuum model residuals ( in standard deviations ) of the letg and rgs spectra of h 2356 - 309 and mkn 501 . in the rgs panels , the green segments mark the spectral intervals where the presence of rgs instrumental features ( bad pixels ) hamper the search for unresolved absorption lines .
once a reasonable description of the broad - band continuum was obtained , we then proceded to search for , and model , the possible presence of unresolved ( fwhm@xmath36 km s@xmath37 and fwhm@xmath38 km s@xmath37 at @xmath39 , for the rgs and the letg , respectively ) absorption lines in the 18 - 24 band . to first identify candidate unresolved absorption lines and then confirm their identification in this spectral range we followed the procedure described in nicastro et al .
this procedure yielded the results summarized in table 1 ( line ids are also labeled in figure [ res ] ) .
for all the lines , we report only statistical errors at the the 1@xmath40 significance level .
@xmath0 i d + & width & & + in & in m & & + + @xmath41 & @xmath42 & oi k@xmath1 & oiv k@xmath1 + @xmath43 & @xmath44 & oii k@xmath1 & n / a + @xmath45 & @xmath46 & oiv k@xmath1 & ovi k@xmath1 + @xmath47 & @xmath48 & oii k@xmath3 & ovii k@xmath1 + 22.02 ( frozen ) & @xmath49 ( at 3@xmath40 ) & ovi k@xmath1 & n / a + @xmath50 & @xmath51 & ovii k@xmath1 & n / a + @xmath5219.21 ( frozen ) & @xmath53 ( at 3@xmath40 ) & n / a & ovii k@xmath3 + + @xmath54 & @xmath55 & oi k@xmath1 & oiv k@xmath1 + @xmath56 & @xmath57 & oii k@xmath3 & ovii k@xmath1 + @xmath58 & @xmath59 & ovi k@xmath1 & n / a + @xmath60 & @xmath61 & ovii k@xmath1 n / a + @xmath5219.25 ( frozen ) & @xmath62 ( at 3@xmath40 ) & n / a & ovii k@xmath3 + + @xmath63 & @xmath64 & oi k@xmath1 & oiv k@xmath1 + @xmath65 & @xmath61 & oii k@xmath3 & ovii k@xmath1 + @xmath66 & @xmath67 & ovi k@xmath1 & n / a + @xmath68 & @xmath69 & ovii k@xmath1 n / a + @xmath5219.26 ( frozen ) & @xmath70 ( at 3@xmath40 ) & n / a & ovii k@xmath3 + + @xmath52 for each spectrum , ovii k@xmath3 3@xmath40 upper limits are given at the redshift of the putative ovii k@xmath1 whim line , as derived from the best - fitting position of the line at @xmath71
. the 18 - 24 letg spectrum of h 2356 - 309 shows the presence of 5 unresolved absorption lines , which can all be identified as @xmath2 local transitions imprinted by a wide ionization range of oxygen ions in our galaxy s multi - phase ism . in particular , the letg data of h 2356 - 309 hint at the oi k@xmath1 ( 2.5@xmath40 ) , oii k@xmath1 ( 2.3@xmath40 ) and k@xmath3 ( 2@xmath40 ) , oiv k@xmath1 ( 2.9@xmath40 ) and ovii k@xmath1 ( 2.3@xmath40 ) transitions ( fig .
[ res ] , top panel ) .
the rgs data of h 2356 - 309 confirm the presence of the local oi k@xmath1 , oii k@xmath3 and ovii k@xmath1 absorption lines ( detected at statistical significances of 3.4@xmath40 , 2.8@xmath40 and 2.8@xmath40 , respectively ) and hint at the presence of a ovi k@xmath1 line ( only 1.6@xmath40 ) , but can not * safely * detect either the oii k@xmath1 or the oiv k@xmath1 lines , due to the presence of instrumental features at the relevant wavelengths ( fig .
[ res ] , middle panel , green horizontal segments ) . .
two of these lines , the @xmath2 ovii k@xmath1 and the line at @xmath72 ( letg ) or @xmath56 ( rgs ) ( table 1 ) were also reported by b09 and f10 ( who use the same letg data that we use here ) , who identified the second of these lines as an ovii k@xmath1 absorber at @xmath73 imprinted by an intervening whim filament at a redshift consistent with that where the sculptor wall concentration of galaxies crosses our line of sight to h 2356 - 309 .
the line at @xmath74 detected in the letg spectrum ( not detectable in the rgs spectrum because of an overlapping instrumental feature ) and identifiable with local oiv k@xmath1 absorption , was not reported instead by f15 , though the line is clearly visible in the 20 - 23.5 portion of the spectrum that fn15 show in their fig .
4 . the 18 - 24 letg spectrum of mkn 501 hints at the presence of the same 4 unresolved absorption lines that are detected in the rgs spectrum of h 2356 - 309 , and that can be identified as @xmath2 local transitions imprinted by oi k@xmath1 ( 4.2@xmath40 ) , oii k@xmath3 ( 2.8@xmath40 ) , ovi k@xmath1 ( 2.2@xmath40 ) and ovii k@xmath1 ( 3.2@xmath40 ) transitions ( fig . [ res ] , bottom panel ) . as for the rgs spectrum of h 2356 - 309
, the rgs spectrum of mkn 501 can not detect either the local oii k@xmath1 or oiv k@xmath1 lines , due to the presence of instrumental features at the relevant wavelengths ( fig .
[ res ] , bottom panel , green horizontal segments ) . also in the case of the rgs spectrum of mkn 501 , the @xmath2 ovii k@xmath1 line and the line at @xmath75 ( rgs ) ( table 1 ) were already reported by rfb14 ( using the same rgs data that we use here ) , who , again , identified the second of these lines as an ovii k@xmath1 absorber at @xmath76 ( consistent with the source systemic redshift ) imprinted by an intervening whim filament at a redshift consistent with that where the hercules supercluster concentration of galaxies crosses our line of sight to mkn 501 .
from the possible identifications listed in columns 3 and 4 of table 1 , it is evident that 3 of the 6 different lines detected ( or hinted at ) in the letg and rgs spectra of h 2356 - 309 and mkn 501 have unambiguous identifications .
these are , the @xmath2 ovii k@xmath1 ( letg and rgs ) , ovi k@xmath1 ( rgs only in h 2356 - 309 , where also letg is available , but consistent with the 30 m 3@xmath40 letg ew upper limit at that position ) and oii k@xmath1 ( letg only , because the rgs spectrum is blocked by an instrumental feature at those wavelengths ) , and can only be imprinted by a multiphase ism / cgm , with one or more high - ionization component producing ovi and ovii absorption and a low - ionization component producing oii ( and oi ) absorption . however , the 3 remaining lines have ambiguous identification
. they could all be imprinted by a multi - phase mildly ionized medium in the galaxy s ism / cgm ( with dominant oi - iv ions ) , but could also be identifiable with @xmath0 transitions imprinted by hotter medium , dominated by ovi - vii ions .
this is because the ovii k@xmath1 , ovi k@xmath1 and oiv k@xmath1 transitions , at @xmath0 overlap with the rest frame wavelengths of the oii k@xmath3 , oiv k@xmath1 and oi k@xmath1 transitions , respectively , expected to be imprinted by the galaxy s low - ionization and medium - ionization metal mediums ( limm - e.g. n15 and references therein - and mimm ) . to discriminate between these two possibilities for these 3 absorption lines ,
we make use of two different versions of our photoionized absorber spectral engine ( phase , krongold et al . ,
2003 ) code : one photo - ionized , for the ism / cgm of our galaxy ( _ galabs _ model hereinafter : see n15 for details ) , needed to model at least the three lines with unambiguous ism / cgm identification , and one hybridly - ionized and optimized for the whim ( _ whimabs _ model , hereinafter : see , e.g. , zappacosta et al . , 2010 ) , to attempt to model at least part of the three controversial and ambiguous lines .
the letg and rgs data of h 2356 - 309 have snre=13 ( letg ) and 26 ( rgs ) at 22 , much larger , when combined , than the snre=24 of the rgs spectrum of mkn 501 .
moreover , the combined letg and rgs spectrum of h 2356 - 309 , detects 6 different lines ( 5 in the letg and 3 in the rgs , 2 of which in common with the letg ) , compared to the 4 lines detected in the lower s / n rgs spectrum of mkn 501 .
h 2356 - 309 is therefore the best suited target to test our physically self - consistent ism / cgm versus whim models . to verify whether all the 6 different lines detected ( or hint at ) in the letg and rgs spectra of h 2356 - 309 could be modeled by absorption due to our own galaxy multi - phase ism / cgm
, we followed the procedure we used in n15 .
namely , we model simultaneously the 6 - 50 letg and the 6 - 30 rgs1 and rgs2 data of h 2356 - 309 with a continuum model consisting of a powerlaw ( native xspec model _ xspowerlaw _ in _ sherpa _ ) attenuated by the column of neutral galactic gas ( xspec native model _ xstbabs _ in _ sherpa _ , and add to each of the three spectra ( letg , rgs1 and rgs2 ) a number of _ galabs _ components till all the lines detected at single - line statistical significance @xmath77 could be adequately modeled . for each
_ galabs _ component we link all parameters to a single variable value in the 3 spectra , except the redshift that is allowed to vary by @xmath78 km s@xmath37 from spectrum to spectrum , to allow for line misalignment within instrument resolution elements .
this procedure required three different _ galabs _ components : ( a ) a low - ionization component , with a temperature of t@xmath79 k and a * poorly constrained * doppler parameter @xmath80 km s@xmath37 ( the limm : dubbed warm ionized metal medium in n15 ) , that models well the lines at @xmath81 and 23.35 , identified as oi k@xmath1 ( best - fitting limm ew@xmath82 m versus best - fitting gaussian @xmath83ew@xmath84 ) and oii k@xmath1 ( ew@xmath85 m versus ew@xmath86 m ) transitions , but is able to reproduce only half of the @xmath87 line , as oii k@xmath3 ( ew@xmath88 m versus @xmath83ew@xmath89 m ) ; ( b ) a warmer mildly ionized ism / cgm component with t@xmath90 k and b@xmath91 km s@xmath37 ( the mimm ) , modeling less than half of the @xmath92 line with the oiv k@xmath1 transitions ( ew@xmath93 m versus ew@xmath94 m ) ; ( c ) a hot ism / cgm component , with t@xmath95 k and b@xmath91 km s@xmath37 ( hereinafter high - ionization metal medium : himm ) , that models about 2/3 of the @xmath96 line , identified as ovii k@xmath1 ( ew@xmath97 m versus @xmath83ew@xmath98 m ) , but fails to model the ovi k@xmath1 line hinted at a single - line statistical significance of only @xmath99 in the rgs1 spectrum .
[ bfmodel ] shows the 21 - 24 portions of the letg spectrum of h 2356 - 309 , with superimposed the best - fitting ism / cgm model ( _ model - b _ ) including the limm , the mimm and the himm ( top panel ) .
the ews of the 5 lines modeled by the three _ galabs _ components , are consistent , within their 1@xmath40 uncertainties , with the best - fitting ews obtained with _ model - a _ and listed in table 1 , but the best - fitting values of the oii k@xmath3 , oiv k@xmath1 and ovii k@xmath1 are all lower than the those obtained with _ model - a _ ( single - line gaussian fit ) .
this could simply reflect the fact that the photo - ionization - equilibrium models that we use are not a sufficiently accurate description of the gas physics , but it might also indicate that additional components are needed .
for example , for the unmodeled ovi k@xmath1 line at @xmath100 and the not fully modeled @xmath2 ovii k@xmath1 , a fourth , ism / cgm phase could be present , with a degree of ionization in between that of the mimm and the himm , while for the partly unmodeled @xmath2 oii k@xmath3 oiv k@xmath1 an extragalactic intervening whim component may be needed , with temperature such to be dominated by ovi and ovii ions .
we conclude that the 3-phase ism / cgm _ model - b _ reproduces in a statistically satisfactory way the combined letg and rgs data of h 2356 - 309 , but leaves some room for ( statistically not required : @xmath101 for 5332 dof ) intervening @xmath0 whim absorption , filling in the 1/3 missing line opacity at the wavelengths of the @xmath2 oii k@xmath3 and the half missing opacity at the wavelengths of the @xmath2 oiv k@xmath1 , with the @xmath0 ovii k@xmath1 and ovi k@xmath1 transitions , respectively . to verify the need for an additional whim component
, we added a _
whimabs_ hybrid - ionization component to our best - fit ism / cgm 3-phase _ model - b _ , and re - fitted the data leaving the temperature , equivalent hydrogen column density , and doppler parameter of the gas , free to vary and constraining its redshift to vary only between @xmath102 ( _ model - c _ ) .
this yielded a negligible improve in statistics ( only @xmath103 for 3 additional degree of freedom over the initial 115 , in the 21 - 24 spectral interval ) , a best - fitting redshift of the putative whim filament of @xmath104 , a temperature t@xmath105 k , a doppler parameter @xmath106 km s@xmath37 and an oxygen column density of n@xmath107 @xmath7 , in much better agreement with theoretical expectations for the whim than the implausible large oxygen column reported by f15 ( n@xmath108 @xmath7 for their best fitting @xmath109 km s@xmath37 ) .
the _ whimabs _ component fills in the missing opacities at the wavelengths of the @xmath2 oii k@xmath3 and oiv k@xmath1 transitions , with @xmath20 ovii k@xmath1 and ovi k@xmath1 absorption , and the best - fitting _ model - c _ molds well the 5 absorption lines detected in the letg+rgs spectrum of h 2356 - 309 ( fig .
[ bfmodel ] , bottom panel ) .
table 2 lists , for these 5 lines , the best - fitting _ model - c _ ews , together with the corresponding best - fitting _ model - a _ ews already reported in table 1 .
.*comparison between _ model - c _ and _ model - a _ absorption line ews , for the h 2356 - 309 sightline . * [ cols="^,^,^,^ " , ] oii absorption is ubiquitously detected through its strongest k@xmath1 transition , along virtually all galactic and extragalactic lines of sight of the n15 samples .
the 8 times weaker k@xmath3 transition is hinted at in 9 out of the 20 lines of sight and 9 out of 21 extragalactic lines of sight of the n15 samples .
this absorption line is detected with average equivalent widths @xmath110 m ( down to a 3@xmath40 detectability threshold of @xmath111 m ) and @xmath112 m ( down to a looser - because of the lower s / n of the agn spectra - 3@xmath40 detectability threshold of 80 m ) , for the galactic and extragalactic sources , respectively ( n15 ) .
the n15 sample of extragalactic sources includes the two blazars h 2356 - 309 and mkn 501 , whose letg and rgs data we also reanalyze here . in our re - analysis of all the high - resolution x - ray spectra available for these two targets , for the absorption line at @xmath113 , we measure @xmath114 and @xmath61 m , for h 2356 - 309 and mkn 501 , respectively , fully consistent with the average ew@xmath115 measured in our samples of galactic and/or extragalactic lines of sight in n15 . based on this comparison ,
we conclude that the @xmath113 absorption line observed in the spectra of h 2356 - 309 and mkn 501 is likely to be produced entirely by oii k@xmath3 absorption .
in this work we reanalyze all the available high - resolution x - ray spectra of the two blazars h 2356 - 309 and mkn 501 , for which claims of intervening @xmath0 whim absorptions were presented by b09 , f10 and rfb14 .
we demonstrate that the presence of metal absorption from the diffuse limm in the galaxy s ism / cgm , casts serious doubts on these two ovii k@xmath1 whim detection claims ( b09 , f10 , rfb14 ) .
in particular , we show that the most likely identification for this putative @xmath116 ovii k@xmath1 whim line , is instead that of the k@xmath3 transition of oii , imprinted by the galaxy s limm ( n15 ) . for the case of h 2356 - 309 , for which we dispose of combined _ chandra _ letg and xmm-_newton _ rgs data that allow us to detect 5 absorption lines between 21 - 24 , at single - line significance @xmath117
, we conclude that any redshifted ( @xmath20 ) ovii k@xmath1 whim contribution to the unavoidable presence of an oii k@xmath3 absorber at @xmath118 , must be lower than n@xmath119 @xmath7 ( for @xmath120 km s@xmath37 ) , in much better agreement with theoretical expectations for the whim , than the implausible large oxygen column reported by f15 ( n@xmath108 @xmath7 for their best fitting @xmath121 km s@xmath37 ) .
we thank the anonymous referee for the useful comments that helped improving the paper .
fn and fs acknowledge support from inaf - prin grant 1.05.01.98.10 .
buote , d.a .
et al . , 2009 ,
apj , 695 , 1351 : b09 cen , r. & ostriker , j.p . , 2006 , 650 , 560 fang , t. et al . , 2010 , apj , 714 , 1715 : f10 falco , e. et al . , 2000 , `` the updated zwicky catalog ( uzc ) '' , vol 1 . , p. 1
freeman , p. , doe s. & siemiginowska a. , 1999 , spie , 4477 , 76 fruscione a. & siemiginowska a. , 1999 stin , 9906596 gabriel c. et al . , 2004 ,
aspc , 314 , 759 gatuzz , e. et al . , 2013 ,
apj , 768 , 60 ( 2013a ) gatuzz , e. et al . , 2013 ,
apj , 778 , 83 ( 2013b ) gupta , a. et al . , 2012 ,
apj , 756 , l8 jones , d.h .
et al . , 2009 , `` the 6df galaxy survey data release 3 '' , vol . 1 , p. 1
krongold , y. et al . , 2003 , apj , 597 , 832 mathur , s. , weinberg , d.h .
& chen , x. , 2003 , apj , 582 , 82 nicastro , f. et al . ,
2015 , mnras , submitted : n15 nicastro , f. et al .
, 2005 , nature , 433 , 495 ( 2005a ) nicastro et al .
, 2005 , apj , 629 , 700 ( 2005b ) nicastro , f. et al . , 2010 ,
apj , 715 , 854 ren , b. , fang , t. & buote , d.a . , 2014 ,
apj , 782 , l6 : rfb14 tananbaum , h. et al . , 2014 ,
rpph , volume 77 , issue 6 , article i d .
066902 : rxiv:1405.7847 wilms , j. , allen , a. , & mccray , r. , 2000 , apj , 542 , 914 zappacosta l. , nicastro , f. , krongold , y. & maiolino , r. , 2012 , apj , 753 , 137 zappacosta , l. et al .
, 2010 , apj , 717 , 74 | in this letter we demonstrate that the two claims of @xmath0 ovii k@xmath1 absorption lines from warm hot intergalactic medium ( whim ) along the lines of sight to the blazars h 2356 - 309 ( buote et al . , 2009 ; fang et al . , 2010 ) and mkn 501 ( ren , fang & buote , 2014 ) are likely misidentifications of the @xmath2 oii k@xmath3 line produced by a diffuse low - ionization metal medium in the galaxy s interstellar and circum - galactic mediums .
we perform detailed modeling of all the available high signal - to - noise chandra letg and xmm - newton rgs spectra of h 2356 - 309 and mkn 501 and demonstrate that the @xmath0 whim absorption along these two sightlines is statistically not required .
our results , however , do not rule out a small contribution from the @xmath0 ovii k@xmath1 absorber along the line of sight to h 2356 - 309 . in our model
the temperature of the putative @xmath4 whim filament is t@xmath5 k and the ovii column density is n@xmath6 @xmath7 , twenty times smaller than the ovii column density previously reported , and now more consistent with the expectations from cosmological hydrodynamical simulations .
absorption lines , ism , galaxy , whim |
the advantage of charged particles , in particular , protons and carbon nuclei , used for radiation therapy of cancer consists in elevated dose delivered at the end of projectile range in tissues .
the plateau of the depth - dose distribution at the entrance of a monoenergetic beam terminates with a sharp bragg peak which can be targeted at the tumor .
such a dose profile helps to spare healthy tissues located in front of the tumor as well as beyond the projectile range . since a set of beam energies
is typically used in treatments to cover the whole tumor volume , the resulting dose distribution is characterized by a spread - out bragg peak ( sobp ) @xcite with a wide domain of elevated dose . the damage of healthy tissues during therapy can be essentially reduced if the ratio between the relative biological effectiveness ( rbe ) values at the sobp region and plateau is favorable . as recognized almost 40 years ago in radiobiological experiments with sobp beams of light nuclei performed at berkeley @xcite , this rbe ratio is greater than 1 and increases with ion charge up to carbon .
it was also found that this ratio decreases for ne and becomes less than 1 . in 1994
first patient treatments with beams of carbon nuclei started in japan at national institute of radiological sciences ( nirs ) @xcite and in 1997 in germany at gesellschaft fr schwerionenforschung ( gsi ) @xcite .
later the advantages of @xmath2c with respect to @xmath4he and @xmath5ne were confirmed in experiments at nirs with these nuclear beams @xcite . as explained @xcite , the rbe of @xmath2c beam with the linear energy transfer ( let ) of was found similar to the rbe of neutrons which have been used for treatment at nirs during last 20 years .
this similarity also motivated the choice of carbon nuclei for treatments at nirs . in the last decades
localized tumors have been successfully treated with beams of carbon nuclei at several facilities constructed in japan and germany @xcite . despite of the broad clinical experience
collected worldwide with proton and carbon - ion beams , other light nuclei can be also considered as future therapy options .
the radiobiological properties of proton and @xmath2c beams were compared in several studies , see e.g. @xcite . however
, less attention has been paid so far to @xmath0he or @xmath1o and to their comparison with protons and @xmath2c . there
exist several clinical rationale behind the use of @xmath6he and @xmath1o for therapy : * @xmath6he and @xmath1o beams have a reduced lateral spread of the dose distribution compared to protons ; * their rbe in the target volume is higher compared to protons ; * lower dose in the tail region and lower rbe in the plateau is expected for @xmath6he compared to @xmath2c due to reduced nuclear fragmentation of @xmath6he ; * @xmath1o is a promising option for hypoxic tumors as it provides a higher dose averaged let in the target volume compared to @xmath2c .
the choice of ion species and their energies at each new particle therapy facility @xcite essentially depends on let and rbe of the projectiles under consideration .
the distributions of dose , let and dose averaged let of @xmath3h , @xmath0he , @xmath7li , @xmath8be , @xmath9b , @xmath2c , @xmath10n and @xmath1o nuclei of therapeutic energies were studied @xcite by means of monte carlo simulations with shield - hit and fluka codes , respectively .
similar distributions for @xmath3h , @xmath4he , @xmath2c , @xmath5ne and @xmath11ni were calculated @xcite with the geant4 toolkit to study their dependence on the ions mass , charge and energy .
it was assumed @xcite that ions with let above should be used for efficient cancer therapy as such species induce on average two and more double strand breaks in dna close to each other .
however , as pointed out in the same work , this limit is not sharp and ought to vary with ion mass and charge .
this indicates that the rbe of respective ions has to be additionally considered for an accurate comparison of their biological action .
indeed , as shown in our recent work @xcite , there is no direct correspondence between rbe and the frequency - mean linear energy @xmath12 , which represents let , for monoenergetic beams of therapeutic energies .
similar rbe values were estimated @xcite at the peak and plateau regions which are characterized , however , by very different @xmath12 . as demonstrated @xcite , microdosimetry spectra for monoenergetic @xmath3h , @xmath0he , @xmath13li and @xmath2c nuclei propagating in a water phantom
can be accurately described with our monte carlo model for heavy - ion therapy ( mchit ) @xcite , and this model coupled with the microdosimetric kinetic ( mk ) model @xcite can be used to calculate the respective rbe profiles .
in the present work we evaluate @xmath0he and @xmath1o for cancer therapy as complementary options to @xmath3h and @xmath2c by considering the biological dose distribution with a sobp delivered by these four projectiles .
this is an important prerequisite for planning radiobiological experiments with @xmath0he and @xmath1o sobp beams and extending existing treatment planning systems to operation with @xmath0he and @xmath1o .
the experimental data @xcite collected at himac for the sobp obtained by the moderation of a @xmath2c beam are used as a reference .
the results of microdosimetry simulations are validated by comparison with microdosimetry data collected for @xmath3h , @xmath0he and @xmath2c beams @xcite .
this makes possible to predict the rbe and cell survival profiles for @xmath1o beams and compare all four ion species in a common framework .
with the monte carlo model for heavy ion therapy ( mchit ) @xcite the propagation of accelerated protons and light nuclei of therapeutic energies in tissue - like media can be simulated .
the model is based on the geant4 toolkit @xcite and takes into account all major physics processes relevant to the interactions of beam particles with these materials . the geant4 version 9.5 with patch 02 is used to build the present version of mchit , which also simulates the interactions of various particles with walled and wall - less tissue equivalent proportional counters ( tepc ) thus providing respective microdosimetry distributions .
the ionization of atoms and multiple coulomb scattering on nuclei of the media are the most important electromagnetic processes to simulate the energy loss and straggling of primary and secondary charged particles .
two predefined physics lists for electromagnetic processes are employed in mchit , namely , g4emstd ( which uses the `` standard electromagnetic physics option 3 '' ) and g4empen ( which uses the penelope models for low - energy processes ) .
the low - energy thresholds for production of @xmath14-electrons are 990 ev for g4emstd and 100 ev for g4empen .
however , in order to reduce the cpu time , cuts in range and energy thresholds for particle production are set differently for water , plastic of the tepc wall and the tepc sensitive volume filled with tissue - equivalent gas @xcite .
a customized physics list , g4empen+iongas , which is based on g4empen and the models describing the ionization of gas media by ions , can be also used in calculations . as demonstrated @xcite , g4emstd and g4empen
provide statistically equivalent results for microdosimetry spectra , which agree well with the distributions measured for @xmath6he @xcite , excluding the domains of low linear energy , @xmath15 , and around the maximum at @xmath16 .
however , the agreement between the data and calculations is improved when g4empen+iongas is used .
therefore , g4empen+iongas is involved in mchit also in the present work to simulate electromagnetic processes inside the tepc volume .
a therapeutic nuclear beam is attenuated in tissues due to the loss of beam nuclei in nuclear fragmentation reactions , which generate secondary projectile and target fragments @xcite .
nuclear reactions are taken into account in mchit to reproduce this effect .
as shown @xcite , the build - up of secondary fragments produced by and @xmath2c beams is generally well described with a customized physics list based on the light ion binary cascade model ( g4bic ) @xcite coupled with the fermi break - up model ( g4fermibreakup ) @xcite responsible for subsequent decays of excited nuclear fragments created at the first fast stage of nucleus - nucleus collisions . in the present work g4bic
is used for proton , helium and lithium beams , while the quantum molecular dynamics model ( g4qmd ) @xcite is involved in simulations with carbon and oxygen beams .
more details on the physics processes and respective geant4 models involved in modeling with mchit are given in our recent publications @xcite .
the design and materials of specific tepc models were thoroughly introduced in mchit .
this made possible to simulate the microdosimetry spectra measured with a walled tepc at several positions inside a water phantom irradiated by @xmath13li and @xmath2c beams @xcite and study the impact of nuclear fragmentation reactions on these spectra @xcite .
the influence of the positioning of the tepc with respect to the beam axis and the distortion of the spectra due to the pile - up of individual events were investigated . after correcting for such effects
the calculated microdosimetry spectra agree well with the experimental data in general . following the validation of mchit for microdosimetry of monoenergetic @xmath13li and @xmath2c beams @xcite
, this model is applied to microdosimetry of sobp dose distributions considered in the present work . in the measurements performed at himac with a sobp for @xmath3h , @xmath0he and @xmath2c @xcite
the data for frequency - mean lineal energy , @xmath12 , dose - mean lineal energy , @xmath17 , and saturation - corrected dose - mean lineal energy , @xmath18 @xcite were collected .
the microdosimetry spectra were measured with a walled tepc corresponding to a tissue - equivalent sphere of in diameter . according to the linear - quadratic ( lq ) model the fraction of cells @xmath19 survived after the impact of the radiation dose @xmath20 is calculated as @xmath21 } \ .
\label{eq : slq}\ ] ] following the modified mk model @xcite applied to human salivary gland ( hsg ) tumor cells the parameter @xmath22 is estimated as @xmath23 with the constant term @xmath24 = 0.13 gy@xmath25 representing the initial slope of the survival fraction curve in the limit of zero let and @xmath26 = 0.05 gy@xmath27 , @xmath281 g/@xmath29 as the density of tissue and @xmath300.42 @xmath31 m as the radius of a sub - cellular domain in the mk model .
the dependence of the @xmath22-parameter of the linear - quadratic model on @xmath18 rather than on @xmath17 reflects the reduction of the rbe known as the saturation effect .
it means that an excessive local energy deposition does not boost biological effects induced by high - let particles @xcite .
as demonstrated @xcite , the same value of parameter @xmath26 = 0.05 gy@xmath27 can be used to fit the data on @xmath19 with eq .
( [ eq : slq ] ) for x - rays and ions .
this justifies the assumption of the mk model that @xmath26 is independent of let . according to the lq model the rbe@xmath32 for 10% survival of hsg cells
is calculated using the following relation @xcite : @xmath33 where @xmath34 is the 10% survival dose of ions and @xmath35 = 5.0 gy is the 10% survival dose of the reference radiation ( 200 kvp x - rays ) for hsg cells @xcite .
finally , the biological dose @xmath36 is calculated from rbe@xmath32 and physical dose : @xmath37 the computing time required for treatment planing in carbon - ion therapy can be reduced by using pre - computed libraries of dose and rbe distributions for monoenergetic beams @xcite .
the aim of the treatment planning is to find an optimum superposition of many beams with their individual energy , position and intensity in order to obtain the prescribed biological sobp dose profile .
it is expected that a similar approach will be also suitable for other therapeutic beams , like @xmath0he and @xmath1o .
therefore , we implemented a common algorithm to calculate the relative weights of pre - defined monoenergetic beams of @xmath3h , @xmath0he , @xmath2c and @xmath1o to obtain flat biological sobp distributions for each projectile as a product of the physical dose and rbe calculated for mixed radiation field .
a library of depth - dose profiles and the corresponding microdosimetry spectra for different beam energies and nuclei were calculated by monte carlo simulations with mchit .
they are used as input data for a procedure similar to one implemented at nirs @xcite . according to this procedure based on the theory of dual radiation action @xcite the survival fraction of cells exposed to mixed radiation
is calculated as : @xmath38 @xmath39 @xmath40 here @xmath41 is the weight coefficient ( fraction ) of the local physical dose of the @xmath42th monoenergetic beam which contribute to the total physical dose @xmath20 , while @xmath43 and @xmath44 are the parameters of the lq model specific to @xmath42th monoenergetic beam . the parameters @xmath43 and @xmath44 are calculated along the beam axis using mchit coupled with the modified mk model as described in sec .
[ sec : monoenergetic ] .
the resulting rbe@xmath45 for the mixed radiation is calculated from the survival fraction of cells @xmath46 and it also depends on the depth .
rbe@xmath32 and rbe@xmath45 for the monoenergetic and sobp ( mixed ) beams , respectively , are considered in the following .
they account for the relative biological effectiveness corresponding to 10% survival of cells .
a dedicated algorithm to obtain @xmath41 for a given biological sobp was developed .
it starts with the determination of the weight at the distal edge of the sobp distribution and then calculates weights for less energetic beams by adjusting their contribution to provide a flat sobp plateau .
simulation results for monoenergetic @xmath3h , @xmath0he , @xmath2c and @xmath1o beams are shown in figure [ fig : fig1 ] .
the beam energies were chosen to place the bragg peaks of all four beams at the depth of @xmath47 . for hsg cells ( middle panel ) and biological dose ( bottom panel ) for @xmath3h , @xmath0he , @xmath2c and @xmath1o beams in water .
the rbe@xmath32 calculated from measured values of @xmath18 @xcite are shown by circles .
the rbe@xmath32 estimated from lq fitting on survival curves of hsg cells @xcite are presented by triangles .
the biological dose distributions for all ions were rescaled to get the same value at the maximum . ]
as expected , the @xmath2c and @xmath1o beams deposit much higher energy at the bragg peak compared to @xmath3h and @xmath0he .
however , higher energy deposition by @xmath2c and @xmath1o is also observed at the entrance and tail region .
therefore , no clear advantages of therapeutic @xmath2c and @xmath1o beams with respect to @xmath3h and @xmath0he beams regarding healthy tissues can be inferred exclusively from the analysis of the considered depth - dose distributions .
the rbe@xmath32 profiles for the considered @xmath3h , @xmath0he , @xmath2c and @xmath1o beams are shown in the middle panel of figure [ fig : fig1 ] and they differ from each other .
the rbe@xmath32 for @xmath2c beam estimated from the parameters of a lq fitting on survival curves of hsg cells with @xmath26 fixed to 0.05 gy@xmath27 @xcite is presented for comparison .
the rbe@xmath32 profile for @xmath2c beam calculated with mchit coupled with the mk model agrees well with the profile which is also calculated with mk , but using the measured microdosimetry data on @xmath18 @xcite .
a prominent difference between rbe@xmath32 profiles for @xmath2c and @xmath1o and the profile for @xmath0he is seen in the insert of figure [ fig : fig1 ] . the backward shift of the maximum of rbe@xmath32 of @xmath2c and @xmath1o with respect to the position of the @xmath0he maximum is due to the saturation effect .
it is also found that @xmath2c and @xmath1o nuclei are characterized by higher rbe values for 10% survival of hsg cells along the whole irradiated medium .
their maximum values reach 2.9 and 3.1 , respectively , close to the bragg peak . at the same time
the rbe@xmath32 values for helium are relatively low at the entrance and tail regions and demonstrate a steep rise to 2.2 at the bragg peak position .
before the bragg peak the rbe@xmath32 values for proton beam are slightly below 1 . and increase to 1.5 well after the distal edge of the proton bragg peak .
this rise of rbe@xmath32 for proton beam is explained by the presence of secondary nucleons produced by beam protons in water and propagating beyond the bragg peak .
finally , the biological dose profiles for @xmath3h , @xmath0he , @xmath2c and @xmath1o ( a.u . ) are presented in the bottom panel of figure [ fig : fig1 ] .
they were rescaled to get the same value at the maximum .
the profiles for @xmath0he , @xmath2c and @xmath1o are very similar to each other .
they are characterized by a more sharp rise and fall of the biological dose in the bragg peak region compared to protons .
after renormalization lower biological dose for @xmath0he , @xmath2c and @xmath1o is predicted at the entrance with respect to protons .
sobp profiles of biological dose for all these projectiles are considered below . as explained in sec .
[ sec : sobpmethod ] , a given biological sobp dose distribution is composed from a set of depth - dose and depth-@xmath18 profiles calculated with mchit for monoenergetic beams . such a library created in the present work contains pristine bragg curves with a increment of the bragg peak positions which are within 90 depth in water .
seven pristine bragg peaks for @xmath2c covering a domain in depth are shown in figure [ fig : fig2 ] to illustrate the content of this library .
the microdosimetry variables which are also stored in the library as functions of depth are used to estimate rbe@xmath32 profiles according to the mk model , see sec .
[ sec : sobpmethod ] .
the respective rbe@xmath32 distributions are shown in figure [ fig : fig2 ] for the same beams .
as seen in figure [ fig : fig2 ] , the height of the bragg peak noticeably diminishes with depth , while the maximum rbe@xmath32 remains almost constant ( @xmath48 ) over the considered depth range .
c in water in steps ( top panel ) and the corresponding rbe@xmath32 profiles for hsg cells calculated with mchit and the mk model ( bottom panel ) . ] a -wide sobp profile of biological dose for @xmath2c beam , which was built according to the above - described procedure is shown in the top panel of figure [ fig : fig3 ]
. here and in the following such profiles are normalized to 1 at the plateau in order to facilitate the comparison of various ion species .
the resulting distribution has a flat sobp plateau with negligible fluctuations due to the presence of individual bragg peaks .
in contrast , the corresponding sobp distribution of the physical dose , which is also shown in figure [ fig : fig3 ] is not flat , but rather decreases with depth .
the respective rbe@xmath32 amounts to @xmath49 at the proximal edge of the sobp , while it is slightly above 2.5 at the distal edge , see the bottom panel of figure [ fig : fig3 ] .
the insert in the bottom panel of figure [ fig : fig3 ] demonstrates the calculated relative weights for monoenergetic beams used to build the sobp distribution of biological dose shown in figure [ fig : fig3 ] .
c ion beam and the respective physical dose ( top panel ) .
rbe@xmath45 profile for hsg cells ( bottom panel ) and the distribution of beam energy used to build the sobp profiles ( insert ) .
] the -wide rbe@xmath45 profiles calculated for hsg cells for @xmath3h , @xmath0he , @xmath2c and @xmath1o beams are presented in figure [ fig : fig4 ] .
they were calculated within the mk model basing on microdosimetry data generated by monte carlo simulations with mchit .
the reliability of these profiles can be proved by comparing them with rbe@xmath45 calculated by two different approaches . in the first case
, rbe@xmath45 is also calculated within the mk model , but on the basis of measured @xmath18 values @xcite . in the second approach
, rbe@xmath45 is calculated using the parameters of lq fitting of survival curves of hsg cells with @xmath26 fixed to 0.05 gy@xmath27 @xcite .
the profiles based on mchit simulations agree very well with the rbe@xmath45 estimated on the basis of the two set of experimental data , see figure [ fig : fig4 ] . in order to make such comparison , experimental values @xcite corresponding to @xmath3h and @xmath0he
were shifted in depth due to the difference of beam energies used in measurements and simulations . a good agreement with data for @xmath3h , @xmath0he and @xmath2c suggests that this method can be also applied to @xmath1o beam .
the distribution of rbe@xmath45 for @xmath1o obtained on the basis of microdosimetry simulations with mchit is shown in figure [ fig : fig4 ] for comparison .
the shapes of rbe@xmath45 profiles for @xmath0he , @xmath2c and @xmath1o are found to be similar with a rise of rbe for @xmath2c and @xmath1o at the proximal edge of the sobp distribution .
a characteristic rise of rbe for @xmath3h beyond depth at the distal region of the sobp is found , similarly to the case of monoenergetic proton beam , sec .
[ sec : pristine ] .
it is found that rbe@xmath50 at the entrance region of @xmath3h and @xmath0he beams .
the ratio between rbe@xmath45 values at the proximal ( depth of @xmath47 ) and distal ( depth of @xmath47 ) regions is larger for @xmath1o compared to @xmath2c .
the rbe@xmath45 profile for @xmath1o demonstrates the most pronounced tail with respect to other projectiles .
distributions calculated with mchit and the mk model for sobp beams of @xmath3h , @xmath0he , @xmath2c and @xmath1o .
the beam energy profiles for @xmath3h , @xmath0he and @xmath1o were defined in such a way that they yield a sobp at the same depth as for @xmath2c ion .
the rbe@xmath45 values estimated with the mk model from measured @xmath18 @xcite are presented by full symbols as explained in the legend .
the rbe@xmath45 values estimated from lq fitting of survival curves of hsg cells are presented by open symbols . ]
the sobp distributions of biological dose @xmath36 for @xmath3h , @xmath0he , @xmath2c and @xmath1o are shown in figure [ fig : fig5 ] .
they are calculated for hsg cell as a product of physical dose calculated with mchit and rbe@xmath45 obtained within the mk model on the basis of microdosimetric modeling with mchit .
all four sobp distributions presented in figure [ fig : fig5 ] are wide , and they are normalized to 1 . at the plateau to facilitate the comparison of their shapes and ratios between the plateau and entry channel .
therefore , in the following the distributions of @xmath36 are discussed in terms of dose relative to the plateau values .
h , @xmath0he , @xmath2c and @xmath1o nuclei obtained as a product of physical dose calculated with mchit and rbe@xmath45 shown in figure [ fig : fig4 ] .
the distributions for all projectiles were rescaled to get the same value at the plateau . ]
the distribution of @xmath36 for @xmath2c is characterized by the lowest values at the entrance channel .
this helps to spare healthy tissues located in front of the target volume in treatments with @xmath2c .
in contrast , the highest entrance dose is predicted for protons .
however , the tail of the proton distribution beyond the distal edge of the plateau is negligible , while it is essential both for @xmath2c and @xmath1o .
this indicates that the proton beam is the best option if very sensitive organs are located behind the tumor volume .
the choice of ion species for each specific treatment can , in principle , provide an optimal ratio between the doses in the entrance and tail regions .
the distribution of @xmath36 for @xmath0he demonstrates a favorably small dose at the tail region , while its entrance value is higher compared to @xmath2c . the biological dose delivered by @xmath1o to normal tissue is slightly increased both in front of the target volume and behind it compared to @xmath2c . from the analysis of @xmath36 distributions
one can conclude that @xmath2c is the best treatment option compared to @xmath3h , @xmath0he and @xmath1o , unless only a very low @xmath36 is acceptable in the tail region . in the latter case protons become the best treatment option .
the central part of our study is devoted to the comparison of survival fractions of cells @xmath51 calculated as a function of depth in the water phantom , which are estimated for tissues of different radiosensitivity .
this makes possible to evaluate the respective therapeutic outcome for such tissues .
the distributions of @xmath51 calculated for cells ( tissues ) with @xmath52 gy , 2 gy and 10 gy after exposing them to @xmath3h , @xmath0he , @xmath2c and @xmath1o sobp beams are shown in figure [ fig : fig6 ] . throughout this text the parameters for hsg cells were taken as @xmath53 gy@xmath25 and @xmath52 gy @xcite .
hereafter the radiosensitivity of such tissues is considered as normal .
this serves as a natural reference point for comparison with two other tissues with their parameters taken following kase _
et al . _
the latter two cases correspond to early responding tissue ( @xmath54 gy@xmath25 , @xmath55 gy ) very sensitive to radiation and late responding tissue ( @xmath56 gy@xmath25 , @xmath57 gy ) which is radioresistant .
after irradiation with @xmath3h , @xmath0he , @xmath2c and @xmath1o sobp beams calculated for tissues with @xmath52 gy ( hsg cells , top panel ) , 2 gy ( early responding tissue , middle panel ) and 10 gy ( late responding tissue , bottom panel ) .
the values of @xmath24 parameter of the mk model are taken as 0.13 gy@xmath25 , 0.44 gy@xmath25 and 0.04 gy@xmath25 , respectively . ]
the dose applied to hsg cells ( normal radiosensitivity ) leads to their 10% survival at the target volume for all four beams .
the beams of @xmath0he and @xmath2c equally well spare tissue in the entrance channel , while the impact of @xmath3h and @xmath1o is stronger there . as one can expect , the main difference between light @xmath3h , @xmath0he and heavier @xmath2c , @xmath1o is revealed beyond the distal edge of sobp profile due to the tail of secondary fragments from @xmath2c and @xmath1o .
one can also note that @xmath1o is the worst option for hsg cells , while @xmath0he is the best one . in the case of early and late responding tissues ,
see figure [ fig : fig6 ] , middle and bottom panels , the relation between the survival fractions estimated for @xmath3h , @xmath0he , @xmath2c and @xmath1o beyond the distal edge of the sobp plateau are quite similar to the case of hsg cells .
this is because of the fact that the dose in the tail region is defined by the presence of secondary fragments , which is essential for @xmath2c and @xmath1o beams . in the case of early responding tissues
less than 5% of cells survive in the tumor volume , but their survival outside it is also unacceptably low ( 1020% ) .
this means that in this case there are no advantages of charged particle therapy with respect to the intensity modulated radiation therapy ( imrt ) with photons . in the case of late responding tissues more cells ( @xmath58% ) survive in the tumor volume after the impact of @xmath3h , @xmath0he , @xmath2c and @xmath1o sobp beams , see figure [ fig : fig6 ] , bottom panel .
the @xmath2c and @xmath1o sobp beams are very effective in killing tumor cells , but the cell survival is also lower in the entrance channel ( @xmath59% ) compared to @xmath60% for @xmath3h and @xmath0he . by considering all three sensitivity cases
one can conclude that the @xmath0he beam is equally suitable for irradiation of tissues with normal and low radiosensitivity as the @xmath2c beam .
moreover , due to the reduced fragmentation of @xmath0he , this option can be even better than @xmath2c when sparing tissues after the tumor volume is crucial . at the same time the @xmath1o beam has no clear advantages compared to @xmath2c .
due to higher ionization in the entrance channel and enhanced fragmentation more cells are killed by @xmath1o beam outside the tumor volume compared to @xmath2c beam .
however , one can consider @xmath1o is a good option for highly resistant tumors , as it effectively kills cells in the tumor volume .
in this work we presented our approach based on monte carlo simulations of microdosimetric spectra of monoenergetic beams of @xmath3h , @xmath0he , @xmath2c and @xmath1o in water .
it provides @xmath18 values as input to the modified mk model @xcite for calculating rbe@xmath45 for hsg cells for sobp distributions composed from monoenergetic beams of these projectiles .
this method gives rbe@xmath45 for @xmath3h , @xmath0he , @xmath2c which are in full agreement with rbe@xmath45 also calculated within the mk model , but from measured @xmath18 @xcite and rbe@xmath45 calculated from the parameters of lq fitting of survival curves of hsg cells @xcite .
this makes us confident in extending our approach to @xmath1o beams for which the respective data are not available .
our approach provides well - adjusted biological dose distributions for @xmath3h , @xmath0he , @xmath2c and @xmath1o with a very flat sobp plateau .
thus basic properties of mixed radiation fields in treatments with these projectiles are emulated .
it is found that the shapes of rbe@xmath45 profiles for @xmath0he , @xmath2c and @xmath1o are similar to each other , while the rbe@xmath45 for protons is almost constant ( @xmath61 ) over the whole depth in water , excluding enhanced rbe@xmath45 ( @xmath62 ) after the distal edge of the sobp plateau .
considerably lower rbe@xmath45 values are estimated in the entrance and tail region for @xmath0he compared to @xmath2c and @xmath1o . in the target volume
the highest rbe@xmath45 values of 1.52.5 are calculated for @xmath2c and @xmath1o . in order to reduce side effects of ion therapy such as radionecrosis @xcite the damage to surrounding healthy tissues
should be reduced as much as possible . with the help of our mchit model connected with the modified mk model the severity of this damage
is evaluated by calculating the cell survival fractions in healthy tissues for several kinds of therapeutic beams ( @xmath3h , @xmath0he , @xmath2c , @xmath1o ) .
we considered the cases of normal ( hsg cells ) , high and low radiosensitivity of tissues in the tumor volume and around it .
the consideration of the impact of @xmath3h , @xmath0he , @xmath2c and @xmath1o sobp beams in these three cases led us to the following conclusions : * in the case of early responding tissues all four charged particle beams induce severe damage not only to the target volume , but also around it . since in this case
the region of high damage is not conformal to the target volume , the treatment with charged particles loses its advantages with respect to treatment with photons . * in the case of tissues with normal radiosensitivity ( hsg cells ) @xmath0he and @xmath2c beams spare tissue in the entrance channel better than @xmath3h and @xmath1o ones . *
@xmath0he and @xmath2c nuclei are equally suitable for irradiation of tissues with normal and low radiosensitivity .
the cell survival fractions calculated , respectively , for the entrance channel and target volume are similar for @xmath0he and @xmath2c .
* however , as soon as it is important to spare healthy tissues after the distal edge of the sobp plateau , @xmath0h can be recommended due to the reduced nuclear fragmentation of these projectiles . *
no definitive advantages of @xmath1o with respect to @xmath2c were found , with the except of an enhanced impact of these heavier projectiles on radioresistant tumors . in a recent work @xcite
the authors studied the possibility to spare healthy tissues by properly selecting ion species for therapy .
in addition to @xmath63h , @xmath6he , @xmath2c they considered lithium , beryllium , boron and neon ions and calculated the dose to normal tissue delivered by these beams . however , the option of @xmath1o was not considered , while the heidelberger ionenstrahl - therapiezentrum ( hit ) in heildeberg , germany , provides @xmath1o beams of therapeutic energies @xcite in addition to @xmath63h , @xmath6he , @xmath2c .
treatments at this facility are performed presently only with protons and carbon ions , but @xmath1o can be also used following respective pre - clinical studies . in this sense
our study complements the results of remmes _ et al .
_ @xcite by considering @xmath1o beams , and also by comparing cell survival profiles in @xmath3h , @xmath0he , @xmath2c and @xmath1o treatments in addition to biological dose profiles .
as suggested @xcite , multi modal irradiations with various nuclei can be used for let - painting . in this method
high - let radiation is used to boost the let in a hypoxic sub - volume of the target ( hypoxic compartments of the tumor ) . at the same time
low - let radiation is applied to the complementary target volume .
such combination may increase tumor control and reduce side effects .
this means that a thorough evaluation of the physical properties and biological effectiveness of different beams is necessary before they can be applied in real treatments .
our approach can be also used for estimating rbe and cell survival fractions for @xmath13li , @xmath8be , @xmath9b and @xmath10n nuclei prior to planning radiobiological experiments with these beams .
this work was partially supported by hic for fair within the hessian loewe - initiative .
our calculations were performed at the center for scientific computing ( csc ) of the goethe university frankfurt .
we are grateful to the csc staff for support . | beams of @xmath0he and @xmath1o nuclei are considered for ion - beam cancer therapy as alternative options to protons and @xmath2c nuclei . spread - out bragg peak ( sobp ) distributions of physical dose and relative biological effectiveness for 10% survival are calculated by means of our geant4-based monte carlo model for heavy ion therapy ( mchit ) and the modified microdosimetric kinetic model .
the depth distributions of cell survival fractions are calculated for @xmath3h , @xmath0he , @xmath2c and @xmath1o for tissues with normal ( hsg cells ) , low and high radiosensitivity . in each case
the cell survival fractions were compared separately for the target volume , behind and in front of it . in the case of normal radiosensitivity @xmath0he and
@xmath2c better spare tissues in the entrance channel compared to protons and @xmath1o .
the cell survival fractions calculated , respectively , for the entrance channel and target volume are similar for @xmath0he and @xmath2c .
when it is important to spare healthy tissues located after the distal edge of the sobp plateau , @xmath0he can be recommended due to reduced nuclear fragmentation of these projectiles .
no definite advantages of @xmath1o with respect to @xmath2c were found , with the except of an enhanced impact of these heavier projectiles on radioresistant tumors . |
qcd predicts the existence of glueballs , the bound states of gluons , and the observation of glueballs would provide a direct test of qcd . in the quenched approximation ,
lattice qcd calculations predict the lightest glueball to be a @xmath7 with the mass being in the region from 1.4 to 1.8 gev/@xmath4 @xcite .
although the identification of a glueball is very complicated , there are several glueball candidates , including the @xmath17 and @xmath13 .
the properties of the @xmath17 and @xmath13 are reviewed in detail in the latest issue of the particle data group ( pdg ) @xcite .
@xmath3 radiative decays have been suggested as promising modes for glueball searches .
the @xmath18 process was analyzed previously in the markiii @xcite , dm2 @xcite and besi @xcite experiments , in which there was evidence for @xmath6 and an additional @xmath19 .
however , the high mass shoulder of the @xmath6 , at about 1.45 gev/@xmath4 , was unsettled .
a revised amplitude analysis of markiii data assigned the shoulder to be a scalar at @xmath20 1.43 gev/@xmath4 , and , in addition , found the peak at @xmath20 1.7 gev/@xmath4 to be scalar rather than tensor @xcite .
the @xmath21 process was also studied by the crystal ball @xcite and besi experiments @xcite , but no partial wave analysis has yet been performed on this channel . in this paper , the results of partial wave analyses on @xmath18 and @xmath22 are presented based on a sample of 58 m @xmath3 events collected by the upgraded beijing spectrometer ( besii ) located at the beijing electron positron collider ( bepc ) .
besii is a large solid - angle magnetic spectrometer that is described in detail in ref .
charged particle momenta are determined with a resolution of @xmath23 = 1.78 % @xmath24 ( @xmath25 in gev/@xmath26 ) in a 40-layer cylindrical main drift chamber ( mdc ) .
particle identification is accomplished by specific ionization ( @xmath27 ) measurements in the drift chamber and time - of - flight ( tof ) measurements in a barrel - like array of 48 scintillation counters .
the @xmath27 resolution is @xmath28 = 8.0 % ; the tof resolution is @xmath29 ps for bhabha events .
outside of the time - of - flight counters is a 12-radiation - length barrel shower counter ( bsc ) comprised of gas tubes operating in limited stream mode .
the bsc measures the energies of photons with a resolution of @xmath30 21 % /@xmath31 ( @xmath32 in gev ) . outside the solenoidal coil , which provides a 0.4 t magnetic field over the tracking volume ,
is an iron flux return that is instrumented with three double layers of counters that are used to identify muons . in this analysis , a geant3 based monte carlo simulation program ( simbes ) @xcite with detailed consideration of detector performance ( such as dead electronic channels )
the consistency between data and monte carlo has been checked in many high purity physics channels , and the agreement is quite reasonable @xcite .
the first level of event selection of @xmath18 requires two charged tracks with total charge zero . each charged track , reconstructed using mdc information ,
is required to be well fitted to a three - dimensional helix , be in the polar angle region @xmath33 , and have the point of closest approach of the track to the beam axis be within 2 cm of the beam axis and within 20 cm from the center of the interaction region along the beam line .
more than one photon per event is allowed because of the possibility of fake photons coming from the interactions of charged tracks with the shower counter or from electronic noise in the shower counter .
a neutral cluster is considered to be a photon candidate when the energy deposited in the bsc is greater than 50 mev , the first hit is in the beginning six radiation lengths , the angle between the nearest charged track and the cluster is greater than 18@xmath34 , and the angle between the cluster development direction in the bsc and the photon emission direction is less than 30@xmath34 . the total number of layers with hits associated with the two charged particles in the muon counter is required to be less than four in order to remove @xmath35 events . to remove the large backgrounds from bhabha events
, we require that ( i ) the opening angle of the two tracks satisfies @xmath36 and ( ii ) the energy deposit by each track in the bsc satisfies @xmath37 gev .
we require @xmath38 to remove @xmath39 conversions that occur at low @xmath1 mass . in order to reduce the background from final states with kaons and electrons , both tracks are required to be identified as pions by tof or @xmath27 when the momenta are lower than 0.7 gev/@xmath26 . in other cases ,
at least one track is required to be identified as a pion by tof .
requirements on two variables , @xmath40 and @xmath41 , are imposed @xcite .
the variable @xmath42 is required to satisfy @xmath43 gev . here ,
@xmath44 and @xmath45 are , respectively , the missing energy and momentum of charged particles .
the variable @xmath46 is required to be @xmath47 ( gev/@xmath26)@xmath48 , where @xmath49 is the angle between the missing momentum and the photon direction .
the @xmath40 cut removes most background from events having multikaon or other neutral particles , such as @xmath50 , @xmath51 events .
the cut on @xmath41 is used to reduce backgrounds with @xmath52s . in order to reduce the dominant @xmath53 background , events with more than one photon satisfying @xmath54 gev/@xmath4
are rejected . here
@xmath55 is the invariant mass of the two isolated photons with the smallest angle between the plane determined by these two photons and the direction of @xmath45 in all possible photon combinations .
@xmath55 is calculated using @xmath56 and the angle between @xmath45 and the @xmath39 direction .
the advantage of this method is that it uses the momenta of the charged tracks measured by the mdc , which has good momentum resolution , and is independent of photon energy measurement . finally , the two charged tracks and photon in the event are kinematically fitted using four energy and momentum conservation constraints ( 4-c ) under the @xmath57 hypothesis to obtain better mass resolution and to suppress backgrounds further by using the requirements @xmath58 and @xmath59 .
if there is more than one photon , the fit is repeated using all permutations and the combination with the best fit to @xmath60 is retained . for @xmath21 ,
the @xmath52 mesons in the event are identified through the decay @xmath61 .
the isolated photon is required to have the energy deposited in the bsc greater than 80 mev and come from the interaction point .
the number of isolated photons is required to be greater than four and less than seven .
a 4-c kinematic fit to @xmath62 is performed , the combination of five photons with the smallest @xmath63 is selected , and a kinematic fit chi - square @xmath64 is required . for five photons , there are 15 combinations from which to construct two @xmath52s .
to select @xmath52s , we choose the combination with the smallest @xmath65 , where @xmath66 and require @xmath67 mev/@xmath4 . to reduce background with @xmath68s , events with the invariant mass of a @xmath52 and one photon in the @xmath68 mass interval @xmath69 mev/@xmath4
are rejected . to further suppress backgrounds with more than one neutral particle recoling to the @xmath2 system ,
the recoiling mass squared of the @xmath2 system is required to be less than 4.8 ( gev/@xmath4)@xmath48 .
[ mass1 ] shows the @xmath1 mass spectrum for the selected events , together with the corresponding background distributions and the dalitz plot .
there is a strong @xmath70 peak mainly due to background from @xmath71 .
a strong @xmath6 signal , a shoulder on the high mass side of the @xmath6 , an enhancement at @xmath20 1.7 gev/@xmath4 , and a peak at @xmath20 2.1 gev/@xmath4 are clearly visible .
the lightly shaded histogram in fig .
[ mass1 ] corresponds to the dominant background @xmath72 .
the data taken at the @xmath73 center of mass energy of 3.07 gev , with a luminosity of @xmath74 nb@xmath75 , are used to determine the continuum background .
the sum of continuum background and the other possible backgrounds , such as @xmath76 ( @xmath77 ) , @xmath78 , @xmath79 , is estimated to be 3.8 % of the data in the whole mass range and is shown as the dark shaded histogram in fig .
[ mass1 ] .
2 shows the @xmath2 mass spectrum and the dalitz plot .
the shaded histogram corresponds to the sum of estimated backgrounds determined using pdg branching ratios @xcite .
the backgrounds are mainly from @xmath80 ( @xmath81 , @xmath82 ) , @xmath83 , ( @xmath84 ) and @xmath85 ( @xmath86 , @xmath82 ) .
the peak below 0.5 gev/@xmath4 is mainly from @xmath87 , ( @xmath88 ) .
the continuum background is also studied using the data taken at the @xmath73 center of mass energy of 3.07 gev .
no events survive selection requirements . in general , the @xmath1 and @xmath2 mass spectra exhibit similar structures above 1.0 gev/@xmath4 . and the dalitz plot for @xmath18 , where the lightly and dark shaded histograms in the upper panel correspond to @xmath72 and other estimated backgrounds , respectively.,title="fig:",scaledwidth=35.0% ] and the dalitz plot for @xmath18 , where the lightly and dark shaded histograms in the upper panel correspond to @xmath72 and other estimated backgrounds , respectively.,title="fig:",scaledwidth=35.0% ] and the dalitz plot for @xmath21 , where the shaded histogram in the upper panel corresponds to the estimated backgrounds.,title="fig:",scaledwidth=35.0% ] and the dalitz plot for @xmath21 , where the shaded histogram in the upper panel corresponds to the estimated backgrounds.,title="fig:",scaledwidth=35.0% ]
we have carried out partial wave analyses for the @xmath14 mass range from 1.0 to 2.3 gev/@xmath4 using relativistic covariant tensor amplitudes constructed from lorentz - invariant combinations of the polarization and momentum 4-vectors of the initial and final state particles , with helicity @xmath89 for @xmath3 initial states @xcite .
cross sections are summed over photon polarizations .
the relative magnitudes and phases of the amplitudes are determined by a maximum likelihood fit . for @xmath18 , the following channels are fitted to the data : @xmath90 @xmath91 constant width breit - wigner functions are used for each resonance .
the form is described as follows : @xmath92 where @xmath93 is the square of @xmath1 invariant mass , @xmath94 and @xmath95 are the mass and width of intermediate resonance @xmath96 , respectively .
the dominant background in @xmath18 comes from @xmath72 . from markiii s analysis on @xmath72 , described in terms of the amplitudes representing the sequential two - body decay process @xmath97 @xcite , the @xmath98 is dominant , but there are also contributions from the excited states of @xmath98 .
a preliminary pwa on besii @xmath72 also shows that a complete description of the data requires not only the dominant @xmath98 , but also the contributions from @xmath99 states @xmath100 , @xmath101 and @xmath102 , as well as @xmath103 state @xmath104 . using the branching fraction measurement of besii @xcite , which agrees with the result from the babar collaboration @xcite , and the generator based on the published results of ref .
@xcite , @xmath72 events are generated and given the opposite log likelihood in the fit to cancel the background events in the data .
the other background contributions are much smaller than @xmath72 background in the 1.0 to 2.3 gev/@xmath4 @xmath1 mass range and not considered in the partial wave analysis . due to the limitation of the present statistics and the complexity of the large @xmath53 background , it is difficult to cleanly distinguish components in the high mass region .
the main goal of this analysis will be to understand the structures below 2.0 gev/@xmath4 .
we use the four states @xmath105 , @xmath106 , @xmath107 , and @xmath108 , which are listed in pdg @xcite in the fit with the masses and widths fixed to those in the pdg , to describe the contribution of the high mass states in the mass range below 2.0 gev/@xmath4 . for the @xmath109 states , relative phases between different helicity amplitudes for a single resonance
are theoretically expected to be very small @xcite .
therefore , these relative phases are set to zero in the final fit so as to constrain the intensities further . after the mass and width optimization
, the resulting fitted intensities are illustrated in figs . 3 and 4 .
angular distributions in the whole mass range are shown in fig .
[ fangcr ] . here , @xmath49 is the polar angle of the photon in the @xmath3 rest frame , and @xmath110 is the polar angle of the pion in the @xmath1 rest frame . from figs . 3 and 5
, we see that the fit agrees well with data .
[ fm2cr ] shows the distributions of the individual components and full @xmath7 and @xmath109 contributions including interferences
. a free fit to @xmath6 gives a fitted mass of @xmath111 mev/@xmath4 and a width of @xmath112 mev/@xmath4 .
the fitted masses and widths of the @xmath17 and @xmath13 are m@xmath113 @xmath114 mev/@xmath4 , @xmath115 mev/@xmath4 and m@xmath116 @xmath117 mev/@xmath4 , @xmath118 mev/@xmath4 , respectively .
the branching fractions of @xmath6 , @xmath17 , and @xmath13 determined by the partial wave analysis fit are @xmath119 , @xmath120 , and @xmath121 , respectively . for the @xmath6
, we find the ratios of helicity amplitudes @xmath122 and @xmath123 with correlation factor @xmath124 , where @xmath125 , @xmath126 , @xmath127 correspond to the three independent production amplitudes with helicity 0 , 1 and 2 .
the errors here are statistical errors .
an alternative fit is tried by replacing @xmath17 with a @xmath109 resonance .
there are three helicity amplitudes fitted for spin 2 , while only one amplitude for spin 0 , which means the the number of degrees of freedom is increased by 2 in the @xmath128 case . however , the log likelihood is worse by 108 .
this indicates that @xmath17 with @xmath129 is strongly favored .
if the @xmath13 is removed from the fit , the log likelihood is worse by 379 , which corresponds to a signal significance much larger than @xmath130 .
the partial wave analysis of @xmath21 is performed independently .
the components used are the same as those of @xmath18 . due to the limited statistics , we take the partial wave analysis fit results of @xmath21 as a cross - check of the ones obtained in the charged channel .
the background distribution is close to flat in the mass interval 1.0 - 2.3 gev/@xmath4 , so we use @xmath131 multiplied by phase space to approximately describe such a flat contribution , where the 2nd order polynomial function @xmath132 is obtained by fitting the @xmath2 mass phase space distribution of @xmath21 monte carlo simulation .
the fitted intensities as a function of @xmath2 mass are illustrated in fig .
. a free fit to @xmath6 gives a fitted mass of @xmath133 mev/@xmath4 and a width of @xmath134 mev/@xmath4 .
the fitted masses and widths of the @xmath17 and @xmath13 are m@xmath113 @xmath135 mev/@xmath4 , @xmath136 mev/@xmath4 and m@xmath116 @xmath137 mev/@xmath4 , @xmath138 mev/@xmath4 , respectively .
the errors shown here are statistical .
besides the above global fit , a bin - by - bin fit is applied to @xmath18 data using the method described in ref .
@xcite . a strong @xmath6
is observed , and the s - wave @xmath1 mass distribution shows a large signal at @xmath20 1.45 gev/@xmath4 , a significant signal at @xmath20 1.75 gev/@xmath4 , and a peak at @xmath20 2.1 gev/@xmath4 . in general ,
the bin - by - bin fit gives similar features as those of the global fit , and the results of these two fits are approximately consistent with each other .
the systematic errors for the partial wave analysis fit to @xmath18 data are estimated by varying the masses and widths of the @xmath6 , @xmath17 , and @xmath13 within the fitted errors ; varying the masses and widths of the @xmath105 , @xmath106 , @xmath107 , and @xmath108 within the pdg errors @xcite ; adding a small component @xmath139 ; varying the background fraction within reasonable limit and replacing the @xmath105 with the @xmath140 .
they also include the uncertainties in the number of @xmath3 events analyzed , the efficiency of photon detection , the efficiency of mdc tracking , and the kinematic fit .
different generators , one is based on the published results of ref .
@xcite and another is from besii preliminary pwa results which include a dominant @xmath98 and its excited states @xmath100 , @xmath101 , @xmath102 and @xmath141 , are used for the estimation of the background @xmath142 .
the pwa results with different backgrounds agree with each other within the error .
their difference is taken as the systematic error too .
the systematic errors in this analysis do not include all model - dependent effects , such as the difference of using simple breit - wigner formalism and k - matrix formalism .
table [ sys1 ] shows the summary of the systematic errors for the global fit . for the @xmath6 ,
the total systematic errors are 0.10 and 0.19 for @xmath143 and @xmath144 , respectively .
the correlation factor between the @xmath143 and @xmath144 systematic errors is 0.29 which is calculated with @xmath145 , where i runs over all systematic errors .
table [ result1 ] shows the mass , width , and branching fraction measurements for @xmath6 , @xmath17 , and @xmath13 , where the first error is statistical and the second is systematic , determined by adding all sources in quadrature . in order to compare the branching fractions of @xmath6 , @xmath17 , and @xmath13 in @xmath146 and @xmath22 , we fix the mass and width of each component in @xmath22 to those of the charged channel and re - calculate the branching fractions and estimate the systematic errors .
the results are shown in table [ result2 ] .
the branching fractions determined from the two channels agree with each other within errors after considering isospin corrections . [ cols="<,^,^,^",options="header " , ] [ result2 ]
the measured mass of the @xmath6 , @xmath147 mev/@xmath4 , is lower than the pdg value , and the branching fraction of @xmath148 is a bit higher than the pdg value @xcite .
a fit with the pdg mass and width is visibly poorer , and the log likelihood is worse than the optimum fit by 44 . in this analysis ,
the s - wave contribution on the high mass shoulder of the large @xmath6 peak is well separated , which may explain this mass difference .
the ratios of the helicity amplitudes of the @xmath6 from the present analysis are @xmath149 and @xmath150 with correlation @xmath151 and @xmath152 .
the values of @xmath143 and @xmath144 are in agreement with predictions @xcite within the errors .
comparing to the results determined by dm2 @xcite , markiii @xcite , and crystal ball @xcite , there is a difference in the value of y. the main reason for this difference may be that we consider influences from other states in the 1.0 to 2.3 gev/@xmath4 @xmath5 mass range , while previous analyses by dm2 and markiii ignored these and only considered the @xmath6 in the 1.15 - 1.4 gev/@xmath4 @xmath5 mass range .
the most remarkable feature of the above results is that three @xmath7 states at the @xmath14 mass @xmath20 1.45 , 1.75 , and 2.1 gev/@xmath4 are observed from the partial wave analysis . for the high mass state at @xmath20 2.1 gev/@xmath4
, we use the @xmath106 , which is listed in the pdg , to describe it .
no further efforts are made on the measurements of its resonant parameters due to the difficulties described in section iv .
the lower @xmath7 state peaks at a mass of @xmath8 mev/@xmath4 with a width of @xmath9 mev/@xmath4 , which is consistent with the scalar glueball candidate , @xmath17 .
spin 0 is strongly preferred over spin 2 .
therefore we interpret the small but definite shoulder on the high mass side of the @xmath6 in the @xmath1 mass distribution as originating from the @xmath17 , which interferes with nearby resonances in the partial wave analysis .
however , due to the uncertainties of the mass and width determinations and large interferences between the s - wave states , the existence of the @xmath153 in @xmath154 can not be excluded by present data .
strong production of the @xmath13 was observed in the partial wave analysis of @xmath155 , with a mass of @xmath156 mev/@xmath4 , a width of @xmath157 mev/@xmath4 , and a branching fraction of @xmath158 of @xmath159 @xcite . interpreting the @xmath7 state in the mass region
@xmath20 1.75 gev/@xmath4 as coming from the @xmath13 and using the branching fraction of @xmath160 determined from @xmath161 after isospin correction and the branching fraction of @xmath162 in ref .
@xcite , we obtain the ratio of @xmath14 to @xmath15 branching fractions for the @xmath13 as @xmath163 the ratio is consistent with the pdg value ( @xmath164 ) @xcite within the errors . an alternative interpretation for this @xmath7 state
is the @xmath165 .
data on @xmath166 and @xmath167 show a definite peak in @xmath168 at 1790 mev/@xmath4 but no significant signal in @xmath169 @xcite .
if the mass and width of the @xmath7 state are fixed to 1790 mev/@xmath4 and 270 mev/@xmath4 found in @xcite , the log likelihood is worse by 47 .
this @xmath7 state may also be a superposition of @xmath13 and @xmath165 .
due to the uncertainty of the high mass region , we do an alternative fit removing the @xmath108 and re - optimizing the masses and widths of @xmath6 , @xmath17 , and @xmath13 for the @xmath18 sample .
the log likelihood is worse by 160 .
the fit gives a @xmath6 mass of @xmath170 mev/@xmath4 and a width of @xmath171 mev/@xmath4 .
the measured masses and widths of the @xmath17 and @xmath13 are m@xmath113 @xmath172 mev/@xmath4 , @xmath173 mev/@xmath4 and m@xmath116 @xmath174 mev/@xmath4 , @xmath175 mev/@xmath4 , respectively .
the branching fractions of the @xmath6 , @xmath17 , and @xmath13 are @xmath176 , @xmath177 , and @xmath178 , respectively .
the light - meson spectroscopy of scalar states in the mass range 1 - 2 gev/@xmath4 , which has long been a source of controversy , is still very complicated .
overlapping states interfere with each other differently in different production and decay channels .
more experimental data are needed to clarify the properties of these scalar states .
in summary , the partial wave analyses of @xmath18 and @xmath21 using 58 m @xmath3 events of besii show strong production of @xmath6 and evidence for two @xmath7 states in the 1.45 and 1.75 gev/@xmath4 mass regions . for the @xmath6 ,
the branching ratio determined by the partial wave analysis fit is @xmath179 .
the ratios of the helicity amplitudes of the @xmath6 are determined to be @xmath180 and @xmath181 with correlations @xmath151 and @xmath152 .
the @xmath17 has a mass of @xmath182 mev/@xmath4 , a width of @xmath9 mev/@xmath4 , and a branching fraction @xmath183 .
the @xmath7 state in the @xmath201.75 gev/@xmath4 mass region has a mass of @xmath184 mev/@xmath4 and a width of @xmath185 mev/@xmath4 .
if this @xmath7 state is interpreted as coming from @xmath13 , the ratio of the @xmath14 to @xmath15 branching fractions is @xmath16 .
this may help in understanding the properties of @xmath17 and @xmath13 .
the bes collaboration thanks the staff of bepc and computing center for their hard efforts .
we wish to thank prof .
david bugg for contributions to the early stage of this analysis .
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 , and the knowledge innovation project of cas under contract nos .
kjcx2-sw - n10 , u-602 , u-34 ( ihep ) , the national natural science foundation of china under contract no .
10225522 ( tsinghua university ) , and the department of energy under contract no.de-fg02-04er41291 ( u hawaii ) .
99 g. bali , k. schilling , a. hulsebos , a. irving , c. michael , and p. stephenson , phys .
b * 309 * , 378 ( 1993 ) ; c. michael , _ hadron spectroscopy _ , aip conf .
432 ( aip .
melville , ny , 1997 ) , p. 657 ; w. lee , and d. weingarten , hep - lat/9805029 ; c. morningstar , and m. peardon , phys .
d * 60 * , 034509 ( 1999 ) . | results are presented on @xmath0 radiative decays to @xmath1 and @xmath2 based on a sample of 58 m @xmath3 events taken with the besii detector .
partial wave analyses are carried out using the relativistic covariant tensor amplitude method in the 1.0 to 2.3 gev/@xmath4 @xmath5 mass range .
there are conspicuous peaks due to the @xmath6 and two @xmath7 states in the 1.45 and 1.75 gev/@xmath4 mass regions .
the first @xmath7 state has a mass of @xmath8 mev/@xmath4 , a width of @xmath9 mev/@xmath4 , and a branching fraction @xmath10 .
spin 0 is strongly preferred over spin 2 .
the second @xmath7 state peaks at @xmath11 mev/@xmath4 with a width of @xmath12 mev/@xmath4 .
if this @xmath7 is interpreted as coming from @xmath13 , the ratio of its branching fractions to @xmath14 and @xmath15 is @xmath16
. 0.2 cm |
for all terms and definitions , not defined specifically in this paper , we refer to @xcite . unless mentioned otherwise , all graphs considered here are simple , finite and have no isolated vertices .
the researches in the field of graph labeling commenced with the introduction of the notion of graceful labeling of graphs in @xcite .
analogous to the number valuations of graphs , the notion of set - valuations of graphs was introduced in @xcite .
let us first consider the following notions of a set - labeling and a set - indexer of a given graph @xmath1 .
[ d0 ] [ d1 ] .
note that if either @xmath8 or @xmath9 is countably infinite , then their sum set will also be countably infinite .
hence , all sets we consider here are finite sets .
we denote the cardinality of a set @xmath8 by @xmath10 . using the concepts of sum sets of finite sets , the notion of an integer additive set - indexer of a given graph
@xmath1 has been introduced as follows .
[ d2 ] .
[ d3 ] [ du ] @xcite an iasi is said to be _ @xmath5-uniform _ if @xmath6 for all @xmath7 .
that is , a connected graph @xmath1 is said to have a @xmath5-uniform iasi if all of its edges have the same set - indexing number @xmath5 .
in particular , we say that a graph @xmath1 has an _
arbitrarily @xmath5-uniform iasi _ if @xmath1 has a @xmath5-uniform iasi for every positive integer @xmath5 . in @xcite ,
we have proved the following results on weakly uniform and arbitrarily uniform iasis .
[ l1 ] @xcite let @xmath8 and @xmath9 be two non - empty finite subsets of @xmath0
. then @xmath11 .
[ r1 ] the following results are a necessary and sufficient condition for an iasi to be a weak iasi and a weakly uniform iasi respectively , of a given graph @xmath1 .
[ l2a ] @xcite an iasi @xmath12 of a graph @xmath1 is a weak iasi if and only if at least one end vertex of every edge of @xmath1 has the set - indexing number @xmath13 .
[ l2b ] @xcite an iasi @xmath12 of a graph @xmath1 is a weakly @xmath5-uniform iasi if and only if one end vertex of every edge of @xmath1 has the set - indexing number @xmath13 and the other end vertex has the set - indexing number @xmath5 , where @xmath5 is a positive integer . in this paper
, we study a special type of integer additive set - indexer of a graph @xmath1 called a strong integer additive set - indexer of @xmath1 .
first note that all the sets we consider during this discussion , are non - empty , finite sets of non - negative integers . in view of lemma [ l1 ] ,
the maximum cardinality of the sum set of two non - empty finite sets is the product of the cardinalities of each individual sets .
that is , for two non - empty finite sets @xmath8 and
@xmath9 , we have @xmath14 . in this context
, the study regarding characteristics of graphs in which the set - indexing number of every edge is the product of the set - indexing numbers of its end vertices arises much interest .
hence , we define [ d6 ] [ t2 ] let @xmath8 , @xmath9 be two non - empty distinct subsets of @xmath0
. then , @xmath15 if and only if the differences between any two elements of one set are not equal to any differences between any two elements of the other . due to lemma [ l1 ] , we have @xmath16
. we have @xmath17 if and only if @xmath18 for some @xmath19 and @xmath20
. that is , when @xmath21 .
therefore , @xmath15 if and only if @xmath22 for any @xmath19 and @xmath20 .
let @xmath23 and let @xmath24 .
then , it is clear that @xmath25 if and only if no element of @xmath26 is equal to an element of @xmath27 .
the set of all positive differences between any two elements of a set @xmath8 is called the _ difference set _ of @xmath8 and can be denoted by @xmath26 .
that is , @xmath28 .
hence , we propose the following theorem which establishes a necessary and sufficient condition for a complete graph @xmath29 to admit a strong iasi .
let @xmath8 and @xmath9 be two non - empty subsets of @xmath0 .
we use the notation @xmath30 in the sense that @xmath31 for all @xmath32 and @xmath33 .
that is , @xmath34 . also , by the sequence @xmath35 , we mean that all the sets @xmath36 are pairwise disjoint .
[ tsk ] let each vertex @xmath37 of the complete graph @xmath29 be labeled by the set @xmath38 .
then , @xmath29 admits a strong iasi if and only if there exists a finite sequence of sets @xmath39 where each @xmath40 is the difference set of the set - label @xmath36 .
assume that @xmath29 admits a strong iasi .
let @xmath41 be the vertex set of @xmath29 .
let each vertex @xmath37 is labeled by the set @xmath36 in @xmath42 .
define the set @xmath40 by @xmath43 which is the set of all differences between any two elements of @xmath36 .
since the vertex @xmath44 is adjacent to @xmath45 , without loss of generality , assign the set @xmath46 to @xmath44 and the set @xmath47 to the vertex @xmath45 .
then , by lemma [ t2 ] , we have @xmath48 .
since @xmath45 is adjacent to @xmath49 , assign @xmath50 to @xmath49 so that @xmath51 .
combining these two , we get @xmath52 . proceeding in this way , for @xmath53 ,
assign the set @xmath36 to @xmath37 and the set @xmath54 to @xmath55 so that @xmath56 .
after all possible assignments , we get @xmath39 .
conversely , assume that each vertex @xmath37 of the complete graph @xmath29 is labeled by a non - empty set @xmath38 such that there exists a finite sequence of sets @xmath39 where each @xmath40 is the set of all differences between any two elements of the set @xmath36 .
then , for each edge @xmath57 we have @xmath56 .
hence , by lemma [ t2 ] , @xmath58 for all @xmath59 .
therefore , @xmath1 admits a strong iasi .
[ tss ] if a graph @xmath1 admits a strong iasi , then its subgraphs also admit strong iasi .
the following result is a general condition for a general graph to admit a strong iasi .
[ tsc ] a connected graph @xmath1 ( on @xmath60 vertices ) admits a strong iasi if and only if each vertex @xmath37 of @xmath1 is labeled by a set @xmath36 in @xmath42 and there exists a finite sequence of sets @xmath61 , where @xmath62 is a positive integer and each @xmath40 is the set of all differences between any two elements of @xmath36 . in view of theorem [ tss ] , an interesting question we need to answer is whether a strong iasi of a graph @xmath1 induces to some other graph structures formed from @xmath1 by removing or identifying some of its elements ( vertices or edges ) .
first , recall the following notion .
the following theorem establishes a necessary and sufficient condition for a graph homeomorphic to a given strong iasi graph @xmath1 to admit a strong iasi .
let @xmath1 be a graph which admits a strong iasi .
then , any graph @xmath63 , obtained by applying a finite number of elementary topological reductions on @xmath1 , admits a strong iasi if and only if there exist the same number of paths @xmath64 , not the part of any triangle in @xmath1 , such that the difference sets of the set - labels of their vertices are pairwise disjoint .
let @xmath1 be a graph which admits a strong iasi .
let @xmath65 be a vertex of @xmath1 with @xmath66 . then @xmath65 is adjacent two vertices @xmath67 and @xmath68 in @xmath1
. then we have @xmath69 and @xmath70 .
let @xmath71 , @xmath72 and @xmath73 be the labeling sets of @xmath74 respectively .
also let @xmath75 be the corresponding sets of all differences between the elements of @xmath76 respectively .
assume that @xmath77 admits a strong iasi . then , @xmath78 .
therefore , by [ t2 ] , the sets all differences of the set - labels of @xmath67 and @xmath68 are pairwise disjoint in @xmath79 .
that is , @xmath80 in @xmath79 .
hence , the path @xmath81 in @xmath1 whose vertices have the set - labels with the difference sets of these vertices are pairwise disjoint .
conversely , assume that the difference sets of the set - labels of whose vertices are pairwise disjoint.that is , @xmath82 .
now , delete @xmath65 from @xmath1 .
let @xmath83 . here , we have @xmath80 .
then , by lemma [ t2 ] , @xmath78 in @xmath79 .
all other vertices and edges are the same in @xmath79 and @xmath1 .
therefore , @xmath79 admits a strong iasi .
this completes the proof .
figure [ ghom ] depicts the existence of strong iasis of the graphs @xmath84 , and @xmath85 which are obtained by applying finite number of elementary topological reductions on @xmath1 . in the following section
, we are going to discuss about the strongly uniform integer additive set - indexers of given graphs .
in other words , we can say that a strong integer additive set - indexer @xmath12 of a graph @xmath1 is said to be a _
strongly @xmath5-uniform integer additive set - indexer _ if @xmath86 for all @xmath7 , where @xmath5 is a positive integer . in view of the above definition , we note the following .
[ r6 ] [ r7 ] let us now proceed to discuss the admissibility of strongly uniform iasis by certain graphs and conditions required for their existence for various graphs .
the following theorem establishes the admissibility of strongly uniform iasis by bipartite graphs .
[ tbs ] for any positive integer @xmath5 , all bipartite graphs admit a strongly @xmath5-uniform iasi .
the following result is a special case when the number @xmath5 , we consider above is a prime number .
[ cbp ] if @xmath87 is a prime number , then a strongly @xmath87-uniform iasi of a bipartite graph @xmath1 is also a weakly @xmath87-uniform iasi for @xmath1 .
invoking theorem [ tbs ] and corollary [ cbp ] , we have the following result .
[ cbc ] a bipartite graph has distinct weakly @xmath5-uniform iasi and strongly @xmath5-uniform iasi if and only if @xmath5 is a positive composite integer .
figure [ su - graphs ] illustrates some strong uniform iasis for bipartite graphs .
it may be noted that the second graph in figure [ su - graphs ] is a 4-uniform iasi as well as a @xmath88-completely uniform iasi .
the following theorem discusses a condition required for a complete graph to admit a strongly @xmath5-uniform iasi .
[ tcg ] a strongly @xmath5- uniform iasi of a complete graph @xmath29 is a @xmath89-completely uniform iasi , where @xmath90 .
since all distinct pairs of vertices are adjacent in @xmath29 , @xmath29 admits a strongly @xmath5- uniform iasi when the set - indexing number of all vertices of @xmath29 are equal , say @xmath91 .
therefore , @xmath29 admits @xmath89-completely uniform iasi . here
@xmath92 . to proceed further
, we need the following results from the number theory .
[ r8 ] @xcite if @xmath93 a non - square integer , then for every divisor @xmath94 of @xmath93 there exists another divisor @xmath95 for @xmath93 .
that is , the number of divisors of a non - square integer is even .
[ r9 ] @xcite if @xmath93 a perfect square integer , then for every divisor @xmath94 of @xmath93 there exists another divisor @xmath95 for @xmath93 except for the divisor @xmath96 .
that is , the number of divisors of a perfect square is odd . in view of proposition [ r8 ] and proposition [ r9 ]
, we have the following theorem . [ m3 ]
let @xmath1 be a graph which admits a strongly @xmath5-uniform iasi .
also , let @xmath60 be the number of divisors of @xmath5
. then if @xmath5 is a non - square integer , then @xmath1 has at most @xmath97 bipartite components and if @xmath5 is a perfect square integer , then @xmath1 has at most of @xmath98 components in which at most @xmath99 components are bipartite components .
let @xmath1 be the graph which admits a strongly @xmath5-uniform iasi .
also let @xmath100 .
* case 1 : * let @xmath5 be a non - square integer . then , by result [ r8 ] , the number of divisors @xmath60 of @xmath5 is even . by remark [ r7 ]
, @xmath101 can be partitioned into at most @xmath60 sets .
let this partition be @xmath102 , where each @xmath103 is the set of vertices having set - indexing number @xmath104 .
since @xmath1 admits a maximal @xmath5-uniform iasi , each vertex of @xmath103 is adjacent to some vertices of @xmath105 so that @xmath106 . also , no two vertices of @xmath103 and no two vertices of @xmath105 are adjacent and no two vertices of both the sets can not be adjacent to the vertices of any other set @xmath107 of @xmath1 . that is , @xmath108 is a component of @xmath1 which is bipartite .
we can find at most @xmath97 such bipartite components @xmath109 of @xmath1 .
* case 2 : * let @xmath5 be a perfect square . then , by result [ r9 ] , the number of divisors @xmath60 of @xmath5 is odd . as in case 1 , for each @xmath103 of @xmath101 , we can find a set @xmath105 such that @xmath109 is a bipartite component of @xmath1 except for the set @xmath110 whose corresponding divisor @xmath111 .
but no vertex of @xmath110 can be adjacent to a vertex of any other set @xmath112 .
hence , the induced subgraph @xmath113 $ ] of @xmath1 is a component of @xmath1 such that the vertices in @xmath110 are adjacent to some vertices of itself .
therefore , @xmath1 has at most @xmath99 bipartite components and the component @xmath113 $ ] .
that is , @xmath1 has at most @xmath98 components of which @xmath99 are bipartite components .
[ csb ] let @xmath5 be a non - square integer .
then a graph @xmath1 admits a strongly k - uniform iasi if and only if @xmath1 is bipartite or a union of disjoint bipartite components .
recall that a _ clique _ of a graph @xmath1 is a maximal complete subgraph of @xmath1 .
[ m4 ] if a graph @xmath1 , which admits a strongly @xmath5-uniform iasi , then it contains at most one component which is a clique .
let the graph @xmath1 has a strongly @xmath5-uniform iasi .
hence , if @xmath1 has a component which is a clique , then @xmath5 is a perfect square . in view of the results discussed above , we establish the following theorem . [ tnb ]
a connected non - bipartite graph @xmath1 admits a strongly @xmath5-uniform iasi if and only if @xmath5 is a perfect square and this iasi is a @xmath89-completely uniform , where @xmath90 . invoking the above results ,
more generally we have a connected graph @xmath1 admits a strongly @xmath5-uniform iasi if and only if @xmath1 is bipartite or this iasi is a @xmath89-completely uniform iasi of @xmath1 , where @xmath114 .
in this paper , we have given some characteristics of the graphs which admit strong iasis and strongly @xmath5-uniform iasis .
more properties and characteristics of weak and strong iasis , both uniform and non - uniform , are yet to be investigated .
more studies may be done in the field of iasi when the ground set @xmath115 is finite instead of the set @xmath0 .
we have formulated some conditions for certain graphs to admit weakly and strongly uniform iasis .
the problems of establishing the necessary and sufficient conditions for various graphs and graph classes to admit these types of uniform iasis have still been unsettled .
certain iasis which assign set - labels having specific properties , to the elements of given graphs also seem to be promising for further investigations .
all these facts highlight the wide scope for intensive studies in this area .
25 b d acharya , * set - valuations and their applications * , mri lecture notes in applied mathematics , the mehta research institute of mathematics and mathematical physics , new delhi,1983 .
b d acharya , _ set - indexers of a graph and set - graceful graphs _ , bull .
allahabad math .
soc . , * 16*(2001 ) , 1 - 23 .
b d acharya , k a germina and t m k anandavally , _ some new perspective on arithmetic graphs _ in * labeling of discrete structures and applications * , ( eds . :
b d acharya , s arumugam and a rosa ) , narosa publishing house , new delhi , ( 2008 ) , 41 - 46 .
b d acharya and k a germina , _ strongly indexable graphs : some new perspectives _ , adv .
modeling and optimisation , * 15*(1)(2013 ) , 3 - 22 .
s alexander ( 2008 ) , * the mathematical coloring book * , springer - verlag .
t m k anandavally , _ a characterisation of 2-uniform iasi graphs _
journal of contemp .
sciences , * 8*(10)(2013 ) , 459 - 462 .
tom m apostol , * introduction to analytic number theory * , springer - verlag , new york , ( 1989 ) .
j a bondy and u s r murty , * graph theory * , springer , ( 2008 ) .
j a bondy and u s r murty , * graph theory with applications*,north - holland , new york , ( 1976 ) .
d m burton , * elementary number theory * , tata mcgraw - hill inc . , new delhi , ( 2007 ) . g chartrand and p zhang , * introduction to graph theory * , mcgraw - hill inc . , ( 2005 ) .
r diestel , ( 2000 ) . * graph theory * , springer - verlag , new york , ( 2000 ) .
j a gallian , _ a dynamic survey of graph labelling _ , the electronic journal of combinatorics ( ds 16 ) , ( 2011 ) . k a germina , * set - valuations of a graph and applications * , final technical report , dst grant - in - aid project no.sr/s4/277/05 , the dept . of science and technology ( dst ) , govt . of india , ( 2011 ) .
k a germina and kumar abhishek , _ set - valued graphs
i _ , ispacs journal of fuzzy set valued analysis , * 2012 * , article idjfsva-00127 , ( 2012 ) , 17 pages .
k a germina and t m k anandavally , _
integer additive set - indexers of a graph : sum square graphs _
, journal of combinatorics , information and system sciences , * 37*(2 - 4)(2012 ) , 345 - 358 .
k a germina , n k sudev , ( 2013 ) . _ on weakly uniform integer additive set - indexers of graphs _ , international mathematical forum , * 8*(37 ) , 1827 - 1834 .
j. gross , j. yellen , * graph theory and its applications * , crc press , ( 1999 ) .
f harary , * graph theory * , addison - wesley publishing company inc . , ( 1969 ) .
k d joshi , * applied discrete structures * , new age international , ( 2003 ) .
m b nathanson ( 1996 ) .
* additive number theory , inverse problems and geometry of sumsets * , springer , new york .
f s roberts , * graph theory and its practical applications to problems of society * , siam , philadelphia , ( 1978 ) .
a rosa , _ on certain valuation of the vertices of a graph _ , in * theory of graphs * , gordon and breach , new york and dunod , paris , ( 1967 ) [ proceedings of the international symposium held in rome ] .
d b west , * introduction to graph theory * , pearson education inc . , ( 2001 ) . | let @xmath0 be the set of all non - negative integers .
an integer additive set - indexer of a graph @xmath1 is an injective function @xmath2 such that the induced function @xmath3 defined by @xmath4 is also injective .
an iasi is said to be _ @xmath5-uniform _ if @xmath6 for all @xmath7 . in this paper
, we introduce the notions of strong integer additive set - indexers and initiate a study of the graphs which admit strong integer additive set - indexers .
keywords : set - indexers ; integer additive set - indexers ; strong integer additive set - sndexers ; strongly uniform integer additive set - indexers .
mathematics subject classification 2010 : 05c78 |
centaurusa ( cena ) is by far the most nearby radio galaxy and one of the brightest radio sources in the sky .
its proximity of only @xmath10mpc @xcite makes it a unique target for studies of supermassive black hole ( smbh ) accretion , jet formation and acceleration , as well as the interaction of the jets and lobes with the interstellar and intergalactic media .
centaurusa ( see * ? ? ?
* for a review of the properties of cena ) is embedded in the giant elliptical host ngc5128 .
unlike most other elliptical galaxies , ngc5128 displays very prominent dust lanes that are likely due to merger activity in the past ( see e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* and references therein ) . ngc5128 may thus be in the process of rebuilding a strongly warped , gaseous disk ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) , which shows signatures of spiral arms @xcite
. these properties may affect the gas accretion process onto the active smbh ( mass : @xmath11m@xmath12 ; @xcite ) .
22ghz water ( h@xmath13o ) masers have proven to be great tools to study the environment close to smbhs .
water masers are collisionally pumped ( e.g. * ? ? ?
* ) and the requisite high gas densities are typically achieved in shock fronts .
thus , h@xmath13o masers can be found in very diverse environments .
typical locations include shells of expanding asymptotic giant branch stars , expanding envelopes of young stellar objects , accretion disks of smbhs and interaction zones of radio jets with the surrounding media ( see * ? ? ?
* for a review ) .
the latter two are often referred to as ` disk masers ' ( e.g. * ? ? ?
* ; * ? ? ?
` jet masers ' ( e.g. * ? ? ?
* ; * ? ? ?
studying nearby jet and disk masers may also lead to a better understanding of smbhs in the high redshift regime . due to their brightness ,
22ghz water masers have been observed up to @xmath14 @xcite . in the past ,
cena was target of a number of unsuccessful h@xmath13omaser emission searches ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) . in the following ,
we report the first detection of a water maser toward the nuclear environment of cena .
our observations and data reduction steps are described in
[ sec : obs ] , followed by a presentation of our results in [ sec : res ] , and a discussion on the nature of the detected maser feature in [ sec : dis ] .
we summarize our findings in [ sec : sum ] .
we observed cena with the australia telescope compact array ( atca ) .
the atca is a 6 antenna interferometer with baselines between 30 m and 6 km .
the observations were performed around the rest frequency of the @xmath15 water transition at 22.23508 ghz . to exclude cross - identification of instrumental features with the water line , we slightly shifted the central frequencies between some of the observations ( see table[tab : obs ] ) .
we used the compact array broadband backend ( cabb ; * ? ? ?
* ) in two different modes : a wideband mode , with a total bandwidth of two times 1ghz and a channel width of 1mhz ( or 13 km s@xmath2 velocity resolution ; cfb1m-0.5k mode ) and cfb64m-32k , where we tuned one of the 64mhz wide zooms to water ( total velocity range of @xmath16 km s@xmath2 ; channel width of 23khz or @xmath17 km s@xmath2 ) . to reduce systematic effects , we observed at different central frequencies ( see table[tab : obs ] ) . after cropping bandpass edges
, the velocity range of the narrow band observations amounts to @xmath18 km s@xmath2 for a central frequency of 22.172ghz and @xmath19 km s@xmath2 for 22.272ghz .
absolute fluxes were calibrated with the atca standard pks1934 - 638 .
bandpasses were typically obtained by observing pks1253 - 055 for @xmath20min , but we used pks0537 - 441 when the former was not available at the start of the observations .
the observational setups and properties are summarized in table[tab : obs ] , where we list the dates of the observations in column ( 1 ) , the atca array configurations in column ( 2 ) , the central frequencies and bandwidths of the instrumental setups in columns ( 3 ) and ( 4 ) , the beam sizes and position angles in columns ( 5 ) and ( 6 ) , the continuum fluxes in column ( 7 ) , and the rms noise values , the channel separations , and the on - source integration times in columns ( 8) , ( 9 ) , and ( 10 ) , respectively .
the data were self - calibrated on the strong continuum emission from the core of cena to correct phase and gain variations as a function of time .
finally , we extracted spectra from the images , after robust weighting and cleaning 1000 iterations per channel .
all data reduction steps were performed in miriad . whereas the narrowband observations show sufficiently flat baselines , fitted well by 0th order polynomials , the wideband observations exhibit variations of order @xmath21 per cent . given the relatively high flux density of the central source ( see table[tab : obs ] ) these variations amount to a @xmath22mjy level , and we fitted and subtracted third order polynomials to a @xmath23 km s@xmath2 range around the water frequency for satisfactory results .
given the spectral variations of the baselines , we estimate @xmath24per cent uncertainty for the absolute fluxes of the well behaved narrowband observations , and a conservative @xmath20 per cent error for the wideband data .
we extracted line emission at the central position of the smbh at ra(j2000)=13@xmath2525@xmath2627@xmath276 , dec(j2000)=@xmath2843@xmath2901@xmath3009@xmath31 and subtracted the continuum emission .
the wideband spectra are shown in fig.[fig : specbroad ] and the narrowband data in fig.[fig : specnarrow ] .
the systemic velocity of cena is at a kinematic local standard of rest ( lsrk ) value of @xmath32 km s@xmath2@xcite and we mark this value in the figures , assuming that the line can be identified as the @xmath15 water maser transition at a rest frequency of 22.23508ghz ( see [ sec : id ] ) .
we clearly detect a spatially unresolved emission feature at @xmath33 km s@xmath2offset from systemic .
no detections are seen at systemic and blueshifted velocities . as the narrowband observations resulted in higher quality spectra , we used these data to constrain the line properties .
although the lines do not appear to be perfectly gaussian in shape , we nevertheless fitted one - component gaussians to these lines .
the results are listed in table[tab : gauss ] and plotted in fig.[fig : specnarrow ] .
the table contains the dates of the observations in column ( 1 ) , followed by the best fit central lsrk velocities @xmath34 , the full widths at half maximum of the fitted line @xmath35 and the peak fluxes @xmath36 in columns ( 2 ) , ( 3 ) , and ( 4 ) . in column
( 5 ) we list the derived isotropic luminosity values in @xmath37 .
the resulting parameters reveal that the h@xmath13o maser line is at a velocity of @xmath1 km s@xmath2 redshifted @xmath38 km s@xmath2 from systemic , with a width of @xmath39 km s@xmath2 .
the amplitude is @xmath40mjy with the exception of the data taken on 2012 may 22 , where we find a three times higher peak flux density of @xmath41mjy .
the continuum flux at this epoch is only about 30 per cent higher than the lowest measured continuum flux at all epochs , whereas the line is @xmath42 times stronger .
although some continuum variations could be due to the different synthesized beam sizes of the individual observations ( see table[tab : obs ] ) , this suggests that the variation is predominantly intrinsic to cena .
our observations are consistent with unresolved line emission emerging from the center of cena . for each of the 6 km array data , the beams ( see table[tab : obs ] ) correspond to physical resolutions of about @xmath43pc @xmath44pc and @xmath45pc @xmath46pc .
the position angles of the two observations are separated by @xmath47 , and we can assume that the line is emitted from within the overlap region with a conservative upper limit of a maximum of @xmath5pc projected distance away from the smbh .
we observe the line feature at a sky frequency of @xmath48mhz .
if the line would originate in cena at its systemic velocity of 545 km s@xmath2 lsrk , the line rest frequency would be expected to be close to a value of @xmath49mhz .
adding a potential velocity offset of @xmath50 km s@xmath2 ( or @xmath51mhz ) , the following astrophysical spectral lines other than the h@xmath13o @xmath15 maser transition with a rest frequency of @xmath52ghz may be candidates for the emission in cena : ch@xmath53ohcho @xmath54 at a rest frequency of 22.143mhz , c@xmath55cs @xmath56 at @xmath57ghz , and cco @xmath56 at 22.258ghz .
none of these lines are expected to be bright enough to explain the observed intensity . and none of them lie near systemic velocity requiring very unusual kinematics if they are within a star forming cloud that is at a projected distance no further than 3pc from the core of cena ( [ sec : loc ] ) .
in addition , given the strong continuum of cena , all of the thermal lines are more likely to be detectable in absorption and not emission ( cena has a complex molecular absorption spectrum as , e.g. , shown in * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
apart from the molecules , the h83@xmath58 radio recombination line has a rest frequency of 22.196ghz and could be a potential candidate for the emission . but given that we do not even see any h@xmath59 recombination lines in our spectra , we can exclude an identification with the much weaker h@xmath58 line .
the only likely line candidate is the redshifted h@xmath13o maser @xmath15 transition , which is by far the brightest astrophysical transition across the observed frequency range . as alluded to in the introduction ( [ sec : intro ] ) , h@xmath13o has also been observed toward the nuclei of other active galaxies , and the maser line is the most likely one observed in emission in the presence of cena s strong continuum emission .
the maser feature has a fwhm of @xmath39 km s@xmath2 , which we can not decompose into more narrow individual maser features down to our @xmath60 km s@xmath2 velocity resolution .
we may thus assume that the line emerges from a single maser spot unless high resolution imaging will break it up into more components .
the isotropic luminosity of the feature is @xmath6@xmath37 ( see table[tab : gauss ] ) .
if the maser line is emitted from a single spot , it falls into the category of kilomasers .
masers of similar strengths are found in nearby , moderately star forming galaxies and mergers such as m51 , ngc253 , m82 , ngc2146 , ngc3556 , ngc3620 or the antennae @xcite .
it is also similar to the galaxy s most luminous water maser in w49n ( e.g. * ? ? ?
. in all but one epoch the peak of the maser feature is about 20mjy .
the 2012 may 22 data , however , reveal a three times stronger line ( see [ sec : res ] ) . the variability may be related to a merely @xmath61per cent stronger continuum flux at the time , which may indicate that the maser itself is unsaturated . to determine this
conclusively , however , requires more epochs of observations , desirably at higher spatial resolution .
most kilomasers are found in star formation regions of galaxies ( see references above and , e.g. * ? ? ?
they typically arise from outflows of massive stars and show velocity spreads of up to a few tens of km s@xmath2 . for cena
, we can pinpoint the maser to within @xmath5pc projected distance from the nucleus ( [ sec : loc ] ) .
we consider the probability of crossing a massive star formation region along the line of sight to be low . even if such a region is located toward that direction , the velocity offset of @xmath62 km s@xmath2 from systemic and the linewidth of @xmath39 km s@xmath2 would be inconsistent with a typical h@xmath13o maser in a star forming region well outside the nuclear region .
strong water masers are also observed toward the accretion disks of some smbhs ( `` disk masers '' ) and close to agn jet - ism interaction regions ( `` jet masers '' ) .
typical examples for disk masers are the archetypical ngc4258 galaxy ( e.g. * ? ? ?
* ; * ? ? ?
* ) and a larger sample of such masers have been observed in dedicated search campaigns ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
disk masers typically have three main components : one at systemic velocity as well as blueshifted and redshifted line features , both offset by a few hundred km s@xmath2 .
some disks , however , lack one or two of these main components . at high velocity resolution ( see , e.g. * ? ? ?
* ) the disk maser components show significant substructures at km s@xmath2linewidth scales .
they can overlap and blend to form broader components .
although we may see some indications of substructure in the wideband data that was observed in 2011 ( see fig.[fig : specbroad ] ) , the overall h@xmath13o spectrum in cena is relatively smooth , even at the km s@xmath2 scales of our high spectral resolution data ( fig.[fig : specnarrow ] ) .
we thus consider a disk maser origin unlikely .
jet masers , have been detected in ngc1068 @xcite , ngc3079 @xcite , circinus @xcite , mrk348 @xcite and possibly in ngc1052 @xcite . in particular , the jet maser in mrk348 resembles cena s maser in many respects .
the line is @xmath63 km s@xmath2 redshifted from systemic , 130 km s@xmath2 wide and relatively smooth .
very long baseline array data shows that the maser is about @xmath60pc away from the core .
monitoring also shows some significant variability on timescales as short as a day and loosely correlated with the continuum strength @xcite .
the isotropic luminosity of the mrk348 maser , however , is about two orders of magnitude higher than the feature in cena .
we also observe similarities with the possible jet maser in ngc1052 .
the h@xmath13o line in ngc1052 is redshifted @xmath62 km s@xmath2 from systemic with a linewidth of @xmath64 km s@xmath2 .
its isotropic luminosity is of the same order as the maser in mrk348 .
given the similarities of the cenah@xmath13o maser emission with the above cases , we discuss the implications of the jet maser model for cena in the following section .
a useful model for a jet maser has been established by @xcite .
they propose that the jet plows into a neighboring molecular cloud and creates a shock front that is propagating away from the jet direction .
the tip of the shock front covers a large opening angle , and the morphology resembles that of a mushroom , with cloud material being compressed toward the front and the the sides of the jet .
the projected velocities of the shocked material may thus combine to the large linewidth that we witness .
shocks propagate according to the density contrast between the the two media at the shock boundary via @xmath65 , where @xmath66 and @xmath67 are the density and velocity of the initial , shocking material and @xmath68 and @xmath69 of the propagating forward shock front in the material that was hit .
@xcite monitored the motions of the parsec - scale jet components in cena and derived velocities of @xmath70c . for a shock that propagates with
the velocity offset of the maser feature in cena , @xmath71 km s@xmath2 , we can then derive a density contrast of @xmath72 .
interstellar h@xmath13o
masers require volume densities of at least @xmath73@xmath9@xcite to be pumped ( see * ? ? ? * for scenarios when radiative pumping may become important ) .
when we apply the density contrast to this number we can infer an incoming jet density of @xmath74@xmath9 .
geometry constraints will change the numbers but the order of magnitude estimate should remain if the shock front has a large opening angle as suggested by @xcite . the jet velocity , however , may be underestimated .
@xcite shows that the northern jet is much brighter than the southern counterpart , which can be explained by doppler boosting the brightness of the approaching jet .
such an effect would be expected at much higher velocities than the @xmath75 measured for the individual components . for
the limit of a jet velocity at light speed the above calculations will reduce the lower limit of the jet density by two orders of magnitude and we thus estimate a more conservative @xmath76@xmath9 .
but even such a reduced value is much larger than typical electron - positron or thermal gas plasma densities expected for relativistic jets .
the radio galaxy 3c120 , for example , has densities @xmath77@xmath9@xcite , which is four orders of magnitude lower than our limit for the jet of cena .
the difference can potentially be explained by surrounding material that is evaporated by and entrained into the jet during its outward propagation .
@xcite find indications for the appearance of narrow components in the otherwise broad maser spectrum of ngc1052 .
these components show flux variabilities and line narrowing on timescales of days .
our 2011 atca broadband h@xmath13o spectra of cena show indications for similar , transient narrow features . in the jet maser model
such variations could be due to the jet shocking small , high density substructures of the molecular cloud .
the maser properties could be the response to the shock dynamics that includes local density enhancements competing with h@xmath13o shock dissociation .
the cena maser line is redshifted to velocities larger than systemic .
the receding , fainter jet may thus be a natural location for the origin of the jet maser feature .
we note , however , that very long baseline array data show that the similarly redshifted h@xmath13o maser in mrk348 originates from the approaching , brighter jet component of this galaxy .
the full location and geometry of the maser spots in cena are therefore uncertain until higher resolution imaging will be available .
we report the detection of a 22ghz @xmath15 water maser emission line toward the inner 3pc of the most nearby radio galaxy centaurusa .
the weak and slightly variable maser feature can be classified as a kilomaser and could have its origin in a star formation region , a nuclear disk around the central supermassive black hole , or from material that is shocked close to the jet base of cena .
given the relative smoothness and width of the line , as well as its central location , we conclude that the `` jet maser '' scenario is the most likely one . in this case , the radio jet is shocking a central molecular cloud . from the shock properties , a lower limit to the density of the jet of @xmath78@xmath9 can be derived . a full characterization of the nature and properties of the maser will require higher spatial resolution .
the discovery of a nuclear water maser in cena opens up further opportunities to probe the extreme environments very close to a smbh .
we thank the australia telescope national facility staff for their assistance , and gratefully acknowledge the allocation of some director s time to confirm this feature .
this research has made use of nasa s astrophysics data system and the nasa / ipac extragalactic database ( ned ) which is operated by the jet propulsion laboratory , california institute of technology , under contract with the national aeronautics and space administration .
auld , r. , smith , m. w. l. , bendo , g. , et al .
2012 , , 420 , 1882 barnes , j. e. 2002 , , 333 , 481 baudry , a. , & brouillet , n. 1996 , , 316 , 188 bennert , n. , barvainis , r. , henkel , c. , & antonucci , r. 2009 , , 695 , 276 braatz , j. a. , wilson , a. s. , & henkel , c. 1996 , , 106 , 51 braatz , j. a. , henkel , c. , greenhill , l. j. , moran , j. m. , & wilson , a. s. 2004 , , 617 , l29 braatz , j. a. , & gugliucci , n. e. 2008 , , 678 , 96 brogan , c. , johnson , k. , & darling , j. 2010 , , 716 , l51 brunthaler , a. , castangia , p. , tarchi , a. , et al .
2009 , , 497 , 103 cappellari , m. , neumayer , n. , reunanen , j. , et al . 2009 , , 394 , 660 castangia , p. , tarchi , a. , henkel , c. , & menten , k. m. 2008 , , 479 , 111 castangia , p. , impellizzeri , c. m. v. , mckean , j. p. , et al .
2011 , , 529 , a150 claussen , m. j. , heiligman , g. m. , & lo , k. y. 1984 , , 310 , 298 claussen , m. j. , & lo , k .- y .
1986 , , 308 , 592 claussen , m. j. , diamond , p. j. , braatz , j. a. , wilson , a. s. , & henkel , c. 1998 , , 500 , l129 eckart , a. , cameron , m. , genzel , r. , et al . 1990 , , 365 , 522 espada , d. , peck , a. b. , matsushita , s. , et al .
2010 , , 720 , 666 espada , d. , matsushita , s. , peck , a. b. , et al .
2012 , , 756 , l10 gallimore , j. f. , baum , s. a. , odea , c. p. , brinks , e. , & pedlar , a. 1996 , , 462 , 740 greenhill , l. j. , jiang , d. r. , moran , j. m. , et al .
1995 , , 440 , 619 greenhill , l. j. , booth , r. s. , ellingsen , s. p. , et al . 2003 , , 590 , 162 harris , g. l. h. , rejkuba , m. , & harris , w. e. 2010 , , 27 , 457 henkel , c. , tarchi , a. , menten , k. m. , & peck , a. b. 2004 , , 414 , 117 henkel , c. , peck , a. b. , tarchi , a. , et al .
2005 , , 436 , 75 ho , p. t. p. , martin , r. n. , henkel , c. , & turner , j. l. 1987 , , 320 , 663 impellizzeri , c. m. v. , mckean , j. p. , castangia , p. , et al .
2008 , , 456 , 927 israel , f. p. 1992
, , 265 , 487 israel , f. p. 1998
, , 8 , 237 kameno , s. , nakai , n. , sawada - satoh , s. , sato , n. , & haba , a. 2005 , , 620 , 145 kondratko , p. t. , greenhill , l. j. , moran , j. m. , et al .
2006 , , 638 , 100 knig , s. , eckart , a. , henkel , c. , & garca - marn , m. 2012 , , 420 , 2263 kylafis , n. d. , & norman , c. a. 1991 , , 373 , 525 lo , k. y. 2005 , , 43 , 625 muller , s. , & dinh - v - trung 2009 , , 696 , 176 nakai , n. , & kasuga , t. 1988 , , 40 , 139 neumayer , n. , cappellari , m. , reunanen , j. , et al .
2007 , , 671 , 1329 peck , a. b. , henkel , c. , ulvestad , j. s. , et al .
2003 , , 590 , 149 quillen , a. c. , brookes , m. h. , keene , j. , et al .
2006 , , 645 , 1092 quillen , a. c. , neumayer , n. , oosterloo , t. , & espada , d. 2010 , , 27 , 396 saviane , i. , & jerjen , h. 2007 , , 133 , 1756 struve , c. , oosterloo , t. a. , morganti , r. , & saripalli , l. 2010 , , 515 , a67 surcis , g. , tarchi , a. , henkel , c. , et al .
2009 , , 502 , 529 tarchi , a. , henkel , c. , peck , a. b. , & menten , k. m. 2002 , , 389 , l39 tingay , s. j. , jauncey , d. l. , reynolds , j. e. , et al .
1998 , , 115 , 960 trotter , a. s. , greenhill , l. j. , moran , j. m. , et al .
1998 , , 495 , 740 walker , r. c. , matsakis , d. n. , & garcia - barreto , j. a. 1982 , , 255 , 128 walker , r. c. , benson , j. m. , & unwin , s. c. 1987 , , 316 , 546 wiklind , t. , & combes , f. 1997 , , 324 , 51 wilson , w. e. , ferris , r. h. , axtens , p. , et al .
2011 , , 416 , 832 yates , j. a. , humphreys , e. m. l. , & richards , a. m. s. 1997 , , 251 , 285 llcccccccc 2011 july 16 & h214 & 22.450 & 1000 & [email protected] & 82.3 & @xmath80 & 10 & 13 & 8 + 2011 october 1 & h75 & 22.450 & 1000 & [email protected] & 87.9 & @xmath81 & 5 & 13 & 9 + 2012 march 27 & h168 & 22.494 & 1000 & [email protected] & 82.7 & @xmath82 & 4 & 13 & 8 + 2012 march 29 & h168 & 22.172 & 64 & [email protected] & 77.5 & @xmath83 & 3 & 0.4 & 10 + 2012 may 22 & 6d & 22.172 & 64 & [email protected] & 18 & @xmath84 & 6 & 0.4 & 8 + 2013 february 21 & 6a & 22.162 & 64 & [email protected] & -26.5 & @xmath85 & 15 & 0.4 & 5.5 | we report the detection of a 22ghz water maser line in the nearest ( @xmath0mpc ) radio galaxy centaurusa using the australia telescope compact array ( atca ) .
the line is centered at a velocity of @xmath1 km s@xmath2 , which is redshifted by about @xmath3 km s@xmath2 from the systemic velocity .
such an offset , as well as the width of @xmath4 km s@xmath2 , could be consistent with either a nuclear maser arising from an accretion disk of the central supermassive black hole , or for a jet maser that is emitted from the material that is shocked near the base of the jet in centaurusa .
the best spatial resolution of our atca data constrains the origin of the maser feature within @xmath5pc from the supermassive black hole .
the maser exhibits a luminosity of @xmath6l@xmath7 , which classifies it as a kilomaser , and appears to be variable on timescales of months .
a kilomaser can also be emitted by shocked gas in star forming regions .
given the small projected distance from the core , the large offset from systemic velocity , as well as the smoothness of the line feature , we conclude that a jet maser line emitted by shocked gas around the base of the agn is the most likely explanation .
for this scenario we can infer a minimum density of the radio jet of @xmath8@xmath9 , which indicates substantial mass entrainment of surrounding gas into the propagating jet material . |
the gamma model of parallel programming was introduced by jean - pierre bentre and daniel le mtayer in @xcite ( see also @xcite for a more accessible account of the model and @xcite for a recent account of it ; also see @xcite for a thorough presentation of the field of multiset processing ) . at the time of its introduction ,
parallel programming as a mental activity was considered more difficult ( one might also say cumbersome ) than sequential programming , something that even today it is still valid to a certain degree .
bentre and le mtayer designed gamma in order to ease the design and specification of parallel algorithms .
thus , making a parallel programming task easier compared to previously available approaches .
gamma was inspired by the chemical reaction model . according to this metaphor
, the state of a system is like a chemical solution in which _ molecules _ , that is , processes , can interact with each other according to _ reaction rules _ , while all possible contacts are possible since a magic hand stirs the solution continuously . in gamma solutions
are represented by
_ multisets _
( i.e. , an extension of sets that is based on the assumption that elements may occur more than one time , see @xcite for more details ) .
processes are elements of a multiset and the reaction rules are multiset rewriting rules .
thus , gamma is a formalism for multiset processing .
the chemical abstract machine ( or cham for short ) is a model of concurrent computation developed by grard berry and grard boudol @xcite .
the cham is based on the gamma model and was designed as a tool for the description of concurrent systems . basically , each cham is a chemical solution in which floating molecules can interact with each other according to a set of reaction rules .
in addition , a magical mechanism stirs the solution so as to allow possible contacts between molecules .
as is evident , a number of interactions happen concurrently .
informally , a process is a program in execution , which is completely characterized by the value of the _ program counter _ ,
the contents of the processor s registers and the process _ stack _ ( see @xcite for an overview ) . naturally , two or more processes may be initiated from the same program ( e.g. , think of a web browser like firefox ) . in general ,
such processes are considered as distinct execution sequences .
although , they have the same text section , the data sections will vary .
interestingly , one can view the totality of processes that run in a system at any moment as a multiset that evolves over time , which is built _ on - the - fly _ and may be viewed as the recorded history of a system .
however , there is a problem that is not covered by this scheme and which i plan to tackle here : the processes are not identical , but _
similar _ and so i need a more expressive mathematical formalism to describe running processes .
fuzzy set theory was introduced by lotfi asker zadef in 1965 ( see @xcite for a thorough presentation of the theory of fuzzy sets ) .
fuzzy set theory is based on the denial of the idea that an element either belongs or does not belong to some set .
instead , it has introduced the notion of a _ membership degree _
@xmath1 according to which an element belongs to a fuzzy subset of an ordinary or _
crisp _ set .
usually , this degree is a real number between zero and one ( i.e. , @xmath2 ) .
for example , i may say that the degree to which @xmath3 belongs to the fuzzy subset @xmath4 is @xmath5 . according to fuzzy set theory
, when an element belongs to a fuzzy subset with membership degree equal to zero , this does not imply that the element does not belong to a set . on the contrary , it simply means that the element belongs to the fuzzy subset with degree that is equal to zero .
formally , given a universe @xmath6 a fuzzy subset @xmath4 is characterized by a function @xmath7 $ ] , where @xmath8 means that the membership degree of @xmath3 is @xmath1 .
[ [ plan - of - the - paper ] ] plan of the paper + + + + + + + + + + + + + + + + + i start by giving a `` formal '' definition of chams .
next , i argue why processes should be viewed as fuzzy entities and then i present a model of fuzzy processes that culminates to a brief introduction of fuzzy @xmath6-machines .
the fuzzy chemical abstract machine is introduced next and this discussion is followed by an exercise in defining a fuzzy version of the @xmath0-calculus from the description of a fuzzified chemical abstract machine that originally described the @xmath0-calculus .
the ingredients of a cham are _ molecules _ @xmath9 that flow in _ solutions
_ @xmath10 .
solutions change according to a number of _ transformation rules _ , which determine a _ transformation relation _ @xmath11 .
a cham is fully specified by defining its molecules , its solutions , and its transformation rules .
there are two types of transformation rules : general rules , applicable to any cham , and local rules , which are used to specify particular chams .
molecules are terms of algebras associated with specific operations for each cham .
solutions are multisets of molecules @xmath12 $ ] . by applying the _ airlock _ constructor``@xmath13 '' to some molecule @xmath14 and a solution @xmath15 , one gets the new molecule @xmath16 .
a specific rule has the following form @xmath17 where @xmath18 and @xmath19 are molecules . also , there are four general transformation rules : 1 . the _ reaction _ rule where an instance of the right - hand side of a rule replace the corresponding instance of its left - hand side . in particular , if there is a rule @xmath20 and if the @xmath21s and @xmath22s are instances of the @xmath18s and the @xmath23s by a common substitution , then @xmath24\rightarrow[m_{1}^{\prime } , m_{2}^{\prime},\ldots , m_{l}^{\prime}].\ ] ] 2 .
the _ chemical _ rule specifies that reactions can be performed freely within any solution : @xmath25 where @xmath26 is the multiset sum operator .
3 . according to the _ membrane _ rule sub - solutions can freely evolve in any context : @xmath27\rightarrow[c(s')]},\ ] ] where @xmath28 is molecule with a hole @xmath29 in which another molecule is placed .
note that solutions are treated as megamolecules .
the _ airlock _ rule has the following form : @xmath30\uplus s\leftrightarrow [ m \lhd s].\ ] ]
at any moment any operating system is executing a number of different processes . quite naturally , many different processes are instances of the same program . for example , in a multi - user environment , different users may run different instances of the same web browser .
cases like this can be naturally modeled by _
multisets _ , which from a generalization of sets .
any multiset may include more than one copy of an element .
typically , a multiset @xmath31 is characterized by a function @xmath32 , where @xmath33 is the set of natural numbers including zero and @xmath34 means that multiset @xmath31 contains @xmath35 copies of @xmath3 .
one may say that the elements of a multiset are tokens of different types .
thus , a multiset consisting of two copies of `` a '' and three copies of `` b '' can be viewed as a structure that consists of two tokens of type `` a '' and three tokens of type `` b '' ( see @xcite for a thorough discussion of this idea ) .
when dealing with an operating system , one may argue that types represent the various programs available to users and tokens represent the various processes corresponding to these programs .
although it does make sense to view programs as types and processes as tokens , still not all tokens are identical .
for example , different people that use a particular web browser view different web pages and have different numbers of tabs open at any given moment .
thus , we can not actually talk about processes that are identical , but we can surely talk about processes that are similar .
the next question that needs to be answered is : how can we compare processes ?
an easy solution is to define _ archetypal _ processes and then to compare each process with its corresponding archetypal process .
the outcome of this procedure will be the estimation of a similarity degree .
but what exactly is an archetypal process ?
clearly , there is no single answer to this question , but roughly one could define it as follows : assume that @xmath36 is an executable program for some computational platform ( operating system , cpu , etc . ) , then an archetypal process @xmath0 of @xmath36 is the program @xmath36 running with minimum resources . here the term _ minimum resources
_ means that the archetypal process consumes the least possible resources ( e.g. , memory , cpu time , etc . ) .
let @xmath37 be some web browser , then an archetypal process of @xmath37 is the web browser which when starts shows a blank page in only one tab / window .
similarly , an archetypal process for the unix ` ls ` command is the command running without arguments in an empty directory .
assume that @xmath38 and @xmath39 are two processes initiated from the same binary @xmath36 .
assume also that @xmath0 is an archetypal process and that @xmath40 is a method that measures the similarity degree of some process @xmath41 to @xmath0 .
in different words , @xmath42 means that @xmath36 is similar with @xmath0 to a degree equal to @xmath1 .
if @xmath43 denotes degree to which the two processes @xmath44 and @xmath45 are similar , then @xmath46 suppose that one has selected a number of criteria to choose and specify archetypal processes .
let @xmath47 denote the set of all possible archetypal processes for a particular system @xmath48 and a particular set of criteria .
without loss of generality , we can think that the elements of @xmath47 are the names of all programs that can possibly be executed on a system @xmath48 .
for instance , for some typical unix system @xmath49 , the set @xmath50 may contain the names of programs under /usr / bin , /usr / local / bin , /opt / sfw / bin , etc .
suppose that @xmath51 and that at some moment @xmath52 , @xmath53 are processes initiated from program @xmath0 .
then this situation can naturally be modelled by fuzzy multisets , that is , multisets whose elements belong to the set to some degree .
a fuzzy multiset is characterized by a higher - order function @xmath54\rightarrow\mathbb{n})$ ] .
fuzzy multisets were introduced by ronald r. yager @xcite . by uncurrying the functional @xmath55
, we get a function @xmath56\rightarrow\mathbb{n}$ ] . thus , in general , at any given moment the processes running in a system can be described by a multiset of pairs @xmath57 , where @xmath58 denotes the membership degree . however , such a structure reflects what is happening in a system a given moment .
thus , to describe what is going on in a system at some time interval we need a structure that can reflect changes as time passes .
the most natural choice that can solve this problem is a form of a _ set through time_. bill lawvere was the first to suggest that sheaves can be viewed as continuously varying sets ( see @xcite and @xcite for a detailed account of this idea ) . since in this particular case
i am interested in _ fuzzy multisets continuously varying through time _
, it seems that sheaves are not the structures i am looking for .
however , as i will show this is not true .
but before i proceed , it is more than necessary to give the general definition of a sheaf .
the definition that follows has been borrowed from @xcite ( readers not familiar with basic topological notions should consult any introductory textbook , e.g. , see
@xcite ) : let @xmath6 be a topological space and @xmath59 its collection of open sets . a sheaf over @xmath6 is a pair @xmath60 where @xmath61 is a topological space and @xmath62 is a local homeomorphism ( i.e. , each @xmath63 has an open neighborhood @xmath64 in @xmath61 that is mapped homeomorphically by @xmath36 onto @xmath65 , and the later is open in @xmath6 ) .
let us now construct a fuzzy multiset through time .
suppose that @xmath7\rightarrow\mathbb{n}$ ] characterizes some fuzzy multiset .
clearly , the function @xmath66\rightarrow\mathbb{n}$ ] also characterizes the same fuzzy multiset . and if this function characterizes some fuzzy multiset , then it is absolutely reasonable to say that the graph of this function characterizes the same fuzzy multiset .
let @xmath67 , @xmath68 , where @xmath69 is a set of indices , be the graphs of all functions @xmath70 that characterize fuzzy multisets .
in addition , assume that each @xmath67 is an open set of a topological space @xmath71 .
obviously , it is not difficult to define such a topological space .
for example , it is possible to define a metric between points @xmath72 and @xmath73 and from this to define a metric topology .
having defined a topology on @xmath74)\times\mathbb{n}$ ] , it is straightforward to define a sheaf over @xmath71 . in particular ,
if @xmath33 denotes the order topology on the set of natural numbers , then @xmath75 , where @xmath76 is a local homeomorphism , is a sheaf over @xmath71 . in general
, such a sheaf characterizes a fuzzy multiset through _ discrete _ time . clearly , this is not the only sheaf over @xmath71 one can define .
in fact , one can build a category @xmath77 with objects all sheaves over @xmath71 and with arrows @xmath78 the continuous maps @xmath79 such that @xmath80{b}\arrow[8]{se , l}{q}\\[10 ] \node{a}\arrow[7]{ne , l}{k}\arrow[15]{e , b}{p}\node[18]{\mathscr{x } } \end{diagram}\ ] ] commutes .
in general , the sheaf @xmath81 , where @xmath82 is the set of the positive real numbers including zero and here denotes also the order topology on this set and @xmath83 is a local homeomorphism , characterizes a fuzzy multiset through _ continuous _ time .
it is not difficult to define a monic arrow @xmath84 and , thus , to show that @xmath85 belongs to @xmath86 , that is , the collection of sub - objects of @xmath87 .
thus , we have an implicit proof of the following statement : sheaf @xmath87 contains more information than @xmath88 . to put it differently ,
no discrete system can fully simulate a continuous system .
it should be obvious , that in the most general case one can not know beforehand all the components of a sheaf representing a fuzzy multiset through time .
when such a structure is used to represent processes , then it is _ noncomputable _ , since one can not construct it using some algorithm as it is not possible to foresee what the users will do .
a model of computation is a precise description of a conceptual , real , or abstract computational device , which can be used to implement certain computational tasks .
consequently , there are tasks that are implementable and tasks that are not feasible . by broadening the notion of computation ( e.g. , by augmenting the computational machinery with external agents ) , it is possible to extend a model of computation .
for example , if one assumes that things happen randomly , then it makes sense to introduce some form of randomness in our computation ( e.g. , think of nondeterministic turing machines ) .
thus , it was more than expected to see the emergence of extended variants of the cham . indeed , a. di pierro , c. hankin , and h. wiklicky introduced in @xcite a probabilistic version of cham .
in this model , multisets and mutliset rewriting rules are associated with probabilities so as to allow a nondeterministic computation . in general ,
nondeterministic conceptual computing devices are not more powerful than their deterministic counterparts .
thus , from the point of view of computability theory such models are not that interesting . on the other hand
, fuzziness seems to be a more promising and interesting direction ( e.g. , see @xcite ) and so models of computation incorporating fuzziness are more interesting and , at the same time , seem to be closer to what happens in reality .
a cham is defined by specifying the molecule syntax and a set of molecule reaction rules .
the molecule syntax describes some algebra and the transformation rules are actually multiset rewriting rules that model transitions in a system .
thus , in order to define a fuzzy cham , or just fucham for simplicity , it is necessary to review known notions from fuzzy algebraic theory and to introduce some notion related to fuzzy labeled transition systems .
[ [ fuzzy - algebras ] ] fuzzy algebras + + + + + + + + + + + + + + back in 1971 , fuzzy groups were introduced by azriel rosenfled @xcite as a natural extension of groups : assume that @xmath89 is a group and that @xmath7 $ ] is a fuzzy subset of @xmath6 , then @xmath4 is a fuzzy subgroup of @xmath6 if @xmath90 by replacing @xmath91 with some other fuzzy @xmath52-norm , one gets a more general structure called a @xmath52-fuzzy group .
also , by following this line of thought it is possible to define more complex fuzzy algebraic structures ( e.g. , fuzzy rings , etc . ) .
but , this is not the only way to fuzzify a group .
indeed , m. demirci and j. recasens proposed in @xcite a fuzzy subgroup to be a group @xmath92 , where @xmath93 denotes the degree to which @xmath94 .
an even more fuzzier algebraic structure is one were both the elements belongs to the underlying set to a degree and the result of the operation is the `` real '' result up to some degree .
assume that @xmath92 is a group and that @xmath7 $ ] is a fuzzy subset of @xmath6 , then @xmath4 is a fuzzy subgroup of @xmath6 if @xmath95 of course when one talks about processes , it does not make sense to speak about real algebraic operations .
for example , although the notion of the inverse of some program @xmath96 has been discussed thoroughly in @xcite , there is no general technique to get the inverse of some program .
fortunately , one can talk about monoidal operations since we can find some identity process .
given two processes @xmath36 and @xmath97 , the expressions @xmath98 and @xmath99 denote a nondeterministic choice between @xmath36 and @xmath97 and the ( parallel ) composition of @xmath36 and @xmath97 , respectively .
in particular , @xmath98 behaves either like @xmath36 or like @xmath97 . in many case , the environment favors one particular alternative , while in others it may favor the other alternative . and
it is this remark that justifies the introduction of fuzzy set theory to model cases like this one .
the expression @xmath99 can be viewed as a system in which @xmath36 and @xmath97 may proceed independently but may also interact in some prescribed way .
as time passes by , the system may evolve but its evolution will depend on the degree @xmath36 and/or @xmath97 can evolve to @xmath100 and/or @xmath101 , respectively .
of course , we can introduce fuzziness by specifying to what degree two processes may proceed independently . [ [ fuzzy - labeled - transition - systems ] ] fuzzy labeled transition systems + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + let us now turn our attention of fuzzy labeled transition systems .
the ideas developed below are generalizations of ideas and results presented in @xcite .
recall that a fuzzy binary relation @xmath102 from a set @xmath49 to a set @xmath61 is a fuzzy subset of @xmath103 ( i.e. , @xmath102 is characterized by a function @xmath104 $ ] ) and it is denoted by @xmath105 .
similarly , one can define fuzzy @xmath35-ary relations . a fuzzy labeled transition system ( flts ) over a crisp set of actions
@xmath106 is a pair @xmath107 consisting of * a set @xmath108 of states ; * a fuzzy relation @xmath109 called the _
fuzzy transition relation_. if the membership degree of @xmath110 is @xmath111 , `` indicates the strength of membership within the relation '' @xcite . ]
we write @xmath112 to denote that the plausibility degree to go from state @xmath97 to state @xmath101 by action @xmath113 is @xmath111 . more generally , if @xmath114
, then @xmath115 is called a _ derivative _ of @xmath97 with plausibility degree equal to @xmath116 . as in the crisp case
, it is quite reasonable to ask when two states in a flts should be considered equivalent to some degree . for example , what can be said about the two fltses depicted in figure [ similar : fltses ] ? in order to be able to answer this question we need to define the notion of similarity between fltses .
lr ( 150,130)(-50,50 ) ( 20,100 ) ( 15,100)@xmath117 ( 31,100)(1,0)30 ( 40,105)a ( 35,90)0.50 ( 70,100 ) ( 65,100)@xmath44 ( 130,130 ) ( 125,130)@xmath45 ( 81,100)(4,3)40 ( 95,120)b ( 105,105)0.75 ( 130,70 ) ( 125,70)@xmath118 ( 81,100)(5,-4)40 ( 95,95)c ( 80,75)0.80 & ( 150,130)(-50,50 ) ( 20,100 ) ( 15,100)@xmath119 ( 70,130 ) ( 65,130)@xmath120 ( 31,100)(4,3)30 ( 45,120)a ( 55,105)0.65 ( 70,70 ) ( 65,70)@xmath121 ( 31,100)(5,-4)30 ( 45,95)a ( 30,75)0.55 ( 120,130 ) ( 115,130)@xmath122 ( 80,130)(1,0)30 ( 90,135)b ( 85,120)0.80 ( 120,70 ) ( 115,70)@xmath123 ( 80,70)(1,0)30 ( 90,75)c ( 85,60)0.80 assume that @xmath107 is an flts and that @xmath124 is a fuzzy binary relation over @xmath108
. then @xmath124 is called a _ strong fuzzy simulation over @xmath107 with simulation degree @xmath125 _ , denoted by @xmath126 if , whenever @xmath127 if @xmath128 , then there exists @xmath129 such that @xmath130 , @xmath131 , and @xmath132 .
we say that @xmath97 _ strongly fuzzily simulates _
@xmath36 with degree @xmath133 $ ] , if there exists a strong fuzzy simulation @xmath126 such that @xmath134 . in order to make the notion just defined more clear , i will present an example of strong fuzzy simulation .
let us consider the two fltses depicted in figure [ similar : fltses ] and the fuzzy binary relation described by the following table .
@xmath135 this fuzzy relation is a strong fuzzy simulation and therefore @xmath117 strongly fuzzily simulates @xmath119 . to verify
this one needs to examine each transition @xmath136 for every pair @xmath137 and show that it is matched by some transition @xmath138 .
for example , consider the pair @xmath139 .
state @xmath121 has one transition @xmath140 which is matched by @xmath141 because @xmath142 and @xmath143 .
therefore , @xmath121 strongly fuzzily simulates @xmath44 with degree 0.80 .
[ sfbs:1 ] a fuzzy binary relation @xmath144 is said to be a strong fuzzy bisimulation over the flts @xmath107 with simulation degree @xmath125 if both @xmath124 and @xmath145 ( i.e. , the inverse of @xmath124 ) are strong fuzzy simulations . any @xmath36 and @xmath97 are _ strongly fuzzily bisimilar with degree _
@xmath111 or _ strongly fuzzily equivalent to degree _
$ ] , @xmath146 , if there is a strong fuzzy bisimulation @xmath126 such that @xmath134 .
some strong fuzzy bisimulations are constructed by others more simple ones as the proof of the following propositions shows .
[ sfb:2 ] suppose that each @xmath147 , @xmath148 , is a strong fuzzy bisimulation with simulation degree @xmath149 .
then the following fuzzy relations are all strong fuzzy bisimulations : @xmath150 1 . for the identity relation @xmath151 it holds that @xmath152 and @xmath153 for all @xmath154 and where @xmath155 .
in addition , it holds that @xmath156 , which trivially shows that @xmath157 is a strong fuzzy bisimulation .
2 . in order to show that @xmath158 is a strong fuzzy bisimulation we need to show that the inverse of @xmath145 ( i.e. , @xmath124 ) is a strong fuzzy bisimulation , which is obvious from definition [ sfbs:1 ] .
3 . recall that if @xmath159 and @xmath160 are two strong fuzzy bisimulations with similarity degrees @xmath161 and @xmath162 , respectively , then @xmath163 is a new fuzzy binary relation such that @xmath164.\ ] ] obviously , @xmath165 , which shows exactly what we were looking for .
assume that @xmath166 and @xmath167 are two strong fuzzy bisimulations .
then their union as fuzzy binary relations is defined as follows .
@xmath168.\ ] ] from this it not difficult to see that @xmath169,\ ] ] proves the simplest case . from this
it is not difficult to see why the general case holds .
the following statement reveals some properties of the strong fuzzy bisimulation . 1 .
@xmath170 is an equivalence relation ; 2 .
@xmath170 is a strong fuzzy bisimulation .
1 . recall that a fuzzy binary relation @xmath171 is a _ fuzzy equivalence relation _ if it is reflexive , symmetric , and transitive . for reflexivity
, it is enough to consider the identity relation @xmath151 , which is a strong fuzzy bisimulation . for symmetry
it is enough to say that given a strong fuzzy bisimulation @xmath126 , its inverse @xmath158 is also a strong fuzzy bisimulation .
finally , for transitivity it is enough for say that the relational composition of two strong fuzzy bisimulations is also a strong fuzzy bisimulation .
this is a direct consequence of proposition [ sfb:2 ] .
[ [ fuzzy - x - machines ] ] fuzzy @xmath6-machines + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + @xmath6-machines are a model of computation that has been introduced by samuel eilenberg @xcite .
roughly , given an arbitrary set @xmath6 and a family of relations @xmath172 where @xmath173 , an @xmath6-machine @xmath174 of type @xmath175 is an automaton over the alphabet @xmath175 .
although a labeled transition system is not an automaton ( e.g. , there are no terminal states ) , still it is very easy to define automata using the data of a labeled transition system as a starting point .
thus , a fuzzy automaton is actually a special case of a fuzzy labeled transition system that includes a set of initial states and a set of final states . given a fuzzy automaton over an alphabet ( a set ) @xmath176 , a partial function
can be defined @xmath177 where @xmath178 is the free monoid with base @xmath176 and @xmath179 is the concatenation of strings @xmath180 and @xmath3 . note that @xmath181 is the _ left multiplication _ and it is defined as @xmath182 .
thus , @xmath183 is the inverse of the left multiplication . now , by replacing each edge @xmath184 of an automaton with an edge @xmath185 the result is a new automaton which is a fuzzy @xmath6-machine .
note that the type of such a machine is @xmath186 .
obviously , one can construct an @xmath6-machine even from an flts , but the result will not be a _ machine _ since it will not have initial and terminal states .
this view is correct when one has in mind the classical view of a machine as a conceptual device that after processing its input in a _
finite _ number of operations terminates .
interestingly , there are exceptions to this view that are widely used even today .
for example , an operating system does not cease to operate and those who stop unexpectedly are considered failures .
thus , one can assume that a machine will not terminate to deliver a result but instead it delivers ( partial ) results as it proceeds ( see @xcite for more details ) .
on the other hand if states are elements of some fuzzy subset , then we can say that there is a termination degree associated with each computation .
in other words , a computation may not completely stop or it may stop at some unexpected moment .
this is a novel idea , since the established view is that a computation must either stop or it will loop forever .
ideas like this one could be used to model the case where an external agent abruptly terminates a computation . however
, a full treatment of these and other similar ideas is out of the scope of this paper .
roughly , a fucham can be identified with a solution with fuzzy molecules and a set of fuzzy reaction rules . a solution with fuzzy molecules
can be modelled by a fuzzy multiset , while fuzzy reaction rules can be described by fuzzy transitions . before presenting a formal definition of fuchams
let us informally examine whether it makes sense to talk about solutions with fuzzy molecules and about fuzzy reaction rules . to begin with consider the following concrete chemical reaction rule : @xmath187 according to the `` traditional '' view , two hydrogen molecules react with one oxygen molecule to create two water molecules .
a fuzzy version of this reaction rule should involve fuzzy molecules and it should be associated with a plausibility degree .
this is justified by the _
fact _ that molecules of some chemical element or compound are not identical but rather similar to a degree with an ideal molecule of this particular element or compound .
in other words , not all hydrogen and oxygen molecules that react to create water are identical .
for example , think of deuterium and tritium as `` hydrogen '' molecules up to some degree that react with oxygen to produce heavy water , tritiated water and/or partially tritiated water , that is , water up to some degree .
thus , the `` water '' molecules produced when millions of `` hydrogen '' molecules react with oxygen molecules are not identical but just similar ( if , in addition , the reader considers hydrogen peroxide , that is , @xmath188 , then things will get really _ fuzzy _ ) . obviously , the higher the similarity degree , the more likely it is that the reaction will take place . and
this is the reason one must associate with each reaction rule a plausibility degree .
although these ideas may seem unnatural , still g.f .
cerofolini and p. amato @xcite have used fuzziness and linear logic to propose an axiomatic theory for general chemistry . in particular , they have developed ideas similar to ours in order to explain how chemical reaction take place , which means that my proposal , which i call the _ fuzzy chemical metaphor _ , is not unnatural at all .
the fuzzy chemical metaphor is essentially identical to the chemical metaphor , nevertheless , it assumes that molecules of the same kind are similar and not identical .
solutions of fuzzy molecules may react according to a number of fuzzy reaction rules , whereas each rule is associated with a feasibility degree that specifies how plausible it is that a particular reaction will take place .
a fucham is an extension of the ( crisp ) cham that is build around the fuzzy chemical metaphor . like its crisp counterpart
, any fucham may have up to four different kinds of transformation rules that are described below .
[ [ fuzzy - reaction - rules ] ] fuzzy reaction rules + + + + + + + + + + + + + + + + + + + + assume that we have a solution with fuzzy molecules that are supposed to react according to some fuzzy reaction rule . then the reaction will take place only when the similarity degree of each molecule is greater or equal to the feasibility of the particular reaction rule . [
feasible : rule ] assume that @xmath189 and @xmath190 are archetypal molecules .
then @xmath191 is an _ ideal _ fuzzy reaction rule with feasibility degree @xmath192 that describes how likely it is that molecules @xmath193 may react to create molecules @xmath194 .
suppose that @xmath195 is an instance of @xmath18 to degree @xmath196 , that is , the molecule @xmath21 is similar to @xmath18 with degree equal to @xmath196 .
then the following fuzzy reaction @xmath197 \overset{}{\underset{\lambda}{\rightarrow } } [ m_{1}^{\prime},\ldots , m_{l}^{\prime}]\ ] ] is _ feasible _ with feasibility degree equal to @xmath192 if @xmath198 the similarity degree of a molecule @xmath199 depends on the similarity degrees of the atoms that make up this particular molecule .
the most natural choice is to assume that it is the minimum of these degrees .
it is quite possible to have a situation where the same reacting molecules may be able to yield different molecules , something that may depend on certain factors . in different words
, we may have a solution where a number of different fuzzy reaction rules are _ potentially applicable_. in this case , the reaction rule with the highest feasibility degree is _ really applicable_. assume that @xmath49 is a solution for which the following reaction rules are potentially applicable @xmath200 and that @xmath201 , @xmath202 are the similarity degrees of the actual molecules that are contained in @xmath49 .
then the really applicable rule is the one that satisfies the conditions of definition ( [ feas : eq ] ) and whose feasibility degree is the largest among the feasibility degrees of all potentially applicable rules . using the perl pseudo - code of algorithm [ alg:1 ]
, one can compute the really applicable and the potentially applicable rules . [ [ the - fuzzy - chemical - rule ] ] the fuzzy chemical rule + + + + + + + + + + + + + + + + + + + + + + + mixing up two solutions @xmath203 and @xmath204 yields a new solution @xmath205 that contains the molecules of both @xmath203 and @xmath204 .
in other words , @xmath205 is the sum of @xmath203 and @xmath204 , or more formally @xmath206 note that in order to find the sum of two or more fuzzy multisets we work as in the crisp case ( see @xcite ) . nonetheless , because of restriction [ feas : eq ] , the fuzzy chemical rule takes the following form : @xmath207 note that this restriction applies to all other general rules . [
[ fuzzy - airlock - rule ] ] fuzzy airlock rule + + + + + + + + + + + + + + + + + + the airlock rule creates a new molecule that consists of a solution and a molecule .
therefore , one needs to define a similarity degree for solutions in order to be able to estimate the similarity degree of the new molecule .
suppose that @xmath96 is a process solution represented by a fuzzy multiset @xmath49 .
the similarity degree of @xmath96 to a solution that contains only prototype molecules is given by : @xmath208 if @xmath49 is a fuzzy solution and @xmath209 a fuzzy molecule , then @xmath210\uplus s\overset{}{\underset{\lambda}{\leftrightarrow } } [ m\lhd s]}.\ ] ] [ [ fuzzy - membrane - rule ] ] fuzzy membrane rule + + + + + + + + + + + + + + + + + + + suppose that @xmath211 denotes the similarity degree of molecule @xmath212
. then the fuzzy membrane rule is formulated as follows : @xmath213\overset{}{\underset{\lambda}{\rightarrow } } [ c(s')]}.\ ] ]
the cham has been used to describe the operations of various process calculi and algebras , which have been proposed to describe concurrency . since the cham is a special case of the fucham
, one can use the fucham to describe the operational semantics of any such formalism .
nevertheless , this is almost meaningless as there is no reason to use a hammer to hit a nail ! on the other hand , it would make sense to ( try to ) describe the operational semantics of a formalism that has been designed to describe concurrency in a vague environment .
the truth is that there is no process algebra or process calculus that is built around the fundamental notion of vagueness .
therefore , it is not possible to perform a similar exercise for the fucham . a different way to demonstrate
its expressive power is to take an existing cham , which has been designed to describe some crisp process calculus or algebra , and then to try to fuzzify the description and , consequently , to come up with a description of a ( hypothetical ? ) fuzzy process calculus or algebra .
naturally , this approach does not lead to a full - fledged theory of vague or imprecise concurrency theory , but rather it can be considered as an exercise in defining behaviors from models . for
this exercise i will use the @xmath0-calculus @xcite and the corresponding cham @xcite .
the @xmath0-calculus is a mathematical formalism especially designed for the description of _ mobile processes _ , that is , processes that live in a virtual space of linked processes with mobile links .
the @xmath0-calculus is a basic model of computation that rests upon the primitive notion of _ interaction_. it has been argued that interaction is more fundamental than reading and writing a storage medium , thus , the @xmath0-calculus is more fundamental than turing machines and the @xmath192-calculus ( see @xcite and the references herein for more details ) . in the @xmath0-calculus processes are described by process expressions that are defined by the following abstract syntax : @xmath214 if @xmath215 , then @xmath216 is the null process that does nothing .
in addition , @xmath217 denotes an _ action prefix _ that represents either sending or receiving a message , or making a silent transition : @xmath218 the expression @xmath219 behaves just like one of the @xmath220 s , depending on what messages are communicated to the composite process ; the expression @xmath221 denotes that both processes are concurrently active ; the expression @xmath222 means that the use of the message @xmath113 is restricted to the process @xmath96 ; and the expression @xmath223 means that there are infinitely many concurrently active copies of @xmath96 .
as it stands the only way to introduce fuzziness in the @xmath0-calculus is to assume that action prefixes are fuzzy .
usually , it is assumed that there is an infinite set of names @xmath88 from which the various names are drawn . in our case
, it can be assumed that names are drawn from @xmath224 $ ] . in other words ,
a name would be a pair @xmath225 which will denote that the name used is similar to the prototype @xmath3 with degree equal to @xmath1 .
skeptic readers may find this idea strange as an @xmath3 is always an @xmath3 and nothing more or less . indeed , this is true , nevertheless , if we consider various @xmath3s drawn from different ( computer ) fonts , then one @xmath3 is more @xmath3 than some other @xmath3 . to fully understand this idea ,
consider the sequence of letters in figure [ typo : sim ] borrowed from @xcite .
clearly , the rightmost symbol does not look like an a at least to the degree the second and the third from the left look like an a. i call this kind of similarity _ typographic similarity _ , for obvious reasons .
thus , one can say that names are typographically similar .
berry and boudol @xcite provide two encodings of the @xmath0-calculus , but for reasons of simplicity i will consider only one of them .
the following rules can be used to describe the functionality of the @xmath0-calculus . [ cols= "
< , < " , ] note that the first four rules are common to the two encodings of berry and boudol .
unfortunately , these rules do not describe summation .
however , one can imagine that a sum is an inactive megamolecule that changes to a simpler molecule in a single step .
of course , this is a crude idea and not a bulletproof solution , so i will not say more on the matter . in order to proceed with this exercise , it is necessary to fuzzify the encoding presented above . basically , the reaction and @xmath113-conversion rules are the most problematic rules
. a fuzzification of these rules can be obtained by attaching to each rule a plausibility degree . in the first case , it is reasonable to demand that the similarity degrees of @xmath3 and @xmath226 are the same and at the same time greater than the feasibility degree of the plausibility degree and also the difference of the similarity degrees of @xmath227 and @xmath228 is not greater than the plausibility degree . in other words ,
the reaction @xmath229,q,\ ] ] is feasible if @xmath230 and @xmath231 .
similarly , the @xmath113-conversion @xmath232\ ] ] if plausible only if @xmath233 .
because of this definition , it is necessary to define the notion of fuzzy structural congruence .
one option is to use a slightly modified version of the definition provided in @xcite .
the slight modification involves @xmath113-conversion plus @xmath234 . from the discussion
so far it should be clear that @xmath235 if and only if @xmath236 . with these redefinitions
it is not difficult to go `` back '' to a fuzzy version of the @xmath0-calculus .
i have tried to merge fuzziness with concurrency theory .
the rationale of this endeavor is that one can view processes as being similar and not just identical or completely different . in order to build a model of concurrency in a vague environment ,
i have introduced fuzzy labeled transition systems and proved some important properties . in passing ,
i have defined fuzzy @xmath6-machines and discussed some interesting ideas .
the i introduced fuchams as a model of concurrent computation in a fuzzy environment where fuzzy processes interact to perform a computation according to some fuzzy reaction rules .
the model was used to device a toy process calculus which is a fuzzy extension of the @xmath0-calculus .
the next step is to use the ideas developed here to develop real fuzzy process calculi and algebras and thus to broaden the study of concurrency .
in addition , these ideas may form the basis for developing fuzzy programming languages and maybe fuzzy computers .
i would like to thank athanassios doupas for his help during the early stages of this work .
alessandra di pierro , chris hankin , and herbert wiklicky . .
in frank s. de boer , marcello m. bonsangue , susanne graf , and willem - paul de roever , editors , _ formal methods for components and objects , 4th international symposium , fmco 2005 amsterdam , the netherlands , november 1 - 4 , 2005 revised lectures _ , number 4111 in lecture notes in computer science , pages 388407 .
springer - verlag , berlin , 2006 . | fuzzy set theory opens new vistas in computability theory and here i show this by defining a new computational metaphor the fuzzy chemical metaphor .
this metaphor is an extension of the chemical metaphor .
in particular , i introduce the idea of a state of a system as a solution of fuzzy molecules , that is molecules that are not just different but rather similar , that react according to a set of fuzzy reaction rules .
these notions become precise by introducing fuzzy labeled transition systems .
solutions of fuzzy molecules and fuzzy reaction rules are used to define the general notion of a fuzzy chemical abstract machine , which is a _ realization _ of the fuzzy chemical metaphor .
based on the idea that these machines can be used to describe the operational semantics of process calculi and algebras that include fuzziness as a fundamental property , i present a toy calculus that is a fuzzy equivalent of the @xmath0-calculus . |
with the discovery of the standard model ( sm ) higgs in the atlas @xcite and cms @xcite experiments , we have taken one step further toward understanding the electroweak symmetry breaking ( ewsb ) through the spontaneous symmetry breaking ( ssb ) mechanism in the scalar sector .
the next mission for the high - luminosity large hadron collider ( lhc ) is to explore not only the detailed properties of the sm higgs , but also the new physics effects .
since problems related to the origin of neutrino mass , dark matter ( dm ) , and matter - antimatter asymmetry can not be resolved in the sm , it is believed that the sm of particle physics is an effective theory at the electroweak scale .
if new physics exists at the tev scale , the lhc can detect it .
some potential events indicating the existence of new effects indeed have been observed in the recent atlas and cms experiments .
for instance , diboson excess of @xmath20 with @xmath21 at around 2 tev was shown by atlas @xcite and cms @xcite ; the branching ratio ( br ) for lepton - flavor - violating higgs decay @xmath22 with a @xmath23 significance was presented by cms @xcite ; a resonance at a mass of 750 gev in the diphoton invariant mass spectrum was reported by atlas @xcite and cms @xcite .
although the results are not conclusive yet , these experimental measurements have inspired theorists to speculate various effects to interpret the excesses . ever since the sm higgs was observed ,
the higgs measurements have approached to the precision level .
it becomes an important issue to uncover the physics beyond the sm through the higgs portal .
precise higgs measurements can also give strict bounds on the new couplings ; for instance , @xmath24 in the two - higgs - doublet model has been limited to be close to the decoupling limit @xcite , and the sm with a fourth generation of chiral fermions has become disfavored @xcite .
although the extension of the sm with chiral fermions has been severely limited , the constraint on the vector - like quark ( vlq ) models may not have the same situation due to the use of different representations and coupling structures . unlike chiral fermion models , where the appearance of chiral quarks has to accompany chiral leptons due to gauge anomaly , the gauge anomaly in vlq models is cancelled automatically .
therefore , it is not necessary to introduce the exotic heavy leptons into the sm when vlqs are added .
due to their interesting properties , the phenomena of some specific vlqs at the lhc have been investigated from a theoretical viewpoint @xcite . in experiments , single vlqs and pairs of vlqs
have been produced at atlas @xcite and cms @xcite .
based on the sm gauge symmetry @xmath25 , the representations of vlqs can basically be any @xmath26 multiplets .
however , in order to consider the vlq decays , the possible representations of vlq couplings to the sm quarks are singlet , doublet , and triplet . to interpret the excesses of dibosons and diphotons indicated by atlas and cms , we proposed a model that contains one higgs singlet and two triplet vlqs with hypercharges of @xmath27 and @xmath28 , respectively @xcite .
since the representations of the vlqs are different from those of the sm quarks , the higgs- and @xmath1-mediated flavor - changing neutral currents ( fcncs ) are induced at the tree level and the cabibbo - kobayashi - maskawa ( ckm ) matrix is non - unitary matrix . in our earlier studies , besides the resolutions of the excesses , we focused on the leading effects , which were from the left - handed flavor - mixing matrices , and found that they led to interesting contributions to top fcncs @xmath29 and the sm higgs production and decays . in this study , we systematically discuss the left- and right - handed flavor - mixing effects together .
we revisit the constraints and present the bounds from @xmath30 processes in detail . with the values of constrained parameters
, it is found that the modified top coupling to the sm higgs , which arises from the right - handed flavor mixing , can diminish the influence of the sm higgs production and the decay to diphotons by around @xmath2 and @xmath31 deviations from the sm results , respectively .
we demonstrate how the changes of the sm ckm matrix elements can be smeared so that the severe bounds from the current measurements of the ckm matrix elements are satisfied @xcite .
in addition to the phenomena in flavor physics , we also investigate the single and pair production of vlqs in this work . in the proposed model ,
the new quarks are @xmath32 , @xmath33 , @xmath34 , and @xmath35 , whose associated electric charges are @xmath36 , @xmath37 , @xmath38 , and @xmath39 , respectively .
therefore , @xmath32 and @xmath33 can be regarded as top and bottom partners , respectively . since vlqs @xmath34 and @xmath35 carry
the unusual electric charges , they do not have fcnc couplings to the sm quarks . as a result , their single production and decays are only through charged weak interactions .
since the top and bottom partners involve more complicated fcnc interactions , we concentrate the study on vlqs @xmath34 and @xmath35 .
it is found that the single production cross sections of @xmath34 and @xmath35 can be larger than the pair production cross sections , which are dominant from quantum chromodynamics ( qcd ) . in order to understand this phenomenon
, we analyze each process @xmath40 for the single production of @xmath34 and @xmath35 .
it is found that with @xmath41 tev , the cross sections for @xmath42 and @xmath43 modes can be of the order of 100 fb while the pair production cross sections are smaller by a factor of around 2 .
we postpone the study of event simulation to another paper .
the rest of this paper is organized as follows .
we establish the model , discuss the new flavor mixing effects , and derive the new higgs and gauge couplings in section ii .
we present the constraints from low - energy and higgs measurements in section iii .
we also study the implications on top - quark fcnc processes @xmath29 . in section iv
, we discuss the single and pair production for vlqs @xmath44 and @xmath45 , and thoroughly analyze the production mechanism in @xmath7 collisions .
the conclusions are given in section v.
we extend the sm by including one real higgs singlet and two vector - like triplet quarks ( vltqs ) , where the representations of vltqs in @xmath46 gauge symmetry are chosen as @xmath47 and @xmath48 @xcite . for suppressing the mixing between higgs singlet and doublet , we impose a @xmath49 discrete symmetry on the scalar potential , where the scalar fields follow the transformations @xmath50 and @xmath51 under the @xmath49 transformation .
thus , the scalar potential is expressed as : @xmath52 we adopt the following representation of @xmath53 : h= ( cc g^+ + ( v+ h + ig^0 ) ) , where @xmath54 and @xmath55 are goldstone bosons , @xmath56 is the sm higgs field , and @xmath57 is the vacuum expectation value ( vev ) of @xmath53 . the @xmath58 field can not develop a non - vanishing vev in the scalar potential of eq .
( [ eq : vhs ] ) when @xmath59 . due to the @xmath49 symmetry
, @xmath56 and @xmath58 do not mix at the tree level ; thus @xmath60 is the mass of @xmath58 , @xmath61 , and @xmath62 gev is the mass of the sm higgs @xcite .
we note that the @xmath49 symmetry is softly broken by some other sector of lagrangian .
the gauge - invariant yukawa couplings of vltqs to the sm quarks , the sm higgs doublet , and the new higgs singlet are expressed as : -l^y_vltq & = & tr(|f_2l f_2r ) s + & + & m_f_1 tr(|f_1l f_1r ) + m_f_2 tr(|f_2l f_2r)+ h.c .
, [ eq : yukawa ] where @xmath63 is the left - handed sm quark doublet , all flavor indices are hidden , @xmath64 , and @xmath65 is the @xmath66 vltq with hypercharge @xmath67 , whose representations of @xmath68 are : f_1 = ( cc u_1/ & x + d_1 & -u_1/ ) , f_2 = ( cc d_2/ & u_2 + y & -d_2/ ) . under @xmath49 transformation , @xmath69 .
the electric charges of @xmath70 , @xmath71 , @xmath34 , and @xmath35 are @xmath72 , @xmath37 , @xmath73 , and @xmath74 , respectively .
therefore , @xmath75 can mix with up ( down)-type sm quarks . the masses of vltqs do not originate from the electroweak symmetry breaking . due to the gauge symmetry ,
vltqs in a given multiplet state are degenerate , and denoted by @xmath76 . since the mass terms of vltqs do not involve the @xmath58 field and the associated operators are dimension-3 , the discrete @xmath49 symmetry is softly broken by @xmath77 terms .
it is worth mentioning that @xmath78 results in the mixture of the sm quarks and vltqs ; consequently , the @xmath56 coupling to the top quark is modified and the @xmath56 couplings to vltqs are induced . as a result , the sm higgs production cross section via gluon - gluon fusion ( ggf ) and its decays will be modified . in the next subsection , we discuss the modifications in detail .
we note that the @xmath49 breaking effects will induce @xmath79 term through one - loop diagrams in the scalar potential .
however , in addition to the suppression factor @xmath80 , the loop effects are suppressed by the small yukawa couplings @xmath78 ( see the detailed discussions later ) . as a result ,
the induced vev of @xmath58 field and the induced mixing between @xmath56 and @xmath58 are small , and the br for @xmath81 decay is a factor of two smaller than that for @xmath82 decay @xcite .
next , we discuss the weak interactions of vltqs . as usual , we write the covariant derivative of @xmath25 as : d_= _ + i ( t^+ w^+_+t^- w^- _ ) + i ( t^3 - s^2_w q ) z_+
i e q a _ , where @xmath83 , @xmath84 , and @xmath85 are the gauge bosons in the sm , @xmath86 is the gauge coupling of @xmath87 , @xmath88 , @xmath89 is the weinberg angle , @xmath90 , and the charge operator @xmath91 , where @xmath35 is the hypercharge of the particle . thus , the gauge interactions of vltqs are written as @xcite : _ vff & = & -g + & - & , [ eq : vff ] where we express the triplet vlqs as @xmath92 and @xmath93 , diag@xmath94 , diag@xmath95 , and diag@xmath96 . to further understand the weak interactions in terms of physical states , we have to investigate the structures of flavor mixings when @xmath78 effects in eq .
( [ eq : yukawa ] ) are involved .
the introduced two @xmath87 triplet vlqs contain the quarks with electric charges of @xmath97 and @xmath37 . from the new higgs yukawa couplings to the sm higgs and vltqs , the mixture between the sm quarks and vltqs is generated after ewsb . in order to get the physical mass eigenstates of quarks and the new flavor mixings , we have to diagonalize the mass matrices of the sm quarks and vltqs . with the yukawa couplings @xmath98 and @xmath99 ( i=1 - 3 ) in eq .
( [ eq : yukawa ] ) , the quark mass terms are given by : -l_mass = |*q*_l * m*_q * q*_r + |*q*_l * y*^q v * f*_r + |*f*_l * m*_f * f*_r + h.c . , where @xmath100 or @xmath101 denotes the sm up- or down - type quarks .
we have chosen the basis such that @xmath102 is a @xmath103 diagonal matrix , @xmath104 or @xmath105 is the vltq with charge @xmath97 or @xmath37 , diag@xmath106 ) , and =( cc y_11/2 & y_21/ + y_12/2 & y_22/ + y_13/2 & y_23/ ) , * y^d*= ( cc y_11/ & -y_21/2 + y_12/ & -y_22/2 + y_13/ & -y_23/2 ) .
[ eq : yud ] we do not have @xmath107 terms due to gauge invariance . the quark mass matrices for electric charges @xmath97 and @xmath37 now become @xmath108 matrices .
one can introduce the @xmath109 unitary matrices @xmath110 and @xmath111 to diagonalize the mass matrices , namely @xmath112 . in order to obtain the information of @xmath113 ,
we consider the multiplications of mass matrices to be @xmath114 and @xmath115 , where @xmath116 and @xmath117 are expressed as : @xmath118 with @xmath119 , and @xmath120 .
it is clear that the off - diagonal matrix elements in @xmath116 are related to @xmath121 while those in @xmath117 are @xmath122 .
due to @xmath123 , the unitary matrices @xmath124 can be expanded with respect to @xmath125 and @xmath126 ; at the leading order approximation , they can be formulated as : v^q _
( cc _ 33 & -(^q_)_32 + ( ^q _ ) _ 23 & _ 22 ) , [ eq : vqlr ] where @xmath127 , @xmath128 , and @xmath129 .
we find that the effects of @xmath130 are suppressed by @xmath131 while those of @xmath132 are associated with @xmath133 .
since the top and bottom quarks are much heavier than other sm quarks , in this study we keep the contributions from @xmath134 and @xmath135 , and ignore other @xmath136 that involve the light quark masses .
we use the flavor mixing matrices of eq .
( [ eq : vqlr ] ) to investigate the new flavor couplings of the higgs and the weak gauge bosons below . from the yukawa couplings in eq .
( [ eq : yukawa ] ) , the sm higgs couplings to the quarks in the flavor space are written as : -l_hqq & = |q_l v^q q_r h + h.c . , [ eq : hqq ] where @xmath137 or @xmath138 , @xmath139 , and @xmath140 are the physical states of vltqs and carry the electric charges of @xmath72 and @xmath37 , respectively , and @xmath141 is the mixing matrix for the @xmath142-type quark and is given by : @xmath143 the small effects , such as @xmath144 , @xmath145 , and @xmath146 , have been dropped . according to eq .
( [ eq : vq ] ) , the @xmath56-mediated fcncs for the sm quarks ( e.g. , @xmath56-@xmath142-@xmath147 ) are proportional to @xmath148 . if the mass effects of the first two generations of quarks are neglected , we have the flavor - changing higgs interactions : @xmath149 where @xmath150 quark , @xmath151 quark , the definition of @xmath152 in eq .
( [ eq : yud ] ) is applied , and @xmath153 .
the @xmath154 and @xmath155 oscillations can be induced via the tree - level higgs mediation in the vltq model .
additionally , the brs for the flavor - changing processes @xmath156 , which are highly suppressed in the sm , become sizable .
in addition to the new fcnc couplings , the flavor - conserving couplings are also modified : -l_hqq
= |t_l t_r h + quark will affect the sm higgs production and decays in the @xmath7 collisions at the lhc .
if we take @xmath157 tev , @xmath158 , and @xmath159 gev , the @xmath56 production cross section by the top - quark loop will be reduced by @xmath160 of the sm prediction .
that is , the influence of new effects can not be ignored arbitrarily . from the flavor mixing matrix in eq .
( [ eq : vq ] ) , we can also obtain the sm higgs interactions with the vltqs as : @xmath161 with @xmath162 . since vltqs are color triplet states in @xmath163 and carry the same color charges as those of the sm quarks , the new couplings @xmath164 also contribute to the @xmath56 production cross section via the ggf channel .
we will study their influence on the process @xmath165 in the numerical analysis . by combining the charged weak interactions of the sm
quarks with those of vltqs in eq .
( [ eq : vff ] ) , the charged current interactions of quarks can be formulated by : _
w & = & - |*u*_l ^v^l_ckm * d*_l w^+ _ - |*u*_r ^v^r_ckm * d*_r w^+_+h.c .
, [ eq : wud ] where @xmath166 and @xmath167 are respectively the physical up- and down - type quarks , and @xmath168 is the @xmath109 ckm matrix for left ( right)-handed quarks , defined by : v^l_ckm & = & v^u_l ( cc ( v_ckm ) _ 33 & 0_32 + 0_23 & _ 22 ) v^d_l , v^r_ckm = v^u_r ( cc 0_33 & 0_32 + 0_23 & _ 22 ) v^d_r .
the @xmath103 matrix @xmath169 is associated with the sm ckm matrix . since the weak isospin of a triplet quark differs from that of a doublet quark , the new @xmath109 ckm matrices @xmath170 are non - unitary . by using the results of eq .
( [ eq : vqlr ] ) , the ckm matrix elements for the three - generation sm quarks are modified to be : ( v^sm_ckm)_i j ( v_ckm)_i j + ( _ 1i _ 1j - _ 2i _ 2j ) .
[ eq : mckm ] with @xmath171 and @xmath172 tev , the changes of the sm ckm matrix elements are roughly estimated as @xmath173 . as indicated by experiments @xcite ,
the value of @xmath174 has the same order of magnitude as @xmath175 and is larger than @xmath176 . to satisfy the constraints of @xmath176 , the possible schemes are : ( a ) @xmath177 is less than @xmath178 , the smallest ckm matrix element , ( b ) @xmath179 so that @xmath180 and ( c ) @xmath181 which leads to @xmath180 .
moreover , if we adopt @xmath182 @xmath1831 - 3 ) , all ckm matrix elements return to the sm ones . with the leading - order approximation for @xmath184 ,
the @xmath185-boson interactions with the sm quarks and vltqs are given by @xcite : @xmath186w^+_\mu \non \\ & - \frac{g}{2 } \left [ \zeta_{2i } \bar d_{il } \gamma^\mu y_l + \frac{m_b \zeta_{23}}{m_{f_2 } } \bar b_r \gamma^\mu y_r - \zeta_{1i } \bar x_l \gamma^\mu u_{il } - \frac{m_t \zeta_{13}}{m_{f_1 } } \bar x_r \gamma^\mu t_r \right ] w^+_\mu + h.c .
\label{eq : wfq } \end{aligned}\ ] ] the charged weak interactions of vltqs can be directly read from eq .
( [ eq : vff ] ) .
we next discuss the neutral weak interactions .
it is known that the left - handed and right - handed quarks in the sm are @xmath187 doublets and singlets , respectively ; however , the vltqs are @xmath87 triplets .
since the isospin of a triplet is different from those of doublets and singlets , in order to combine the vltqs with the sm quarks into the same representation in the flavor space , we need to rewrite the vertex structure of the @xmath1-boson , @xmath188 , in eq .
( [ eq : vff ] ) to fit the cases of doublets and singlets , such as @xmath189 , where @xmath190 for doublets and @xmath191 for singlets . due to the isospin difference ,
@xmath1-mediated fcncs are induced at the tree level . since vltqs @xmath34 and @xmath35 carry the electric charges of @xmath73 and @xmath192 , respectively , they can not mix up with other quarks in the neutral current interactions . in terms of weak eigenstates
, we write the weak neutral current interactions in eq .
( [ eq : vff ] ) as : @xmath193\ , , \label{eq : zff } \end{aligned}\ ] ] where @xmath194 and @xmath195 are composed of vltqs with electric charges of @xmath72 and @xmath37 , respectively , @xmath196 , @xmath190 for @xmath197 , @xmath198 , and @xmath199 .
we succeed in expressing the @xmath1 couplings to vltqs by using the sm @xmath1 couplings .
it is clear that the first two terms in eq .
( [ eq : zff ] ) lead to the flavor - conserving couplings when the sm quarks and vltqs form a representation in the dimension-5 flavor space . since the sm quarks
do not have the interaction structures , as shown in the last three terms of eq .
( [ eq : zff ] ) , as a result , fcncs via @xmath1 mediation are generated .
hence , the @xmath1-boson interactions with quarks , which carry electric charges of @xmath72 and @xmath37 , can be formulated as : _
zqq & = & - c^q_l_ij |q_il ^q_j l z_- c^q_r_ij |q_i r ^q_jr z _ , + [ eq : zqq ] c^q_l_ij & = & ( i^q_3 -s^2_w q_q ) _
ij + ( -v^q_li4 v^q*_lj4 + v^q_li5 v^q*_lj5 ) , + c^q_r_ij & = & -s^2_w q_q _ ij + _
q ( v^q_r)_i _ q ( v^q*_r)__q j where @xmath200 or @xmath138 , @xmath201 are defined in eq . ( [ eq : vqlr ] ) , @xmath202 , and @xmath203 .
using eq .
( [ eq : vqlr ] ) and the leading - order approximation , the new gauge couplings of the @xmath1-boson to the sm quarks are given by : _
zq_iq_j = - ( a_q _ 1i _ 1j - b_q _ 2i _
2j ) |q_il ^q_jl z _ , [ eq : zqiqj ] where @xmath204 denote the up- or down - type sm quarks , @xmath205 , and @xmath206 .
it can be seen that the fcnc effects can contribute to @xmath30 neutral meson mixings . a comparison with the results in eq .
( [ eq : mckm ] ) indicates that the induced new coupling structures in charged and neutral currents are different .
it is interesting to investigate the possible schemes that can simultaneously satisfy the constraints from the ckm matrix elements and the data of neutral meson oscillations .
the interactions of the @xmath1-boson couple to one vltq and one sm quark are shown as : @xmath207 one can get the @xmath1 couplings to vltqs from eq .
( [ eq : vff ] ) .
in this section , we discuss the constraints from low - energy @xmath30 processes and from the data of the sm higgs production and decay into diphotons . from eqs .
( [ eq : hqq ] ) and ( [ eq : zqiqj ] ) , we know that the @xmath56- and @xmath1-mediated fcncs appear and contribute to the @xmath30 processes , such as @xmath209 and @xmath210 mixings , where the current experimental data can give strict constraints on the free parameters . since the fcnc couplings in the up - type quarks are the same as those in the down - type quarks and the hadronic effects in the @xmath211-meson system are dominated by unclear non - perturbative effects , we focus on @xmath212 and @xmath213 processes . following the notations in previous studies @xcite ,
the transition matrix elements for @xmath209 and @xmath210 mixings are given by : @xmath214\ , , \\
m^{k}_{12}(z ) & = \frac{(\delta^{sd}_l(z))^2}{2 m^2_z } c^{\rm vll}_1 ( \mu_z ) \bar p^{\rm vll}_1(k,\mu_z)\ , , \\
m^{b_q}_{12}(z ) & = \frac{(\delta^{bd}_l(z))^2}{2 m^2_z } c^{\rm vll}_1 ( \mu_z ) \bar p^{\rm vll}_1(b_q , \mu_z)\ , .
\end{aligned}\ ] ] @xmath215 is the wilson coefficient with @xmath216 qcd corrections , and @xmath217 denotes the hadronic effects that include the renormalization group ( rg ) evolution from high energy to low energy , whose expressions are @xcite : @xmath218_{\rm sll } r_{b_q } b^{\rm sll}_{1}(\mu_b ) -\frac{3}{2 } [ \eta_{21}(\mu_b)]_{\rm sll } r_{b_q } b^{\rm sll}_{2}(\mu_b ) \ , , \non \\
p^{\rm sll}_2(b_q,\mu_b ) & = - \frac{5}{8 } [ \eta_{12}(\mu_b ) ] _ { \rm sll } r_{b_q } b^{\rm sll}_{1}(\mu_b ) -\frac{3}{2 } [ \eta_{22}(\mu_b)]_{\rm sll } r_{b_q } b^{\rm sll}_{2}(\mu_b ) \ , , \end{aligned}\ ] ] @xmath219 , where @xmath220 and @xmath221 are the mass and decay constant of the @xmath222-meson , respectively , @xmath223 gev for the @xmath224-meson , @xmath225 , and the values of other hadronic effects and rg evolution effects are given in table [ tab : values ] .
@xmath226 are from the short - distance interactions of eqs .
( [ eq : hqq ] ) and ( [ eq : zqiqj ] ) and are written as : @xmath227 since we have ignored the effects of light quark masses , the @xmath56-mediated fcnc has no contribution to @xmath209 mixing . to constrain the parameters ,
we assume that the obtained @xmath228 in the model should be less than the experimental measurements . to understand the individual influences of @xmath56 mediation and @xmath1 mediation , we show their constraints separately . with @xmath229 , @xmath230 , and the inputs of table [
tab : values ] , we obtain the constraints as : @xmath209 mixing : @xmath231 @xmath154 mixing : @xmath232 @xmath155 mixing : @xmath233 from these results , we find that the constraint from @xmath234 is only a factor of 2 stronger than that from @xmath235 . since the ratio @xmath236 in experiments is very close to the wolfenstein s parameter @xmath237
@xcite , the difference of a factor of @xmath238 between eq .
( [ eq : bdc ] ) and eq .
@xmath239 is reasonable . .
[ cols="^,^,^,^,^",options="header " , ] finally , we briefly discuss the new physics in connection to the flavor physics . in this study
, we do not introduce new couplings to the lepton sector , therefore , the contributions to the lepton flavor - changing processes are similar to the sm predictions . however , the introduced vlqs lead to fcncs at the tree level in the quark sector , where the strict constraints from @xmath30 processes have been considered in section iii .
besides the rare decays @xmath240 and @xmath241 that were discussed earlier , it is also interesting to investigate the fcnc effects in the low energy physics .
for instance , the coupling @xmath242 can contribute to the @xmath243 and @xmath244 decays , where the sm predicted brs are of @xmath245 , both are sensitive to the new physics effects , and the theoretical uncertainties are well - controlled @xcite .
furthermore , the na62 experiment at cern can achieve the @xmath246 to be a precision of @xmath2 @xcite ; and the koto experiment at j - parc for @xmath247 decays can reach the sm sensitivity .
thus , it is important to search for new physics in rare @xmath248 decays . in @xmath249-meson physics ,
the tree - level couplings @xmath250 with @xmath251 can contribute to @xmath252 decays .
although the measured @xmath253 is consistent with the sm prediction @xcite , a @xmath254 deviation from the sm prediction in the angular analysis of @xmath255 is observed @xcite .
it is worthy to explore the excess in our model .
since the detailed analysis of flavor physics is beyond the scope of this paper , a complete analysis will be studied elsewhere @xcite .
we have studied the phenomenology of which two triplet vlqs with @xmath0 and a higgs singlet are embedded in the sm . because the isospin of vlqs is different from that of the sm quarks , higgs- and @xmath1-mediated fcncs are generated at the tree level and the new ckm matrix becomes a non - unitary matrix .
we find that the modifications of the ckm matrix elements coupled to the sm quarks can be smeared out if two triplet vlqs are introduced and the scheme @xmath256 is adopted , where @xmath257 are the parameters from flavor mixings .
although the tree - level fcncs can not be removed , it was found that when the constraints from @xmath30 processes are applied , the upper limits of brs for @xmath258 decays are @xmath259 , which is two orders of magnitude smaller than the current experimental bounds . with the values of constrained parameters
, we examined the influence of the model on the sm higgs production and its diphoton decay ; we found that @xmath260 and @xmath261 can have @xmath262 and @xmath31 deviations from the sm results , respectively . as a result ,
the signal strength for @xmath165 is thus changed by @xmath2 .
the main purpose of this work was to explore the single production of exotic vlqs @xmath34 and @xmath35 in the @xmath7 collisions at @xmath8 tev .
we gave a detailed analysis for each possible @xmath6 scattering , where @xmath204 and @xmath263 are the possible initial quarks .
it was found that the contributions of @xmath264-channel annihilations are much smaller than those of @xmath265-channel annihilations . from this study , we comprehend the contribution of each subprocess to the production cross section of a specific vlq . the interesting production channels are @xmath266 , @xmath267 , and @xmath268 , where the corresponding production cross sections for @xmath12 tev are @xmath13 , @xmath14 , and @xmath15 fb , respectively . from our analysis , it is clear to see that the single production cross sections of vlqs are much larger than the pair production cross sections , which are through qcd processes .
the dominant decay modes of the vlqs are @xmath16 and @xmath17 .
each br can be 1/2 in our chosen scheme . for illustration
, we estimate the significances for the channels proposed in eq .
( [ eq : signal ] ) .
it is found that the significance for @xmath18 channel can be over @xmath19 . + * acknowledgments * g. aad _ et al . _ [ atlas collaboration ] , `` observation of a new particle in the search for the standard model higgs boson with the atlas detector at the lhc '' , phys .
b * 716 * , 1 ( 2012 ) [ arxiv:1207.7214 [ hep - ex ] ] .
s. chatrchyan _ et al . _
[ cms collaboration ] , `` observation of a new boson at a mass of 125 gev with the cms experiment at the lhc '' , phys .
b * 716 * , 30 ( 2012 ) [ arxiv:1207.7235 [ hep - ex ] ] .
g. aad _ et al . _ [ atlas collaboration ] , jhep * 1512 * , 055 ( 2015 ) doi:10.1007/jhep12(2015)055 [ arxiv:1506.00962 [ hep - ex ] ] .
v. khachatryan _ et al .
_ [ cms collaboration ] , jhep * 1408 * , 173 ( 2014 ) [ arxiv:1405.1994 [ hep - ex ] ] ; v. khachatryan _ et al . _
[ cms collaboration ] , jhep * 1408 * , 174 ( 2014 ) [ arxiv:1405.3447 [ hep - ex ] ] ; v. khachatryan _ et al . _
[ cms collaboration ] , phys .
b * 740 * , 83 ( 2015 ) [ arxiv:1407.3476 [ hep - ex ] ] .
v. khachatryan _ et al . _
[ cms collaboration ] , phys .
b * 749 * , 337 ( 2015 ) [ arxiv:1502.07400 [ hep - ex ] ] . the atlas collaboration , atlas - conf-2015 - 081 . m. aaboud _ et al . _ [ atlas collaboration ] , arxiv:1606.03833 [ hep - ex ] .
cms collaboration [ cms collaboration ] , cms - pas - exo-15 - 004 .
v. khachatryan _ et al . _
[ cms collaboration ] , arxiv:1606.04093 [ hep - ex ] .
a. arhrib , r. benbrik , c. h. chen , m. gomez - bock and s. semlali , eur .
j. c * 76 * , no .
6 , 328 ( 2016 ) doi:10.1140/epjc / s10052 - 016 - 4167 - 9 [ arxiv:1508.06490 [ hep - ph ] ] .
f. del aguila , m. perez - victoria and j. santiago , jhep * 0009 * , 011 ( 2000 ) [ hep - ph/0007316 ] .
a. atre , m. carena , t. han and j. santiago , phys .
d * 79 * , 054018 ( 2009 ) [ arxiv:0806.3966 [ hep - ph ] ] .
j. mrazek and a. wulzer , phys .
d * 81 * , 075006 ( 2010 ) [ arxiv:0909.3977 [ hep - ph ] ] .
g. cacciapaglia , a. deandrea , d. harada and y. okada , jhep * 1011 * , 159 ( 2010 ) [ arxiv:1007.2933 [ hep - ph ] ] .
s. gopalakrishna , t. mandal , s. mitra and r. tibrewala , phys .
d * 84 * , 055001 ( 2011 ) [ arxiv:1107.4306 [ hep - ph ] ] .
f. j. botella , g. c. branco and m. nebot , jhep * 1212 * , 040 ( 2012 ) [ arxiv:1207.4440 [ hep - ph ] ] .
y. okada and l. panizzi , adv .
high energy phys . * 2013 * , 364936 ( 2013 ) [ arxiv:1207.5607 [ hep - ph ] ] .
h. cai , jhep * 1302 * , 104 ( 2013 ) [ arxiv:1210.5200 [ hep - ph ] ] .
g. cacciapaglia , a. deandrea , l. panizzi , s. perries and v. sordini , jhep * 1303 * , 004 ( 2013 ) [ arxiv:1211.4034 [ hep - ph ] ] .
a. atre , m. chala and j. santiago , jhep * 1305 * , 099 ( 2013 ) [ arxiv:1302.0270 [ hep - ph ] ] . j. a. aguilar - saavedra , r. benbrik , s. heinemeyer and m. perez - victoria , phys .
d * 88 * , no .
9 , 094010 ( 2013 ) [ arxiv:1306.0572 [ hep - ph ] ] .
s. beauceron , g. cacciapaglia , a. deandrea and j. d. ruiz - alvarez , phys .
d * 90 * , no .
11 , 115008 ( 2014 ) [ arxiv:1401.5979 [ hep - ph ] ] .
a. k. alok , s. banerjee , d. kumar and s. u. sankar , arxiv:1402.1023 [ hep - ph ] .
d. karabacak , s. nandi and s. k. rai , phys .
b * 737 * , 341 ( 2014 ) [ arxiv:1405.0476 [ hep - ph ] ] .
a. k. alok , s. banerjee , d. kumar , s. u. sankar and d. london , phys .
d * 92 * , no . 1 , 013002 ( 2015 ) [ arxiv:1504.00517 [ hep - ph ] ] .
n. vignaroli , phys .
d * 91 * , no . 11 , 115009 ( 2015 ) [ arxiv:1504.01768 [ hep - ph ] ] . c. h. chen and t. nomura , phys .
d * 92 * , no .
11 , 115021 ( 2015 ) [ arxiv:1509.02039 [ hep - ph ] ] .
g. aad _ et al .
_ [ atlas collaboration ] , phys .
b * 718 * , 1284 ( 2013 ) [ arxiv:1210.5468 [ hep - ex ] ] .
g. aad _ et al .
_ [ atlas collaboration ] , jhep * 1411 * , 104 ( 2014 ) [ arxiv:1409.5500 [ hep - ex ] ] .
g. aad _ et al .
_ [ atlas collaboration ] , jhep * 1508 * , 105 ( 2015 ) [ arxiv:1505.04306 [ hep - ex ] ] .
g. aad _ et al . _ [ atlas collaboration ] , phys .
d * 91 * , no .
11 , 112011 ( 2015 ) [ arxiv:1503.05425 [ hep - ex ] ] . g. aad _ et al . _
[ atlas collaboration ] , jhep * 1510 * , 150 ( 2015 ) [ arxiv:1504.04605 [ hep - ex ] ] .
g. aad _ et al . _ [ atlas collaboration ] , phys .
d * 92 * , no .
11 , 112007 ( 2015 ) [ arxiv:1509.04261 [ hep - ex ] ] .
g. aad _ et al .
_ [ atlas collaboration ] , arxiv:1510.02664 [ hep - ex ] .
g. aad _ et al . _ [ atlas collaboration ] , arxiv:1602.06034 [ hep - ex ] . s. chatrchyan _ et al .
_ [ cms collaboration ] , phys .
b * 729 * , 149 ( 2014 ) [ arxiv:1311.7667 [ hep - ex ] ] .
s. chatrchyan _ et al .
_ [ cms collaboration ] , phys .
* 112 * , no .
17 , 171801 ( 2014 ) [ arxiv:1312.2391 [ hep - ex ] ] . v. khachatryan
_ et al . _
[ cms collaboration ] , arxiv:1507.07129 [ hep - ex ] .
v. khachatryan _ et al . _
[ cms collaboration ] , phys .
d * 93 * , no . 1
, 012003 ( 2016 ) [ arxiv:1509.04177 [ hep - ex ] ] . v. khachatryan _ et al . _
[ cms collaboration ] , jhep * 1506 * , 080 ( 2015 ) [ arxiv:1503.01952 [ hep - ex ] ] .
olive et al .
( particle data group ) , chin .
c , 38 , 090001 ( 2014 ) .
a. j. buras and j. girrbach , jhep * 1203 * , 052 ( 2012 ) [ arxiv:1201.1302 [ hep - ph ] ] .
g. aad _ et al . _ [ atlas collaboration ] , arxiv:1507.04548 [ hep - ex ] .
s. chatrchyan et al . .
[ cms collaboration ] , cms - pas - hig-14 - 009 g. aad _ et al . _ [ atlas collaboration ] , arxiv:1508.05796 [ hep - ex ] .
s. chatrchyan _ et al . _
[ cms collaboration ] , phys .
lett . * 112 * , no .
17 , 171802 ( 2014 ) [ arxiv:1312.4194 [ hep - ex ] ] .
a. belyaev , n. d. christensen and a. pukhov , comput .
phys . commun .
* 184 * , 1729 ( 2013 ) [ arxiv:1207.6082 [ hep - ph ] ] .
m. nadolsky , h. l. lai , q. h. cao , j. huston , j. pumplin , d. stump , w. k. tung and c .- p .
yuan , phys .
d * 78 * , 013004 ( 2008 ) [ arxiv:0802.0007 [ hep - ph ] ] .
j. alwall _ et al .
_ , jhep * 1407 * , 079 ( 2014 ) [ arxiv:1405.0301 [ hep - ph ] ] . a. alloul , n. d. christensen , c. degrande , c. duhr and b. fuks , comput . phys
. commun .
* 185 * , 2250 ( 2014 ) [ arxiv:1310.1921 [ hep - ph ] ] . c. s. deans [ nnpdf collaboration ] , arxiv:1304.2781 [ hep - ph ] .
t. sjostrand , s. mrenna , p.
z. skands , jhep * 0605 * , 026 ( 2006 ) .
a. j. buras , d. buttazzo and r. knegjens , jhep * 1511 * , 166 ( 2015 ) [ arxiv:1507.08672 [ hep - ph ] ] . f. newson _ et al .
_ , arxiv:1411.0109 [ hep - ex ] .
a. romano , arxiv:1411.6546 [ hep - ex ] .
v. khachatryan _ et al .
_ [ cms and lhcb collaborations ] , nature * 522 * , 68 ( 2015 ) [ arxiv:1411.4413 [ hep - ex ] ] . r. aaij _
[ lhcb collaboration ] , jhep * 1602 * , 104 ( 2016 ) [ arxiv:1512.04442 [ hep - ex ] ] .
c.h . chen and t. nomura ,
study of rare @xmath248 , @xmath211 , and @xmath249 meson decays in this model is in progress . | two triplet vector - like quarks ( vlqs ) with hypercharges of @xmath0 and one singlet scalar boson are embedded in the standard model ( sm ) to resolve the 750 gev diphoton excess .
the constraints on the tree - level higgs- and @xmath1-mediated flavor - changing neutral currents are discussed in detail . besides the resolution of excess , it is found that the signal strength of diphoton higgs decay can have a @xmath2 deviation from the sm prediction and that the upper limits of the branching ratios for rare top - quark decays are @xmath3 .
we find that the production cross section of a single vlq by electroweak processes is larger than that of vlq - pair by qcd processes .
to explore the signals of the heavy vlqs at the lhc , we throughly analyze the production of single @xmath4 and @xmath5 via @xmath6 annihilations in @xmath7 collisions at @xmath8 tev .
it is found that the electroweak production cross sections for @xmath9 , @xmath10 , and @xmath11 channels with @xmath12 tev can be @xmath13 , @xmath14 , and @xmath15 fb , respectively ; and the dominant decay modes are @xmath16 and @xmath17 . with adopting kinematic cuts , the significance for @xmath18 channel can be over @xmath19 .
kias - p16026 |
in this paper , we perform the first benchmark of finite - volume lattice methods for the low - energy scattering of composite objects .
our results will have immediate applications to lattice studies of elastic neutron - nucleus scattering . in the analysis presented here we consider scattering between a fermion and a bound dimer composed of two fermions . in order to test the precision of our lattice calculations
, we also repeat the same calculations using the skorniakov - ter - martirosian ( stm ) integral equation @xcite . along the way we also provide the most accurate calculation to date for the fermion - dimer effective range parameter .
some of the results presented were summarized in a letter publication @xcite , and we present the full details of the calculations here .
we consider two component fermions .
we will refer to the two fermion components as spin up and spin down and consider the case when the masses are equal , @xmath4 .
we assume finite - range attractive interactions and consider the universal shallow binding limit .
if @xmath5 is the range of the interactions and @xmath6 is the binding momentum of the dimer , then the shallow binding limit corresponds to @xmath7 .
we note that much of the literature on universal three - body systems has focused on the efimov effect @xcite for three bosons , three - component fermions , unequal mass fermions , or mixed bose - fermi systems @xcite . for equal mass two - component fermions , however , there are no short - distance three - body instabilities such as the thomas collapse @xcite
. hence there are no relevant momentum scales other than the dimer binding momentum @xmath6 , and all low - energy scattering parameters can be expressed as dimensionless constants times the corresponding power of @xmath6 . in the shallow binding limit
@xmath6 is the same as the reciprocal of the fermion - fermion scattering length .
there have been numerous calculations of the fermion - dimer scattering length .
the first goes back to the early work of skorniakov and ter - martirosian who found @xmath8 @xcite .
an overview of the skorniakov - ter - martirosian ( stm ) integral equation method will be presented in our discussion below . the same value @xmath8 has been confirmed several times using integral equations @xcite .
a value of @xmath9 was obtained using an epsilon expansion in dimensions @xcite .
a recent correlated gaussian expansion calculation obtained @xmath10 @xcite .
this agrees with integral equation studies which found @xmath11 @xcite and @xmath12 @xcite .
the fermion - dimer results at shallow binding should approximately describe neutron - deuteron scattering in the spin - quartet channel .
experimental measurements find a quartet scattering length @xmath13 fm @xcite .
this corresponds with @xmath14 .
the agreement is better when expressed as fraction of the spin - triplet proton - neutron scattering length , @xmath15 .
the 30% difference between the two values gives an indication of higher order effective range effects .
a more detailed calculation including interaction range effects obtains @xmath16 fm @xcite , in full agreement with experimental values .
in contrast with the scattering length , there is only one previously reported determination of the fermion - dimer effective range parameter .
the correlated gaussian expansion calculation in ref .
@xcite found @xmath17 .
neutron - deuteron scattering data also favors a small value for @xmath18 .
however the sign of @xmath18 has remained an open question . in our continuum calculations presented here
, we use the stm equation to calculate the scattering length and effective range parameter .
we find the values @xmath2 and @xmath3 .
our main focus though is to benchmark lattice calculations of the fermion - dimer scattering length and effective range parameter . for the lattice calculation we apply the finite - volume phase - shift analysis of lscher .
we show that finite - volume topological corrections to the dimer binding energy must be considered in order to obtain accurate results .
once these topological corrections are included in the finite - volume analysis , we find that the lattice and continuum calculations are in full agreement .
we use two different lattice hamiltonian formulations , that agree in the continuum limit . as a final result
we find : @xmath19
few - body systems of two - component fermions with short - range interactions and large scattering lengths in comparison to interparticle distances show universal properties .
physics in such systems does not depend on the structure of the interactions at short distances .
the problem of three two - component fermions at low energies can be described by a local quantum field theory whose only interaction term is a two - body contact interaction . in the following
, we will always consider equal mass fermions with mass @xmath20 .
the extension to unequal masses is straightforward .
the free non - relativistic effective hamiltonian in momentum space can be written as @xmath21 where @xmath22 and @xmath23 are annihilation and creation operators . in position space
these operators can be expressed as @xmath24 combining eqs .
( [ free - non - relativistic ] ) and ( [ operators ] ) the free hamiltonian in configuration space is then given by @xmath25 now we introduce an interaction between the fermions via the potential @xmath26 where @xmath27 denotes normal ordering . at low energies , the potential ( [ potential ] )
can be replaced by a delta - function interaction @xmath28 and the the lowest order effective hamiltonian for two - component fermions is @xmath29 @xmath30 denotes the two - body coupling constant and is directly related to the fermion - fermion scattering length .
it is assumed to be negative so that the interaction is attractive .
the exact value of @xmath30 depends on the scheme used to regulate the short distance behavior . in the next step
we consider the two - body and three - body systems of two - component fermions in hamiltonian lattice formalism and use the lanczos method @xcite to find the lowest eigenvalues .
further details of the hamiltonian lattice formulation can be found in @xcite .
let @xmath31 denote spatial lattice points on a three - dimensional @xmath32 periodic cube .
we use lattice units where physical quantities are multiplied by powers of the spatial lattice spacing @xmath33 to make the combination dimensionless .
the two - component fermions are labelled as spin - up and spin - down , the lattice annihilation operators are written as @xmath34 and @xmath35 .
the free non - relativistic hamiltonian of two - component fermions with only short - range interaction corresponding to the hamiltonian ( [ freehamiltonian ] ) on the three dimensional lattice is @xmath36 , \label{h0}\ ] ] where @xmath37 is the spatial lattice unit vector .
we define the spin - density operators @xmath38 and consider two different kinds of hamiltonians . in the first hamiltonian
we have only a single - site interaction .
this hamiltonian is @xmath39 we consider a second hamiltonian using a contact interaction as well as nearest - neighbour interaction terms in order to eliminate the two - body effective range parameter , @xmath40.\end{aligned}\ ] ] the finite lattice spacing error in these two hamiltonians is of order @xmath41 . the next step is to determine the interaction coefficients , @xmath42 , @xmath43 and @xmath44 using lscher s formula @xcite . at present lscher s formula
is a standard tool in lattice quantum chromodynamics and in lattice effective field theory .
it relates the two - body energy levels in a finite volume to the s - wave phase shift .
@xmath45 where @xmath46 is the three - dimensional zeta - function , @xmath47 is the length of the box and @xmath48 is the center - of - mass momentum . the zeta - function in three dimensions is defined as @xmath49 .
\label{s(eta)}\ ] ] for @xmath50 we can expand @xmath46 in powers of @xmath51 , @xmath52,\nonumber\label{expand}\\ & = -\frac{1}{\eta}+s_{0}+s_{1}\eta+s_{2}\eta^{2}+s_{3}\eta^{3}+\cdots,\end{aligned}\ ] ] where @xmath53,\quad s_{i}=\sum_{\vec{n}\neq0}\frac{1}{(\vec{n}^{2})^{i+1}}. \label{coefficients}\ ] ] the first few coefficients are @xmath54 lscher s formula does not include the contribution from higher partial waves but at asymptotically small momenta we can neglect such corrections . for small momenta
we have the effective range expansion , @xmath55 where @xmath56 is the scattering length and @xmath57 is the effective range .
thus the effective range parameters can be extracted from the finite volume energy levels using eq .
( [ pcot ] ) . in terms of @xmath51 ,
the energy of the dimer is @xmath58 the interaction coefficient @xmath42 is tuned to construct two - body binding states ( dimers ) comprised of one spin - up and one spin - down fermion of energies @xmath59 mev , @xmath60 mev , @xmath61 mev , @xmath62 mev , @xmath63 mev and @xmath64 mev in the large volume ( @xmath65 ) using the lanczos method . in such a large volume
the finite volume corrections to the dimer binding energy are negligible . in our calculation
we take @xmath66 mev and @xmath67 mev . to find the interaction coefficients of lattice hamiltonian @xmath68 we proceed as follows .
setting the effective range to zero requires that the following relation should be satisfied near threshold @xmath69 the interaction coefficients @xmath43 and @xmath44 are tuned in order to give the binding energies listed above for the ground state in a large volume ( @xmath65 ) and to fullfil eq .
( [ condition ] ) for the first excited state .
the plot for @xmath70 versus @xmath71 for the first excited state is shown in fig .
[ pcot_c0c1 ] .
the different values of @xmath48 are generated by calculating for different box sizes .
we note that @xmath70 has zero slope near threshold since we set the effective range to zero .
both hamiltonians reproduce the same continuum limit of fermions with attractive zero - range interactions . the corresponding values for the interaction coefficients
are summarized in table [ tab1 ] . | we present continuum and lattice calculations for elastic scattering between a fermion and a bound dimer in the shallow binding limit . for the continuum calculation we use the skorniakov - ter - martirosian ( stm ) integral equation to determine the scattering length and effective range parameter to high precision . for the lattice calculation we use the finite - volume method of lscher .
we take into account topological finite - volume corrections to the dimer binding energy which depend on the momentum of the dimer . after subtracting these effects
, we find from the lattice calculation @xmath0 and @xmath1 .
these results agree well with the continuum values @xmath2 and @xmath3 obtained from the stm equation .
we discuss applications to cold atomic fermi gases , deuteron - neutron scattering in the spin - quartet channel , and lattice calculations of scattering for nuclei and hadronic molecules at finite volume . |
in recent years there has been a great interest in the study of absorption effects on transport properties of classically chaotic cavities @xcite ( for a review see ref .
this is due to the fact that for experiments in microwave cavities @xcite , elastic resonators @xcite and elastic media
@xcite absorption always is present .
although the external parameters are particularly easy to control , absorption , due to power loss in the volume of the device used in the experiments , is an ingredient that has to be taken into account in the verification of the random matrix theory ( rmt ) predictions . in a microwave experiment of a ballistic chaotic cavity connected to a waveguide supporting one propagating mode , doron _
@xcite studied the effect of absorption on the @xmath0 sub - unitary scattering matrix @xmath1 , parametrized as @xmath2 where @xmath3 is the reflection coefficient and @xmath4 is twice the phase shift .
the experimental results were explained by lewenkopf _
_ @xcite by simulating the absorption in terms of @xmath5 equivalent parasitic channels " , not directly accessible to experiment , each one having an imperfect coupling to the cavity described by the transmission coefficient @xmath6 . a simple model to describe chaotic scattering including absorption was proposed by kogan _ et al . _
it describes the system through a sub - unitary scattering matrix @xmath1 , whose statistical distribution satisfies a maximum information - entropy criterion . unfortunately the model turns out to be valid only in the strong - absorption limit and for @xmath7 . for the @xmath0 @xmath1-matrix of eq .
( [ s11 ] ) , it was shown that in this limit @xmath4 is uniformly distributed between 0 and @xmath8 , while @xmath3 satisfies rayleigh s distribution @xmath9 where @xmath10 denotes the universality class of @xmath1 introduced by dyson @xcite : @xmath11 when time reversal invariance ( tri ) is present ( also called the _ orthogonal _ case ) , @xmath12 when tri is broken ( _ unitary _ case ) and @xmath13 corresponds to the symplectic case . here , @xmath14 and @xmath15 , is the ratio of the mean dwell time inside the cavity ( @xmath16 ) , where @xmath17 is the mean level spacing , and @xmath18 is the absorption time .
this ratio is a measure of the absorption strength .
( [ rayleigh ] ) is valid for @xmath19 and for @xmath20 as we shall see below .
the weak absorption limit ( @xmath21 ) of @xmath22 was calculated by beenakker and brouwer @xcite , by relating @xmath3 to the time - delay in a chaotic cavity which is distributed according to the laguerre ensemble .
the distribution of the reflexion coefficient in this case is @xmath23 in the whole range of @xmath24 , @xmath22 was explicitly obtained for @xmath12 @xcite : @xmath25 , \label{beta2}\ ] ] and for @xmath13 more recently @xcite .
( [ beta2 ] ) reduces to eq .
( [ laguerre ] ) for small absorption ( @xmath21 ) while for strong absorption it becomes @xmath26 notice that @xmath27 approaches zero for @xmath3 close to one .
then the rayleigh distribution , eq .
( [ rayleigh ] ) , is only reproduced in the range of few standard deviations i.e. , for @xmath28 .
this can be seen in fig .
[ fig : fig1](a ) where we compare the distribution @xmath27 given by eqs .
( [ rayleigh ] ) and ( [ bigbeta2 ] ) with the exact result given by eq .
( [ beta2 ] ) for @xmath29 .
as can be seen the result obtained from the time - delay agrees with the exact result but the rayleigh distribution is only valid for @xmath7 . since the majority of the experiments with absorption are performed with tri ( @xmath11 )
it is very important to have the result in this case . due to the lack of an exact expression at that time
, savin and sommers @xcite proposed an approximate distribution @xmath30 by replacing @xmath24 by @xmath31 in eq .
( [ beta2 ] ) .
however , this is valid for the intermediate and strong absorption limits only .
another formula was proposed in ref .
@xcite as an interpolation between the strong and weak absorption limits assuming a quite similar expression as the @xmath12 case ( see also ref .
more recently @xcite , a formula for the integrated probability distribution of @xmath32 , @xmath33 , was obtained .
the distribution @xmath34 then yields a quite complicated formula . due to the importance to have an `` easy to use '' formula for the time reversal case ,
our purpose is to propose a better interpolation formula for @xmath22 when @xmath11 . in the next section
we do this following the same procedure as in ref .
we verify later on that our proposal reaches both limits of strong and weak absorption . in sec .
[ sec : conclusions ] we compare our interpolation formula with the exact result of ref . @xcite .
a brief conclusion follows .
from eqs . ( [ rayleigh ] ) and ( [ laguerre ] )
we note that @xmath24 enters in @xmath22 always in the combination @xmath31 .
we take this into account and combine it with the general form of @xmath27 and the interpolation proposed in ref . @xcite . for @xmath11
we then propose the following formula for the @xmath3-distribution @xmath35 , \label{beta1}\ ] ] where @xmath36 , @xmath37 is a hyper - geometric function @xcite , and @xmath38 is a normalization constant @xmath39 where @xmath40 and @xmath41 is the incomplete @xmath42-function @xmath43 in the next sections , we verify that in the limits of strong and weak absorption we reproduce eqs .
( [ rayleigh ] ) and ( [ laguerre ] ) .
, for ( a ) @xmath12 ( unitary case ) and ( b ) @xmath11 ( orthogonal case ) . in ( a )
the continuous line is the exact result eq .
( [ beta2 ] ) while in ( b ) it corresponds to the interpolation formula , eq .
( [ beta1 ] ) .
the triangles in ( a ) are the results given by eq .
( [ bigbeta2 ] ) for @xmath12 and in ( b ) they correspond to eq .
( [ bigbeta1 ] ) .
the dashed line is the rayleigh distribution eq .
( [ rayleigh ] ) , valid only for @xmath44 and @xmath45 .
, width=302 ] in the strong absorption limit , @xmath46 , @xmath47 , and @xmath48 .
then , @xmath49 therefore , the @xmath3-distribution in this limit reduces to @xmath50 which is the equivalent of eq .
( [ bigbeta2 ] ) but now for @xmath11 . as for the @xmath12 symmetry ,
it is consistent with the fact that @xmath51 approaches zero as @xmath3 tends to one .
it reproduces also eq .
( [ rayleigh ] ) in the range of a few standard deviations ( @xmath52 ) , as can be seen in fig .
[ fig : fig1](b ) .
for weak absorption @xmath53 , the incomplete @xmath42-function in @xmath38 reduces to a @xmath42-function @xmath54 [ see eq . ( [ gammafunc ] ) ] .
then , @xmath51 can be written as @xmath55 .\end{aligned}\ ] ] by keeping the dominant term for small @xmath56 , eq .
( [ laguerre ] ) is reproduced .
, 2 , 5 , and 7 . the continuous lines correspond to the distribution given by eq .
( [ beta1 ] ) . for comparison
we include the exact results of ref .
@xcite ( dashed lines).,width=302 ] in fig .
[ fig : fig2 ] we compare our interpolation formula , eq .
( [ beta1 ] ) , with the exact result of ref .
@xcite . for the same parameters used in that reference
we observe an excellent agreement . in fig .
[ fig : fig3 ] we plot the difference between the exact and the interpolation formulas for the same values of @xmath24 as in fig .
[ fig : fig2 ] .
the error of the interpolation formula is less than 4% . ) , for the @xmath3-distribution for @xmath11 for the same values of @xmath24 as in fig .
[ fig : fig2].,width=302 ]
we have introduced a new interpolation formula for the reflection coefficient distribution @xmath22 in the presence of time reversal symmetry for chaotic cavities with absorption .
the interpolation formula reduces to the analytical expressions for the strong and weak absorption limits .
our proposal is to produce an `` easy to use '' formula that differs by a few percent from the exact , but quite complicated , result of ref .
we can summarize the results for both symmetries ( @xmath11 , 2 ) as follows @xmath57,\ ] ] where @xmath58 is a normalization constant that depends on @xmath14 .
this interpolation formula is exact for @xmath12 and yields the correct limits of strong and weak absorption .
the authors thank to dgapa - unam for financial support through project in118805 .
we thank d. v. savin for provide us the data for the exact results we used in figs .
[ fig : fig2 ] and [ fig : fig3 ] and to j. flores and p. a. mello for useful comments . | we propose an interpolation formula for the distribution of the reflection coefficient in the presence of time reversal symmetry for chaotic cavities with absorption .
this is done assuming a similar functional form as that when time reversal invariance is absent .
the interpolation formula reduces to the analytical expressions for the strong and weak absorption limits .
our proposal is compared to the quite complicated exact result existing in the literature . |
* simulation . * to find the transmission amplitude and phase of a multi - element metasurface , the rigorous coupled wave analysis technique was used @xcite .
a normally incident plane wave at each wavelength was used as the excitation , and the amplitude and phase of the transmitted wave were extracted . since the lattice is subwavelength for normal incidence at both wavelengths , only the zeroth order diffracted light is nonzero .
this justifies the use of only one transmission value at each wavelength to describe the behavior of meta - atoms .
the lattice constant was chosen as 720 nm , and the a - si posts were 718-nm tall .
refractive indices of 3.56 and 3.43 were assumed for a - si at 915 nm and 1550 nm , respectively .
the paraxial focal distance of the two lenses were calculated to be 286 @xmath24 m and 495 @xmath24 m for the lenses that focus light from the fiber to 400 @xmath24 m and 1000 @xmath24 m respectively , by fitting a parabola to the phase profiles of the lenses . for a fitted parabola @xmath28
, the paraxial focal distance can be calculated using @xmath29 .
the corresponding numerical apertures can then found to be 0.46 and 0.29 for the two lenses .
the perfect phase mask ( that also served as the goal phase profile for the designed devices ) was calculated from the illuminating field and the aspherical desired phase profile using the method described in supplementary information of @xcite .
the illuminating field was calculated by propagating the output fields of single mode fibers at each wavelength using plane wave expansion ( pwe ) method up to the metasurface layer .
the perfect phase mask was then applied to the field , and the result was propagated using the pwe method to the focal point .
the diffraction limited fwhm was then calculated from the intensity profile at the focal plane .
full wave simulation of a full lens was done using finite difference time domain method ( fdtd ) in meep @xcite . a lens with a diameter of 75 @xmath24 m and a focal length of 100 @xmath24 m was designed with the same method as the fabricated device .
to the lens focuses the light emitted from a single mode fiber ( with mode diameters of 10.4 @xmath24 m at 1550 nm and 6 @xmath24 m at 915 nm ) placed 150 @xmath24 m away from a 125 @xmath24 m thick fused silica substrate ( all of the geometrical dimensions were chosen 4 times smaller than the values for the experimentally measured device ) .
the distances to fibers were chosen such that more than 99@xmath0 of the total power emitted by the fiber passes through the lens aperture . at both wavelengths ,
the light from the fibers was propagated through air , air - glass interface , and through glass up to a plane about a wavelength before the metasurface using a plane wave expansion ( pwe ) code .
electric and magnetic field distributions at this plane were used as sources for fdtd simulation of the lenses , and fields were calculated at about a wavelength after the metasurface using meep .
the pwe code was used again to further propagate these fields to the focal plane and beyond ( fig .
the focusing efficiencies were calculated by dividing the power in a 20-@xmath24m - diameter disk around the focus , to the total power incident on the lens .
* sample fabrication . * a 718-nm - thick hydrogenated a - si layer was deposited on a fused silica substrate using the plasma enhanced chemical vapor deposition ( pecvd ) technique with a 5% mixture of silane in argon at @xmath30 .
a vistec ebpg5000 + electron beam lithography system was used to define the metasurface pattern in the zep-520a positive resist ( @xmath31300 nm , spin coated at 5000 rpm for 1 min ) .
the pattern was developed in a resist developer for 3 minutes ( zed - n50 from zeon chemicals ) .
an approximately 100-nm - thick aluminum oxide layer was deposited on the sample using electron beam evaporation , and was lifted off reversing the pattern .
the patterned aluminum oxide hard mask was then used to dry etch the a - si layer in a 3:1 mixture of @xmath32 and @xmath33 plasma .
after etching , the mask was removed using a 1:1 solution of ammonium hydroxide and hydrogen peroxide at @xmath34 c. * measurement procedure .
* devices were measured using a fiber placed @xmath311100 @xmath24 m away from the metasurface ( 500 @xmath24 m substrate thickness plus 600 @xmath24 m distance between the fiber and the substrate ) , and a custom built microscope with @xmath31100x magnification ( supplementary information fig .
s2 ) . at 915 nm ,
a fiber coupled semiconductor laser with a single mode fiber with an angled polished connector was used for illumination .
fiber tip angle was adjusted to correct for the angled connector cut .
a 100x objective lens ( olympus umplanfl , na=0.95 ) and a tube lens ( thorlabs ac254 - 200-b - ml ) with a focal distance of 20 cm were used to image intensity at planes of interest to a ccd camera ( coolsnap k4 , photometrics ) .
a calibration sample with known feature sizes was measured to find the pixel - size transferred to the object plane .
the objective was moved with a translation stage to image different planes around the focus .
the plotted axial plane intensities are upsampled 2:1 in the axial direction ( 4 @xmath24 m adjacent measurement planes distance to 2 @xmath24 m ) to achieve a smoother graph . for focusing efficiency measurement at 915 nm ,
a 20-@xmath24m - diameter pinhole was placed in the focal plane of the metasurface lens to only let the focused light pass through .
the pinhole was made by wet etching a 20 @xmath24 m hole in a thick layer of chrome deposited on a fused silica substrate .
a power meter ( thorlabs pm100d with photodetector head thorlabs s122c ) was then used to measure the power after the pinhole , and the output power of the fiber .
the efficiency was calculated as the ratio of these two powers .
the reported measured efficiency is therefore a lower bound on the actual efficiency as it does not include reflection from the substrate , and two reflections from the two sides of the pinhole glass substrate .
a similar setup was used for measurements at 1550 : a tunable 1550 nm laser ( photonetics tunics - plus ) was used with a single mode fiber for illumination .
the same 100x objective was used with a 20 cm tube lens ( thorlabs ac254 - 200-c - ml ) to image the intensity in the object plane to a camera ( digital camir 1550 by applied scintillation technologies ) .
the camera has a significantly non - uniform sensitivity for different pixels which leads to high noise level of the images captured by the camera ( as seen in fig .
the nonphysical high frequency noise of the images ( noise with frequencies higher than twice the free space propagation constant ) was removed numerically to reduce the noise in the axial intensity patterns .
the intensity pattern was also upsampled in the axial direction from the actual 4 @xmath24 m distance between adjacent measurement planes , to 2 @xmath24 m to achieve a smoother intensity profile . to find the focused power ,
the focal plane of the lens was imaged using the microscope to a photodetector . a 2 mm iris in the image plane ( corresponding to 20 @xmath35 in the object plane ) was used to limit the light reaching the photodetector .
the input power was measured by imaging the fiber facet to the photodetector using the same setup and without the iris .
the efficiency was obtained by dividing the focused power by the input power . *
supplementary information .
* discussion of 915 nm efficiencies , measurement setup figure , measurement results for the lower na lens , simulation results of two 915 nm blazed gratings .
this work was supported by samsung electronics .
a.a and y.h were also supported by darpa , and s.k was supported as part of the department of energy ( doe ) light - material interactions in energy conversion energy frontier research center under grant no .
de - sc0001293 .
the device nanofabrication was performed at the kavli nanoscience institute at caltech . * author contributions * + e. a. , a. a. , and a. f. conceived the experiments .
e. a. , a. a. , s. k , and y. h. performed the simulations and fabricated the devices .
e. a. and a. a. performed the measurements , and analyzed the data .
e. a. , a. a. , s. k. , and a. f. co - wrote the manuscript .
all authors discussed the results and commented on the manuscript .
10 url # 1`#1`urlprefix[2]#2 [ 2][]#2 , , , & .
_ _ * * , ( ) . , , , & .
_ _ * * , ( ) . , , , & .
_ _ * * , ( ) . , , & .
_ _ * * , ( ) . , & .
_ _ * * ( ) . & .
_ _ * * , ( ) .
_ et al . _ . _
_ * * , ( ) .
_ _ * * ( ) . , , & .
_ _ * * , ( ) . .
_ _ * * , ( ) . , & _ _ ( , ) . & _ _ ( , ) . .
_ _ * * , ( ) . .
_ _ * * , ( ) . .
_ _ * * , ( ) . & .
_ _ * * , ( ) . & .
_ _ * * , ( ) . . ,
( ) . , & .
_ _ * * , ( ) . , , & .
_ _ * * , ( ) .
_ et al . _ . _
_ * * , ( ) . & .
_ _ ( ) . , , & .
_ _ * * , ( ) .
_ et al . _ . _
_ * * , ( ) .
_ et al . _ . _
_ * * , ( ) . & .
_ _ * * , ( ) .
_ et al . _ . _
_ * * , ( ) . & .
_ _ * * , ( ) .
to understand the reasons behind the low efficiency of the lenses at 915 nm , two double wavelength blazed gratings were designed using the proposed meta - molecule scheme .
one grating with a small deflection angle ( 5 degrees ) and another one with a larger angle ( 20 degrees ) were simulated at 915 nm using meep , and power loss channels were analyzed in both cases ( fig .
both gratings were chosen to be 2 meta - molecules wide in the @xmath36 direction , so that periodic boundary conditions in this direction can be used in fdtd .
the 5 degree grating is 322 lattice constants long in @xmath37 direction , while the 20 degree one is 146 lattice constants .
the lengths are chosen such that the grating phases at 915 nm and 1550 nm are both almost repeated after the chosen lengths ( fig .
an x - polarized plane - wave normally incident from the fused silica side was used as excitation in both simulations , and the transmitted and reflected electric and magnetic field intensities were calculated about a wavelength apart from the meta - molecules .
the transmitted fields were further propagated using a plane wave expansion program , and the resulting fields in an area of length 30 @xmath24 m around the center can be seen in fig .
s3b and s3c for 5 degree and 20 degree gratings , respectively .
the field distributions outside of the areas shown here look similar to the ones shown . in both cases ,
a dominant plane wave propagating in the design direction is observed , along with some distortions .
angular distribution of power in transmission and reflection is analyzed using the fourier transform of the fields .
the resulting power distributions are shown in fig .
s3d and s3e for 5 degrees and 20 degrees , respectively . while the average power transmission of meta - molecules used in both gratings ( found from the data in fig .
2 g ) is slightly above 73@xmath0 , only 36@xmath0 and 22@xmath0 of the incident power is directed to 5 and 20 degrees for the corresponding gratings .
the actual total transmitted powers are 56@xmath0 and 50@xmath0 for the 5 and 20 degree gratings , showing that an additional @xmath3120@xmath0 of the power gets reflected as a result of the introduced aperiodicity . because of the relatively large lattice constant
, even a small aperiodicity can result in generation of propagating modes in the substrate , thus the reflection is considerably higher for the gratings than for a perfectly periodic lattice . from
the 56@xmath0 transmitted power in the 5 degree grating , 20@xmath38 is lost to diffraction to other angles . from fig .
s3b and s3c we can see there are distortions in the transmitted field .
these distortions , mainly due to the low transmission amplitude of some of the meta - molecules and their phase errors result in the transmitted power being diffracted to other angles .
besides , it is seen that power loss to other angles both in reflection and transmission is higher for larger grating angles .
this is due to the need for finer sampling of the wave front for waves with steeper angles . the lower efficiency for gratings with larger angles results in lower efficiency of lenses with higher numerical apertures which need bending light with larger angles . | metasurfaces are nano - structured devices composed of arrays of subwavelength scatterers ( or meta - atoms ) that manipulate the wavefront , polarization , or intensity of light . like other diffractive optical devices ,
metasurfaces suffer from significant chromatic aberrations that limit their bandwidth . here
, we present a method for designing multiwavelength metasurfaces using unit cells with multiple meta - atoms , or meta - molecules .
transmissive lenses with efficiencies as high as 72@xmath0 and numerical apertures as high as 0.46 simultaneously operating at 915 nm and 1550 nm are demonstrated . with proper
scaling , these devices can be used in applications where operation at distinct known wavelengths is required , like various fluorescence microscopy techniques . over the last few years
, a new wave of interest has risen in nano - structured diffractive optical elements due to advances in nano - fabrication technology @xcite . from the multiple designs proposed so far ,
dielectric transmitarrays @xcite are some of the most versatile metasurfaces because they provide high transmission and subwavelength spatial control of both polarization and phase .
several diffractive optical elements , including high numerical aperture lenses and simultaneous phase and polarization controllers have recently been demonstrated with high efficiencies @xcite .
these devices are based on subwavelength arrays of high refractive index dielectric nano - resonators ( scatterers ) with different geometries , fabricated on a planar substrate .
scatterers with various geometries impart different phases to the transmitted light , shaping its wavefront to the desired form
. one main drawback of almost all of metasurface devices , particularly the ones with spatially varying phase profiles like lenses and gratings , is their chromatic aberration : their performance changes as the wavelength is varied @xcite .
refractive optical elements also suffer from chromatic aberrations ; however , their chromatic aberrations , which stem from material dispersion , are substantially smaller than those of the diffractive elements @xcite .
an ideal refractive lens made of a dispersionless material will show no chromatic aberration .
on the other hand , the chromatic aberration of diffractive elements mainly comes from the geometrical arrangement of the device .
early efforts focused on making achromatic diffractive lenses by cascading them in the form of doublets and triplets @xcite , but it was later shown that it is fundamentally impossible to make a converging achromatic lens which has a paraxial solution ( i.e. is suitable for imaging ) by only using diffractive elements @xcite .
although diffractive - refractive combinations have successfully been implemented to reduce chromatic aberrations , they are mostly useful in deep uv and x - ray wavelengths where materials are significantly more dispersive @xcite . more recently , wavelength and polarization selectivity of metallic meta - atoms have been used to fabricate a fresnel zone plate lens that operates at two distinct wavelengths with different orthogonal polarizations @xcite .
besides undesired multi - focus property of fresnel zone plates and efficiency limitations of metallic metasurfaces @xcite , the structure works only with different polarizations at the two wavelengths .
the large phase dispersion of dielectric meta - atoms with multiple resonances has also been exploited to compensate for the phase dispersion of metasurfaces at three wavelengths @xcite , but the cylindrical lens demonstrated with this technique is polarization dependent and has low numerical aperture and efficiency .
multiwavelength metasurfaces based on elliptical apertures in metallic films are demonstrated in @xcite , but they are also polarization dependent and have a multi - focus performance . an achromatic metasurface design is proposed in @xcite based on the idea of dispersionless meta - atoms ( i.e. meta - atoms that impart constant delays ) . unfortunately , this idea only works for metasurface lenses with one fresnel zone limiting the size and numerical aperture of the lenses . for a typical lens with tens of fresnel zones , dispersionless meta - atoms will not reduce the chromatic dispersion as we will shortly discuss . in the following
we briefly discuss the reason for chromatic dispersion of metasurface lenses , and then propose a method for correcting this dispersion at distinct wavelengths .
we also present experimental results demonstrating corrected behavior of a lens realized using the proposed method .
in diffractive lenses , chromatic dispersion mainly manifests itself through a significant change of focal length as a function of wavelength @xcite .
this change is schematically shown in fig .
1a , along with a metasurface lens corrected to have the same focal distance at a few wavelengths . to better understand the underlying reasons for this chromatic dispersion
, we consider a hypothetical aspherical metasurface lens .
the lens is composed of different meta - atoms which locally modify the phase of the transmitted light to generate the desired wavefront .
we assume that the meta - atoms are dispersionless in the sense that their associated phase changes with wavelength as @xmath1 like a piece of dielectric with a constant refractive index . here
@xmath2 is the wavelength and @xmath3 is an effective parameter associated with the meta - atoms that controls the phase ( @xmath3 can be an actual physical parameter or a function of physical parameters of the meta - atoms ) .
we assume that the full 2@xmath4 phase needed for the lens is covered using different meta - atoms with different values of @xmath3 .
the lens is designed to focus light at @xmath5 ( fig .
1b ) to a focal distance @xmath6 , and it s phase profile in all fresnel zones matches the ideal phase profile at this wavelength . because of the specific wavelength dependence of the dispersionless meta - atoms ( i.e. proportionality to 1/@xmath2 ) , at a different wavelength ( @xmath7 ) the phase profile of the lens in the first fresnel zone follows the desired ideal profile needed to maintain the same focal distance ( fig .
1b ) .
however , outside the first fresnel zone , the actual phase profile of the lens deviates substantially from the desired phase profile . due to the jumps at the boundaries between the fresnel zones , the actual phase of the lens at @xmath7 is closer to the ideal phase profile at @xmath5 than the desired phase profile at @xmath7 . in the inset of fig .
1b the effective meta - atom parameter @xmath3 is plotted as a function of distance to the center of the lens @xmath8 .
the jumps in @xmath3 coincide with the jumps in the phase profile at @xmath7 . in fig .
1c , the simulated intensity profile of the same hypothetical lens is plotted at a few wavelengths close to @xmath5 .
the focal distance changes approximately proportional to @xmath9 .
this wavelength dependence is also observed in fresnel zone plates @xcite , and for lenses with wavelength independent phase profiles @xcite ( the @xmath9 dependence is exact in the paraxial limit , and approximate in general ) .
this behavior confirms the previous observation that the phase profile of the lens at other wavelengths approximately follows the phase profile at the design wavelength .
therefore , the chromatic dispersion of metasurface lenses mainly stems from wrapping the phase , and the dependence of the phase on only one effective parameter ( e.g. @xmath3 ) whose value undergoes sudden changes at the zone boundaries .
as we show in the following , using two parameters to control metasurface phase at two wavelengths can resolve this issue , and enable lenses with the same focal lengths at two different wavelengths .
the metasurface platform we use in this work is based on amorphous silicon ( a - si ) nano - posts on a fused silica substrate ( fig .
2a ) .
the nano - posts are placed on the vertices of a hexagonal lattice , and locally sample the phase to generate the desired phase profile @xcite . for a fixed height , the transmission phase of a nano - post can be controlled by varying its diameter .
the posts height can be chosen such that at a certain wavelength the whole 2@xmath4 phase shift is covered , while keeping the transmission amplitude high . to design a metasurface that works at two different wavelengths , a unit cell consisting of four different nano - posts ( fig .
2b ) was chosen because it has more parameters to control the phases at two wavelengths almost independently . as molecules consisting of multiple atoms
form the units for more complex materials , we call these unit cells with multiple meta - atoms _ meta - molecules_. the meta - molecules can also form a periodic lattice ( in this case hexagonal ) , and effectively sample the desired phase profiles simultaneously at two wavelengths .
the lattice is subwavelength at both wavelengths of interest ; therefore , the non - zero diffraction orders are not excited .
in general , the four nano - posts can each have different diameters and distances from each other .
however , to make the design process more tractable , we choose three of the four nano - posts with the same diameter @xmath10 and the fourth post with diameter @xmath11 , and place them in the centers of the hexagons at a distance @xmath12 ( as shown in fig .
2b ) .
therefore , each meta - molecule has two parameters , @xmath10 and @xmath11 , to control the phases at two wavelengths . for this demonstration , we choose two wavelengths of 1550 nm and 915 nm , because of the availability of lasers at these wavelengths .
a periodic array of meta - molecules was simulated to find the transmission amplitude and phase as shown in fig . 2c
( see methods for simulation details ) .
the simulated transmission amplitude and phase for 1550 nm ( @xmath13 and @xmath14 ) and 915 nm ( @xmath15 and @xmath16 ) are plotted as functions of @xmath10 and @xmath11 in fig .
2d and 2e . in these simulations
the lattice constant ( @xmath17 ) was set to 720 nm and the posts were 718 nm tall .
since the two wavelengths are not close , the ranges of @xmath10 and @xmath11 must be very different in order to properly control the phases at 1550 nm and 915 nm .
for each desired combination of the phases @xmath14 and @xmath16 in the @xmath18 range at the two wavelengths , there is a corresponding @xmath10 and @xmath11 pair that minimizes the total transmission error which is defined as @xmath19 .
these pairs are plotted in fig .
2f as a function of @xmath14 and @xmath16 .
using the complex transmission coefficients ( i.e. @xmath20 and @xmath21 ) in error calculations results in automatically avoiding resonance areas where the phase might be close to the desired value , but transmission is low .
the corresponding transmission amplitudes for the chosen meta - molecules are plotted in fig .
2 g , and show this automatic avoidance of low transmission meta - molecules . in the lens design process , the desired transmission phases of the lens are sampled at the lattice points at both wavelengths resulting in a @xmath22 pair at each lattice site .
using the plots in fig .
2f , values of the two post diameters are found for each lattice point .
geometrically , the values of the two diameters are limited by @xmath23 . besides , we set a minimum value of 50 nm for the gaps between the posts to facilitate the metasurface fabrication .
a double wavelength aspherical lens was designed using the proposed platform to operate at both 1550 nm and 915 nm .
the lens has a diameter of 300 @xmath24 m and focuses the light emitted from single mode fibers at each wavelength to a focal plane 400 @xmath24 m away from the lens surface ( the corresponding paraxial focal distance is 286 @xmath24 m , thus the numerical aperture is 0.46 ) .
the lens was fabricated using standard nanofabrication techniques : a 718-nm - thick layer of a - si was deposited on a fused silica substrate , the lens pattern was generated using electron beam lithography and transferred to the a - si layer using aluminum oxide as a hard mask ( see methods for more details on fabrication ) .
optical and scanning electron microscope images of the lens and nano - posts are shown in fig .
3 . for characterization ,
the fabricated metasurface lens was illuminated by light emitted from the end facet of a single mode fiber , and the transmitted light intensity was imaged at different distances from the lens using a custom built microscope ( see methods and supplementary information fig .
s1 for measurement details ) .
measurement results for both wavelengths are plotted in fig .
4a-4d .
figures 4a and 4b show the intensity profiles in the focal plane measured at 915 nm and 1550 nm , respectively .
the measured full width at half maximum ( fwhm ) is 1.9 @xmath24 m at 915 nm , and 2.9 @xmath24 m at 1550 nm .
the intensity measured at the two axial plane cross sections are plotted in fig . 4c and 4d for the two wavelengths . a nearly diffraction limited focus
is observed in the measurements , and no other secondary focal points with comparable intensity is seen . to confirm the diffraction limited behavior ,
a perfect phase mask was simulated using the same illumination as the measurements .
the simulated fwhm s were 1.6 @xmath24 m and 2.75 @xmath24 m for 915 nm and 1550 nm respectively ( see methods for details on the simulation ) .
focusing efficiencies of 22@xmath0 and 65@xmath0 were measured for 915 nm and 1550 nm , respectively . focusing efficiency
is defined as the ratio of the power passing through a 20-@xmath24m - diameter disk around the focus to the total power incident on the lens .
anther lens with a longer focal distance of 1000 @xmath24 m ( thus a lower na of 0.29 ) was fabricated and measured with the same platform and method .
measurement results for those devices are presented in supplementary information fig .
s2 .
slightly higher focusing efficiencies of 25@xmath0 and 72@xmath0 were measured at 915 nm and 1550 nm for those devices . for comparison ,
a lens designed with the same method and based on the same metasurface platform is simulated using finite difference time domain ( fdtd ) method with a freely available software ( meep ) @xcite . to reduce the computational cost ,
the simulated lens is four times smaller and focuses the light at 100 @xmath24 m distance . because of the equal numerical apertures of the simulated and fabricated devices , the focal intensity distributions and the focal depths are comparable .
the simulation results are shown in fig .
4e-4h .
figures 4e and 4f show the simulated focal plane intensity of the lens at 915 nm and 1550 nm , respectively .
the simulated fwhm is 1.9 @xmath24 m at 915 nm and 3 @xmath24 m at 1550 nm , both of which are in accordance with their corresponding measured values .
also , the simulated intensity distributions in the axial cross section planes , which are shown in fig . 4 g and 4h , demonstrate only one strong focal point .
the focusing efficiency was found to be 32@xmath0 at 915 nm , and 73@xmath0 at 1550 nm .
we attribute the difference in the simulated and measured efficiencies to fabrication imperfections and measurement artifacts ( see methods for details about measurements ) . the efficiency at 915 nm is found to be lower than what expected both in measurement and fdtd simulation .
while the average power transmission of the selected meta - molecules is about 73@xmath0 as calculated from fig .
2 g , the simulated focusing efficiency is about 32@xmath0 . to better understand the reasons for this difference , two blazed gratings with different angles
were designed and simulated for both wavelengths using the same meta - molecules ( see supplementary information section 1 and fig .
s3 ) .
it is observed that for the gratings ( that are aperiodic ) , a significant portion of the power is diffracted to other angles both in reflection and transmission . besides , the power lost into diffractions to other angles is higher for the grating with larger deflection angle .
the main reason for the large power loss to other angles is the the relatively large lattice constant .
the chosen lattice constant of @xmath25 720 nm is just slightly smaller than 727 nm , the lattice constant at which the first order diffracted light starts to propagate in the glass substrate for a perfectly periodic structure .
thus , even a small deviation from perfect periodicity can result in light diffracted to propagating orders .
besides , the lower transmission of some meta - molecules reduces the purity of the plane wave wavefronts diffracted to the design angle .
furthermore , the desired phase profile of high numerical aperture lenses can not be sampled at high enough resolution using large lattice constants .
therefore , as shown in this work , a lens with a lower numerical aperture has a higher efficiency .
there are a few methods to increase the efficiency of the lenses at 915 nm : the lattice constant is bound by the geometrical and fabrication constraint : @xmath26 , hence the smallest value of @xmath27 that gives full phase coverage at the longer wavelength sets the lower bound for the lattice constant .
this limit can usually be decreased by using taller posts , however , that would result in a high sensitivity to fabrication errors at the shorter wavelength .
thus , a compromise should be made here , and higher efficiency designs might be possible by more optimum selections of the posts height and the lattice constant
. the lattice constant can also be smaller if less than the full 2@xmath4 phase shift is used at 1550 nm ( thus lower efficiency at 1550 nm ) .
in addition , as explained earlier , in minimizing the total transmission error equal weights are used for 915 nm and 1550 nm . a higher weight for 915 nm might result in higher efficiency at this wavelength , probably at the expense of 1550 nm efficiency . for instance , if we optimize the lens only for operation at 915 nm , devices with efficiencies as high as 80@xmath0 are possible @xcite .
the approach presented here can not be directly used to correct for chromatic dispersion over a continuous bandwidth ; the multiwavelength lenses still have chromatic dispersion much like normal metasurface lenses in narrow bandwidths around the corrected wavelengths . for achieving zero chromatic dispersion over a narrow bandwidth
, the meta - atoms should independently control the phase at two very close wavelengths .
high quality factor resonances must be present for the meta - atom phase to change rapidly over a narrow bandwidth , and such resonances will result in high sensitivities to fabrication errors that would make the metasurface impractical .
the meta - molecule platform , used here to correct for chromatic aberration at specific wavelengths , can also be used for applications where different functionalities at different wavelengths are desired .
for instance , it can be used to implement a lens with two given focal distances at two wavelengths , or a lens converging at one wavelength and diverging at the other .
multi - wavelength operation is necessary in various microscopy applications where fluorescence is excited at one wavelength and collected at another . in this work
we have used only two of the degrees of freedom of the meta - molecules , but increased functionalities at more than two wavelengths can be achieved by making use of all the degrees of freedom .
operation at more than two wavelengths enables applications in color display technologies or more complex fluorescence imaging techniques . |
the fractional fourier transform is becoming increasingly more important due to an ever - growing list of applications in signal processing , optics and quantum mechanics and in other areas of science @xcite . in order to define this object
one needs first to diagonalize the fourier transform operator @xmath0 , and that is where hermite functions @xmath1 become indispensable .
it is well - known that hermite functions are the eigenfunctions of the fourier transform operator @xmath2 and that they form a complete orthonormal basis of @xmath3 . in other words ,
the orthogonal basis of hermite functions diagonalizes the fourier transform operator @xmath0 and this allows us to define the fractional power @xmath4 the fractional fourier transform .
of course , there are infinitely many ways to choose an eigenbasis of the fourier transform , each of them would lead to a different version of the fractional fourier transform .
what makes hermite functions
so special among all other eigenfunctions is that they are analytically tractable and they have many useful properties .
this analytical tractability and mehler s formula @xcite ( which gives an explicit expression for the integral kernel of the fractional fourier transform ) have sealed the standing of hermite functions as the canonical set of the eigenfunctions for fourier transform .
.dimensions of the eigenspaces @xmath5 of the discrete fourier transform @xmath6 , corresponding to the eigenvalues @xmath7 . [ cols="<,^,^,^,^",options="header " , ] compared to the continuous fourier transform , the situation with the eigenvectors of the discrete fourier transform ( dft ) is much more complex .
let us first introduce several necessary definitions and notations .
the ( centered ) discrete fourier transform @xmath6 is a linear map that sends a vector @xmath8 into a vector @xmath9 according to the rule @xmath10 where @xmath11 denotes the set of integers @xmath12 everywhere in this paper we will follow the convention that all vectors will be denoted by bold lower case letters and the elements of a vector will be labeled by the set @xmath11 .
before we begin discussing the eigenvectors of the dft , let us say a few words about its eigenvalues .
the eigenvalues of both continuous and discrete fourier transform are given by @xmath13 , @xmath14 .
in the continuous case each eigenspace is infinite dimensional , whereas the multiplicities of eigenvalues in the discrete case are presented in the table [ tab_multiplicities ] .
this result was obtained in 1921 by schur , however it can be easily derived from a much earlier result of gauss on the law of quadratic reciprocity , which is essentially a statement about the trace of the matrix corresponding to the linear transformation @xmath6 .
the formula for the multiplicities of eigenvalues was rediscovered by mcclellan and parks @xcite in 1972 , and since then there have appeared many other proofs ( see @xcite for a very simple proof based on the vandermonde determinant formula ) .
returning to our discussion of eigenvectors of the dft , it seems that the first example of an eigenbasis was constructed by mcclellan and parks @xcite as a by - product of their proof of the results in table [ tab_multiplicities ] .
other constructions of eigenvectors of the dft include a number - theoretic construction @xcite and a representation theory approach @xcite .
mehta @xcite has constructed an eigenbasis based on theta functions .
one important approach for finding orthogonal eigenvectors of the dft is based on certain symmetric tridiagonal ( or almost tridiagonal ) matrices which commute with the dft .
the eigenvectors of these matrices give an orthogonal eigenbasis of the dft .
such matrices were first discovered by dickinson and steiglitz @xcite and by grunbaum @xcite ( see also @xcite for more recent developments ) .
the above discussion clearly shows that there is an abundance of different sets of eigenvectors of the dft .
how can we choose the best " among them ?
first we need to define what we mean by the best " .
an ideal set of eigenvectors of the discrete fourier transform @xmath6 should satisfy the following three requirements : the eigenvectors should be orthogonal , they should converge to the corresponding hermite functions as @xmath15 and they should have at least some degree of analytical tractability ( so that we are able to compute them ) . judging by these standards , most of the above candidates can be disqualified . for example , the eigenbasis of mcclellan and parks is explicit but not orthogonal and it does not approximate hermite functions ; the eigenbasis of mehta @xcite does approximate hermite functions but it is not orthogonal ; the eigenbasis constructed in @xcite is orthogonal , but it requires @xmath16 to be a certain prime number and it does not seem to approximate hermite functions .
in fact , the only sets of eigenvectors that do satisfy the above requirements come from the commuting symmetric matrix approach of dickinson and steiglitz @xcite and grunbaum @xcite , though in both cases the eigenvectors are not given explicitly and to find them one needs to diagonalize a tridiagonal symmetric matrix . not much is known about _
explicit _ hermite - type eigenvectors of the dft .
in fact , it seems that the only result in this direction comes from the recent paper by kong @xcite , who gives two examples of such eigenvectors .
more precisely , when @xmath17 the vector @xmath18 with @xmath19 is an eigenvector of @xmath6 , and the same is true when @xmath20 for the vector @xmath21 with @xmath22 numerical evidence given in @xcite suggests that these two vectors approximate the first two hermite functions .
what is also interesting about these two eigenvectors is that they are unique in a certain sense .
it follows from the proof of the main result in @xcite ( though it is not explicitly stated there ) that the only eigenvector of the dft of the dimension @xmath17 which has zero elements @xmath23 for @xmath24 is the vector @xmath25 defined in . in this paper
we provide the first known example of an explicit orthogonal eigenbasis of the dft for which the individual eigenvectors seem to approximate the corresponding hermite functions .
first of all , we construct a basis of @xmath26 which consists of even and odd vectors similar to kong s vectors @xmath25 and @xmath27 given above . these vectors behave nicely under the dft and we prove that they are unique in a certain sense .
the proof of these results uses the methods of kong @xcite combined with the @xmath28-binomial theorem . from this basis of @xmath26
we construct an eigenbasis of the dft following the same procedure as mcclellan and parks @xcite . as a corollary of this construction
, we obtain a new proof of the result on the multiplicities of the eigenvalues of the dft presented in table [ tab_multiplicities ] .
finally , we apply the gramm - schmidt orthogonalization procedure to the basis of each eigenspace of the dft .
this gives us an orthonormal basis of @xmath26 , consisting of the eigenvectors of the discrete fourier transform @xmath6 .
we prove that the first eight eigenvectors converge to the corresponding hermite functions as @xmath15 and we conjecture that this convergence holds true for all eigenvectors .
the paper is organized as follows . in section [ section_results ]
we present our results for the case @xmath29 .
the proofs of these results are collected in section [ section_proofs ] and in the appendix we give the corresponding formulas for all remaining cases @xmath30 .
we begin by stating some definitions and notations that will be used throughout this paper .
we denote by @xmath31 the eigenspaces of the dft corresponding to eigenvalues @xmath32 ( their dimensions are given in table [ tab_multiplicities ] ) .
the dot product of two vectors @xmath25 and @xmath27 will be denoted by @xmath33 and the euclidean norm of a vector in @xmath34 is defined as @xmath35 .
we denote by @xmath36 and @xmath37 the floor and the ceiling function . from now on we will drop the subscript @xmath16 in the notation of the discrete fourier transform and will write simply @xmath0 instead of @xmath6 there should be no confusion as the continuous fourier transform will not be used anymore .
we say that a vector @xmath38 is _ even _
( _ odd _ ) if @xmath39 ( respectively @xmath40 ) for all @xmath41 such that @xmath42 .
an equivalent definition can be given as follows : when @xmath43 the even ( odd ) vectors satisfy the condition @xmath44 ( respectively , @xmath45 ) for @xmath46 .
when @xmath47 the even vectors satisfy @xmath44 for @xmath48 ( note that there is no restriction on @xmath49 ) and the the odd vectors satisfy @xmath50 for @xmath48 and @xmath51 .
the following two properties are well known and will be used extensively later : if @xmath18 is an even vector and @xmath52 is an odd vector , then * @xmath53 ( @xmath54 ) is a real even ( respectively , real odd ) vector ; * @xmath55 and @xmath56 ; the following object will appear frequently in this paper : the sequence @xmath57 is defined by @xmath58 and @xmath59 .
this sequence satisfies @xmath60 property follows at once from the definition while property can be derived by writing the polynomial @xmath61 as a product of @xmath62 linear factors and then taking the limit as @xmath63 .
we denote by @xmath64 _ the support _ of a vector @xmath65 , which is the set of all indices @xmath66 such that @xmath67 .
following kong @xcite we define _ the signal length _ ( or simply , _ the length _ ) of a vector @xmath68 as @xmath69 we record the following important relationship between the length of a vector and its dft : for any nonzero vectors @xmath70 we have @xmath71 this result ( which reminds one of the uncertainty principle ) follows from theorem 1 in @xcite .
we include the proof of this result in section [ section_proofs ] for convenience of the reader .
the next theorem is our first main result .
it gives the exact number of even / odd vectors that are extreme " in the sense that the quantity @xmath72 is the smallest possible among all even / odd vectors .
[ thm_support ] @xmath73 assume that @xmath74 . * if @xmath43 there exist only @xmath75 nonzero even vectors satisfying @xmath76 and only @xmath77 nonzero odd vectors such that @xmath78 . *
if @xmath47 there exist only @xmath79 nonzero even vectors satisfying @xmath80 and @xmath81 and only @xmath79 nonzero odd vectors such that @xmath82 . the results and definitions that we have presented so far are valid for all integer values of @xmath83 . from now on we will concentrate on the case when @xmath84 .
the corresponding results for all remaining cases @xmath85 are collected in the appendix .
our next goal is to find explicit expression for the vectors described in theorem [ thm_support ] and to compute their discrete fourier transform .
[ theorem_4n_plus1_n1 ] @xmath73 * the even vectors @xmath86 described in theorem [ thm_support](i ) are given by @xmath87 these vectors are linearly independent and they satisfy @xmath88 * the odd vectors @xmath89 described in theorem [ thm_support](i ) are given by @xmath90 these vectors are linearly independent and they satisfy @xmath91 using the sum - to - product formula for the cosine function and identity it is easy to check that formula is equivalent to @xmath92 in particular , the vector @xmath93 is a scalar multiple of the vector @xmath25 discovered by kong @xcite ( see ) .
let us highlight some important properties of the vectors @xmath94 and @xmath95 .
first of all , property implies that @xmath96 in particular , we have @xmath97 and it is clear that the vectors @xmath94 and @xmath95 satisfy @xmath98 so these are precisely the vectors described in theorem [ thm_support ] .
the vectors @xmath94 and @xmath95 should be viewed as the discrete counterparts of the gaussian functions @xmath99 and @xmath100 ( see figure [ fig_l40 ] ) .
there are indeed many analogies .
first of all , the elements of these vectors , @xmath101 and @xmath102 , converge to the corresponding gausian functions as @xmath15 ( this follows from lemma [ lemma_v0 ] in section [ section_proofs ] ) .
second , the identities and are discrete analogues of the integral formulas @xmath103 finally , the vectors @xmath94 and @xmath95 are connected by the following identity @xmath104 which we will establish in section [ section_proofs ] .
this result is the finite difference counterpart of the formula @xmath105 our next goal is to construct a basis for each eigenspace @xmath106 , @xmath107 , @xmath108 and @xmath109 of the dft .
this construction is based on theorem [ theorem_4n_plus1_n1 ] and the following property : if @xmath25 is an even vector ( @xmath27 is an odd vector ) then @xmath110 ( respectively , @xmath111 ) is an eigenvector of the discrete fourier transform @xmath0 with the corresponding eigenvalue @xmath13 ( respectively , @xmath112 ) .
this property can be easily established using the fact that @xmath55 and @xmath56 for every even vector @xmath25 and every odd vector @xmath27 .
[ prop_4n_plus1_n2 ] define @xmath16 vectors @xmath113 , @xmath114 , @xmath115 , @xmath116 as follows : @xmath117 the vectors @xmath113 ( respectively , @xmath114 , @xmath115 and @xmath116 ) form a basis for the eigenspace @xmath106 ( respectively , @xmath107 , @xmath108 and @xmath109 ) .
the results of corollary [ prop_4n_plus1_n2 ] imply that the eigenvalues @xmath118 of the discrete fourier transform have corresponding multiplicities @xmath119 , which agrees with table [ tab_multiplicities ] .
the eigenbasis constructed in proposition [ prop_4n_plus1_n2 ] is not orthogonal .
our next step is to build an orthonormal basis via the gramm - schmidt algorithm .
[ def_phi ] define @xmath16 vectors @xmath120 as follows : @xmath121 where the vectors @xmath122 , @xmath123 , @xmath124 , @xmath125 are obtained by gramm - schmidt algorithm starting from @xmath113 , @xmath114 , @xmath115 , @xmath116 : @xmath126 @xmath127 the following theorem is our main result in this paper .
recall that hermite functions @xmath128 are defined by .
[ thm_convergence ] @xmath73 * the vectors @xmath129 form an orthonormal basis in @xmath130 consisting of eigenvectors of the discrete fourier transform : for all @xmath131 we have @xmath132 * for @xmath133 we have @xmath134 where @xmath135 .
+ the first six ( respectively , twelve ) eigenvectors @xmath136 and the corresponding scaled hermite functions are shown on figure [ fig_l10 ] for @xmath137 ( respectively , figure [ fig_l100 ] for @xmath138 ) .
the results presented on figure [ fig_l100 ] suggest that the convergence in is not restricted to the first eight eigenvectors , and that it should hold for all of them .
thus we formulate * conjecture 1 : * for all @xmath139 we have @xmath140 our proof of theorem [ thm_convergence ] which is presented in section [ section_proofs ] is essentially a proof by verification , and it does not provide an intuitive reason as to why this convergence should hold . our approach would probably work for @xmath141 , though larger values of @xmath142 would require a prohibitive amount of algebraic computations .
thus our method of proof of theorem [ thm_convergence ] is probably not the right way for proving conjecture 1 .
we hope that some other properties of the vectors @xmath94 and @xmath95 ( which we may have overlooked ) will lead to a simple and insightful proof of this conjecture .
figure [ fig_l40 ] provides a bird s eye view " of the eigenbasis @xmath143 and compares it with hermite functions .
in particular , on figure [ l40_f3 ] we show the values of @xmath142 and @xmath144 where @xmath145 is either positive , or negative or equal to zero . from these results
it seems that the vectors @xmath136 have the same number of zero crossings as the corresponding hermite functions @xmath146 .
let us make this statement precise .
we define the number of zero crossings of a vector @xmath147 as the number of pairs @xmath148 such that ( i ) @xmath149 , ( ii ) @xmath150 , ( iii ) @xmath151 and ( iv ) @xmath152 for all @xmath153 .
based on numerical evidence obtained from high - precision computations for many different values of @xmath16 ( see figure [ l40_f3 ] for @xmath154 ) we arrive at * conjecture 2 : * for @xmath131 the vector @xmath136 has exactly @xmath142 zero crossings .
figure [ l40_f3 ] also illustrates the following result @xmath155 which follows easily from and the above construction of @xmath136 .
in particular , this result implies that about 25% of all elements
@xmath145 of the eigenbasis are equal to zero . in view of theorem [ thm_support ]
the following question naturally arises : are there other sets of eigenvectors @xmath156 of the dft such that @xmath157 satisfies and has exactly @xmath142 zero crossings ?
in other words , is the eigenbasis @xmath143 unique in a similar sense in which the vectors @xmath94 and @xmath95 are unique ?
for a nonzero vector @xmath158 we denote by @xmath159 the rational function @xmath160 we also denote @xmath161 .
it is clear that @xmath162 if and only if @xmath163 for all @xmath164 .
the following simple observation will play the main role in the proof of theorem [ thm_support ] : @xmath165 to establish this fact we define @xmath166 and note that @xmath167 is a polynomial of degree @xmath168 . to give a preview of our method of the proof of theorem [ thm_support ] ,
let us prove kong s result ( see ( * ? ? ?
* theorem 1 ) ) using the above property .
assume that for some nonzero vector @xmath158 we have @xmath169 and @xmath170 where @xmath171 .
since @xmath170 it follows that @xmath172 for at least @xmath173 indices @xmath164 . in view of identity @xmath163 , we see that the nonzero rational function @xmath159 has at least @xmath173 distinct zeros on the unit circle , which is impossible because of .
* proof of theorem [ thm_support ] : * everywhere in this proof we will denote @xmath171 . when we say that a vector or a function is unique " we will mean unique , up to multiplication by a nonzero constant " . let us first prove part ( i ) for even vectors .
assume that @xmath43 for some @xmath174 , so that @xmath175 .
it is clear that there exists a unique even vector @xmath176 of length one ( its only nonzero element is @xmath177 ) .
the corresponding dft vector @xmath65 has elements @xmath178 and has length @xmath16 .
having dealt with this special case , we now assume that @xmath179 for some @xmath180 .
if we impose the constraint @xmath181 , then @xmath182 and we must have @xmath183 for @xmath184 .
this gives us @xmath185 zeros of @xmath159 , and since @xmath186 is a polynomial of degree @xmath185 we can uniquely identify @xmath187 which in turn uniquely identifies the vector @xmath188 .
thus we have shown that for every @xmath180 there exists a unique even vector @xmath188 of length @xmath189 satisfying @xmath181 .
since there is also exactly one even vector of length one and exactly one even vector of length @xmath16 satisfying the same property ( these were discussed above ) , we conclude that the total number of such vectors is @xmath75 .
consider next an odd vector @xmath38 of length @xmath189 .
note that there are no odd vectors of length one , thus we can assume that @xmath190 . since @xmath188 is an odd vector , the function @xmath159 can be written in the form @xmath191 in particular , we see that @xmath192 . assuming that @xmath193 we obtain @xmath194 . since @xmath65 is an odd vector of length @xmath195 we must have @xmath183 for @xmath196 and @xmath197 .
this gives us @xmath198 distinct zeros of @xmath159 .
taking into account the zero at @xmath199 and using the fact that @xmath200 is a polynomial of degree @xmath185 , we can uniquely identify @xmath201 thus we have proved that for every @xmath202 there exists a unique odd vector @xmath188 of length @xmath189 satisfying @xmath203 . from our proof
it should be clear that there can not exist an odd vector satisfying @xmath204 , because in this case the function @xmath159 would have @xmath205 zeros , which is impossible due to .
this completes the proof of part ( i ) .
next , let us prove part ( ii ) for even vectors .
assume that @xmath47 and that @xmath38 is a nonzero even vector .
in this case @xmath206 and assuming that @xmath51 we see that this function satisfies @xmath207 .
this identity implies @xmath208 there is a unique even vector of length one satisfying @xmath51 ( its only nonzero elements is @xmath177 ) , however for this vector we have @xmath209 , thus we have to exclude this vector from consideration and we can assume that @xmath179 with @xmath180 . if @xmath210 then @xmath211 . since @xmath65 is also an even vector with @xmath212 , the equality @xmath211 implies that @xmath183 for @xmath213 and for @xmath214 . note that @xmath215 , thus @xmath159 has a zero at @xmath199 , and from condition we see that this zero must have multiplicity two .
thus we have determined all @xmath185 zeros of the polynomial @xmath216 of degree @xmath185 , which allows us to identify uniquely @xmath217 to summarize : we have proved that for every @xmath180 there exists a unique even vector @xmath188 of length @xmath189 satisfying @xmath218 and @xmath219 . from our proof
it is clear that there can not exist an even vector with the same properties satisfying @xmath181 , because in this case the function @xmath159 would have @xmath189 zeros , which is impossible due to .
this completes the proof of part ( ii ) for even vectors .
the proof of theorem [ theorem_4n_plus1_n1 ] ( and the derivation of the corresponding formulas in the appendix a ) is based on the following lemma .
we emphasize that this result is true for _ all _ @xmath220 ( not only those satisfying @xmath221 ) .
let us define the @xmath28-pochhammer symbol as @xmath224 and @xmath225 .
we begin with the @xmath28-binomial theorem @xmath226 when @xmath227 , the above equality gives us @xmath228 where we have also used the identity @xmath229 .
next , we set @xmath230 in , we check that this choice of @xmath28 implies @xmath231 and we obtain @xmath232 e^{2\pi \i km / n}.\ ] ] the expression in the right - hand side is the inverse fourier transform of the sequence given in square brackets . applying the discrete fourier transform results in @xmath233 formula
follows by simplifying the above expression and applying the identity .
* proof of theorem [ theorem_4n_plus1_n1 ] : * in order to prove formula , we set @xmath17 and replace variables @xmath144 , @xmath173 and @xmath142 in the equation with @xmath234 , @xmath235 and @xmath236 .
this gives us the following identity @xmath237 which is valid for @xmath238 , @xmath239 and @xmath240 .
extending the above identity by periodicity for the values of @xmath241 in the range @xmath242 gives us the desired result .
the fact that these vectors are the same ones that were discussed in theorem [ thm_support](i ) follows from and the fact that @xmath243 . in order to prove that these vectors are linearly independent ,
let us assume to the contrary that @xmath244 where @xmath245 denotes the zero vector , @xmath246 for all @xmath247 and @xmath248 .
note that , by construction , @xmath249 for @xmath250 and @xmath251 for @xmath252 .
therefore we have @xmath253 while @xmath254 for all other indices @xmath255 .
equation then forces @xmath256 to be zero and we arrive at a contradiction . to prove formula , we use and and check that @xmath257 subtracting the above two equations and simplifying the result gives us formula . applying the discrete fourier transform to both sides of we
obtain @xmath258 in deriving the above identity we have used the fact that @xmath259 for @xmath260 .
formula follows from and .
the linear independence of the vectors @xmath95 can be established by the same argument as we used for @xmath94 .
first of all , we use property and check that @xmath266 let us denote @xmath267 for @xmath268 . applying the euler - maclaurin summation formula we obtain @xmath269 where the remainder term has an upper bound @xmath270 we compute the first term in the right - hand side in @xmath271 where the clausen function is defined by @xmath272 the function @xmath273 is odd and periodic with period @xmath274 ( see ( * ? ? ?
* chapter 4 ) ) , which gives us @xmath275 the clausen function can be written as taylor series @xmath276 , where @xmath277 , @xmath278 , @xmath279 and the series converges if @xmath280 . using this result
we obtain @xmath281 assuming that @xmath282 we have @xmath283 when @xmath284 , each term in the series in the right - hand side of the above formula can be estimated as follows @xmath285 for all @xmath286 . combining all the above two formulas and using the fact that the series @xmath287 converges we arrive at our final estimate of @xmath288 , which is valid for all @xmath284 and all @xmath282 : @xmath289 dealing with the remaining terms in the right - hand side of is much easier .
we check that the second term satisfies @xmath290 similarly , the third term can be estimated as @xmath291 finally , the remainder can be bounded as follows @xmath292 combining with and the above four asymptotic results ends the proof of .
our first goal is to prove the following upper bound : there exists an absolute constant @xmath296 such that @xmath297 in order to establish we note that @xmath298 which shows that @xmath299 for @xmath300 . by symmetry
we also have @xmath301 for @xmath302 .
lemma [ lemma_v0 ] implies that when @xmath303 is close to @xmath304 then @xmath305 .
since @xmath306 decreases as @xmath307 increases , this bound holds true for all @xmath308 , and we have proved for @xmath309 . for other values of @xmath142
we use the same method coupled with the identity @xmath310 which follows at once from .
next , we introduce @xmath311 . note that as @xmath312 we have @xmath313 . using the double - angle formula for the cosine function we rewrite in the form @xmath314 we emphasize that the implied constant in the error term @xmath315 does not depend on @xmath144 .
applying we obtain @xmath316 the identity @xmath317 coupled with give us @xmath318 using formulas and and the fact that @xmath319 we arrive at the following result @xmath320 again , in all of the above formulas the implied constants in the error terms @xmath321 do not depend on @xmath144 .
now we restrict @xmath144 to the interval @xmath322 .
for @xmath323 we have @xmath324 , thus @xmath325 .
we also check that @xmath326 and @xmath327 which follows from and lemma [ lemma_v0 ] . combining all of the above results with the following formulas for hermite polynomials
@xmath328 we conclude that for @xmath329 @xmath330 let us consider , for example , vector @xmath331 .
we use the above result and and the estimate to conclude @xmath332 thus , as @xmath15 we have @xmath333 recall that @xmath334 . then combined with the last formula in show that for all @xmath323 we have @xmath335 in the range @xmath336 both the left hand side and the right - hand side are bounded by @xmath337 ( see ) , thus as @xmath15 both of these expressions converge to zero , uniformly in @xmath336 .
this concludes the proof of the convergence result for @xmath338 . the proof in the cases @xmath339 follows exactly the same steps .
* fact : * _ assume that the orthonormal vectors @xmath341 and @xmath342 are obtained from two linearly independent vectors @xmath188 and @xmath65 by gramm - schmidt orthogonalization @xmath343 then the vectors @xmath341 and @xmath342 would be the same if instead of @xmath188 and @xmath65 we started with @xmath344 and @xmath345 for some @xmath346 and @xmath347 . _ now , armed with this result , let us prove the convergence result for @xmath348 . according to the definition [ def_phi ] , the vectors @xmath349 and
@xmath350 are obtained by the gramm - schmidt orthogonalization starting from vectors @xmath351 and @xmath352 . according to the above fact
, we will obtain the same result if we start with vectors @xmath353 it is clear from that @xmath354 for @xmath323 . using and we compute @xmath355 combining this result with , and the following expression for the hermite polynomial @xmath356 we find @xmath357 where the error term satisfies @xmath358 . for @xmath329 we can estimate @xmath359 .
thus we have proved that for all @xmath329 @xmath360 it is clear from and that for @xmath308 we have an upper bound @xmath361 , thus @xmath362 converges to zero as @xmath15 uniformly in @xmath308 .
now , the vector @xmath350 is given by @xmath363 , where we set @xmath364 applying , we conclude ( in the same way as we did above for @xmath365 ) that as @xmath15 @xmath366 thus the quantity @xmath367 converges to zero as @xmath368 , which shows that @xmath369 as @xmath15 ( uniformly in @xmath66 ) .
this result , combined with proves that the convergence result holds true for @xmath348 .
the proof of for @xmath370 follows the same steps , though the computations become more tedious ( the use of a symbolic computation package is highly recommended for checking these expressions ) .
for example , when @xmath371 we use and to find @xmath372 we define @xmath373 and check that @xmath374 where @xmath375 .
starting with this result and following exactly the same argument as we have used in the case @xmath348 one can show that the convergence result holds true for @xmath371 .
the details and the proof of the remaining cases @xmath376 are left to the reader . in this section
we present the analogues of theorem [ theorem_4n_plus1_n1 ] and corollary [ prop_4n_plus1_n2 ] for the remaining cases @xmath377 . in each case
we give formulas for the even and odd vectors @xmath94 and @xmath95 ( which were described in theorem [ thm_support ] ) and compute their corresponding discrete fourier transforms . the proof of these results
is based on lemma [ main_lemma ] and it is very similar to the proof of theorem [ theorem_4n_plus1_n1 ] , we omit all the details . we construct a basis for each eigenspace @xmath106 , @xmath107 , @xmath108 and @xmath109 , from which one can build the hermite - type eigenvectors @xmath136 in exactly the same way as in definition [ def_phi ] .
we define @xmath379 vectors @xmath86 and @xmath380 vectors @xmath381 as follows : for @xmath382 we set @xmath383 and @xmath384 these @xmath16 vectors are linearly independent and they satisfy @xmath385 and @xmath386 .
the vectors @xmath387 and @xmath381 are precisely the vectors described in theorem [ thm_support](ii ) .
an equivalent expression for the vectors @xmath94 is given by @xmath388 define @xmath16 vectors @xmath113 , @xmath114 , @xmath115 , @xmath389 as follows : @xmath390 the vectors @xmath113 ( respectively , @xmath114 , @xmath115 and @xmath389 ) form a basis for the eigenspace @xmath106 ( respectively , @xmath107 , @xmath108 and @xmath109 ) .
we define @xmath392 vectors @xmath393 and @xmath394 vectors @xmath395 as follows : for @xmath396 we set @xmath397 and @xmath398 these @xmath16 vectors are linearly independent and they satisfy @xmath399 and @xmath400 . the vectors @xmath401 and @xmath395 are precisely the vectors described in theorem [ thm_support](ii ) .
an equivalent expression for the vectors @xmath94 is given by @xmath402 define @xmath16 vectors @xmath113 , @xmath114 , @xmath403 , @xmath116 as follows : @xmath404 the vectors @xmath113 ( respectively , @xmath114 , @xmath403 and @xmath116 ) form a basis for the eigenspace @xmath106 ( respectively , @xmath107 , @xmath108 and @xmath109 ) .
we define @xmath392 vectors @xmath393 and @xmath379 vectors @xmath406 as follows : for @xmath407 we set @xmath408 these @xmath16 vectors are linearly independent and they satisfy @xmath409 the vectors @xmath410 and @xmath411 are precisely the vectors described in theorem [ thm_support](i ) .
an equivalent expression for the vectors @xmath94 is given by @xmath412 define @xmath16 vectors @xmath113 , @xmath413 , @xmath403 , @xmath116 as follows : @xmath414 the vectors @xmath113 ( respectively , @xmath413 , @xmath403 and @xmath116 ) form a basis for the eigenspace @xmath106 ( respectively , @xmath107 , @xmath108 and @xmath109 ) . | the search for a canonical set of eigenvectors of the discrete fourier transform has been ongoing for more than three decades .
the goal is to find an orthogonal basis of eigenvectors which would approximate hermite functions
the eigenfunctions of the continuous fourier transform .
this eigenbasis should also have some degree of analytical tractability and should allow for efficient numerical computations . in this paper we provide a partial solution to these problems .
first , we construct an explicit basis of ( non - orthogonal ) eigenvectors of the discrete fourier transform , thus extending the results of @xcite . applying the gramm - schmidt orthogonalization procedure we obtain an orthogonal eigenbasis of the discrete fourier transform .
we prove that the first eight eigenvectors converge to the corresponding hermite functions , and we conjecture that this convergence result remains true for all eigenvectors .
0.15 cm _ keywords _ : eigenvectors , discrete fourier transform , orthogonal basis , hermite functions , q - binomial theorem + _ 2010 mathematics subject classification _ : primary 42a38 , secondary 65t50 |
consider the following situation .
alice writes a message to bob consisting of the numbers of several bank accounts to which bob has to send some money .
she writes in a hurry ( she just got to know that the transfers are urgent if they do not want to pay delay punishment , but currently she has little time ) .
therefore her characters are not very well legible , so bob may misread some numbers .
however , there are some rules for the possible mistakes , e.g. , a @xmath0 may be thought to be a @xmath1 but never a @xmath2 .
this relation between the possible digits need not be symmetric : it is possible that a @xmath3 is sometimes read as a @xmath2 but a @xmath2 may not be decoded as a @xmath3
. these rules of possible confusions are known both by alice and by bob . as
alice is aware of the possibility that bob misread her message , later in the day she sends another message to bob , the goal of which is to make bob certain whether he read ( decoded ) the first message correctly or not . if he did he can transfer the money with complete confidence that he sends it to the right accounts .
if he did not he will know that he does not know the account numbers correctly and so he better wait and pay the punishment than transfer the money to the wrong place .
the second message will be received by bob correctly for certain , but it uses an expensive device , e.g. , alice sends it as an sms from another country after she has arrived there .
( now we understand why she was in a hurry : she had to arrive to the airport in time . ) for some reason , every character sent from this foreign country costs a significant amount of money for her .
so she wants to send the shortest possible message that makes it sure ( here we insist on zero - error ) that bob will know whether his decoding of the original handwritten message was error - free or not .
the problem is to determine the best rate of communication over the second channel as the length of the original message received tends to infinity . in section [ sec : dil ]
we describe the abstract communication model for this scenario and show that the best achievable rate is a parameter of an appropriate directed graph .
we will see that this parameter of a directed graph is a generalization of the parameter ( of an undirected graph ) called witsenhausen rate .
( this means that we also obtain a new interpretation of witsenhausen rate . ) in section [ sec : bounds ] we investigate the relationship with other graph parameters .
these include sperner capacity and the dichromatic number .
the former is a generalization of shannon s graph capacity @xcite to directed graphs .
though originally defined to give a general framework for some problems in extremal set theory ( see @xcite ) , sperner capacity also has its own information theoretic relevance , see @xcite .
the dichromatic number is a generalization of the chromatic number to directed graphs introduced in @xcite . using the above mentioned relations we determine our new parameter for some specific directed graphs .
in section [ sec : comp ] we consider a compound channel type version of the problem parallel to @xcite . in section [ sec : extr ]
some connections to extremal set theory are pointed out . in section [ sec :
ambit ] we will consider the setup where the requirement is more ambitious and we want that alice s second ( the error - free but expensive ) message make bob able to decode the original message with zero - error . (
that is , he will know the message itself not only the correctness or incorrectness of his original decoding of alice s handwriting . )
we will see that this setting leads to the witsenhausen rate of an undirected graph related to the problem .
this gives a further new interpretation of witsenhausen rate .
all logarithms are meant to be of base @xmath4 .
the abstract setting for our communication scenario is the following .
we have a source whose output is sent through a noisy channel .
( this belongs to alice s handwriting . )
the input and output alphabets of this channel are identical and they coincide with the output alphabet of the source . it is known how the noisy channel can deform the input , in particular we know what ( input ) letters can become a certain , possibly different ( output ) letter on the other side .
( we always assume though that every letter can result in itself , that is get through the noisy channel without alteration . )
later another message is sent ( by the same sender ) to the same receiver .
this second message is sent via a noiseless channel and its goal is to make zero - undetected - error decoding possible , i.e. , after having received this second message the receiver should be able to decide whether it decoded the first message correctly .
the use of the noiseless channel is expensive , so the second message should be kept as short as possible .
let the shortest possible message that satisfies the criteria have length @xmath5 when @xmath6 characters of the source output are encoded together .
let @xmath7 denote the noisy channel .
the efficiency of the communication is measured by the quantity @xmath8 that we call the dilworth rate of the noisy channel @xmath7 .
( for an explanation of the name see remark [ rem : name ] in subsection [ subsect : dilwdef ] . ) [ rem : cvsr ] note the special feature of the problem that we characterize a channel by a rate , that is with a parameter that , unlike channel capacity , we want to be as small as possible .
the reason is that we measure the reliability of a channel not by the amount of information it can safely transfer but with the amount of information needed _ to be added _ for making the communication reliable .
@xmath9 the relevant properties of @xmath7 are described by a directed graph @xmath10 having the ( common input and output ) alphabet as its vertex set and the following edge set .
an ordered pair @xmath11 of two letters forms a directed edge of @xmath10 if and only if @xmath12 and the output of @xmath7 can be @xmath13 when it is fed by @xmath14 at the input .
[ rem : ve ] as usual we will use @xmath15 to denote the vertex set and @xmath16 to denote the edge set of a directed graph @xmath17 .
we will use similar notation for undirected graphs that we always consider to be the same as a _ symmetrically directed graph_. in such a graph an ordered pair @xmath18 of two vertices forms a directed edge if and only if the reversely ordered pair @xmath19 is also present in the digraph as a directed edge .
we will use the term _ oriented graph _ for directed graphs that do not contain any edge together with its reversed version .
that is @xmath17 is an oriented graph if @xmath20 implies @xmath21 .
as it is also customary , the term _ digraph _ will be used as a synonym for `` directed graph '' .
@xmath9 to express @xmath22 as a graph parameter we need the following notion . the and product @xmath23 of two directed graphs @xmath17 and @xmath24 is defined as follows .
the vertex set of @xmath25 is the direct product @xmath26 and vertex @xmath27 sends a directed edge to @xmath28 iff either @xmath29 and @xmath30 or @xmath29 and @xmath31 or @xmath32 and @xmath30 .
the @xmath6-th and power of a digraph @xmath24 , denoted by @xmath33 is the @xmath6-wise and product of digraph @xmath24 with itself .
observe that this graph exponentiation extends to sequences of letters the relation between individual letters @xmath34 and @xmath35 expressing that feeding @xmath34 to the noisy channel @xmath7 may result in observing letter @xmath35 at the output .
a sequence of letters at the input of @xmath7 can result in another such sequence at the output of @xmath7 if at each coordinate the character in the first sequence can result in the corresponding character of the second sequence .
( this includes the possibility that the character does not change when sent through @xmath7 . )
[ rem : ao ] the terminology of graph products is not completely standardized .
the and product we just defined is also called normal product @xcite , strong direct product @xcite , or strong product @xcite .
we follow the paper of alon and orlitsky @xcite when use the name and product , because we find this name informative .
a similar remark applies to the or product that we will introduce later in definition [ defi : or ] .
@xmath9 recall that the chromatic number @xmath36 of a graph @xmath37 is the minimal number of colors that suffice to color the vertices of @xmath37 so that adjacent vertices get different colors .
if @xmath17 is a digraph , its chromatic number @xmath38 is understood to be the chromatic number of the underlying undirected graph .
[ prop : chi ] @xmath39 it is easy to see that the above limit always exists .
( the reason is the submultiplicative behaviour of the chromatic number under the and product ) .
@xmath9 * proof .
* alice and bob can agree in advance in a proper coloring of @xmath40 with @xmath41 colors .
alice can send bob the color of the vertex belonging to the original source output using @xmath42 bits .
bob compares this to the color of the vertex representing the sequence he obtained as a result of his decoding .
if the latter color is identical to the one alice has sent him , then he can be sure that his decoding was error - free .
this is because any other sequence that could result in his decoded sequence is adjacent ( in @xmath43 ) to this decoded sequence , so its color is different . on the other hand , if alice sent a shorter message through the noiseless channel , then she could not have @xmath41 distinct messages and thus there must exist two adjacent vertices in @xmath43 that are encoded to the same codeword by alice ( for the noiseless channel ) .
then one of the two sequences represented by these two adjacent vertices could result in the other one , while this other one could also result in itself .
thus bob can not make the difference between these two sequences , one of which is the correct source output sequence while the other one differs from it .
so receiveing this message bob could not be sure whether his decoding was error - free or not .
@xmath44 the right hand side expression in proposition [ prop : chi ] can be considered as a digraph parameter that we will call the dilworth rate of the digraph @xmath10 .
[ defi : dilrate ] for a directed graph @xmath24 we define its ( logarithmic ) _ dilworth rate _ to be @xmath45 the non - logarithmic dilworth rate is @xmath46{\chi(\vec{g}^{\wedge t}}).\ ] ] obviously , @xmath47 [ rem : name ] let @xmath48 be the directed graph on @xmath4 vertices with a single directed edge .
if we consider the vertices of @xmath49 as characteristic vectors of subsets of a @xmath6-element set then @xmath50 can be interpreted as the asymptotic exponent of the minimum number of antichains ( sets of pairwise incomparable elements ) in the boolean lattice of these subsets that can cover all the subsets .
the exact value of this minimum number is given ( easily ) by a special case of what is called the `` dual of dilworth s theorem '' @xcite ( also called mirsky s theorem , see @xcite ) .
this connection to dilworth s celebrated result is the reason for calling our new parameter dilworth rate .
note that the name sperner capacity was picked by the authors of @xcite for analoguous reasons : the sperner capacity of the digraph @xmath48 has a similar relationship with sperner s theorem @xcite .
@xmath9 the and product is also defined for undirected graphs .
considering undirected graphs as symmetrically directed graphs the definition is straightforward .
witsenhausen considered the `` zero - error side - information problem '' that led him to introduce the quantity @xmath51 that is called the witsenhausen rate of ( the undirected ) graph @xmath52 .
it is straightforward from the definitions that if @xmath24 is a symmetrically directed graph and @xmath52 is the underlying undirected graph ( that we consider equivalent ) , then @xmath53 .
thus dilworth rate is indeed a generalization of witsenhausen rate to directed graphs .
[ rem : nrfam ] we note that nayak and rose @xcite defines what they call `` the witsenhausen rate of a set of directed graphs '' .
though formally this gives the dilworth rate of a directed graph , the focus of @xcite is elsewhere .
when its motivating setup results in a family consisting of a single digraph , then this digraph is symmetrically directed .
( see also theorem [ thm : wm ] in section [ sec : comp ] . )
sperner capacity was introduced by gargano , krner and vaccaro @xcite .
traditionally this parameter is defined by using the or product .
[ defi : or ] the or product @xmath54 of directed graphs @xmath17 and @xmath24 has vertex set @xmath26 and @xmath27 sends a directed edge to @xmath28 iff either @xmath55 or @xmath56 .
the @xmath6-th or power @xmath57 is the @xmath6-wise or product of digraph @xmath24 with itself .
let @xmath58 denote the complete directed graph on @xmath59 vertices , that is the one we obtain from a(n undirected ) complete graph @xmath60 when substituting each of its edges @xmath61 by the two oriented edges @xmath11 and @xmath62 .
the ( directed ) complement of a digraph @xmath24 is the directed graph @xmath63 on vertex set @xmath64 having edge set @xmath65 now we note the straightforward relation of the and and or powers that @xmath66
. the ( logarithmic ) sperner capacity of digraph @xmath24 is defined ( see @xcite ) as @xmath67 where @xmath68 denotes the _ symmetric clique number _ , that is the cardinality of the largest symmetric clique in digraph @xmath17 : the size of the largest set @xmath69 where for each @xmath70 both @xmath71 and @xmath72 are edges of @xmath17 . using the above relation of the and and or products , sperner capacity ( of the complementary graph @xmath73 )
can also be defined as @xmath74 where @xmath75 stands for the independence number ( size of the largest edgeless subset of the vertex set ) of graph @xmath17 .
this is the definition given in @xcite .
( the authors of @xcite call this value the sperner capacity of @xmath24 . ) when @xmath52 is an undirected ( or symmetrically directed ) graph , then @xmath76 , the _ shannon capacity _ of graph @xmath52 ( see @xcite ) .
we will need a sort of probabilistic refinement of our capacity - like parameters called their `` within - a - type '' versions , see @xcite .
first we need the concept of @xmath77-typical sequences , cf .
@xcite .
[ defi : peps ] let @xmath78 be a finite set , @xmath79 a probability distribution on @xmath78 , and @xmath80 .
a sequence @xmath81 in @xmath82 is said to be @xmath77-typical if for every @xmath83 we have @xmath84 , where @xmath85 .
we denote the set of @xmath77-typical sequences in @xmath82 by @xmath86 . when @xmath87 we also write @xmath88 for @xmath89 .
if @xmath90 we say that the _ type _ of @xmath81 is @xmath79 . let @xmath91 stand for either @xmath33 or @xmath57 .
for a directed ( or undirected ) graph @xmath17 and @xmath69 we denote by @xmath92 $ ] the digraph induced by @xmath17 on the subset @xmath93 of the vertex set .
we also use the shorthand notation @xmath94 $ ] .
let @xmath95 be either of the following graph parameters of the directed graph @xmath24 : independence number , clique number , chromatic number , clique cover number ( which is the chromatic number of the complementary graph ) , symmetric clique number , or transitive clique number .
( the latter is the size of the largest subset @xmath93 of @xmath64 the elements of which can be linearly ordered so that if @xmath96 precedes @xmath97 then the oriented edge @xmath18 is present in @xmath98 . )
let the asymptotic parameter @xmath99 be defined as @xmath100 while @xmath101 stands for @xmath102 .
[ defi : captyp ] the parameter @xmath103 of a digraph @xmath24 within a given type @xmath79 is the value @xmath104 we note that for several of the allowed choices of @xmath95 and @xmath91 we obtain a graph parameter that already exists in the literature . for example , when @xmath105 and the power we look at is the or power , we get sperner capacity within a given type , that has an important role in the main results of the papers @xcite .
if we choose @xmath106 and the or power , we obtain the functional called graph entropy , which is defined in @xcite and has several nice properties , see @xcite , as well as important applications , see e.g. @xcite .
when @xmath106 but the exponentiation is the and power , then we arrive to the within a type version of dilworth rate @xmath107 .
the special case of this for an undirected graph @xmath52 was already known under name `` complementary graph entropy '' that could justifiably be called `` witsenhausen rate within a given type '' .
this parameter was introduced by krner and longo @xcite and further investigated by marton @xcite .
although this within - a - type version of witsenhausen s invariant was introduced earlier than the non - probabilistic version ( cf .
@xcite ) , for the sake of consistancy we denote it by @xmath108 . note that marton @xcite proved the important identity @xmath109 where @xmath110 is the entropy of the probability distribution @xmath111 .
this holds for any probability distribution @xmath79 on @xmath112 .
along the same lines one can also prove the following theorem .
we give its proof for the sake of completeness .
we will use the notion of fractional chromatic number @xmath113 in the proof .
let @xmath114 denote the set of independent sets in @xmath52 .
a function @xmath115 is a fractional coloring of @xmath52 if for every vertex @xmath116 we have @xmath117 , that is the sum of the weights @xmath118 puts on independent sets containing @xmath97 is at least @xmath1 .
( a proper colorig is also a fractional coloring : the color classes get weight @xmath1 , the other independent sets get weight @xmath3 . )
the fractional chromatic number is @xmath119 , that is the minimum ( taken over all fractional colorings ) of the total weight put on independent sets by a fractional coloring @xmath118 .
( formally we should write infimum but it is known that the minimum is always attained . see the book @xcite for a detailed account on fractional graph parameters . )
we will need the following properties of the fractional chromatic number .
[ defi : vtrans ] a directed graph @xmath24 is vertex - transitive if for any two vertices @xmath120 it admits an automorphism that maps @xmath96 to @xmath97 . if @xmath37 is a vertex - transitive graph , then @xmath121 ( for a proof see @xcite , proposition 3.1.1 on page 41 . ) for every graph @xmath37 we have @xmath122{\chi(f^{\odot t}})=\lim_{t\to\infty}\sqrt[t]{\chi_f(f^{\odot t}}).\ ] ] the latter follows from lovsz s result @xcite stating that @xmath123 and the obvious inequality @xmath124 that holds for all ( finite simple ) graphs .
[ thm : sumhp ] let @xmath24 be a directed graph and @xmath79 an arbitrary fixed probability distribution on @xmath64 . then @xmath125 * proof .
* note that by the well - known ( and more or less trivial ) inequality @xmath126 for every graph @xmath37 , we have @xmath127 .
clearly , this relation also holds if we have a directed graph @xmath17 in place of the undirected graph @xmath37 .
this is straightforward since @xmath38 and @xmath75 are defined to be identical to the corresponding parameter of the underlying undirected graph @xmath37 .
it is also well - known ( cf .
e.g. @xcite ) that @xmath128 .
the last two relations immediately give @xmath129 for the reverse inequality let us fix a sequence of probability distributions @xmath130 on the vertex set of our graph so that @xmath131 and @xmath132 notice that @xmath133 is a vertex - transitive graph , since every sequence forming an element of @xmath134 can be transformed into any other such sequence by simply permuting the coordinates .
thus @xmath135 @xmath44 using standard techniques of the method of types , cf . @xcite we can already state our lower bound on @xmath136 .
we need the fact that fixing the length @xmath6 the number of distinct types of a sequence over some fixed alphabet is only a polynomial funcion of @xmath6 ( cf . the type counting lemma 2.2 in @xcite ) , while the parameters we investigate are asymptotic exponents of some graph parameters that grow exponentially as @xmath6 tends to infinity . with this in mind
we can write @xmath137 [ thm : lowerb ] @xmath138 * proof .
* using the above equalities , we obtain @xmath139 this gives the lower bound in the statement .
@xmath44 note that sperner capacity is unkown for many graphs , so the lower bound above usually does not give a known numerical value . still , there are some examples of graphs where sperner capacity is known and is non - trivial .
a basic example is the cyclically oriented triangle , or more generally , any cyclically oriented cycle .
first we formulate a consequence of the above formula .
[ cor : vtrans ] if @xmath24 is a vertex - transitive digraph then @xmath140 * proof .
* let @xmath141 denote the uniform distribution on the vertex set of @xmath24 .
if @xmath24 is vertex - transitive then by symmetry @xmath142 and @xmath143 .
combining these equalities with theorem [ thm : lowerb ] we obtain @xmath144 and thus the statement .
@xmath44 now we will use this corollary to determine the dilworth rate of the cyclically oriented @xmath145-length cycle @xmath146 for every @xmath145 .
note that the complement of a cyclically oriented cycle is a cyclically oriented cycle of the same length together with all diagonals as bidirected ( or equivalently , undirected ) edges . for
@xmath147 there are no diagonals , so the cyclic triangle is isomorphic to its complement .
the sperner capacity of the cyclic triangle was determined in @xcite , cf .
also @xcite , and its value is @xmath148 .
this result was generalized by alon @xcite , who proved that the sperner capacity of a digraph @xmath24 is bounded from above by @xmath149 where @xmath150 and @xmath151 stand for the maximum outdegree and maximum indegree of @xmath17 , respectively .
the indegree and outdegree of a vertex @xmath97 is the number of edges at @xmath97 that are oriented towards or outwards @xmath97 , respectively .
@xcite for a further generalization of alon s result . ) on the other hand , sperner capacity is bounded from below by ( the logarithm of ) the transitive clique number , the number of vertices in a largest transitively directed complete subgraph , denoted by @xmath152 .
( this is an easy observation which implies that substituting @xmath153 by @xmath154 in the definition of sperner capacity gives the same value , i.e. it gives an alternative definition of sperner capacity , see @xcite and also proposition 4 and the remark following it in @xcite . )
note that a transitively directed complete subgraph meant here is not necessarily induced .
it is allowed that some reverse edges are also present on the same subset of vertices .
[ cor : ck ] the dilworth rate of the cyclically oriented @xmath145-cycle is @xmath155 * proof .
* let the directed complement of @xmath146 be denoted by @xmath156 . since @xmath157
, alon s above mentioned result implies that the sperner capacity of @xmath156 is at most @xmath158 .
it is easy to see that @xmath159 , so the lower bound mentioned above is also @xmath158 .
since the above two bounds coincide , the sperner capacity of @xmath160 is equal to @xmath161 . using that @xmath162 is vertex transitive corollary [ cor : vtrans ] implies the statement . @xmath44
note that corollary [ cor : ck ] shows that the dilworth rate is a true generalization of witsenhausen rate since @xmath163 if @xmath164 .
[ defi : acyc ] call a subset of the vertex set of a directed graph @xmath24 acyclic if it induces an acyclic subgraph .
the latter means that there is no oriented cycle on these vertices .
the acyclicity number @xmath165 of a directed graph @xmath24 is the number of vertices in a largest acyclic subset of @xmath64 . note that unlike for a transitive clique we do not allow reverse edges in an acyclic subgraph .
let @xmath166 be an odd number .
the following tournaments ( oriented complete graphs ) are also generalizations of the cyclic triangle .
let @xmath167 and @xmath168 is an edge iff @xmath169 for some @xmath170 .
( figure 1 shows the tournament @xmath171 . ) note that it holds for every directed graph that reversing all of its edges does not change the value of either its sperner capacity or of its dilworth rate .
this implies that if @xmath172 is a tournament then we have @xmath173 and @xmath174 . by @xmath175
all the four values are equal . .
] [ cor : tourn ] for all odd integers @xmath176 we have @xmath177 * proof .
* we know that @xmath178 holds for all directed graphs ( cf .
@xcite and also the discussion before corollary [ cor : ck ] for an equivalent statement concerning the complementary digraph ) .
this gives @xmath179 by @xmath180 ( see the note before stating the corollary ) alon s result can be applied implying that our lower bound is sharp .
since @xmath181 is vertex - transitive we can apply corollary [ cor : vtrans ] to complete the proof .
@xmath44 observe that corollary [ cor : tourn ] shows not only that the value of the dilworth rate of an oriented graph may differ from the witsenhausen rate of the underlying undirected graph , but also that the difference can be arbitrarily large .
indeed , denoting the complete graph on @xmath182 vertices by @xmath183 we have @xmath184 for every @xmath185 . the left hand side of the inequality is bounded above by @xmath186 , while the right hand side goes to infinity with @xmath182 .
now we show that the ( logarithm of the ) _ dichromatic number _ defined in @xcite is an upper bound on the dilworth rate .
[ def : dic ] the _ dichromatic number _
@xmath187 of a directed graph @xmath24 is the minimum number of acyclic subsets that cover @xmath64 . a partition of @xmath64 into acyclic subsets will be called a _ directed coloring _ or _
dicoloring_. we note that an undirected edge ( meaning a bidirected edge ) is considered to be a @xmath4-length cycle , therefore its two endpoints can not be both contained in an acyclic set .
this shows that for undirected ( equivalently , symmetrically directed ) graphs the dichromatic number is equal to the chromatic number .
we do not use the term `` acyclic coloring '' , because it is already used for a completely different concept , see @xcite .
@xmath9 [ thm : upb ] for any directed graph @xmath24 @xmath188 * proof .
* let us fix a directed coloring of digraph @xmath24 consisting of @xmath189 acyclic subsets ( `` color classes '' ) .
for each @xmath190 let @xmath191 denote the color class that contains @xmath97 . now consider @xmath33 .
it has @xmath192^t$ ] vertices . for each sequence
@xmath193 we attach the sequence of colors @xmath194 .
there are @xmath195 such color sequences , so this gives a partition of @xmath196 into @xmath195 partition classes .
we also give another partition of @xmath196 according to types .
two vertices are in the same partition class if their type is the same . by the type counting lemma 2.2 in @xcite , we know that this latter partition has at most @xmath197 , that is a polynomial number ( in @xmath6 ) of classes .
now let @xmath198 be the common refinement of these two partitions .
we have @xmath199 by the foregoing .
now we show that each partition class @xmath200 induces an independent set in @xmath33 .
let two sequences @xmath201 and @xmath202 belong to the same @xmath200 , that is , they have the same type and @xmath203 .
let @xmath204 be an index for which @xmath205 .
since @xmath206 and @xmath207 are in the same color class of a valid dicoloring , we can not have both @xmath208 and @xmath209 present in the graph as a directed edge .
if neither is present then there is no edge between @xmath210 and @xmath211 . if @xmath212 , then we know that @xmath213 , so the oriented edge @xmath214 can not be an edge of @xmath33 .
we need to show that neither the opposite oriented edge @xmath215 can be present in @xmath33 .
this is because if @xmath212 , then there should be another position @xmath216 for which @xmath217 . if this is true , then @xmath218 and so @xmath219 .
the existence of @xmath216 with the above property holds for the following reason .
consider all coordinates @xmath220 , for which @xmath221 .
denote the set of these @xmath220 s @xmath222 . since vertices @xmath223 with the same `` color '' @xmath191 induce an acyclic subdigraph
, we can put these vertices into a linear order so , that @xmath224 implies that @xmath97 precedes @xmath225 in this linear order .
so if @xmath212 , then for our edge @xmath208 @xmath206 precedes @xmath207 .
however , since @xmath210 and @xmath211 has the same type it can not happen that for each @xmath226 @xmath227 precedes @xmath228 in this linear order .
so there must be a coordinate @xmath216 where @xmath229 precedes @xmath230 and this implies the claimed properties .
thus each partition class @xmath200 is independent indeed , so the undirected graph underlying @xmath33 can be properly colored with @xmath231 colors .
this implies that @xmath232{\chi([\vec{g}_h]^{\wedge t})}\le \liminf_{t\to\infty}\sqrt[t]{(t+1)^{|v(\vec{g})|}k^t}=k=\chi_{\rm dir}(\vec{g}).$ ] @xmath44 as a strengthening of the previous theorem we will show that we can also write a natural fractional relaxation of the dichromatic number on the right hand side above .
( we could not prove this right away , as the weaker statement will be used in the proof . ) to prove this stronger statement we need some preparation , in particular we will use the following observations .
first note , that @xmath233 holds for any digraph @xmath17 .
this is simply because independent sets in @xmath17 are special acyclic sets , so any proper coloring of @xmath17 is also a directed coloring of @xmath17 .
[ prop : submult ] the dichromatic number is submultiplicative with respect to the and product , i.e. @xmath234 in particular , @xmath235^t.\ ] ] a straightforward consequence of proposition [ prop : submult ] is that the limit @xmath236{\chi_{\rm dir}(f^{\wedge t})}$ ] exists .
* proof of proposition [ prop : submult ] * let @xmath237 and @xmath238 be optimal directed colorings of the digraphs @xmath17 and @xmath24 , respectively . using these colorings we define the function @xmath239 as follows . for @xmath240 let @xmath241 .
observe that the and product of two acyclic ( sub)graphs results in an acyclic ( sub)graph .
assume for contradiction that @xmath242 and @xmath243 are acyclic subsets of @xmath15 and @xmath64 , respectively , and @xmath244 $ ] contains a directed cycle .
( recall that @xmath245 $ ] denotes the digraph @xmath246 induces on @xmath247 . )
let its vertices be @xmath248 in the ( cyclic ) order the cycle defines , i.e. @xmath249 is an edge of @xmath250 for all @xmath251 where addition is intended modulo @xmath145 .
we may assume without loss of generality that not all @xmath252 s are equal .
then in the sequence @xmath253 we have for all @xmath254 either @xmath255 or @xmath256 ( addition is again modulo @xmath145 ) and for some @xmath257 the second case occurs .
but then there must be a directed cycle in @xmath258 $ ] contradicting the assumption that @xmath242 is acyclic .
the above implies that @xmath259 is a directed coloring of @xmath25 .
as it uses @xmath260 colors the statement is proved .
[ lem : nemno ] for any digraph @xmath17 and positive integer @xmath145 we have @xmath261^k.\ ] ] * proof .
* fix an arbitrary positive integer @xmath145 .
we can write @xmath262{\chi(\vec{f}^{\wedge mk})}\\ = & \lim_{m\to\infty}\sqrt[k]{\sqrt[m]{\chi([\vec{f}^{\wedge k}]^{\wedge m})}}\\ = & \sqrt[k]{\lim_{m\to\infty}\sqrt[m]{\chi([\vec{f}^{\wedge k}]^{\wedge m})}}\\ = & \sqrt[k]{r_{\rm d}(\vec{f}^{\wedge k } ) } , \end{array}\ ] ] that implies the statement
. @xmath44 [ prop : asympt ] for any digraph @xmath17 we have @xmath122{\chi_{\rm dir}(\vec{f}^{\wedge t})}=r_{\rm d}(\vec{f}).\ ] ] * proof . * by @xmath263 we have
@xmath122{\chi_{\rm dir}(\vec{f}^{\wedge t})}\le \lim_{t\to\infty}\sqrt[t]{\chi(\vec{f}^{\wedge t})}=r_{\rm d}(\vec{f}).\ ] ] for the reverse inequality we can write @xmath264{r_{\rm d}(\vec{f}^{\wedge t})}\le\lim_{t\to\infty}\sqrt[t]{\chi_{\rm dir}(\vec{f}^{\wedge t})},\ ] ] where the equality follows by noticing that lemma [ lem : nemno ] above is valid for all positive integers @xmath6 and the inequality is a consequence of @xmath265 applied for @xmath266 .
@xmath44 [ def : fracdic ] let the set of subsets of the vertex set inducing an acyclic subgraph in a digraph @xmath24 be @xmath267 .
a function @xmath268 is called a _
fractional directed coloring _ ( or fractional dicoloring ) if for @xmath269 we have @xmath270 the _ fractional dichromatic number _ of @xmath24 is @xmath271 where the minimum is taken over all fractional directed colorings @xmath118 of @xmath24 .
note the obvious inequality @xmath272 for any digraph @xmath24 .
we will need the following lemma .
[ lem : dicfsub ] for any digraphs @xmath17 and @xmath24 we have @xmath273 * proof .
* let @xmath34 and @xmath118 be optimal fractional directed colorings of @xmath17 and @xmath24 , respectively .
we use the observation , already verified in the proof of proposition [ prop : submult ] , stating that if @xmath274 and @xmath275 then the direct product @xmath276 is in @xmath277 , i.e. @xmath276 induces an acyclic subdigraph in @xmath25 .
now give the following weights @xmath278 to the acyclic sets of @xmath25 .
if @xmath279 has a product structure , i.e. @xmath280 for some @xmath281 and @xmath282 , then let @xmath283 .
if @xmath7 is not of this form , then let @xmath284 .
for any @xmath285 we have @xmath286 thus @xmath278 is a fractional dicoloring of @xmath25 .
now we have @xmath287 this completes the proof .
@xmath44 [ cor : powfsub ] for any digraph @xmath24 and any positive integer @xmath6 we have @xmath288^t.\ ] ] @xmath44
we also need the following result .
[ prop : asyfrac ] for any digraph @xmath17 we have @xmath122{\chi_{{\rm dir},f } ( \vec{f}^{\wedge t})}=r_{\rm d}(\vec{f}).\ ] ] for the proof we need some preparation .
a hypergraph @xmath289 consists of a vertex set @xmath290 and an edge set @xmath291 , where the elements of @xmath291 are subsets of @xmath78 .
a covering of hypergraph @xmath292 is a set of edges the union of which contains all elements of @xmath293 .
let @xmath294 denote the minimum number of edges in a covering of @xmath295 .
a fractional covering of a hypergraph @xmath289 is a function @xmath296 satisfying for every @xmath297 that @xmath298 .
the fractional covering number is @xmath299 where the minimization is over all fractional covers @xmath118 . clearly , @xmath300 .
lovsz proved in @xcite ( cf . also @xcite ) that @xmath301 where @xmath302 , that is the cardinality of a largest edge in @xmath292 . for a directed graph @xmath24 let @xmath303 where @xmath304 , i.e. it consists of the acyclic subsets of vertices in @xmath24 .
it is straightforward that @xmath305 and @xmath306 while @xmath307 .
thus the above result implies that @xmath308 * proof of proposition [ prop : asyfrac ] * we have @xmath236{\chi_{{\rm dir},f}({\vec{f}}^{\wedge t})}\le \lim_{t\to\infty}\sqrt[t]{\chi_{\rm dir}({\vec{f}}^{\wedge t})}=r_{\rm d}({\vec{f}})$ ] by proposition [ prop : asympt ] and the obvious inequality @xmath309 applied to @xmath310 . for the reverse inequality we write @xmath311{\chi_{\rm dir}(\vec{f}^{\wedge t})}\\ \le&\lim_{t\to\infty}\sqrt[t]{\chi_{{\rm dir},f}(\vec{f}^{\wedge t})(1+\log a(\vec{f}^{\wedge t}))}\\ = &
\left(\lim_{t\to\infty}\sqrt[t]{\chi_{{\rm dir},f}(\vec{f}^{\wedge t})}\right)\times\\&\times\left(\lim_{t\to\infty}\sqrt[t ] { ( 1+\log a(\vec{f}^{\wedge t}))}\right)\\ = & \lim_{t\to\infty}\sqrt[t]{\chi_{{\rm dir},f}(\vec{f}^{\wedge t})}. \end{array}\ ] ] @xmath44 [ thm : frupb ] for any directed graph @xmath24 @xmath312 * proof .
* from the above we have @xmath313{\chi_{{\rm dir},f}(\vec{g}^{\wedge t})}\\&\le \lim_{t\to\infty}\sqrt[t]{[\chi_{{\rm dir},f}(\vec{g})]^t } = \chi_{{\rm dir},f}(\vec{g } ) .
\end{array}\ ] ] @xmath44 there are several directed graphs @xmath24 for which the above upper bound is sharp . in particular
, corollaries [ cor : ck ] and [ cor : tourn ] can be proved using theorem [ thm : frupb ] instead of vertex - transitivity .
( an optimal fractional dicoloring of @xmath171 is shown on figure 2 . ) . ]
note however , that theorem [ thm : frupb ] is not always tight : for the graph ( symmetrically directed digraph ) @xmath314 , we have @xmath315 by results in @xcite and @xcite , while @xmath316 we present another such example which is not symmetrically directed .
let the @xmath317-length cycle be oriented in an ( as much as possible ) alternating manner , that is so that only one of its vertices will have outdegree @xmath1 ( implying that two of the @xmath318 others will have outdegree @xmath4 , and the remaining @xmath4 have outdegree @xmath3 ) .
denote this oriented graph by @xmath319 .
( see figure 3 . ) . ]
we know that @xmath320 .
( this is proven as proposition 4 in @xcite , see @xcite for more details on the sperner capacity of oriented self - complementary graphs .
all other orientations of the @xmath317-cycle have sperner capacity @xmath186 , see @xcite and @xcite . by theorems [ thm : lowerb ] and [ thm : frupb ]
this implies that the dilworth rate of their complements is @xmath321 . )
thus by theorem [ thm : lowerb ] for the complement of @xmath319 we have @xmath322 or equivalently @xmath323 .
( the digraph @xmath324 is shown on figure 4 . )
( complement of @xmath319 ) .
bidirected edges are shown as undirected ones . ]
[ prop : gyok6 ] @xmath325 * proof .
* the first inequality was already given above .
to prove the second inequality we give @xmath2 acyclic sets of vertices in the second power @xmath326^{\wedge 2}$ ] of our graph , that cover all vertices in @xmath327^{\wedge 2})$ ] .
the existence of this covering implies that @xmath328^{\wedge 2})\le\chi_{\rm dir}([\vec{a}_5^c]^{\wedge 2})\le 6 $ ] thus by lemma [ lem : nemno ] we get @xmath329 .
let us denote the vertices of @xmath330 by @xmath331 in their cyclic order , so that in @xmath319 we have @xmath332 and the unique outneighbor of @xmath333 is @xmath4 .
( that is , the outdegree @xmath1 vertex is @xmath333 , and thus the outdegree @xmath4 vertices in @xmath319 are @xmath318 and @xmath1 , while @xmath4 and @xmath3 have outdegree @xmath3 . ) the following six subsets of @xmath334 induce acyclic subgraphs of @xmath335^{\wedge 2})$ ] that entirely cover its vertex set : @xmath336 @xmath337 @xmath338 @xmath339 @xmath340 @xmath341 one can check that within all these five sets if @xmath342 is to the left of @xmath343 in the same line above ( @xmath344 may not all be different ) , then either @xmath345 or @xmath346 ( or both ) form an edge of @xmath319 , thus this is a missing edge in @xmath324 .
this implies that as a vertex of @xmath326^{\wedge 2}$ ] the pair @xmath347 does not send an edge to vertex @xmath348 , therefore the corresponding set of vertices induces an acyclic subgraph in @xmath326^{\wedge 2}$ ] .
this completes the proof of the second inequality . to see that @xmath349 it is enough to realize that any @xmath333 vertices of @xmath324 contains a bidirected edge , thus any acyclic induced subgraph has at most two vertices . to see equality
we can put weight @xmath350 on all the five @xmath4-element acyclic subsets .
@xmath44 [ rem : coinc ] getting the same upper bound for the values determined in corrollaries [ cor : ck ] and [ cor : tourn ] in two different ways above is not pure coincidence .
it follows from the fact that if @xmath24 is vertex - transitive then @xmath351 note that this is a generalization of the relation , that for every vertex - transitive ( undirected ) graph @xmath52 @xmath352 that we already referred to right after definition [ defi : vtrans ] in subsection [ subsec : sperner ] .
this latter equality is presented in @xcite ( proposition 3.1.1 on page 41 ) as a consequence of a more general equality concerning vertex - transitive hypergraphs ( see proposition 1.3.4 on page 7 of @xcite ) .
a vertex - transitive hypergraph is a hypergraph that attains for every pair @xmath353 of its vertices an automorphism that maps @xmath96 to @xmath97 .
proposition 1.3.4 in @xcite states that if @xmath289 is a vertex - transitive hypergraph then @xmath354 where ( as before ; cf .
the discussion after stating proposition [ prop : asyfrac ] ) @xmath355 for a directed graph @xmath24 we attach again the hypergraph @xmath303 where @xmath304 .
it is straightforward that if @xmath24 is vertex - transitive then so is @xmath356 .
the equality quoted for @xmath357 from @xcite gives the stated equality @xmath358 for vertex - transitive digraphs @xmath24 .
imagine that the handwritten message is left to bob by one of his three secretaries but it is not known in advance which one .
their handwriting is rather different and this has two consequences that are important for us .
one is that the possible mistakes bob can make when decoding the message are different depending on which secretary wrote him the message .
( for example , in the first secretary s handwriting a @xmath0 can be thought to be a @xmath1 , while the second secretary `` crosses '' the leg of @xmath0 , so it can never look like a @xmath1 , however it can be confused with a @xmath318 , etc . )
this means that in place of the noisy channel @xmath7 we had so far , now there are three distinct channels @xmath359 , and @xmath360 and we do not know in advance which one will be used .
the other important consequence of the secretaries handwriting being different is that bob recognizes who wrote the message , i.e. , he will know which one of the three noisy channels model the actual situation .
the relevant characteristics ( the graphs @xmath361 ) of each of these channels are known by bob and also by his bank .
now it is the bank that will send the second , error - free but expensive message to bob .
although the bank knows the characteristics it does not know which secretary left the first message . so the second message should make bob able to decide whether his decoding ( of the first message ) was correct irrespective of which secretary wrote it .
as before , we are interested ( asymptotically ) in the shortest possible message the bank can send to satisfy the requirements .
notice that this scenario is basically that of having a compound channel for the first communication .
see @xcite for more on compound channels from a zero - error point of view . here is the abstract setting for the above situation .
we have @xmath145 distinct noisy channels described by the family @xmath362 .
the relevant properties of this set are characterised by the family of directed graphs @xmath363 .
[ defi : comp ] ( cf .
@xcite and @xcite ) the dilworth rate of a family of directed graphs @xmath364 all having the same vertex set @xmath78 , is @xmath365 where @xmath366 denotes the graph on the common vertex set @xmath82 of the graphs @xmath367 with edges set @xmath368 .
[ prop : multi ] if @xmath369 is the shortest possible message the bank should send to inform bob about the correctness of his decoding of the handwritten message for @xmath6 consecutive rounds , then @xmath370 * proof . *
it is enough to prove @xmath371 .
let a proper coloring of the graph @xmath372 be fixed and agreed on by bob and the bank in advance .
bob knows that he received the first message via , say , @xmath373 . since the fixed coloring is a proper coloring of @xmath374 , proposition [ prop : chi ] implies that the right hand side is an upper bound .
if @xmath369 would be smaller , then there is some @xmath204 for which @xmath375 has two adjacent vertices for which the bank sends the same message .
if the channel in use is just @xmath373 then proposition [ prop : chi ] implies that the right hand side above is also a lower bound .
@xmath44 the interesting fact about the above quantity is that it is not more then its obvious lower bound .
[ thm : wm ] ( @xcite , cf .
also @xcite ) for every finite family of directed graphs @xmath364 we have @xmath376 the analogous result for witsenhausen rate ( that is the special case of the above when all graphs are undirected ) is proven in @xcite .
the above general form is already stated by nayak and rose in @xcite ( cf .
remark [ rem : nrfam ] of the present paper ) .
they write that the proof uses essentially the same argument as in @xcite and they omit it for the sake of brevity .
we do the same .
as is the case with sperner capacity , dilworth rate also has relevance in extremal set theory .
( recall that both notions got their name from this relationship , cf .
remark [ rem : name ] . )
these connections are uncovered when we consider the @xmath6-length sequences of vertices of a ( di)graph @xmath24 as characteristic vectors of partitions of a @xmath6-element set .
we already mentioned that if @xmath24 is the digraph consisting of two vertices and a single oriented edge between them , then the dilworth rate is just the asymptotic exponent of the minimum number of sperner systems ( antichains in the boolean lattice ) that cover all subsets of a @xmath6-element set ( the elements of the boolean lattice ) .
this is known ( and easy to prove ) to be @xmath377 , that is the asymptotic exponent is @xmath3 .
( the situation with sperner capacity is similar : its value for the above mentioned single edge graph is the asymptotic exponent of the size of a largest sperner system on a @xmath6-element set which is easy to see to be @xmath1 . ) here we present another example that we believe to be interesting .
let us call a family of pairs of disjoint subsets @xmath378 of a @xmath6-element set cross - intersecting if for every two pairs @xmath378 and @xmath379 both of the intersections @xmath380 and @xmath381 are nonempty .
( in other words , @xmath382 iff @xmath383 . )
bollobs @xcite proved that for such a family @xmath384 .
now we ask , what is the minimum number of cross - intersecting families that can cover all possible pairs of disjoint subsets of a @xmath6-element set . if we are satisfied with determining the asymptotic exponent ( i.e. not the exact value ) of this number , then this question is equivalent to asking the dilworth rate of an appropriate graph . [
prop : bolfam ] let @xmath385 denote the minimum number of cross - intersecting families that cover all pairs of disjoint subsets of a @xmath6-element set . then @xmath386 * proof . *
let @xmath17 be the following directed graph .
the vertex set of @xmath17 is @xmath387 and the edge set is @xmath388 .
that is @xmath17 has two undirected ( bidirected ) edges connecting @xmath3 to the other two vertices and one oriented edge from @xmath1 to @xmath4 . if we encode pairs of disjoint sets of a @xmath6-element set by ternary sequences ( the positions of @xmath1 s are the elements of @xmath389 and the positions of @xmath4 s are the elements of @xmath390 in the ternary sequence encoding the pair @xmath378 ) , then it is immediate to see that @xmath385 is just the chromatic number of @xmath391 .
thus @xmath392 can indeed be interpreted as the limit in the statement .
now we have to show that @xmath393 .
we have @xmath394 , so we have @xmath395 by theorem [ thm : upb ] .
since @xmath17 contains an undirected edge , @xmath391 contains a symmetric clique of size @xmath396 .
this implies @xmath397 and thus @xmath398 .
the two inequalities prove @xmath399 @xmath44
here we consider the more ambitious setup , where bob , otherwise in the same situation as described in the introduction , should decode the actual message with zero - error .
( not only getting to know whether his earlier decoding was correct or not . )
it remains true that all the relevant information to solve this problem is contained by the directed graph @xmath10 defined at the beginning of subsection [ subsect : dilwdef ] .
we will need the following operation on directed graphs .
[ defi : hacek ] let @xmath17 be a directed graph on vertex set @xmath78 . let the _ closure graph _ @xmath400 of @xmath17 be the following undirected graph . @xmath401 and @xmath402 @xmath403 note that if @xmath404 then @xmath405 is the graph where two vertices @xmath14 and @xmath13 are connected if and only if the input letters they represent can result in the same output letter .
this output letter can be one of @xmath14 and @xmath13 but also a third element @xmath97 of the alphabet .
( recall that the input and output alphabets of the noisy channel @xmath7 are identical . )
the last possibility means that @xmath406 may have edges the two endpoints of which are not adjacent in @xmath10 in either direction .
for example , if @xmath10 has three vertices , @xmath407 and only two ( directed ) edges @xmath408 and @xmath409 , then @xmath10 is a bipartite graph , while @xmath406 is the complete ( undirected ) graph on @xmath333 vertices .
now we are ready to state the graph theoretic solution of the problem considered here .
[ thm : w ] let @xmath410 denote the minimum number of bits alice should send to bob via the noiseless channel for making bob able to decode a @xmath6-length sequence of the source output with zero - error .
( the subscript @xmath411 stands for `` complete '' . ) then @xmath412 the witsenhausen rate of the closure graph @xmath406 . *
* assume that a @xmath6-length source output is sent through channel @xmath7 , and the second message sent by alice is shorter than @xmath413^{\wedge t})$ ] .
then there are two @xmath6-length source outputs , that is two sequences @xmath414 in @xmath415^{\wedge t})$ ] that are adjacent in @xmath416^{\wedge t}$ ] and for which alice sends the same message when encoding either of them for the noiseless channel .
the adjacency of @xmath81 and @xmath417 in @xmath416^{\wedge t}$ ] means that for every @xmath257 there is a @xmath418 such that both @xmath419 and @xmath420 can result in @xmath421 when sent through the noisy channel @xmath7 .
( the reason of this can be that @xmath422 or that @xmath423 is an edge of @xmath10 , in which case @xmath424 or @xmath425 is an edge of @xmath10 and @xmath426 or we have @xmath427 , where @xmath421 differs from both @xmath419 and @xmath420 . ) thus if bob s original decoding of alice s ( first ) message was @xmath428 then he knows that the message sent could be either of @xmath81 or @xmath417 .
since alice s second message for @xmath81 is identical to that for @xmath417 , bob will not know even after receiving the second message whether the original message was @xmath81 or @xmath417 . on the other hand ,
if the length of alice s second message is at least @xmath413^{\wedge t})$ ] then alice can make bob able to decide for sure what the original message was
. indeed , fix a proper coloring of @xmath416^{\wedge t}$ ] with @xmath429^{\wedge t})$ ] colors in advance that is known by both parties .
encode each color by a ( distinct ) sequence of @xmath430^{\wedge t})\rceil$ ] bits .
if the original message was @xmath431 then send bob the ( codeword for the ) color of @xmath432 . since as a vertex of @xmath416^{\wedge t}$
] @xmath433 is connected to all those sequences that could result in the same sequence when sent through @xmath7 what @xmath432 can result in , all these sequences have a different color than @xmath432 in our coloring of @xmath416^{\wedge t}$ ] .
thus when bob gets to know the color of @xmath432 from alice s second message he will know that whatever he saw at the output of @xmath7 could only arise from @xmath432 as the input .
so he will decode @xmath432 with zero - error .
thus we proved that @xmath434^{\wedge t})\rceil.\ ] ] so @xmath435^{\wedge t})=r_{\rm w}({\rm cl}(\vec{g}_h))$ ] as stated .
@xmath44 not every graph can appear as the closure graph @xmath436 of some directed graph @xmath24 .
let @xmath52 be a(n undirected ) bipartite graph with @xmath437
. then @xmath52 can not be the closure graph of any directed graph .
* let @xmath400 be the closure graph of a directed graph @xmath17 .
observe that if @xmath400 has an edge @xmath438 connecting two vertices that were not adjacent ( in either direction ) in @xmath17 , then @xmath438 is contained in a triangle in @xmath400 .
let @xmath52 be a bipartite graph with more edges than vertices .
by bipartiteness @xmath52 contains no triangle , so if it is a closure graph of some graph @xmath24 , then @xmath24 is just a directed version of @xmath52 .
if any vertex have indegree at least @xmath4 in @xmath24 that would generate a triangle in @xmath436 , so the closure graph could not be @xmath52 itself .
since the sum of indegrees equals the number of edges , we can not avoid having a vertex with indeegree two if @xmath439 for any finite simple undirected graph @xmath52 , there exists a directed graph @xmath17 such that @xmath400 contains @xmath52 as an induced subgraph . * proof .
* let @xmath52 be an arbitrary finite simple undirected graph .
for every edge @xmath440 consider a new vertex @xmath441 .
we add the oriented edges @xmath442 and @xmath443 to our graph @xmath52 .
now delete the edges of @xmath52 thus obtaining a graph @xmath17 on vertex set @xmath444 containing only the @xmath445 oriented edges leading to some vertex @xmath441 .
it is straightforward to see , that @xmath400 contains graph @xmath52 as an induced subgraph . @xmath44
the general problem concerning the dilworth rate is to determine it for specific directed graphs .
since this is a difficult and mostly open problem to the related notions of shannon and sperner capacities as well as for the witsenhausen rate , we can not expect that this problem is easy .
nevertheless , we have seen some digraphs for which it was solvable ( at least when using some non - trivial results already established for sperner capacity ) . still , there are some directed graphs for which determining the dilworth rate seems particularly interesting .
[ prob : altot ] what is the dilworth rate of the graph @xmath324 we presented in subsection [ subsec : dichrom ] ? recall that we know @xmath446 tournaments play a special role in our setting , because they are exactly those oriented graphs the complement of which is also an oriented graph ( that is one without bidirected edges ) .
so it may have some particular interest how their dilworth rate behave .
is there a tournament @xmath172 for which @xmath447 is strictly smaller than @xmath448 ?
useful discussions with imre csiszr are gratefully acknowledged .
j. krner , coding of an information source having ambiguous alphabet and the entropy of graphs , in : _ transactions of the 6th prague conference on information theory , etc .
_ , 1971 , academia , prague , ( 1973 ) , 411425 .
g. simonyi , graph entropy : a survey , in : _ combinatorial optimization _ , ( w. cook , l. lovsz , p. seymour eds . ) , dimacs series in discrete mathematics and theoretical computer science , volume 20 , ams , providence , ri , 1995 , 399 - 441 . | we investigate a communication setup where a source output is sent through a free noisy channel first and an additional codeword is sent through a noiseless but expensive channel later . with the help of the second message the decoder should be able to decide with zero - error whether its decoding of the first message was error - free .
this scenario leads to the definition of a digraph parameter that generalizes witsenhausen s zero - error rate for directed graphs .
we investigate this new parameter for some specific directed graphs and explore its relations to other digraph parameters like sperner capacity and dichromatic number .
when the original problem is modified to require zero - error decoding of the whole message then we arrive back to the witsenhausen rate of an appropriately defined undirected graph .
keywords : zero - error , graph products , sperner capacity , dichromatic number , witsenhausen rate |
the counter - intuitive phenomenon of effective attractions between like - charges immersed in an electrolyte has recently attracted considerable attention , and questions one of the fundamental tenets of colloid science @xcite .
the understanding of the underlying mechanism is crucial for a correct description of a vast variety of industrial and natural processes , in particular the stability and phase behaviour of colloidal dispersions . in this article
, we investigate the stability of electrostatically swollen stacks of regularly spaced planar colloids or membranes of infinite lateral extension in an electrolyte .
such a one - dimensional geometry describes the early stages of clay swelling @xcite but is also relevant for lamellar phases of charged bilayers @xcite . at the simplest level of mean - field description , with neglect of excluded volume and coulomb correlation effects ,
the inhomogeneous density profiles of microscopic co- and counter - ions forming the electric double layers around the planar macroions can be obtained analytically @xcite .
the resulting poisson - boltzmann osmotic ( or disjoining ) pressure in the stack , is invariably found positive @xcite , which is the signature of the tendency to swell .
this behaviour is reminiscent of the repulsive nature of pair interactions ( as we shall see below , the two phenomena are however distinct ) , established within the same mean - field framework @xcite and recently extended beyond mean - field @xcite , with the possible inclusion of approximate correlation contributions .
note that more refined incorporation of discrete solvent effects by adding a bilinear non local term to the free energy of standard poisson - boltzmann theory has shown the possibility of a net attraction ( negative pressure ) at high surface charges @xcite .
the purpose of the present investigation is twofold .
first , we apply the general local density functional formalism of ref .
@xcite to provide a prediction for the stability of stacked or lamellar phases .
we show in section [ sec : swelling ] that the interactions of electric double layers generically lead to swelling whereas describing an equilibrium spacing between the membranes either requires the inclusion of non electrostatic forces ( usually van - der - waals like ) or the consideration of more refined theories .
second , the pair potential problem is considered in section iii with a resulting effective repulsion at all distances , within the same framework that encompasses in particular the non - linear poisson - boltzmann theory .
this statement is consequently in contradiction with the attractive behaviour reported by sogami , shinohara and smalley @xcite .
we show that their controversial finding is attributable to a confusion in the thermodynamic potential describing the electrostatic situation under study .
before setting the framework of the analysis , it is worthwhile to point out that the results obtained in @xcite for a pair of colloids do not apply to the regular stacked situation under scrutiny here ( the precise definition of the geometry is given at the beginning of section iia )
. first of all , the mechanical route followed in refs .
@xcite ( deriving the interactions by integration of the stress tensor over the colloids surfaces ) would yield a vanishing effective force in a regular stack , with cancelling contributions acting on both sides of the platelets , and is thus uninformative about the stability of the array .
moreover , the situation considered in @xcite is that of two colloids immersed in an electrolyte , confined in a cylinder with an axis parallel to the colloids line of centers , and of infinite extension along this axis [ see figure [ figure]-a ) ] .
the assumption of infinite length is crucial for the validity of the effective repulsion ( see appendix [ app : a ] ) and the corresponding confined two - body problem does not include the multi - body stack geometry , for which the confining cylinder ( wigner - seitz cell ) would be a slab of finite length .
this point is in contradiction with one erroneous conclusion reached in ref .
@xcite .
we consider a regular succession of rigid infinite parallel plates located at @xmath0 ( @xmath1 ) .
each plate carries a uniform surface charge @xmath2 , and releases its counter - ions in the electrolyte solution , considered to be a mixture of @xmath3 microions , where species @xmath4 has charge number @xmath5 and local density @xmath6 .
we write the free energy@xcite of the total charge distribution in the wigner - seitz cell around a given platelet ( e.g. that situated at @xmath7 for which the cell is a slab extending between @xmath8 and @xmath9 ) as @xmath10\,d z \,+\ , \frac{1}{2 } \ , \int_{-h}^h \rho_c(z)\,\psi(z ) \
, d z , \label{eq : lda}\ ] ] where @xmath11 is the local total charge distribution , including the microions and the plate ( with global electroneutrality @xmath12 ) .
@xmath13 is expressed per unit area of the macroscopic surfaces . within the primitive model of electrolytes
@xcite , whereby the solvent is regarded as a mere dielectric continuum of permittivity @xmath14 , the total electrostatic potential @xmath15 is the solution of poisson s equation @xmath16 \label{eq : poisson}\ ] ] and can be cast in the form @xmath17 where @xmath18 is the appropriate green s function .
the boundary condition of vanishing electric field at @xmath19 is fulfilled by @xmath20 .
this mean - field like reduction of the problem materialized by the introduction of the wigner - seitz slab , may be corrected by correlation terms included in the free energy density @xmath21 . moreover , even if the van der waals energy term @xmath22 is of mean - field form , correlation terms can be accounted for provided they translate into a local correction to the free energy , in the spirit of the approaches depicted in refs @xcite .
consequently , the term @xmath23 in equation ( [ eq : lda ] ) does not reduce in general to the entropic microions contribution ( as in poisson - boltzmann theory @xcite ) but may also include both non mean - field energetic and entropic effects .
of course , the present formalism encompasses the non - linear poisson - boltzmann and modified poisson - boltzmann theories @xcite . for both
canonical ( fixed ionic content of the electrolyte solution ) and semi grand - canonical ( when the solution is in osmotic equilibrium with a salt reservoir ) descriptions , the optimal density profiles @xmath24 are obtained by minimizing the functional @xmath25 where @xmath26 is either the lagrange multiplier associated with the constraint of fixed average concentration , or the chemical potential of species @xmath4 in the reservoir .
the resulting stationary equations read for all species @xmath4 @xmath27 these implicit relations between the electrostatic potential and the densities \{@xmath28 } allow to solve poisson s equation ( [ eq : poisson ] ) and compute the optimal profile @xmath24 ( hereafter noted @xmath6 without ambiguity ) , from which we deduce the helmholtz free energy @xmath29 , with @xmath30 ( @xmath31 is the total internal energy , @xmath32 the absolute temperature and @xmath33 the entropy of the total micro - ions charge distribution in the cell ) . from the knowledge of @xmath34 , the definition of the thermodynamic potential @xmath35 ( which variations correspond to the reversible work performed by an operator and thus define the osmotic pressure ) requires the specification of the thermodynamic situation under consideration . in the canonical case with constant charge plates , @xmath36 @xcite .
if on the other hand , the platelets are held at constant potential ( e.g. by an external generator ) , @xmath35 is the legendre transform of @xmath34 with respect to the surface charge @xcite , namely @xmath37 where @xmath38 denotes the surface potential . in the opposite case of a system in equilibrium with a salt reservoir , @xmath39 in any case ,
the osmotic pressure is defined as @xmath40 where @xmath35 depends on the electrostatic situation considered , as explained above .
it is however important to stress that the osmotic pressure should not depend on the electrostatic situation under study , as becomes clear below .
theories that do not result in the above fundamental invariance of @xmath41 can consequently be disposed of .
the free energy variation induced by changing the inter - membrane distance is computed in appendix [ app : b ] with the result @xmath42_{z = h } 2 \delta h\,+\ , \sum_{\alpha}\mu_\alpha\delta n_{\alpha}\,+\ , \psi_{_{\cal p } } \delta \sigma , \label{eq : fvar}\ ] ] so that the osmotic pressure reads @xmath43_{z
= h},\ ] ] independently of the situation of constant potential or constant charge considered .
not surprisingly , the pressure is related to the legendre transform of the free energy density , as for ordinary homogeneous gases @xcite and equals the local stress @xmath44 at the mid - plane between the membranes : @xmath45+\sum_{\alpha } n_{\alpha}(z ) \frac{\partial f}{\partial n_\alpha}(z ) .
\label{eq : piofz}\ ] ] when the stack is in osmotic equilibrium with a salt reservoir , the comparison of @xmath41 obtained in section [ ssec : osmotic ] with the pressure exerted by the reservoir quantifies the tendency towards swelling ( @xmath46 ) or collapse ( @xmath47 ) . for consistency ,
the ( neutral ) reservoir needs to be described within the same framework as the electrolyte around the platelet .
the remainder of this section is devoted to the proof that @xmath41 is extremal in the reservoir , and that this extremum is a minimum under fairly general conditions .
we first analyze the @xmath48 dependence of the pressure @xmath49 , defined by @xmath50 where @xmath51 is the solution of poisson s equation ( [ eq : poisson ] ) . from eq .
( [ eq : piofz ] ) we have @xmath52 that can be recast making use of the stationary condition ( [ eq : station ] ) @xmath53 @xmath49 thus goes through an extremum in the reservoir ( @xmath54 by definition ) where the charge density vanishes ( unlike at the mid - plane @xmath55 where @xmath56 ) .
relation ( [ eq : dpi ] ) together with poisson s equation implies that the local stress introduced in ( [ eq : piofz ] ) fulfills the mechanical equilibrium condition for the fluid of microions : @xmath57 which is the simple form taken in a one - dimensional problem by the condition of vanishing divergence for the generalized maxwell stress tensor in the dielectric medium @xcite .
the second derivative of @xmath49 can be obtained by introducing the hessian matrix @xmath58 @xmath59 that obeys the relation @xmath60 as can be seen from eq .
( [ eq : station ] ) .
we thus have @xmath61 which is a positive quantity when the matrix of second derivatives @xmath62 is positive definite . under this assumption , @xmath49 is a convex - up function of the potential and the extremum attained in the reservoir is therefore a minimum , so that @xmath46 . in spite of the differences outlined at the beginning of section [ sec : swelling ] between the pair potential and the multi - body stacked problems , the method employed here is close to that of ref .
@xcite , and we shall also distinguish two different situations to conclude with the stability analysis * _ case a ) _ : the free energy density does not depend on the elementary charge @xmath63 ( as in mere mean - field treatments @xcite ) . in the limit @xmath64 ( at fixed valency @xmath65 ) , we obtain a locally neutral mixture where the total free energy functional reduces to the contribution @xmath23 .
the thermodynamic stability criterion of this neutral mixture implies the positive definiteness of the matrix @xmath62 that is independent on @xmath63 .
* _ case b ) _ : correlation or fluctuation effects are taken into account with a resulting @xmath63-dependent free energy density .
it is no longer possible to find an uncharged mixture of micro - species described by the same density @xmath21 .
the thermodynamic stability condition of the full functional ( [ eq : lda ] ) involves the green s function @xmath66 and does not imply the positive definiteness of the hessian @xmath67 .
however , the convexity of @xmath21 with respect to density variations is generally fulfilled by the approaches proposed in the literature , either in the full density range @xcite or for the small plasma coupling parameters relevant to colloidal dispersions @xcite . from the above discussion and the positive definiteness of the hessian matrix @xmath67 ,
we conclude that the stack generically exhibits a tendency towards swelling .
in this section , we consider within the density functional formalism of eq .
( [ eq : lda ] ) , the problem of the effective interactions between a pair of parallel charged plates immersed in an electrolyte solution of infinite volume ( no confinement ) . the two rigid plates with distance @xmath68
are supposed to be of negligible thickness and uniform surface charge ( as in section [ sec : swelling ] ) ; they divide the electrolyte solution into two disconnected regions ( inner region with @xmath69 and outer region with @xmath70 ) , which does not correspond to the situation analyzed in refs .
however , the generalization to the present case is straightforward and yields an effective repulsion as long as both regions are in contact with salt reservoirs imposing the same chemical potential for micro - species . indeed , from the computation of the free energy variation detailed in section [ ssec : osmotic ] , the effective force can be written @xmath71 from the vanishing of the electric field at @xmath7 and @xmath72 , we have @xmath73 and @xmath74 . for @xmath75 ,
the charge density vanishes so that @xmath76 equals the osmotic pressure in the salt reservoir under consideration .
the argument of section [ ssec : stability ] indicates that @xmath77 and that the interactions are repulsive under the assumption of positive definiteness for the stability matrix .
this last condition is obeyed by poisson - boltzmann ( pb ) theory independently of the valency of the microions [ see eq .
( [ eq : freepb ] ) below ] .
our result is consequently in contradiction with the `` long - range weak attractive part of the free energy '' reported in @xcite for the same system treated at the level of poisson - boltzmann . the work of sogami _ et al .
_ has already been criticized @xcite , but in our opinion , the subsequent controversy @xcite dwells on ambiguities on the thermodynamic potential that should be considered , which to our knowledge , have not been explicitly pointed out so far .
it thus seems worthwhile to restrict to pb theory and devote the remainder of this article to briefly revisit the model of @xcite , introduced to describe the swelling behaviour of n - butyl - ammonium vermiculite gels . within pb mean - field theory ,
the microions are considered as an ideal gas and density fluctuations discarded , so that the free energy density does not include any correlation term and reduces to the entropy of an ideal mixture : @xmath78 , \label{eq : freepb}\ ] ] where the ( irrelevant ) lengths @xmath79 involve the masses of microions and @xmath80 is the inverse temperature .
the stationary condition ( [ eq : station ] ) translates into @xmath81 and the local osmotic stress in ( [ eq : piofz ] ) is given by the ideal equation of state @xmath82 .
the electrostatic potential @xmath48 is chosen to vanish for @xmath72 , so that @xmath83 is the density of species @xmath4 far from the plates with a corresponding chemical potential @xmath84 as a result of the global electroneutrality condition , it can be checked that the final free energy [ expression ( [ eq : freepbmod ] ) below ] is independent on an arbitrary shift of the potential @xmath48 [ in which case the densities appearing in eq .
( [ eq : mualpha ] ) are simply the prefactors of the exponentials in ( [ eq : density ] ) ] .
it is convenient to use the relation between the local ionic densities and the electrostatic potential to recast the helmholtz free energy @xmath30 in the form : @xmath85 where eq .
( [ eq : equilmech ] ) and the ideal equation of state for @xmath44 have been used in going from the first to second line . note that it is understood that the osmotic term @xmath41 in ( [ eq : freepbmod ] ) takes the value @xmath86 ( resp .
@xmath76 ) for @xmath87 ( resp .
@xmath70 ) . strictly speaking ,
expression ( [ eq : freepbmod ] ) diverges ( if salt is added to the electrolyte , some of the quantities @xmath88 are extensive with system size ) .
this feature can be circumvented by computing the excess free energy with respect to a well chosen reference system ( for instance the system with same bulk densities in the absence of the plates ) . once the solution of poisson s equation is known ( see @xcite for a review of the standard solutions , including the present geometry ) , @xmath34 may be computed from equation ( [ eq : freepbmod ] ) . following this route
, we readily recover the helmholtz free energy obtained by sogami _
by means of a charging process . in equation ( [ eq : freepbmod ] ) ,
@xmath89 stands for the total surface charge on a platelet , including both the different inner ( @xmath90 from @xmath91 ) and outer ( @xmath92 from @xmath93 ) contributions , denoted respectively @xmath94 and @xmath95 in @xcite . in the model of @xcite ,
the electrostatic potential is imposed to be continuous throughout the system and the surface potential @xmath96 is independent on the distance @xmath68 between the plates .
moreover , the numbers of microions between the plates also depend on @xmath9 with a fixed chemical potential , given by ( [ eq : mualpha ] ) where the @xmath97 can be considered as the ionic densities of a reservoir in chemical equilibrium with both the inner and outer parts of the electrolyte solution around the plates .
consequently , the thermodynamic potential @xmath98 defined in eq .
( [ eq : constantpot ] ) should be used in computing the force whereas @xmath34 has been considered in @xcite .
the definition of the model imposes that when @xmath9 changes at fixed @xmath96 , @xmath95 is constant while @xmath94 varies , so that the two situations of constant charge and constant potential are equivalent for the outer part of the system ( @xmath70 ) , but not for the inner part . the reversible work performed by an operator changing the distance @xmath68 between the charged plates
is therefore given by the variations of @xmath99 whereas @xmath100 has been considered in @xcite .
omission of the chemical potential terms @xmath101 in the right hand - side of eq .
( [ eq : potpot ] ) leads the force computed in @xcite to depend on the masses ( through the lengths @xmath102 ) , which is impossible in an equilibrium statistical mechanics theory , as already pointed out in @xcite . with the aid of the relations given in @xcite [ e.g. eqs ( 63 ) , ( 64 ) and ( 65 ) valid in the case of counter - ion dominance between the plates ] , an explicit computation of the force @xmath103
yields the standard expression @xmath104 \label{eq : potea}\\ & = & \frac{\varepsilon}{8 \pi } \left [ \left(\frac{\partial \psi}{\partial z}\right)^2_{z = h^+}- \left(\frac{\partial \psi}{\partial z}\right)^2_{z = h^- } \right ] , \label{eq : poteb}\end{aligned}\ ] ] where the continuity of the co- and counter - ion charge density across the membrane ( resulting from the imposed continuity of the potential ) has been used in going from ( [ eq : potea ] ) to ( [ eq : poteb ] ) .
the compatibility of eq .
( [ eq : potea ] ) with ( [ eq : force ] ) is transparent . in terms of the variables used in @xcite , where @xmath105 denotes the reduced potential at mid - distance between the plates
, we get @xmath106 \,\geq 0,\ ] ] hence an effective repulsion at all distances , that is in the present case entirely due to the electrostatic pressure ( the osmotic contribution cancelling on both sides of the plates ) .
it can be checked that in the dual situation where both the inner @xmath90 are outer @xmath92 surface charges are held constant , the effective force is still given by ( [ eq : potea ] ) .
this translates into the legendre identity @xmath107_{\sigma_{\hbox{\scriptsize in}}\,,\,\sigma_{\hbox{\scriptsize out } } } \,=\ , \frac{\partial \left ( { \cal r}_\psi + \sigma_{\hbox{\scriptsize in } } \psi_{_{\cal p}}\right ) } { \partial h}\biggl|_{\sigma_{\hbox{\scriptsize in}}\,,\,\sigma_{\hbox{\scriptsize out } } } \label{eq : identity}\ ] ] that can be considered as a test for the consistency of the thermodynamic potential used . in the limit of counter - ion dominance between the plates investigated in @xcite , the identity ( [ eq : identity ] ) can be checked explicitly with the thermodynamic potential @xmath108 used here .
describing the interactions between the electric double - layers around charged planar colloids with a local density functional theory for ionic screening and the primitive model of electrolytes , we have shown that a regular stack of such plates generally displays a swelling behaviour when electrostatic forces alone are taken into account . within the same framework , the effective pair potential is found to be repulsive at all distances , in contradistinction with the results derived in @xcite .
the repulsive interactions are evidenced without resort to an explicit solution of poisson s equation , and are related to the convexity of the underlying free energy functional , as already noted in @xcite .
of course , the present result does not preclude the possibility of effective attractive pair potentials .
in particular , a drawback of the theories encompassed by eq . ( [ eq : lda ] ) is that the direct correlation function @xmath109 defined as the second functional derivative of @xmath13 is necessarily mean - field like : @xmath110 with a @xmath111-correlated short range part .
this shortcoming may be circumvented by the inclusion of non - local terms in the theory , e.g. in the spirit of the weighted density approximation @xcite .
however , we expect the local formalism considered here to be instructive for inter - plate separations much larger than the ionic size .
references @xcite proved that the effective interactions between a pair of like - charged colloids immersed with counter - ions and salt in a confining cylinder of infinite extension [ see figure [ figure]-a ) ] were repulsive .
we show here that the result does not hold for a finite length cylinder .
we consider two colloids confined in the cylinder of length @xmath68 represented in figure [ figure]-b ) , in the specific case of neumann boundary conditions on the surface @xmath112 ( vanishing normal electric field ) . due to the mirror symmetry of the charge distribution
, the electric field has no @xmath113-component in the plane @xmath7 , and the problem is equivalent to that of a unique colloid ( say colloid 1 ) in a sub - cell cylinder of length @xmath9 , with again neumann boundary conditions [ the right half part of the cylinder , indicated by a dotted rectangle in fig . [
figure]-b ) ] .
we assume the effective force acting on colloid 1 repulsive ( @xmath114 ) .
then , we translate along the @xmath113 axis the left half cell ( @xmath115 ) by a distance @xmath68 , keeping the right half fixed .
the electrostatic potential around colloid 1 is unaffected , so that the effective force is unchanged , corresponding now to an effective attraction . any repulsive configuration with neumann boundary conditions on the confining cylinder
can then be mapped onto an attractive one ( from the construction of the mapping , it appears that this feature disappears in the limit @xmath116 ) .
the hypothesis of infinite length is thus a key ingredient of the proofs in refs .
it is moreover worthwhile to note that numerical solutions of the non - linear poisson - boltzmann equation for finite - size disc - like clay platelets confined in a finite length wigner - seitz cylinder , show repulsive effective pair forces when the distance between the clay particles is smaller than the half length @xmath9 of the cylinder @xcite .
it will be shown that upon modifying the distance between the plates ( @xmath118 ) , the helmholtz free energy changes according to eq .
( [ eq : fvar ] ) .
the present derivation bears some similarities with other ones in the related context of poisson - boltzmann cell theory @xcite . from the definition of the free energy , equation ( [ eq : lda ] ) and the symmetry @xmath119
, we have @xmath120 consider first the energetic contribution .
making use of poisson s equation ( [ eq : poisson ] ) , two integrations by parts [ with vanishing terms @xmath121 yield @xmath122 the stationary condition ( [ eq : station ] ) was used in going from ( [ eq : appb1 ] ) to ( [ eq : appb2 ] ) .
similarly , with @xmath123 and since @xmath124 reduces to the microions charge distribution @xmath125 outside the platelets ( in particular between @xmath9 and @xmath126 ) @xmath127 gathering results , we obtain equation ( [ eq : fvar ] ) . 99 -1.4 cm g.m . kepler and s. fraden , phys .
* 1994 * , _ 73 _ , 356 ; m.d .
carbajal - tinoco , f. castro - romn and j.l .
arauz - lara , phys .
e * 1996 * , _ 53 _ , 3745 ; a.m. larsen and d.g .
grier , nature * 1997 * , _ 385 _ , 230 ; j.c .
crocker and d.g .
grier , phys .
1996 * , _ 77 _ , 1897 . n. grnbech - jensen , k.m . beardmore and p. pincus , physica a * 1998 * , _ 261 _ , 74 ; e. allahyarov , i. damico and h. lwen , phys .
e * 1999 * , _ 60 _ , 3199 ; d. goulding and j .- p .
hansen , europhys .
lett * 1999 * , _ 46 _ , 407 ; p. linse and v. lobaskin , phys .
lett . * 1999 * , _ 83 _ , 4208 ; j.j .
gray , b. chiang and r.t .
bonecaze , nature * 1999 * , _ 402 _ , 750 ; j.j .
arenzon , j.f .
stilck and y. levin , eur .
j. b * 1999 * , _ 1 _ , 79 .
r. van roij , j.p .
hansen and m. dijkstra , phys .
e. * 1999 * , _ 59 _ , 2010 .
warren , j. chem .
phys . * 2000 * , _ 112 _ , 4683 .
squires and m.p .
brenner , phys .
lett . * 2000 * , _ 85 _ , 4976 ; i.v .
ivlev , cond - mat/0012212 . h. van olphen , `` clay colloid chemistry '' ( john wiley , new york , 1977 ) .
m. dubois , t. zemb , l. belloni , a. delville , p. levitz and r. setton , j. chem .
1992 * , _ 75 _ , 944 .
g. gouy , j. phys . *
1910 * , _ 9 _ , 457 ; d.l .
chapman , phil . mag . * 1913 * , _ 25 _ , 475 .
d. andelman in `` membranes , their structure and conformation '' , edited by r. lipowsky and e. sackmann ( elsevier , amsterdam 1996 ) .
neu , phys .
rev . lett .
* 1999 * , _ 82 _ , 1072 . j.e .
sader and d.y .
chan , j. colloid interface sci . * 1999 * , _ 213 _ , 268 ; j.e . sader and d.y .
chan , langmuir * 2000 * , _ 16 _ , 324 .
e. trizac , phys .
e * 2000 * , _ 62 _ , r1465 .
y. burak and d. andelman , j. chem . phys . *
2001 * , _ 114 _ , 3271 . i.s .
sogami , t. shinohara and m.v .
smalley , molec .
* 1991 * , _ 74 _ , 599 . i.s .
sogami , t. shinohara and m.v .
smalley , molec .
1992 * , _ 76 _ , 1 . s. levine and d.g . hall , langmuir * 1992 * , _ 8 _ , 1090 .
r. ettelaie , langmuir * 1993 * , _ 9 _ , 1888 .
overbeek , molec . phys .
* 1993 * , _ 80 _ , 685 .
e. ruckenstein , adv .
colloid interface sci . *
1998 * , _ 75 _ , 169 .
m.v . smalley and i.s .
sogami , molec . phys . * 1995 * , _ 85 _ , 869 ; m.v .
smalley , langmuir * 1995 * , _ 11 _ , 1813 ; m.v .
smalley in `` ordering and phase transition in charged colloids '' , edited by a.k .
akora and b.v.r tata ( vch , 1996 ) .
e. trizac and j.l .
raimbault , phys .
e * 1999 * , _ 60 _ , 6530 . in this article ,
equation ( 14 ) [ and therefore ( 16 ) ] are erroneous as extra terms arise from the application of green s theorem .
consequently , while the effective interactions remain repulsive when the specific size of the micro - ions is taken into account via the modified poisson - boltzmann theory put forward in @xcite , a similar conclusion can not be reached for the situation of complete confinement ( finite length of the confining cylinder ) .
appendix [ app : a ] provides a simple counter - example showing that effective attraction can result from complete lateral confinement .
the above error has been corrected in the more general analysis of ref .
for a review article , see j .-
hansen and h. lwen , annu .
phys . chem . *
2000 * , _ 51 _ , 209 . m.j .
stevens and m.o .
robbins , europhys .
lett . * 1990 * , _ 12 _ , 81 . h. lwen , j .-
hansen and p.a .
madden , j. chem .
1993 * , _ 98 _ , 3275 . m.c .
barbosa , m. deserno and c. holm , europhys .
lett . * 2000 * , _ 52 _ , 80 . for a review ,
hansen and e. smargiassi , in `` monte carlo and molecular dynamics of condensed matter systems '' , edited by k. binder and g. ciccotti ( sif , bologna , 1996 ) , or r. evans , in `` fundamentals of inhomogeneous fluids '' , edited by d. henderson ( marcel dekker , new york , 1992 ) .
m. eigen and e. wicke , j. phys .
1954 * , _ 58 _ , 702 ; v. kralj - igli@xmath128 and a. igli@xmath128 , j. phys .
ii france * 1996 * , _ 6 _ , 477 ; i. borukhov , d. andelman and h. orland , phys .
1997 * , _ 79 _ , 435 and electrochimica acta * 2000 * , _ 46 _ , 221 .
l. lue , n. zoeller and d. blankschtein , langmuir * 1999 * , _ 15 _ , 3726 .
a. diehl , m.n .
tamashiro , m.c .
barbosa and y. levin , physica a * 1999 * , _ 274 _ , 433 .
groot , j. chem . phys . * 1991 * , _ 95 _ , 9191 .
r.j.f leote de carvalho , e. trizac and j .- p .
hansen , phys .
e * 2000 * , _ 61 _ , 1634 . | we show that describing the screened electrostatic interactions in a periodic stack of rigid parallel identical charged plate - like colloids within a local density functional theory approach , generically leads to a swelling behaviour . within the same framework
, we find that the effective potential between a pair of such plates immersed in an electrolyte is repulsive at all distances .
this general result is in contradiction with a theory put forward by sogami , shinohara and smalley , that we criticize and thoroughly reconsider to show how the two approaches can be reconciled . |
the superposition of states is a fundamental feature of quantum mechanics . recently
, the arbitrary superposition of fock states @xcite has been produced in a superconducting resonator with a josephson phase qubit @xcite .
this offers novel ways to directly study the quantum coherence of the photon field , i.e. , superposition of number states .
also , this strongly coupled qubit - resonator system @xcite , may be useful for quantum information processing ( qip ) .
we theoretically study the quantum phase measurement of the photon field in a superconducting resonator coupled to two phase qubits .
we consider probing the quantum coherence of the superposition state @xmath7 and @xmath8 are the vacuum and the multi - photon state , respectively , and @xmath9 is the relative phase .
this superposition of states leads to interference fringes .
however , in a dissipative environment , the quantum coherence of the superposition in eq .
( [ multi - photon ] ) decays rapidly as @xmath6 grows large @xcite . systematically studying such superpositions
should provide a better understanding of the decoherence process @xcite . to measure the quantum state of the photon field ,
wigner tomography can be used @xcite .
the relative phase between two number states can only cause a rotation in wigner phase space without changing the shape of the wigner function @xcite .
alternatively , here we propose a method to transfer the phase information of the photon field to the qubit , such that we can determine the relative phase precisely by measuring the quantum state of the qubit . in this way
, the qubit can be used to store and detect the quantum coherence of an arbitrary superposition of two photon number states .
the superposition of the vacuum and the single - photon state @xmath10 for @xmath11 in eq .
( [ multi - photon ] ) in a microwave cavity has been used for the quantum memory of an atomic qubit @xcite .
this experiment has demonstrated that the quantum information of the qubit can be transferred to the photon field .
however , here we find that it is necessary to use two qubits to transfer the phase information of the superposition of the vacuum and the multi - photon state @xmath12 in eq .
( [ multi - photon ] ) .
the first qubit is used for storing the quantum information of the photon field , whereas the second qubit is used as an auxiliary qubit to disentangle the first qubit from the resonator @xcite by repeatedly applying a controlled - not ( cnot ) quantum gate ( see e.g. , @xcite ) .
several proposals ( e.g. , @xcite ) have been made to implement cnot gates using superconducting qubits .
our proposed quantum phase detection method can measure the degree of quantum coherence of the photon field , which can be determined by the visibility of the detection signal @xcite .
however , the visibilities are greatly reduced due to decoherence , and depend on the quality factor @xcite ( i.e. , ratio of the frequency and the damping rate of the resonator ) .
we find that _ the visibilities can be greatly enhanced if the kerr nonlinearity of the cavity mode _
( e.g. , @xcite ) _ is exploited_. thus , the interference of the superposition of multi - photon states can be observed clearly even in the presence of a dissipative environment .
notably , the production of extremely strong kerr nonlinear strength via coupled cooper pair boxes ( cpbs ) @xcite using the effect of electromagnetically induced transparency ( eit ) ( e.g. , @xcite ) has recently been proposed @xcite . the eit effect in a lossless medium , such as quantum dots embedded in a solid - state substrate , has recently been studied @xcite .
the big problem with shor s algorithm @xcite is that it is very difficult to implement physically for numbers larger than 15 .
thus , here we use the gauss sum approach @xcite because this is implementable and could provide a very valuable test - bed or stepping stone for more powerful future implementations . our proposed phase detection scheme can be used for implementing the gauss sum , which can find the factors of a number using the periodicity of the sum of the quadratic phase factors @xcite .
the gauss sum has been realized with nmr @xcite , cold atoms @xcite and using short laser pulses @xcite .
here we study the implementation of the gauss sum in superconducting circuits .
using our proposed superconducting circuits , factors of integers @xmath13 should be obtainable by the gauss sum if the relaxation times were to be several @xmath2s .
the size of the factors are mainly limited by the coherence time of the photon field in the superconducting resonator .
thus , the factors of much larger numbers could be obtained in the future , when coherence times improve .
we also show that the quantum phase measurement approach proposed here can be applied to precision measurements .
this approach enables to precisely determine the frequency of the resonator and the strength of the kerr nonlinearity .
we show that the superposition of multi - photon states can increase the accuracy of precision measurements .
this can act as a `` frequency standard '' for the other qubits in the circuit .
this paper is organized as follows : in sec .
ii , we describe the model of the system studied . in sec .
iii , we present a method to detect the relative phase of the superposition of the vacuum and the multi - photon state in eq .
( [ multi - photon ] ) . in sec .
iv , we show that the phase detection method can be applied to both the gauss sum for factorization and precision measurements .
we close the paper with a summary . in appendix
a , we study the interaction between a qubit and a resonator in the far - detuning regime . in appendix b
, we discuss the effect of imperfect cnot gate operations on the disentanglement process .
we consider two josephson phase qubits capacitively coupled to a superconducting resonator @xcite as shown in fig .
the hamiltonian of the qubit - resonator system can be written as @xcite @xmath14 @xmath15,\\ & = & \omega{a^\dag{a}}+\sum^2_{j=1}\big\{\frac{\omega_{0j}}{2}\sigma_{jz}+g_j(a\sigma_{j+}+\sigma_{j-}a^\dag)\nonumber\\ & & \!\!-\frac{\omega_j}{2}\big[\exp{(-i\phi_j - i\omega_{qj}{t})}\sigma_{j+}+{\rm h.c.}\big]\big\},\end{aligned}\ ] ] where @xmath16 is the annihilation operator of the photon field , @xmath17 and @xmath18 are the transition and population ( pauli @xmath19 ) operators of the qubit @xmath20 respectively ( @xmath21 ) . here
@xmath22 and @xmath23 are the frequencies of the qubit @xmath20 and the resonator , respectively .
the parameters @xmath24 act as coupling strengths between the photon field and the qubits . also , @xmath25 and @xmath26 are the amplitude and the phase of the coupling between the ground state @xmath27 and the excited state @xmath28 in the @xmath20-th qubit .
the frequency @xmath29 is the frequency of the microwave drive of the qubit @xmath20 .
the qubit @xmath20 can be accurately controlled by adjusting the frequency @xmath22 and the parameter @xmath25 via a classical signal @xcite .
( color online ) circuit diagram of the qubits and the resonator for quantum phase measurements .
the josephson qubits are capacitively coupled to the coplanar waveguide resonator with couplings @xmath30 and @xmath31 .
the frequency of the qubit can be adjusted by the flux bias pulse and can be measured by a superconducting quantum interference device ( squid ) @xcite . here
@xmath32 and @xmath33 are the amplitudes of the couplings between the ground and excited states.,height=219 ] it is convenient to work in the interaction picture . by considering the unitary transformation @xmath34,\ ] ]
the transformed hamiltonians are @xmath35,\end{aligned}\ ] ] and @xmath36,\end{aligned}\ ] ] where @xmath37 and @xmath38 hereafter , the two transformed hamiltonians @xmath39 and @xmath40 at resonance ( i.e. , when @xmath41 and @xmath42 ) , @xmath43,\\ \label{qubit_resonator } h^{(j)}_{2}&=&g_j(a\sigma_{j+}+\sigma_{j-}a^\dag),\end{aligned}\ ] ] will be used frequently . also , the time evolution operators @xmath44
$ ] and @xmath45 $ ] can be obtained explicitly as @xcite @xmath46,\end{aligned}\ ] ] and @xmath47 where @xmath48 is the unit operator . to transfer the phase information of the photon field to the qubit
, it is required to switch - on and off the interaction between the qubit and the resonator .
the coupling between the qubit and the resonator is fixed , but the frequency of the qubit can be adjusted by the bias current @xcite .
the qubit - resonator interaction can thus be turned - off by far - tuning the frequency @xmath22 of the phase qubit @xcite . in appendix
a , we give a detailed discussion of the qubit - resonator coupling in the far - detuning regime .
we now present a procedure to use the qubits to probe and store the quantum coherence of the superposition of two fock states . the relative phase between the superposition of the vacuum and the _ single_-photon state in eq .
( [ multi - photon ] ) can be completely transferred to the qubit such that the qubit can act as a probe of the quantum coherence of the photon field @xcite .
the phase can be determined by measuring the qubit s state .
we require two qubits to measure the phase of the superposition state of the vacuum and the _ multi_-photon state in eq .
( [ multi - photon ] ) .
one qubit is used for probing the phase and the other qubit is used to disentangle the qubit from the resonator . in the following subsections , we will describe the different schemes for the single- and multi - photon cases .
let us now consider the quantum phase measurement of the superposition , @xmath50/\sqrt{2}$ ] , of the vacuum and the single - photon fock state in the resonator .
we can create the superposition state @xmath10 in eq .
( [ multi - photon ] ) by just using one qubit @xcite .
initially , the product state of the vacuum of the resonator and the ground state of the qubit 1 is prepared , i.e. , @xmath51 .
we first produce an equal superposition of the states @xmath27 and @xmath28 of the qubit 1 by applying a @xmath52-pulse to the qubit 1 ( i.e. , turning on the drive of the qubit 1 for a time @xmath53 ) . by applying the time - evolution operators @xmath54 in eqs .
( [ u ] ) and ( [ u1 ] ) to the state @xmath27 , we have @xmath55.\nonumber\\\end{aligned}\ ] ] now consider @xmath56 , then @xmath57,\nonumber\\ & = & \frac{1}{\sqrt{2}}\exp\big(\frac{i\omega_{01}t}{2}\big)\big[|g\rangle+i\exp{(-i\phi_1')}|e\rangle\big],\end{aligned}\ ] ] where @xmath58 .
the states can thus be written as @xmath59|0\rangle.\nonumber\\\end{aligned}\ ] ] now for @xmath60 , then @xmath61 , @xmath62 the energy of excited state of qubit 1 then lowers one level , to its ground state , and a photon is created in the resonator .
this can be derived by applying the evolution operator @xmath63 in eqs .
( [ u],[u2 ] ) giving @xmath64 and the ground state @xmath65 has not changed , @xmath66 combining eqs .
( [ phi_1],[g1],[g0 ] ) , we obtain the state @xmath67}|1\rangle\big\}.\end{aligned}\ ] ] here we have ignored the global phase factor @xmath68 in eq .
( [ pi/2 ] ) .
we then switch - off the qubit - resonator interaction and let the system evolve freely for a short period @xmath69 , such that a relative phase is acquired between the two number states @xmath70 and @xmath71 .
the total state , at the time @xmath72 , becomes @xmath73}|1\rangle\big\}.\nonumber\\\end{aligned}\ ] ] we can now transfer the phase information to the qubit 1 by switching - on the qubit - resonator interaction for the time @xmath74 in eq .
( [ timeqr ] ) .
the state @xmath75 will swap to @xmath76 , while the ground state @xmath65 remains unchanged .
the state now reads @xmath77 and @xmath78 .
here we have assumed that the period @xmath69 is much greater than the time @xmath74 . the relative phase information between the two number states is now imprinted in the qubit 1 and the qubit 1 is disentangled from the photon field in the superconducting resonator .
let us apply a @xmath52-pulse to the qubit 1 so that the final state can be written as @xmath79|e\rangle\big\}|0\rangle,\end{aligned}\ ] ] where @xmath80 , @xmath81 and @xmath53 .
we have also assumed that the time @xmath53 is much shorter than the time @xmath82 and we approximate @xmath83 .
the excited state of the qubit 1 ( with a much higher tunneling rate than that of the ground state ) can now be measured .
this can be done by applying a measurement pulse and read - out by a squid @xcite .
the phase factor can thus be determined from the probability of the excited state , which is @xmath84 the photon field inevitably suffers from the dissipation present in realistic situations .
the thermal average photon number is about zero ( @xmath85 ) for a high - frequency resonator ( @xmath86ghz ) at low temperatures ( @xmath87 @xmath88 mk ) @xcite .
the time evolution of the density matrix @xmath89 of the photon field can be described by the master equation @xcite @xmath90+\gamma(2a\rho{a^\dag}-a^\dag{a}\rho-{\rho}a^\dag{a}),\end{aligned}\ ] ] where @xmath91 is the damping rate .
we assume that the number of thermal photons is negligible . here
we ignore the decoherence effect during the qubit and qubit - resonator operations because their time durations ( @xmath92 and @xmath93 @xmath94ns in @xcite ) are extremely short compared to the dissipation time @xmath95 ( several @xmath2s in @xcite ) .
therefore , we consider the dissipation of the photon field during the free time - evolution @xmath69 .
the density matrix @xmath89 of the photon field at the time @xmath69 can then be found as @xcite @xmath96\big\}.\end{aligned}\ ] ] we then follow the same procedures discussed above .
the probability @xmath97 that the qubit is in its excited state can be readily obtained @xmath98,\end{aligned}\ ] ] where @xmath99 .
the visibility of the quantum coherence can be defined as @xmath100 where @xmath101 and @xmath102 denote the maximum and minimum values of the coherence factor @xmath103 , @xmath104 which characterizes the quantum coherence of the superposition state .
note that the visibility @xmath105 is unity when dissipation is absent .
the visibility @xmath105 can be obtained as @xmath106,\end{aligned}\ ] ] where @xmath107,\ ] ] and @xmath108 .
clearly , a higher visibility can be obtained for larger resonator quality factors ( i.e. , ratios of @xmath23 and @xmath91 ) .
( color online ) circuit diagram of the coupled qubits and the resonator .
the josephson qubits can be coupled to each other via a capacitor ( shown in dark red line ) . here
@xmath110 is the coupling strength between the two qubits.,height=219 ] we now investigate the quantum phase detection of the superposition of the vacuum and multi - photon states in eq .
( [ multi - photon ] ) , where @xmath111 .
this superposition state @xmath12 can be generated by applying an appropriate sequence @xcite of evolution operators @xmath112 to the initial state @xmath113 , where @xmath114 and @xmath115 are the @xmath116-th time steps for the time evolution operators @xmath117 and @xmath118 , respectively , and @xmath20 is the total number of steps .
we assume that we can adjust the time duration of the interactions in each step .
the time steps @xmath114 and @xmath115 can be found inversely @xcite by applying @xmath119 to the state @xmath12 in eq .
( [ multi - photon ] ) .
we now consider two qubits ( qubit 1 and qubit 2 ) for the quantum measurement .
we assume that both qubits have the same form , eq .
( [ ham1 ] ) , of the interaction with the resonator .
qubit 1 is used as an auxiliary qubit to disentangle qubit 2 from the resonator .
a cnot gate @xmath120 can be applied to the two qubits .
details for implementing a cnot gate can be found elsewhere ( e.g. , refs .
@xcite ) .
note that we now need to add a qubit - qubit coupling term in the hamiltonian in eq .
( [ ham1 ] ) ( see red capacitor in fig .
[ fig2 ] ) .
the qubits in fig .
[ fig1 ] can be coupled @xcite to each other via a capacitor so that `` spin - spin '' interactions can be produced .
this qubit - qubit coupling hamiltonian , with the coupling strength @xmath110 , can be written as @xmath121 alternatively , the qubits can also be coupled via a high - excitation - energy quantum circuit such as a superconducting resonator @xcite .
a cnot gate can be realized , e.g. , by either applying an external field @xcite , pulses @xcite or applying iswap gates [ turning - on the qubit - qubit interaction in eq .
( [ qbcoup ] ) for the time @xmath122 appropriately @xcite . by adjusting the frequencies of the two qubits , the cnot gate and
the qubit - resonator interaction can be independently operated .
we choose qubit 1 as the control qubit and qubit 2 as the target qubit such that : @xmath123 .
let us first prepare the resonator and the two qubits in a product state as @xmath124 we let the resonator evolve freely for a time @xmath69 , and the state then becomes @xmath125,\end{aligned}\ ] ] where @xmath9 is the accumulated phase factor , @xmath126 now we study how to transfer the relative phase between the two fock states to the qubit 1 .
we switch - on the interaction between the two qubits and the resonator sequentially , for the times @xmath127 and @xmath128 given by @xmath129 where we set @xmath130 .
the state thus becomes @xmath131,\ ] ] where @xmath132 .
we have assumed that the time duration @xmath69 is much greater than the times @xmath127 and @xmath128 . to perform the cnot gate ,
we detune the qubits from the resonator and set the two qubits at the same frequency .
we then apply the cnot gate such that the state @xmath133 will change to @xmath134 .
the state then becomes @xmath135,\ ] ] where @xmath136 , and @xmath137 is the time duration for the cnot gate .
we only turn - on the interaction between the qubit 2 and the resonator for a time @xmath138 , and this gives @xmath139,\ ] ] where @xmath140 . in the interaction picture
, we repeatedly apply the evolution operator @xmath141 $ ] until the number of photons in the resonator becomes zero .
we summarize this procedure in the interaction picture as @xmath142 where the time @xmath143 is @xmath144 and @xmath145 is a positive number for @xmath146 .
qubit 1 now completely disentangles from qubit 2 and the resonator . the final state becomes @xmath147}|e\rangle\bigg\}|g\rangle|0\rangle,\ ] ] where @xmath148 and @xmath149 is the total time for disentanglement .
we assume that the free - evolution time @xmath69 is much larger than the time @xmath149 so that we can ignore the relative phase accumulated during the time @xmath149 .
note that the relative phase @xmath9 is encoded on the qubit 1 .
afterwards , we apply the @xmath52 pulse to the qubit 1 and then measure the excited state of qubit 1 . we can determine the phase factor from the probability @xmath97 of the excited state of qubit 1 .
for simplicity , we set @xmath150 in eq .
( [ qubit_drive ] ) .
the probabilities of the excited state of qubit 1 then become @xmath151 where @xmath152 . in realistic situations ,
the cnot gate is not perfect due to decoherence or experimental constraints .
let us briefly examine quantum phase measurements using imperfect cnot gates .
here we assume that the fidelity of this imperfect cnot gate is very high .
the small imperfections of this cnot gate can be characterized by a parameter @xmath153 which is positive and close to one .
if this cnot gate would be perfect , then the parameter @xmath153 would be equal to one . a more detailed discussion of the effects of non - ideal cnot gate operations on the disentanglement process is given in appendix
b. we repeat the same procedure in eq .
( [ dent_prod ] ) using a number of imperfect cnot gate operations .
the system can be described by the density matrix @xmath154 @xmath155{\otimes}|g\rangle\langle{g}|0\rangle\langle{0}|,\end{aligned}\ ] ] where @xmath156 .
here we have taken the leading order approximation of the density matrix ( see appendix b ) .
we notice that qubit 1 can not be fully disentangled from qubit 2 and the resonator by using imperfect cnot gates .
we now apply a @xmath52 pulse to qubit 1 and then measure the excited state of qubit 1 . for simplicity , we set @xmath150 in eq .
( [ qubit_drive ] ) .
the probabilities of the excited states of qubit 1 are @xmath157 where @xmath152 .
the coherence factors , @xmath158 or @xmath159 in eq .
( [ prob_multi_1 ] ) , contain a coefficient @xmath160 which is smaller than one .
this means that the imperfect cnot gate operations lead to dephasing of the qubit 1 .
we now take into account the dissipation effect of the photon field in eq .
( [ master ] ) during the free evolution . after a time @xmath69
, the density matrix @xmath89 of the photon field can be written as @xcite @xmath161^{n - k}}{k!(n - k)!}|k\rangle\langle{k}|\bigg\}.\nonumber\\\end{aligned}\ ] ] we have also assumed that the decoherence of the qubit is negligible during the disentanglement process and the time @xmath149 is much smaller than the dissipation timescale @xmath162 .
the probabilities of the excited state of qubit 1 then become @xmath163 where @xmath152 , and @xmath164 and @xmath165 are two functions given by @xmath166 we note that the superposition state in eq .
( [ multi - photon ] ) can also be used to measure the phase due to the kerr nonlinearity @xcite .
the hamiltonian of the nonlinear interaction is given by @xcite @xmath167 where @xmath168 is the interaction strength .
the two coupled cpb qubits can form a nonlinear medium of the photon field @xcite ( see also ref .
@xcite ) . the strength @xmath168 of the nonlinear interaction in eq .
( [ nonlinear ] ) can attain @xmath169 ghz or even higher @xcite .
we only turn - on the nonlinear interaction during the free evolution .
thus , the phase factor @xmath170 can be rewritten as @xmath171 where @xmath172 is an effective frequency due to the kerr nonlinearity .
we then apply the same procedure to detect the phase of the photon field .
now we investigate the visibility of the quantum coherence .
the visibilities @xmath173 and @xmath174 denote the odd and even photon numbers , respectively . from eqs .
( [ visibility ] ) and ( [ prob_multi_2 ] ) , the visibility @xmath173 for an odd number of photons can be found as @xmath175 , \\
\tau_{1}&=&\frac{1}{\tilde{\omega}}\bigg\{\arccos\bigg[-\bigg(1+\frac{{\gamma}^2}{\tilde{\omega}^2}\bigg)^{-\frac{1}{2}}\bigg]\bigg\},\end{aligned}\ ] ] where @xmath176 is the parameter that characterizes the quality of the imperfect cnot gate and @xmath177 is the effective frequency in eq .
( [ effomega ] ) . for an even number of photons
, we take the form of the coherence factor @xmath178 the visibility @xmath174 for even photon numbers can be found as ( @xmath179 ) @xmath180.\end{aligned}\ ] ] the visibility @xmath173 is slightly higher than @xmath174 . for @xmath181 and @xmath182
, we find that the visibilities have no difference between the cases of single - photon state @xmath10 and multi - photon superposition state @xmath12 in eq .
( [ multi - photon ] ) .
the degree of visibility depends on the ratio of @xmath23 and @xmath91 .
however , the visibilities of the superposition of multi - photon states can be greatly enhanced with the nonlinear interaction strength @xmath168 . for @xmath183 ,
the degree of visibility depends on the ratio @xmath184 from eq .
( [ effomega ] ) such that higher visibilities can be obtained for larger number of photons ( @xmath185 ) .
this means that the superposition of multi - photon states can show a high contrast of interference fringes , even in the presence of dissipation .
we have shown that the quantum phase of the photon states in a superconducting resonator can be measured with qubits .
these measurement methods are useful for the implementation of the gauss sum ( this section ) and metrology ( the following section ) .
now we study a physical realization of the gauss sum algorithm using superconducting circuits .
the gauss sum can verify whether a number @xmath186 is a factor of another number @xmath187 or not .
the gauss sum @xcite is defined by @xmath188 if @xmath186 is a factor of @xmath187 , then @xmath189 is equal to one .
otherwise , the value of @xmath189 is less than one . to find the factors @xmath186 of the integer @xmath187 , the trial factor @xmath186 scans through all numbers from 2 to @xmath190 @xcite
thus , in this manner , it can factorize integers . however
, this gauss sum algorithm does not provide a speed - up over classical computation . to save considerable experimental resources and minimize the effects of decoherence , it is advantageous to use as few terms as possible in the gauss sum in eq .
( [ gauss_sum ] ) @xcite .
thus , the truncated gauss sum can be employed @xcite , @xmath191 where @xmath192 is a positive integer which is smaller than @xmath186 .
we only need to sum over @xmath193 terms instead of the total @xmath186 terms so that considerable experimental resources can be saved .
we can apply the same method in determining the phase factor of the photon field in the resonator using the superposition of the vacuum and single - photon state in eq .
( [ multi - photon ] ) .
we follow the same method in sec .
iii.a to generate the superposition of the vacuum and the single - photon state with equal weights , i.e. , @xmath194 waiting the following times @xmath195 for the free - evolution , allows the state to accumulate a relative phase @xmath196 between the two fock states @xmath70 and @xmath71 .
the state then becomes @xmath197.\ ] ] using the method described in sec .
iii.a , we can transfer the relative phase to qubit 1 .
the state of qubit 1 should then be measured after applying a @xmath52-pulse to it .
the phase factor @xmath198 can be determined from the probability of the excited state of qubit 1 in eq .
( [ prob_e ] ) . by repeating the same procedure @xmath193 times and setting the time duration @xmath199 for @xmath200
, we take the average of the total sum .
we readily obtain the real part of the truncated gauss sum in eq .
( [ tgs ] ) as @xcite @xmath201 indeed , the real part of the truncated gauss sums has been shown to experimentally find the factors of integers @xcite .
( color online ) truncated gauss sum for @xmath202 , plotted versus the trial factors @xmath186 .
results for a damping rate @xmath203 ( no dissipation ) are shown with black circles ; while @xmath204 are shown with red crosses .
we have summed over @xmath205 terms for @xmath206 .
the factors of @xmath187 are marked by the vertical lines located at @xmath207 and 13 .
all factors are above the threshold @xmath208 , indicated by a thin horizontal dotted line . ,
height=302 ] now we study how to apply the truncated gauss sum for checking factors . in fig .
[ fig3 ] , the truncated gauss sum @xmath209 for @xmath202 is plotted against the trial factors @xmath186 .
the truncated gauss sums are represented by black circles .
we have only used @xmath205 terms in the truncated gauss sum for @xmath186 by scanning @xmath186 from 2 to @xmath210 . in ref .
@xcite , tefak _ et al .
_ gave a threshold to discriminate factors from non - factors .
they found that the truncated gauss sum for non - factors is bounded from above by @xmath211 in the limit of large @xmath192 @xcite .
as shown in fig .
[ fig3 ] , we can see that all factors are above the threshold of the gauss sums , @xmath212 @xcite , whereas all non - factors are below the threshold , @xmath212 , shown by the horizontal dotted line .
therefore , this enables us to clearly distinguish the factors from nonfactors .
we now examine the truncated gauss sum in the presence of dissipation of the photon field .
we adopt the same method as discussed in the previous subsection to determine the phase factor . from eq .
( [ p_es ] ) , the truncated gauss sums can be written as @xmath213 the performance of the gauss sums can decrease due to dissipation .
the terms with higher @xmath214 can become vanishingly small in eq .
( [ tgs1 ] ) .
this limits the size of the number @xmath187 to be vertified by the gauss sum . in fig .
[ fig3 ] , we plot the truncated gauss sum @xmath215 as a function of the trial numbers @xmath186 for @xmath202 . here
we consider the damping rate @xmath216 ( we have taken the values of @xmath217 ghz and @xmath218 hz from the experiment in @xcite ) .
we then sum over @xmath205 terms for checking the 31 numbers @xmath219 .
as shown in fig .
[ fig3 ] , the truncated gauss sums slightly decrease due to damping ( denoted by the red cross ) .
however , we can see that the truncated gauss sum can still be used for distinguishing the factors from nonfactors if the relaxation times of resonators are about several @xmath2s @xcite .
( color online ) the truncated gauss sum with @xmath205 terms for @xmath220 is plotted versus the trial factors @xmath186 , for the nonlinear kerr interaction strength @xmath221 and damping rate @xmath222 .
the blue diamonds , red crosses and black circles represent the different numbers of photons : @xmath11 , 3 and 5 , respectively .
the factors of @xmath187 ( here : 5 , 7 , 11 , 13 , 35 , 55 and 65 ) are indicated by the vertical lines .
all factors are above the threshold @xmath208 , indicated by a thin horizontal dotted line . , height=302 ]
next we consider the multi - photon superposition states @xmath12 in eq .
( [ multi - photon ] ) for the truncated gauss sum .
we first produce the superposition of the vacuum and the multi - photon state with equal weights , i.e. , @xmath223 we let the system freely evolve for the times @xmath224 where @xmath225 is the effective frequency in eq .
( [ effomega ] ) if the kerr nonlinearity is used .
then , a relative phase between the two fock states is acquired .
the state becomes @xmath226.\ ] ] we now apply the phase detection method described in sec .
iii.b for the multi - photon case . after applying a @xmath52-pulse to qubit 1
, we measure the state of qubit 1 .
then , for an odd number of photons , we can determine the phase factor @xmath198 from the probabilities of qubit 1 in eq .
( [ prob_multi_1 ] ) .
the truncated gauss sum is given by @xmath227 where @xmath160 is a parameter originating from imperfect cnot gate operations .
in the presence of dissipation of the photon field , we can determine the phase factor from the probabilities in eq .
( [ prob_multi_2 ] ) .
the truncated gauss sum becomes @xmath228 now we study the performance of the truncated gauss sum for checking factors using the kerr nonlinearity . in fig .
[ fig4 ] , the truncated gauss sum is plotted versus the trial factors @xmath186 for @xmath220 . the truncated gauss sums with various numbers of photons [ @xmath11 ( blue diamonds ) , @xmath229 ( red crosses ) and @xmath230 ( black circles ) ]
are shown in fig .
we can see that the truncated gauss sum for the factors are much closer to unity if the larger photon numbers are used .
the truncated gauss sum using the kerr nonlinearity can show a clearer pattern to discriminate the factors from nonfactors and a much larger number can be verified @xcite .
this can be easily understood by rewriting the truncated gauss sum [ from eqs .
( [ tau ] ) and ( [ tgs2 ] ) ] as @xmath231 the exponential functions in eq .
( [ tgs3 ] ) are closer to one when the effective frequency @xmath177 becomes higher .
the larger number of photons leads to a higher value of @xmath232 .
therefore , the use of kerr nonlinearities enhances the performance of the truncated gauss sum for finding the factors of an integer , even in the presence of dissipation in the photon field . we can roughly estimate the size of the factorized number @xmath187 for which the exponential factor in eq .
( [ tgs3 ] ) is around @xmath233 . in this case , the fully factorizable numbers are then multiplied by a factor of @xmath6 and it can attain @xmath234 if the nonlinear strength @xmath168 is about the frequency of the resonator , and with the same values for the parameters discussed above and @xmath235 . if the relaxation time of the resonator were to be @xmath236s , @xmath237 ms and @xmath237 s , then the largest partially - factorizable numbers would be @xmath238 , @xmath239 and @xmath240 , respectively , where @xmath6 is the number of photons in the resonator .
( color online ) truncated gauss sum with @xmath205 terms for @xmath220 plotted versus the trial factors @xmath186 , where the nonlinear kerr interaction strength @xmath221 , the damping rate @xmath222 , and the number of photons @xmath6 is 3 .
different values of @xmath153 are shown : @xmath183 ( black circles ) , @xmath241 ( blue squares ) and @xmath242 ( red diamonds ) .
the factors of @xmath187 are indicated by the vertical lines .
all factors are above the threshold @xmath208 , indicated by a thin horizontal dotted line . , height=302 ]
note that we require more cnot gate operations to disentangle qubit 1 from both qubit 2 and the resonator .
this requirement is for the multi - photon states involving higher number of photons in eq .
( [ multi - photon ] ) . from eq .
( [ tgs2 ] ) , the truncated gauss sum scales with the coefficient @xmath160 .
thus , the imperfect cnot gates unavoidably affect the performance of the truncated gauss sum . in fig .
[ fig5 ] , we plot the truncated gauss sum for @xmath220 against the trial factors @xmath186 , using the multi - photon superposition state for @xmath229 in eq .
( [ multi - photon ] ) .
we find that the performance of the truncated gauss sum is very sensitive to small variations of @xmath153 . as shown in fig .
[ fig5 ] , the values of the truncated gauss sum for @xmath242 ( red diamonds ) are much lower than the case for @xmath241 ( blue squares ) and @xmath183 ( black circles ) .
this may limit the use of multi - photon states @xmath12 for the truncated gauss sum if the photon number @xmath6 becomes large .
we can determine the quantum phase factor of the photon states by detecting the state of the qubit .
this should enable to precisely measure the frequency of the resonator @xmath23 .
the uncertainty @xmath243 of the frequency @xmath23 of the resonator is given by @xcite @xmath244 is the number of measurements .
we now consider the detection with the superposition of the vacuum and the single - photon state .
using eq .
( [ prob_e ] ) , the minimum value of the uncertainty @xmath245 , for @xmath203 , is @xmath246 is the time duration of each measurement and @xmath247 is an odd integer . in the presence of dissipation ( @xmath248 ) ,
the uncertainty is given by latexmath:[\ ] ] where @xmath285 .
the leading order approximation is the density matrix containing the first four terms with the coefficients @xmath160 .
the remaining terms involving a large number of entangled states of qubit 1 , qubit 2 and the resonator are of the order of @xmath286 . using imperfect cnot gates ,
qubit 1 can not be completely disentangled from qubit 2 and the resonator in eq .
( [ npfstate ] ) . | we study the implementation of quantum phase measurement in a superconducting circuit , where two josephson phase qubits are coupled to the photon field inside a resonator .
we show that the relative phase of the superposition of two fock states can be imprinted in one of the qubits .
the qubit can thus be used to probe and store the quantum coherence of two distinguishable fock states of the single - mode photon field inside the resonator .
the effects of dissipation of the photon field on the phase detection are investigated .
we find that the visibilities can be greatly enhanced if the kerr nonlinearity is exploited .
we also show that the phase measurement method can be used to perform the gauss sum factorization of numbers ( @xmath0 ) into a product of prime integers , as well as to precisely measure both the resonator s frequency and the nonlinear interaction strength .
the largest factorizable number is mainly limited by the coherence time .
if the relaxation time of the resonator were to be @xmath1 @xmath2s ( @xmath3 ms ) , then the largest factorizable number can be @xmath4 ( @xmath5 ) , where @xmath6 is the number of photons in the resonator . |
in a recently published letter we have shown that the measurement of the magnetic moment of samples with ring geometry enables the identification of regions with degraded superconducting regions and can be used as a sensitive method to investigate the critical current density @xmath4 of high temperature superconducting thin films @xcite .
the ac field amplitude dependence of the ac susceptibility of structured narrow rings provides the experimental foundation for this method . within the bean model and via the determination of the so - called penetration field @xmath5 ( at which the perfect diamagnetic shielding is lost ) and geometry parameters we obtain @xmath4 . in this work
we have exploited this technique at zero dc magnetic field @xmath6 to investigate the influence of the structuring method used to fabricate the high - temperature superconducting rings on @xmath4 .
our results indicate clearly that ion beam etching is a far less destructive structuring technique than laser ablation . we have also obtained the dc field dependence of @xmath4 and analysed the effects of flux creep with an extended bean model which enables the interpretation of the frequency dependence of the ac susceptibility associated with a finite resistivity due to flux motion @xcite . we have analyzed the shift of the ac susceptibility with frequency and applied a scaling relation which has been predicted by brandt @xcite . within this theory
we have determined the flux creep exponent @xmath7 and compared it with the one determined by relaxation measurements with a squid .
the paper is organized as follows . in the next section
we give a brief summary of the measured samples and experimental details . in section
iii we provide a theoretical resume of the ac response of narrow superconducting rings as well as the determination of the flux creep exponent from ac susceptibility .
the results are described in section iv .
a brief summary is given in section v.
the ac susceptibility measurements were performed with a lake shore 7000 ac susceptometer designed for operation at low - level ac magnetic field amplitudes @xmath8 2 mt . we applied dc magnetic fields up to 3 t by a superconducting solenoid .
we have studied rings made from yba@xmath1cu@xmath2o@xmath3 high - temperature superconducting 200 nm thin films prepared by pulsed laser deposition on 1 mm thick @xmath9 substrates with a critical temperature between and 90 k @xcite .
a 30 nm @xmath10 buffer layer was used .
all the films were characterized by ac susceptibility measurements before structuring .
results on the films can be found in @xcite .
the films were structured by two methods : ( a ) by laser ablation ( la ) with an excimer workstation having an optical resolution of @xmath11 and ( b ) by ion beam etching ( ie ) . for
the ion beam etching procedure a spun - on @xmath12 m thick resist layer ( az1450 ) was structured by excimer laser ablation using a scanning method .
the resist was etched by laser apart from a thin , @xmath13 nm thick remaining film to prevent any damage of the superconducting layer .
we used ion beam sputtering with argon to transfer the mask structure into the film .
the ion etching was performed in a non - commercial ibe system with a base pressure better than @xmath14 mbar equipped with a kaufman type ion beam source operating at a beam energy of 700 ev and current density of 0.2 ma/@xmath15 .
the beam was neutralized by means of a hot filament to prevent charge accumulation on the sample surface .
the samples were mounted on a water - cooled and rotating sample holder to avoid an excessive increase of temperature during etching .
table [ samples ] shows the main characteristics of the measured samples .
in this section we briefly review the ac response of narrow superconducting rings as described in @xcite .
we discuss the limit of a thin narrow ring of width @xmath16 much smaller than the mean radius @xmath17 , . within the bean model
the virgin magnetization curve completely determines the hysteresis loop which has the shape of a parallelogram @xcite . from these loops one obtains the complex susceptibility of a superconducting narrow ring with constant @xmath4 . for cycled magnetic field
we define the nonlinear susceptibility , the quantity we measure , as @xmath18 where @xmath19 is the magnetic moment and @xmath20 the sample volume .
the virgin magnetization curve of such a ring is composed of two straight lines , for and for , since the screening supercurrent in the ring is limited to a maximum value ( @xmath21 is the ring thickness ) . as long as holds for the current induced in the ring by the applied field @xmath22 , no magnetic flux can penetrate through the ring into the hole of the ring .
when @xmath23 is reached the ring becomes transparent to magnetic flux .
therefore , when the applied field is increased further , flux lines will move through the ring as described in @xcite .
these authors treat the superconducting strip with transport current in an applied field . with the moving flux lines in the ring
, magnetic flux enters the ring hole until the screening supercurrent decreases again to the value @xmath24 .
the magnetic moment of the ring thus saturates at the value .
the applied field value @xmath5 at which this saturation is reached follows from the inductivity @xmath25 of the flat narrow ring @xcite , with @xmath26 @xcite .
the magnetic flux generated in the hole by a ring current @xmath27 is @xmath28 . as long as @xmath29 one has ideal screening , thus .
equating these two fluxes one obtains for the flat ring . at @xmath30
one reaches , thus .
therefore , one can express @xmath4 as @xmath31 } h_p\,.\ ] ] the formulae given above are accurate to corrections of order @xmath32 since the precise values of @xmath19 and @xmath25 depend on the current distribution across the ring width @xmath16 which changes during the magnetization process . with eq .
( [ eq : chi ] ) and the expressions for @xmath33 which are given in @xcite we obtain the susceptibility of the ring normalized to the initial value @xmath34 , i.e. , to @xmath35 , @xmath36 with @xmath37 and @xmath38 .
these theoretical results have been confirmed experimentally in a previous paper @xcite .
we use expression ( [ eq : jc ] ) to determine @xmath4 from the measured penetration field @xmath5 . for high enough dc magnetic fields , the influence of flux creep should be taken into account in the ac response of the superconducting ring . in our analysis
we use a current voltage law @xmath39 that is connected to a logarithmic dependence @xcite of the pinning potential on the current density and describes flux creep as an activated motion of vortices which have to overcome this potential .
following @xcite we have @xmath40 where @xmath41 is the so - called flux creep exponent .
the power law in eq .
( [ ecvlaw ] ) includes the limiting cases of bean - like field distribution for which @xmath42 and of ohmic dissipation for which @xmath43 . by investigating the frequency ( @xmath44 ) dependence of narrow rings under the application of a harmonic ac field @xmath45 during flux
creep , , brandt @xcite derived a scaling law for the frequency and magnetic field amplitude which states the following : the complex ac susceptibility at fixed temperature remains unchanged under the simultaneous transformation of time by a constant factor @xmath46 , e.g. @xmath47 , and magnetic field by a factor @xmath48 , e.g. @xmath49 , with @xmath50 .
this is equivalent to state that the following equation holds : @xmath51 using this property we determined the flux creep exponent @xmath52 at different @xmath53 and applied dc fields @xmath6 by measuring the ac susceptibility at different frequencies and by equating the ac fields at which the same susceptibility is measured , i.e. @xmath54 where @xmath46 is determined by the frequency ratio @xmath55 .
equation ( [ esusc1 ] ) is a power law for the magnetic field amplitude .
plotting @xmath56 vs. @xmath57 should then give a straight line of slope @xmath50 .
+ jnsson - kerman et al .
recently confirmed the scaling relation for the nonlinear ac susceptibility response of hgba@xmath1cacu@xmath1o@xmath58 thin films @xcite . in a previous work @xcite
the flux creep exponent @xmath59 was determined from the frequency dependence of @xmath60 in the limit of large ac amplitudes . in what follows we present a further confirmation of the scaling relation ( [ etransform ] ) by applying it on structured thin rings .
furthermore , we use a straightforward analysis procedure to obtain the field dependence of the pinning potential @xmath61 for fields up to 2 t .
in @xcite a drastic decrease of @xmath4 after structuring the films into rings was found .
this behavior has been explained by the existence of microcracks in the path of the ring . in this paper
we show that the structuring technology used to fabricate the superconducting ring has a large influence on its @xmath4 .
figure [ jcla2 ] shows the temperature dependence of the critical current density @xmath4 for sample la2 before and after structuring by laser ablation . for this sample and before structuring @xmath4 is 500 times larger than its value after structuring . for comparison the @xmath4 values for the sample ie1 before and after structuring by ion beam etching
are given in fig .
[ jcie1 ] . for the ring of this sample @xmath4
is reduced only 30 % compared to the unstructured sample . for all measured samples
we have found that @xmath4 of the measured la - rings is at least 100 times smaller than @xmath4 before structuring . on the other hand
the results for the ie - rings vary only a few percent . on the origin of the remarkable difference in @xmath4 of the rings structured by laser ablation we speculate as follows .
during treatment by the laser the tiny region within the laser spot is heated up to several thousands kelvin for a few milliseconds . because of the bad thermal conductivity of the substrate the temperature of the film rises to several hundred degrees in a small region around the spot as well .
it is known that heating can change the oxygen concentration of @xmath62 . as a result
the superconducting properties at the edges of the structured path may be altered up to the complete loss of superconductivity . also , thermally induced shock waves may produce microcracks within the width of the ring and degrade the maximum critical current density .
our measurements of the thickness profile of the ring after structuring indicate a region of @xmath63 at the edges of the ring where the film appears to be completely damaged .
in contrast we found sharp edges on films structured by ion beam etching . to compare both structuring methods we can use a better way to present the data by plotting the in phase component of the magnetic moment @xmath64 as a function of the ac magnetic field amplitude .
this is shown for the rings la2 , fig .
[ mla2 ] and ie1 , fig .
[ mie1 ] . the la - ring in fig .
[ mla2 ] shows a clear two - step - transition , one in the low - field - range and the other in the high - field range .
we recognize a crossover from the ring with a lower @xmath4 to the bulk signal for a long strip @xcite , see fig .
[ mla2 ] . on the other hand the ie - ring in fig .
[ mie1 ] shows only a slight deviation from the expected ideal ring behavior ( continuous line in fig .
[ mie1 ] ) very probaly due to small inhomogeneities within the ring . to check whether the laser ablation structuring process leads to a `` homogeneous '' or `` inhomogeneous '' @xmath4 over the width of the ring we structured sample
la1 once more by ion beam etching from a width of 0.2 mm to 0.05 mm .
whereas in the `` homogeneous '' case we do not expect a change in the measured @xmath4 in the `` inhomogeneous '' case a different value for @xmath4 should be measured due to the existence of islands with higher or lower critical current density . in fig .
[ jcla1 ] the results for the two consecutive structuring procedures are shown .
we assign the observed decrease in @xmath4 after the second structuring to tiny islands of lower @xmath4 which are along the path of the ring with smaller width and were created after the first laser ablation structuring , indicating strongly inhomogeneous film properties . in a similar experiment we reduced the width of the ie - ring ( notch ) by cutting a notch into it by ion beam etching , leaving only a small ( 16 @xmath65 m ) path where the current can flow . by this
means we basically measure @xmath4 in the path region .
figure [ jcnotch ] presents the critical current density corresponding to bulk , ring and ring - notch signal . within experimental uncertainty @xmath4
remains unchanged .
this result supports the assumption of a homogeneous @xmath4 in the film .
ion beam etching is therefore a far less destructive structuring technique than laser ablation .
for the measurements presented in this section we have chosen rings with highest critical current density and with different radius to width ratios , e.g. rings ie1 and ie2 . together with a dc bias field the small ac perturbation @xmath66mt was applied . in fig .
[ pdcshift ] one recognizes that the dc bias field mainly shifts @xmath67 to lower ac amplitudes and leaves the dependence of @xmath68 on the ac field amplitude essentially unaltered .
this behavior is due to a decrease in the critical current density that determines the field @xmath69 at which the magnetic flux enters into the center of the ring . during the increase of the applied dc field
a current is induced in the ring . as described in section [ s : ac ] perfect shielding occurs as long as the current density is below the critical one . at currents larger than @xmath4 flux can move into the center until the critical value is reached again . at large dc fields the magnetic moment of the ring
thus saturates at a value that is determined by the shielding capability of the ring which is given by the critical current related to the specific dc field . by applying then a harmonic ac signal to the ring
, its magnetic moment has a hysteresis loop around the static dc bias field similar as for @xmath70 .
since the application of the dc field leads to flux penetration it will influence the pinning of vortices in the ring material as well . by this
means the flux creep exponent @xmath71 is altered . in fig .
[ pdcshift ] an increase in flux creep can be found for large dc fields . as predicted by brandt @xcite a slight smoothing of or @xmath60 around the penetration field @xmath5
is observed for increasing dc field , i.e. decreasing @xmath59 .
from the shift of @xmath5 with dc field one can determine the field dependence of @xmath4 shown in fig .
[ pfieldjc ] .
the field dependence of @xmath4 we obtain is similar to that measured by vibrating reed experiments in y123 thin films @xcite .
the observed field dependence of @xmath4 can be understood within the collective pinning theory for three dimensional pinning mechanism @xcite .
three dimensional pinning is assumed since the correlation length of the vortex lines along the field direction is smaller than the thickness of the films @xcite .
the magnetic field independence of @xmath4 observed at low fields indicate that the vortices are pinned independently , i.e. the single vortex pinning regime . at higher fields
the interaction between flux lines leads to the formation of flux bundles which are pinned collectively and therefore @xmath4 decreases with field .
in what follows we concentrate on the field dependence of the pinning potential @xmath61 . for a series of dc fields we measured the amplitude dependence of @xmath60 at four different frequencies .
such a scan can be seen in fig .
[ psuscf](a ) for the ring ie2 . in a step by step procedure
we determine the ac field amplitudes corresponding to the frequencies for a fixed value of @xmath60 where the slope is largest .
this is indicated by the arrow in fig .
[ psuscf](a ) .
then we plot the logarithm of the obtained magnetic ac field amplitudes versus the logarithm of the respective frequencies , see fig .
[ pfshift ] , and by performing a linear fit we determine the corresponding slope @xmath50 for different dc fields . within the available frequency range
this plot gives straight lines for fixed dc field and temperature as theoretically @xcite expected .
this gives the flux creep exponent @xmath59 for fixed temperature and dc field , see eq .
( [ hsca ] ) .
figure [ psuscf](b ) shows the rescaling of all four curves in ( a ) following eqs .
( [ ecvlaw ] ) and ( [ etransform ] ) and using the obtained value for @xmath59 . in routine measurements
one only needs to measure @xmath68 around the maximum in @xmath72 to determine the shift of @xmath60 with frequency ( only 15 to 20 points per scan ) .
the necessary information on the ac field amplitude range to be measured at each dc field can be obtained by a preliminary scan . within 20 hours
we were thus able to measure the scans required to determine @xmath7 for 10 dc fields at one fixed temperature .
scans for the two selected rings and temperatures and have been performed and analyzed in this manner .
the dependence of the flux creep exponent @xmath7 and the pinning potential @xmath73 on dc field are shown in fig .
[ pufield ] . in fig .
[ pufield](a ) one clearly sees the strong decrease of @xmath59 with dc field .
it can be also noted that @xmath59 increases the lower the temperature of the ring .
the large values of @xmath74 at @xmath70 for both rings are consistent with the little frequency dependence of the ac response at low dc fields .
for this reason the bean model is a good approximation in this limit . at the same @xmath53 and @xmath6 we find .
we assign this difference in @xmath59 between the two rings to variations in the film properties from which the rings were structured .
the same qualitative statement holds for @xmath4 as well , see fig .
[ pfieldjc ] . in fig .
[ pufield](b ) one notes for both rings a similar dependence of @xmath61 on @xmath6 .
we distinguish a low- and a high - field region in which we obtain straight lines in a double logarithmic plot . assuming a power law for the field dependence of @xmath61 , e.g. @xmath75
, we obtain exponents @xmath76 and @xmath77 in the low- and high - field regions , respectively . at @xmath78k and in a dc field range between 20 and 100 oe a similar dc field dependence for the flux creep
exponent @xmath79 has been recently observed for hgba@xmath1cacu@xmath1o@xmath58 thin films ( @xmath80k ) @xcite .
figure [ pufield](b ) also shows the values for @xmath61 obtained from relaxation measurements .
we used this technique to verify our results by an independent method . in the presence of flux
creep the magnetic moment @xmath33 relaxes after ramping of the external field @xmath6 to some value . after a transient time @xmath81 one
observes the universal relaxation @xcite @xmath82 with a squid magnetometer we have performed three relaxation measurements with the ring ie2 .
the results are shown in fig .
[ prelax](a ) .
following eq .
( [ erelax ] ) we obtain the flux creep exponent @xmath59 and hence the pinning potential @xmath61 from a linear fit in the double logarithmic plot
. with the respective values for @xmath83 , @xmath84 vs. @xmath85 should be time independent .
this is nicely confirmed in fig .
[ prelax](b ) .
a similar time dependence with @xmath86 at @xmath87 t has been obtained for the irreversible magnetization of bi@xmath1sr@xmath1ca@xmath1cu@xmath2o@xmath88 samples below 20 k @xcite . as shown in fig .
[ pufield](b ) , the relaxation measurements provide values for the pinning potential @xmath61 which are similar to those obtained with the nonlinear ac susceptibility scaling procedure .
in this work we have used the nonlinear ac susceptibility method on structured yba@xmath1cu@xmath2o@xmath3 high - temperature superconducting rings to study the influence of different patterning methods to the critical current density of the films .
the nonlinear susceptibility method applied on superconducting rings of small enough widths is a suitable method to identify regions with degraded superconducting properties .
we have found that laser ablation structuring degrades the superconducting properties of the film , decreasing strongly its critical current density and introducing inhomogeneities within @xmath89 m from the patterned edges .
in contrast , the ion beam etching procedure is a far less destructive technique . measuring the nonlinear susceptibility of the rings at different frequencies we reconfirmed the recently proposed scaling relation for the frequency and field amplitude dependence .
the scaling relation allows us to obtain the field dependence of the flux creep exponent or pinning potential at different temperatures .
our results show a power law dependence for the flux creep exponent @xmath90 for fields below @xmath91 t and at reduced temperatures @xmath92 .
the authors thank m. lorenz for providing us with the thin films .
this work was supported by the german - israeli foundation for scientific research and development ( grant g-0553 - 191.14/97 ) and by the innovationskolleg `` phnomene an den miniaturisierungsgrenzen '' ( dfg ik 24/b1 - 1 ) .
f. mrowka , m. wurlitzer , p. esquinazi , e. h. brandt , m. lorenz and k. zimmer , appl .
. lett . * 70 * , 898 ( 1997 ) . for a review
see , e.g. , e. h. brandt , rep .
prog . phys . * 58 * , 1465 ( 1995 ) .
e. h. brandt , phys .
b * 55 * , 14513 ( 1997 ) .
m. lorenz , h. hochmuth , d. natusch , h. brner and k. kreher , _ mrs proceedings _ , vol . 341 ( mrs , pittsburgh , 1994 ) , p. 189 .
m. lorenz , h. hochmuth , d. natusch , h. brner , g. lippold , k. kreher and w. schmitz , appl .
phys . lett . *
68 * , 3332 ( 1996 ) .
m. wurlitzer , m. lorenz , k. zimmer and p. esquinazi , phys .
b * 55 * , 11816 ( 1997 ) . t. ishida and h. mazaki , j. appl . phys . *
52 * , 6798 ( 1981 ) .
j. gilchrist and m. konczykowski , physica c * 212 * , 43 ( 1993 ) . e. h. brandt and m. v. indenbom and a. forkl , europhys
. lett . * 22 * , 735 ( 1993 ) . e. h. brandt and m. indenbom , phys .
b * 48 * , 12893 ( 1993 ) .
j. gilchrist and e. h. brandt , phys .
b * 54 * , 3530 ( 1996 ) . l. d. landau and e. m. lifshitz , _ electrodynamics of continous media _ , vol .
iii of theoretical physics ( pergamon , oxford , 1963 ) .
e. zeldov , n. m. amer , g. koren , a. gupta , m. w. mcelfresh , and r. j. gambino , appl .
* 56 * , 680 ( 1990 ) . v. m. vinokur , m. v. feigelman and v. b. geshkenbein , phys .
* 67 * , 915 ( 1991 ) .
b. j. jnsson - kerman , r. v. rao and e. h. brandt , phys .
b * 60 * , 14913 ( 1999 ) .
b. j. jnsson , k. v. rao , s. h. yun and u. o. karlsson , phys .
b * 58 * , 5862 ( 1998 ) .
e. h. brandt , phys .
b * 49 * , 9024 ( 1994 ) ; * 50 * , 4034 ( 1994 ) .
m. ziese and p. esquinazi , z. phys .
b * 94 * , 265 ( 1994 )
. m. u. feigelman , v. m. vinokur , phys .
b * 41 * , 8986 ( 1990 ) .
e. h. brandt , phys .
lett . * 76 * , 4030 ( 1996 ) . v. v. makarov , y. kopelevich and s. moehlecke , physica c * 264 * , 213 ( 1996 ) .
.characteristics of the measured rings .
the structuring methods were laser ablation ( la ) and ion beam etching ( ie ) .
@xmath17 denotes the radius of the ring , @xmath16 its width , @xmath0 and @xmath4 ( 77 k ) the critical temperature and critical current density at zero applied field .
@xmath4 was calculated using eq .
( [ eq : jc ] ) . [ cols="^,^,^,^,^,^,^,^ " , ] | we have studied the influence of the patterning procedure on the critical current density of high-@xmath0 yba@xmath1cu@xmath2o@xmath3 thin rings using the nonlinear ac susceptibility method . at no applied dc magnetic field
we have found that laser ablation degrades strongly the critical current density whereas ion beam etching has only a weak influence on it . from the measurements at different frequencies and
dc magnetic fields we analyzed the influence of flux creep and obtained the field dependence of the flux creep exponent .
our data reconfirm the recently observed scaling relation for the nonlinear susceptibility response of type - ii superconductors . |
evidence for efimov three - body bound states , which were proposed originally in the context of nuclear physics @xcite , has been observed in a number of ultracold atom gas experiments @xcite . in principle , near a magnetic - field feshbach resonance @xcite there exists an infinite number of three - body bound states that follow a discrete scaling law .
the primary signature of these three - body states in cold atom gases has been resonantly enhanced three - body loss of trapped atoms .
a loss resonance occurs at a negative value of the two - body scattering length @xmath2 , which is denoted @xmath3 , where the energy of the efimov state coincides with the scattering threshold energy for three atoms @xcite , as shown schematically in fig .
[ fig : efimov ] ( a ) .
several experiments have observed multiple efimov loss features whose locations follow discrete scaling , with each @xmath3 larger than that of the last by a factor of @xmath4 , where @xmath5 is a universal parameter @xcite .
an additional signature of efimov states can be found when the energy of an efimov state coincides with the threshold scattering energy for a feshbach molecule and an atom .
this occurs at a positive value of @xmath2 denoted @xmath6 and results in resonant collisional loss in a trapped gas mixture of feshbach molecules and atoms .
atom - molecule loss resonances have been observed for @xmath7li @xcite , @xmath8cs @xcite , and the mixture of @xmath0k and @xmath1rb @xcite .
in addition , unanticipated resonances in the loss of trapped atoms at positive @xmath2 values , without initially creating molecules , have been seen for @xmath9li @xcite , @xmath10k @xcite , and the mixture of @xmath11k and @xmath1rb @xcite .
the observed loss features are relatively small , with the increase in the atom loss rate ranging from about a factor of two to a factor of five .
the features can be quite narrow , with the width ranging from a few @xmath12 to a few hundred @xmath12 , where @xmath12 is the bohr radius .
these resonances are believed to be related to @xmath6 and have been attributed to an avalanche mechanism @xcite , whereby feshbach molecules that are produced by non - resonant three - body recombination eject atoms from the trap via resonant , secondary atom - molecule collisions .
however , there has not yet been an observation of an avalanche feature and an atom - molecule loss resonance in the same system .
in addition , a recent theoretical simulation suggests that the avalanche mechanism fails to produce a narrow atom loss feature near the atom - dimer resonance @xcite .
the observation of an atom loss feature does not require the preparation of feshbach molecules and therefore can be a simpler method for experimentally locating @xmath6 .
however , it is important to verify the connection of observed avalanche peaks with atom - molecule efimov resonances . in previous work @xcite
, we measured an atom - molecule loss resonance for @xmath0k@xmath1rb feshbach molecules and @xmath1rb atoms , but did not see any corresponding loss feature for an atom gas prepared without creating a population of trapped feshbach molecules .
the measured atom - molecule loss rate coefficient @xmath13 is presented in fig .
[ fig : efimov ] ( b ) , showing resonant loss around @xmath14 . because this data could have missed a narrow or a small amplitude avalanche peak , we present here additional atom loss measurements on the positive @xmath2 side of the @xmath0k@xmath15rb feshbach resonance .
in particular , in order to search for an efimov - related avalanche feature , we take many more data points with a finer spacing in @xmath2 . in addition
, we ensure uniformity of the initial atom gas conditions as we change @xmath2 , since variation of the densities or temperature could shift or broaden a resonance feature @xcite .
finally , we look for features in both the atom loss as well as the ratio of the number loss for rb and k , since the avalanche mechanism should result in additional loss of rb atoms from the resonant , secondary collisions . efimov bound states . at @xmath16 , resonant enhancement of three - body recombination
is observed . at @xmath17 , resonant enhancement of atom - molecule inelastic collisions
is observed and enhanced atom loss due to the avalanche mechanism has been postulated . in the scattering process cartoons ,
red and blue collision partners represent @xmath1rb and @xmath0k atoms , respectively .
( b ) measured atom - molecule loss rate coefficient as a function of @xmath2 in a mixture of rb and rbk @xcite .
resonant atom - molecule loss was observed near @xmath18 ; the line shows a fit to a theoretical lineshape @xcite.,width=288 ] the rest of paper is organized as : sec .
[ sec - ii ] describes how we prepare the ultracold bose - fermi mixture and measure atom loss , sec .
[ sec - iii ] presents the experimental results , which are compared against predictions based on the avalanche mechanism near an atom - molecule efimov resonance , and sec .
[ sec - iv ] gives conclusions .
our measurements start with an ultracold mixture of bosonic @xmath1rb atoms in the @xmath19=@xmath20 state and fermionic @xmath0k atoms in the @xmath21 state , where @xmath22 corresponds to the atomic angular momentum and @xmath23 is its projection .
an @xmath24-wave feshbach resonance is used to control the interactions between @xmath1rb and @xmath0k atoms , where @xmath2 as a function of magnetic field @xmath25 is given by @xmath2=@xmath26-@xmath27 , @xmath28=@xmath29 , @xmath30=@xmath31 , @xmath32=@xmath33 @xcite .
the atom gas is initially prepared at a magnetic field @xmath34 g below @xmath35 , which corresponds to @xmath36 .
we keep the temperature @xmath37 of the gas greater than @xmath38 as well as greater than @xmath39 , where @xmath40 is the transition temperature for bose - einstein condensation of @xmath1rb and @xmath41 is the fermi temperature of @xmath0k . we have investigated atom loss in a single - beam optical dipole trap characterized by trapping frequencies for rb of @xmath42 radially and @xmath43 axially . in our far - detuned optical dipole trap ,
the trapping frequencies for k are larger than those for rb by a factor of 1.4 .
the optical trap beam propagates along a horizontal direction , with a beam waist of @xmath44 and a wavelength of @xmath45 .
the atom gas mixture is prepared with an initial number of rb atoms , @xmath46 , between @xmath47 and @xmath48 , an initial number of k atoms , @xmath49 , between @xmath50 and @xmath51 , and an initial temperature between 0.7 @xmath52k and 1.0 @xmath52k .
three - body recombination produces a feshbach molecule with a kinetic energy determined by the binding energy . in order to make sure that our trapping potential does not confine these krb molecules or the energetic atoms resulting from scattering with krb molecules , we want the trap depth to be lower than the binding energy of krb molecules @xcite . as a consequence ,
we take measurements for @xmath53 in a trap just deep enough to hold an atom gas with a temperature @xmath54 . to extend our measurement to larger values of @xmath2 where the binding energy of krb is smaller , for @xmath55
we lower our trap depth to be just deep enough to hold an atom gas with a temperature of @xmath56 , with @xmath57 radially and @xmath58 axially . for all of the data ,
the binding energy of the molecules is greater than 1.5 times @xmath59 , where @xmath60 is the boltzmann constant . to measure loss
, we use a magnetic - field sweep to quickly increase @xmath2 and then wait for fixed amount of time @xmath61 as shown in the inset of fig .
[ fig : alpha ] . the magnetic field
is then returned to the original value where @xmath36 and both atom species are imaged a few milliseconds after release from the optical trap .
the final atom numbers for rb and k , which we denote @xmath62 and @xmath63 , respectively , are determined from fits to gaussian distributions . combining this with the measured initial atom numbers , @xmath64 and @xmath65 , yields the loss rate .
we take data for values of @xmath25 during the hold time @xmath61 that correspond to @xmath2 from @xmath66 to @xmath67 .
the hold time @xmath61 is changed for different ranges of @xmath2 in order to keep the fractional number loss @xmath68 between @xmath69 and @xmath70 , where @xmath71 and @xmath72 .
the value of @xmath61 varies from @xmath73 at small @xmath2 to @xmath74 at large @xmath2 . for @xmath75 s. the dashed lines
show the measured initial rb and k atom numbers .
the lower panel shows the extracted three - body recombination rate coefficient @xmath76 based on eq .
( [ eq : alpha ] ) .
inset shows magnetic - field sweep : the magnetic field @xmath25 is increased to a value near the feshbach resonance in @xmath77 ms , held at that value for a time @xmath61 , and then swept back to the original value in @xmath77 ms.,width=288 ] fig .
[ fig : alpha ] shows a subset of our loss measurement data for a hold time , @xmath61 , of 1 s. this two - point measurement approach ( measuring the number at time 0 and at time @xmath61 ) trades accuracy for precision .
specifically , a full measurement of the loss curve , where the atom numbers are measured at many different times , allows for a more accurate determination of the three - body rate coefficient at a particular value of @xmath25 , however the faster two - point measurement minimizes the effect of drifts in experiment parameters and therefore enhances the precision and our ability to detect any small loss peaks as we vary @xmath25 . in order to combine data taken for different hold times @xmath61
, we extract an approximate three - body rate coefficient @xmath76 . for three - body recombination of @xmath1rb+@xmath1rb+@xmath0k , @xmath76
is defined by @xmath78 @xcite , where @xmath79 and @xmath80 are number densities of @xmath1rb and @xmath0k , respectively . to simplify this differential equation
, we can use the fact that k and rb share almost the same polarizability in our optical dipole trap and ignore the small relative sag between rb and k clouds .
assuming a gaussian density profile consistent with a harmonically trapped maxwell - boltzmann gas , we can re - write the integral in terms of the total number @xmath81 and temperature @xmath37 as @xmath82 @xcite . here ,
@xmath83 , @xmath84 , @xmath85 is the number ratio @xmath86 , @xmath87 is the atom masses of @xmath1rb .
although the number ratio @xmath85 can change during a measurement , the parameter @xmath88 is only weakly dependent on @xmath85 . in the approximation
that the temperature and the parameter @xmath88 are constant during @xmath61 , @xmath76 can be solved for analytically , @xmath89\frac{t^3}{6a\bar{\omega}^6\delta t}.\ ] ] using the average initial number ratio @xmath90 and initial temperature @xmath37 , we obtain @xmath76 using eq .
( [ eq : alpha ] ) ( see fig . [
fig : alpha ] , lower panel ) . as a check ,
we have compared the results from our two - point measurements using eq .
( [ eq : alpha ] ) with previous data where @xmath76 was extracted from many measurements of the number of atoms as function of time @xcite , and we find that they agree to within a factor of @xmath91 .
we previously identified an efimov - like resonance between @xmath0k@xmath1rb feshbach molecules and @xmath1rb atoms at the scattering length @xmath92 with width of roughly @xmath93 @xcite . according to refs .
@xcite , this atom - molecule resonance can also result in enhanced atom loss at scattering lengths near @xmath6 for a gas initially consisting of atoms only . here
, non - resonant three - body recombination of atoms produces energetic @xmath0k@xmath1rb feshbach molecules that then collide with atoms multiple times to result in atom loss .
in addition to enhanced total atom loss , our two - species atom gas could provide an additional signature for this avalanche scenario .
namely , the number loss ratio @xmath94 should also show a resonant increase that coincides with the enhanced atom loss feature , since the collision channel @xmath0k@xmath1rb+@xmath1rb is enhanced while @xmath0k@xmath1rb+@xmath0k is not @xcite . here
, the number loss ratio is defined by @xmath95 fig .
[ fig : avalanche1 ] shows our measurement results , where each point on the plots shows the average of four repeated measurements within each dataset having the same value of @xmath61 .
the vertical error bars indicate the standard deviation of the mean , while the horizontal error bars indicate the range of @xmath2 values used in the averaging .
the upper panel in fig .
[ fig : avalanche1 ] shows the loss rate coefficient @xmath76 extracted using eq .
( [ eq : alpha ] ) while the lower panel shows the number loss ratio @xmath94 .
we see no clear evidence for an avalanche peak . specifically , aside from the deviation from @xmath96
scaling at small values of @xmath2 , the dominant features in @xmath76 appear to be small systematic shifts that occur when we combine datasets taken with different values of @xmath61 , and we can easily rule out the presence of any feature where @xmath76 is increased by a factor of two or more .
the measured number loss ratio @xmath94 has an average value of approximately @xmath91 , which is the expected value for three - body recombination with no additional avalanche mechanism loss .
the measurement of @xmath94 has a lower signal - to - noise ratio than @xmath76 and one can identify some possible peaks in the data .
however , these peaks have no corresponding feature in @xmath76 . in addition , our measured number loss ratio is qualitatively inconsistent with predictions from an avalanche mechanism model as shown by dashed and dot - dashed lines in the lower panel of fig .
[ fig : avalanche1 ] . the amplitude of these potential peaks in our data is smaller than that of the model by a factor of 2 or more and the width is narrower by a factor of 10 or more .
the avalanche model is described in detail below . in ref .
@xcite , machtey _ et al .
_ present an improved avalanche model based on ref .
the results of this model show qualitative agreement with the two @xmath9li experiments @xcite and the @xmath10k experiment @xcite , although the predicted widths for enhanced loss were typically several times larger than the observed avalanche peak widths .
monte - carlo methods have also been used to simulate the avalanche loss for homonuclear systems @xcite .
these simulations were applied to the @xmath9li system and result in an even wider avalanche loss feature , with widths that are 10 to 20 times larger than the observed atom loss features @xcite . for comparison with our data , we have applied the model of machtey _ et al .
_ , which we modify for the heteronuclear case , to calculate the expected avalanche peak for our experiment parameters . in the machtey _
_ model , the elastic and inelastic atom - dimer cross sections are used to calculate the probability that a dimer created by three - body recombination undergoes a specific number of secondary elastic collisions with atoms before exiting the trap
. a weighted sum of these probabilities then yields the expected number of extra atoms lost due to the avalanche mechanism . to extend this model to a two - species gas
, we use the scattering length for a rb atom and a krb feshbach molecule given in ref .
@xcite : @xmath97a$ ] with constants @xmath98 , @xmath99 , and @xmath100 .
the efimov parameters , @xmath101 and @xmath92 , are taken from the fit to atom - dimer trap loss data in ref .
@xcite . in terms of @xmath102 ,
the atom - molecule elastic and inelastic cross sections are given by @xmath103 and @xmath104 , respectively @xcite .
for the initial collision , the relative wave number @xmath105 is calculated assuming the atom is essentially at rest and the molecule has a kinetic energy given by @xmath106e_\text{b}$ ] , where @xmath107 is the binding energy of the molecule . in each subsequent collision with an atom at rest , the mean energy of the dimer is multiplied by a factor of @xmath108/\left(m_{\text{k}}+2m_{\text{rb}}\right)^2\approx0.52 $ ] . for secondary collisions ,
a key parameter is the column density of the trapped atom gas , @xmath109 . in the calculation
, we use the average density of the rb atoms for @xmath110 and the geometric mean root - mean - squared width of the trapped gas for @xmath111 . for our data , @xmath112 and @xmath113 . with these parameters ,
we calculate the mean number of rb atoms lost per three - body recombination event , which can be directly compared to our number loss ratio data .
the red dashed curve in the lower panel of fig .
[ fig : avalanche1 ] shows the model result .
because a much lower efimov resonance inelasticity parameter , @xmath114 , can be extracted from fits to three - body loss data @xcite , we also show the model result for @xmath115 ( black dot - dashed curve in fig .
[ fig : avalanche1 ] ) . for either value of @xmath114 ,
the model predicts a wide resonance feature in atom loss and in @xmath94 , which is clearly inconsistent with our measurements .
given that we do not observe a feature consistent with these predictions for an efimov avalanche peak , it is useful to compare the parameters for our system to those of experiments where avalanche peaks have been observed . in particular ,
compared to the @xmath9li experiment of ref .
@xcite , our temperature is very similar ( within a factor of 2 ) , the mean size of the trapped gas @xmath111 is similar ( within @xmath116 ) , and our atom density is an order of magnitude larger , which should be favorable for observing an efimov avalanche peak . in addition , the efimov resonance parameters @xmath6 and @xmath114 used to model the @xmath9li loss feature @xcite are very similar to those for the @xmath0k-@xmath1rb case .
our trap aspect ratio is larger than that of ref .
@xcite by a factor of 15 , but similar to that of ref .
@xcite , which also reported an avalanche peak for @xmath9li .
finally , we note that we have also taken measurements in a crossed - beam optical dipole trap with aspect ratio of 30 , and again no clear avalanche loss feature was observed .
we have measured the three - body recombination loss rate and the number loss ratio for a @xmath0k-@xmath1rb atom gas mixture at positive scattering length over the range from @xmath66 to @xmath67 in a search for a feature connected to the previously observed atom - dimer efimov resonance at @xmath92 .
while an avalanche model has been used to interpret atom loss features seen in other systems as being a consequence of an atom - dimer resonance , our measurements do not show a loss feature consistent with this model .
the fact that there remains no single system in which both resonant loss in an atom - dimer gas mixture and a corresponding loss feature for an atom gas have been observed is problematic for validation of this explanation of these atom loss peaks .
t. kraemer , m. mark , p. waldburger , j. g. danzl , c. chin , b. eneser , a. d. lange , k. pilch , a. jaakkola , h .- c .
ngerl , and r. grimm , nature ( london ) * 440 * , 315 ( 2006 ) .
t. b. ottenstein , t. lompe , m. kohnen , a. n. wenz , and s. jochim , phys .
* 101 * , 203202 ( 2008 ) .
j. h. huckans , j. r. williams , e. l. hazlett , r. w. stites , and k. m. ohara , phys .
lett . * 102 * , 165302 ( 2009 ) .
g. barontini , c. weber , f. rabatti , j. catani , g. thalhammer , m. inguscio , and f. minardi , phys .
rev . lett . * 103 * , 043201 ( 2009 ) . s. e. pollack , d. dries , and r. g. hulet , science * 326 * , 1683 ( 2009 ) .
s. nakajima , m. horikoshi , t. mukaiyama , p. naidon , and m. ueda , phys .
lett . * 105 * , 023201 ( 2010 ) .
n. gross , z. shotan , s. kokkelmans , and l. khaykovich , phys .
* 105 * , 103203 ( 2010 ) .
m. berninger , a. zenesini , b. huang , w. harm , h .- c .
ngerl , f. ferlaino , r. grimm , p. s. julienne , and j. m. hutson , phys .
* 107 * , 120401 ( 2011 ) .
r. j. wild , p. makotyn , j. m. pino , e. a. cornell , and d. s. jin , phys .
lett . * 108 * , 145305 ( 2012 ) .
r. s. bloom , m .-
hu , t. d. cumby , and d. s. jin , phys .
lett . * 111 * , 105301 ( 2013 ) .
t. lompe , t. b. ottenstein , f. serwane , k. viering , a. n. wenz , g. zrn , and s. jochim , phys .
* 105 * , 103201 ( 2010 ) .
s. knoop , f. ferlaino , m. mark , m. berninger , h. schbel , h .- c .
ngerl , and r. grimm , nat . phys . * 5 * , 227 ( 2009 ) . c. chin , r. grimm , p. julienne , and e. tiesinga , rev .
* 82 * , 1225 ( 2010 ) .
e. braaten and h .- w .
hammer , ann .
phys . * 322 * , 120 ( 2007 ) .
b. huang , l. a. sidorenkov , r. grimm , and j. m. hutson , phys .
lett . * 112 * , 190401 ( 2014 ) .
tung , k. jimenez - garcia , j. johansen , c. v. parker , and c. chin , arxiv : 1402.5943 .
r. pires , j. ulmanis , s. hfner , m. repp , a. arias , e. d. kuhnle , and m. weidemller , phys .
lett . * 112 * , 250404 ( 2014 ) .
o. machtey , d. a. kessler , and l. khaykovich , phys .
lett . * 108 * , 130403 ( 2012 ) .
j. schuster , a. marte , s. amtage , b. sang , g. rempe , and h. c. w. beijerinck , phys .
lett . * 87 * , 170404 ( 2001 ) . c. langmack , d. h. smith , and e. braaten , phys .
a * 86 * , 022718 ( 2012 ) ; * 87 * , 023620 ( 2013 ) . | the avalanche mechanism " has been used to relate efimov trimer states to certain enhanced atom loss features observed in ultracold atom gas experiments .
these atom loss features are argued to be a signature of resonant atom - molecule scattering that occurs when an efimov trimer is degenerate with the atom - molecule scattering threshold .
however , observation of these atom loss features has yet to be combined with the direct observation of atom - molecule resonant scattering for any particular atomic species .
in addition , recent monte - carlo simulations were unable to reproduce a narrow loss feature .
we experimentally search for enhanced atom loss features near an established scattering resonance between @xmath0k@xmath1rb feshbach molecules and @xmath1rb atoms .
our measurements of both the three - body recombination rate in a gas of @xmath0k and @xmath1rb atoms and the ratio of the number loss for the two species do not show any broad loss feature and are therefore inconsistent with theoretical predictions that use the avalanche mechanism . |
disclination lines @xcite are characteristic defects of uniaxial nematic phases .
these lines break the orientational order of the director field .
they can easily form , for instance by stirring a nematic drop deposited on a glass plate .
observation in the microscope reveals a _ threaded texture _ composed of thin and thick threads which are primarily @xmath0 and @xmath1 wedge disclination lines .
the former have singular cores whereas the latter form continuous configurations since the director can escape along the disclination axis ( the so - called _ escape in the third dimension _ @xcite ) .
these lines cost an energy per unit length of the order of k ( the relevant frank constant ) , so they tend to spontaneously collapse when the flow vanishes
. they can even all disappear at the end if the constraints at the boundaries and in the bulk are negligible ( no external field ) .
by contrast , some of them may subsist if the boundary conditions impose topological constraints which are incompatible with a continuous director field .
such a phenomenom occurs when the liquid crystal is confined inside a sphere with strong homeotropic anchoring ( nematic molecules anchored parallel to the local normal of the surface ) . in this case
, a point defect occurs at the center of the sphere ( hedgehog configuration ) @xcite .
this configuration is stable if the anchoring energy per unit surface area , @xmath2 , ( see below for a more detailed definition of this quantity ) is strong enough to prevent the homeotropic anchoring to be broken .
more precisely , a defect will appear if the anchoring penetration length @xmath3 @xcite is smaller than the sphere radius @xmath4 .
such configurations can be obtained by dispersing a nematic phase in a non miscible isotropic liquid @xcite .
they are also found in polymer dispersed liquid crystals ( pdlc ) @xcite .
these materials , of great practical interest , are obtained by a polymerization - induced phase separation with an average droplet radius typically ranging between 0.1 @xmath5 m and 10 @xmath5 m .
another way to nucleate a disclination line is to confine the nematic liquid crystal inside a capillary tube with a strong homeotropic anchoring .
if the tube is circular , the line is a + 1 disclination @xcite . on the other hand , two + 1/2
disclinations must occur near both sides of a flat rectangular capillary tube @xcite .
in all these examples the nematic phase forms a simply connected medium .
conversely , it is possible to disperse a non - miscible liquid into a nematic liquid crystal and form this way an emulsion .
this problem has recently attracted attention @xcite because the elastic distorsions produced by the point defects that nucleate near each droplet create long range dipolar forces between the droplets . in this article , we have chosen to study the nucleation and the physical properties of the -1/2 wedge disclination line that forms at the edge of a nematic sample confined between two parallel glass plates treated chemically to induce a homeotropic anchoring . at the edge ,
the nematic sample is in contact with an isotropic fluid .
the interface between the nematic phase and the surrounding isotropic fluid is generally curved across the sample thickness because the two phases wet the glass plates differently .
the first experimentally addressed example was that of the nematic interface in contact with its own melt .
such a situation is achieved in directional solidification when a homeotopic nematic sample is placed in a temperature gradient so that the nematic / isotropic phase transition temperature lies somewhere in the sample .
previous experiments in directional solidification have shown that a disclination line can detach from the interface in the form of triangular domains @xcite . to explain this observation ,
it was suggested that a disclination line lies behind the meniscus separating the two phases and that a dust particle crossing the line could detach it from the meniscus , leading to the observed triangles .
this interpretation was reinforced by the fact that energy considerations alone suggest this line can exist near the meniscus .
another reason to suppose its existence was that the wavelength measured at the onset of the mullins - sekerka instability was five times larger than the wavelength predicted by the linear stability analysis @xcite , suggesting a possible renormalization of the apparent surface tension due to the disclination line and to the associated elastic effects @xcite .
unfortunately the line was not visible in the microscope in the thin samples used in directional growth .
for this reason we tried to see the line by increasing the sample thickness ( its distance to the meniscus should essentially scale with the sample thickness ) .
surprisingly , even though we increased the sample thickness up to 400 @xmath5 m , we never saw the line and concluded that it might be virtual .
prompted by this failure , we attempted to observe a disclination line at a free nematic interface ( i.e. , in contact with air ) , and we discovered that the line may or may not exist in that system , depending on both the strength of the anchoring energy and , much more important , on the value of the preferred tilt angle of the molecules at the free interface .
this article is structured as follows : in section ( [ sec : experiment ] ) we detail the basic experimental setup and procedure , together with the main phenomenology involved in the formation of a disclination line near a free interface in a confined nematic phase . in section ( [ sec : model ] ) we outline a planar model for the director field in the presence of a disclination line which we use to do a numerical study . in section ( [ sec : model_3d ] ) a more realistic , three - dimensional structure of the director field is considered and we show it is a good approximation to use the planar model to calculate the position of the disclination line . finally , in section (
[ sec : anchoring_transition ] ) we consider the effect of the anchoring angle on the disclination line , and use our planar model to study a temperature - induced anchoring angle transition .
we have prepared thin nematic ( n ) liquid crystal samples confined between two parallel glass plates . in order to favor a homeotropic anchoring of the liquid crystal on the glass plates , thus inducing a homeotropic alignment of the liquid crystal in the bulk , we have treated the surface of the glass plates by chemically adsorbing a surface active layer on them @xcite .
prior to the chemical treatment , the glass plate is thoroughly cleaned by gently washing it with soap and water in order to remove any dirt spots , followed by soaking in soapy water for about 60 minutes , then rinsed in distilled water and quickly dried , first with clean air , and then baked under vacuum at 110@xmath6c for about 30 minutes .
we have tested the use of ctab ( hexadecyl trimethyl ammonium bromide ) and the use of a silane compound from merck ( zli-2510 ) to align the liquid crystal molecules . in order to treat the plates with ctab ,
we dip them into a 0.002% aqueous solution of the surface - active agent .
the plates are subsequently lifted slowly out of the solution in order to avoid precipitation of ctab . for the treatment with the silane compound
, we prepare a saturated solution of the surface active agent in toluene .
the glass plates are dipped in the solution for a few seconds and quickly lifted out to avoid creating dry silane spots .
the excess silane is removed from the glass plates by gently rubbing with lens cleaning tissue .
the cells are mounted with the desired gap between plates using either nylon thread ( for gaps in the range 100 @xmath5 m - 400 @xmath5 m ) or nickel wire ( for gaps smaller than 100 @xmath5 m ) as spacer .
a precision better that 5% in the cell gap is obtained this way .
the cells are partially filled by capillarity , keeping them at a temperature in the nematic range of the liquid crystal .
since the observations are to be made in the nematic phase , we make sure that the sample does not go into the isotropic phase once the cells are filled .
even though homeotropic alignment of the nematic phase is achieved faster if the product in the cell is first melted into the isotropic phase , we have observed in some cases that the motion of the n / air meniscus caused by the melting of the nematic phase crucially alters the observable phenomena .
sealing of the cell can also be performed to avoid degradation of the material or to prevent motion of the n / air meniscus .
different liquid crystal mesophases have been used in our experiments : 5cb ( k15 from merck ) , 8cb ( k24 from merck ) , mbba ( from acros ) , and i52 ( from merck ) .
samples with gaps in the range @xmath7 have been used .
when a disclination line is present , it appears equidistant from the two glass plate boundaries and parallel to the meniscus , suggesting a structure that is translationally invariant along the interface ( see fig .
[ fig : sample_configuration ] ) . when the sample is observed between crossed polarizers ( with the polarizers being parallel or perpendicular to the disclination line ) , the region between the disclination line and the homeotropic bulk reveals the presence of a distortion of the director field which indicates that it has a component in the direction of the disclination line ( see fig .
[ fig:3d - deffect ] ) .
the presence of three - dimensional defects on the disclination line , and the fact that the position of the line is altered in the vicinity of such defects suggest that the three - dimensional structure of the director field may play a significant role in the position of the disclination line .
as we will see below , however , a simple planar model for the disclination line ( in which the director field has no component along the line ) is able to reproduce both qualitatively and quantitatively our experimental observations on the line position ( see section [ sec : model ] ) .
a three - dimensional extension of the planar model shows that even though the component of the director field along the disclination line is not negligible , its effects on the position of the line are not very significant ( see section [ sec : model_3d ] ) . in order to measure the distance of the disclination line to the meniscus , we always select a region that is reasonably far from any such defects and that does not appear distorted by them . for gaps
@xmath8 m , both the limit of our optical resolution and the presence of the three - dimensional defects make our observations difficult .
we are able to observe the formation of the disclination line , but we are unable to meaningfully measure its distance to the meniscus .
therefore , we will restrict our measurements to samples with @xmath9 m .
we first studied the dependence of the position of the disclination line on the thickness of the sample .
the measurements are performed at a constant temperature in a region of the nematic phase where temperature variations of about @xmath10c have a negligible effect on the position of the disclination line .
typically , this is the situation when the temperature is more that @xmath10c above the nematic / smectic - a phase transition ( provided a smectic - a phase exists ) and more that @xmath10c below the nematic / isotropic phase transition . in fig .
[ fig : d_vs_b ] we plot the sample - thickness dependence of the position of the disclination line for 5cb .
our liquid crystal has @xmath11c , and we perform our measurements at @xmath12c ( @xmath13c ) . as we see in fig .
[ fig : d_vs_b ] , the distance of the disclination line to the meniscus scales linearly with the cell gap over the range of gaps we have explored . clearly , the shape of the meniscus , particularly the wetting angle with the glass plates , will have a strong influence on the position of the disclination line . assuming a constant radius of curvature ( circular profile ) , we are able to experimentally estimate the wetting angle on the plates ( see fig . [
fig : d_vs_b ] ) . given that the n / air surface energy ( @xmath14 jm@xmath15 ) is orders of magnitude larger than the nematic anchoring energy on the free interface ( @xmath16 jm@xmath15 ) @xcite , a shape with a circular profile , which both minimizes the surface area between the two media and meets the glass plates at a given contact angle , should be expected .
similar results are observed with 8cb , mbba , and i52 , with a combination of @xmath17 ratios and wetting contact angles @xmath18 as shown in table [ table : lines ] .
.ratio of distance of disclination line to n / air meniscus , @xmath19 , to the cell gap , @xmath20 , and wetting contact angle on the glass plates , @xmath18 , for several liquid crystals and surface treatments .
the wetting properties of the glass plates are noticeably different with the two different surface treatments we have used .
[ table : lines ] [ cols="<,^,^,^,^ " , ] it is customary to construct the dimensionless parameter @xmath21 , which compares the length scale imposed by the confined geometry ( in our case the cell gap , @xmath20 ) , with an intrinsic length scale of the material , @xmath3 .
note that throughout this article we will suppose that the anchoring energy on the glass plates ( which induces the homeotropic orientation of the bulk ) is large compared to @xmath2 , so that the presence of the free interface does not alter the anchoring condition on the plates .
the perturbation caused on the homeotropic director field configuration by the presence of the free interface will be less apparent for samples thinner than @xmath3 .
one expects the formation of a disclination line for @xmath22 , when surface anchoring effects dominate and impose a rigid anchoring angle .
if @xmath23 , however , elastic effects will dominate and a balance between elastic and surface anchoring torques , rather that a rigid anchoring angle , is expected to determine the anchoring condition at the meniscus . in this case
, the distortion caused by the free interface will be less important , and the equilibrium structure may be one without a disclination line .
the results reported in fig . [ fig : d_vs_b ] and table [ table : lines ] , namely , a disclination line whose position is completely determined by the geometry of the system , prove that the only relevant length scale is @xmath20 , the cell gap .
we are clearly in a regime with @xmath22 , or _ infinite _ anchoring energy , where the anchoring angle on the meniscus is rigidly set . if we were able to make clear observations with a small enough sample thickness ( typically @xmath24 m ) , we should presumably see the end of the @xmath25 regime , and observe the collapse of the disclination line . in the next section , we outline a model to relate the equilibrium position of the disclination line to the different experimental parameters .
our model predicts the disclination line to start collapsing noticeably for @xmath26 ( see fig .
[ fig : numeric_basic ] ) , which we may use to give a lower bound for the anchoring energy , @xmath2 .
we have confidently measured the position of the disclination line for @xmath27 m , and still observe the same @xmath17 ratio ( within error bars ) .
we can , thus , estimate @xmath28 ( @xmath29 for 5cb at @xmath30c @xcite ) .
a similar estimate is obtained from our measurements with 8cb . in an attempt to observe experimentally the collapse of the disclination line due to small values of @xmath21
, we have considered altering the value of @xmath2 by replacing the air interface with an interface with another isotropic medium , such as an isotropic fluid or glass wires with different surface treatments . even though we have observed an effect of the different interfaces on the anchoring conditions , the configurations had poor reproducibility and dubious stability .
we also tried to increase @xmath21 by approaching the nematic / smectic - a phase transition temperature .
indeed , we know from previous work that the bend elastic constant @xmath31 diverges at this transition @xcite so that the disclination line energy also diverges .
the experiment was performed by placing a @xmath32m - thick sample of 8cb inside a constant temperature gradient parallel to the liquid - crystal / air interface ( see fig . [
fig : neartna ] ) .
the immediate observation is that while in the n phase a disclination line is observed , it disappears as we enter the sma phase .
more precisely , @xmath17 shifts from its asymptotic value ( close to @xmath33 ) when @xmath34c and typically equals @xmath35 at @xmath36c . at this temperature @xmath37 while @xmath38 when @xmath39 ( see the following section ) .
this allows us to roughly estimate w in 8cb : @xmath40 .
in order to better relate the phenomena described in the previous section to the experimental parameters , we modeled the director - field configuration in a confined nematic geometry in the presence of a free interface .
even though our observations show a three - dimensional director - field structure around the disclination line ( see section [ sec : observations ] ) , we propose a simple , two - dimensional model with a planar disclination line ( the director field is assumed to be perpendicular to the line ) .
this model has allowed us to obtain numerical estimates for the position of the disclination line that match very well the experimental observations .
finally , we consider a more accurate model , where the three - dimensional structure of the director field is taken into account . we verify that , in fact , the predictions from the planar model quite accurately predict the position of the disclination line . if we consider that the distortion caused by the presence of the free interface is translationally invariant along the interface , and ignore the component of the director field along the disclination line ( planar disclination line ) , then a typical director field configuration will appear as shown in fig .
[ fig : sample_configuration ] , and it will be characterized by the angular field @xmath41 , with @xmath42 . given the two - dimensional nature of this approximation , there will be no contribution from the twist of the director field to the elastic free energy ( @xmath43 ) .
likewise , the saddle - splay term ( @xmath44 ) will vanish identically @xcite
. moreover , for the experimental conditions described in the previous section in which the samples are kept at a temperature away from both the n / i and n / sma transitions , the splay and bend elastic constants take on roughly the same values , @xmath45 .
a final approximation is made in which the mixed splay - bend term ( @xmath46 ) is neglected @xcite . with all this , we can write , @xmath47,\ ] ] which gives @xmath48,\ ] ] where @xmath49 .
the glass walls are taken to be at @xmath50 , and we assume that a strong anchoring fixes @xmath51 there .
the interaction with the n / air meniscus contributes a surface energy that is customarily approximated by @xmath52 with @xmath53 . here , @xmath54 is the angle between the local normal to the meniscus and the horizontal direction , and @xmath55 introduces a favored anchoring angle between the director field and the local normal to the interface for which the surface energy density is minimum ( see fig . [
fig : sample_configuration ] ) .
the energy introduced in the system by the presence of the n / air interface will be @xmath56 where the volume integral spans over the bulk nematic phase and the surface integral spans over the n / air meniscus .
we will seek the director - field distribution ( characterized by @xmath57 ) by minimizing the energy functional with respect to @xmath57 .
integrating by parts and imposing that the director field be homeotropic both on the glass boundaries and far from the interface , one obtains [ eq : configuration ] @xmath58 & , & \text { on the free interface.}\end{aligned}\ ] ] eq .
( [ eq : anchoring_bc ] ) establishes a torque balance at the interface ( here , @xmath59 represents the derivative along the local normal direction , and @xmath54 is the angle between the local normal to the interface ( @xmath60 ) and the horizontal direction ) .
the shape of the interface is completely determined by the experimentally measured wetting contact angle ( see fig . [
fig : sample_configuration ] ) .
these equations can be made dimensionless by rescaling all length scales by @xmath20 and defining the anchoring parameter @xmath21 , which contains all the material properties . with this
, one would expect the distance from the disclination line to the meniscus to have a functional dependence of the form @xmath61 . in the strong anchoring regime ( @xmath62 )
, @xmath63 will be rigidly set on the interface , and @xmath25 is expected .
note that there strictly exists in our model a discontinuity in the director field at the tip of the meniscus when @xmath64 is rigidly imposed at the interface .
this may lead to values of @xmath19 smaller than the real ones . the use of eq .
[ eq : anchoring_bc ] with a large , but finite , value of @xmath21 effectively removes this discontinuity and shows that the error introduced by the _
infinite _ anchoring approximation is very small . in order to find the configuration which minimizes the total energy , we proceed to solve eq .
[ eq : configuration ] by relaxing the fixed boundary problem obtained by setting the position of the disclination line at a distance @xmath19 from the meniscus .
therefore , @xmath65 is set for @xmath66 , while @xmath67 is set for @xmath68 . in order to avoid the @xmath69-discontinuity at @xmath70 ,
we restrict our calculation to the @xmath71 region .
the value of the wetting angle ( which can be determined experimentally , see fig .
[ fig : d_vs_b ] and table [ table : lines ] ) and the value of the favored anchoring angle , @xmath55 , together with the condition of homeotropic configuration both far from the meniscus and on the glass walls , completely determine the director field distribution .
typically , we use discretization grids of about 400x200 points .
once the configuration is obtained , we compute its total energy and then minimize this energy with respect to the configurations obtained by changing the value of @xmath19 .
note that we neglect the contribution of the core of the disclination to the total energy .
we assume that it will be constant , independent of @xmath19 and , therefore , will not be relevant in the determination of the configuration with the least energy . in fig . [
fig : numeric_basic ] , we show the results of these calculations .
first , we explore @xmath17 as a function of @xmath21 .
this corresponds to the measurements performed in section ( [ sec : d_vs_b ] ) .
we observe that the model predicts that @xmath17 converges asymptotically to @xmath72 as @xmath62 , for @xmath73 , and @xmath74 .
note that this is , within error bars , consistent with the results reported in table [ table : lines ] . only our measurements with 5cb ( shown in fig .
[ fig : d_vs_b ] ) are slightly smaller that the prediction from our model .
our numerical calculations predict as well that , as @xmath21 gets smaller , @xmath17 starts to decrease . in order for @xmath17 to decrease more than , say , 10% below the infinite anchoring position ( which would be experimentally measurable with our optical resolution )
one would need @xmath26 ( see fig . [
fig : numeric_basic ] ) . we have also explored the effect that @xmath18 and @xmath55 have on the disclination line .
we can see in fig .
[ fig : numeric_basic ] that @xmath17 increases with increasing values of @xmath18 . indeed ,
if we held @xmath17 fixed , a flatter meniscus ( larger @xmath18 ) would require larger distortions of the director field in order to match both the strong anchoring on the glass walls and on the meniscus .
this would increase the elastic energy , which can be prevented by increasing @xmath19 .
the experimental observations reported in table [ table : lines ] are consistent with this result , even though the dispersion in the data , combined with the fact that in the @xmath75 region @xmath17 is not very sensitive to changes in @xmath18 , do not allow for any conclusive trend to be extracted from our measurements . on the other hand
, we may exploit the modeled effects of the anchoring angle , @xmath55 , to study the known phenomenon of temperature - induced anchoring transitions ( see section [ sec : anchoring_transition ] ) .
we observe ( see fig . [
fig : numeric_basic ] ) that , within our numerical resolution , @xmath19 decreases roughly linearly with increasing values of @xmath55 . using this result , one can immediately correlate changes in the position of the disclination line with changes in the anchoring condition , provided @xmath2 remains unchanged .
observe that , extrapolating the calculations reported in fig .
[ fig : numeric_basic ] , we can expect that , even in the presence of a strong anchoring , a configuration without a disclination line may be observed , provided @xmath55 is large enough ( @xmath76 for the configuration in fig .
[ fig : numeric_basic ] ) .
this is the reason why a real disclination line is not formed spontaneously near a nematic / isotropic front , since on that interface large values for @xmath55 are reported ( @xmath77 for 8cb @xcite ) .
we recall that the anchoring length at this interface @xmath3 is too small ( of the order of @xmath78 m with @xmath79 jm@xmath15 @xcite and @xmath80 @xcite)to explain alone the absence of a real disclination line in thick samples ( @xmath81 m ) .
we also emphasize that the formation of planar triangles in directional solidification @xcite requires that a dust particle crosses the n - i interface , suggesting the disclination line is not real but virtual ( its core lies in the isotropic liquid ) .
this observation fully agrees with our predictions . as we have seen in section ( [ sec :
observations ] ) , our experimental observations reveal a three - dimensional structure in the director field in the vicinity of the disclination line .
the model described above assumes a planar configuration of the director field , in which the component along the disclination line is ignored . in this section ,
we study the modifications introduced by a three - dimensional model of the director field .
we have used the full landau - ginzburg - de gennes tensorial expansion for the elastic free energy density in power series of the quadrupolar order parameter ( which we define as @xmath82 , assuming a uniaxial system ) .
we take @xmath83 , which is an excellent approximation for 5cb and 8cb in the region of the nematic phase we study .
moreover , the saddle - splay term is included as well ( @xmath84 ) and the mixed splay - bend term is set to zero ( @xmath85)@xcite . we have only considered the case of rigid homeotropic anchoring , in which the liquid crystal molecules are forced to anchor on the meniscus at @xmath86 .
a detailed description of the model and the numerical methods involved are given in @xcite .
indeed , we obtain an equilibrium configuration where the director field has a significant projection in the direction of the disclination line , especially in the vicinity of the line ( see fig .
[ fig:3d - model ] ) .
surprisingly , though , when the position of the disclination line is estimated , the result is , within error bars , the same as the one found with the much simpler planar model .
therefore , the three - dimensional calculations validate the use of the planar model to predict the position of the disclination line as a function of the relevant physical parameters .
we have seen on section [ sec : model ] that , based on the results from our computations , we expect the disclination line to approach the meniscus as the anchoring angle is larger than zero and that , presumably , configurations without a disclination line might be possible , provided the anchoring angle is large enough .
it is well - known that the equilibrium anchoring angle of the nematic molecules at a n / air interface can depend on temperature @xcite .
we have performed measurements with mbba and i52 , both of which are known to have such a behavior , in order to study the effect the anchoring angle has on the formation of a disclination line .
samples are prepared as explained above and placed in a temperature - regulated oven and the position of the disclination line as a function of temperature is measured . for i52 ,
the disclination line starts to approach continously the n / air meniscus as we _ increase _ the temperature beyond @xmath87c , until it reaches a final configuration where the distance remains unchanged by further increasing the temperature .
no hysteresis is observed as we alternate heating and cooling cycles , and the transition temperature is independent of sample thickness .
in fact , measurements performed with different values of the cell gap are collapsed by rescaling the distance of the disclination line with the cell gap ( see fig . [
fig : d_over_b_vs_t ] ) .
this suggests that what we are seeing is not due to changes in the anchoring energy , since its influence in the position of the disclination line is introduced by the parameter @xmath21 , which depends on the sample thickness . for mbba ,
the distance from the disclination line to the n / air meniscus starts to decrease continuously as we _
lower _ the temperature below @xmath88c . just as with i52 , no hysteresis is observed here .
the transition temperature is slightly different for different samples , however , and it may depend on sample thickness , even though we have been unable to find any meaningful trend .
shifting the temperature axis with the value of @xmath89 for each experiment , and rescaling the position of the disclination line by @xmath20 effectively brings onto a single curve the data taken with different sample thicknesses ( see fig .
[ fig : d_over_b_vs_t ] ) suggesting as well that we are always in the _ infinite _ anchoring regime . both for i52 and for mbba
, we see that the departure from the upper plateau is well - described by a continuous ( or supercritical ) bifurcation of the form @xmath90 , which is analogous to what has been observed elsewhere for anchoring - angle transitions @xcite .
fitting such an analytical profile to our data , we can get a more precise estimate for the transition temperature , @xmath89 .
note that , even though a second stable position for the disclination line is observed for i52 , the line keeps approaching the meniscus as the mbba samples are cooled to lower temperatures , until the disclination line finally collapses . as we describe below , there is hysteresis in this collapse , and the dynamics that leads to it is reminiscent of a first - order transition . based on the observations reported above and on the results found in the literature , we would like to argue that the changes we see in the position of the disclination line when we vary the temperature of the samples can be accounted for by changes in the anchoring angle at the interface .
the existence of an anchoring angle transition is easily demonstrated by observing between crossed polarizers a micro - droplet of the liquid crystal placed on a substrate that induces a homeotropic anchoring ( see fig .
[ fig : i52_drop ] ) . for temperatures at which the disclination line is at the upper plateau in fig .
[ fig : d_over_b_vs_t ] , we have observed no point defects in the droplets . as we vary the temperature entering the region where a transition in the position of the disclination is observed in fig .
[ fig : d_over_b_vs_t ] , a characteristic point defect structure starts to become manifest , and the distortion in the director field increases in the temperature region where the disclination line is closest to the meniscus .
we have used our planar model for the structure of the director field near the n / air interface in order to relate the position of the disclination line to the anchoring angle ( see fig . [
fig : numeric_basic ] ) .
the numerical model considers a planar disclination line and a strong anchoring condition at the meniscus with an anchoring angle that can be different from the homeotropic orientation ( @xmath91 ) .
note that our numerical model assumes @xmath92 .
for mbba , in the relevant temperature range ( @xmath93c ) , one finds in the literature that @xmath94 .
the same information , however , is not available for i52 .
the fact that our planar model successfully predicts the position of the disclination line in the homeotropic anchoring situation is a good indication that the above approximation is valid .
moreover , in the vicinity of the continuous anchoring angle transition , the angular dependence of the surface energy may have its leading order term cancelled .
indeed , the surface energy in the vicinity of the second order anchoring angle transition can be expanded as @xmath95 where @xmath96 and @xmath97 are two positive constants .
the equilibrium angle is @xmath74 for @xmath98 and @xmath99 for @xmath100 ( mbba case ) .
the torque originated by an anchoring at an angle @xmath101 at @xmath102 is @xmath103 we see that , at @xmath104 , the leading order term vanishes .
we will assume that the first non - vanishing term , @xmath105 still increases steeply when @xmath106 , so that the anchoring angle remains close to its value at equilibrium .
the numerical calculations show that the dependence of the ratio @xmath17 on the anchoring angle can be roughly approximated by a straight line , whose slope depends on the wetting angle of the meniscus on the glass plates ( see fig . [
fig : numeric_basic ] ) . using our numerical results ,
we have been able to transform the @xmath17 vs. @xmath107 data in fig .
[ fig : d_over_b_vs_t ] into @xmath55 vs. @xmath107 ( see fig . [
fig : phi_vs_t ] ) . among other features ,
our @xmath55 vs. @xmath107 data maintains a continuous bifurcation compatible with @xmath108 , consistent with previous measurements by other groups .
our estimations for the dependence of the anchoring angle on temperature for mbba are well in agreement with all the results reported in the literature , both at the onset of the bifurcation @xcite and at much lower temperatures @xcite , extending almost @xmath109c below the transition .
we observe a homeotropic anchoring above @xmath89 and a steadily increasing anchoring angle whose saturation we are unable to observe as we enter a regime where the disclination line collapses .
we see that , provided the anchoring angle is large enough ( @xmath110 ) a configuration without a disclination line is possible . on the other hand , our estimates for i52
depart from the homeotropic anchoring configuration for @xmath111 above @xmath89 , with values for @xmath55 that are about 1.7 times lower than what was reported by sonin _
_ @xcite . for @xmath111 well above @xmath89 , we find a stable configuration with @xmath112 tilted from the homeotropic orientation .
the limited available information on the physical properties of i52 may account , at least partially , for these discrepancies , even though our only assumptions are strong anchoring and one - constant approximation for the elastic energy . 0.5 cm when the temperature is lowered below a thickness - dependent value ( typically 10 degrees below @xmath89 ) , regions where the disclination line has collapsed start to nucleate around dust particles on the meniscus ( see fig .
[ fig : hysteresys_line ] ) . these collapsed regions
propagate until they eventually extend along all the interface , and the equilibrium configuration has no disclination line . before collapsing
, the disclination line in the now metastable regions keeps approaching the meniscus as the temperature is lowered even further .
we have already explained that , as long as the disclination line does not collapse , there is no hysteresis between heating and cooling cycles .
once the line has collapsed , however , heating the sample does not regenerate the line continuously .
instead , there is a heterogeneous nucleation process similar to the one that leads to the collapse of the line upon cooling , and the temperature at which the line is regenerated is well above the temperature at which the line started to collapse . the diagram in fig .
[ fig : d_over_b_vs_t ] illustrates this hysteresis cycle .
this dynamics suggests the coexistence of two possible configurations in the structure of the director field for mbba , one presenting a disclination line and another with a continuous distortion of the director field .
the observed nucleation mechanism for the collapse and regeneration of the disclination line shows that the transition between these two configuration is energy activated .
a detailed study of this process would yield a good quantitative characterization of the energy associated with the core of the disclination line , which has yet to be done .
in this article , we have studied the nucleation and the physical properties of a -1/2 wedge disclination line which forms near the open edge of a nematic liquid crystal sample confined between two glass plates that induce a homeotropic orientation of the nematic phase .
the position of the disclination line , in particular its distance to the edge of the n / air meniscus is directly measurable , and we have related it to the physical properties of the liquid crystal with the help of a planar model for the structure of the director field in the presence of the n / air interface . since the three - dimensional nature of the structure of the director field in the vicinity of the disclination line is experimentally evidenced , we have used a three - dimensional model to obtain a more accurate description of this structure .
indeed , we have observed a significant _ escaped _ component of the director field , even though the position of the disclination line that minimizes the total energy agrees with the predictions of the planar model , which we have therefore used since it allows to implement realistic boundary conditions in a simple way .
we have performed different attempts at altering the position of the disclination line by changing the anchoring parameter @xmath21 . for large values of this parameter ,
the position of the disclination line is completely determined by the geometry of the problem , and therefore @xmath25 . by changing the sample thickness ,
we have tried to reduce the magnitude of @xmath21 so that the line aproaches the meniscus faster than the linear dependence on @xmath20 .
our attempts at observing this regime and extracting a value for @xmath2 this way have failed , most likely due to the experimental limitation in the smallest values for @xmath20 we have been able to use . under usual experimental conditions ,
the disclination line will form and its position will agree with the predictions from the _ infinite _ anchoring approximation . altering @xmath2 in a way that can affect
the position of the disclination line has proven to be a difficult task .
given the fact that @xmath2 is an intrinsic property of the interface , one would need to consider different isotropic media or even chemically altering the anchoring .
we have been unable to alter the nature of the anchoring strength in a stable and reproducible way . another promising way to control
the magnitude of @xmath21 is approaching the nematic / smectic - a phase transition temperature . the bend elastic
constant , @xmath31 , can be made arbitrarily large by approaching that temperature . by placing our samples in a temperature gradient ,
we have been able to observe the collapse of the disclination line , which we see to start approaching the n / air meniscus when the temperature is less than a few hundred millikelvin above @xmath113 . from this experiment
, we estimated @xmath2 in 8cb near @xmath113 .
nevertheless , this method has important experimental limitations ( presence of a vertical temperature gradient in the samples , existence of a zig - zag instability of the disclination line at very small temperature gradient , bad knowledge of the elastic constants near the transition , etc . ) .
finally , our calculations show that a departure of the anchoring condition at the n / air meniscus from the homeotropic orientation will significantly alter the position of the disclination line .
indeed , we have been able to observe the disclination line to approach the meniscus when materials known to have a temperature - induced anchoring angle transition are used .
moreover , we have shown that a configuration without a disclination line is obtained for values of the anchoring angle above @xmath114 ( a value that depends slightly on the material properties ) , even though energy considerations alone would suggest the possibility of the formation of a disclination line ( @xmath21 is large ) .
we believe this is the situation we encounter in the vicinity of a nematic / isotropic liquid crystal front in directional growth experiments .
even though we are in a regime of infinite anchoring ( large @xmath21 ) , a large value of the anchoring angle ( about @xmath115 ) guarantees that the disclination line will be virtual in those cases . | we have studied the nucleation and the physical properties of a -1/2 wedge disclination line near the free surface of a confined nematic liquid crystal .
the position of the disclination line has been related to the material parameters ( elastic constants , anchoring energy and favored anchoring angle of the molecules at the free surface ) .
the use of a planar model for the structure of the director field ( whose predictions have been contrasted to those of a fully three - dimensional model ) has allowed us to relate the experimentally observed position of the disclination line to the relevant properties of the liquid crystals .
in particular , we have been able to observe the collapse of the disclination line due to a temperature - induced anchoring angle transition , which has allowed us to rule out the presence of a real disclination line near the nematic / isotropic front in directional growth experiments . |
qcd is nowadays the main theoretical framework for understanding hadronic processes , but its non - abelian character makes low energy calculations unfeasible . the usual strategy for overcoming this difficulty consists in working with effective theories , constructed in such a way as to include , as much as possible , the main features of qcd . as in nuclear physics
most processes involve only the quarks @xmath1 and @xmath2 , one requires these theories to be poincar invariant and to possess approximate @xmath3 chiral symmetry .
the breaking of the latter , due to the quark masses , is parametrized by the small pion mass @xmath4 at the effective level . in the last twenty years
, effective theories incorporating chiral symmetry have been successfully applied to hadronic interactions , including or not electro - weak probes . as far as the hadronic sector is concerned it
is useful to distinguish two classes of processes , involving only mesons or both mesons and baryons . in the case of purely pionic systems , effective lagrangians
are treated relativistically and yield well defined power counting procedures , in both pion mass and momenta @xcite .
this gives meaning to the idea of chiral perturbation theory .
when nucleons are present , things become more complicated @xcite and calculations were performed in the simplified framework of the so called heavy baryon chiral perturbation theory ( hbchpt ) , in which nucleons are treated non - relativistically @xcite .
only recently a well defined power counting scheme was proposed for the relativistic approach @xcite . in the case of two - nucleon systems , interactions involve processes with a marked spatial hierarchy , whose dynamical complexity increases rapidly when one moves inward . in the best semi - phenomenological potentials existing at present , that can reproduce low - energy observables accurately ,
the interaction is determined by the undisputed one pion exchange , supplemented by a theoretical two - pion exchange potential ( tpep ) and parametrized at short distances @xcite .
the tpep is the locus of chiral symmetry in this problem , since it is closely related to the pion - nucleon ( @xmath5n ) amplitude @xcite .
after the works of weinberg in the early nineties restating the role of chiral symmetry in nuclear interactions @xcite , the interest in applications was renewed and several authors have tackled the problem of constructing the tpep . initially , non - linear lagrangians containing only pions and nucleons were employed @xcite .
however , these lagrangians do not suffice to describe experimental @xmath6 data @xcite and the corresponding potential misses even the scalar - isoscalar medium range nn attraction . in a later stage ,
other degrees of freedom were introduced , coherence with @xmath6 information was restored and descriptions compatible with asymptotic nn scattering data were produced @xcite .
it is possible to distinguish two partially overlapping theoretical frameworks in these chiral calculations .
one of them , adopted in refs.@xcite and @xcite , is based on hbchpt . in this theory
, a rule exists for counting powers of the typical three - momenta exchanged between nucleons , assumed to be of the order of the pion mass .
this allows the construction of non - relativistic effective lagrangians , that include unknown couterterms and are used to derive amplitudes in which contributions from loops and these counterterms add up coherently .
a rather puzzling aspect of these counting schemes , in the case of nn interactions , is that they predict a leading contribution in the @xmath7 expansion with the following spin - isospin structure @xcite : @xmath8 , where @xmath9 , @xmath10 and @xmath11 are the usual isospin , spin and tensor operators .
this means that , in hbchpt , the scalar - isoscalar central potential , which is by far the most important phenomenological component of the tpep , corresponds to just a subleading effect .
the other approach to the tpep is typically relativistic @xcite and emphasis is given to the tail of the interaction , which is determined by the well known @xmath5n amplitude .
no attention is paid to counterterms in the effective lagrangian , that correspond to zero - range interactions and represent processes such as multimeson or quark exchanges .
tail and conterterm contributions do not overlap in space and hence can be treated separately in nn potentials . for systems containing just a single nucleon ,
relativistic and hbchpt calculations were compared and found to be consistent , provided the dimensional regularization scale is set equal to the nucleon mass @xcite . in this work
we repeat this kind of comparison , for the leading contribution to the two - pion exchange @xmath0 potential , and show that relativistic and hbchpt predictions fail to agree by 30% . the reason for this important discrepancy
may be traced back to the fact that the intermediate @xmath6 amplitude can not be represented by a series around the point that determines the potential at large distances , as shown by becher and leutwyler @xcite .
the construction of the tpep was discussed in detail in ref .
@xcite and here we just sketch the main steps . the potential is based on the on - shell nn scattering amplitude containing two intermediate pions , from which one subtracts the iterated opep , in order to avoid double counting .
the isoscalar component , represented by @xmath12 , is given by @xmath13^{(1)}\;[t^+]^{(2 ) } } { [ k^2-\mu^2]\;[k^{\prime\,2}-\mu^2]}-(opep)^2 \ ; , \label{1}\ ] ] where @xmath14 is the isospin symmetric @xmath6 scattering amplitude for pions with momenta @xmath15 and @xmath16 , @xmath17 , and the factor @xmath18 accounts for the symmetry under the exchange of the intermediate pions @xcite .
in the sequence we represent initial and final nucleon momenta by @xmath19 and @xmath20 , their mass by @xmath21 and also use the variables @xmath22 , @xmath23 , @xmath24 and @xmath25 .
in general , the amplitude @xmath14 can depend on four independent variables , @xmath26 , @xmath27 , @xmath28 and @xmath29 .
as the exchanged pions are off shell , one should , in principle , keep @xmath28 and @xmath29 unconstrained everywhere . however , when these factors appear in the numerator of amplitudes , we may write @xmath30 or @xmath31 and use the terms within parentheses to cancel pion propagators in eq.([1 ] ) .
this kind of cancellation is associated with short range terms , which do not contribute to the asymptotic potential .
therefore , whenever possible , we neglect them and replace @xmath28 and @xmath29 by @xmath32 in the numerators of our expressions . concerning the variables @xmath26 and @xmath27 , the conditions of integration in eq.([1 ] ) are such that the main contributions come from the unphysical region @xmath33 , @xmath34 @xcite .
in particular , the vicinity of the point @xmath35 determines the potential at very large distances . in order to understand the structure of the tpep , it is convenient to isolate in @xmath14 a term @xmath36 , involving only chiral pion - nucleon interactions at tree level and written as @xmath37 \not\!q \right\ } u \ ; , \label{2}\ ] ] where @xmath38 is the @xmath6 coupling constant .
the remainder is represented by @xmath39 and encompasses other degrees of freedom , either in the form of low energy constants or by means of specific models , that may include the @xmath40 and other states .
one writes symbolically @xmath41=[t_n^+ ] + [ t_r^+]$ ] for each nucleon and the potential is then proportional to @xmath41^{(1)}[t^+]^{(2)}=[t_n^+]^{(1)}[t_n^+]^{(2 ) } + \ { [ t_n^+]^{(1)}[t_r^+]^{(2 ) } + [ t_r^+]^{(1)}[t_n^+]^{(2)}\ } + [ t_r^+]^{(1)}[t_r^+]^{(2)}$ ] .
this product involves box , crossed box , triangle and bubble diagrams , that should be taken into account in a complete calculation . as pointed out by gross @xcite long ago , the @xmath42^{(1)}[t_n^+]^{(2)}$ ] contribution is very small , due to chiral cancellations .
the detailed numerical study of this structure performed in ref.@xcite has confirmed this conclusion and shown that the crossed tem , within curly brackets , is largely dominant .
it is due to a @xmath39 which may be expressed as @xmath43 where the @xmath44 are subthreshold coefficients @xcite . in this multipolar expansion
, @xmath45 represents the fact that a nucleon can be deformed when interacting with pions and is associated with the so called axial polarizability parameter , given by @xmath46 .
this allows the asymptotic central potential to be written as @xmath47^{(1 ) } \;\left[\frac{\sigma_n(t)}{\mu^2}\;\bar{u}\;u\right]^{(2)}+ \left(1 \leftrightarrow 2\right)\ ; , \label{4}\ ] ] where @xmath48 is the nucleon contribution to the scalar form factor.$ ] , which tends to the present one for very long distances . ]
quite generally , this form factor is defined in terms of the symmetry breaking lagrangian @xmath49 as @xmath50 . as discussed by gasser , sainio and varc @xcite , @xmath48
is associated with the diagram of fig.[fig.1]a and given by @xmath51 \ ; , \label{5}\end{aligned}\ ] ] where the loop integrals @xmath52 are written as @xmath53\;[(q\!+\!\q/2)^2\!-\!\mu^2]}\ ; , \label{6}\\[2 mm ] & & j_{c , sn}^{(1)}(t ) = \frac{(4\pi)^2}{i}\;\frac{1}{v^2 } \nonumber\\[2 mm ] & & \times \int\frac{d^4q}{(2\pi)^4}\ ; \frac{2mv\!\cdot\ ! q}{[(q\!-\!\q/2)^2\!-\!\mu^2]\;[(q\!+\!\q/2)^2\!-\!\mu^2]\ ; [ q^2\!+\!2mv\!\cdot\ ! q\!-\!\q^2/4]}\;. \label{7}\end{aligned}\ ] ] the integral given by eq.([6 ] ) is divergent and has to be regularized .
however , this procedure affects only the zero - range properties of the potential and is irrelevant for the present discussion , which concentrates on asymptotic properties .
therefore , in the sequence , we consider only differences between functions calculated at generic values of @xmath27 and at @xmath54 , denoted by a caret .
the non - relativistic potential is obtained by going to the center of mass frame , where @xmath55 , and making @xmath56 \rightarrow1 $ ] in eq.([4 ] ) .
one then has @xmath57 \frac{m}{\pi \mu } \left [ \hat{j}_{c , c}(m ; \mbox{\boldmath $ \q$ } ) - \hat{j}_{c , sn}^{(1)}(m ; \mbox{\boldmath $ \q$ } ) \right ] \right\ } \ ; , \label{8}\ ] ] where we now indicate explicitly the dependence of the potential on the nucleon mass
. the minus sign in front of this expression was introduced in order to account for our convention of the relativistic t - matrix .
it is worth noting that this contribution was already included in the general formulation of ref.@xcite .
we define the `` small '' dimensionless quantities @xmath58 , @xmath59 , and use the standard techniques for loop integration in order to write @xmath60 \nonumber\\[2 mm ] & = & 2- \sqrt{1 + 4\ ; \alpha^2/\kappa^2}\ ; \ln\left [ \frac { \sqrt{1 + 4 \;\alpha^2/\kappa^2}+1 } { \sqrt{1 + 4 \;\alpha^2/\kappa^2}-1}\right ] \ ; , \label{9}\\[2 mm ] \hat{j}_{c , sn}^{(1)}(m ; \mbox{\boldmath $ \q$ } ) & = & 2 \int_0 ^ 1 da \;(1\!-\!a)^2 \int_0 ^ 1 db\ ; ( 1\!-\!b ) \nonumber\\[2 mm ] & \times & \left\ { \left [ a(1\!-\!a)b\ ; \kappa^2 + ( a+b - ab)\ ; \alpha^2 + ( 1\!-\!a)^2 ( 1\!-\!b)^2 \right]^{-1 } \right .
\nonumber\\[2 mm ] & - & \left.\left [ ( a+b - ab)\ ; \alpha^2 + ( 1\!-\!a)^2 ( 1\!-\!b)^2\right]^{-1 } \right\ } \nonumber\\[2 mm ] & = & \hat{j}_{c , c}(m ; \mbox{\boldmath $ \q$ } ) - \left(\frac{\pi\mu}{m}\right)\ ; \hat{i}_n(m ; \mbox{\boldmath $ \q$ } ) \ ; , \label{10}\end{aligned}\ ] ] where the integral @xmath61 is given by @xmath62 & & \times \left.\left [ \tan^{-1}\left ( \frac{(1\!-\!a ) - ( \alpha^2+a\;\kappa^2)/2}{\sqrt{\alpha^2+a(1\!-\!a)\;\kappa^2 -(\alpha^2+a\;\kappa^2)^2 /4 } } \right ) \right.\right .
\nonumber\\[2 mm ] & & + \left.\left .
\tan^{-1}\left ( \frac { ( \alpha^2+a\;\kappa^2)/2}{\sqrt{\alpha^2+a(1\!-\!a)\;\kappa^2 -(\alpha^2+a\;\kappa^2)^2/4}}\right ) \right ] \right .
\nonumber\\[2 mm ] & & - \left.\frac{\alpha^2}{\sqrt{\alpha^2 - \alpha^4 /4 } } \left [ \tan^{-1}\left ( \frac{(1\!-\!a ) - \alpha^2/2}{\sqrt{\alpha^2 - \alpha^4/4 } } \right ) + \tan^{-1}\left ( \frac { \alpha^2/2}{\sqrt{\alpha^2 - \alpha^2/4}}\right ) \right ] \right\ } \;.
\label{11}\end{aligned}\ ] ] the function @xmath48 and the asymptotic central potential can then be written simply as @xmath63 \ ;
\sigma_n(t ) \nonumber\\[2 mm ] & = & - \;\frac{2}{f_\pi^2\mu^2}\ ;
\left [ f_\pi^2 \left ( a_{00}^+ -\mbox{\boldmath $ \q$}^2 \;a_{01}^+ \right ) \right ] \left [ \frac{3 \mu^3}{32\pi}\left ( \frac{g}{m}\right)^2 \hat{i}_n(m ; \mbox{\boldmath $ \q$})\right ] \;. \label{12}\end{aligned}\ ] ] the first result is very general and independent of specific models or approximation schemes used to calculate the scalar form factor .
it has a rather transparent physical interpretation : the factor proportional to @xmath48 represents , as one will show in the sequence , the pion cloud of one of the nucleons , which is extended in space ; the term involving the subthreshold coefficients in the square bracket represents the other nucleon , acting as a scalar source expanded in multipoles ; the factor 2 arises because the potential is due to the crosed term in @xmath64 and the scale @xmath65 corresponds to the energy density of the quark condensate .
@xmath66 & @xmath67 & @xmath68 + @xmath69 & @xmath70 & @xmath71 & @xmath72 & 0.72 & -1.89 & -1.20 + @xmath73 & @xmath74 & @xmath75 & @xmath76 & 0.12 & 0.94 & 1.12 + @xmath77 ( mev ) & @xmath78&@xmath79 & @xmath80 & & & + [ tab.1 ] empirical values of the subthreshold coefficients can be obtained by means of dispersion relations and some results available in the litterature are given in table [ tab.1 ] .
as pointed out long ago@xcite , they may be well represented by a model including the delta and supplemented by information about the scalar form factor . in order to use this information
, one writes @xmath81 , where @xmath82 is the @xmath6 @xmath82-term .
the parameter @xmath83 , the slope at the origin , may also be obtained by means of dispersion relations .
for instance , a recent analysis @xcite , based on @xmath84 mev , has produced @xmath85 mev , which corresponds to @xmath86 .
the scalar form factor is then related to the subthreshold coefficients by @xmath87 and @xmath88 , where @xmath89 is the pion decay constant .
as far as the delta is concerned , one uses the results of ref.@xcite and obtains @xmath90 a_{01}^+)_\delta & = & \;g_\delta^2 \left(\frac{4}{9}\right)\ ; \frac{(m+m_\delta)}{m_\delta^2-m^2-\mu^2}\ ; \left\ { \left ( 1-\;\frac{\mu^2}{m_\delta^2}\right ) \right .
\nonumber\\[2 mm ] & + & \left .
\frac{\mu^2}{2 m_\delta^2 ( m_\delta^2-m^2-\mu^2 ) } \left [ m_\delta(3 m_\delta - m ) - \mu^2 \frac{(3 m_\delta+2m)}{(m_\delta+m)}\right ] \right\}\ ; , \label{14}\end{aligned}\ ] ] where @xmath91 is the delta mass , @xmath92 is the pion - nucleon - delta coupling constant and we have used @xmath93 .
the constant @xmath92 can be related to @xmath38 by means of the large @xmath94 result @xmath95 @xcite and the leading delta contributions may be written as @xmath96 a_{01}^+)_\delta & = & \frac{1}{2}\ ; \left(\frac{g}{m}\right)^2 \ ; \frac{1}{m_\delta - m}\;. \label{16}\end{aligned}\ ] ] adopting @xmath97 and @xmath98 mev , one has the values displayed in table 1 .
most of the dependence of the potential on the momentum transferred is embodied on the dimensionless function @xmath61 , which is also influenced by the value of the nucleon mass . in the heavy baryon ( hb )
limit , defined as @xmath99 , this function can be evaluated analytically and one has @xcite
@xmath100 \nonumber\\[2 mm ] & = & - \left [ \frac{\mu^2+\mbox{\boldmath $ \q$}^2/2}{\mu\ ; |\mbox{\boldmath $ \q$}|}\;\tan^{-1}\frac{|\mbox{\boldmath $ \q$}|}{2 \mu } -\frac{1}{2 } \right ] \;. \label{17}\end{aligned}\ ] ] the expression for the heavy baryon potential in momentum space is then @xmath101\ ; , \label{18}\ ] ] where we have used the goldberger - treiman relation @xmath102 .
this result reproduces , up to irrelevant polynomial terms , the potential derived in ref.@xcite using hbchpt , when one identifies the subthreshold coefficients and the low energy constants @xmath103 as follows : @xmath104 and @xmath105 .
it is worth pointing out that the numerical values recommended in that work yield @xmath106 and @xmath107 , which deviate from those quoted in table 1 .
this happens because the constants @xmath103 include both leading terms and first order corrections@xcite .
the @xmath103 then take the latter into the potential and give rise to fourth order corrections to otherwise third order expressions .
when these corrections are eliminated for the sake of a consistent power counting , one recovers values compatible with table 1 .
heavy baryon chiral perturbation theory was also used in other calculations @xcite , based on time ordered perturbation theory . in order to allow comparison with our results ,
we perform cauchy integrations in eqs.([6 ] ) and ( [ 7 ] ) and obtain @xmath108 & & \left .
-\ ; \frac{2m}{v^2}\left [ \frac{(mv^0+e_q)v^0+\mbox{\boldmath $ v$}\!\cdot\!\mbox{\boldmath $ q$ } } { e_q [ ( mv^0+e_q)^2\!-\!\omega_+^2 ] [ ( mv^0+e_q)^2\!-\!\omega_-^2 ] } \right.\right . \nonumber\\[2 mm ] & & \left.\left.+ \frac{1}{(\omega_+^2-\omega_-^2 ) } \left ( \frac{\omega_+v^0+\mbox{\boldmath $ v$}\!\cdot\!\mbox{\boldmath $ q$}}{\omega_+\;[(mv^0-\omega_+)^2-e_q^2 ] } - \frac{\omega_-v^0+\mbox{\boldmath $ v$}\!\cdot\!\mbox{\boldmath $ q$}}{\omega_-\;[(mv^0-\omega_-)^2-e_q^2]}\right ) \right]\right
. \nonumber\\[2 mm ] & & \left .
-\left(\omega_+\rightarrow\sqrt{\mu^2+\mbox{\boldmath $ q$}^2},\;\ ; \omega_- \rightarrow\sqrt{\mu^2+\mbox{\boldmath $ q$}^2}\right ) \right\ } \ ; , \label{19}\end{aligned}\ ] ] where @xmath109 and @xmath110 .
going to the limit @xmath99 , one has @xmath111
\label{20}\ ] ] and a potential given by @xmath112 \;. \label{21}\ ] ] , eq.([11 ] ) , on the nucleon mass , for @xmath113 ( continuous line ) , @xmath114 ( dot - dashed line ) , @xmath115 ( dashed line ) , @xmath116 ( dotted line ) , where @xmath117 is the empirical nucleon mass.,width=566 ] , eq.([11 ] ) and @xmath118 , eq.([17 ] ) , on the nucleon mass ; conventions as in fig.[fig.2].,width=566 ] in order to check the consistency of this result , we have integrated analytically eq.([20 ] ) and recovered eq.([17 ] ) exactly . in fig.[fig.2 ]
we display the behaviour of @xmath61 for @xmath119 and @xmath120 , where @xmath121 mev is the empirical nucleon mass .
the ratios @xmath122 for the same values of @xmath21 are shown in fig.[fig.3 ] , where it is possible to see that typical deviations from the heavy baryon limit are respectively about 30% , 10% , 2% and less than @xmath123 .
thus , one learns that the heavy baryon limit is about 30% far from the real world . as we discuss in the next section ,
this is related to the fact that the heavy baryon expansions of loop integrals sometimes fail to converge @xcite .
the potential in coordinate space is given by @xmath124\ ; \sigma_n(m ; \mbox{\boldmath $ r$ } ) \ ; , \label{22}\ ] ] where @xmath125 is the form factor in configuration space .
it is convenient to write it in terms of a dimensionless function @xmath126 as @xmath127 with @xmath128 in the heavy baryon limit this expression can be evaluated explicitly @xcite and , using @xmath129 , one writes @xmath130 ^ 2 \ ; , \label{25}\\[2 mm ] \mbox{\boldmath $ \nabla$}_{\!x}^2 \hat{f}_{hb}(\mbox{\boldmath $ r$})&= & 4 \left ( 1+\frac{3}{x}+\frac{11}{2\;x^2}+\frac{6}{x^3}+\frac{3}{x^4}\right ) \frac{e^{-2x}}{x^2 } \;.
\label{26}\end{aligned}\ ] ] for generic values of @xmath21 , it is more convenient to perform the fourier transform using directly eqs.([9 ] ) and ( [ 10 ] ) and one has @xcite @xmath131 \ ; , \label{27}\ ] ] where @xmath132 \ ; , \label{28}\\[2 mm ] \eta^2 & = & \left [ a+b - ab + ( 1\!-\!a)^2 ( 1\!-\!b)^2 \;m^2/ \mu^2 \right ] \;\lambda^2 \ , .
\label{29}\end{aligned}\ ] ] the behaviour of the ratio @xmath133 as function of @xmath21 is displayed in fig.[fig.4 ] , where one finds the same qualitative pattern as in figs.[fig.2 ] and [ fig.3 ] . in configuration space ,
however , the gap between the heavy baryon limit and results with the empirical value of the nucleon mass varies with distance and is about 25% asymptotically .
the difference between the sizes of the gaps in momentum and configuration spaces can be understood by noting that the former may contain pure polynomials , which correspond to zero - range functions and hence are not present in the latter .
, eq.([27 ] ) and @xmath134 , eq.([25 ] ) , on the nucleon mass ; conventions as in fig.[fig.2].,width=566 ]
the behaviour of the triangle integral that determines the scalar form factor has been extensively discussed in a recent work by becher and leutwyler @xcite .
the leading term in their chiral expansion of @xmath135 can be obtained by neglecting terms proportional to @xmath136 in eq.([27 ] ) and one has @xmath137 this result is equivalent to the dispersive representation of ref.@xcite , given by @xmath138 with @xmath139 \;.
\label{32}\ ] ] in order to study the structure of this integral , we rewrite it as @xmath140 for large values of @xmath141 , the exponential in the integrand falls off very rapidly as a function of @xmath142 and the integral tends to be dominated by a small region around @xmath35 . on the other hand ,
the function @xmath143 vanishes at this point and hence the net result is due to a compromise between these two effects . in the work of becher and leutwyler
one learns that it is possible to write accurately @xmath144_{\!hb } -\left [ \frac{\alpha(t-2\mu^2)}{2\sqrt{t\!-\!4\mu^2}}\right]_{\!a } + \left [ \frac{\alpha\sqrt{t}}{2\sqrt{t\!-\!4\mu^2 } } - \frac{\sqrt{t}}{2\mu } \tan^{-1}\frac{\alpha\mu}{\sqrt{t\!-\!4\mu^2}}\right]_{\!b}\ ; , \label{34}\ ] ] where @xmath145 .
the first two terms in this expression correspond to the standard heavy baryon series whereas the last one implements the correct analytic behaviour around the point @xmath35 .
in particular , the @xmath146 term gives rise to eq.([25 ] ) and has been used in hbchpt calculations of the @xmath0 potential .
( dashed line ) , @xmath147 ( dotted line ) and @xmath148 ( continuous line ) , obtained by using eq.([34 ] ) into eq.([31 ] ) , by @xmath149 , eqs.([30]-[32]).,width=566 ] in fig.[fig.5 ] we display the ratios of the functions @xmath150 , @xmath147 and @xmath148 , obtained by using eq.([34 ] ) into eq.([31 ] ) , by @xmath149 .
the first thing to be noted is that there is a gap between @xmath150 and @xmath149 , which is about @xmath151 in the region of physical interest and can be as large as @xmath152 asymptotically .
the inclusion of @xmath153 brings the approximate result much closer to the exact one , indicating that the last term in eq.([34 ] ) is important mainly at rather large distances .
thus , in principle , hbchpt could allow a precise numerical description of the @xmath0 potential , but this would require the extension of existing calculations to higher orders .
this illustates the advantage of the relativistic framework proposed recently by becher and leutwyler , namely that it gives rise automatically to leading contributions which are already rather accurate .
the leading chiral contribution to the central potential is obtained by using eqs.([31 ] ) and ( [ 32 ] ) into eqs.([23 ] ) and ( [ 22 ] ) . in order to assess the phenomenological implications of this result
, one could evaluate phase shifts and compare them with experiment .
however , one should bear in mind that , in contributing to phase shifts , the central tpep is superimposed to both isospin dependent components and a large opep background .
its effects are completely blurred in the @xmath154 wave , but may be identified in the waves @xmath155 and @xmath156 . in this situation ,
it becomes possible for different chiral models to accomodate experimental data@xcite .
in fact , in the case of the tpep , phase shifts do not represent the most stringent test possible . ,
eqs.([22 ] , [ 23 ] ) amd ( [ 30]-[32 ] ) , by the central medium - range argonne potential @xcite , for the values of the subthreshold coefficients of refs .
@xcite ( continuous line ) , @xcite ( dashed line ) and @xcite ( dotted line).,width=566 ] therefore , in this work we compare the predictions of the asymptotic @xmath157 with the argonne phenomenological potential@xcite , which is rather accurate and constrained by a large data base , that also includes information about the @xmath154 channel . in fig.[fig.6 ] we plot the ratio of the leading chiral potential by the corresponding argonne component , in the range 1@xmath1584fm , for the sets of subthreshold coefficients considered in table 1 .
typical discrepancies are of the order of 20% , indicating that the very simple mechanism described in the previous section is indeed responsible for the leading contribution to the central potential .
the existing gap has the order of magnitude expected from chiral corrections .
it is worth pointing that , in the case of @xmath0 interactions , these corrections would affect not only the scalar form factor , but also require the evaluation of other amplitudes . in particular , in the case of the delta contributions , one has also to include box and crossed box diagrams .
a comprehensive discussion of this problem will therefore be postponed .
we begin by considering first order corrections to the leading result for @xmath135 , which come from two different sources .
one of them corresponds to the terms neglected when passing from eq.([27 ] ) to eq.([30 ] ) , whereas the other one is implemented by subthreshold coefficients , representing the remainder @xmath159 discussed in the previous section .
thus , the scalar form factor , including both leading and first order corrections , is given by @xmath160 where @xmath161 is the result of ref.@xcite , given by eq.([29 ] ) and @xmath162 is written in terms of the subthreshold coefficients @xmath163 as@xcite @xmath164 \right\}\ ; \hat{f}_{cc } ( \mbox{\boldmath $ r$ } ) \ ; , \label{36}\ ] ] with @xmath165 the main contributions to the coefficients entering this result come from the @xmath82-term and the @xmath40 and one may also write @xmath166 the term @xmath167 is given by the diagram of fig.[fig.1]b . using the results of ref.@xcite
one finds , after neglecting terms @xmath168 @xmath169 where @xmath170 is obtained from the @xmath171 of eq.([30 ] ) by replacing @xmath172 with @xmath173\ ; \lambda^2 \;. \label{40}\ ] ] eq.([36 ] - continuous line ) , @xmath167 eq.([39 ] - dashed line ) and @xmath174 eq.([38 ] - dotted line ) by the leading chiral contribution @xmath175 eqs.([30]-[32 ] ) and ( [ 23]).,width=566 ] as the function @xmath176 determines the effective mass of the two - pion system exchanged between the nucleon and the scalar source , this result indicates that @xmath167 has shorter range than @xmath161 . in fig.[fig.7 ] , we display the ratios of @xmath162 , @xmath167 and @xmath174 by the leading chiral contribution @xmath175 and one notes that these corrections remain quite visible at relatively large distances .
the fourier transform of the scalar form factor corresponds to a mass density , that surrounds the nucleon and is due , in part , to the pion cloud .
as the density of the quark condensate is negative , the sign of the scalar density indicates that the pions , as goldstone bosons , destroy the condensate in order to exist . in this picture
, the energy density of the scalar cloud can not exceed that of the condensate . in fig.[fig.8 ] , we plot the function @xmath177 as function of the distance , and it is possible to see that this critical situation occurs close to 0.6 fm . on the other hand , the value 1/100 , for which the pion field may be considere as weak , occurs around 1.5 fm , suggesting that this is the order of magnitude of the distances beyond which perturbative calculations may be considered as reliable .
eq.([23 ] - dashed line ) , @xmath167 eq.([39 ] - dotted line ) and @xmath178 ( continuous line).,width=566 ] in order to explore this picture further , we recall that the nucleon @xmath82-term is related to @xmath179 by @xmath180 the evaluation of this integral , using @xmath181 and replacing this function by @xmath182 whenever @xmath183 , yields @xmath184 mev , quite close to the value quoted in ref.@xcite . this somewhat surprising result is associated with the factor @xmath185 in eq.([34 ] ) , that suppresses the integrand at small distances , and seems to indicate that other mechanisms contribute little to scalar dynamics .
another interesting aspect of expression ( [ 39 ] ) is that it allows one to study the behaviour of the scalar form factor in the large @xmath94 limit , where the nucleon and the delta become both heavy and degenerate . in that limit one has @xmath186 and hence , @xmath187 ^ 2 \;. \label{42}\ ] ]
this expression is identical with that obtained in the framework of the skyrme model , where @xmath188 and , for large distances , one has@xcite @xmath189 this result , when combined with the fact that the model accounts well for the opep , shows that skyrmions do have asymptotic properties consistent with precise perturbative calculations in the large @xmath94 limit . on the other hand , in the perturbative approach
, the departure from this limit is very important , suggesting that some of the quantitative limitations of the skyrme model may be associated with the intrinsic @xmath190-@xmath40 degeneracy .
in this work we studied the central @xmath0 potential at large distances and have emphasized the special role played by the nucleon scalar form factor , which describes the mass density of its pion cloud .
the effect of this cloud over the other nucleon , represented as a multipole expansion , dominates the asymptotic interaction .
this rather simple picture accounts for about 75% of the potential , as given by the precise phenomenological fit by the argonne group .
we have evaluated this leading contribution , due to the exchange of two pions , and determined its dependence on the nucleon mass , using the relativistic formalism .
our results coincide with those of heavy baryon chiral perturbation theory when the nucleon mass is taken to infinity , but become about 20% smaller when the empirical mass is used .
this shows that it is important to keep the nucleon mass finite in precise descriptions of the @xmath0 system . in a recent work , becher and leutwyler
have shown that the scalar form factor is determined by an intermediate @xmath6 amplitude , close to the threshold at @xmath35 , which lies almost on top of a branch point . as a consequence
, the heavy baryon expansion does not make sense for the scalar form factor .
we have investigated the implications of this result for the @xmath0 interaction and found out that threshold effects can be seen at very large distances only .
therefore , in principle , hbchpt could account numerically for the discrepancy between the heavy baryon limit and reality found in the central potential , but this would require the extension of existing calculations to higher orders .
this stresses the advantage of the relativistic framework , namely that it does not deal with truncated expressions and hence gives rise automatically to leading contributions which are already rather accurate .
a fully covariant description of the @xmath0 interaction , including other degrees of freedom , will be presented elsewhere .
w.n.cottingham and r.vinh mau , phys.rev . * 130 * , 735 ( 1963 ) ; w.n.cottingham , m.lacombe , b.loiseau , j.m.richard and r.vinh mau , phys.rev .
d * 8 * , 800 ( 1973 ) ; m. lacombe , b. loiseau , j. m. richard , r. vinh mau , j. cot , p. pires and r. de tourreil , phys . rev .
c * 21 * , 861 ( 1980 ) . c. ordez and u. van kolck , phys .
lett . * b291 * , 459 ( 1992 ) ; l.s.celenza , a.pantziris and c.m.shakin , phys.rev.*c46 * , 2213 ( 1992 ) ; j.l.friar and s.a.coon , phys.rev . * c49 * , 1272 ( 1994 ) ;
c.a.da rocha and m.r.robilotta , phys.rev . * c49 * , 1818 ( 1994 ) ; m.c.birse , phys.rev . * c49*,2212 ( 1994 ) . | the central two - pion exchange @xmath0 potential at large distances is studied in the framework of relativistic chiral symmetry and related directly to the nucleon scalar form factor , which describes the mass density of its pion cloud .
this relationship is well supported by phenomenology and allows the dependence of the asymptotic potential on the nucleon mass to be assessed .
results in the heavy baryon limit are about 25% larger than those corresponding to the empirical nucleon mass in the region of physical interest .
this indicates that it is very important to keep this mass finite in precise descriptions of the @xmath0 system and supports the efficacy of the relativistic chiral framework .
one also estimates the contribution of subleading effects and presents a simple discussions of the role of the quark condensate in this problem .
m.r .
robilotta _ _ |
the total number of models pretending to explain the origin of the knee exceeds twenty and the number of people who propose these models is about thirty .
the number of ideas is definitely larger , but i deliberately distinguish _ ideas _ and _ models _ , because sometimes the ideas are not developed at all to be considered as a model .
the quality of the development is also quite different for different models . to the great extent
it is due to the lack of astrophysical or particle physics data which could constrain the models , but there are also some evident subjective reasons . certainly it is our duty to examine as much consequences of the proposed idea as we can and to some extent it is our fault that we ignore some evident experimental or observational facts which contradict the proposed idea .
there is a tradition in our p.n.lebedev physical institute that if you propose the idea you should also propose the test which can * kill * your idea .
we must admit that there is not a lot of authors who propose such tests .
in this talk i examine some features which should be explained by models of the knee and requirements which should be satisfied .
usually we consider just three main characteristics of cosmic rays ( cr ) which should be predicted by the model and checked by the comparison with experimental data - they are _ the energy spectrum , the mass composition and the anisotropy_. among them there might be some phenomenological sub - features , which are sometimes crucially important .
quite a good survey of models with their regard to the observed features of the energy spectrum and the mass composition is given in @xcite . here
i focus only on the problems related to the cr energy spectrum .
usually we examine such features of the spectrum as the absolute intensity and slopes below and above the knee .
i should like to draw the attention to some other sub - features of the spectrum shape viz .
its negligible _ irregularity _ , then _ the sharpness and the fine structure _ of the knee . since the discovery of extensive air showers ( eas ) it has been found that their size spectrum has the power law character in wide energy ranges below and above the knee .
the theory of shock acceleration explained this power law and since then the educated guess was that it is supernova explosions which create the shocks , which in turn accelerate particles and produce a regular power law of the cr energy spectrum .
however , we should remark that a stochastic character of supernova explosions in space and time prevents the formation of the regular spectrum .
the subsequent propagation of cr in the interstellar medium ( ism ) only emphasizes the irregularity since it results in the dominant contribution of nearby and recent supernova remnants ( snr ) to cr observed at earth .
the accurate monte - carlo realization of the scenario , where only galactic snr give birth to observed cr , shows the extremely irregular character of the spectrum with the very low probability to observe the good power law ( figure 1a ) .
we consider the irregularity as a very important phenomenological feature and propose as the measure of the irregularity the parameter @xmath0 , which is the standard deviation from the mean intensity taken in a logarithmic scale . as an example
, figure 1b shows the irregularity as the function of primary energy for the scenario of standard supernova exploding sporadically in our galaxy and producing different spectra , shown in figure 1a .
it is seen that due to the decreasing lifetime of higher energy cr in the galaxy the smaller number of recent snr contributes to the intensity and the irregularity of the spectra increases .
the remarkable sharpness of the knee has been mentioned even in the first publication by kulikov and khristiansen in 1958 @xcite .
they have written : _ the presented data indicate with a great probability the possible * sharp * change of the spectrum in the studied range of shower sizes @xmath1_. in @xcite we suggested to use as the measure of sharpness the parameter @xmath2 , where @xmath3 is the shower intensity and @xmath4 is the eas size .
the advantage of this second derivative is that it depends _ neither _ on the absolute intensity _ nor _ on the slopes of the spectrum below and above the knee .
the world survey of 40 spectra @xcite showed that inspite of the big spread of sharpness values nearly all of them exceed the value 0.3 expected for the original galactic modulation model ( figure 2 ) .
this important phenomenological characteristic should be taken into account in every reasonable model of the knee and never ignored .
fluctuations in the eas development smear out the irregularity of the primary cr energy spectrum , but not down the regular power law with @xmath5 . due to the stochastic nature of the cr sources there
* must * be structure_at some level_. the well known examples of such structures are the widely discussed _ knee _ and _
the ankle_. the search for other structures is difficult since the deviations from the power law are evdently small .
one must have a good statistics to find the structure in the single experiment .
to use the data accumulated in many experiments one has to eliminate their difference in the background and calibration .
we satisfied the first requirement using _ the deviation from the running mean_. to satisfy the second requirement we referred the position of the deviation to the position of the knee in each individual experiment .
the position of the knee has been determined as the point of the maximum sharpness in the spectrum .
this approach is absolutely legitimate if you study the shape of the spectrum .
we processed 40 eas size spectra and 5 cherenkov light spectra available at that time and the result is shown in figure 3 .
besides the apparent excess at the knee there is a second excess ( peak ) at @xmath6 seen both in the cherenkov light and in the eas electron size spectrum .
we consider this result as the strong evidence for the fine structure of the knee , which should be taken into account by any reasonable model of the knee .
the origin of this peak will be discussed later .
however it is clear that such structure is nearly impossible to get in the classic galactic modulation model with its smooth leakage of cr from the galaxy .
all models of the knee can be tentatively divided into three groups : _ astrophysical _ models , _ interaction _ models and _ phenomenological _ models .
the astrophysical models in turn can be subdivided into _ source _ models and _ propagation _ models .
the models of the first group explain the origin of the knee by astrophysical reasons : either by the change or the evolution of cr sources , or by the change of the propagation and the leakage from the galaxy .
the interaction models claim that there is no sharp change of the primary cr energy spectrum .
they refer the steepening of the observed eas size spectrum and the change of their characteristics to the change of the particle interactions at pev energies .
the phenomenological models do nt specify new sources or new processes above the knee , but try to explain the steepening by the existing features of the primary cr or an eas development .
sometimes the association of the model with any group is ambiguous , and this classification is given just for the orientation .
below i ll give some examples of the models , which were discussed in recent years .
certainly the list of these examples is not complete and i ask those authors who will not be mentioned to apologize me for my ignorance .
the most developed conception is that sources of cr below the knee are supernova ( sn ) explosions .
the shock waves produced by the explosions can accelerate protons up to maximum energies @xmath7 0.1 - 0.4 pev .
the knee is due to the impossibility for sn to accelerate particles up to higher energies .
this cut - off at @xmath8 is rather sharp and by this way one can explain the visible sharpness of the knee .
however the immediate questions to these models are : + ( i ) are sn standard enough to give about the same @xmath8 to give the sharp knee ? + ( ii ) what is the origin of cr above the knee ?
different source models give different answers to these questions .
i should like to speak about our single source model later and here just mention the model of p.l.biermann @xcite .
he allows that there might be just a few ordinary sn remnants ( snr ) which give dominant contribution at the knee and ensure the formation of the sharp knee .
he even assumes that their explosions due to special conditions created during the collapse can be standard @xcite .
the cosmic rays above the knee are produced by another type of sn , which progenitors are more massive , hot and emit powerful particle wind before the explosion .
the typical representative of such type of sources are well known wolf - raye stars .
the powerful shock wave propagating through the bubble created by the hot wind accelerates particle up to eev energies .
similar idea is developed by l.g.sveshnikova @xcite , who examined the variety of sn types and concluded that the dominant contribution to cr below the knee is given by just sni@xmath9 type and above the knee - by explosions of more massive and hot stars which she called _ ` hypernovae ' _ though this name is often associated with sn explosions accompanied by gamma - ray bursts ( grb ) .
the interesting model of the cosmic ray acceleration by blobs of relativistic plasma is proposed by a.dar , r.plaga and a.de rujula @xcite .
its main subject is the origin of gamma ray bursts , but there are important points in this scenario which have relevance to our discussion .
the sources of cosmic rays in this model are cannonballs - relativistic plasmoids which are emitted by the core collapsed supernovae and stopped in the galactic halo or in the intergalactic medium ( igm ) . by this way
authors avoid difficulties connected with the need to introduce different sources below and above the knee - the sources are the same and the knee is connected with the increasing number of reflections from the cannonball , needed to accelerate particles to higher energies . since the model of the knee is not sufficiently developed it is difficult to say whether it explains the sharpness of the knee or not . as for the irregularity the problem seems not to be severe since cr sources can be even more numerous than snr since each of them can emit a few cannonballs .
there is no problem with the isotropy and the absence or presence of the gzk cutoff , since the sources are mainly distributed in the galactic halo @xcite or in the igm @xcite , but not in the galactic disk .
there are other models which claim that there is no need to introduce another kind of sources beyond the knee .
snr are able to accelerate particles up to hundreds pev due to compression and magnification of magnetic fields inside the shell @xcite
. however these strong fields decrease with time as soon as the snr expands .
snr turned out to accelerate particles to higher energies during the shorter time at the very beginning of their history .
it gives rise to the steepening of the energy spectrum and imitating the knee @xcite .
there are so called compact source models which are the combination of the source and interaction models .
they examine the possibility for protons emitted by the source to lose the energy in photopion reactions and for nuclei - to disintegrate after the interaction with intense radiation fields around these compact sources @xcite .
though the knee formed by these process is sharper than in the propagation models considered below ( s @xmath10 0.35 instead of 0.3 ) it is not sufficient to explain the observations .
moreover these models predict the extremely light mass composition beyond the knee , which contradicts the observations .
i think all the mentioned source models have problems when compared with the shower phenomenology .
models which require different sources below and above the knee usually propose more energetic and less abundant sources of higher energy cr above the knee .
they inevitably meet problems with the irregularity .
the small number of high energy sources or shorter time of the acceleration to higher energies increase the sensitivity of the cosmic ray energy spectrum to the particular space - time distribution of the sources and reduce the possibility to form the regular power - law spectrum .
source models need the additional mechanism ( halo or igm ) which could smooth the irregularity of the cr spectrum below and above the knee . on the other hand
the models with snr accelerating particles up to sub - eev and eev energies give just the smooth steepening of the cr energy spectrum and are not able to explain the sharpness of the knee .
propagation models were the first suggested after the discovery of the knee @xcite .
they associate the steeper spectrum above the knee with the increased leakage of cr particles from the galaxy when their gyroradius becomes comparable with the transverse size of the galactic disk .
latest developments of this idea included models with introduction of the hall diffusion @xcite and an anomalous diffusion in the ism of quasi - fractal type @xcite .
all they have problems with the sharpness of the observed spectrum .
original models had the sharpness of the constituent spectra not greater than 0.6 . for the mixed mass composition with the abundance of constituent nuclei equal to their abundance at tev energies the model sharpness decreased down to 0.3 . in more sophisticated models with the hall diffusion and a special geometry of regular magnetic field in the galactic disk and halo
the authors @xcite created the small intensity bump in the knee region and by this way increased the sharpness of constituent spectra up to 1.0 .
the composite spectrum with a mixed composition will also have a greater sharpness but no more than 0.7 .
it is much less than the required value of about 3 .
the second problem is the observed fine structure of the spectrum .
if the propagation models predict the smooth constituent spectra with a continuous steepening , it is impossible to construct a total spectrum with discontinuities which are seen as the fine structure .
the interaction models introduce new processes in the interaction of cr particles at pev energies to justify the sharp steepening of the eas size spectrum .
some of these models require the new physics either to create new heavy particles with large cross - sections , which decay into yet unobservable high energy muons and neutrinos @xcite , some required emission of technihadrons , gravitons @xcite or very many low energy particles @xcite .
the model with an enhanced production of charmed particles @xcite slows down the development of the atmospheric cascade so that the bulk of its energy is dissipated not in the atmosphere but reaches the mountain or even the sea level and is absorbed in the earth .
the experimental confirmation of neutrino oscillations and therefore an existence of the neutrino mass inspired the creation of the models where cosmic ray particles interact with relic neutrinos in the gravitationally attracted neutrino halos surrounding galaxies and galaxy clusters @xcite .
by assuming the neutrino mass in the region of ev it is easy to obtain the energy threshold for the inelastic @xmath11p interaction in the pev region . in my mind
the drawback of these models is not only that most of them require the new physics in the energy region very close to that already studied in fermilab , but that they often do nt specify the new process and do nt develop their interaction model to incorporate and test it with well developed simulation codes such as corsika , like it has been done for more conservative models ( qgsjet , sybill etc . ) .
these models can explain the existence of the knee , sometimes even its sharpness by introducing a rapid rise of the cross - section above the threshold , but never other phenomenology of eas accumulated hitherto . in this respect
i should better attribute nearly all of them to ideas and not to models .
phenomenological models of the knee just extrapolate the observed phenomenological characteristics of cr or eas into the high energy region and demonstrate that with certain approximations they give the steepening of the energy spectrum which they associate with the knee . to such type of models
i attribute the model of tsallis et al @xcite .
they applied the formulae of generalized non - extensive statistics to the observed cr energy spectrum and have found that it is possible to fit its behaviour with the bend at sub - gev and the steepening at pev energies in the total range of 34 decades of the differential intensity .
the knee in this model appears as a crossover between two fractal thermal regimes . to the phenomenological class of models
belongs also the multi - component model of biermann and ter - antonyan @xcite .
they obtain parameters of the constituent nuclei spectra by multivariate fitting of the electron and muon size spectra measured by kascade and maket ani experiments .
another phenomenological model is proposed by stenkin @xcite , who connects the steepening of the eas size spectrum with the non - linearity of the @xmath12 dependence which in turn is connected with the change of the eas structure as soon as the shower maximum approaches the observation level . if @xmath13 where @xmath14 is the decreasing function of @xmath15 then even for the power law primary energy spectrum @xmath16 with a constant @xmath17 the slope index @xmath18 of the eas size spectrum @xmath19 increases with @xmath20 and the spectrum steepens imitating the knee even for the constant slope of the primary spectrum .
i would attribute to phenomenological models also the so - called _ poly - gonato model _ of j.r.hrandel @xcite .
he noticed that spectra of constituent nuclei in primary cr become flatter with an increasing primary mass .
he extrapolates these spectra up to their rigidity cutoff and found that it is possible to describe not only the knee but the general behaviour of the primary energy spectrum up to sub - eev energies . at these high energies
the substantial fraction of cosmic rays can be trans - iron elements .
since these models are purely phenomenological and not pretend to explain the origin of cosmic rays they have no problems with the spectrum irregularities .
however they may have problems with the observed fine structure of the spectrum and definitely with the sharpness of the spectrum if they will not assume the sharp cutoff of the constituent spectra .
it would be unfair if i ll tell nothing about the present status of the single source model ( ssm ) .
me and prof.a.w.wolfendale proposed it in 1997 @xcite and updated in 2001 @xcite .
the main idea was that the knee is due to the contribution of just one single , recent and nearby snr .
since it is nearby and recent , cr from this snr have a flat energy spectrum with a sharp cutoff @xcite . at pev energies
their intensity becomes comparable with the total cr background from all other distant and old snr and their contribution creates the bump which forms the knee . due to the sharp cutoff of the spectrum beyong its maximum rigidity @xmath21
the knee is sharp and different constituent nuclei which have different maximum energy @xmath22 give rise to the fine structure of the spectrum .
actually the energy spectrum of cr created by the single source is the result of the fit of theoretical spectra @xcite to the existing experimental data from 40 independent eas electron size spectra , 9 muon size spectra , 3 hadron size and energy spectra and 5 cherenkov light spectra . comparing this spectrum with the spectrum of cr expected from snr of different ages and placed at different distances from the earth we estimated the most likely intervals of distances and ages of the single source @xcite .
they appeared to be rather narrow around 300 pc and 100 kyear . soon after that american astronomers using the triangulation technique have found that the pulsar b0656 + 14 associated with the snr monogem ring is situated exactly at the indicated range of distances and ages as required for our single source @xcite . in @xcite
we presented arguments why the nearby snr is not seen in gamma rays .
it is because being nearby it is situated in our local superbubble with its low density of ism which is about @xmath23 @xmath24 .
moreover it is not a discrete , but an extended source with the angular radius of 20@xmath25 and the problem of the correct subtraction of the background from such a big area is not trivial .
many experiments have already looked for an excessive intensity of cr from this region and found nothing @xcite .
nevertheless the situation is controversial , since recently the moscow university @xcite and tien - shan @xcite groups revealed weak eas excesses in the monogem ring area .
if there are indeed cr excesses at sub - pev and pev energies from the monogem ring region they can evidence for the activity of the pulsar b0656 + 14 which emitted high energy cr soon after its birth and which were confined within the snr for a long time of about 80 kyear @xcite .
high energy gamma - quanta which create these excesses are produced by cr on the local fluctuations of the interstellar gas density .
the energy spectum of cr produced by the pulsar b0656 + 14 has the sharp peak at the rigidity of 0.25 pv which is very close to the @xmath26 pv , required by the sn model of berezhko et al @xcite and can contribute to the observed knee sharpness , but its contribution to the cr intensity at the knee does not exceed 15% .
however , this pulsar can be responsible for cr beyond the knee at energies above 10 pev .
recently we analysed the region beyond the knee and have found an extended excess of eas intensity in the broad region slightly above @xmath27 gev @xcite ( figure 4 ) .
the existence of this excess gives an argument in favor of the assumption that the pulsar b0656 + 14 contributes to the cr intensity above @xmath28 gev . so that there is an impression that the snr monogem ring with its associated pulsar b0656 + 14 can be a serious contender for the single source responsible for the pev cr intensity bump and the knee itself .
it is not easy to kill the ssm , since its parameters were taken to fit the existing phenomenology .
it is possible , however , to check and to kill the initial mass composition predicted by ssm .
originally we attributed intensity peaks at 3 and 12 pev to @xmath29 and @xmath30 .
if we are right then we can not expect another peaks beyond 12 pev .
however if the observed peaks are actually due to @xmath31 and @xmath29 , then we can expect the existence of the sharp @xmath30 peak at 40 pev .
so far there is no indication of this peak in the experimental data from kascade .
we look forward at the new data from kascade - grande .
in the summary i stress again that the models which pretend to explain the origin of the knee should explain not only the high isotropy , the tendency of the primary mass composition to become heavier above the knee , but also such phenomenological features as the _ sharpness _ , the _ fine structure _ of the knee and the relatively _ small irregularity _ of the energy spectrum both _ far below _ and _ far above _ the knee . at the moment none of the models
can satisfy these requirements .
source models which require less abundant and more energetic sources have problems with the large irregularity of the spectrum in particular above the knee .
the same difficulty have the models with relatively abundant sources like pulsars which emit high energy cr for a short time after their creation .
definitely we have to incorporate the object or the mechanism which level out irregularities originated from the stochastic space - time distributions of sources .
the first candidates for such an object could be the galactic halo or igm .
all propagation models have not the sufficient sharpness of the knee even for the constituent spectra . as the consequence
they should have problems with the observed fine structure of the spectrum .
they take for granted the regular power law spectra with no explanation of the origin of this regularity , therefore they can also have hidden problems with the irregularity of the spectrum .
the phenomenological models do nt pretend to explain the origin of cr and the knee - they use different input spectra and find the best fit to the observed phenomenology of the showers .
the sharpness , the fine structure depend on the shape of these input spectra so that there is no actual explanation of the observed phenomenology . the interaction models , besides that they require the new physics , are not sufficiently developed to be tested by the phenomenology .
it is clear that modifying the threshold behaviour of the new process it is possible to fit both the sharpness and the fine structure of the spectrum , but the value of this exercise is not very high . since the main subject of interaction models is the knee they do not explain the small irregularity of the cr spectrum outside the knee region .
this table shows that the _ astrophysical _ models are the most developed and some of them can match the observed phenomenology . among astrophysical models _ source _ models are the most plausible . for nearly all models we need to introduce a mechanism which could smear out the irregularity of the energy spectrum due to the stochastic nature of the sources , viz .
galactic halo , intergalactic medium or so .
51 j. r. hrandel , _ astropart .
_ , * 21 * , 241 ( 2004 ) . g. v. kulikov and g. b. khristiansen , _ jetp _ , * 35 * , 635 ( 1958 ) .
a. d. erlykin and a. w. wolfendale , _ j. phys .
g : nucl . part .
_ , * 23 * , 979 ( 1997 )
. a. d. erlykin and a. w. wolfendale , _ j. phys .
g : nucl . part .
_ , * 27 * , 1005 ( 2001 ) .
p. l. biermann , _ new ast .
_ , * 48 * , 41 ( 2004 ) ; astro - ph/0309810
. p. l. biermann et al . , astro - ph/0302201 ( 2003 ) .
l. g. sveshnikova , _ a&a _ , * 409 * , 799 ( 2003 ) .
a. dar , astro - ph/9809163 ( 1998 ) .
a. dar and r. plaga , astro - ph/9902138 ( 1999 ) .
r. plaga , _ new astronomy _ ,
* 7 * , 317 ( 2002 ) .
a. de rujula , astro - ph/0207033 ( 2002 ) .
a. dar , astro - ph/0408310 ( 2004 ) . a. de rujula , this conference .
a. r. bell and s. g. lucek , _ mnras _ , * 321 * , 433 ( 2001 ) .
drury et al .
, astro - ph/0309820 ( 2003 ) . v. s. ptuskin and v. n. zirakashvili , 2003 , a&a , * 403 * , 1 ; astro - ph/0408025 .
w. tkaczyk _ 27th icrc , hamburg _ , * 5 * , 1979 ( 2001 ) .
j. candia et al .
, _ astropart .
_ , * 17 * , 23 ( 2002 ) .
b. peters , _ 6th icrc , moscow _ , * 3 * , 157 ( 1959 ) . v. s. ptuskin et al . , _ a&a _ , * 1 * , 1 ( 1993 )
. n. n. kalmykov and a. i. pavlov , _
26th icrc , salt lake city _ , * 4 * , 263 ( 1999 ) .
e. roulet , astro - ph/0310367 .
a. a. lagutin and v. v. uchaikin , _
27th icrc , hamburg _ , * 5 * , 1900 ( 2001 ) .
a. a. petrukhin , _
27th icrc , hamburg _ , * 1 * , 1 ( 2001 ) ; _ phys . atomic nucl .
_ , * 66 * , 517 ( 2003 ) .
m. t. dova et al . , astro - ph/0112191 ( 2001 ) .
d. kazanas and a. nikolaidis , _
27th icrc , hamburg _ , * 5 * , 1760 ( 2201 ) ; astro - ph/0103147 , hep - ph/0109247 ( 2001 ) . s. i. nikolsky , _ nucl .
b ( proc .
_ , * 39a * , 228 ( 1995 ) ; _ 27th icrc , hamburg _ , * 4 * , 1389 ( 2001 ) . v. i. yakovlev , _ 25th icrc , rome _ , * 1 * , 446 ( 1995 ) .
r. wigmans , _ astropart .
_ , * 19 * , 379 ( 2003 ) . c. tsallis et al . , _ phys
. lett .
a _ , * 310 * , 372 ( 2003 ) .
p. l. biermann and s. v. ter - antonyan , _ 28th icrc , tsukuba _ , * 1 * , 235 ( 2003 ) .
v. stenkin , _ 28th icrc , tsukuba _ , * 1 * , 267 ( 2003 ) ; _ mod .
lett . a _ , * 18 * , 1225 ( 2003 ) . j. r. hrandel , _ astropart .
_ , * 19 * , 193 ( 2003 )
. e. g. berezhko et al . , _ jetp _ , * 82 * , 1 ( 1996 ) .
a. d. erlykin and a. w. wolfendale , _ j. phys .
g : nucl . part .
_ , * 29 * , 709 ( 2003 ) . s. thorsett et al . , _ apj lett .
_ , * 592 * , l71 ( 2003 ) .
d. e. alexandreas et al .
_ apj _ , * 383 * , l53 ( 1991 ) .
t. mckay et al .
_ apj _ , * 417 * , 742 ( 1993 ) . l. sun and s. sun _ 25th icrc , durban _ , * 4 * , 165 ( 1997 )
. k. wang et al .
_ apj _ , * 558 * , 477 ( 2001 ) .
f. a. aharonian et al .
_ a&a _ , * 390 * , 39 ( 2002 ) .
s. w. cui et al .
28th icrc , tsukuba , * 4/7 * , 2315 ( 2003 ) . t. antoni et al .
_ apj _ , * 604 * , 687 ( 2004 ) . t. antoni et al . , astro - ph/0402656 ( 2004 ) .
r. atkins et al .
, astro - ph/0403097 ( 2004 ) .
a. a. chilingaryan et al . , this conference .
g. v. kulikov and m. yu .
zotov , 28th rcrc , moscow , 2004 ; astro - ph/0407138 .
d. benko et al . , 28th rcrc , moscow , 2004 ( to be published ) .
a. d. erlykin and a. w. wolfendale , _ astropart .
_ , * 22/1 * , 47 ( 2004 ) .
a. d. erlykin and a. w. wolfendale ( in preparation ) .
fomin et al . , _ 28th icrc , tsukuba _ , * 1 * , 119 ( 2003 ) . | it is evident that models of the knee should match the observational phenomenology . in this talk
i discuss a few aspects of phenomenology , which are important not only for the understanding of the knee origin , but also for the general problem of the origin of cosmic rays . among them
are the shape of the energy spectrum , its irregularity , the sharpness of the knee and its fine structure .
the classification of models is given and some examples of the most recent models are discussed .
the most probable conclusion deduced from this examination is that the knee has an astrophysical origin and the so called source models of the knee are most likely among them . |
the transverse degree of freedom of an optical field is the fundamental aspect of light that contains spatial information .
utilization of this information is the basic resource in traditional imaging systems and in applications such as microscopy , lithography , holography or metrology .
in addition , use of the transverse modes of light has recently been demonstrated to be an important resource in optical communication @xcite , high - dimensional entanglement studies @xcite , and quantum key distribution @xcite .
having control of the spatial coherence properties of a light beam provides an additional degree of control compared to using fully coherent light only , and has been shown to be advantageous for a number of applications .
beams of decreased coherence allow access to spatial frequencies that are twice those available in a purely coherent system @xcite .
greater spatial frequencies can enable improvements in imaging based systems and has been shown to be particularly useful in lithography @xcite .
partial coherence also allows for the suppression of unwanted coherent effects by decreasing the coherence , such as suppression of speckle @xcite which enables lower noise and opens the door to novel imaging modalities @xcite .
it has also been suggested that partial coherence can improve the deleterious effects of optical propagation through random or turbulent media @xcite .
in addition , the coherent property of optical beams can be used for novel beam shaping @xcite as well as a method for control over soliton formation due to modulation instabilities in the study of nonlinear beam dynamics @xcite . the ability to generate arbitrary optical beams
could also be used as a tool in basic research , such as in optical propagation @xcite or testing of novel methods in quantum state tomography dealing with the transverse wavefunction of light that has seen a great deal of interest recently @xcite .
traditional methods used to generate partially coherent beams of light often rely on imprinting a changing pattern of random phase or speckle onto a coherent beam , such as with a spatial light modulator ( slm ) @xcite or rotating diffuser @xcite .
it has even been demonstrated that slms allow the statistics of the speckle patterns to be varied across the beam to give spatially varying coherence properties @xcite .
however none of these methods have been shown to allow for complete arbitrary control over the spatial coherence of an optical beam . in this paper
we demonstrate how to generate any arbitrary quasi - monochromatic partially coherent field that can be specified by a cross - spectral density function @xmath1 , i.e. for fields fully specified by their two point spatial correlations .
this is done by first computing the coherent mode decomposition of @xmath1 , which is an incoherent mixture of orthogonal coherent modes . for each of these coherent modes a computer generated hologram ( cgh )
is computed for a digital micromirror device ( dmd ) that acts as a binary amplitude spatial light modulator with rapid modulation speeds .
the dmd then switches between each coherent mode on timescales slower than the coherence time of the source laser , but long relative to the detection time of the ccd .
this creates an incoherent averaging that physically reproduces the coherent mode decomposition .
section [ sec : coherentmodedecomposition ] details computation of the coherent mode decomposition , section [ sec : binarygratings ] describes the algorithm to compute binary amplitude cghs for the generation of coherent modes and section [ sec : experiment ] details the experimental demonstration of this technique .
the transverse wavefront of a deterministic and coherent scalar beam is described by a complex field , @xmath2 . for a stochastic beam
, @xmath2 is a random variable and it becomes necessary to represent the field in a more sophisticated way .
the standard way of doing this is with the cross - spectral density function . at a single frequency
the cross - spectral density function is defined as @xmath3 and represents the average intensity ( @xmath4 ) , as well as the correlations ( up to second order ) of such a partially coherent field @xcite .
@xmath0 can be decomposed into an incoherent sum of orthogonal spatial modes @xmath5 , written as @xmath6 where @xmath7 are real and nonnegative , and @xmath8 is the relative weight of the field in mode @xmath5 @xcite .
the modes @xmath5 can be computed as the eigenfunctions with corresponding eigenvalues @xmath7 from the fredholm integral equation @xmath9 this representation is often referred to as a coherent mode decomposition of @xmath0 .
mathematically eq .
[ eq : decomposition ] is a sum over an infinite number of modes , but in practice @xmath10 is bounded by the maximum spatial frequency content of @xmath1 , i.e. there is some maximum @xmath11 such that for @xmath12 , @xmath13 will be negligibly small .
for example , gaussian schell - model beams are a common example of a partially coherent beam .
such a beam is defined by having a gaussian intensity @xmath14 , as well as a gaussian degree of coherence @xmath15 , which gives a cross - spectral density function @xmath16 a coherent mode decomposition of such a gaussian schell - model beam shows that the number of coherent modes necessary to describe eq .
[ eqn : gaussianschell ] is given by the number of independent coherent regions within the beam which is quantified by @xmath17 @xcite .
physically eq .
[ eq : decomposition ] can be realized if one can create a beam that alternates between the coherent modes @xmath5 in time with relative frequency weighted by @xmath13 . for measurement to yield the intended field
, the switching time @xmath18 must be much faster than any detector integration time @xmath19 in order to create the intended averaging over the inputs .
in addition , for the mixture to be an incoherent mixture , the various modes must not have any correlations in time .
thus the switching time must be slower than the coherence time @xmath20 of the source .
together these form the condition @xmath21 if eq .
[ eqn : coherencetimes ] is met , then one has a physically realized implementation of @xmath1 .
in order to generate an arbitrary partially coherent field @xmath1 , one only needs to find a way to create the coherent fields @xmath5 in rapid succession .
such rapid mode generation was recently demonstrated by using dmds to create quickly addressable binary amplitude modulated cghs @xcite , though this comes at the cost of having a maximum efficiency around 10% .
dmds are devices that provide both the speed and resolution desired for rapid generation and switching of coherent fields @xcite .
a dmd consists of a 2-dimensional array of mirrors that can be in one of two positions , which can be used to act as an on or off state at each pixel .
each pixel can be individually addressed and changed very rapidly , at frame rates exceeding @xmath22 .
( 1,0.35 ) ( 0,0 ) ( 0.16937818,0.07624671)(0,0)[lb ] ( 0.15969989,0.33046017)(0,0)[b ] ( 0.24935869,0.28885149)(0,0)[lb ] ( 0.07004105,0.28885149)(0,0)[rb ] ( 0.50968643,0.07624671)(0,0)[lb ] ( 0.58924682,0.28885149)(0,0)[lb ] ( 0.8499949,0.07624705)(0,0)[lb ] ( 0.92955516,0.28885149)(0,0)[lb ] the fact that dmds have 2 settings , allows us to make a binary grating .
any periodic structure acts as a diffraction grating .
a transverse shift in this diffraction grating will induce a phase shift or detour phase in the diffracted orders , even if the grating is an amplitude only structure .
in addition , the form of each period will determine the scattering efficiency into the diffracted order .
taken together , modulating the grating position and each periodic form locally within the hologram allows one to control both the amplitude and phase , and thus create any field , @xmath23 in the diffracted order .
a well known method of encoding binary holograms is to create a periodic array of binary fringes or rectangular ` pulses . '
a one dimensional representation of this is shown in fig .
[ fig : binarygrating ] .
a shift in the location of these pulses will change the overall phases into the diffracted orders , while changing the widths or duty cycles of the pulses will change the diffracted efficiency .
these two methods are known as pulse position and pulse width modulation respectively @xcite , and such a modulation represents a generalization of the moir technique @xcite .
mathematically , a periodic binary grating can be written as a fourier series @xmath24 where @xmath25 is the grating wavevector .
the grating consists of rectangular pulses of width @xmath26 spaced at a period of @xmath27 and @xmath28 $ ] is the relative location of the array within each period . looking only at the first diffraction order @xmath29 the field
is given by @xmath30 where @xmath31 is the input field , which we ll assume to be a constant plane wave .
in addition all optics after the dmd are aligned along the axis of the first diffraction order , allowing us to ignore any phase tilt from @xmath31 as well as the @xmath32 tilt from the grating in our description of @xmath33 .
we can allow @xmath34 and @xmath35 to become functions of position and the previous results still hold so long as @xmath36 and @xmath37 vary much slower than the grating period @xmath27 .
then any complex field @xmath38 can be generated by allowing @xmath39 where the phase is @xmath40,$ ] which is defined symmetrically around 0 to avoid encoding errors in the presence of a varying amplitude @xcite .
this full procedure can be represented in the following fashion .
first one chooses the field @xmath41 that one wishes to create .
then @xmath36 and @xmath37 are computed from eq .
[ eqn : pulsepostionandwidth ] and a periodic sinusoidal function is computed to give @xmath42 to convert this into a binary hologram , this function is thresholded by @xmath43 to create a binary pulse train with local pulse width @xmath44 .
this can be written in the compact form @xmath45 , \label{eqn : gratingmodulation}\ ] ] where @xmath46 is the heaviside step function defined as @xmath47
a schematic of the experimental setup is shown in fig . [
fig : setup ] .
a hene laser is spatially filtered using a 4f system to provide an initial coherent plane wave incident on the dmd .
the various coherent modes , @xmath48 , are created in rapid succession with a spatially modulated binary diffraction grating on the dmd that gives the desired field in the first diffraction order .
a second 4f system and pinhole are used to filter out all other diffraction orders and the resultant beam is imaged onto a ccd camera . .
a fully spatially coherent plane wave is prepared by spatial filtering of a hene laser .
this collimated beam is reflected off a cgh generated by the dmd and the desired diffracted order is filtered by a 4f system and imaged onto a ccd.,scaledwidth=47.0% ] the dmd is a type of micro - electronic mechanical system , commonly known as a mems , that can function as an amplitude only slm @xcite .
the device consists of a two dimensional pixelated array of micromirrors each mounted on an individually addressed mems that can be in one of two positions . in order to use the device as a slm
, the device is aligned such that the light is reflected and collected by the optics after the dmd if the micromirrors are in the on position , but scattered out of the system if the mirrors are in the off position . the device used in the experiment was a texas instrument dlp3000 .
this device has a display resolution of @xmath49 pixels , a micromirror size of @xmath50 , and the pixels can be switched at rates up to 4khz which is much faster than a typical phase based slm @xcite .
the ccd operates at 60hz , thus the detector integration time is @xmath51 .
the dlp3000 dmd used in this experiment has a switching rate of 4khz , thus @xmath52 , which fulfills the first inequality in eq .
[ eqn : coherencetimes ] .
the bandwidth of the hene is 1.5ghz which gives @xmath53 which meets the second part of the inequality in eq .
[ eqn : coherencetimes ] . as a demonstration of the ability to generate a single coherent state
the field @xmath54 was generated .
this represents a coherent superposition of two plane wave states , which form a sinusoidal interference pattern as shown in fig [ fig : coherentdata ] . for this experiment
the mode was generated using a grating with wavevector @xmath55{px}}(\mathbf{\hat{x } } + \mathbf{\hat{y}}),\ ] ] which represents a period of @xmath56{pixels } \approx \unit[275]{\mu m}$ ] oriented at @xmath57 .
this value of @xmath58 was chosen to be large enough to allow enough separation in the fourier plane to allow for filtering of the 1st diffracted order with an iris .
in addition a nonzero value was chosen for both the @xmath59 and @xmath60 components of @xmath58 in order to minimize the noise by ensuring that the diffracted order did not overlap with any specular reflection due to the dmd s imperfect pixel fill - fraction .
the underlying grating can be seen in the left image in fig .
[ fig : coherentdata ] which have the appearance of the small diagonally oriented slivers . the plane wave transverse wavenumber was chosen to be @xmath61{px } } \approx \frac{2\pi}{\unit[780]{\mu m}}.\ ] ] @xmath62 and thus is slowly varying enough to allow us to use the procedure in section [ sec : binarygratings ] to construct the cgh to create this state .
since we are perfectly interfering 2 plane waves , the intensity varies as @xmath63 .
therefore @xmath64 , while @xmath65 .
next we created a superposition of the plane waves @xmath66 and @xmath67 as before , but this time the degree of coherence between the two beams was spatially varied , creating a partially coherent mix of modes .
the coherent modes used to represent this is given by @xmath68 where the relative probability weightings are given as @xmath69 , and where @xmath70 is related to the fringe visibility @xmath71 by @xmath72 the intensity for this beam is @xmath73 which is the sum of an incoherent and a coherent term which can be continuously tuned from fully coherent ( @xmath74 ) to incoherent ( @xmath75 ) .
the visibility function chosen for the experiment is given by @xmath76 where @xmath77{px } } \approx \frac{2\pi}{\unit[580]{\mu m}}.\ ] ] since @xmath78 was chosen to be real , eq .
[ eqn : visibility ] also represents our spectral degree of coherence at @xmath79 .
the cghs necessary to create the modes @xmath80 and @xmath81 ( eq . [ eqn : partiallycoherentmodes ] ) for this spatially varying fringe visibility are shown in the top row of fig .
[ fig : partiallycoherent ] .
the cgh parameters are @xmath82 where @xmath83 is the maximum value of @xmath84 and @xmath85 along the 1d slice of @xmath86 for @xmath87 , i.e. along the @xmath59-axis ( solid blue line ) .
also shown as the black dotted line is the theoretical envelope of the maximum and minimum intensities based on the intended visibility function @xmath88.,scaledwidth=47.0% ] in order to compare the intended visibility given by eq .
[ eqn : visibility ] with the image shown in fig .
[ fig : partiallycoherent ] , a one dimensional slice of the intensity is plotted in fig .
[ fig : fringevisibility ] .
this slice is a radial slice @xmath86 along the @xmath59 axis ( i.e. at an orientation of @xmath87 ) , and is plotted over an entire period of @xmath89 of the visibility .
in addition the theoretical envelope of the visibility equal to @xmath90 is plotted for comparison . as can be seen in both the original coherent and partially coherent cases ,
the intended and measured patterns are in excellent agreement with one another .
in this paper we have demonstrated a novel method of generating arbitrary fields of light .
any partially coherent field that is described by the cross - spectral density function @xmath1 can be generated by computing the coherent mode decomposition into an incoherent sum of coherent modes .
this incoherent mix of modes was physically realized by rapid generation of spatial holograms on a dmd and was temporally averaged in detection .
we acknowledge mayukh lahiri , joe vornehm and alex radunsky for helpful discussions .
our work was supported by the defense advanced research projects agency ( darpa ) inpho program and osml also acknowledges support from the conacyt .
j. wang , j .- y .
yang , i. m. fazal , n. ahmed , y. yan , h. huang , y. ren , y. yue , s. dolinar , m. tur , and a. e. wilner , `` terabit free - space data transmission employing orbital angular momentum multiplexing , '' nature photonics * 6 * , 488496 ( 2012 ) .
b. rodenburg , m. p. j. lavery , m. malik , m. n. osullivan , m. mirhosseini , d. j. robertson , m. j. padgett , and r. w. boyd , `` influence of atmospheric turbulence on states of light carrying orbital angular momentum , '' * 37 * , 3735 ( 2012 ) .
a. c. dada , j. leach , g. s. buller , m. j. padgett , and e. andersson , `` experimental high - dimensional two - photon entanglement and violations of generalized bell inequalities , '' nature physics * 7 * , 677680 ( 2011 ) .
r. w. boyd , a. jha , m. malik , c. osullivan , b. rodenburg , and d. j. gauthier , `` quantum key distribution in a high - dimensional state space : exploiting the transverse degree of freedom of the photon , '' proceedings of spie * 7948 * , 79480l6 ( 2011 ) .
m. malik , m. n. osullivan , b. rodenburg , m. mirhosseini , j. leach , m. p. j. lavery , m. j. padgett , and r. w. boyd , `` influence of atmospheric turbulence on optical communications using orbital angular momentum for encoding , '' * 20 * , 13195 ( 2012 ) .
z. chen , s. m. sears , h. martin , d. n. christodoulides , and m. segev , `` clustering of solitons in weakly correlated wavefronts , '' proceedings of the national academy of sciences * 99 * , 52235227 ( 2002 ) .
c. rickenstorff , e. flores , m. olvera - santamara , and a. ostrovsky , `` modulation of coherence and polarization using nematic 90 degree - twist liquid - crystal spatial light modulators , '' revista mexicana de fsica * 58 * , 270273 ( 2012 ) . p. zhang , z. zhang , j. prakash , s. huang , d. hernandez , m. salazar , d. n. christodoulides , and z. chen , `` trapping and transporting aerosols with a single optical bottle beam generated by moir techniques .
'' optics letters * 36 * , 14913 ( 2011 ) . | we describe an experimental technique to generate a quasi - monochromatic field with any arbitrary spatial coherence properties that can be described by the cross - spectral density function , @xmath0 .
this is done by using a dynamic binary amplitude grating generated by a digital micromirror device ( dmd ) to rapidly alternate between a set of coherent fields , creating an incoherent mix of modes that represent the coherent mode decomposition of the desired @xmath0 .
this method was then demonstrated experimentally by interfering two plane waves and then spatially varying the coherence between them .
it is then shown that this creates an interference pattern between the two beams whose fringe visibility varies spatially in an arbitrary and prescribed way . |
nls1 sources are defined as a subclass of broad line objects at the basis of the optical spectra , as sources with the full width at half maximum ( fwhm ) of h@xmath2 line , produced in broad line region , smaller than 2100 kms@xmath3 . in bls1s fwhm(h@xmath2 ) is @xmath4 2100 kms@xmath3 .
it was later found that nls1 generally have smaller black hole masses for a given luminosity , and strong soft x - ray excesses ( mathur et al .
2001 ; crummy et al , 2006 ) .
soft x - ray photon indices , @xmath1 , of nls1 sources have a mean value 3.1 , while bls1 sources are harder , with a mean of 2.1 ( e.g. boller et al .
it suggests that nls1 and bls1 differ with respect to the accretion pattern .
we have considered a sample of 21 nls1 galaxies ( e.g. mrk 110 , 335 , 478 , ark 564 , ngc 4051 , 5506 , ton s180 , i zw 1 , pg 1211 + 143 ) and a sample of bls1 galaxies ( e.g. ngc 4151 , 5548 , 7469 , 3c 120 , f9 ) studied by nikoajuk et al .
all sources were monitored both in optical and in x - ray band .
the masses of supermassive black hole for all our galaxies were taken from literature , @xmath5 .
they were obtained with the reverberation method ( e.g. peterson et al .
2004 ) or with the stellar velocity dispersion method ( e.g. woo & urry 2002 ) . those masses were used to compare with values obtained from x - ray variability .
we have determined the masses using the variance method described by nikoajuk et al .
it is based on the scaling of the normalized variance of x - ray light curves , @xmath6 , measured in range 2 - 10 kev with the @xmath7 value .
generally , values of @xmath8 require a multiplication by some factor .
mean value of this factor is 20 ( see e.g. papadakis , 2004 ) .
surprisingly , this factor for four galaxies ( mrk 42 , mrk 507 , mrk 766 , mcg -6 - 30 - 15 ) is smaller than 3 and for hb89 1557 + 272 is smaller than 4 ( fig .
[ fig : mg ] ) .
the variability of x - ray light in the 2 - 10 kev range of those 5 nls1 galaxies is thus very similar to all bls1 sources .
we find that this subset have values of soft x - ray photon index , @xmath1 , less than or equal to @xmath0 2 .
once again it is behaviour of bls1 galaxies .
spectra in the soft x - ray range of those 5 nls1 are similar to spectra of bls1 . on the other hand when we compare accretion rates
, we can not see difference between those 5 nls1 and other nls1 galaxies .
this small subset of nls1 accrete , like other nls1 , with @xmath9 higher than 0.1 or even close to eddington limit .
this is opposite to bls1 sources , which accrete with @xmath10 . | we estimate black hole masses in narrow line seyfert 1 ( nls1 ) galaxies at the basis of their x - ray excess variance .
we apply the standard approach appropriate for broad line seyfert 1 ( bls1 ) galaxies . in general
, we find that the obtained masses are by a factor @xmath0 20 too small to agree with values obtained from other methods ( reverberation , stellar dispersion )
. however , a small subset of our nls1 objects does not require that multiplication , or the correction factor is less than 4 .
we find that this subset have a soft x - ray photon index , @xmath1 , smaller than 2 .
we thus postulate that this subclass of nls1 actually belongs to bls1 . |
early empirical correlations between the preponderance of young stars and gas in galaxies ( e.g. , @xcite ) confirmed in a qualitative sense the simple power - law dependence between star formation rate surface density ( @xmath15 ) and gas surface density ( @xmath16 ) first suggested by @xcite who found @xmath17 for hi gas . once the h@xmath18 component as traced by co lines was identified in galaxies , the gas surface density could be related to both hi and h@xmath18 , i.e. , @xmath19 @xcite . in a seminal paper , @xcite established this relation , hereafter called the schmidt - kennicutt ( s - k ) relation , to be : @xmath20 , averaged over entire galaxy disks .
further studies by @xcite and @xcite found a nearly linear s - k relation for the molecular gas on kpc scales ( see also @xcite and @xcite ) , with the sfr surface density having a much closer correspondence with the molecular gas surface density reflecting the well - established fact that stars form out of molecular rather than atomic gas . much theoretical effort has gone into obtaining the exponents and normalization of this relation as unique outcomes of various physical processes occurring in star forming galaxies , with various models capable of yielding ( s - k)-type relations ( e.g. , @xcite ) . it became evident that , while no deterministic microphysics of the interstellar medium ( ism ) and star formation ( sf ) can be linked to a given s - k relation , the high - density gas component ( @xmath21 ) plays a crucial role in ultimately anchoring such relations to the star formation taking place deep inside supersonically turbulent molecular clouds in disks .
the s - k relations for high - density gas are particularly challenging to establish since determining the dense gas mass fraction within a galaxy requires observations of co from @xmath3 ( a total molecular gas mass tracer ) up to at least @xmath22 along with the much fainter lines of _ bona fide _ dense gas tracers like cs and heavy - rotor molecules such as hcn . a multi - component analysis of such co , hcn , and cs spectral line energy distributions ( sleds )
can then yield dense gas masses , @xmath23 ( e.g. , @xcite ) .
however , to do so for a large number of galaxies in order to obtain even a surface - integrated sfr@xmath24 s - k relation has been prohibitively expensive in telescope time . at high redshifts
the situation is made worse due to a lack in sensitivity and angular resolution .
nonetheless , pioneering efforts have been made at discerning @xmath25 at high redshifts using h@xmath26 maps obtained with integral field unit cameras , and high - resolution interferometric co ( @xmath3 to @xmath27 ) observations of massive star forming galaxies at @xmath28 @xcite .
obviously , this situation will now improve dramatically with the advent of the atacama large millimeter / sub - millimeter array ( alma ) . with the dense gas mass fraction distribution currently inaccessible for any statistically significant number of galaxies one must fall back to the integrated ( s - k)-proxy relations : @xmath29 ( where @xmath30 is the line luminosity of a dense gas tracer and @xmath31 a linear proxy of sfr ) , and then invoke theoretically determined links to an underlying s - k relation @xcite .
hcn(@xmath32 ) observations of statistically significant samples of local ir luminous galaxies ( lirgs ) and normal spiral galaxies yielded the first of such ( s - k)-proxy relations using gas tracers other than co lines @xcite , finding the ir@xmath1hcn relation to be linear and with much less scatter than the previously determined ir@xmath1co low-@xmath8 relations .
this was interpreted as hcn(@xmath32 ) , with its high critical density ( @xmath33 ) , being a more direct tracer of a dense , star forming gas component with a nearly constant underlying star formation efficiency ( sfe ) @xcite .
furthermore , with the tight , linear ir@xmath1hcn relation extending down to individual galactic molecular clouds where @xmath34 , thus covering over @xmath35 orders of magnitude in luminosity , its origin could be attributed to the existence of fundamental ` units ' of cluster star formation @xcite .
this view is now further supported by the linear @xmath29 relations found also for the hcn(@xmath36 ) and cs(@xmath37 ) lines @xcite , which for cs(@xmath37 ) also extends ( linearly ) down to galactic cores @xcite . some contentious points do remain however , especially towards the high-@xmath31 end which is dominated by mergers / starbursts where a slightly super - linear ir@xmath1hcn relation has been claimed and argued to be due mostly to an increase in the dense gas sfe in such galaxies @xcite . in this paper
we present the first fir@xmath1co luminosity relations and the corresponding global co sleds that extend above @xmath38 and up to @xmath39 using _ herschel _ spire - fts data for local ( u)lirgs . the fir@xmath1co relations and co sleds presented in this work ( from @xmath3 up to @xmath4 ) , besides a significant extension of the @xmath8-ladder , benefit also from the inclusion of ( u)lirgs from the low- and the high-@xmath40 universe .
this robustly extends the sample towards the important high-@xmath41 end ( as numerous galaxies with ulirg - like , or higher , luminosities have been found in the high-@xmath40 universe ) where very different conditions may prevail for the molecular gas , possibly leaving an imprint on the fir@xmath1co relations and the co sleds .
our new high-@xmath8 co line data - set is uniquely sensitive to such an imprint since these lines need both high densities ( @xmath42 ) _ and _ ( in most circumstances ) high temperatures ( @xmath43 ) to be significantly excited .
the high - density and warm gas necessary for exciting them is the most difficult phase to maintain energetically in appreciable quantities in galaxies .
however , it is one that would leave no easily discernible signature in the low-@xmath8 co and low-/mid-@xmath8 sleds of heavy rotor molecular lines ( e.g. , hcn , cs ) that typically have been available for ( u)lirgs up to now . throughout
, we adopt a flat cosmology with @xmath44 , and @xmath45 @xcite .
for the purposes of this work we first compiled high-@xmath8 ( @xmath46 up to @xmath4 line data from the _ herschel _ comprehensive ( u)lirg emission survey ( hercules , @xcite ) an open time key program on the esa _ herschel _ space observatory @xcite which measured co @xmath46 to @xmath4 for 29 local ( @xmath47 ) ( u)lirgs using the fourier transform spectrometer ( fts ) of the spire instrument @xcite .
the hercules sources were selected from the @xmath48 flux - limited iras revised bright galaxy sample ( @xmath49 ; @xcite ) with separate flux cuts applied to ulirgs and lirgs ( @xmath50 and @xmath51 , respectively ) .
a detailed description of the spire - fts observations , calibration modes , extraction of co line fluxes , and final line luminosities , are given in a dedicated paper ( rosenberg et al .
, in prep . ) .
briefly , the high spectral resolution mode was used with a resolution of @xmath52 over both observing bands .
a reference measurement was used to subtract the emission from the sky , telescope , and instrument .
the spectra were reduced using the herschel interactive processing environment ( hipe ) , ver .
9.0 . at the time of writing ,
fully reduced spire - fts co spectra were available for only 26 sources , and of these three had extended , multi - component morphologies and were discarded . since the spire - fts beam ranges from @xmath53 to @xmath54 ( fwhm ) across the bandpass @xcite , it is essential to perform a beam correction in cases where the sources are extended with respect to the beam .
all spectra ( and thus co line fluxes ) were scaled to a common spatial resolution of @xmath54 using laboca @xmath55 or saboca @xmath56 maps ( see rosenberg et al .
, in prep . for details ) .
obviously , this assumes that the corrections are perfectly mono - chromatic in the fir and sub - millimeter ( sub - mm ) regime , which is a good assumption to within @xmath57 @xcite . for the hercules sources , which are all ( u)lirgs and thus nearly all relatively compact , and well within the beam sizes of the co observations , this correction was minor .
for very extended sources , however , this correction is crucial , and failing to apply it can skew the observed fir@xmath1co relation ( i.e. , @xcite , @xcite , and see discussion in @xcite ) . )
luminosity vs.redshift for the galaxy samples considered in this paper ( after agn - dominated systems have been removed ) , along with histograms of the fir luminosity and redshift distributions ( top and right inserts , respectively ) .
the local ( @xmath47 ) sources include sub - sets of the ( u)lirg samples from hercules ( 20 sources , red symbols ) and @xcite ( 42 sources , grey symbols ) the high-@xmath40 ( @xmath58 ) sources are unlensed , or weakly lensed , dsfgs ( 16 sources , green symbols ) and strongly lensed dsfgs ( 19 sources , blue symbols ) uncovered from various ( sub)-mm surveys ( see [ section : samples - and - data ] ) .
all fir luminosities have been been corrected for lensing using the magnification factors in table [ table : high - z - sample ] . ]
we also included ground - based co line data presented by @xcite for a sample of 45 local ( u)lirgs from the _ iras _ rbgs .
these data consisted of low-@xmath8 co transitions , i.e. , @xmath3 ( all 45 sources ) , @xmath59 ( 17 ) , @xmath27 ( 44 ) , as well as @xmath46 ( 3 ) and @xmath60 ( 12 ) observations
. this allowed us to both fill - in the @xmath3 , @xmath59 , @xmath27 transitions for the 11 hercules sources that overlapped with this sample ( except for one source which did not have @xmath61 measurement ) , and bring in additional co low-@xmath8 and @xmath62 lines ( the only mid-/high-@xmath8 co lines accessible with the ground - based telescopes used
see @xcite for details ) to the sample .
we stress that the co line fluxes given in @xcite are _ total _ line fluxes , and so no additional beam correction is required for these sources . of our sample of 68 local ( u)lirgs ( listed in table [ table : low - z - sample ] ) ,
30 sources ( 20 + 10 from hercules and @xcite sub - samples , respectively ) are also part of the great observatories all - sky lirg survey ( goals ; @xcite ) .
to weed out active galactic nuclei ( agn ) , the sample was cross - correlated against estimates of the agn contribution to the bolometric luminosity based on several mir diagnostics such as the equivalent width of the @xmath63 pah feature , the [ nev]@xmath64[neii ] and [ oiv]@xmath64[neii ] emission line ratios as well as @xmath65-to-@xmath66 continuum flux ratios @xcite .
only six sources ( indicate by a @xmath67 in table [ table : low - z - sample ] ) were found to have an agn contribution @xmath68 and were omitted from our analysis ( although , including them in our analysis did not alter the findings of this paper ) .
the fir / sub - mm continuum data were obtained from a number of studies ( see @xcite and references therein ) as well as from the nasa / ipac extragalactic database ( ned ) .
all the @xmath69 and @xmath70 available fluxes were corrected for co @xmath22 , and @xmath59 line contamination ( @xmath71 ) , respectively .
we also corrected for any non - thermal radio continuum contributions whenever radio data were available , allowing for a power - law extrapolation to the sub - mm wavelengths .
the fir ( @xmath72 ) luminosities derived from the continuum data ( see [ section : analysis ] for details ) span the range @xmath73 ( fig.[figure : fig0 ] ) .
the two samples are well matched in luminosity , although no hercules sources are found at @xmath74 .
the fact that the more luminous sources tend to have higher redshifts merely reflects the flux - limited selection of the two samples .
high - redshift dusty star forming galaxies ( dsfgs ) are thought to resemble the local ( u)lirg population and most have @xmath75 .
moreover , typically multiple high-@xmath8 co lines and fir/(sub)-mm continuum observations are available for them .
these were the main reasons for including them in our analysis . in order to achieve the best possible uniformity , a meticulous compilation of the aforementioned observations ( co line and continuum observations ) for all published dsfgs
was extracted from the literature ( guided by major review papers by @xcite and @xcite ) .
sources with clear signs of agn ( e.g. , from optical spectroscopy showing strong ly@xmath26 , civ , and ciii emission lines and a power - law continuum , or radio - loudness ) were not included in our sample . in cases where multiple observations of the same co transition existed ,
we adopted the weighted mean of the velocity - integrated line flux after discarding any outliers and measurements with low signal - to - noise .
many of the high-@xmath40 co detections are of strongly lensed dsfgs , which we here take to mean a gravitational magnification factor ( @xmath76 ) @xmath77 , and in those cases we adopted the best estimates of @xmath76 available at the time of writing ( e.g. , @xcite ) .
a total of 74 dsfgs constituted our initial high-@xmath40 sample . for 39 ( 53% ) of the dsfgs
, however , we were unable to put reliable constraints on their fir luminosities ( see [ section : analysis ] ) , and these were therefore discarded for the analysis presented in this paper .
this left us with a final sample of 35 high-@xmath40 sources ( listed in table [ table : high - z - sample ] ) , spanning the redshift range @xmath78 with a median redshift of @xmath79 ( see also fig.[figure : fig0 ] ) .
the lensed dsfgs ( 19 in total ) , _ after magnification correction _ , span a similar range in fir luminosity as the non - lensed dsfgs ( @xmath80 , see fig .
[ figure : fig0 ] ) , which is about an order of magnitude higher than that of the local samples .
the only exception is smmj163555.2@xmath81661150 ( @xmath82 ) , which has an intrinsic luminosity similar to that of local lirgs @xcite . within the high-@xmath40 samples , we see no strong dependence of fir luminosity on redshift , which is due to the well - known flat selection function at sub - mm wavelengths for @xmath83 @xcite .
finally , we stress that while the dsfgs , as a sample , cover all co transitions from @xmath3 to @xmath84 , no individual galaxy has continuous coverage across this transition range .
lccc + i d & @xmath40 & @xmath85 } / { \mbox{$l_{\odot}\,$}})$ ] & @xmath86 } / { \mbox{$l_{\odot}\,$}})$ ] + iras00085@xmath11223 ( ngc34 ) & 0.0196 & 11.21 & 11.47 + iras00506@xmath817248 ( mcg@xmath8112 - 02 - 001 ) & 0.0157 & 11.26 & 11.53 + iras01053@xmath11746 ( ic1623 ) & 0.0201 & 11.47 & 11.74 + iras04315@xmath10840 ( ngc1614 ) & 0.0159 & 11.32 & 11.59 + iras05189@xmath12524@xmath67 & 0.0426 & 11.73 & 12.12 + iras08354@xmath812555 ( ngc2623 ) & 0.0185 & 11.33 & 11.60 + iras10257@xmath14339 ( ngc3256 ) &
0.0094 & 11.40 & 11.60 + iras11506@xmath13851 ( eso320@xmath1g030 ) & 0.0108 & 11.09 & 11.30 + iras12540@xmath815708 ( mrk231)@xmath67 & 0.0422 & 12.14 & 12.56 + iras13120@xmath15453 ( wkk2031 ) & 0.0308 & 12.07 & 12.34 + iras13183@xmath813423 ( arp193 ) & 0.0233 & 11.44 & 11.68 + iras13229@xmath12934 ( ngc5135 ) & 0.0137 & 11.13 & 11.33 + iras13242@xmath15713 ( eso173@xmath1g015 ) & 0.0097 & 11.38 & 11.65 + iras13428@xmath815608 ( mrk273)@xmath67 & 0.0378 & 11.91 & 12.17 + iras16504@xmath810228 ( ngc6240 ) & 0.0245 & 11.61 & 11.87 + iras15107@xmath810724 ( zw049.057 ) & 0.0130 & 11.05 & 11.28 + iras17208@xmath10014 & 0.0428 & 12.20 & 12.47 + iras18093@xmath15744 ( ic4687 ) & 0.0173 & 11.12 & 11.39 + iras18293@xmath13413 & 0.0182 & 11.62 & 11.84 + iras23007@xmath810836 ( ngc7469 ) & 0.0163 & 11.32 & 11.60 + iras23134@xmath14251 ( ngc7552 ) & 0.0054 & 10.84 & 11.05 + iras23488@xmath812018 ( mrk331 ) & 0.0185 & 11.26 & 11.53 + iras23488@xmath811949 ( ngc7771 ) & 0.0143 & 11.26 & 11.43 + _
iras00057@xmath814021 _ & 0.0445 & 11.34 & 11.60 + _
iras00509@xmath811225 _ & 0.0611 & 11.40 & 11.67 + _
iras01077@xmath11707 _ & 0.0351 & 11.42 & 11.69 + _ iras01418@xmath811651 _ & 0.0274 & 11.29 & 11.56 + _
iras02114@xmath810456 _ & 0.0297 & 11.26 & 11.43 + _
iras02401@xmath10013 _ & 0.0037 & 10.96 & 11.23 + _
iras02483@xmath814302 _ & 0.0514 & 11.67 & 11.85 + _
iras02512@xmath811446 _ & 0.0312 & 11.43 & 11.70 + _ iras03359@xmath811523 _ & 0.0353 & 11.27 & 11.45 + _
iras04232@xmath811436 _ & 0.0795 & 11.81 & 12.08 + _
iras05083@xmath817936 _ & 0.0543 & 11.88 & 12.06 + _ iras08572@xmath813915_@xmath67 & 0.0582 & 11.73 & 12.11 + _
iras09126@xmath814432 _ & 0.0398 & 11.48 & 11.65 + _
iras09320@xmath816134 _ & 0.0393 & 11.77 & 11.95 + _
iras09586@xmath811600 _ & 0.0080 & 10.37 & 10.64 + _
iras10035@xmath14852 _ & 0.0648 & 11.83 & 12.10 + _ iras10039@xmath13338_@xmath67 & 0.0341 & 11.47 & 11.74 + _
iras10173@xmath810828 _ & 0.0489 & 11.53 & 11.80 + _
iras10356@xmath815345 _ & 0.0033 & 10.08 & 10.35 + _
iras10565@xmath812448 _ & 0.0428 & 11.75 & 12.01 + _
iras11231@xmath811456 _ & 0.0341 & 11.40 & 11.57 + _
iras12001@xmath810215 _ & 0.0066 & 10.11 & 10.29 + _
iras12112@xmath810305 _ & 0.0727 & 12.07 & 12.34 + _
iras12224@xmath10624 _ & 0.0263 & 11.05 & 11.23 + _
iras12243@xmath10036 _ & 0.0073 & 10.77 & 11.04 + _
iras13001@xmath12339 _ & 0.0215 & 11.24 & 11.41 + _
iras13102@xmath811251 _ & 0.0112 & 10.48 & 10.66 + _
iras13188@xmath810036 _ & 0.0186 & 11.00 & 11.17 + _
iras13362@xmath814831 _ & 0.0278 & 11.18 & 11.45 + _
iras13470@xmath813530 _ & 0.0168 & 10.96 & 11.10 + _
iras13564@xmath813741 _ & 0.0125 & 10.88 & 11.05 + _
iras14003@xmath813245 _ & 0.0145 & 10.77 & 10.95 + _
iras14178@xmath814927 _ & 0.0256 & 11.08 & 11.35 + _ iras14348@xmath11447 _ & 0.0825 & 12.08 & 12.42 + _
iras15163@xmath814255 _ & 0.0402 & 11.59 & 11.94 + _
iras15243@xmath814150 _ & 0.0089 & 10.38 & 10.65 + _ iras15327@xmath812340 _ & 0.0182 & 11.80 & 11.98 + _ iras15437@xmath810234_@xmath67 & 0.0128 & 10.84 & 11.01 + _
iras16104@xmath815235 _ & 0.0292 & 11.37 & 11.63 + _
iras16284@xmath810411 _ & 0.0245 & 11.23 & 11.40 + _
iras17132@xmath815313 _ & 0.0507 & 11.62 & 11.89 + _
iras19458@xmath810944 _ & 0.1000 & 12.28 & 12.45 + _
iras20550@xmath811656 _ & 0.0363 & 11.63 & 11.97 + _
iras22491@xmath11808 _ & 0.0773 & 11.92 & 12.19 + _
iras23365@xmath813604 _ & 0.0644 & 11.88 & 12.15 + & & & + & & & + [ table : low - z - sample ] [ cols="<,^,^,^,<,<",options="header " , ] [ table : critical - densities ] two trends regarding the @xmath87 relations become apparent from figs .
[ figure : fig2 ] and [ figure : fig3 ] ( see also table [ table : fits ] ) .
firstly , the slopes are linear for @xmath3 to @xmath9 but then start becoming increasingly sub - linear , the higher the @xmath8 level .
secondly , the normalization parameter @xmath88 remains roughly constant ( @xmath89 ) up to @xmath46 , @xmath90 , but then increases with higher @xmath8 level , reaching @xmath91 for @xmath4 , which for a given co luminosity translates into @xmath92 orders of magnitude higher @xmath41 .
we stress that although the @xmath87 relations are linear , and @xmath88 roughly constant , up to @xmath9 , it does not in general imply that the co lines are thermalized ( i.e. , @xmath93 ) up to this transition .
there is significant scatter within the samples , and while a few sources do have nearly - thermalized @xmath61 , @xmath27 , and/or @xmath36 lines , in general , @xmath94 .
in fact , re - writing the @xmath87 relations as : @xmath95 and inserting the fitted values from table [ table : fits ] yields @xmath96 over the range @xmath97 . in the following sections we discuss these empirical relations in the context of existing theoretical models and the new observational studies of such relations using heavy rotor molecules like hcn , and cs .
before comparing our derived fir@xmath1co slopes with those from the literature we must add two cautionary notes , namely : a ) many studies examine the @xmath98 relation , rather than @xmath87 , and one can not compare the two simply by inferring the inverse relation , b ) often , only the errors in one variable ( typically @xmath99 ) are taken into account when fitting such relations , when in fact the uncertainties in both @xmath41 and @xmath99 must be considered ( see @xcite for a further discussion ) .
failing to do so can result in erroneous estimates of the slope . for these reasons ,
we have re - fitted the data from a number of studies ( see below ) using the method described in
[ section : analysis ] , i.e. , with errors in both @xmath41 and @xmath99 and including only sources with @xmath100 in our analysis .
finally , not all studies use the fir definition used here to infer @xmath41 , while other studies use the full ( @xmath101 ) luminosity .
these differences can result in a different overall normalization ( i.e. , @xmath88 ) , but are not expected to affect the determination of @xmath26 ( see table [ table : fits ] where there is little change in @xmath26 when switching between @xmath102}$ ] and @xmath103}$ ] ) . from fig .
[ figure : fig2 ] we note the overall good agreement between the fir@xmath1co slopes derived here and values from the literature . for co(@xmath32 ) , however , one set of measurements found super - linear slopes ( @xmath104 ; @xcite ) , while most others favor a slope of unity ( * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* this work ) .
note , our re - analysis of the @xcite and @xcite data revised their slopes from super - linear to linear : @xmath105 and @xmath106 , respectively ( a similar result was found by @xcite ) .
@xcite finds a super - linear fir@xmath1co slope ( @xmath107 ) from their entire sample ( combining @xmath108 objects with lirgs and ulirgs ) , yet including only the lirgs and ulirgs in the analysis , we obtain a linear slope ( @xmath109 ) .
re - fitting the data presented in @xcite and @xcite we reproduce their super - linear slopes . for co(@xmath59 )
our slope of unity is consistent within the errors with @xcite , who finds a slightly super - linear slope based on 17 sources .
our re - analysis of the co@xmath110 data by @xcite yields a sub - linear slope of @xmath111 .
however , as pointed out by the authors themselves , a non - negligible fraction of the total co@xmath110 emission is likely to have been missed due to the smaller telescope beam at higher frequencies thus biasing the fir@xmath1co relation to a shallower value of @xmath26 . in the case of co(@xmath27 ) , the existing slope - determinations @xcite , including our own , are all in agreement and favor a value of unity within the errors .
this includes a re - analysis of the @xcite data , which yielded @xmath112 ( see also @xcite ) . for co(@xmath36 ) to co(@xmath37 ) there is agreement within the errors between our results and the slopes found by @xcite , which however were determined using observed as well as model - extrapolated co luminosities of 7 low-@xmath40 and 10 high-@xmath40 sources . a departure from linear towards sub - linear is also found by the latter study ,
albeit we find this turn - over to occur at @xmath12 rather than at @xmath46 deduced by @xcite . here
we must note , however , that the use of models to extrapolate to high-@xmath8 co luminosities is not safe and artificial turnovers can be introduced because of the inability of such models in the absence of appropriate line data to reliably account for the existence of warmer and denser gas components .
this further underscores the value of our observed @xmath87 relations from @xmath3 to @xmath4 in safely determining such departures from linearity and/or in normalization before proceeding towards any interpretation based on ism / sf physics .
the few super - linear slopes of fir@xmath1co luminosity relations for low-@xmath8 co lines that survive careful re - analysis @xcite could be a byproduct of dense molecular gas being the direct sf fuel in all galaxies ( with a constant sfe ) and a @xmath113 ( where x could be total h@xmath18 gas mass traced by co @xmath3 , @xmath61 lines ) that varies within the galaxy sample with @xmath114 .
variations of this simple scenario have been suggested throughout the literature @xcite , and unlike more sophisticated interpretations of such super - linear slopes offered by the two theoretical treatises available on this matter @xcite , the only assumption here is that @xmath114 .
the latter is a well - documented fact as starbursts / mergers , which dominate the high-@xmath41 end , are observed to have larger dense / total gas mass fractions than lower-@xmath41 isolated disks ( e.g. , @xcite ) . within this scheme , the further the gas phase x is from the dense , star forming phase in terms of physical conditions and relevance to the star formation , the higher the @xmath115 value is and the deduced super - linear slope of the @xmath116 ( or corresponding s - k ) relation . on the other hand for galaxy samples with a smaller range of ir - luminosities that have a nearly constant @xmath117 such as the ( u)lirg+dsfg sample considered in this paper ( see [ subsection : beta ] ) a linear slope of an @xmath116 relation can still be recovered for e.g. , co @xmath3 that traces all metal - rich molecular gas mass rather than only the dense sf one .
we return to this point later in our discussion .
as already mentioned in the introduction , the s - k relation , and especially the important one involving the dense gas component in galaxies , is not easily accessible observationally .
so one falls back to the much more observationally accessible proxy @xmath116 relations .
their index can then be linked to that of an assumed underlying s - k relation in galaxies using two theoretical models @xcite .
both models posit the same intrinsic s - k relation of @xmath119 ( or @xmath120 ( for disks of near - constant scale - height ) , justified under the assumption of a constant gas fraction transformed into stars per free fall time ( @xmath121 ) .
the same s - k relation emerges also if the sf timescale is instead set by the dynamical timescale of a marginally toomre - stable galactic disk with sf converting a fixed fraction of gas into stars over such a timescale @xcite .
both models give expected values of @xmath26 versus @xmath122 for @xmath116 luminosity relations over a large range of line critical densities , but both models are applicable only for lines that require low temperatures to excite @xmath123 ( for the @xcite model this limit is @xmath124 ) .
the reasons behind this limitation are explicit assumptions about isothermal gas states at a set temperature @xcite , or the tracking of such states over a small range of @xmath125 @xcite .
the low-@xmath8 lines of heavy rotor molecular line data - sets found in the literature and our low-@xmath8 co lines are certainly within the range of applicability of these models .
the indices of the corresponding @xmath116 power law relations can thus be compared to these theoretical predictions .
@xcite found a linear fir@xmath1hcn correlation for the @xmath3 transition ( see also @xcite ) , which extends from local ulirgs / lirgs ( @xmath126 ) to normal , star forming galaxies ( @xmath127 ) , down to individual galactic molecular clouds with @xmath128 @xcite .
a weakly super - linear fir@xmath1hcn(@xmath32 ) slope ( @xmath129 ) was found by @xcite and @xcite over a combined sample of local normal galaxies and lirg / ulirgs .
a careful analysis by @xcite , however , demonstrated that a bimodal fit ( i.e. , a different normalization parameter @xmath130 ) is better , with each galaxy sample well fit by a linear relation .
finally , a weakly super - linear @xmath131 appears when high-@xmath40 observations of the most ir - luminous starburst galaxies and qsos are included in the locally - established relation @xcite .
this could also be bimodal instead , but with an otherwise linear @xmath132 relation ( and with insufficient high-@xmath41 objects to decide the issue ) . a physical reason for such bimodalities is discussed in
[ subsection : beta ] .
extending hcn observations to include many more objects in the crucial @xmath133 regime is necessary for deciding such issues .
even then one must eventually obtain the underlying sfr@xmath134 relation before arriving at secure conclusions about a varying @xmath135 of the dense gas in ( u)lirgs . the latter is the crucial physical quantity underlying the normalization of such @xmath136 relations and e.g. , a rising or bimodal @xmath137 factor towards high-@xmath31 systems can easily erase purported sfe trends obtained by using single - line proxies of dense gas . a sub - linear fir@xmath1hcn slope ( @xmath138 ) for the @xmath22 transition has been reported @xcite but is very likely biased low due to not having performed any beam correction ( see [ section : samples - and - data ] ) for some of their very nearby extended objects , where the hcn beam does not cover the entire ir emitting region . for these sources , the hcn measurements do not match the ir luminosities , and since they all reside at the lower end of the hcn luminosity distribution , the net effect will be to bias the relation towards shallower values .
for this reason we have chosen to ignore the sub - linear fir@xmath1hcn slopes from @xcite and @xcite . a recent survey of hcn @xmath46 and cs @xmath139 @xcite , and cs @xmath3 , @xmath59 , @xmath27 , and @xmath90 ( zhang et al .
, in prep .
) , towards nearby star forming galaxies ( @xmath140 ) , where such effects have been adequately accounted for , establishes a slope @xmath141 for these transitions ( see also @xcite and @xcite ) .
many of these cs transitions have higher critical densities than hcn , and cs is furthermore less prone to ir pumping effects than hcn is ( cs is pumped at @xmath142 compared to @xmath143 for hcn ) .
pumping of hcn ( and also hnc ) , however , typically only becomes important at dust temperatures @xmath144 @xcite , and would typically require even higher temperatures for cs .
this is important since pumping could affect the cs / hcn luminosities ( especially at high-@xmath8 ) , and thus in principle result in linear ir - cs / hcn relations . in fig .
[ figure : fig2 ] we summarize all the observationally determined @xmath145 and @xmath146 slopes from the literature along with those derived from our co lines .
overall , the data suggest @xmath141 for our @xmath147 relations from @xmath3 up to @xmath9 , @xmath60 and for the heavy - rotor molecular lines .
the latter cover a range of @xmath148 , i.e. , reaching up well into the high - density regime of the star forming gas phase . from fig .
[ figure : fig2 ] it becomes clear that most observations are incompatible with current model predictions ( shown as the grey - shaded area ) both for the heavy rotor and the low-@xmath8 co lines ( where such models remain applicable ) .
super - linear slopes do appear for some co @xmath3 data - sets but then , unlike model predictions , the slopes remain linear for lines with much higher critical densities , including those of mid-@xmath8 co lines @xmath149 ( the fir@xmath1co luminosity relation for @xmath12 is also compatible with a linear one ) . in summary , we conclude that the global @xmath41-@xmath150 relations in ( u)lirgs and dsfgs , as parametrized by @xmath151 , are linear for heavy rotor lines , and our co line data - set up to @xmath12 , at which point the relations become increasingly sub - linear for higher @xmath8 . moreover , the normalization factor @xmath130 shows a similar behavior by being nearly constant @xmath11 up to @xmath152 but then starting to increase systematically with increasing @xmath8 , reaching @xmath153 for @xmath4 . the simplest scenario outlined in
[ subsection : s - k simple ] seems to work both for the low-@xmath8 co and the heavy rotor molecular lines with much higher critical densities . given that our sample solely consists of ( u)lirgs ( i.e. , @xmath6}\ge 10^{11}\,{\mbox{$l_{\odot}\,$}}$ ] ) for which the dense gas fraction ( i.e. , @xmath154 ) is nearly constant ( see discussion in [ section : alpha - and - beta ] ) , linear slopes are to be expected for its low-@xmath8 fir@xmath1co relations . for other samples in the literature ( e.g. , @xcite ) that reach lower ir luminosities ( @xmath155 ) and thus span a wider range in @xmath31 ( over which @xmath156 changes appreciably ) the super - linear slopes of their low-@xmath8 fir@xmath1co relations seen in fig .
[ figure : fig2 ] are also expected . in this simple picture
neither the occasional super - linear nor the linear slope of the @xmath87 low-@xmath8 relations carry any profound ism physics other than more dense gas mass corresponds to proportionally higher sfrs ( a picture also suggested by @xcite and @xcite ) . for dense gas tracer lines
this shows itself directly with @xmath157 relations that always have linear slopes . in this picture
there is actually more ism physics to be found in exploring what sets the value of the normalization parameter @xmath88 rather than the slope of @xmath157 relations . the low- and high-@xmath40 ( u)lirgs studied here are highly dust - obscured galaxies , and radiation pressure exerted by the strong absorption and scattering of fuv light by dust grains could be an important feedback mechanism , possibly setting the value of the normalization ( and ultimately regulating the sf ) .
the maximum attainable @xmath158 ratio of a star forming region before radiation pressure halts higher accretion rates is ultimately set by the eddington limit giving : @xmath159 @xcite .
@xcite expressed the expected @xmath160 and @xmath145 relations in the case of eddington - limited sfrs and found that , for co luminosity tracing only the actively star forming gas , the maximal possible luminosity is given by @xmath161 , where @xmath162 is the rosseland - mean opacity , and @xmath163 is the @xmath99-to-@xmath164 conversion factor .
a similar expression holds for hcn , albeit with different @xmath162 and @xmath165 values ( see @xcite for details ) .
although the exact normalization of this relation for each molecular line depends on poorly constrained quantities like @xmath162 and @xmath165 , the eddington limit set by the strong fuv / optical radiation from embedded sf sites acting on the accreted dust and dense gas can naturally provide the normalization of the observed @xmath147 and @xmath166 relations .
in fact , adopting @xmath167 and @xmath168 , which are perfectly reasonable values for ( u)lirgs @xcite , we find @xmath169 .
we note that the high @xmath162-value ( @xmath170 ) , which corresponds to a three - fold increase in the dust - to - gas mass ratio for the rosseland - mean opacity ( see @xcite for details ) , as might be expected in ( u)lirgs , yields @xmath171 , which is close to the observed normalisation values obtained ( @xmath172 ) for the low-@xmath8 co lines in [ section : analysis ] ( table [ table : fits ] ; see also fig . [ figure : fig3 ] ) . given that the eddington limit is ultimately set within individual sf sites embedded deep inside molecular clouds , it will operate on all galaxies , not just ( u)lirgs . for ordinary star forming spirals
, the global @xmath88 normalization value of the @xmath87 relations for low-@xmath8 co lines will be lower than its ( eddington limit)-set value by a factor approximately equal to the logarithm of its dense gas fraction , i.e. , @xmath173 . by the same token , the offset in the @xmath174 plane between two populations with significantly different dense gas fractions ( @xmath175 and @xmath176 , say )
can be shown to be @xmath177 .
thus , an increasing @xmath178 function can cause the super - linear fir@xmath1co(low-@xmath8 ) relations seen in some galaxy samples ( which in reality is a varying @xmath179 rather than a superlinear @xmath26 ) .
local ( u)lirgs and high-@xmath40 dsfgs on the other hand form stars closer to the eddington limit on a global scale @xcite .
thus , the linear fir@xmath1co ( low - j ) and fir@xmath1hcn / cs relations observed for local ( u)lirgs and high-@xmath40 dsfgs is consistent with the notion that radiation pressure is an important physical mechanism that underlies the observed star formation laws in highly dust - obscured galaxies . in effect , the extreme merger / starbursts that dominate the ( u)lirgs and high-@xmath40 dsfg population resemble dramatically scaled - up versions of dense gas cores hosting sf deep inside giant molecular clouds ( gmcs ) , with the balance between radiation pressure and self - gravity setting their equilibrium during their ir - luminous phase . in this framework the failure of the available theoretical models to account for the observed @xmath87 relations of low-@xmath8 co and heavy rotor molecular lines might be attributed to the role radiation pressure feedback plays in ultimately determining such relations .
this has not be taken into account in all current theoretical considerations that either seek to explain the s - k relation in galaxies ( e.g. , @xcite ) , or use an ( s - k ) relation of @xmath180 to determine the emergent @xmath181 relations for molecular gas @xcite .
these use self - gravity , the associated time - scale of free - fall time , along with models on how the sf efficiency ( gas mass fraction converted into stars per free fall time ) varies per phase in turbulent gas as the main ingredients towards a complete understanding of sf , s - k relations , and the proxy @xmath116 relations .
a non - gravitational force like that exerted by radiation pressure on accreted gas and dust near sf sites can greatly modify such a picture by reducing or eliminating the dependence on the free fall time , especially for the high - density gas ( which presumably is the one closest to active sf sites ) .
alternatively , is has been suggested that in lower luminosity systems the star formation may be regulated by feedback - driven turbulence ( kinetic momentum feedback ) rather than by radiation pressure @xcite . assuming a continuum optical depth at fir wavelengths ( @xmath182 ) of order unity for our sample of starburst / merger ( u)lirgs and typical dust temperatures of @xmath183
, we can make a rough estimate of the expected radiation pressure , namely @xmath184 ( where @xmath185 is stefan - boltzmann s constant and @xmath186 the speed of light ) .
this is comparable to the turbulent pressure @xmath187 , obtained assuming a turbulent velocity dispersion of @xmath188 and an average gas mass density of @xmath189 corresponding to @xmath190 .
both of these greatly exceed the expected thermal pressure @xmath191 ( for @xmath192 and @xmath193 ) thus highlighting the point made above that a complete physically model of the s - k relations has to incorporate the effects of radiation pressure and/or turbulence .
higher than @xmath12 neither the slope , @xmath26 , nor the normalization , @xmath130 , of the @xmath87 relations remain constant but @xmath26 decreases while @xmath88 increases towards higher @xmath8 levels .
this can be understood using a simple argument first put forth ( in a slightly different form than here ) by @xcite .
consider that @xmath194 can be expressed as : @xmath195 where @xmath196 is the slope of the fir@xmath1hcn @xmath3 relation , which as mentioned previously , is near unity .
the last term , @xmath197 , is a convenient parametrization of deviations in @xmath198 from unity , and depends on both the dense gas content ( as traced by hcn ) and the global co line excitation . for co @xmath3
, @xmath199 is a linear proxy of dense gas mass fraction .
this is simply due to the linearity of the @xmath200 relation ( a fit to the @xmath201 } > 10^{11}\,{\mbox{$l_{\odot}\,$}}$ ] sources in the @xcite sample yields @xmath202 ) , and the fact that co @xmath3 provides a good linear measure of @xmath203 .
the same applies also for the @xmath61 line . for the higher @xmath8 co lines @xmath204 becomes similar to that of hcn @xmath3 ( or only slightly surpasses it ) while their @xmath205 ( @xmath206 ) significantly exceed that of hcn @xmath3 ( @xmath207 ) .
the high-@xmath8 co lines are significantly excited ( see [ subsection : co - sled ] ) and , following the argument first made by @xcite , this is unlikely to be a pure density effect , as this would imply too large co @xmath3 and @xmath59 optical depths and , in turn , @xmath208co@xmath209co line ratios well below the typical values ( @xmath210 ) observed for local ( u)lirgs ( e.g. , @xcite ) .
instead , we argue that the high-@xmath8 co lines are produced by a dense and warm ( @xmath211 ) phase .
the @xmath212 then becomes a measure of the @xmath213 modulo gas excitation differences between the dense ( d ) and the dense and warm ( d - w ) molecular gas reservoirs .
the derivative inside the parenthesis in eq.[equation : slope ] will be nearly zero for both low- and high-@xmath8 co lines as long as : a ) the dense gas mass fraction remains nearly constant within our galaxy sample ( i.e. , the sample is homogeneous in terms of @xmath156 and its proxies @xmath199 , @xmath214 ) , and b ) the @xmath215 ratio also remains constant .
the latter means that the relative excitation conditions and mass between the dense gas component ( d ) and its sub - component of dense and warm gas ( d - w ) remain invariant across the sample .
the trend of @xmath87 relations above @xmath12 towards increasing sub - linearity for higher @xmath8 levels is due to a decrease of @xmath212 with increasing high-@xmath8 co luminosity , thus resulting in @xmath216 .
this behavior is indeed obvious in fig.[figure : fig4 ] , which shows @xmath217 as a function of @xmath218 for a sub - set of our local ( u)lirg sample with hcn(@xmath32 ) detections from @xcite .
note , we have not included dsfgs in this plot since most detections of hcn at high redshifts are of qsos and agn dominated dsfgs . in conjunction with eq . [ equation : slope ] , fig .
[ figure : fig4 ] can account for our established fir@xmath1co slopes in fig .
[ figure : fig2 ] and table [ table : fits ] . a decreasing @xmath217 with increasing high-@xmath8 co luminosity ( yielding a negative derivative inside the parenthesis in eq.[equation : slope ] ) indicates an increasing mass and/or excitation conditions of the warm and dense ( d - w ) gas component relative to the dense gas reservoir ( d ) that presumably contains it .
this is possible if galaxies with increasingly larger high-@xmath8 co line luminosities ( and thus also sfrs ) increasingly have a warm and dense gas component no longer tied to their sf via the average fuv / optical radiation field .
such examples have been found for individual starbursts or star forming galactic nuclei @xcite , while the presence of large masses of such a molecular gas component was recently suggested as a general feature of the ism in extreme merger / starbursts @xcite .
high cosmic ray ( cr ) energy densities and/or the dissipation of galaxy - wide shocks due to strong supersonic turbulence can maintain @xmath219 for large amounts of high - density gas even in the absence of fuv radiation fields ( e.g. , @xcite ) .
appreciable fractions of dense gas mass per gmc above such temperatures demand different heating mechanisms that can strongly heat the gas without readily dissociating co as fuv radiation does , and without being attenuated by dust ( i.e. , cr- and turbulent heating ) .
the onset of increasing normalization factors , @xmath88 , of the @xmath87 relations above @xmath12 is then simply another result of the weakening link between the fuv - powered @xmath41 and the thermal state of dense gas for systems with high sfrs ( and high-@xmath8 co line luminosities ) .
the rapid rise of @xmath130 with @xmath8-level is expected if co lines at increasingly higher-@xmath8 levels probe ever higher gas thermal states with smaller mass per ir luminosity .
the above picture retains the simple explanation for the observed @xmath220 for the @xmath87 relations from @xmath3 to @xmath9 , and for those found for several heavy rotor molecular lines - as long as all these relations refer to a near @xmath156-homogeneous galaxy sample , with sf powering both the dust continuum and the molecular line luminosity via fuv radiation .
highly super - linear slopes can only occur for galaxy samples with significantly different dense gas fractions , or different star formation relation normalizations ( e.g. , @xcite ) .
finally , in this overall scheme , and for good ( i.e. , linear ) dense sf gas tracers such as hcn and cs lines it is rather hard to envisage how sub - linear slopes can come about @xcite , since even the high-@xmath8 transitions of these heavy - rotor molecules will trace the dense , cold star forming gas .
thus , the second term in eq . [ equation : slope ] will remain close to zero , leaving the fir@xmath1hcn ( or fir@xmath1cs ) relation linear .
in fact , linear slopes are observed for transitions as high as cs @xmath139 @xcite .
a more direct indication of significant amounts of warm and dense gas in our ( u)lirg - dominated sample , and to what extent its thermal state is likely to be maintained by the sfr - powered average fuv radiation fields , is provided by the co sleds . in fig .
[ figure : fig5 ] we show the fir- and co(@xmath32)-normalised co sleds ( top and middle panels , respectively ) , as well as the raw co luminosities ( bottom panel ) .
the first version allows for an assessment of the co sleds for the full samples ( not all of our sources have co @xmath3 measurements ) , and shows the cooling power of the co lines with respect to the continuum .
the co @xmath3 normalized representation of the co sleds makes for a direct comparison with observed co line ratios in the literature , and is furthermore what is usually used to constrain the excitation conditions of the gas . a detailed analysis of the co sleds , in conjunction with the multi-@xmath8 hcn
, cs and hco@xmath221 line data - sets available for many of the ( u)lirgs in fig.[figure : fig5 ] , is needed for a full understanding of the heating and cooling mechanisms of the molecular gas and for quantifying the relative mass - fractions of the gas phases .
nevertheless , the marked contrast between their co sleds and that of the milky way disk ( where most of bulk of the molecular gas is warmed by photoelectric heating induced by the ambient fuv radiation field ) , already indicates the presence of a different heating source . while intense x - ray radiation fields ( @xmath222 ) generated by agn can penetrate and heat gas up to @xmath223 at column densities of @xmath224 ( and without dissociating all of the co ) , it is unlikely to be the case here since great care has been taken in removing agn from our sample .
the integrated power emitted in the co @xmath139 to @xmath4 transitions for all the ( u)lirgs in our sample , constitutes the bulk ( about 60% ) of the total energy output of all the co lines . exploring the effects of different cr and mechanical heating rates on the thermal structure of clouds
, @xcite found that even for extreme cr fluxes ( @xmath225 the milky way value ) it is difficult to maintain temperatures @xmath223 , and the effect on the high-@xmath8 co lines appears to be minor .
mechanical heating , such as supernova driven turbulence and shocks , however , was found to heat the gas more efficiently , and we favor this as the most likely explanation for hot gas and the ` boosted ' high-@xmath8 co lines observed in our ( u)lirgs . highly excited co sleds have been found for the merger / starburst ngc6240 , where a recent analysis found fuv photons ( and the resulting photoelectric heating ) to be inadequate as the main heating source for the high temperatures of its dense gas ( @xcite ) , and for mrk231 where x - rays from the agn are thought to heat the dense molecular gas reservoir @xcite ( although , mrk231 has recently been shown to be much less x - ray luminous than previously though , see @xcite ) . while such co sleds , and the need for alternative heating mechanisms than fuv - photons to explain them , might be linked to the unusually high co line - to - continuum ratios of both of these two sources , a similar conclusion was reached for m82 and ngc253 based on analyses of their full co sleds @xcite .
our work is the first to demonstrate such highly excited high-@xmath8 co sleds as a near generic characteristic of merger / starbursts ( the galaxies that dominate the sample shown in fig .
[ figure : fig5 ] ) . as mentioned in
[ section : samples - and - data ] incorrect fir@xmath1co relations may be inferred if the fir and co measurements cover different regions within galaxies .
this can be a serious problem for local extended sources where single - dish co beams can be smaller than the extent of the ir emission ( see discussion in @xcite ) .
we are confident that this is not an issue for our hercules sample , where all spire - fts co line fluxes were scaled to a common @xmath226 angular resolution , which is beyond then the extent of the ir emission in these sources ( as traced by laboca @xmath227 maps ) . however ,
if this was not the case , and the spire - fts measurements did not capture all the co emission , it would imply that the derived fir@xmath1co slopes are biased high ( since the co luminosity will be underestimated relative to the total fir luminosity , see fig .
[ figure : fig1 ] ) .
in short , the sub - linear fir@xmath1co slopes at high-@xmath8 transition found here are robust against the ( unlikely ) possibility that some ( small ) fraction of the co emission is unaccounted for .
the effects that the presence of strong agns would have on the sf relations are two - sided . on the one hand it could lead to an overestimate of the ir luminosity attributed to star formation , thus biasing the fir@xmath1co slopes high . on the other hand , agn - dominated environments , where penetrating x - rays may be dominating the gas heating , tend to have ` boosted ' high-@xmath8 co lines compared to star forming regions ( e.g. , @xcite ) .
if the agn was deeply buried it would not be easily detectable in x - rays and would be optically thick in the ir , possibly down to mm wavelengths .
the effect could be just what is observed here a change of slope in the correlation and the high-@xmath8 co lines reflecting a hot , deeply embedded agn . as mentioned in
[ section : samples - and - data ] , we used @xmath228}$ ] instead of @xmath6}$ ] in the fir@xmath1co relations in order to minimize the effects of agn .
more importantly , we only included local ( u)lirgs for which the bolometric agn contribution was deemed to be @xmath71 , as estimated from several mir diagnostics ( [ section : samples - and - data ] ) . in the case of the high-@xmath40 sample , obvious agn - dominates systems were discarded from the sample to begin with ( [ section : samples - and - data ] ) .
furthermore , we note that deep x - ray observations as well as mir spectroscopy of millimeter and sub - millimeter selected dsfgs ( which show no obvious signs of harboring an agn ) have shown that any agn that might be present typically contribute @xmath229 to the total ir luminosity @xcite .
based on the above , we feel confident that neither the fir nor the co luminosities are biased high due to agn , and that therefore our findings are not systematically affected by agn .
galaxies that are gravitationally lensed are prone to differential magnification , an effect in which regions within a galaxy are magnified by different amounts due to variations in their location within the galaxy , and/or spatial extent @xcite .
this can significantly skew the observed relative contributions from hot vs. cold dust to the ir luminosity , as well as low- vs. high-@xmath8 co line luminosity ratios @xcite .
furthermore , a flux - limited sample of strongly lensed sources will tend to preferentially select compact sources @xcite , which may be more likely to have extreme co excitation conditions . from fig .
[ figure : fig5]a , however , there is nothing to suggest that the lensed dsfgs have markedly different @xmath230 values than the non - lensed and local ( u)lirgs . in fig .
[ figure : fig5]b , however , we do see a few lensed dsfgs which have markedly higher @xmath231 ratios at high-@xmath8 than the other samples .
this is exactly what we would expect to see if these sources were differentially lensed , and the high-@xmath8 lines tracing more compact regions than the @xmath3 line , or if the lensing preferentially selects compact sources ( which would tend to have more extreme excitation conditions ) .
the strongest argument against our analysis being affected by differential magnification effects is the fact that the lensed dsfgs make up only a minor fraction of our total number of galaxies . to verify that this was indeed the case we fitted the fir@xmath1co relations without the lensed dsfgs .
this resulted in slopes nearly identical to the ones given in table [ table : fits ] , and fully consistent within the errors .
thus , we conclude that our findings are not affected in any significant way by differential magnification effects .
utilizing _ herschel_/spire - fts observations of a statistically significant sample of 23 local ( u)lirgs , simultaneously covering the co @xmath9 to @xmath4 lines in one single spectrum , and combining these with co @xmath3 , @xmath59 , @xmath27 and @xmath36 , @xmath60 data from our comprehensive ground - based co survey of the same sample , as well for an additional 44 local ( u)lirgs , we have presented fir@xmath1co luminosity relations for the full co rotational ladder from @xmath3 to @xmath4 .
included in our analysis is also a carefully groomed sample of 35 high-@xmath40 lensed and unlensed dsfgs ( spanning the redshift range @xmath232 ) with robust fir and co luminosity measurements . due to their high redshifts many of these sources
have been observed in the mid- to high-@xmath8 co lines from the ground , thus allowing us to extend the mid- to high-@xmath8 fir@xmath1co relations to the highest redshifts . for this data - set of low- and high-@xmath40
merger / starburst dominated galaxies we find linear ( @xmath233 ) fir@xmath1co relations for co @xmath3 up to @xmath9 , and with nearly constant normalization ( @xmath11 ) . in light of the linear star formation relations found for hcn and cs ( e.g. , @xcite , both of which are _ bona fide _
tracers of dense star forming gas , we have shown that our results are to be expected provided the dense gas mass fraction does not change significantly within the sample .
our findings are also qualitatively consistent with models in which the star formation in ( u)lirgs is regulated on a global scale by radiation pressure as these predict linear @xmath234 slopes for any molecule / transition that traces star forming gas in a homogeneous sample ( i.e. , constant normalization ) @xcite . for co @xmath12 and up to @xmath4
we find increasingly sub - linear slopes and higher normalization constants , which we argue is due to these lines effectively being detached from the star formation as they trace gas that is dense ( @xmath235 ) but also radically warmer ( @xmath223 ) than what is typical for star forming gas .
this dense and warm ism component is reflected in the global co sleds of the ( u)lirgs , and indeed of the high-@xmath40 dsfgs , which remain highly excited from @xmath12 up to @xmath4 .
this suggests that star formation powered by fuv radiation fields is unlikely to be responsible for maintaining the gas temperature , but instead alternative heating sources are required .
mechanical heating via shocks / turbulence seems to be the most plausible alternative given its effectiveness ( compared to crs ) at driving the temperatures in clouds to the required levels ( @xmath236 ) .
finally , we note that our derived fir@xmath1co relations are sufficiently tight , especially for the high-@xmath8 lines , that they can predict the expected co line brightness of high-@xmath40 dsfgs , which in turn might be useful for planned observations with alma .
+ the authors gratefully acknowledge financial support under the `` demogas '' project .
the project `` demogas '' is implemented under the `` aristeia '' action of the `` operational programme education and lifelong learning '' .
the project is co - funded by the european social fund ( esf ) and national resources .
trg acknowledges support from an stfc advanced fellowship .
trg was also supported by chinese academy of sciences fellowship for young international scientists ( grant no .
2012y1ja0006 ) .
zyz acknowledges support from the european research council ( erc ) in the form of advanced grant , cosmicism .
we are indebted to p.p .
papadopoulos for extensive discussions and comments on the paper ( _ he hookele waa no ka la ino _ ) .
basic research in infrared astronomy at the naval research laboratory is funded by the office of naval research .
jf also acknowledges support from the nhsc / jpl .
the research presented here has made use of the nasa / ipac extragalactic database ( ned ) which is operated by the jet propulsion laboratory , california institute of technology , under contract with the national aeronautics and space administration .
finally , we would like to thank the anonymous referee for a useful and constructive referee report , which helped improve the paper .
aalto , s. , booth , r. s. , black , j. h. , johansson , l. e. b. 1995 , a&a , 300 , 369 aalto , s. , spaans , m. , wiedner , m. c. , httemeister , s. 2007 , a&a , 464 , 193 ao , y. , henkel , c. , menten , k. m. , et al .
2013 , a&a , 550 , 135 alaghband - zadeh , s. , chapman , s. c. , swinbank , a. m. , et al .
2013 , , 435 , 1493 alexander , d. m. , bauer , f. e. , chapman , s. c. , et al .
2005 , , 632 , 736 akritas , m. g. & siebert , j. 1996 , , 278 , 919 andrews , b. h. & thompson , t. a. , , 727 , 97 aravena , m. , murphy , e. j. , aguirre , j. e. , et al .
2013 , , 433 , 498 baan , w. a. , henkel , c. , loenen , a. f. , baudry , a. , wiklind , t. 2008 , a&a , 477 , 747 bayet , e. , gerin , m. , phillips , t. g. , contoursi , a. 2009 , , 399 , 264 bigiel , f. , leroy , a. , walter , f. , et al .
2008 , , 136 , 2846 blain , a. w. & longair , m. s. 1993 , , 264 , 509 blain , a. w. 1999 , , 304 , 669 bothwell , m. s. , smail , i. , chapman , s. c. , et al .
2013a , , 429 , 3047 bothwell , m. s. , aquirre , m. s. , chapman , s. c. , et al . 2013b , , 779 , 67 bradford , c. m. , nikola , t. , stacey , g. j. , et al .
2003 , , 586 , 891 burgarella , d. , baut , v. , iglesias - pramo , j. 2005 , , 360 , 1413 bussmann , r. s. , narayanan , d. , shirley , y. l. , et al .
2008 , , 681 , l73 bussmann , r. s. , gurwell , m. a. , fu , h. , et al . 2012 , , 756 , 134 bussmann , r. s. , prez - fournon , i. , amber , s. , et al .
2013 , , 779 , 25 calzetti , d. , kinney , a. l. & storchi - bergmann , t. 1994 , , 429 , 582 carilli , c. l. & walter , f. 2013 , , 51 , 105 chary , r. & elbaz , d. 2001 , , 556 , 562 casoli , f. , dupraz , c. & combes , f. 1992 , a&a , 264 , 55 conley , a. , cooray , a. , vieira , j. d. , et al .
2011 , , 732 , l35 cox , p. , krips , m. , neri , r. , et al .
2011 , , 740 , 63 daddi , e. , dannerbauer , h. , krips , m. , et al .
2009 , , 695 , l176 dale , d. a. & helou , g. 2002 , , 576 , 159 danielson , a. l. r. , swinbank , a. m. , smail , i. , et al .
2011 , , 410 , 1687 dopita , m. a. & ryder , s. d. 1994 , , 430 , 163 downes , d. & solomon , p. m. 2003 , , 582 , 37 elmegreen , b. g. 2002 , aspc , 285 , 425 engel , h. , tacconi , l. j. , davies , r. i. , et al .
2010 , , 724 , 233 fixsen , d. j. , bennett , c. l. , mather , j. c. 1999 , , 526 , 207 frayer , d. t. , ivison , r. j. , scoville , n. z. , et al . 1999 , , 514 , 13 frayer , d. t. , koda , j. , pope , a. , et al .
2008 , , 680 , l21 frayer , d. t. , harris , a. i. , baker , a. j. , et al .
2011 , , 726 , l22 freundlich , j. , combes , f. , tacconi , l. j. , et al .
2013 , a&a , 553 , a130 fritz , j. , franceschini , a. , hatziminaoglou , e. 2006 , , 366 , 767 fu , h. , cooray , a. , feruglio , c. , et al .
2013 , , 498 , 338 galametz , m. , kennicutt , r. c. , calzetti , d. , et al .
2013 , , 431 , 1956 gao , y. & solomon , p. m. 2004a , , 152 , 63 gao , y. & solomon , p. m. 2004b , , 606 , 271 gao , y. , carilli , c. l. , solomon , p. m. , vanden bout , p. a. 2007 , , 660 , l93 genzel , r. , tacconi , l. j. , graci - carpio , j. , et al .
2010 , , 407 , 2091 gerritsen , j. p. e. 1997 , phd thesis , rijksuniversiteit groningen garca - burillo , s. , usero , a. , alonso - herrero , a. , et al .
2012 , a&a , 539 , a8 graci - carpio , j. , garca - burillo , s. & planesas , p. 2008a , apss , 313 , 331 graci - carpio , j. , garca - burillo , s. , planesas , p. , fuente , a. , usero , a. 2008b , a&a , 479 , 703 greve , t. r. , bertoldi , f. , smail , i. , et al .
2005 , mnras , 359 , 1165 greve , t. r. , papadopoulos , p. p. , gao , y. , radford , s. j. e. , 2009 , , 692 , 1432 griffin , m. j. , abergel , a. , abreu , a. , et al .
2010 , a&a , 518 , l3 hailey - dunsheath , s. , nikola , t. , stacey , g. j. , et al .
2008 , , 689 , l109 harris , a. i. , baker , a. j. , zonak , s. g. , et al .
, , 723 , 1139 harris , a. i. , baker , a. j. , frayer , d. t. , et al . , 2012 , , 752 , 152 hezaveh , y. d. , marrone , d. p. & holder , g. p. 2012
, , 761 , 20 hodge , j. a. , carilli , c. l. , walter f. , et al .
2012 , , 760 , 11 iono , d. , wilson , c. d. , yun , m. s. , et al .
2009 , , 695 , 1537 ivison , r. j. , smail , i. , papadopoulos , p. p. , et al .
2010 , , 404 , 198 ivison , r. j. , papadopoulos , p. p. , smail , i. , et al .
2011 , , 412 , 1913 ivison , r. j. , swinbank , a. m. , smail , i. , et al . 2013 , , 722 , 137 juneau , s. , narayanan , d. t. , moustakas , j. , et al .
2009 , , 707 , 1217 kelly , b. c. 2007 , , 665 , 1489 kennicutt , jr , r. c. 1989 , , 344 , 685 kennicutt , jr , r. c. 1998 , , 498 , 541 kim , c. g. , ostriker , e. c. & kim , w .- t .
2013 , , 776 , 1 knudsen , k. k. , neri , r. , kneib , j .-
p . , van der werf , p. p. 2009 ,
a&a , 496 , 45 kovcs , a. , chapman , s. c. , dowell , c. d. , et al .
2006 , , 650 , 592 krumholz , m. & thompson , t. a. 2007 , , 669 , 289 leroy , a. , walter , f. , brinks , e. , et al .
2008 , , 136 , 2782 leroy , a. , walter , f. , sandstrom , k. , et al . 2013 , , 146 , 19 lupu , r. e. , scott , k. s. , aretxaga , i. , et al . 2012 , , 757 , 135 makiwa , g. , naylor , d. a. , ferlet , m. , et al .
2013 , , 52 , 3864 mao , r. q. , henkel , c. , schulz , a. , et al .
2000 , a&a , 358 , 433 mao , r. q. , schulz , a. , henkel , c. , et al .
2010 , , 724 , 1336 maraston , c. 2005 , , 362 , 799 meijerink , r. , spaans , m. & israel , f. p. 2007
, a&a , 461 , 793 meijerink , r. , spaans , m. , loenen , a. f. , van der werf , p. p. 2011
, a&a , 525 , a119 meijerink , r. , kristensen , l. e. , wei , a. , et al .
2013 , , 762 , l16 mendez - delmestre , k. , blain , a. w. , alexander , d. m. , et al .
2007 , , 655 , l65 narayanan , d. , groppi , c. e. , kulesa , c. a. , walker , c. k. 2005 , , 630 , 269 narayanan , d. , cox , t. j. , shirley , y. , et al .
2008 , , 684 , 996 neri , r. , genzel , r. , ivison , r. j. , et al .
2003 , , 597 , l116 noll , s. , burgarella , d. , giovannoli , e. , et al .
2009 , a&a , 507 , 1793 ostriker , e. c. , shetty , r. 2011 , , 731 , 41 panuzzo , p. , rangwala , n. , rykala , a. , et al .
2010 , a&a , 518 , l37 papadopoulos , p. p. , van der werf , p. p. , isaak , k. , xilouris , e. m. 2012 , , 715 , 775 papadopoulos , p. p. , zhang , z .- y . ,
xilouris , e. m , et al .
2014 , , 788 , 153 petric , a. o. , armus , l. , howell , j. , et al .
2011 , , 730 , 28 pilbratt , g. l. , riedinger , j. r. , passvogel , t. et al .
2010 , a&a , 518 , l1 planck collaboration , ade , p. a. r. , aghanim , n. , armitage - caplan , c. et al .
2013 , in press .
( arxiv:1303.5076 ) rangwala , n. , maloney , p. r. , glenn , j. , et al .
2011 , , 743 , 94 riechers , d. a. , walter , f. & carilli , c. l. 2007 , , 671 , l13 riechers , d. a. , hodge , j. a. , walter , f. , carilli , c. l. , bertoldi , f. 2011a , , 739 , l31 riechers , d. a. , cooray , a. , omont , a. , et al . 2011b , , 733 , l12 riechers , d. a. , bradford , c. m. , clements , d. l. , et al .
2013 , , 496 , 329 robson , e. i. , ivison , r. j. , smail , i. , et al .
2014 , , in press rosenberg , m. j. f. , kazandjian , m. v. , van der werf , p. p. , et al .
2014 , a&a , 564 , a126 sanders , d. b. , mazzarella , j. m. , kim , d .- c . , et al .
2003 , , 126 , 1607 sanduleak , n. 1969 , , 74 , 47 schmidt , m. 1959 , , 129 , 243 schier , f. l. , van der tak , f. f. s. , van dishoeck , e. f. , black , j. h. 2005 , a&a , 432 , 369 schruba , a. , leroy , a. k. , walter , f. , et al .
2011 , , 142 , 37 scott , k. s. , lupu , r. e. , aguirre , j. e. , et al .
2011 , , 733 , 29 scoville , n. z. & polletta , m. 2001 , aspc , 249 , 591 serjeant , s. 2012 , , 424 , 2429 sharon , c. e. , baker , a. j. , harris , a. i. , thomson , a. p. 2013
, , 765 , 6 shetty , r. & ostriker , e. c. 2012 , , 754 , 2 soifer , b. t. , sanders , d. b. , madore , b. f. , et al .
1987 , , 320 , 238 solomon , p. m. , downes , d. & radford , s. j. e. , barrett , j. w. 1997 , , 478 , 144 solomon , p. m. , downes , d. & radford , s. j. e. , et al .
1992 , , 387 , l55 solomon , p. m. & vanden bout , p. a. 2005 , , 43 , 677 stierwalt , s. , armus , l. , surace , j. a. , et al .
2013 , , 206 , 1 swinbank , a. m. , smail , i. , longmore , s. , et al .
2010 , , 464 , 733 tacconi , l. j. , neri , r. , chapman , s. c. , et al .
2006 , , 640 , 288 tacconi , l. j. , neri , r. , genzel , r. , et al .
2013 , , 768 , 74 teng , s. h. , brandt , w. n. , harrison , f. a. , et al .
2014 , , 785 , 19 thompson , t. a. , quataert , e. & murray , n. 2005 , , 630 , 167 thomson , a. p. , ivison , r. j. , smail , i. , et al .
2012 , , 425 , 2203 van der werf , p. p. , isaak , k. g. , meijerink , r. , et al .
2010 , a&a , 518 , l42 veilleux , s. , rupke , d. s.n . , kim , d .- c .
, et al .
2009 , , 182 , 628 walter , f. , decarli , r. , carilli , c. l. , et al .
2012 , , 486 , 233 wang , j. , zhang , z. & shi , y. 2011 , , 416 , l21 ward , j. s. , zmuidzinas , j. , harris , a. i. , isaak , k. g. 2003 , , 587 , 171 wei , a. , ivison , r. j. , downes , d. , et al .
2009 , , 705 , l45 wong , m. & blitz , l. 2002 , , 569 , 157 wright , e. l. , mather , j. c. , bennett , c. l. , et al .
1991 , , 381 , 200 wu , j. , evans , ii , n. j. , gao , y. , et al .
2005 , , 635 , l173 wu , j. , evans , ii , n. j. , shirley , y. l. , knez , c. 2010 , , 188 , 313 yao , l. , seaquist , e. r. , kuno , n. , dunne , l. 2003 , , 588 , 771 zhang , z .- y . , gao , y. , henkel , c. , et al .
2014 , , 784 , l31 | we present fir[@xmath0@xmath1co luminosity relations ( i.e. , @xmath2 ) for the full co rotational ladder from @xmath3 up to @xmath4 for a sample of 62 local ( @xmath5 ) ( ultra ) luminous infrared galaxies ( lirgs ; @xmath6 } > 10^{11}\,{\mbox{$l_{\odot}\,$}}$ ] ) using data from _ herschel _ spire - fts and ground - based telescopes .
we extend our sample to high redshifts ( @xmath7 ) by including 35 ( sub)-millimeter selected dusty star forming galaxies from the literature with robust co observations , and sufficiently well - sampled fir / sub - millimeter spectral energy distributions ( seds ) so that accurate fir luminosities can be deduced .
the addition of luminous starbursts at high redshifts enlarge the range of the fir@xmath1co luminosity relations towards the high - ir - luminosity end while also significantly increasing the small amount of mid-@xmath8/high-@xmath8 co line data ( @xmath9 and higher ) that was available prior to _
herschel_. this new data - set ( both in terms of ir luminosity and @xmath8-ladder ) reveals linear fir@xmath1co luminosity relations ( i.e. , @xmath10 ) for @xmath3 up to @xmath9 , with a nearly constant normalization ( @xmath11 ) .
in the simplest physical scenario this is expected from the ( also ) linear fir@xmath1(molecular line ) relations recently found for the dense gas tracer lines ( hcn and cs ) , as long as the dense gas mass fraction does not vary strongly within our ( merger / starburst)-dominated sample .
however from @xmath12 and up to the @xmath4 transition we find an increasingly sub - linear slope and higher normalization constant with increasing @xmath8 .
we argue that these are caused by a warm ( @xmath13 ) and dense ( @xmath14 ) gas component whose thermal state is unlikely to be maintained by star formation powered far - uv radiation fields ( and thus is no longer directly tied to the star formation rate ) .
we suggest that mechanical heating ( e.g. , supernova driven turbulence and shocks ) , and not cosmic rays , is the more likely source of energy for this component .
the global co spectral line energy distributions ( sleds ) , which remain highly excited from @xmath12 up to @xmath4 , are found to be a generic feature of the ( u)lirgs in our sample , and further support the presence of this gas component . |
in this note , all graphs are finite , simple and undirected .
the complete graph on @xmath1 vertices is denoted by @xmath2 .
the number of graph vertices in the largest complete subgraph of @xmath0 , denoted by @xmath3 is called the clique number of @xmath0 .
the girth of a graph is the number of edges in its shortest cycle .
a proper ( vertex ) coloring is an assignment of labels or colors to each vertex of a graph so that no edge connects two identically colored vertices .
the smallest number of colors needed for the proper coloring of a graph @xmath0 is the chromatic number , @xmath4 . for a given graph @xmath5 , a graph @xmath0
is called @xmath5-free if no induced subgraph of @xmath0 is isomorphic to @xmath5 . in particular , a @xmath6-free graph is called a triangle - free graph . it is obvious that @xmath7 .
the chromatic and clique numbers of a graph can be arbitrarily far apart .
there are various constructions of triangle - free graphs with arbitrarily large chromatic number .
probably the best known is due to mycielski @xcite . in 1955
, he created a construction that preserves the property of being triangle - free but increases the chromatic number . for more references , see also blanche descartes @xcite , john kelly and leroy kelly @xcite , and alexander zykov @xcite .
erds @xcite with a deeper insight showed the existence of graphs that have high girths and still have arbitrarily large chromatic number by probabilistic means . indeed , exploring the relation between clique number and other properties of graphs such as chromatic number , maximum degree , etc . , is still an active and fascinating research area within mathematics .
there are still a lot of open questions waiting to be solved in this area , such as bounding the chromatic number of triangle - free graphs with fixed maximum degree @xcite , or more generally reed s conjecture @xcite . for more problems
see @xcite .
the insights gained from this literature review indicate that we need a deeper understanding of these spaces .
thus , it is of interest to know new ways of constructing such spaces . in @xcite , compatibility graphs and @xmath0-tucker lemma
were introduced to obtain a new topological lower bound on the chromatic number of a special family of graphs .
indeed , via the bound , a new method for finding test graphs was proposed by the author . in this note , we give a new way , based on compatibility graphs and @xmath0-tucker lemma , for constructing graphs with high chromatic number and small clique . finally , as a corollary of this method
, we give a new proof of the famous kneser s conjecture .
the organization of the note is as follows . in section @xmath8
, we set up notation and terminology , and we repeat the relevant material from @xcite that will be needed throughout the note .
finally , in section @xmath9 , our main results are stated and proved .
in this section we review definitions and results as required for the rest of the note . here and subsequently , @xmath0 stands for a non - trivial finite group , and its identity element is denoted by @xmath10 .
a partially ordered set or poset is a set and a binary relation @xmath11 such that for all @xmath12 : @xmath13 ( reflexivity ) ; @xmath14 and @xmath15 implies @xmath16 ( transitivity ) ; and @xmath14 and @xmath17 implies @xmath18 ( anti - symmetry ) .
a pair of elements @xmath19 of a partially order set are called comparable if @xmath14 or @xmath17 .
a subset of a poset in which each two elements are comparable is called a chain .
a function @xmath20 between partially ordered sets is order - preserving or monotone , if for all @xmath21 and @xmath22 in @xmath23 , @xmath24 implies @xmath25 .
if @xmath26 is a set , then a group action of @xmath0 on @xmath26 is a function @xmath27 denoted @xmath28 , such that @xmath29 and @xmath30 for all @xmath31 in @xmath0 and all @xmath32 in @xmath26 .
a @xmath0-poset is a poset together with a @xmath0-action on its elements that preserves the partial order , i.e , if @xmath33 then @xmath34 . to provide topological lower bounds on the chromatic numbers of graphs ,
hom complex was defined by lovsz .
for a recent account of the theory , we refer the reader to @xcite .
we need the following version of this concept .
let @xmath35 be a graph with vertex set @xmath36 .
for a graph @xmath5 , we define hom poset @xmath37 whose elements are given by all @xmath1-tuples @xmath38 of non - empty subsets of @xmath39 , such that for any edge @xmath40 of @xmath35 we have @xmath41 .
the partial order is defined by @xmath42 if and only if @xmath43 for all @xmath44 .
let @xmath45 be the cyclic group of order @xmath46.the cyclic group @xmath47 acts on the poset @xmath48 naturally by cyclic shift .
more precisely , for each @xmath49 and @xmath50 , define @xmath51 . to find a new bound on the chromatic number of a special family of graphs , a combinatorial analog of the borsuk - ulam theorem for @xmath0-spaces , @xmath0-tucker lemma ,
wsa introduced in @xcite . to recall the lemma
, we need to make some definitions .
consider the @xmath0-poset @xmath52 with natural @xmath0-action , @xmath53 , and the order defined by @xmath54 if @xmath55 . also , let @xmath56 be the @xmath0-poset whose action is @xmath57 , and the order relation is given by : @xmath58 suppose that @xmath1 is a positive integers , @xmath0 is a finite group , and @xmath59 is a map such that @xmath60 for all @xmath61 and all @xmath32 in @xmath62 .
then there exist two sets @xmath63 and @xmath64 such that @xmath65 . throughout this note
, such a pair is called the bad pair .
it is worth noting that , there is a more general result for the case @xmath66 , the cyclic group of prime order @xmath67 , see the @xmath68-tucker lemma of ziegler @xcite .
let us finish this section by recalling the definition of compatibility graph from @xcite .
let @xmath23 be a @xmath0-poset .
the compatibility graph of @xmath23 , denoted by @xmath69 , has @xmath23 as vertex set , and two elements @xmath70 are adjacent if there is an element @xmath71 such that @xmath32 and @xmath72 are comparable in @xmath23 .
in this section we state and discuss the main results of this note . for every graph @xmath5 and @xmath73 ,
the graph @xmath74 is @xmath75-free . for an @xmath46-tuple @xmath76 ,
define @xmath77 .
suppose that vertices @xmath78 form a clique of size @xmath79 in @xmath74 .
without loss of generality assume that @xmath80 for each @xmath81 .
now , by the definition , there are @xmath82 such that @xmath83 , for each @xmath84 .
we claim that @xmath85 for each @xmath86 .
suppose , contrary to our claim , that @xmath87 for some @xmath88 .
since @xmath89 is connected to @xmath90 there is a , @xmath91 such that @xmath92 is comparable to @xmath90 . without loss of generality assume that @xmath93 .
therefore @xmath94 which contradicts the fact that any two distinct entries of @xmath95 are disjoint ( note that , since @xmath91 , @xmath96 which demonstrates that @xmath97 and @xmath98 are two distinct entries of @xmath95 ) , therefore @xmath99 .
now , the proof is completed .
if @xmath100 is a graph homomorphism , we associate it to a new map @xmath101 by sending each @xmath102 to @xmath103 . since a complete bipartite subgraph of @xmath5 is sent by @xmath104 to a complete bipartite subgraph of @xmath105 , the map @xmath106 is a graph homomorphism .
moreover , the construction commutes with the composition of maps and the identity homomorphism is mapped to the identity homomorphism .
so , using the terminology of category theory , one can say that @xmath106 is a functor from the category of graphs to the category of @xmath75-free graphs .
furthermore , we have an obvious graph homomorphism from @xmath107 to @xmath5 by sending each vertex @xmath102 to @xmath108
. this give us the following lower bound on chromatic number .
for any graph @xmath5 , @xmath109 .
recalling that the kneser graph @xmath110 is the graph whose vertices correspond to the @xmath79-element subsets of the set @xmath111 , and two vertices are adjacent if and only if the two corresponding sets are disjoint .
these graphs were introduced by lovsz @xcite in his famous proof of the kneser s conjecture @xcite . in the context of graph theory ,
kneser s conjecture is @xmath112 , whenever @xmath113 . beside the lovsz s proof , there exist several different proofs for kneser s conjecture , see for instance @xcite .
now we are in a position to state and prove our main theorem . for all integers @xmath114 , @xmath115 .
for any @xmath116 and each @xmath117 , define @xmath118 .
also , denote by @xmath119 the @xmath46-tuple @xmath120 where @xmath89 is the set of all @xmath79-subsets of @xmath121 .
now , assume that @xmath122 is a proper coloring of @xmath123 with @xmath124 colors .
we define a @xmath125 coloring .
let @xmath126 . if @xmath127 for all @xmath117 , set @xmath128 where @xmath129 is the first nonzero element in @xmath26 . if @xmath130 and @xmath131 for some @xmath132 , set @xmath133 where @xmath129 is the first nonzero element in @xmath26 such that @xmath134 .
finally , consider the case that all of @xmath130 .
let @xmath135 be the one of @xmath136 , with @xmath137 .
now , set @xmath138 note that , the vertices @xmath139 form a clique of size @xmath46 in @xmath123 , @xmath140 .
thus , @xmath141 takes its values in @xmath142 .
it is easy to check that @xmath143 for all @xmath144 and all @xmath145 .
so , to use @xmath47-tucker lemma , its enough to show that @xmath141 can not have a bad pair .
let @xmath63 , @xmath146 and @xmath147 .
we consider three cases .
* the first case is @xmath148 . in this case
, we have @xmath149 for all @xmath150 and @xmath151 .
these imply that @xmath152 .
thus , @xmath153 . *
the second case is @xmath154 . in this case , the number of @xmath150 for which @xmath130 is equal to the number of @xmath150 for which @xmath155 .
this fact beside the fact that @xmath156 for all @xmath150 , imply that @xmath130 if and only if @xmath155 .
therefore , @xmath153 . *
finally , assume that @xmath157 and @xmath158 .
on the one hand , by the definition of @xmath141 , @xmath159 . on the other hand , since @xmath63 and @xmath158 , the vertex @xmath160 is connected to the vertex @xmath161 which contradicts the proper coloring of @xmath162 . applying @xmath47-tucker s lemma we have @xmath163 , thus @xmath164 .
in fact , the existence of graphs with high chromatic number and small clique is a direct consequence of theorem @xmath165 and theorem @xmath166 . as another application
, one could deduce a new proof of kneser s conjecture .
more precisely , it is well - known that the kneser graph @xmath110 has a coloring with @xmath167 colors ( see ( * ? ? ?
* section3.3 ) ) , thus by theorem @xmath166 and corollary @xmath165 we have @xmath168 in summary we have the following corollary . for all @xmath169 and @xmath113 , the chromatic number of @xmath170 is @xmath167 . in particular , @xmath171 .
this is part of the author s ph.d .
thesis , under the supervision of professor hossein hajiabolhassan .
i would like to express my sincere gratitude to him for the continuous support of my ph.d .
study and related research , for his patience , motivation , and immense knowledge . 10 alishahi , meysam , and hossein hajiabolhassan .
`` on the chromatic number of general kneser hypergraphs . '' journal of combinatorial theory , series b 115 ( 2015 ) : 186 - 209 .
brny , j. `` a short proof of kneser s conjecture . '' journal of combinatorial theory , series a 25.3 ( 1978 ) : 325 - 326 .
daneshpajouh , hamid reza .
`` a topological lower bound for the chromatic number of a special family of graphs . ''
arxiv preprint arxiv:1611.06974 ( 2016 ) .
descartes , blanche . `` a three colour problem . ''
eureka 9.21 ( 1947 ) : 24 - 25 .
erds , paul ( 1959 ) , `` graph theory and probability '' , canadian journal of mathematics,11 : 3438 , doi:10.4153/cjm-1959 - 003 - 9 greene , joshua e. `` a new short proof of kneser s conjecture . ''
the american mathematical monthly 109.10 ( 2002 ) : 918 - 920 .
kostochka , a. v. , degree , girth and chromatic number . combinatorics ( proc .
fifth hungarian colloq .
, keszthely , 1976 ) , vol .
ii , pp . 679696 , colloq
jnos bolyai , 18 , north - holland , amsterdam - new york , 1978 .
kozlov , dimitry .
combinatorial algebraic topology .
springer science & business media , 2007 .
kelly , john b. , and l. m. kelly .
`` paths and circuits in critical graphs . ''
american journal of mathematics 76.4 ( 1954 ) : 786 - 792 .
kneser , martin .
`` aufgabe 360 . ''
jahresbericht der deutschen mathematiker - vereinigung 2 ( 1955 ) : 27 .
lovsz , lszl .
`` kneser s conjecture , chromatic number , and homotopy . ''
journal of combinatorial theory , series a 25.3 ( 1978 ) : 319 - 324 . j. matouek . a combinatorial proof of kneser s conjecture .
combinatorica , 24(1):163170 , 2004 .
matouek , ji . using the borsuk - ulam theorem :
lectures on topological methods in combinatorics and geometry .
springer science & business media , 2008 .
mycielski , jan .
`` sur le coloriage des graphes . ''
161 - 162 .
b. reed ( 1998 ) : @xmath172 , @xmath173 and @xmath174 .
journal of graph theory , vol .
27 , 177 - 212 .
ziegler , gnter m. `` generalized kneser coloring theorems with combinatorial proofs . ''
inventiones mathematicae 147.3 ( 2002 ) : 671 - 691 .
a. a. zykov , on some properties of linear complexes ( russian ) .
sbornik 24 ( 1949 ) 163 - 188 . | in this note , we introduce a new method for constructing graphs with high chromatic number and small clique . indeed , via this method ,
we present a new proof for the well - known kneser s conjecture . borsuk - ulam theorem , chromatic number , @xmath0-tucker lemma , triangle - free graphs |
this is the first in a series of papers about the numerical treatment of collapsing gravitational waves using a new pseudospectral code developed for the problem . in the early @xmath1s critical phenomena
were discovered in gravitational collapse @xcite , in spherical symmetry , with general relativity minimally coupled to a massless scalar field .
one aspect of the finding was that , amazingly , the critical solution dividing the formation of a blackhole from dissipation of the field , was unique , in the sense that if one takes any one parameter family of initial data , with the parameter controlling somehow the strength of the data , and tunes this parameter to the threshold of blackhole formation , one finds that the same solution is always obtained , regardless of the family ! shortly thereafter similar phenomenology was reported in axisymmetric , vacuum general relativity @xcite , or in other words in the collapse of gravitational waves . since then
multiple studies have been performed to reproduce this finding , albeit with different initial data and numerical approaches , but without success . perhaps most strikingly , in @xcite , numerical evidence of a different critical solution was presented .
even if one completely accepts the available evidence for criticality in vacuum collapse , this obviously begs the question whether or not the naive expectation of uniqueness of the critical solution in axisymmetric , rather than spherical , collapse holds .
roughly speaking there are two types of code being used used in 3d numerical relativity .
the first uses the moving puncture method @xcite , which consists , in essence , of a clever choice of evolved variables and gauge conditions , normally treated numerically by finite differencing . secondly is the pseudospectral method , most prevalently used with a first order generalized harmonic formulation of general relativity by the spec code @xcite .
recently , we presented a study of the collapse of gravitational waves using the moving puncture method @xcite , in part to establish how close to the critical regime one can get with this standard approach .
the conclusion being ; not very .
major difficulties included the formation of coordinate singularities and a lack of accuracy .
therefore one would like to tackle the problem using the pseudospectral approach to establish what can be achieved in that setting .
we have thus developed a new pseudospectral code along the lines of spec , specializing the continuum and numerical method as much as possible towards the problem of vacuum gravitational collapse .
the present paper represents the first outcome of this maneuver .
herein we describe the formulation of gr employed , our boundary conditions , the code , calibration of the method , our initial data , our approach to axisymmetric apparent horizons , plus a suite of validation tests for gauge waves , gravitational waves , blackhole and collapse spacetimes . throughout
we compare our results carefully with those in the literature .
we aim to give a body of evidence for the correctness of the method that the reader will find compelling . with this out of the way , in subsequent papers we turn to the problem of critical collapse .
a summary follows before the main text . in section [ section : ghg ]
we look at a slightly modified version of the first order generalized harmonic formulation of @xcite .
we consider constraint preserving , radiation controlling boundary conditions , paying special attention to the constraint preserving boundaries . by considering the reflection of outgoing waves in the linear approximation
we ultimately suggest modified conditions that should reduce spurious reflections caused by the use of constraint damping .
we also suggest alternative gauge boundary conditions .
next , in section [ section : bamps ] , we outline the ` bamps ` code , including our carefully constructed cubed - sphere grids , which avoid clustering of grid - points in unfortunate positions of the domain .
for the discretization we employ a pure chebyschev approach .
we also discuss our ` octant ' symmetry implementation , the crucial patching - penalty approach for communicating data between neighboring coordinate patches , and finally the boundary implementation . in the follow - up section [ section : energypenalty ]
we complete the presentation of the penalty method by computing the penalty parameters appropriate for the semi - discrete system . given the difficulties in the literature in reproducing the results of @xcite it seems necessary to solve the problem in axisymmetry before moving to examine the collapse of fully 3d waves without symmetry . in our moving puncture gauge study
@xcite a major disadvantage in using the bam code was that 3d grids were employed to evolve axisymmetric data . in section [ section :
axi ] we present our approach to evolving axisymmetric spacetimes with the ` bamps ` code , for which we employ the cartoon method @xcite to reduce from the standard ` bamps ` 3d domains to a plane , by using the killing vector to evaluate any angular derivatives .
we discuss various flavors of axisymmetric moment - of - time - symmetry initial data and their numerical construction .
these initial data sets are evolved in a forthcoming study .
we also give a detailed description of our formulation of the apparent horizon conditions in axisymmetry . to the best of our knowledge
this is the first time that the regularity conditions on the symmetry axis have been carefully taken care of .
this is important in later work as the search for apparent horizons will be our key diagnostic tool .
the next three sections ( [ section : experiments]-[section : supercritical ] ) contain a write - up of our development and validation tests .
the tests include evolutions with the proposed gauge boundary conditions , which we find are helpful when using large gauge source parameters , as desired .
they also include runs comparing the fully 3d , octant symmetry and cartoon evolutions , demonstrating that the various symmetry setups are well - behaved . in the evolution of single blackholes we test different gauges and boundary conditions , and following @xcite ,
look at evolutions in which the blackhole is perturbed by a gravitational wave injected through the outer boundary .
our results are in good agreement with the earlier studies .
we then examine the evolution of supercritical brill waves , where , after the formation of an apparent horizon the run is continued after interpolation onto an excision grid , as used to evolve a single blackhole , which is needed to evolve data with a horizon for long - times .
finally we conclude in section [ section : conclusions ] .
[ [ the - evolution - system ] ] the evolution system : + + + + + + + + + + + + + + + + + + + + + we use the first order reduction of the generalized harmonic formulation with several free parameters . the full reduction from the second order einstein equations
is presented in detail elsewhere @xcite so here we give only a brief overview to establish our notation . throughout the paper in continuum equations we use the latin @xmath2 for four dimensional indices , but @xmath3 for spatial indices , with the exception of @xmath4 and @xmath5 , whose meaning when used as indices will be described shortly .
greek indices are used to refer to the position in a state - vector , grid indices , or where otherwise needed .
we start from the vacuum generalized harmonic formulation in second order form , @xmath6\,,\label{eqn : ghg_22}\end{aligned}\ ] ] for the unknown spacetime metric @xmath7 with christoffels @xmath8 .
the constraints of the system are @xmath9 , plus the standard hamiltonian and momentum constraints of gr .
the gauge source functions @xmath10 are freely specifiable , provided that they do not include derivatives of the metric , which would affect the principal part of the pde .
the terms involving @xmath11 are included so as to damp away high - frequency constraint violations @xcite .
the parameters @xmath12 and @xmath13 control whether or not the constraint addition made in the construction of the formulation is done either with the covariant or the partial derivative , or some combination .
the latter choice has the effect of simplifying the constraint subsystem . in the code
we use a first order reduction by introducing the variables @xmath14 and @xmath15 .
the equations of motion are , @xmath16c_c\ , , \label{eqn : ghg11evsys}\end{aligned}\ ] ] with shorthands to be defined momentarily
. the formulation here agrees with that of @xcite except for the inclusion of the @xmath12 and @xmath13 parameters .
we will either take the new parameters to vanish , or choose @xmath17 .
the lapse and shift are denoted @xmath18 and @xmath19 respectively .
the unit normal to the spatial slices of constant coordinate time @xmath20 is written @xmath21 .
when the normal is contracted with a tensor we sometimes use the abbreviation @xmath22 , and likewise for the arbitrary unit spatial vector @xmath23 .
the induced metric on the slice is written @xmath24 . in matrix notation
this system can be written as @xmath25 with @xmath26 , and principal matrix , @xmath27 and @xmath28 containing all source terms .
we use the shorthand for the christoffel symbols under the first order reduction , @xmath29 and will frequently use the abbreviation @xmath30 .
the system is symmetric hyperbolic , having the same principal part as a particular first order reduction of the wave equation .
the characteristic variables are given by , @xmath31 with the projection operator @xmath32 , and speeds , @xmath33 respectively .
for future reference let us also note that a convenient way to transform to the characteristic variables is to write @xmath34 , where here the indices represent the position in the state - vector @xmath35 and where the similarity matrix is , @xmath36 which has left inverse @xmath37 .
but note however that @xmath38 .
the strength of this representation in practical terms is in avoiding special cases in the numerical implementation , like for example @xmath39 , in the characteristic decomposition .
[ [ gauge - source - functions ] ] gauge source functions : + + + + + + + + + + + + + + + + + + + + + + + for the gauge source functions @xmath10 we choose , @xmath40 our convention differs from that of both @xcite and @xcite in a trivial normalization of the spatial part with respect to the lapse function .
writing the resulting gauge conditions in terms of the lapse and shift we get , @xmath41 with @xmath42 the trace of the extrinsic curvature and @xmath43 the contracted christoffel symbol of the spatial metric . before blackhole formation for the scalar functions
@xmath44 we choose , @xmath45 with @xmath46 some constants . by default
we choose @xmath47 and @xmath48 , which naturally maintains the shift damping term even if the lapse function is close to zero , in contrast to the standard condition employed in spec @xcite , which takes @xmath49 .
since we wish to study near - singular gravitational effects in the computational domain and avoid run - away growth of the shift vector this seems reasonable .
we will report in later work on adjustments to these choices when evolving near - critical data .
when evolving blackholes by excision we follow @xcite taking instead @xmath49 , although so far we have not found it necessary to use the @xmath50 form of @xmath51 .
[ [ the - constraint - subsystem ] ] the constraint subsystem : + + + + + + + + + + + + + + + + + + + + + + + + + the first order reduced harmonic constraints are , @xmath52 the terms without @xmath10 are simply @xmath53 . in these variables
the vacuum adm hamiltonian and momentum constraints can be expressed as @xmath54k}+\tfrac{1}{2}d_{j}\phi_{kin } -\tfrac{1}{2}d_{i}\phi_{jkn } \nonumber\\ & \quad\quad-\tfrac{1}{2}\pi_{j[i}\phi_{j]nn}+\gamma^{lm}\phi_{mk[j}\phi_{i]ln}\nonumber\\ & \quad\quad+2\gamma_{an[i}\,\gamma^a{}_{k]j}\big)\,.\end{aligned}\ ] ] as stated above we use a subscript @xmath4 to denote contraction with the normal vector @xmath21 , but with the convention that @xmath55 stands for the partial derivative , but _ with any such contraction outside of the derivative .
_ we can put the hamiltonian and momentum constraints together as a four - vector of constraints , @xmath56 working with the first order system creates the reduction and closely related ordering constraints , @xmath57ab}=0\,.\end{aligned}\ ] ] the constraints @xmath58 and @xmath59 evolve according to , @xmath60\nonumber\\ & \quad-\gamma_1\gamma_2\beta^i\big(\tfrac{1}{2}g^{cd}c_{icd}n_a - c_{ina}\big)\,,\nonumber\\ \p_tc_{iab}&=\beta^j\big(\p_jc_{iab}+\gamma_1\p_ic_{jab}\big ) + \alpha\,\big[(1+\gamma_1)d_ig_{jn}\,c^j{}_{ab}\nonumber\\ & \quad-\gamma_2c_{iab}+\phi^j{}_{ab}\,c_{ijn}+\frac{1}{2}c_{inn}\pi_{ab}\big]\,,\end{aligned}\ ] ] where we have introduced the constraint , @xmath61 and where the notation @xmath62 means take the partial derivative , and afterwards replace all first derivatives of the metric with the reduction variable @xmath14 . up to lower derivatives in the contraints we find , @xmath63 where @xmath64 denotes equality up to non - principal terms , the remainder having been suppressed for brevity .
the equation of motion for @xmath65 is readily derived by taking derivatives of that of @xmath59 .
notice that the parameter @xmath66 serves to damp the reduction constraint . in the description of @xcite the equivalent reduction variable
is called @xmath67 , with , including @xmath12 and @xmath13 in the natural way , @xmath68 in our variables .
the difference is not substantial , being only that @xmath69 appears slightly more naturally in the second order form of the equations .
note that in , the final term contains a piece which is simply the harmonic constraint in the pure harmonic case , but will act as a non - zero coefficient otherwise .
[ [ first - order - reduction - of - the - constraint - subsystem ] ] first order reduction of the constraint subsystem : + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + following @xcite , a first order reduction of the constraint subsystem is formally introduced by defining the new variable @xmath70 with , @xmath71 the principal part of this formal reduction is given by , @xmath72 the characteristic variables of the constraint subsystem are then found to be @xmath73 with speeds @xmath74 and @xmath75 respectively , where we use upper case latin indices to denote those projected by @xmath76 . a suitable norm of the constraint violation is given by the constraint monitor which is defined as @xmath77 [ [ the - gravitational - wave - degrees - of - freedom ] ] the gravitational wave degrees of freedom : + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + in vacuum the weyl scalars @xmath78 can be expressed as , @xmath79\,,\nonumber\\ \psi_4&=m^am^b[\perp^{(p)bd}\!\!\!{}_{ab}k^ak^cr_{abcd}]\,,\label{eqn : weyl_scalars}\end{aligned}\ ] ] respectively .
here we have introduced the null tetrad @xmath80 with @xmath81 and @xmath82 mutually orthogonal unit spatial vectors , and the projection operator , @xmath83 in terms of the first order ghg variables we can express the principal part of the riemann tensor as , @xmath84}{}^i -\gamma^j{}_{b}\p_i\phi_{ja[c}\gamma_{d]}{}^i + n_a\p_i\pi_{b[c}\gamma_{d]}{}^i\nonumber\\ & \quad - n_b\p_i\pi_{a[c}\gamma_{d]}{}^i + \gamma^i{}_{a}\p_i\pi_{b[c}n_{d ] } -\gamma^i{}_{b}\p_i\pi_{a[c}n_{d]}\nonumber\\ & \quad- n_a \gamma^{ij}\p_i\phi_{jb[c}n_{d ] } + n_b \gamma^{ij}\p_i\phi_{ja[c}n_{d]}\nonumber\\ & \quad-\gamma_1\gamma_2 n_an^k\p_kg_{b[c}n_{d ] } + \gamma_1\gamma_2 n_bn^k\p_kg_{a[c}n_{d]}\nonumber\\ & \quad-\gamma_2\gamma^i{}_a\p_ig_{b[c}n_{d ] } + \gamma_2\gamma^i{}_b\p_ig_{a[c}n_{d]}\,.\end{aligned}\ ] ] of course this expression is unique only up to constraint additions .
note that upon contraction with @xmath85 and @xmath86 to form the weyl scalar @xmath87 , and after a single addition of @xmath65 , we naturally form a projection of the incoming characteristic variable @xmath88 .
this is used in the construction of the boundary condition .
the spatial vector @xmath89 is taken to be the unit spatial normal to the boundary .
[ [ boundary - conditions ] ] boundary conditions : + + + + + + + + + + + + + + + + + + + + at the outer boundary we need to control incoming constraint violation , gauge perturbations and physical radiation . by default
we initially impose , @xmath90 on the constraint subsystem assuming that the characteristic variable @xmath91 is always incoming .
these conditions are essentially those of @xcite , with just the additional @xmath92 term .
other conditions for this variable will be motivated and tested in what follows .
the remaining constraint subsystem characteristic variables may or may not be incoming , and are dealt with on this basis as described in section [ subsection : bc_imp ] , but always according to the same prescription . for the gravitational wave degrees of freedom we choose , @xmath93 the lowest order member of a cascade of conditions on incoming radiation @xcite , with given data @xmath94 .
examining it is obvious that this is equivalent to setting , @xmath95 which is in practice how the conditions are implemented . for the remaining gauge degrees of freedom
we choose either the improved gauge boundary conditions of @xcite , @xmath96&\,\hat{=}\,0\,,\label{eqn : g_bcs}\end{aligned}\ ] ] or the alternative , @xmath97+\gamma_2\phi_{scd}\nonumber\\ & \qquad\qquad+r^{-1}(u_{cd}^{\hat{+}}-2n_{(c}h_{d)}+\gamma_2g_{cd})\ , \big]\,\hat{=}\,0\,,\label{eqn : g_bcs_freeze}\end{aligned}\ ] ] with given data @xmath98 , which we will often take to vanish , and where the overbar derivative notation has the same meaning as in equation .
these conditions are similar to the ` freezing ' gauge boundary conditions employed in @xcite , but taking into consideration the discussion of gauge reflections given in @xcite , and constructed so that the conditions are naturally applied to metric components ( in adm form ) and their derivatives , but excluding contributions from the gauge sources .
we will typically try to choose the given data to be fixed in time , and such that initially the time derivatives vanish for these quantities .
here we have introduced the gauge projection operator , @xmath99 the above boundary conditions are implemented in ` bamps ` using the bjrhus method @xcite as in spec .
details of the method are explained in section [ subsection : bc_imp ] . for completeness here
the constraint projection operator @xmath100 is , @xmath101 and also plays an important role in the implementation of the boundary conditions , as they are again naturally written in the form @xmath102 .
[ [ generalized - harmonic - constraint - subsystem ] ] generalized harmonic constraint subsystem : + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + we already saw the constraint subsystem of the first order reduction of the ghg system . but to get a better idea of the effect of the different constraint preserving boundary conditions let us consider now the subsystem without the reduction .
we have , @xmath103 the shorthand @xmath104 and the variable @xmath69 that follows will be related to quantities present in the first order reduction of the ghg formulation shortly .
we can equivalently express this as , @xmath105+(n^b\nabla_bn^c)\big[\nabla_cc_a + 2\gamma_4\gamma^d{}_{ac } c_d\nonumber\\ & -(\gamma_4-\gamma_5)g_{ac}\gamma^d c_d - 2 \gamma_0 n_{(a}c_{c)}\big ] + \gamma^c{}_{ab}n^bc_c\nonumber\\ & + r_{ab}c^b\,,\label{eqn : ft2s_cons}\end{aligned}\ ] ] where the variable , @xmath106 is used to allow for the most convenient form of these expressions , and the final term of is in fact of second polynomial order in the constraints because of the vacuum field equations .
different choices of the constraint addition parameters @xmath107 result in different behavior in terms of growth of the constraint fields .
it is also obvious that different choices of these parameters can simplify the constraint subsystem , the natural choice apparently being @xmath17 .
[ [ linearization ] ] linearization : + + + + + + + + + + + + + + let us linearize and consider the behavior of a set of fields that satisfies these equations on a fixed constraint satisfying background .
we start with equation and use the tetrad consisting of the null vectors @xmath108 defined in to decompose the first index of @xmath109 . from this
we obtain , @xmath110 for the linearization , where we are free to use the notation @xmath58 for the linearized violation because the constraints are satisfied in the background .
[ [ boundary - conditions-1 ] ] boundary conditions : + + + + + + + + + + + + + + + + + + + + taking the standard setup at the outer boundary so that @xmath23 , used in the construction of the tetrad , denotes the outward pointing spatial unit vector normal to the boundary . restricting our attention to boundary conditions that contain at most one derivative of the constraints , _
geometrically _ the most natural choice seems to be @xmath111 in the first order ghg language these conditions are , @xmath112 whereas , discarding the first order reduction , those of are instead , @xmath113 with either conditions one might guess that the choice @xmath114 reduces reflections from the boundary , especially when using a non - harmonic @xmath115 gauge .
incidentally this choice also makes the two conditions almost coincident .
suppose all derivatives of @xmath116 tangent to the boundary vanish , and that the background is flat .
then we can analyze the solutions in a plane wave approximation . [ [ mode - solutions - on - flat - space ] ] mode solutions on flat - space : + + + + + + + + + + + + + + + + + + + + + + + + + + + + + when linearized around flat - space this system takes the form , @xmath117 the right - travelling mode solutions are , @xmath118 with eigenfrequencies , @xmath119 a very desirable property for our boundary conditions would be that they absorb outward going waves perfectly , that is , without reflection . with this motivation
high - order derivative boundary conditions on the gravitational wave degrees of freedom have been studied @xcite , and implemented in the spec code @xcite in order to absorb higher spherical harmonics of the weyl scalar @xmath120 . in the current context absorption
means that outgoing mode solutions , those associated with an @xmath121 , lie in the kernel of the boundary conditions .
this is only the case if we switch off the damping @xmath122 .
since the low order spherical harmonics might be expected to dominate in the gauge and constraint subsystems , optimizing against this phenomena may be more important than using high - order conditions for the gauge and constraint subsystems whilst neglecting the damping terms .
[ [ remainder - of - mode - solutions ] ] remainder of mode solutions : + + + + + + + + + + + + + + + + + + + + + + + + + + + + substituting these mode solutions into the boundary conditions , or the natural geometric conditions , each after appropriate linearization , and expansion at large frequency @xmath123 gives remainders of order @xmath124 , indicating that neither is the optimal that can be obtained by adding source terms to the constraint boundary conditions .
taking instead , @xmath125 the remainder is rather of order @xmath126 .
there is some freedom in expressing these conditions in the first order ghg language , but we choose , @xmath127 the conditions can be similarly rewritten .
a similar analysis can be performed using the pure gauge subsystem presented in @xcite , but we currently find that existing gauge boundary conditions are sufficient for our needs , so we do not present these calculations here .
tests with the various boundaries are presented in section [ section : experiments ] .
having discussed the continuum system in the previous section , we now discuss details of our numerical implementation of the ghg system . for this
we present the ` bamps ` code , which uses a pseudospectral method on cubed - sphere grids .
the basic idea of the code is based on spec @xcite , but in many details , such as the actual grid implementation and the outer boundary treatment , ` bamps ` differences are present .
[ [ grid - types ] ] grid types : + + + + + + + + + + + the numerical domain on which we solve the evolution equations in ` bamps ` is either a cubed - ball or a cubed - shell grid . each type is built up of multiple deformed cubes .
each patch is described by two fundamental overlapping charts . in _ local coordinates _
@xmath128 and @xmath129 it is a rectangular box @xmath130\times[-1,1]\times[-1,1]$ ] . in _ global cartesian coordinates _
@xmath131 and @xmath132 the cubes are transformed and rotated in such a way that when added together they build the desired domain .
we give a detailed description in the following .
the cubed - ball - grid includes the origin and has a spherical outer boundary .
it consists of @xmath133 coordinate patches : the central cube : : : is centered around the origin and ranges from @xmath134 to @xmath135 in the global cartesian coordinate directions .
the transition shell : : : transfers the grid from the inner cube grid to a spherical shell with radius @xmath136 .
it contains six patches .
the outer shell : : : consists of six patches which extends the grid with additional cubed - shells up to the outer grid boundary at @xmath137 .
the cubed - shell - grid is an excision grid , meaning that it does not include the origin .
it is a special case of the cubed - ball - grid , consisting only of the six outer shell coordinate patches .
[ [ cubed - sphere - coordinate - transformation ] ] cubed - sphere coordinate transformation : + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + the coordinate transformation used in ` bamps ` to construct the grids introduced above relies on the so called `` cubed sphere '' construction .
it was introduced in @xcite and first applied in the context of numerical relativity in @xcite . since
then this idea was implemented in multi - patch approaches @xcite .
in contrast to many of the earlier examples , the numerical method of ` bamps ` does not require overlapping grids , which simplifies the discussion . in @xcite ,
the coordinates are constructed by considering great arcs parametrized by equidistant angles .
such angle coordinates are used in @xcite , while @xcite use an intermediate set of coordinates also given in @xcite that does not have the equidistant angle property . in ` bamps ` the latter type of coordinates are employed . the concrete coordinate transformation is the following .
first , the local coordinates of each patch are transformed to temporary global coordinates @xmath138 this patch , which is orientated in positive @xmath139 direction , will later be referred to as the master patch . from here ,
cyclic permutation is used to rotate the patches to their location in the sphere .
the denominator @xmath140 depends on where the coordinate transformation happens . for the patches of the outer shell it is @xmath141 in the transition shell its definition includes a transition function @xmath142
@xmath143 this coordinate transformation is constructed to transition from the inner cube to the outer shells .
note that this transformation is uniform along the 3d diagonals , where the distance between inner and outer shell boundary is smallest .
this significantly improves the time - stepping restriction in the transition shell .
[ [ subpatches ] ] subpatches : + + + + + + + + + + + each coordinate patch can be further divided into subpatches .
subpatches are helpful for increasing resolution , and form the backbone of the parallelism of ` bamps ` .
each master patch can be split into @xmath144 subpatches with coordinates @xmath145\,,\end{aligned}\ ] ] with @xmath146 and @xmath147 . in practice
we ensure subdivisions are made in such a way that subgrids of two neighboring patches match , and that neighboring patches and subpatches share grid - point positions on their respective boundaries .
this is necessary because our current penalty - communication method does not deal with interpolating penalties .
concretely we split the inner cube into @xmath148 subpatches . the transition and outer shell
are divided in @xmath149 or @xmath150 subpatches in the radial direction . for the angular direction
we choose the number of subpatches to be @xmath151 . in fig .
[ fig : bamps_subpatch ] we show a 2d sketch of the ` bamps ` cubed - ball grid subdivided into subpatches . [ [ discussion ] ] discussion : + + + + + + + + + + + it is straightforward to specify a mapping between a rectangular master patch and a cubed sphere , although some book keeping for the different patches and different types of shell transitions is involved .
it may be useful to examine different such mappings in terms of a numerical quality criterion , say the size of the jacobian , and to minimize the distortions associated with the coordinate transformation .
[ [ spatial - discretization ] ] spatial discretization : + + + + + + + + + + + + + + + + + + + + + + + ` bamps ` uses the method of lines with a standard ode integrator to integrate in time .
the right - hand - sides are approximated using a pseudospectral method .
we use a linear transformation to map the local coordinates of each subpatch @xmath152 into a unit cube @xmath153 ^ 3 $ ] .
we discretize the subpatch by choosing gauss - lobatto collocation points in each dimension , for example , @xmath154 with @xmath155 , and similarly in the other directions .
the number of of grid points @xmath156 depends on the patch location in the grid .
the central cube is discretized with @xmath157 points .
the radial directions of the transition and outer shell are filled with @xmath158 and @xmath159 points respectively . the number of angular points we chose to be the same as in the central cube to assure that we have matching grids . in fig .
[ fig : bamps_subpatch ] we show on the right the gauss - lobatto discretization of a subpatch .
[ [ basis - expansion ] ] basis expansion : + + + + + + + + + + + + + + + + on the collocation points we expand all evolution fields @xmath160 in each dimension in a spectral basis using chebyshev polynomials @xmath161 @xmath162 and analogously in the remaining two directions .
we use the pseudospectral approach and store not the expansion coefficients @xmath163 but the function values @xmath164 at the collocation points @xmath165 .
[ [ derivatives ] ] derivatives : + + + + + + + + + + + + the spatial derivatives of the evolution fields are computed by a matrix multiplication .
for example in the @xmath166-direction we have @xmath167 with the gauss - lobatto derivative matrix , @xmath168 where @xmath169 at boundary points and @xmath170 elsewhere . in practice
we do not compute diagonal terms of the derivative matrix by the analytic formulas stated above but use the identity @xmath171 this _ negative - sum - trick _ maps a constant function explicitly to zero and is known to give the derivative matrix better stability as regards rounding errors @xcite . in preliminary experiments we found that this gives slightly more accurate derivatives , but have not studied the influence on the accuracy of the simulations presented later in the paper . [
[ filtering ] ] filtering : + + + + + + + + + + we find that a crucial ingredient for numerical stability is the use of a filter against high - frequency growth . for this we follow @xcite exactly .
after every full time - step we apply the filter in each dimension .
the filter is easily implemented as a matrix multiplication .
for example , in the @xmath166 direction , we filter the function values by , @xmath172 with the filter matrix @xmath173 where @xmath174 and @xmath175 and @xmath176 are the chebyshev synthesis and analysis matrices respectively .
[ [ time - integration ] ] time integration : + + + + + + + + + + + + + + + + + we integrate the fields forward in time using a 4th order runge - kutta scheme . unless otherwise stated we fix the time - step , @xmath177 , with @xmath178 being the minimal cartesian spatial grid spacing of the whole domain .
empirically we find that this choice for the time step always leads to stable numerical evolutions , in the sense that increasing resolution results in smaller errors .
we have not not carried out a stability analysis of the fully discrete system .
[ [ bamps - octant - grid ] ] bamps octant grid : + + + + + + + + + + + + + + + + + + + when evolving octant symmetric data in ` bamps ` , it is possible to only evolve one eighth of the cubed ball grid .
this saves computational and memory costs . in the ` bamps ` octant mode we choose an odd number of subpatches @xmath179 and a odd number of grid points @xmath180 and
reduce the numerical domain to @xmath181 , @xmath182 and @xmath183 .
this means that all subpatches containing one of the cartesian axes are cut in half along them . in these patches we use the symmetry conditions to construct special matrices which compute the derivatives and filters . to glue all subpatches together
, we have to impose appropriate conditions at the connecting boundaries of the subpatches .
for this we apply the penalty method as is described in @xcite .
the main idea of this method is to add penalty terms for each incoming characteristic variable at the boundary to the right hand side of the evolution equations .
we use the characteristic variables of the evolution system to formulate boundary conditions .
on the boundary surface we define the outward pointing spatial normal vector @xmath89 .
the characteristic variables of the evolution system are given in equation ( [ eqn : ghg11_char_vars ] ) with speeds ( [ eqn : ghg11_char_speeds ] ) . in vector notation
we write @xmath184 incoming characteristic variables to the subpatch boundary have positive speeds . on these
we want to impose the condition that they are equal to the outgoing characteristic variables of the neighboring patch .
table [ tab : in_out_char ] summarizes all incoming and outgoing characteristic depending on the lapse function @xmath18 and the shift in @xmath89 direction , @xmath75 .
[ cols="^,^,^,^,^,^,^",options="header " , ] where the numbers in parentheses denote the number of those grids that were cut in half ( at the axis ) once , twice , or three times respectively , for the 3d grids , and once or twice for the cartoon grids .
note that our current non - octant cartoon implementation is not optimal because we evolve the whole @xmath139-@xmath132 plane , wasting effectively a factor of two .
currently we use the code most often in cartoon octant mode , so fixing this does not have a high priority . looking at the table
the main observation is that the expected reduction factor of eight ( four ) in the total number of grid points is present between the 3d ( cartoon ) and octant grids , but that this number is not so closely reflected in the grid breakdown , where we get only a factor six ( three ) in the total number of grids .
this is obviously because there are many grids with fewer points . since our parallelization
does not take this fact into account , it is possible that one mpi process is given _ all _ non - cut grids , and so we can expect that the speedup rate is determined to a large extent by ratio in the number of grids .
as we make the domain larger the relative number of cut grids decreases , so we might expect that asymptotically the full speedup factors of eight or four can are attained .
[ [ basic - dynamics-1 ] ] basic dynamics : + + + + + + + + + + + + + + + although irrelevant for the octant cartoon comparison , since these data have not been used before , we give a brief description of their evolution .
initially the peak of the kretschmann scalar occurs at @xmath185 with a value @xmath186 .
this profile then oscillates about three times at the origin , attaining a peak value of @xmath187 before rapidly dispersing .
looking at the lapse we see the familiar behavior that at the origin it oscillates slightly before presenting a longer decrease , although at the minimum is only @xmath188 , having started from @xmath189 everywhere .
afterwards this pulse propagates out , roughly following the disturbance in the kretschmann .
looking at the shift component @xmath190 along the @xmath139-axis we find that early on there is a growth which peaks at @xmath191 , with value @xmath192 .
the development of the shift looks more like a slowly oscillating standing wave than a localized propagating feature .
[ [ d - octant - cartoon - and - octant - cartoon - comparison ] ] 3d , octant , cartoon and octant - cartoon comparison : + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + taking first the 3d and octant evolutions , we see near perfect agreement throughout the evolution
. there are small differences however , starting from the beginning of the simulation at the level of round - off ; differences of @xmath193 in metric components , which slowly drift as the evolution goes on .
this behavior is expected because the derivative approximation differ at this level .
similar differences were found between the other setups .
these differences are never larger than the constraint violation , in for example @xmath194 , and we have looked at convergence ( see section [ section : convergence ] for more discussion ) with each setup , although not for this data , and find no indication of a problem . for the speed comparison we ran the code with each setup on 24 cores ( with hyperthreading ) of our local cluster _ core12 _ with intel xeon x5650 processors . the octant run was a little more than @xmath195 times faster than the 3d run , as expected given the foregoing discussion .
the octant cartoon run was about @xmath196 times faster than the pure cartoon test , which is a little disappointing .
going from @xmath197 to @xmath198 radial subdivisions in the outer shells , this value increases to @xmath199 , demonstrating the expected dependence . comparing the full 3d and octant cartoon runs
, there was a gratifying speed up of nearly a factor @xmath200 .
the ` bamps ` numerical method gives us two options for increasing resolution .
the first is to add grid - points in every domain , the second is to subdivide grids further , keeping the number of points inside each subpatch fixed . given fixed finite computational resources it is not obvious what is the optimum strategy to achieve the smallest possible error , because although we would expect that adding points brings spectral convergence , it also comes with a @xmath201 dependence in the allowed time - step , whereas on the other hand , as we will see , adding more subpatches allows the code to scale up to a large number of processors .
probably the optimal strategy relies on a balance between each . to examine the effect of each strategy in the simplest possible way , we evolved gauge wave initial data on the minkowski spacetime , which was setup by choosing @xmath202 $ ] , @xmath203 , with @xmath204 as usual , and otherwise the flat spatial cartesian metric and vanishing extrinsic curvature .
the results are plotted in the four panels of fig .
[ fig : bamps_conv ] and confirm our expectations .
+ to demonstrate the necessity of the filter we evolved a centered @xmath205 brill wave .
the results are plotted in fig .
[ fig : bamps_filter ] . in the left panel
we see that without filtering the constraint violation starts to grow exponentially in time , whereas with filter the growth is completely absent and the norm of constraints remains steady at a very low value . in the right panel we plot the magnitude of the fourth highest spectral coefficient of @xmath206 in the transition shell as a function of time .
this coefficient is the first that is directly unaffected by the filter .
we see that the growth in the constraints seems to be associated with an explosion in the higher spectral coefficients .
interestingly we tried the same experiment with gauge wave initial data , and did not see the effect , at least in the same time - frame .
we expect that the same behavior would manifest if we were to evolve long enough .
the obvious conclusion we draw from this is that it is important to test these methods with several data types to get a reliable picture of their properties .
[ [ strong - scaling ] ] strong - scaling : + + + + + + + + + + + + + + + the current ` bamps ` parallelization strategy is to obtain perfect scaling using many subpatches , and splitting these subpatches across many processors .
the key is that , contrast to buffer zones required in the decomposition of a finite differencing grid , only @xmath207d surfaces of points need be passed by network communication , making the relative time spent there negligible . in a finite differencing approach the relative size of the buffer zones decreases with resolution , but in practice can still be significant in production runs . in fig .
[ fig : bamps_scaling ] we present strong scaling plots performed on the _ supermuc _ cluster located in lrz garching , with intel xeon e@xmath208-@xmath209c processors .
we ran the code in @xmath210d .
we took a grid with @xmath211 total subpatches , and increased the number of cores used until we were computing one patch per core .
we find perfect scaling .
on the other hand ` bamps ` is currently not parallelized whatsoever at the subpatch level , which means that the maximum number of points per subpatch is in principle determined by the amount of memory available to one core .
at least when running the code in cartoon mode however we do not find , in practical terms , that this is problematic .
instead the @xmath201 restriction in the time step makes increasing the number of points infeasible long before we are close to filling the available memory . in @xmath210d
this may no longer be the case .
we leave such considerations for future work . on the supermuc cluster .
here a grid with @xmath212 sub patches was used . in total
this grid consists of @xmath211 patches .
[ fig : bamps_scaling ] ]
the main thrust of our development has been towards treating collapsing axisymmetric gravitational waves accurately . for super - critical data
the cubed - ball grid is unsuitable after the formation of an apparent horizon .
therefore the strategy for long - term evolution is to take the data and interpolate onto a cubed - shell grid , with the excision surface suitably positioned , changing the lapse and shift to be sure that the excision surface is a true outflow boundary .
a necessary requirement is to treat a single blackhole , which is what we discuss here .
[ [ kerr - schild - coordinates ] ] kerr - schild coordinates : + + + + + + + + + + + + + + + + + + + + + + + + we evolve the schwarzschild solution in kerr - schild coordinates as was done with an earlier version @xcite of the present code . although the current numerical method is not particularly close to that used previously , some components of the older code were inherited . importantly evolving this data allows a simple comparison with the previous method and results . in
spherical polar coordinates the metric and extrinsic curvature take the form , @xmath213 with @xmath214 the flat metric on the two - sphere , and @xmath215\,,\end{aligned}\ ] ] respectively . inside the code the line - element is written in cartesian coordinates in the standard way .
more discussion of kerr - schild coordinates can be found in @xcite .
[ [ harmonic - killing - coordinates ] ] harmonic killing coordinates : + + + + + + + + + + + + + + + + + + + + + + + + + + + + + we additionally evolve starting from the harmonic killing slicing described in @xcite , which serves as a convenient starting point when transitioning from one generalized harmonic gauge to another . for this initial data , in spherical polar coordinates ,
the metric and extrinsic curvature are @xmath216 and @xmath217 with the remaining components vanishing . for this data
spatially harmonic coordinates are obtained by building cartesians according to , @xmath218 the resulting metric has a coordinate singularity at @xmath219 , with @xmath220 implicitly defined in the obvious way from the new coordinates .
the coordinate singularity is not a principle problem as we could just put the excision surface outside this radius .
but ` bamps ` relies on standard cartesian coordinates in several places .
so in the code we could transform in the standard way but then choose the gauge source function , @xmath221 with @xmath222 the jacobian between the standard @xmath223-index cartesians and harmonic cartesian @xmath224 index coordinates , the compound object @xmath225 is defined by , @xmath226 with @xmath227 and where indices are manipulated in the obvious way with @xmath7 to obtain .
instead we just choose the gauge source function to be fixed at its initial value , as will momentarily be discussed .
in this section we use the code exclusively in cartoon mode , on a cubed sphere grid .
we start with the excision surface at @xmath228 , and the outer boundary at @xmath229 . in our base setup
we take @xmath230 radial subpatches each with @xmath231 points per direction .
the runs were performed on a desktop machine with an eight - core intel i7 cpu , which was able to compute at about @xmath232 , the base run requiring about @xmath233 mb of ram .
[ [ killing - gauge - sources ] ] killing gauge sources : + + + + + + + + + + + + + + + + + + + + + + given initial data which admit a time - like killing vector , we can ensure that the evolution of the system is trivial , at the continuum level , neglecting the effect of outer boundary conditions , by choosing the killing lapse and shift , and taking the gauge source functions @xmath10 so that @xmath234 initially .
in particular we must choose , @xmath235 [ [ kerr - schild - evolutions - with - spec - ghg ] ] kerr - schild evolutions with spec ghg : + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + we began by evolving the kerr - schild initial data with the standard formulation parameters of @xcite , namely @xmath236 and @xmath237 on our base grid as just described , using the gauge boundary conditions .
immediately we see that the innermost subpatch has the largest constraint violation , peaked at around @xmath238 in the @xmath194 component of the harmonic constraint .
this is not surprising because the innermost subpatch contains the part of the solution with the largest derivatives .
the evolution successfully continues until the final time @xmath239 .
but after the initial expansion to @xmath238 , a slow expansion in @xmath194 is visible , and this growth becomes more rapid as the simulation continues . by the end , the maximum value of @xmath194 is around @xmath240 , with peaks appearing at the inner and outer boundary of roughly the same size .
we then increased resolution from the base grid to @xmath241 and @xmath242 .
the @xmath243 point grid runs at about @xmath244 , and the initial peak in the @xmath194 constraint violation is reduced by a factor of about two , with this ratio of improvement slowly declining until the end of the evolution .
the @xmath245 grid runs at @xmath246 , with both the initial magnitude of the violation and the ` slow expansion ' of the @xmath194 constraint quashed , the peak being a factor @xmath247 smaller than in the base run at the end of the simulation .
the highest resolution @xmath242 point grid ran at @xmath248 , with the final improvement in @xmath194 against the base run being a factor of @xmath249 .
since the largest constraint violation occurs in the excision subpatch an obvious question is whether or not the excision and outer boundaries would interact badly if they were on the same grid .
although the issue is of little practical concern for production runs , for development it deserves a little attention , and therefore we evolved our base grid from before , but cutting the outer two subpatches so that the outer boundary lies at @xmath250 .
this test is not completely fair because the outer boundary conditions are expected to perform better as they are applied further out .
we find that the initial peak in the violation of the @xmath194 constraint is about five times greater than in the base run at @xmath251 . at the end of the evolution again at @xmath239 by coincidence the constraint violation in the restricted domain is smaller , but this is just because the slow oscillations in each simulation are out of phase . [
[ kerr - schild - incoming - wave - evolutions - with - spec - ghg ] ] kerr - schild incoming wave evolutions with spec ghg : + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + next we evolved the same initial data and gauge , but this time with the same domain as in fig .
@xmath210 of @xcite . to do this
we took @xmath252 radial subpatches , with the same base resolution as before , so that the outer boundary is placed at @xmath253 .
we similarly specify exactly the same given data for an incoming gravitational wave as in that study , taking in particular , @xmath254 with the vectors here the coordinate vectors defined in the obvious way .
we take , @xmath255\,,\end{aligned}\ ] ] with @xmath256 , @xmath257 and @xmath258 . in fig .
[ fig : incoming_wave ] we show the results from these experiments , obtained with a sequence of different resolutions .
we plot the weyl scalar @xmath120 , averaged over the outer boundary , @xmath259 the surface area of the outer boundary is @xmath260 .
fitting the highest resolution data between @xmath261 and @xmath262 we find a ring - down frequency of @xmath263\sim0.372 $ ] as expected @xcite .
in this evolution we found that the apparent horizon oscillates slightly as the gravitational wave is absorbed , increasing the horizon mass by about @xmath264 , with @xmath265 the adm mass of the analytic initial data .
note that the gauge boundary condition being employed here is not identical to that used in @xcite , so the agreement is remarkable .
the effect of much larger pulses of gravitational radiation falling onto a blackhole using similar gauge conditions was studied in @xcite . in the outer boundary in the evolution of schwarzschild perturbed by a small gravitational wave injected through the boundary . in the left panel
we see convergence of the constraints as resolution is increased . at lower resolutions
a drift is present in the ring - down .
there is good agreement with fig .
@xmath210 of @xcite , and the ring - down frequency agrees well with the analytical computation @xcite . at the end of the test
there is some disagreement with @xcite , but since square - roots of very small quantities are being taken we expect this is caused by round - off error .
it seems that on the cubed - sphere grid more resolution is needed to obtain clean results than with the spherical harmonic discretization used in @xcite .
this is perhaps not surprising , since the latter discretization is well - suited to the given data .
[ fig : incoming_wave ] ] [ [ discussion - of - and - comparison - with ] ] discussion of and comparison with @xcite : + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + the prior ` bamps ` study focussed on obtaining numerical stability in the evolution of a single schwarzschild blackhole with the kerr - schild slicing .
the numerical method used a chebyschev - fourier - fourier spatial discretization on a single shell with a spin weighted spherical harmonic filter to prevent high frequency growth of the error . in that study
the outer boundary condition employed simply fixed the incoming characteristic variables to some given data , namely their initial values .
this approach is possible only when the analytic solution is known , otherwise incoming constraint violations are generated .
placing the inner boundary at @xmath228 and the outer boundary at @xmath266 , very long evolutions , say until at least @xmath267 , could be performed with little resolution , in accordance with @xcite . on the other hand , using this method , the naive boundary conditions rapidly deteriorated as the outer boundary was pushed out , and , crucially resolution did not help but rather made the problem worse .
a possible explanation for the latter effect is that no filter was being applied in the radial ( chebyschev discretized ) direction , which have already seen is a crucial ingredient for stability with the current method .
the likely cause of the boundary problem is that , as explained in @xcite , boundary conditions that just freeze the incoming ghg characteristic variables are orders of magnitude more reflecting than the sommerfeld like choice contained in .
evidence for this is obtained in the current code by changing from the gauge boundary condition to use instead , @xmath268&\,\hat{=}\,0\ , , \label{eqn : g_bcs_ruined}\end{aligned}\ ] ] evolving once more the kerr - schild initial data on the base grid . placing the outer boundary further out then results in greater reflections .
however rather than trying to improve a condition only suitable for evolving known data , we immediately moved to the constraint preserving , radiation controlling conditions , with which this issue is completely absent .
the first attempted implementation of a regular center in the ` bamps ` code was to use the chebyschev - fourier - fourier discretization with a double covering in the radial direction , similar to that employed in @xcite .
the approach was not successful , as we always eventually found irregularities in the numerical solution at the origin .
an exponential filter was applied to the chebyschev coefficients in the radial direction , but to little effect .
eventually we settled on the cubed sphere approach , in part because of the expectation that they will later be more convenient for mesh - refinement .
other possible solutions to the problems we faced would be to use one - sided jacobi polynomials as in spec @xcite or to employ a filter that projects the solution in another basis onto these polynomials .
[ [ kerr - schild - evolutions - with - simplified - constraint - subsystem ] ] kerr - schild evolutions with simplified constraint subsystem : + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + using our standard choice for the formulation parameters @xmath17 , and taking @xmath269 , returning to our base resolution from the tests with the spec version of ghg , we find that by @xmath262 the @xmath194 constraint is about @xmath208 times larger than that we obtained before , and by the end of the simulation the new run has accrued a @xmath194 constraint violation with a sharp peak at the outer boundary of order @xmath270 .
this result seems to be in contradiction to those of section [ section : c_ex ] , until we remember that there the gauge boundary condition was employed .
increasing the constraint damping to @xmath237 , the initial violation is comparable to the spec ghg evolution previously described throughout the evolution , and the spike at the outer boundary is suppressed by roughly an order of magnitude . at the end of this run the maximum of the @xmath194 constraint occurs at the excision boundary with a value around @xmath240 .
this experiment thus highlights that the choice of the damping parameters and boundary conditions can be rather subtle . .
in the upper panel we plot the logarithm of the constraint violation @xmath194 . in the latter case
the violation is greatly reduced . in the lower two panels we show the lapse and shift ;
the drift present when using is practically absent with .[fig : bh_bndy_cmp],title="fig : " ] . in the upper panel
we plot the logarithm of the constraint violation @xmath194 . in the latter case
the violation is greatly reduced . in the lower two panels we show the lapse and shift ;
the drift present when using is practically absent with .[fig : bh_bndy_cmp],title="fig : " ] [ [ kerr - schild - evolutions - with - alternative - boundary - conditions ] ] kerr - schild evolutions with alternative boundary conditions : + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + next we returned to the base grid , and switched to the alternative gauge boundary conditions , with @xmath17 and @xmath237 .
we find that the aforementioned growth in the constraints is completely eradicated , and the drift in the lapse and shift is also suppressed .
evolving the same data with the same formulation and gauge boundary condition , but using the modified constraint preserving boundary condition gives almost identical results .
using instead the reflection reducing conditions we see a small improvement in the violation throughout the simulation . repeating the experiment with the incoming gravitational wave injected through the boundary with the standard constraint preserving condition and the gauge boundary conditions , the growth visible in fig . [
fig : incoming_wave ] is also completely absent , even on the base resolution @xmath231 grid .
[ [ harmonic - killing - slice - evolutions ] ] harmonic killing slice evolutions : + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + we now returned to our base grid and resolution , taking the formulation parameters @xmath17 , and @xmath237 , evolving the harmonic killing slice with the gauge boundary condition .
the test successfully runs to @xmath239 . comparing with the equivalent evolution of kerr - schild data , we see that initially near the excision boundary the @xmath194 constraint violation is significantly greater in the harmonic killing test . by @xmath251
this difference has accrued to around two orders of magnitude . later however , as the violation in the kerr - schild killing evolution starts to grow , it overtakes that of the harmonic killing evolution . at
@xmath239 the peak of the constraint violation in the harmonic killing run is about an order of magnitude smaller than in the earlier test . as remarked before , in the kerr - schild test
the inner and outer boundaries have roughly the same magnitude in the @xmath194 constraint violation .
interestingly the twin peaks are not present in the harmonic killing data because the outer boundary is hugely improved .
this finding is consistent with the gauge wave tests presented in section [ section : exp_g_bcs ] , although this test is somewhat easier for the gauge boundary conditions because of the complete lack of dynamics present in the gauge wave test . in the harmonic killing evolution
we are evolving with pure harmonic slicing , and some non - zero spatial gauge source functions , which suggests perhaps that the growth at the outer boundary is predominantly caused by the use of a non - trivial gauge source function for the lapse function , as it interacts with the boundary .
indeed looking once more at the lapse function towards the end of the kerr - schild evolution we see that it is drifting from its initial value , but that this effect converges away with resolution . in any case , the peak in the constraint violation at the outer boundary in the killing kerr - schild data is suppressed as the outer boundary is placed further out .
[ [ harmonic - killing - slice - with - gauge - perturbation ] ] harmonic killing slice with gauge perturbation : + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + a desirable property for a set of dynamical coordinates is that in the presence of a , perhaps approximate , timelike killing vector they quickly asymptote to a time - independent state . for an arbitrary physical or gauge perturbation
there is no hope that this will occur , and nor can any finite set of numerical experiments prove that that there is a basin of attraction to a stationary state
. we can however look for some indication of this behavior . to do
so we start by taking the initial data for the killing harmonic coordinates , and then perturb the initial lapse function by gaussian as in the previous gauge wave evolutions . in terms of the first order ghg variables
this is a slightly fiddly procedure , as compared with the use of lapse , shift and spatial metric , so we give a quick summary : * set spatial metric and extrinsic curvature from the exact solution . *
take the killing lapse and shift .
use the conditions @xmath271 and @xmath272 to set the gauge source functions @xmath10 . *
add the desired perturbation to the lapse ( or shift ) and then transform to the first order ghg variables .
we perturbed the lapse by a gaussian , @xmath273\,,\end{aligned}\ ] ] with @xmath274 and @xmath275 .
a similar experiment was made in @xcite , but starting from a maximal slice of schwarzschild to test the gauge driver system .
we find that the perturbation in the lapse propagates away , rapidly leaving behind the solution with the harmonic killing data with unperturbed spatial coordinates , or at least negligibly perturbed .
the greatest danger to the evolution is probably that the excision boundary fails to be outflow , but at least with this perturbation that does not occur .
[ [ harmonic - evolutions - with - incoming - gravitational - wave ] ] harmonic evolutions with incoming gravitational wave : + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + giving the same gravitational wave data as previously , evolving with the standard boundary conditions and but using the harmonic killing gauge source functions .
it is not obvious how , if at all the spacetime computed is related to that considered before , but in any case we find a very similar decay in @xmath120 .
remarkably the growth present in fig .
[ fig : incoming_wave ] is absent even in this low resolution @xmath231 test .
[ [ the - transition - function ] ] the transition function : + + + + + + + + + + + + + + + + + + + + + + + + as elsewhere , we follow @xcite to transform from one generalized harmonic gauge @xmath276 to another @xmath277 . the composite source function is simply , @xmath278\,h^2_a\,.\end{aligned}\ ] ] the transition function is , @xmath279 in the following experiments we choose @xmath280 and @xmath281 .
note that care must be taken to construct the time and space derivatives of @xmath10 with the transition function .
this choice results in gauge source functions that are only @xmath282 at @xmath283 , which could be avoided with a different transition function .
it is not clear if this finite differentiability will have a large effect on extracted physical quantities from a simulation .
[ [ kerr - schild - initial - slice ] ] kerr - schild initial slice : + + + + + + + + + + + + + + + + + + + + + + + + + + for our first phase - in test , we started with the kerr - schild slicing of schwarzschild and evolved with @xmath17 and @xmath237 , on our base resolution grid .
we took the gauge boundary condition and the constraint preserving condition ( including a @xmath92 term ) .
we used the wave gauge parameters @xmath284 and @xmath285 .
the value of @xmath286 here is much smaller than in our wave collapse evolutions .
the reason for this is that when evolving a blackhole it is crucial that the excision boundary is pure outflow in the pdes sense . in other words
the characteristic speeds must all have the same outward pointing sign .
since the speeds in the @xmath89 direction are like @xmath287 this means that the shift can not become too small or else the excision boundary will fail , which in turn means that @xmath286 can not be chosen too large .
we therefore place the excision boundary deeper into the blackhole so that @xmath288 and carefully monitor the coordinate lightspeeds at the inner boundary . note that this requirement is likely to cause difficulties when computing extreme gravitational waves , because on the one hand large shifts can result in poor resolution of important features , but on the other they may be required in some other region so that we may successfully excise the blackhole region . in the evolution
we immediately see significant dynamics and that for example the peak of the @xmath194 constraint violation along the @xmath139-axis is two orders of magnitude greater than in our initial kerr - schild base run with killing gauge sources .
the reason for this is presumably the presence non - trivial dynamics , plus the fact that we are excising nearer the physical singularity similar to the effect we saw with the harmonic killing slice .
regardless , by @xmath289 the data seem very close to stationary .
the simulation then evolves to the target time @xmath239 , and remarkably at the end of simulation the constraint violation in @xmath194 along the @xmath139-axis has a maximum value which is an order of magnitude smaller than in the base run . at no point
does the excision boundary fail to be outflow . as a check of the axisymmetric apparent horizon finder
we compare the results obtained with the simpler algebraic condition , @xmath290 which characterizes the position of the apparent horizon in spherical symmetry .
we find near perfect agreement throughout .
the apparent horizon moves from its initial radius @xmath291 inwards until it reaches @xmath292 around @xmath293 .
from there the horizon starts to grow again and seems to settle down to @xmath294 .
however in our lowest resolution run , a small drift of the horizon outwards is visible . at late time of the simulation , around @xmath295 , this drift accelerates and we observe that the horizon becomes aspherical .
higher resolution runs show that this effect converges away . , and the radius at which the outward lightspeed vanishes @xmath296 , computed on our base grid with inner boundary at @xmath297 .
to successfully excise , the speed must be negative at the inner boundary .
observe that excision _ exactly _ on the apparent horizon is not possible throughout all of the run.[fig : harm_phase - in ] ] [ [ harmonic - initial - slice ] ] harmonic initial slice : + + + + + + + + + + + + + + + + + + + + + + + since the stationary fully harmonic coordinates are singular at @xmath219 , one might guess that the stationary spatial generalized harmonic coordinates with gauge source functions are also singular at some radius on the killing slice , at least for some range of the parameters @xmath44 .
given the broad experience in using these coordinates in binary blackhole simulations , the naive expectation would be that , if present , this coordinate singularity is pushed further towards the physical singularity rather than out towards the event horizon for standard choices of the gauge source functions .
but this behavior is not clear . to truly resolve the issue one could simply solve for such coordinates along the lines of @xcite , but this we defer for the future .
instead we performed simulations varying the initial excision surface from the base grid excision radius @xmath298 down to @xmath299 in steps of @xmath300 .
unsurprisingly we find that initially the constraint violation , is greater in the excision subpatch as the inner boundary is placed closer to the singularity , amounting to about an order of magnitude in the @xmath194 constraint between the @xmath288 and @xmath301 boundary runs by @xmath302 .
besides this there is little to distinguish between the five runs , and at least down to this excision radius no sign of a coordinate singularity forming . by eye
the lapse function in the shared part of the domain agrees very well throughout the evolution .
although a slight drift between them is present towards the end of the test , this is acceptable since the outer boundary conditions are being imposed at different radii , the solutions need not agree everywhere .
there is however a time around @xmath303 above which the runs with inner boundary @xmath304 fail to be outflow at the excision surface . assuming that this is not caused by numerical error this means that boundary conditions are required at the surface .
it furthermore means that convergence of the numerical scheme as resolution is increased is impossible .
the fact that this does not correspond to a catastrophic failure of the code is inconvenient , because it indicates that great care must be taken in monitoring the excision surface . on the other hand , since placing the excision boundary very far in has a large cost in accuracy , a careful balance must be struck . in the spec code
this is taken care dynamically of by a control mechanism @xcite which ` bamps ` does not yet have . in fig .
[ fig : harm_phase - in ] the relationship between the character of the excision boundary and the apparent horizon is examined . comparing the initially harmonic and kerr - schild slice evolutions with excision radius @xmath288
we find that although the lapse functions initially disagree , by about @xmath305 they have exactly the same profile and lie almost on top of one another . after this time
the agreement is maintained .
in this section we present the numerical evolution of a centered brill wave , see section [ subsection : brill ] , with @xmath306 .
this highly supercritical initial data is used as a test case for our excision algorithm for a dynamically forming blackhole .
our dynamical excision method currently consists of the following steps : * 1 .
evolve to collapse : * evolve on cubed ball grid , running the apparent horizon finder in ` daemon ' mode .
the finder then triggers a ` bamps ` checkpoint once a horizon is found . *
2 . go - to excision grid : * next interpolate the checkpoint data onto a cubed - sphere grid . in this step
we want to place excision boundary just inside the apparent horizon , but as we have already seen in the single blackhole evolutions this may not always be possible , as some wiggle room is needed to allow for dynamical behavior of the horizon .
this can require some experimentation , although fine - tuning does not seem necessary .
regauge : * adjust the lapse and shift to ensure that the excision boundary is pure outflow . as a particular choice , we take lapse and shift from kerr - schild slicing of schwarzschild , @xmath307 and translate to cartesian components in the obvious way .
it is desirable that the radial coordinate light - speeds are close to zero , preferably positive , at the apparent horizon , since this determines the direction of motion of the horizon .
therefore we choose the @xmath308 parameter to satisfy this condition reasonably well , although again without particular fine tuning .
safety - net evolution : * we then use single blackhole gauge source parameters like @xmath309 and @xmath310 . during the evolution we use a safety net .
if any coordinate light - speed on the excision boundary reaches a given threshold , typically @xmath311 we again regauge to guarantee the outflow character is maintained .
we monitor the apparent horizon , and if it falls off of the numerical domain we return to an earlier checkpoint , regauging with a smaller @xmath308 to avoid this behaviour .
as the horizon expands we monitor the position and periodically return to the * go - to * step above , excising further out and regauging with a greater @xmath308 .
[ [ discussion-1 ] ] discussion : + + + + + + + + + + + as currently implemented this procedure requires that some steps be performed by hand .
the numerical results in the following subsection serve to demonstrate ` proof of principle ' of this algorithm . on the other hand it seems at least clear how those steps should be automated . at the regauge step
the use of the first order ghg variables is again a little fiddly .
much more convenient would be if the lapse and shift were readily available as variables .
but the procedure is similar to that described in the gauge perturbation tests in section [ subsection : frozen_sources ] , so we do not give full details .
also at the regauge step , it might be good to choose lapse and shift by abandoning the spherical ansatz and imposing that the coordinate light - speeds at the apparent horizon vanish .
the spec approach to controlling the excision surface is much more sophisticated , employing a control mechanism @xcite , we hope to avoid that investment in the near future . because we are interested in the collapse of waves to form , presumably , a single blackhole it seems reasonable to use a simple approach if at all possible .
one aspect of the method that is not very aesthetically appealing , is that by changing the lapse and shift in discrete steps we are computing a spacetime , or patch of spacetime in coordinates that are not globally smooth .
another issue associated with this is that of geometric uniqueness , which for the ibvp is an open question .
nevertheless one expects that the differences to the computed spacetime with one choice of regauging parameters or another will be rather small in practice , so this does not represent an immediate practical concern . [ [ initial - data - and - grid - setup ] ] initial data and grid setup : + + + + + + + + + + + + + + + + + + + + + + + + + + + + we evolved a centered brill wave as described in section [ subsection : brill ] , with seed function .
we chose a centered @xmath312 wave with @xmath306 .
the adm mass of this initial data is @xmath313 .
the maximum of the kretschmann scalar in the initial data occurs at the origin , taking the value @xmath314 . following the algorithm just outlined , we began on a cubed - ball grid with @xmath315 , and @xmath316 points per cube , with internal boundaries @xmath317 , @xmath318 and the outer boundary placed at @xmath319 .
we ran the code in cartoon mode on our local cluster _ quadler _ with @xmath320 cores .
we evolved with the generalized harmonic gauge , as in section [ section : experiments ] in the evolution of a _ much _ weaker @xmath321 brill wave , now with the gauge parameters @xmath322 and @xmath323 . at coordinate time
@xmath324 we first found an apparent horizon with mass @xmath325 .
[ [ continuation - to - code - crash ] ] continuation to code crash : + + + + + + + + + + + + + + + + + + + + + + + + + + + if we continue this evolution without going to an excision grid after the apparent horizon forms , we find that the constraints inside the apparent horizon rapidly grow along with the kretschmann scalar .
the run then crashes at roughly @xmath326 .
this gives the clear signal that if we are to examine the final masses of blackholes formed during collapse , using the ghg formulation , a robust excision algorithm will be essential .
in fact at @xmath327 the horizon has a mass of @xmath328 on the cubed - ball grid , but at the end of our excision simulation , to be described momentarily , we find that @xmath329 after apparent horizon formation it has mass @xmath330 . in the first critical gravitational wave collapse paper @xcite ,
the blackhole masses were evaluated roughly @xmath331 after apparent horizon formation , according to a prescription based on the quasinormal modes of schwarzschild . comparing those values with ours
is difficult because we use different time coordinates , but the basic expectation is that the maximal slicing condition is more `` singularity avoiding '' than one of our generalized harmonic gauges , and therefore we might expect to obtain comparable results if we can evolve for a similar coordinate time after the appearance of a horizon .
this is , however , not clear and deserves further investigation . in any case without excising the blackhole region , the meager @xmath332 after collapse is clearly insufficient .
we have seen in @xcite that with the moving - puncture method this type of data also did not result in successful evolutions beyond apparent horizon formation . but here at least a concrete improvement has been made , in that we find an apparent horizon _ before _ the method fails ! [ [ evolution - on - excision - grid ] ] evolution on excision grid : + + + + + + + + + + + + + + + + + + + + + + + + + + + checkpointing the solution at @xmath333 we then interpolating , again with barycentric lagrange interpolation as used in the apparent horizon finder , onto a cubed - sphere grid with excision radius at @xmath334 with the outer boundary position fixed , and with @xmath335 with @xmath336 angular patches , now with @xmath337 points per cube , naturally again evolving in cartoon mode . in the regauge step
we choose here @xmath338 .
this step immediately removes most of the constraint violation from the computational domain , and the largest spatial derivatives , so that the constraint monitor is @xmath339 as compared to @xmath340 on the original cubed - ball .
this difference seems very troublesome until we take into account that , for example the peak of the kretschmann scalar on the cubed ball grid is @xmath341 whereas on the cubed sphere it is @xmath342 .
so the reduction in the constraints obviously occurs because we are removing the most extreme part of the domain .
note also that our definition of the constraint monitor does not include a normalization by the size of the solution , as in for example @xcite and subsequent papers .
in view of this our reduction in resolution is justified .
the evolution then proceeded , now on @xmath343 cores using @xmath344 .
the regauge safety - net was triggered @xmath210 times up to @xmath345 , having fixed @xmath346 , but the apparent horizon remains on the computational domain throughout the calculation . at @xmath345
we perform the `` * go - to * '' step of our algorithm again , this time excising at @xmath347 choosing @xmath348 . after this the regauge safety - net
was not called before @xmath331 , when we changed cubed - sphere grid once more , keeping the same grid parameters but excising at @xmath349 , and regauging with @xmath350 .
the evolution continued @xmath351 , at which time we changed grid for the final time , before which the safety - net was again not called . in the last grid we took the excision radius to be @xmath352 and regauged with @xmath353 .
after this the regauge safety - net was not called , and the evolution was terminated at @xmath354 after apparent horizon formation . note that in this evolution the `` * go - to * '' step also employed the phase - in for the generalized harmonic gauge , as described in our single blackhole evolutions in section [ subsection : phasing ] , taking the same parameters employed in those earlier tests , but now with the initial source functions chosen so that the lapse and shift were frozen as the evolution starts on the new grid .
other experiments show that this procedure is not strictly necessary
. it may be that some refinement is required to this method to allow the evolution of supercritical data _ indefinitely _ after the collapse , but examining the mass of the apparent horizon , we interpret the solution as having mostly settled down , which should be good enough to diagnose a final mass of the blackhole .
centered brill wave .
the green planes indicate the times at which the `` * go - to * '' step was applied , and what parameter @xmath308 was chosen in that procedure .
the left plot shows a successful choice , and on the right what happens if this parameter is chosen less carefully . in the upper part of the right hand plot
one sees that the horizon contracts , and also sees that the ` regauge ' step is frequently applied , resulting in kinks in the horizon .
[ fig : ah_dynamics ] ] [ [ dynamics - of - the - apparent - horizon ] ] dynamics of the apparent horizon : + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + in the computation described above , as can be seen in in the left panel of fig .
[ fig : ah_dynamics ] , the apparent horizon is always present on the computational domain .
the horizon mass initially rapidly grows to a value around @xmath355 where it remains roughly constant . throughout
we see that when the regauge safety - net is triggered a slight oscillation in the horizon mass follows . on the other hand
when we change grid we see that the horizon mass exhibits a more prominent kink . in the right - hand panel of fig .
[ fig : ah_dynamics ] we plot the apparent horizons obtained when , less - wisely , the parameter @xmath356 is chosen in the last `` * go - to * '' at @xmath351 . with this choice the apparent horizon rapidly contracts , although the code fails before it leaves the domain .
the safety - net is called ever - more frequently as the method insists on forcing the inner boundary to remain pure outflow , until eventually the code crashes at @xmath357 .
the physical interpretation of this is that the excision boundary is falling off of the domain , which starts to drift outside the blackhole region , and that the safety - net then forces the worldline of the excision boundary to be spacelike
. this interpretation would be clearer if we had an event horizon finder , but is given credence by performing evolutions of a schwarzschild blackhole with the @xmath308 gauge parameter similarly poorly chosen .
in such tests we see that the areal radius of the excision boundary can indeed fall outside of the event horizon at @xmath358 .
we have developed a pseudospectral numerical relativity code , ` bamps ` , and in so doing have made a series of improvements and investigations into the approach employed in the spec code .
we presented a set of experiments carefully performed so that direct comparison with either published work , or independent computations of the bam finite differencing code could be made .
these included evolutions of gauge waves , convergence tests , the use of different constraint damping and ghg formulation parameters , scaling tests , perturbed blackhole evolutions and the treatment of supercritical gravitational waves .
ultimately we conclude that the ` bamps ` code is working efficiently , scales as desired up to large numbers of processors , and works on sufficiently general grid setups to evolve initial data of interest .
particularly surprising to us was the sensitivity of the method to our modifications of the ghg boundary conditions , even within the class of constraint preserving conditions .
this was the case even in our simple evolutions of the schwarzschild spacetime , so it would be very interesting to see the extent to which such results carry over to compact binary evolutions , be it in spec , or in the more distant future in ` bamps ` . from the physics point of view , however , our focus is presently on the collapse of axisymmetric gravitational waves .
much of the development reflects this fact .
most notably the implementation of octant symmetry with the cartoon method gives orders of magnitude speedups over evolving the same data in full 3d .
for a recent complimentary approach see @xcite .
we have additionally written a bespoke axisymmetric apparent horizon finder , which already proved a valuable diagnostic tool , crucial in the evolution of supercritical data , where the existence of an apparent horizon was used as the criterion for moving to an excision grid .
naturally further developments to the code may be desirable . for physical interpretation ,
an event horizon finder would complement our apparent horizon finder .
a control system like that of spec @xcite would be useful in controlling the positions of the apparent horizons .
but the highest priority will likely be in generalizing available grid setups to enable dynamical mesh - refinement .
we have also considered various different types of axisymmetric moment of time - symmetry gravitational wave initial data . in forthcoming work
we use ` bamps ` to evolve this initial data , close to the critical amplitude separating dispersion and collapse to a blackhole .
we are grateful to sebastiano bernuzzi , david garfinkle , enno harms , sascha husa , nathan kieran johnson - mcdaniel , harald pfeiffer and hannes rter for interesting discussions .
dh would like to express special gratitude to helmut friedrich for interesting discussions and for his warm encouragement .
this work was supported in part by the deutsche forschungsgemeinschaft ( dfg ) through its transregional center sfb / tr7 `` gravitational wave astronomy '' , by the dfg research training group 1523/1 `` quantum and gravitational fields '' , and the graduierten - akademie jena .
computations were performed primarily at the lrz ( munich ) . | we present a new pseudospectral code , ` bamps ` , for numerical relativity written with the evolution of collapsing gravitational waves in mind .
we employ the first order generalized harmonic gauge formulation . the relevant theory is reviewed and the numerical method is critically examined and specialized for the task at hand . in particular
we investigate formulation parameters , gauge and constraint preserving boundary conditions well - suited to non - vanishing gauge source functions .
different types of axisymmetric twist - free moment of time symmetry gravitational wave initial data are discussed . a treatment of the axisymmetric apparent horizon condition is presented with careful attention to regularity on axis .
our apparent horizon finder is then evaluated in a number of test cases . moving on to evolutions ,
we investigate modifications to the generalized harmonic gauge constraint damping scheme to improve conservation in the strong field regime .
we demonstrate strong - scaling of our pseudospectral penalty code .
we employ the cartoon method to efficiently evolve axisymmetric data in our @xmath0 dimensional code .
we perform test evolutions of schwarzschild perturbed by gravitational waves and by gauge pulses , both to demonstrate the use of our blackhole excision scheme and for comparison with earlier results .
finally numerical evolutions of supercritical brill waves are presented to demonstrate durability of the excision scheme for the dynamical formation of a blackhole . |
the si - w ecal physics prototype is composed of 30 layers of @xmath1 wafers , each wafer having an array of @xmath2 pixels of @xmath3 @xmath4 .
the two top rows of wafers are completed for the full depth in july 2006 .
the mechanical structure consists of tungsten sheets wrapped in carbon fibre , providing 15 alveola where slabs are inserted .
one slab is made of two pcbs on each side of a tungsten layer , with the wafers conductively glued to the pcb .
the very front end electronics ( vfe ) provide preamplification and are located outside the active area , but mounted on the same pcb as the silicon wafers .
the prototype is built of three stacks , each of ten layers of alternating tungsten and silicon .
each stack has a different tungsten thickness : 1.4 mm or @xmath5 per layer in the first stack , 2.8 mm or @xmath6 per layer in the second stack and 4.2 mm or @xmath7 per layer in the rear stack .
this choice should ensure r0.4 a good resolution at low energy , due to the thin tungsten in the first stack , combined with a good containment of the electromagnetic showers , with an overall thickness of about 20 cm or @xmath8 at normal incidence . to rotate to angles of @xmath9 , @xmath10 , @xmath11 and @xmath12 with respect to the beam axis ,
the three stacks , mechanically separate , are also translated laterally so that the beam still passes through all of them .
+ the purpose of the testbeam phase is to validate the simulation against a realistic detector , as well as allowing to detect the potential hardware problems .
once the simulation is trusted , full detector studies will lead to the optimisation of the calorimeter for a particle flow approach .
the test setup at cern is presented on figure [ fig : setup ] , and is simulated using mokka @xcite .
+ three drift chambers are used for the tracking at cern and four at desy .
the ecal , hcal @xcite and tcmt @xcite follow , with the ecal mounted on a movable stage for angle scans .
eight million triggers were taken at desy during 14 days in may 2006 , for seven beam energies ( 1 , 1.5 , 2 , 3 , 4 , 5 , 6 gev ) , five angles ( 0@xmath0 , 10@xmath0 , 20@xmath0 , 30@xmath0 , 45@xmath0 ) and three positions of the beam on the ecal front face ( center , border and corner of a wafer ) , with a minimum of 200,000 events per configuration .
six layers were not instrumented : the last eight layers had one dummy slab , one instrumented slab , and two dummy slabs . + the august 2006 , cern beams allowed to take another 8.6 million triggers in ecal only mode , with the full depth instrumented , for six beam energies between 6 and 45 gev , and four angles .
pions were taken between 30 and 80 gev in combination with the hcal and tcmt . for calibration purposes ,
30 million muon events were also taken parasitically to an experiment upstream . + the setup in october 2006 was slightly different , with ecal and hcal at 6 cm from each other .
3.8 million triggers were taken with electrons and positrons from 6 to 45 gev , and 22 million with pions from 6 to 80 gev .
another 40 million muon events were added for calibration purposes .
the current calibration of the ecal prototype is performed by using a set of @xmath13 high - statistics muon runs ( @xmath14 events each ) , taken during october @xmath15 with another experiment upstream , providing a wide spread of the muon beam over the front face of the prototype .
the events are triggered with a 1 m@xmath16 scintillator counter .
+ r0.45 after pedestal substraction ( see section [ peds ] ) , the runs are reconstructed using a fixed global noise cut of half a minimum ionising particle ( mip ) , 1 mip being estimated to 50 adc counts by former studies . to reject any remaining noise hits , it is required that the hits of one event form tracks characteristic of a mip .
+ the distributions obtained by channel are described by a convolution of a gaussian and a landau function .
the calibration constant is defined as the most probable value ( mpv ) of the landau function , while the standard deviation of the gaussian defines the noise value for each cell . for 6403 out of the 6480 channels of the prototype ,
a calibration constant can be obtained via this method without further investigations .
the remaining cells show a noise value that is unusually high , and are treated by estimating the additional noise contribution , or by applying the calibration constant from one of their neighbours .
+ an entire wafer ( 36 channels ) seems to not be fully depleted at the applied voltage of 200v , resulting in a mip peak at half the normal value
. a relative value between the mean mip signals of the wafer and its neighbours can be estimated , allowing a relative calibration of the cells .
0.14 % of the channels give no output and were considered as dead .
the calibration constants for all calibrated channels are histogrammed in figure [ fig : calib ] .
the distribution is narrow , with almost all pads in the range 40 to 50 adc counts per mip .
the small peak at 23.5 corresponds to the single incompletely depleted wafer . for all beam tests performed ,
the data acquisition consists of a fixed sequence of 500 pedestal events , 500 events with charge injection via the calibration chips , and then 20,000 beam events .
the pedestal events are used to make a first estimate of the pedestal ( mean value ) and noise ( rms ) per channel .
+ it has been observed that the pedestals are not necessarily stable , but subject to a random shift affecting all channels of one layer with the same drift .
this effect concerns several particular pcbs , is time dependant , and is attributed to the instabilities of the power supplies giving the pedestal lines , which are not isolated . to correct for these instabilities ,
the pedestals are recalculated on an event by event basis , by discarding all cells recording a signal , and iterating until the rms of the distribution obtained with the remaining channels is of the order of the expected noise . in order to identify coherent noise , the correlation between pairs of channels
is calculated in signal events .
no clear correlation is seen in an entire pcb , except the one coming from the pedestal drift discussed above .
the results are thus shown only on a wafer basis .
figure [ fig : noise ] displays the correlation factors ( colour scale ) as a function of the channel indices , numbered from 0 to 35 , for a particular wafer affected by the pedestal drifts described above , and a run recorded at desy with an energy of 6 gev
. channels numbered 0 to 17 and 18 to 35 correspond to two different chips . the corresponding noise level per layer , for all wafers ,
is also presented in figure [ fig : noise ] , on the right column .
+ the results before any corrections are shown in the upper row of figure [ fig : noise ] .
the left plot represents the module in which the beam was directed .
it can be seen that the region with signal shows less correlations , due to the fact that most pixels are discarded in the noise calculation .
the middle plots show a neighbouring module , affected only by the global pedestal drift .
the difference between these two wafers allows the identification of a crosstalk issue .
all the channels of some wafers recording a high signal , like the one presented in figure [ fig : noise ] on the left , suffer from a pedestal shift towards negative values .
this effect does not propagate to the neighbouring wafers .
this seems to be random in space and in time , but it is clearly correlated with the intensity of the signal recorded .
this effect is not yet understood , but under investigation . in order to correct for
the induced correlated noise , the mean and standard deviation are calculated on an event - by - event basis , per wafer , after discarding the signal hits , and iterating over the channels taken into account in the sum .
the bottow row on figure [ fig : noise ] shows the results after all the pedestal corrections described above .
the corrections are performing well , bringing the noise back to the normal level of 6 adc counts .
for this particular layer , affected by both problems ( i.e. coherent noise on the pcb level due to an instability in the power supply , and crosstalk affecting the wafer recording a high signal ) , the wafer recording a signal still shows a remaining 20% of correlations .
the correlation is however completely removed for wafers affected only by the crosstalk problem , which confirms that the corrections are performing well .
r0.75 the selection of the single electron events is loose , in order to avoid bias : * @xmath17 removes the noise .
the threshold is about five times the average noise measured per cell .
* the total energy recorded in the ecal , @xmath18 , should be in the range @xmath19 .
@xmath20 is computed with the three stacks weighted in the ratios 1:2:3 , according to the tungsten thickness .
further cuts are applied to some particular samples : the significant pion content of some high energy electron runs at cern is reduced by using the threshold erenkov counter , whereas the low energy halo coming with some low energy desy beams is rejected with additional cuts on the shower barycentre .
the effect of these two additional cuts is indicated by the shaded regions in figure [ fig : eselection ] .
the total response of ecal is computed by summing the hit energies in the three sections of the detector . if @xmath21 , @xmath22 , @xmath23 are the recorded energies in the first , second and third stack respectively , the total response is @xmath24 .
the nave choice for the weights @xmath25 is generally used .
it reflects the relative thicknesses of the tungsten layers in each of the stacks , and hence the relative sampling fractions .
however , a weighting scheme optimisation for energy resolution was performed as well , leading to the slightly different values of @xmath26 .
the normalisation @xmath27 has been arbitrarily fixed to 250 mip / gev .
the guard rings create 2 mm non - active inter - wafer gaps , causing non - uniformities in the ecal response , as illustrated in figure [ fig : edistrib ] , left , where the mean energy is plotted as a function of the shower barycentre , .
dips in response are clearly visible r0.7 and account for the asymmetric tail on the low side of the distribution of total energy ( figure [ fig : edistrib ] , right ) , which is reasonably well modelled by the simulation .
the correction of these non - uniformities will be discussed in section 6 .
the beam profile , and thus the fraction of beam particles traversing the inter - wafer gaps , depends on the beam energy .
therefore , in order to avoid bias , a cut is applied on the shower barycentre position such as to select showers not affected by the gaps .
r0.7 to estimate the energy resolution and the mean calorimeter response , the distribution from figure [ fig : energy ] , left , is fitted by a gaussian in the asymmetric range of @xmath28 $ ] in order to reduce sensitivity to pion background and to radiative effects upstream of the calorimeter , as well as to any residual influence of the inter - wafer gaps .
r0.45 the ratio of the reconstructed energy to the beam energy , as function of the beam energy is shown in figure [ fig : energy ] , right , for the two choices of weights .
non - linearities are at the @xmath29 level .
the linearity is somewhat better for the optimised weights .
the energy resolution , similar for both weightings , is shown in figure [ fig : energy_res ] .
the monte carlo prediction , with the nave weights , also shown , is in reasonably good agreement with the data .
the resolution can be parametrised , for the nave choice of weights and , respectively , the optimised one , as + @xmath30
r0.7 the method used for correcting the interwafer gaps operates at the event level and relies only on the calorimeter information : it parametrises the mean calorimeter response as function of the shower position in the calorimeter and applies subsequently corrective factors for each event according to this parametrisation .
it is only geometrical , independent in @xmath31 and @xmath32 and without any explicit dependence on the shower energy .
the impact of the corrections is clearly illustrated in figure [ fig : gaps ] , left , where the @xmath33 scan of ecal is shown for the raw and , respectively the corrected data .
the low energy tail of the energy distribution is greatly reduced ( figure [ fig : gaps ] , right ) .
the resolution loss when going from the out of gap events to all the events ( with corrections applied on the energy ) is of the order of 10% .
the corrections do not degrade the linearity . when tracking information is available , it is possible to precisely calculate the shower position within each wafer .
subsequently , the ratio of the active to non - active areas crossed by the shower according to a mean shower shape can be estimated and used to correct the energy recorded in each layer .
, normal incidence , centre of wafers ) . _ right : _ radii for 90% and 95% signal containment of e@xmath34 as function of the beam energy . ] , normal incidence , centre of wafers ) .
_ right : _
radii for 90% and 95% signal containment of e@xmath34 as function of the beam energy . ]
only events outside the interwafer gaps were used to characterise the longitudinal development of the shower .
the mean energy distribution is well fitted by the standard parametrisation , @xmath35 , where @xmath36 is the calorimeter depth , @xmath37 is an overall normalisation , @xmath38 and @xmath27 are constants ( figure [ fig : long_dev ] ) .
the position of the shower maximum grows logarithmically with the beam energy .
an important issue in the development of a calorimeter is to achieve the smallest possible effective molire radius , in order to provide the best shower separation .
it requires the use of an absorber with a small intrinsic molire radius ( @xmath39 ) , but also the minimisation of the gaps between the absorber layers .
figure [ fig : rad_dev ] , left shows the event distribution for 90% and 95% levels of signal containment with respect to the radius .
the results for the various energies studied are summarised in figure [ fig : rad_dev ] , right .
the points correspond to the peak position of each radius distribution . at 90% ( 95% )
shower containment the corresponding radius , often quoted as 1 @xmath39 , is about 20 ( 28 ) mm .
the geometry of the ecal prototype , with 2.2 mm thick interlayer gaps leads to an effective molire radius which is expected to be larger by a factor of about 2 with respect to @xmath39 of solid tungsten .
the results from the test beam studies are therefore in agreement with expectations .
r&d effort towards the use of si pads with integrated readout is under way and will hopefully lead to a significant decrease of the interlayer gap and therefore of the ecal effective molire radius .
the spatial and angular resolution of the ecal are studied with the desy data at normal incidence .
the shower direction and position at the ecal front face are constructed on an event - by - event basis using a linear two - parameter chi - square fit to the shower barycentre positions in each layer for the @xmath40 and @xmath33 coordinates separately .
the correlation matrix is determined from simulations for each beam energy .
r0.7 the fit results are compared with the position and angle measured by the tracking system .
the expected e@xmath34 position and direction at ecal front face is obtained from a linear fit of the drift chambers .
sources of systematic uncertainties as residual misalignement , material modelling , and background rate are estimated for the extrapolation to the ecal front face .
the ecal resolution , deconvoluted from the tracking errors , is displayed in figure [ fig : tracking ] .
the si / w ecal prototype was presented , as well as the results of the first beam tests at desy and cern during 2006 .
the prototype calorimeter was further exposed to beam at cern in summer 2007 .
analysis of the data collected is in progress .
mokka , a detailed geant4 simulation for the ilc detectors .
available from + ` http://polywww.in2p3.fr/activities/physique/geant4/tesla/www/mokka/mokka.html ` f. sefkow , `` tile hadron calorimeter '' , calorimeter r&d review , these proceedings . | the calice @xcite si - w electromagnetic calorimeter @xcite has been tested with electron beams ( 1 to 6 gev ) at desy in may 2006 , as well as electrons ( 6 to 45 gev ) and hadrons ( 6 to 80 gev ) at cern in august and october 2006 .
several millions of events have been taken at different incident angles ( from 0@xmath0 to 45@xmath0 ) and three beam impact positions .
the ecal calibration is performed with muon beams and shows a good uniformity for nearly all channels .
the large statistics available allows not only to characterise the ecal physics performance , but also to identify subtle hardware effects . |
a one - sided infinite sequence @xmath1 over the alphabet @xmath2 is called a ( classical ) _ kolakoski sequence _ ( named after w. kolakoski who introduced it in 1965 , see @xcite ) , if it equals the sequence defined by its run lengths , e.g. : @xmath3 here , a _ run _ is a maximal subword consisting of identical letters .
the sequence @xmath4 is the only other sequence which has this property .
one way to obtain @xmath1 of ( [ eq : kol ] ) is by starting with @xmath5 as a seed and iterating the two substitutions @xmath6 alternatingly , i.e. , @xmath7 substitutes letters on even positions and @xmath8 letters on odd positions ( we begin counting at @xmath9 ) : @xmath10 clearly , the iterates converge to the kolakoski sequence @xmath1 ( in the obvious product topology ) , and @xmath1 is the unique ( one - sided ) fixed point of this iteration
. one can generalize this by choosing a different alphabet @xmath11 ( we are only looking at alphabets with @xmath12 ) .
such a ( generalized ) kolakoski sequence , which is also equal to the sequence of its run lengths , can be obtained by iterating the two substitutions @xmath13 alternatingly . here , the starting letter of the sequence is @xmath14 .
we will call such a sequence a kolakoski-@xmath15 sequence , or kol@xmath15 for short .
the classical kolakoski sequence @xmath1 of ( [ eq : kol ] ) is therefore denoted by kol@xmath16 ( and @xmath17 by kol@xmath18 ) .
while little is known about the classical kolakoski sequence ( see @xcite ) , and the same holds for all kol@xmath15 with @xmath14 odd and @xmath19 even or vice versa ( see @xcite ) , the situation is more favorable if @xmath14 and @xmath19 are either both even or both odd . if both are odd , one can , in some cases , rewrite the substitution as a substitution of pisot type ( see @xcite ) , which can be described as ( limit ) aperiodic model sets . by this method ,
kol@xmath20 is studied in @xcite and shown to be a deformed model set .
the case where both symbols are even will be studied below .
it is the aim of this article to determine structure and order of the sequences kol@xmath21 .
this will require two steps : first we establish an equivalent substitution of constant length for kol@xmath21 and analyze it with methods known from the theory of dynamical systems
. then we conclude diffractive properties from this .
remark : every kol@xmath15 can uniquely be extended to a bi - infinite ( or two - sided ) sequence .
the one - sided sequence ( to the right ) is kol@xmath15 as explained above .
the added part to the left is a reversed copy of kol@xmath22 , e.g. , in the case of the classical kolakoski sequence of ( [ eq : kol ] ) , this reads as @xmath23 where `` @xmath24 '' denotes the seamline between the one - sided sequences .
note that , if @xmath25 ( or @xmath26 ) , the bi - infinite sequence is mirror symmetric around the first position to the left ( right ) of the seamline .
the bi - infinite sequence equals the sequence of its run lengths , if counting is begun at the seamline .
alternatively , one can get such a bi - infinite sequence by starting with @xmath27 and applying the two substitutions to get @xmath28 in the first step and so forth .
this also implies that kol@xmath15 and kol@xmath22 will have the same spectral properties , and it suffices to study one of them .
if both letters are even numbers , i.e. , @xmath29 and @xmath30 ( with @xmath31 , where we can concentrate on @xmath32 by the above discussion ) , one can build blocks of two letters and obtain an ( ordinary ) substitution . setting @xmath33 and @xmath34 , these substitutions and their _ substitution matrix _ @xmath35 ( sometimes called _ incidence matrix _ of the substitution )
are given by @xmath36 where the entry @xmath37 is the number of occurrences of @xmath38 in @xmath39 ( @xmath40 ; sometimes the transposed matrix is used ) .
a bi - infinite fixed point can be obtained as follows : @xmath41 this corresponds to the unique bi - infinite kol@xmath21 according to our above convention .
a substitution @xmath42 is _ primitive _ if the corresponding substitution matrix @xmath35 is primitive , i.e. , @xmath43 has positive entries only for some @xmath44 .
equivalently , @xmath42 is primitive if there exists a positive integer @xmath44 such that every @xmath45 occurs in @xmath46 for all @xmath47 .
the vector @xmath48 with components @xmath49 , for @xmath45 , is called the _ length _ of the substitution @xmath42 .
if all @xmath50 are equal , @xmath42 is a substitution of _ constant length_. for the substitution @xmath51 of ( [ eq : subs1 ] ) , we have @xmath52 which is therefore not of constant length ( recall that @xmath53 ) . we will also need some notions from the theory of _ dynamical systems _
, see @xcite and ( * ? ? ?
* chapters 1 , 5 and 7 ) for details .
let @xmath42 be a primitive substitution over @xmath54 and @xmath55 a bi - infinite fixed point of @xmath42 ( i.e. , @xmath56 for some @xmath57 ) . denote by @xmath58 the @xmath59th letter of @xmath60 ( @xmath61 ) and by @xmath62 the ( i.e. , @xmath63 )
. let @xmath54 be equipped with the discrete and @xmath64 with the corresponding product topology .
if we set @xmath65 then @xmath66 is a _
dynamical system_. since we require @xmath42 to be primitive , this dynamical system is _ minimal _
( i.e. , @xmath67 is dense in @xmath68 for all @xmath69 ) , does not depend on the chosen fixed point @xmath60 ( if more than one exists , which is possible in the two - sided situation ) and has a unique probability measure @xmath70 associated with it .
in other words , it is _
strictly ergodic_. on the hilbert space @xmath71 , we have the unitary operator @xmath72 if @xmath73 for some @xmath74 , we call @xmath75 an _ eigenvalue _ of @xmath66 and @xmath76 the corresponding _ eigenfunction_. the _ spectrum _ ( of the dynamical system ) is said to be a _ pure point dynamical spectrum _ ( or _ discrete spectrum _ ) , if the eigenfunctions span @xmath71 . if @xmath77 is the only eigenvalue and the only eigenfunctions are the constants , the spectrum is _
continuous_. it is also possible that it has pure point and continuous components . in that case
it is called _ partially continuous_. two dynamical systems @xmath78 and @xmath79 are _ isomorphic _ ( or _ measure - theoretically isomorphic _ ) , if there exists an invertible measurable map @xmath80 , almost everywhere defined , such that @xmath81 preserves the measure and the dynamics ( i.e. , @xmath82 ) . the spectral theory of primitive substitutions of constant length is well understood . by the following criterion , we know that the substitutions @xmath51 of ( [ eq : subs1 ] ) are related to substitutions of constant length . [ lem : dek - v1 ] ( * ? ? ?
* section v , theorem 1 ) let @xmath42 be a substitution of nonconstant length @xmath48 . if @xmath48 is a right eigenvector of the corresponding substitution matrix @xmath35 , then @xmath83 is isomorphic to a substitution dynamical system generated by a substitution of constant length . since the substitutions @xmath51 of ( [ eq : subs1 ] ) fulfillnote that from ( [ eq : subs0 ] ) , one can also construct a primitive substitution by distinguishing odd and even positions , e.g. , for kol@xmath84 we would get ( we use @xmath85 as mark for even positions ) @xmath86 instead of ( [ eq : subs1 ] ) , we would get a substitution with substitution matrix @xmath87 which is also primitive ( @xmath88 has positive entries only ) , but does not fulfill the requirements of lemma [ lem : dek - v1 ] .
the eigenvalues of this @xmath35 are @xmath89 . ] the requirements of this lemma , the next task is now to construct the corresponding substitutions of constant length .
this is achieved by numbering the @xmath90 s and @xmath91 s in ( [ eq : subs1 ] ) , i.e. , we make the substitutions @xmath92 , respectively @xmath93 .
then , the former substitutions ( [ eq : subs1 ] ) induce @xmath94 from this we get substitutions of constant length @xmath95 ( the eigenvalue of the substitution matrix @xmath35 in ( [ eq : subs1 ] ) ) by parting the right sides in blocks of @xmath95 letters . for example , let @xmath96 and @xmath97 .
then @xmath98 and one extracts the following substitution of constant length @xmath99 : @xmath100 in the same way , we get substitutions of constant length @xmath95 from ( [ eq : subs2 ] ) .
note that these substitutions are all primitive since ( compare to ( [ eq : subs2 ] ) ) in every block of @xmath101 successive letters ( note that we use @xmath102 ) every letter of @xmath103 occurs , so if @xmath104 , then @xmath105 has positive entries only ( this holds for @xmath106 ) .
note also that we can reduce the alphabet by one letter by identifying @xmath107 ( @xmath108 in the example ( [ eq : kol42h2 ] ) ) , because both @xmath90 s always yield the same substitution .
let us now determine the positions of @xmath109 in the sequence @xmath60 generated by ( [ eq : subs2 ] ) .
they are given by @xmath110 for some @xmath111 ( e.g. , @xmath112 ) .
therefore we get @xmath113 .
the _ height _
@xmath114 of a primitive substitution @xmath42 of constant length @xmath0 which generates a sequence @xmath60 is defined as @xmath115 then the following lemma holds .
[ lem : dek - ii13 ] let @xmath66 be a dynamical system , where @xmath42 is a primitive substitution of constant length @xmath0 and height @xmath114 .
then the pure point part of this dynamical system is isomorphic to the dynamical system @xmath116 , where @xmath117 is the addition of @xmath118 on the abelian group @xmath119 , i.e. , the direct product of the @xmath0-adic integers @xmath120 and the cyclic group @xmath121 of order @xmath114 .
therefore the pure point dynamical spectrum is given by @xmath122 note that @xmath123 where @xmath124 are the distinct primes dividing @xmath0 , see ( * ? ? ?
* section 3.10 ) .
the lemma is just a reformulation of ( * ? ? ?
* theorem ii.13 ) , compare with ( * ? ? ?
* section vi . ) and ( * ? ? ?
* section 7.3 ) .
[ prop : spectrum ] suppose @xmath125 has pure point dynamical spectrum , where @xmath51 is the substitution of ( [ eq : subs1 ] ) .
then @xmath126 where @xmath117 is the addition of @xmath118 and @xmath127 the addition of @xmath77 . for the substitution ( [ eq : subs2 ] ) of constant length @xmath128
, we have already seen that @xmath129 .
therefore , using ( [ eq : def_height ] ) , the height of this substitution is @xmath5 if @xmath0 is odd and @xmath77 if @xmath0 is even .
the dynamical system of the substitution ( [ eq : subs2 ] ) is isomorphic to @xmath125 by lemma [ lem : dek - v1 ] , therefore they have the same spectrum .
the remaining statement follows from lemma [ lem : dek - ii13 ] .
we want to show that the spectrum of @xmath51 is indeed pure point . for this
we use slightly different substitutions of constant length that we deduce from @xmath51 .
we substitute @xmath130 and @xmath131 ( so in ( [ eq : subs2 ] ) we build essentially blocks of two , e.g. , @xmath132 ) .
we get @xmath133 which again gives substitutions of constant length @xmath128 .
they are all primitive substitutions by the same argument as before ( in every block of @xmath134 successive letters every letter occurs ) . in the case @xmath135 , we can reduce the alphabet by one letter by identifying @xmath136 .
so we have two cases , @xmath97 with substitutions @xmath137 and @xmath135 with substitutions @xmath138 ( it is cumbersome to write down such a @xmath138 in general form , but we will investigate its structure in the next section ) . now the height of @xmath138 and @xmath139 is always @xmath77 , because if @xmath135 we get @xmath140 , and if @xmath97 we get @xmath141 .
let @xmath42 be a primitive substitution of constant length @xmath0 and height @xmath142 .
one says that @xmath42 admits a _ coincidence _ , if there exist a @xmath44 and @xmath143 such that @xmath144 is the same for all @xmath45 ( the @xmath38th letter of each @xmath145 is the same , i.e. , @xmath146 admits a column of identical values ) .
[ lem : dekiii7 ] ( * ? ? ? * section iii , theorem 7 ) let @xmath66 be a substitution dynamical system of constant length and height @xmath142 .
then @xmath66 has pure point dynamical spectrum if and only if @xmath42 admits a coincidence .
if a substitution has height @xmath147 , one gets a substitution of height @xmath77 by combining letters into blocks of @xmath148 letters .
if this new substitution has pure point dynamical spectrum , so has the original substitution of height @xmath148 , see @xcite . obviously , we get the following : if the substitutions @xmath138 and @xmath139 ( which arise from ( [ eq : subs3 ] ) ) admit coincidences , then the dynamical systems defined by @xmath51 of ( [ eq : subs1 ] ) have pure point dynamical spectrum .
let us first check @xmath139 of ( [ eq : kol2m2 ] ) for coincidences . for this
we begin by exploring the structure : @xmath149 has two @xmath150 s at position @xmath9 and @xmath151 , @xmath152 has an @xmath150 at position @xmath153 , etc .
we get an @xmath150 in @xmath154 at position @xmath155 for @xmath156 .
similar arguments show that there is an @xmath157 in @xmath154 at position @xmath158 for @xmath159 and at @xmath153 in @xmath160 .
now , @xmath161 has @xmath162 s at all positions @xmath163 .
furthermore , @xmath160 has a @xmath162 at position @xmath151 and shares the first @xmath151 letters with @xmath149 .
schematically , we get the following structure of @xmath139 : @xmath164 here we have omitted ( @xmath165 ) all letters that are not necessary and by @xmath166 we denote the part that @xmath149 and @xmath160 share .
we now check for _ pairwise coincidences _ , i.e. , for @xmath167 we check whether there is a @xmath44 and a @xmath168 such that @xmath169 .
so we pick @xmath170 , @xmath171 .
suppose @xmath172 ( otherwise we interchange @xmath173 and @xmath174 ) .
then @xmath174 either equals @xmath157 or at least @xmath175 has an @xmath157 ( every @xmath176 , @xmath45 has one ) . in the first case
take @xmath177 , otherwise @xmath178 .
observe that there are @xmath151 successive @xmath162 s in @xmath161 .
so , if we look at @xmath179 and @xmath180 , we get the following : on the one hand , there are @xmath151 successive @xmath162 s somewhere in @xmath180 , say at positions @xmath181 . on the other hand , in @xmath179 , there is at one of these positions @xmath181 either a @xmath162 , and we have a pairwise coincidence , or an @xmath150 .
say there is an @xmath150 at @xmath182 with @xmath183 .
then in @xmath184 and @xmath185 we have pairwise coincidences at positions @xmath186 ( the @xmath166 s of ( [ eq : kol2m2_e ] ) ) . from this pairwise coincidences
we get a coincidence inductively : we start with two letters @xmath187 and after @xmath188 substitutions we have a pairwise coincidence , say at @xmath189 .
now a third letter @xmath190 may have something else at @xmath191 , but whatever it is , in @xmath192 ( @xmath193 ) all three coincide somewhere at a position @xmath194 with @xmath195 .
since there are @xmath196 letters , we get a coincidence after at most @xmath197 substitutions ( i.e. , there is a @xmath198 such that all @xmath199 are the same for all @xmath45 ) .
the structure of @xmath138 is different .
we have @xmath200 and therefore @xmath201 .
let us first show an example , with @xmath202 and @xmath203 : @xmath204 since the positions of two consecutive @xmath150 s in the sequence differ by at most @xmath151 , there is an @xmath150 in every @xmath205 with @xmath45 ( note that @xmath138 is a substitution of constant length @xmath206 ) .
again we look for pairwise coincidences , so choose @xmath207 . then there is ( at least ) one @xmath150 in @xmath208 , say at position @xmath189 , and ( at least ) one in @xmath209 , say at position @xmath194 .
since there can be more than one @xmath150 in either , we choose @xmath210 such that @xmath211 is minimal .
we further choose @xmath187 such that @xmath212 ( in the case @xmath213 , e.g. , @xmath214 and @xmath215 in the above example , we are already done ) .
if we look at @xmath208 and @xmath209 , there are two cases each ( and therefore four cases , if we look at the combinations ) : either @xmath216 or @xmath217 and either @xmath218 or @xmath219 .
this is all that can occur by the chosen minimality of @xmath220 .
now we examine the case where @xmath221 and @xmath218 .
we want to show that @xmath222 and @xmath223 have a pairwise coincidence .
let us look at the @xmath150 s in @xmath224 only ( we use again the above example , but the reasons given apply for arbitrary @xmath151 , @xmath225 ) : @xmath226 first we number the @xmath150 s with @xmath227 ( left to right in @xmath224 and top ( @xmath228 ) to bottom ( @xmath229 ) ) and we will speak of the @xmath59th @xmath150 ( with @xmath230 ) according to that number .
we observe the following : * let @xmath231 .
if the @xmath59th @xmath150 occurs at position @xmath232 in @xmath224 , then the @xmath233-st @xmath150 occurs at position @xmath234 in @xmath235 .
if the @xmath59th @xmath150 occurs at position @xmath236 , then the @xmath233-st @xmath150 occurs at @xmath237 in the same @xmath224 .
* let @xmath238 .
if the @xmath59th @xmath150 occurs at position @xmath232 , then the @xmath239-st @xmath150 occurs at @xmath234 in the same @xmath224 .
if the @xmath59th @xmath150 occurs at position @xmath240 in @xmath224 , then the @xmath239-st @xmath150 occurs at position @xmath237 in @xmath241 . and the position of the corresponding @xmath150 .
we call this the `` contrary line break property '' . ] * the second and the @xmath242-th @xmath150 occur at the same position @xmath151 in @xmath243 , respectively @xmath244 . with the previous two observations we get :
the @xmath59th and the @xmath245-th @xmath150 occur at the same position for @xmath246 .
* the first and the @xmath247-st @xmath150 occur in @xmath224 where there is at least one more @xmath150 .
this is obvious for the first @xmath248 , for the @xmath247-st observe that if it occurs at position @xmath249 , then there is also one at @xmath250 , and if it occurs at position @xmath251 , then there is also one at @xmath252 .
these observations are based on the facts that the length of the substitution is @xmath95 and that the position of the @xmath150 s in the sequence are separated by @xmath151 or @xmath225 only .
now the fact that the @xmath248 always occur in ascending order ( i.e. , we have @xmath253 and not @xmath254 or something else ) together with the first two observations essentially gives us an algorithm , which always yields a pairwise coincidence in @xmath222 and @xmath223 .
let us explain it in our example ( [ eq : ex - kol106 ] ) : suppose we have @xmath255 and @xmath256 .
then we have @xmath257 and @xmath258 .
the first step is always to reduce @xmath194 by one , so we have @xmath259 .
we have @xmath260 , but there is a second @xmath150 in @xmath244 ( @xmath157 occurs at position @xmath261 in @xmath174 ! ) , so we can increment @xmath189 by @xmath77 and get @xmath262 .
we have @xmath263 , and @xmath264 and @xmath265 both have an @xmath150 at position @xmath5 ( the third respectively the ninth @xmath150 ) .
therefore we get a pairwise coincidence in @xmath266 and @xmath267 .
this algorithm relies on the `` contrary line break property '' .
the other three cases are mutatis mutandis the same , see the positions of the @xmath150 s in ( [ eq : kol106 ] ) .
so , starting with any two @xmath207 we get a pairwise coincidence in @xmath222 and @xmath223 .
inductively like before , we get a coincidence after at most @xmath268 substitutions .
therefore we have established the following .
[ thm : spectrum ] @xmath125 with @xmath51 of ( [ eq : subs1 ] ) has pure point dynamical spectrum .
also , the dynamical system of the substitutions of constant length as defined implicitly in ( [ eq : subs2 ] ) and @xmath269 of ( [ eq : subs3 ] ) and ( [ eq : kol2m2 ] ) have pure point dynamical spectrum .
we have @xmath270 where @xmath117 is the addition of @xmath118 and @xmath127 the addition of @xmath77 . remark : let @xmath42 be a primitive substitution of constant length @xmath0 and height @xmath77 over the alphabet @xmath271
. then we can define the coincidence matrix @xmath272 , which is a quadratic @xmath273 matrix .
the entries are defined as follows ( where @xmath274 and @xmath275 ) : @xmath276 note that the substitution matrix @xmath35 is a submatrix of @xmath272 , since @xmath277 . also , @xmath272 has row sums @xmath0
. with this definition , lemma [ lem : dekiii7 ] reads as follows .
* proposition x.1 ) different from @xcite .
the coincidence matrix there has dimension @xmath278 and has the form ( with the proper enumeration of the pairs ) @xmath279 while the one defined above has the form @xmath280 here @xmath281 , @xmath282 , @xmath283 are quadratic matrices and @xmath281 is the substitution matrix .
the proposition is true for both matrices , the proof is analogous . ] for @xmath66 are equivalent : 1 .
@xmath66 has pure point dynamical spectrum .
2 . @xmath0 is a simple eigenvalue of the corresponding coincidence matrix @xmath272 .
obviously , @xmath0 is an eigenvalue of @xmath272 ( @xmath284 is a stochastic matrix with row sum @xmath77 ) .
now the above proof of theorem [ thm : spectrum ] translates to the following statements for @xmath272 : * for @xmath139 , the third power of the coincidence matrix , @xmath285 , has a column ( @xmath286 ) with nonzero entries only . * for @xmath138 ,
the square of the coincidence matrix , @xmath287 , has a column ( @xmath288 ) with nonzero entries only .
* lemma x.3 ) let @xmath289 be a quadratic matrix with nonnegative integral entries and row sums @xmath0 . if @xmath290 for all @xmath291 and a fixed @xmath38 , then @xmath0 is a simple eigenvalue of @xmath289 .
this establishes the desired result that @xmath0 is a simple eigenvalue for the coincidence matrix @xmath272 of @xmath138 ( @xmath139 ) , since @xmath292 ( @xmath293 ) is a simple eigenvalue of @xmath287 ( @xmath285 ) .
we end this section with an example .
we take @xmath139 for @xmath96 , @xmath97 and therefore the substitution @xmath294 we get the coincidence matrix @xmath295 here , already @xmath287 has columns with positive entries only .
the eigenvalues of @xmath272 are @xmath296 .
a _ model set _ @xmath297 ( or _
cut - and - project set _ ) in _ physical space _
@xmath298 is defined within the following general cut - and - project scheme , see @xcite , @xmath299 where the _ internal space _
@xmath300 is a locally compact abelian group , and @xmath301 is a _ lattice _ ,
i.e. , a co - compact discrete subgroup of @xmath302 .
the projection @xmath303 is assumed to be dense in internal space , and the projection @xmath304 into physical space has to be one - to - one on @xmath305 .
the model set @xmath297 is @xmath306 where the _ window _
@xmath307 is a relatively compact set with nonempty interior .
let @xmath60 be a bi - infinite sequence over @xmath308 and @xmath309 be a ( bounded ) function which assigns to every letter a complex number ( the _ scattering strength _ ) .
then the _ autocorrelation coefficients _
@xmath310 are given by @xmath311 provided the limits exist .
we write @xmath312 for the dirac measure at @xmath313 , i.e. , @xmath314 for @xmath76 continuous .
then the _ correlation measure _ @xmath315 of @xmath60 is given by @xmath316 and the _ _ diffraction spectrum _ _ as an atomic chain , where there is an atom of type @xmath317 at position @xmath225 with scattering strength @xmath318 .
we represent this atom as @xmath319 and therefore get a ( countable ) sum of weighted dirac measures with autocorrelation @xmath315 . ]
is given by the fourier transform @xmath320 of this measure . if @xmath320 is a sum of dirac measures only , i.e. , @xmath321 with a countable set @xmath322 ( for any choice of complex numbers @xmath323 ) , then @xmath60 is _ pure point diffractive _
, i.e. , the diffraction spectrum consists of _ bragg peaks _ only .
also , the @xmath324 s are the square of the absolute value of the corresponding _ fourier - bohr coefficient _ at @xmath59 and therefore non - negative ( real ) numbers .
if there is no dirac measure in @xmath320 , except one at position @xmath9 , @xmath325 , which is determined by the density of the structure only , then the diffraction spectrum is _
continuous_. for substitutive systems , the diffraction spectrum and the spectrum of the corresponding dynamical system are closely related , see @xcite .
[ prop : lm1 ] ( * ? ? ?
* corollary 1 . )
let @xmath42 be a primitive substitution of constant length @xmath0 with height @xmath77 over @xmath271 , where @xmath60 is a fixed bi - infinite word of @xmath42 .
define @xmath326 for all @xmath45 .
we have @xmath327 , where @xmath328 denotes disjoint union
. then the following are equivalent : 1 .
@xmath42 admits a coincidence .
2 . the @xmath329 s are model sets for @xmath330 3 .
the sequence @xmath60 and the sets , where we set @xmath331 and @xmath332 for all @xmath333 . ]
@xmath329 are pure point diffractive .
note that properties `` one - to - one '' and `` dense '' are obvious , the interesting part is that there exist relatively compact windows ( with respect to the @xmath0-adic topology ) with nonempty ( ! ) interior .
we will discuss this point in the next section . with this proposition
, we know that @xmath138 and @xmath139 ( as defined implicitly in ( [ eq : subs3 ] ) ) generate sequences which are pure point diffractive .
but we got every letter @xmath45 for the appropriate alphabet for @xmath138 , @xmath139 by building four - letter blocks in the substitution rule ( [ eq : subs0 ] ) , e.g. , in ( [ eq : kol42 ] ) we have @xmath334 , @xmath335 and @xmath336 .
such a deterministic substitution rule ( i.e. , kol@xmath21 is _ local derivable _ from the sequence generated by @xmath138 , respectively @xmath139 ) does not change the nature of the diffraction spectrum , only the fourier - bohr coefficients .
the diffraction spectrum of kol@xmath21 can be calculated from the one generated by @xmath138 , @xmath139 as follows : for kol@xmath21 we only have scattering strengths @xmath337 and @xmath338 .
if we therefore choose @xmath339 ( @xmath45 with respect to @xmath138 , @xmath139 ) according to its four - letter - composition in @xmath340 , then the diffraction spectrum @xmath320 of @xmath138 , @xmath139 is also a diffraction spectrum of kol@xmath21 , where kol@xmath21 is realized as an atom chain with atoms not on @xmath341 but on @xmath342 ( we get the diffraction spectrum of kol@xmath21 realized on @xmath341 by a simple rescaling with the factor @xmath343 ) .
for the example ( [ eq : kol42 ] ) , this means that we choose @xmath344 so in this case , the diffraction spectrum of kol@xmath84 is given by the one of @xmath345 only .
all kol@xmath21 are pure point diffractive . for a sequence @xmath346 , where @xmath347 is a bi - infinite fixed point of a primitive substitution and @xmath348 is a morphism ( where @xmath60 ( @xmath347 ) is a sequence over @xmath349 ( @xmath350 ) ) , the following statements are equivalent : 1 .
the dynamical system of @xmath60 has pure point dynamical spectrum .
@xmath60 has pure point diffraction spectrum .
this is a ( weak ) conclusion of ( * ? ? ?
* theorem 3.2 ) .
we therefore obtain the following . all kol@xmath21 have pure point diffraction and pure point dynamical spectrum .
remarks : the dynamical system of kol@xmath21 is isomorphic to @xmath351 where @xmath117 is the addition of @xmath118 and @xmath127 the addition of @xmath77 .
this can be seen by the fact that we get kol@xmath21 from the one generated by @xmath51 of ( [ eq : subs1 ] ) by the substitution @xmath352 , @xmath34 , which corresponds just to doubling each letter in the latter one ( see the substitution matrices in ( [ eq : subs1 ] ) and footnote [ foo : note ] ) . if we consider how to get from ( [ eq : subs2 ] ) , respectively ( [ eq : subs3 ] ) , to kol@xmath21 and compare this to proposition [ prop : lm1 ] , we get ( compare with ( [ eq : dynspec ] ) ) : kol@xmath21 , respectively @xmath353 , @xmath354 are model sets for @xmath355 where @xmath356 ( as before , @xmath357 , where @xmath124 are the distinct primes dividing @xmath95 . ) the diffraction spectrum calculated from this cut - and - project scheme is consistent with the previous one , since the fourier - bohr coefficients which arise in each @xmath358 separately are weighted by factors @xmath359 or @xmath360 which depend on the element of the cyclic group @xmath361 ( similar for the case @xmath362 ) , compare with @xcite but this is just how we calculated the @xmath339 s from @xmath337 and @xmath338 .
let us show how one calculates the diffraction spectrum of @xmath345 of ( [ eq : kol42 ] ) explicitly .
this substitution can be written in recursive equations for @xmath363 , @xmath364 and @xmath345 by observing at which position in which substitution a certain letter occurs ( e.g. , @xmath162 occurs in @xmath152 at positions @xmath77 and @xmath5 and in @xmath160 at position @xmath5)-adic space , because multiplication by a factor of @xmath99 is a contraction in the @xmath99-adic topology .
the closure of the windows in the @xmath99-adic internal space is therefore given by the unique compact solution of this ifs by ( a generalized version of ) hutchinson s theorem ( * ? ? ?
* section 3.1(3 ) ) .
this method is well known for unimodular substitutions of pisot - type , see @xcite , @xcite and the vast literature about rauzy fractals .
similar results apply for all primitive substitutions of constant length @xmath0 in @xmath0-adic space . ] : @xmath365 where @xmath366 .
iterating these equations and using @xmath367 , one gets : @xmath368 the fourier transform of each lattice coset @xmath369 is easy to calculate : @xmath370 every fourier - bohr coefficients of @xmath371 are then given by the sum of the fourier - bohr coefficients of the corresponding @xmath372 .
the structure of @xmath345 ( similar for all @xmath329 that occur for substitutions of constant length ) as a union of a countable but infinite set of ( periodic ) lattice cosets @xmath373 gives rise to the name _ limit - periodic _ , see @xcite .
the _ support _ of the bragg peaks of kol@xmath21 is given by @xmath374 where @xmath375 are the distinct primes dividing @xmath95 and @xmath376 is the cyclic group of ( [ eq : cyclicgroup ] ) .
however , there need not be a bragg peak on every point of the support , e.g. , kol@xmath377 is equivalent to a substitution @xmath138 of constant length @xmath378 , but the positions of a letter @xmath248 , @xmath379 are separated by multiples of @xmath5 : the support in this case is better described by @xmath380 than by @xmath381 .
so far , we have talked in an `` abstract '' way about the @xmath0-adic internal space .
usually the discussion ends at this point , but we want to `` visualize '' this @xmath0-adic space .
we hope that by doing this , we also gain some intuition for such spaces and the meaning of @xmath14-adic internal spaces for model sets .
recall that a @xmath14-adic integer can be written as a formal series @xmath382 with integral coefficients @xmath383 satisfying @xmath384 ( _ hensel expansion _ ) . for the following ,
we identify a @xmath14-adic integer @xmath385 with the sequence @xmath386 of its coefficients .
the set of all @xmath14-adic integers ( a ring ) is written as @xmath387 , while the field of @xmath14-adic numbers is written as @xmath388 and can be seen as the set of all laurent series @xmath389 with @xmath390 .
there is a @xmath14-adic _ valuation _ @xmath391 defined by @xmath392 , which gives rise to the @xmath14-adic _ metric _ with @xmath393 and @xmath394 ( and @xmath388 and @xmath387 are the completions of @xmath395 and @xmath341 with respect to the @xmath14-adic metric ) .
so with respect to the @xmath14-adic metric , two numbers in @xmath341 are close if their difference is divisible by a high power of @xmath14 . note that this is a _ non - archimedean _ absolute value ( i.e. , @xmath396 for all @xmath397 ) and we therefore get some `` strange '' properties : all triangles are isosceles , every point inside a ball @xmath398 is the center of this ball , all balls are open and closed , etc . , see @xcite . for euclidean models ( see ( * ? ? ?
* section 1.2 ) ) of @xmath387 we only need to know the formal series @xmath399 .
we can even show models for @xmath120 , where we do not make use of ( [ eq : ladic ] ) .
for this , we use the addressing scheme known for fractals , for example in the sierpinsky gasket , see ( * ? ? ? * chapter iv . ) : = 0.65 now the interesting thing here is that each point in the sierpinsky gasket has a unique address ( at least if we do not take the usual connected sierpinsky gasket but the totally disconnected version ; this can be obtained by using a contraction factor less than @xmath400 in the ifs for the sierpinsky gasket ) .
so each point in the sierpinsky gasket corresponds to a sequence @xmath386 with elements @xmath401 this is just the hensel expansion of the @xmath99-adic integers .
similarly , the cantor set is such a geometric encoding of the @xmath5-adic integers .
`` reasonable '' geometric representations of @xmath120 in @xmath402 are those , where the sets @xmath403 of points starting with the same address are represented by objects of the same size for a fixed @xmath404 . therefore we get that in @xmath405-dimension , @xmath120 with @xmath406 can reasonably be represented , if we do not make use of ( [ eq : ladic ] ) .
note that we can represent @xmath407 either in @xmath408 or @xmath409 .
this geometric representation surely fails for some @xmath14-adic ( or @xmath0-adic ) properties ( all triangles are isosceles , every point inside a ball is its center , etc . ) , but some are also `` preserved '' : points which are close in the @xmath14-adic topology are also close in this geometric representation and the representation as totally disconnected fractal corresponds to the totally disconnected field @xmath388 , @xmath387 and its geometric models are both compact sets . and balls in the @xmath14-adic topology correspond to scaled down copies of the whole fractal .
we like to conclude this section with our example from ( [ eq : kol42 ] ) .
the @xmath99-adic geometric models are given in figure [ fig : kol42 ] .
observe that , in the two - dimensional representation , the parts ( according to our above addressing scheme for the sierpinsky gasket ) @xmath410 , @xmath411 and @xmath412 are colored by only one color .
this corresponds to the fact that at positions @xmath413 are @xmath150 s , on @xmath414 are @xmath415 s and on @xmath416 are @xmath162 s only in the bi - infinite sequence .
so , large patches of the same color in the geometric representations correspond to lattice cosets @xmath417 with small @xmath418 .
a similar addressing scheme can be used for the one - dimensional representation ( and in fact for all @xmath0-adic representations ) .
= 0.5 = 0.5
it is a pleasure to thank m. baake , r.v .
moody and o. redner for discussions and the german research council ( dfg ) for financial support .
fogg , `` substitutions in dynamics , arithmetics and combinatorics '' , _ lecture notes in mathematics _ * 1784 * , edited by v. berth , s. ferenczi , c. mauduit and a. siegel , springer , 2002 ; available at . e.a .
robinson , jr .
, `` the dynamical theory of tilings and quasicrystallography '' , in _ ergodic theory of @xmath419-actions _ , edited by m. pollicott and k. schmidt , cambridge u. p. , cambridge , 1996 , pp . 451473 . | we consider ( generalized ) kolakoski sequences on an alphabet with two even numbers . they can be related to a primitive substitution rule of constant length @xmath0
. using this connection , we prove that they have pure point dynamical and pure point diffractive spectrum , where we make use of the strong interplay between these two concepts .
since these sequences can then be described as model sets with @xmath0-adic internal space , we add an approach to `` visualize '' such internal spaces . |
the dynamics and pinning of vortices in type - ii superconductors is arguably one of the most widely investigated phenomena in the field of superconductivity . recently
, particular attention has been devoted to the possibility of controlling the strength of the vortex pinning using different magnetic templates@xcite . within the london formalism , the interaction between a vortex line and a finite size permanent magnet
is given by@xcite , @xmath0 where the integration is carried out over the volume of the ferromagnet , @xmath1 indicates the position of the vortex line , @xmath2 its spatial dependent magnetic moment , and @xmath3 is the field generated by the vortex line .
this equation shows that the pinning potential not only depends on the size of the dots@xcite but also on their exact magnetic state@xcite .
in particular , if the superconducting penetration depth @xmath4 is on the order of the lateral size of the dot , a weak average pinning is expected when the magnetic dot contains multiple domains .
in contrast to that , a maximum pinning should be achieved for single domain structures . in order to retain these properties in big arrays of magnetic particles
it is necessary to ensure a minimum distance between them in such a way that the field profiles of neighboring elements do not overlap and can be resolved at scales of the magnetic penetration length @xmath4 .
notice that according to eq.([uvm - dot ] ) it is possible to increase strongly the pinning by just growing continuously the magnetic moment of the dots .
this growth is limited by the fact that above certain magnetic moment , a vortex - antivortex pair will be induced by the magnetic element with the consequent change in the pinning properties@xcite .
for instance , dots with perpendicular magnetic moment induce a vortex on top of the dots whereas the accompanying antivortex is located at interstitial positions , in between the dots . as a result
, vortices and antivortices feel a different pinning potential which leads to a distinct dynamic response@xcite .
since the weakly pinned interstitial antivortices limit the maximum critical current @xmath5 , their annihilation with an external field produces a shift of the field position of the @xmath5 towards @xmath6 , where @xmath7 is the number of vortex - antivortex pairs generated by a single magnetic element and @xmath8 is the field corresponding to one external vortex per unit cell . since @xmath8 is a geometrically predefined parameter , the field loci of the @xmath5 is a direct measurement of the number @xmath7 of bounded pairs .
unfortunately , in the case of micromagnets with in - plane magnetic moment , the field polarity symmetry of the system conceals this information from dc electro - transport experiments . in this work ,
we introduce alternative methods to accurately determine the number @xmath7 of vortex - antivortex pairs induced by the magnetic template , based on the ac - dynamic response of the hybrid system .
indeed , the fact that the critical force to bring the vortex and antivortex closer differs from that necessary to pull them apart@xcite , gives rise to a clear rectification effect when the system is submitted to a zero average ac - signal .
this rectified voltage is maximum at zero external field@xcite and progressively decreases with increasing field until all vortices ( or antivortices ) are compensated . at rf frequencies the mode locking or synchronized motion of vortices produces distinct voltage steps in the current - voltage characteristics with separation determined by the number of moving vortices .
both techniques indicate that for the particular system here investigated , the magnetic template generates one single vortex - antivortex pair per unit cell . the crucial conditions for the success of these methods are ( i ) a periodic landscape , i.e. all vortices and antivortices feeling the same environment , and ( ii ) a free flux flow motion of vortices .
although the predefined topological landscape is by definition periodic , the magnetic landscape for the as - grown state consists of a highly disorder distribution of magnetic poles .
this disordered state can be switched to a perfectly ordered magnetic lattice by magnetizing the sample with an in - plane field .
the order - disorder transition , already reported for other s / f systems@xcite , produces a profound modification in the vortex dynamics .
more specifically , in the disordered state , non - linear current - voltage characteristics with high critical currents are observed , whereas a very broad flux flow regime accompanied by a sharp decrease of the critical currents is obtained in the ordered state .
this behavior can be attributed to the reduction of the average hopping distance determined by the separation between neighboring magnetic poles of identical polarity .
in addition , the presence of a flux flow regime at zero applied field reinforces the idea that the investigated magnetic structure is able to induce vortex - antivortex ( v - av ) pairs .
the sample investigated consists of a periodic array of permalloy ( ni@xmath9fe@xmath10 ) equilateral triangles of 1 @xmath11 m side and a thickness of 25 nm .
along one of the principal directions of the array ( @xmath12-axis ) the separation between the triangles is 1 @xmath11 m whereas in the perpendicular direction ( @xmath13-axis ) the triangles are touching [ see fig . [ fig - sample](a ) ] .
this geometry gives rise to two well defined directions of the structure , parallel ( @xmath13 ) and perpendicular ( @xmath12 ) to the line connecting nearby triangles , as indicated in fig .
[ fig - sample](a ) .
a 50 nm thick superconducting al bridge deposited on top of the magnetic template also covers the unpatterned substrate [ see fig . [ fig - sample](b ) ] thus allowing us to compare directly patterned vs reference plain film and identify the influence of the magnetic triangles on the superconductor .
both , the array of triangles and the transport bridge are aligned at submicrometer scale and were prepared by standard lithographic techniques .
the al plain film has a critical temperature at zero field @xmath14 k , with residual resistivity ratio @xmath15k@xmath16k@xmath17 , and superconducting coherence length @xmath18(0)@xmath19 nm .
it is a crucial point that neighboring triangles overlap in one direction in order to eventually achieve two well distinguished phases : disordered in the as - grown state and ordered in the magnetized state . indeed , magnetic force microscopy ( mfm ) images shown in fig .
[ fig - sample](c ) for non touching triangles demonstrate that the remanent state , irrespective of the magnetic history of the triangles , is always a vortex state , in agreement with previous reports on individual micrometer size magnetic structures@xcite .
in contrast to that , if there is a physical contact between neighboring triangles a new set of possible states can be found .
[ fig - sample](d ) shows a mfm image of the sample investigated in the as - grown state . here a clearly disordered arrangement of the , so called , buckle states@xcite is observed .
interestingly , after magnetizing the sample with an homogeneous in - plane field perpendicular to the chain of triangles [ fig .
[ fig - sample](e ) ] , these buckle states can be set in a perfectly ordered arrangement .
the in - plane distribution of magnetic moments calculated by oommf micromagnetic simulations@xcite is indicated in fig .
[ fig - sample](f ) with black arrows whereas the @xmath20-component of the stray field ( @xmath21 ) is shown in fig .
[ fig - sample](g ) . in all cases the magnetic state of each triangle
is characterized by a strong out of plane magnetic pole which is located at one of the two bottom corners of the py triangles and a weaker magnetic pole , with opposite polarity , located at the top corner of the triangle [ fig .
[ fig - sample](e ) ] .
it is precisely this magnetic pole which can be controlled to lie on either side of the triangle by simply changing the direction of magnetization .
notice that this magnetic monopole - like structure breaks the field polarity symmetry of the s / f hybrid system .
the question now arises as whether the distribution of the magnetic dipoles ( ordered versus disordered ) has any effect on the properties of the adjacent superconducting film .
let us first analyze the nucleation of the superconducting order parameter for the two different magnetic states of the hybrid system .
[ fig - phaseboundary ] summarizes the obtained superconductor / normal metal phase boundaries for a dc - current of 100 @xmath11a and using 90@xmath22 criterion of the normal state resistance @xmath23 for different magnetic states and current orientations .
higher criteria exhibit similar results . for the as - grown state ( open circles ) a pronounced decrease of the critical temperature at low fields with respect to the plain film is observed . a further suppression of @xmath24 is obtained after the sample has been magnetized with an in - plane field parallel to the @xmath12 direction ( solid triangles ) .
the fact that the phase boundary of the as - grown state lies in between those corresponding to the plain film and the fully polarized buckle state , suggests that in the virgin sample , randomly oriented buckle states coexist with magnetic vortex states .
this has been indeed confirmed by the mfm images taken at different locations of the pattern .
resistance criterion of the normal resistance @xmath23 and dc - current of 100 @xmath11a for the as - grown state ( open circles ) and different magnetized states and current orientations .
the dashed line shows the corresponding transition lines for the co - evaporated al plain film . the solid line is a fit assuming a lateral confinement of the superconducting condensate ( see text).,scaledwidth=45.0% ] it is worth noticing that for the magnetized sample the @xmath25 boundary exhibits a parabolic background similar to that reported in earlier investigations@xcite .
this behavior can be ascribed to the change of dimensionality when @xmath26 exceeds the size @xmath27 of the regions where superconductivity first nucleates . within this regime
the phase boundary can be approximated by @xmath28 , with @xmath29 . by using this expression to fit the data in fig .
[ fig - phaseboundary ] ( solid line ) we found @xmath30 @xmath11 m which is consistent with the largest separation between neighboring magnetic poles .
this behavior suggests that for the magnetized sample the stray fields confine the superconducting condensate in between the magnetic poles and consequently the superconducting order parameter will be modulated into a sequence of quasi one - dimensional stripes .
this prediction is in agreement with complementary transport measurements for two orthogonal orientations of the applied current , perpendicular to the chain of triangles ( solid triangular symbols ) and parallel to them ( open triangular symbols ) as shown in fig .
[ fig - phaseboundary ] indicating that the onset of superconductivity is independent on the direction of the applied current .
in the previous section we showed that by switching from the as - grown ( disordered phase ) to the magnetized ( ordered phase ) state , a proliferation of magnetic poles takes place which in turn gives rise to a decrease of @xmath31 . in addition , the dimensional crossover from a 2d disordered phase to a 1d ordered phase has little or no influence on the nucleation temperature .
however , deeper into the superconducting state where the electric response is dominated by the vortex dynamics , the one dimensional nature of the field modulation should become more apparent . in order to address this issue
we have measured the voltage - current characteristics [ @xmath32 with the bias current applied in the @xmath12-direction at @xmath33 and @xmath34 k for the two different magnetic states : as - grown ( open circles ) and magnetized ( solid squares ) states as shown in fig .
[ vi ] ( a ) .
the most obvious feature in this figure is the enormous difference ( more than an order of magnitude ) between the critical current of the magnetized state ( 24 @xmath11a ) with respect to the demagnetized state ( 490 @xmath11a ) .
furthermore , the vortex dynamics is severely modified after magnetizing the sample . indeed , whereas in the as - grown state there is a narrow window of currents exhibiting a highly non - linear @xmath35 dependence , where the vortex lattice transit from pinned phase to the normal state , in the magnetized sample a broad flux - flow regime is observed immediately above the critical current . fig .
[ vi](b ) and ( c ) summarize the field dependence of the different dynamical phases for the as grown sample [ panel ( b ) ] and after magnetization [ panel ( c ) ] .
the onset of vortex motion is delimited by a resistance criterion of 10@xmath22 r@xmath36 while a 90@xmath22 r@xmath36 criterion is used to indicate the transition to the normal state .
characteristic with current applied in the @xmath12-direction , at @xmath37 and a temperature of 1.1 k for the as - grown and the magnetized states .
the lower panels show the magnetic field dependence of the critical current and the transition to the normal state as determined by criterion of 10@xmath22 r@xmath36 and 90@xmath22 r@xmath36 , respectively , for the as - grown ( b ) and magnetized ( c ) states.,scaledwidth=45.0% ] notice that the presence of matching features , which require two - dimensional phase coherence , together with the fact that in the magnetized state a finite critical current is needed to depin the vortices , rule out the possibility of having normal metal stripes in the hybrid sample .
instead , the observed difference between both phases can be attributed to the formation of easily moving channels of vortices and antivortices created by the magnetic structure .
indeed , in the ordered phase the potential wells created by the magnetic dipoles are perfectly aligned and strongly overlap , thus creating an easy channel for flux motion .
in contrast to that , in the disordered phase , the average distance between two neighboring magnetic pinning centers of equal polarity is larger thus impeding the easy motion of vortices . from now on we will focus on the dynamic response of the system in the ordered state .
a rough criterion to determine whether the stray field emanating from the magnetic structure is strong enough to induce v - av pairs can be obtained by comparing it to the penetration field @xmath38 resulting from bean - livingston barriers@xcite .
the highest barrier possible is limited by the thermodynamic critical field @xmath39 . using @xmath40 nm and @xmath41
nm we obtain @xmath42 mt which lies well below the field intensity ( @xmath43 100 mt ) produced by the magnetic template at the superconducting film . based on this estimation we can safely conclude that the magnetic structure is able to induce , at least , one vortex - antivortex per unit cell .
red ( green ) circles represent positive ( negative ) vortices .
, scaledwidth=45.0% ] this finding suggests that the ordered phase consists of two well defined channels with opposite vortex polarity as schematically shown in fig .
[ schematics](a ) . since the magnetic template breaks the field polarity symmetry , in principle vortices and antivortices should experience a different depinning current .
moreover , as shown in fig .
[ schematics](b ) , under these circumstances an external field will compensate vortices in the channel with opposite polarity ( thus introducing disorder in this particular channel and increasing the average hopping distance ) and as a result the high vortex mobility will be suppressed .
in contrast to that , the row with vortices of opposite polarity would remain unaltered .
this scenario predicts two anomalous behaviors : the existence of a flux - flow response at zero field and a field independent flux flow regime as long as none of the two rows of vortices is fully compensated by the external field .
both predictions are in agreement with the field evolution of the @xmath35 measurements done at 1.1 k. the slope of the linear regime , proportional to the amount of particles contributing to the easy flow , is plotted as a function of field in fig .
[ comparison](a ) . in comparison to the slope expected for the free flux flow model @xmath44 [ straight line in fig . [ comparison](a ) ] , where @xmath45 is the normal state resistance and @xmath46 is the upper critical field , a weak field dependence of the flux flow slope is seen at low field values .
= 1.1 k. the straight lines indicate the expected dependence according to free flux flow model .
( b ) critical current as a function of magnetic field at a temperature of 1.1 k for both current directions .
a clear anisotropic response is observed.,scaledwidth=45.0% ] further evidence in favor of the existence of channels populated with v - av pairs comes from the measurements of the critical current for currents applied parallel ( @xmath47 ) and perpendicular ( @xmath48 ) to the channels of v - av pairs .
these results are shown in fig .
[ comparison](b ) , with a critical current determined using a criterion of 100 @xmath11v .
this figure shows a strong in - plane anisotropy @xmath49 induced by the magnetic template which results from the easier motion of vortices along the line of magnetic poles than perpendicular to them .
this anisotropy amounts to a factor of 5 at non matching fields and is minimum at the commensurability fields .
the scenario described in the previous section where rows of vortices and antivortices generated by the magnetic template coexist , makes our system similar to the array of elongated magnetic bars studied by de souza silva _
et al._@xcite .
this system has been modeled within the london approximation by carneiro@xcite for the case where no v - av pairs are induced , and more recently by lima and de souza silva@xcite assuming that v - av pairs emanate from the dipoles .
irrespective of whether the magnetic template is able to induce v - av pairs or not , both situations lead to a rectification of the net vortex motion @xmath50 when a zero average ac - force is applied parallel to the dipolar moment ( i.e. along the line connecting a vortex with its corresponding antivortex)@xcite .
however , clear differences in the field dependence of the average velocity are expected . indeed ,
if v - av pairs are present then ( i ) a maximum rectification is expected at zero field , ( ii ) unlike standard ratchet signal , the sign of the average voltage readout should be field polarity independent , ( iii ) as vortices ( antivortices ) are progressively compensated by increasing the external negative ( positive ) field , the rectified signal should decrease . in the temperature - field @xmath51 plane for the magnetized sample subjected to a 1 khz sinusoidal
ac current in the @xmath13 direction of amplitude 100 @xmath11a.,scaledwidth=45.0% ] the measured rectification effect , shown in fig.[ratchet ] , confirms all the above mentioned facts and indicates that the v - av pairs are more easily brought together than separated .
furthermore , if the applied field is increased above the first matching field a sign reversal occurs .
this sign reversal at @xmath8 can not be explained assuming more than one v - av pair created by the magnetic template since in this case the v - av annihilation process would be unlikely interrupted by the partial compensation of one row of vortices .
in contrast to that , if there is a single v - av pair per unit cell , once a whole line is compensated by the external field , a dramatic change in the rectification properties is expected . although this finding seems to indicate that the magnetic structure is able to induce a single v - av per triangular element , a more convincing evidence of this prediction can be obtained only by monitoring the individual motion of vortices at higher frequencies .
the motion of vortices in a periodic potential gives rise to a variety of interesting dynamical effects such as the generation of rf radiation@xcite and synchronization / resonance effects of the vortex lattice@xcite when a rf current @xmath52 is superimposed onto a dc drive @xmath53 .
the latter effect gives rise to a series of plateaus in the @xmath35 measurements very similar to the shapiro steps observed in joshepson junctions@xcite .
the relevant timescale of these processes is determined by the washboard frequency , @xmath54 with @xmath55 the period of the potential and @xmath50 the average velocity of the particles .
the investigation of the above mentioned phenomena can be used to obtain additional information on the dynamics of the vortex lattice@xcite , the periodicity of the underlying potential and the density of particles in the system@xcite . in other words ,
high frequency experiments provide further insight into the microscopic motion of vortices not available with the averaged response presented in previous sections . the synchronous motion of vortices resulting from the resonant condition @xmath56 , known as mode locking effect , allows one to obtain the exact number of v - av pairs created by the magnetic template .
indeed , the value of the average velocity is given by : @xmath57 with @xmath55 the periodicity of the lattice in the direction of motion , @xmath58 the frequency of the rf current and @xmath59 the number of cells that a vortex has traveled during a rf cycle . as we mentioned above , the fingerprint of this type of resonant motion is the presence of clear plateaus in the @xmath60 characteristics separated by voltage steps , @xmath61 with @xmath62 the number of vortices in each unit cell and @xmath63 the number of rows between two voltage contacts .
measuring the voltage separation between consequtive steps and using eq.([discrete_steps ] ) we can estimate the density of vortices moving coherently in the flux flow regime .
measurement ( filled symbols ) at @xmath64 = 0.5 mt and @xmath65 = 1.1 k with a small ac excitation of 100 mhz .
panel ( b ) shows a @xmath66 , the discrete peaks are labeled according to eq.([discrete_steps]).,scaledwidth=45.0% ] fig . [
single measurement shapiro](a ) shows a @xmath35 scan for @xmath670.5 mt and @xmath681.1 k with a small ac excitation @xmath69 of 100 mhz superimposed .
the presence of shapiro steps in the linear part of the @xmath35 curve becomes more apparent by plotting @xmath70 versus @xmath66 [ see fig . [ single measurement shapiro](b ) ] .
these steps represent a clear indication of the coherent flow of vortices . in fig .
[ single measurement shapiro](b ) we assign the present peaks to values of @xmath71 with @xmath59 ranging from 1 to 4 ( @xmath72 , @xmath73 , @xmath74 and @xmath75 ) . from their linear dependence on @xmath59 , shown in the inset of fig .
[ frequency dependence ] , we calculate a value of @xmath76 . an independent estimation can be obtained by tracking the frequency dependence of each voltage plateau @xmath77 .
this is shown in fig .
[ frequency dependence ] for the first two steps .
the linear fit using eq.([discrete_steps ] ) gives @xmath78 and 1.18 for @xmath79 and @xmath80 , respectively .
it is important to notice that , in general , two ingredients are necessary for the presence of mode locking effects .
firstly , the existence of a flux flow regime , which ensures a narrow distribution of average vortex velocities@xcite .
secondly , there should be a periodic perturbation of the pinning potential , otherwise the distortion of the vortex lattice becomes too strong and its coherent motion is destroyed . in principle , this second condition is only fulfilled at the matching field , @xmath81 mt .
this seems to be in contradiction with the fact that clear shapiro steps are present at 0.5 mt , a field which would corresponds to 0.42 vortices per unit cell .
this apparent paradox can be solved invoking once again the presence of v - av pairs induced by the triangles . in this case
, the applied field will annihilate the vortices in the channel with opposite polarity , leading to an increase of their critical current and consequently suppressing their contribution to the coherent motion . on the other hand
, the row with equal polarity will be unaltered by the applied magnetic field and therefore should exhibit a region of field independent shapiro steps .
in other words , the incoming vortices generated by the external magnetic field play no role in the coherent oscillatory motion of vortices thus giving rise to a constant voltage separation between two consecutive shapiro steps .
( circular symbols ) and @xmath80 ( square symbols ) at a field of 0.5 mt and temperature of 1.1 k. inset : @xmath82 measured at 100 mhz.,scaledwidth=45.0% ] versus @xmath70 at a temperature of 1.1 k and an applied rf frequency of 100 mhz.,scaledwidth=45.0% ] in order to check this hypotheses of self organized commensurability we measured the field dependence of the shapiro steps , which are better visualized as a peak in the @xmath66 as previously shown in fig .
[ single measurement shapiro](b ) .
the results of these measurements for @xmath681.1 k are summarized in fig .
[ magnetic field dependence ] as a contour plot in the field - voltage plane , with maximum intensity representing the exact location of the @xmath71 peaks . for magnetic field values ranging from @xmath83
to @xmath84 a clear field independence peak position is observed .
this result confirms that the amount of particles , moving coherently , is independent of the applied external field .
it has been predicted by molecular dynamics simulations that a vortex lattice in a periodic landscape of asymmetric pinning sites can unveil the discrete hopping of vortices when the system is excited with a zero mean rf drive@xcite .
so far , most of the experimental reports evidence a vortex rectification effect obtained in the low frequencies limit which results from an averaging in time and are not able to resolve the individual motion@xcite . in this section
, we will try to bridge this gap by constructing an artificial vortex rectifier resulting from a dc tilt , which breaks the inversion symmetry of our system@xcite .
additionally , this tilt allows us to probe the discrete motion of the vortex lattice without the need of a time resolved signal read out .
= 100 mhz ) with a dc offset of 10 @xmath11a applied .
the indexing ( @xmath85 ) indicates the number of steps forward @xmath86 and backward ( b).,scaledwidth=45.0% ] fig .
[ signle_ac_sweep ] shows the dc voltage as a function of the applied ac drive with an additional dc current on top ( @xmath53 = 10 @xmath11a ) . due to the applied dc tilt the critical current , needed to depin vortices , is smaller for positive than for negative currents ( @xmath87 ) .
depending on the strength of the applied ac amplitude @xmath88 we can determine three different dynamic regimes .
for small ac excitations ( @xmath89 ) the driving force is unable to depin the vortices and as a result the dc voltage is zero . by increasing the ac amplitude above @xmath90 the vortex lattice is set in the flux flow regime and mode locking effects result in an oscillatory dependence of the dc voltage as a function of the ac amplitude .
finally at high ac amplitudes the normal state is reached and @xmath91 .
each maxima and minima in dc voltage can be attributed to the discrete motion of the vortex lattice in the positive and negative cycle of the driving amplitude . to identify the motion
, we introduce the indexing ( @xmath85 ) in fig .
[ signle_ac_sweep ] , with @xmath92 and @xmath93 the number of steps in the positive and negative direction respectively . at @xmath88 just above @xmath90 the driving force is strong enough to move the vortex lattice one unit cell in the positive direction while no motion occurs in the negative cycle of the ac drive and a maxima in the dc voltage output occurs .
if the ac amplitude passes @xmath94 a single back and forth step of the vortex lattice occurs and as such a strong reduction of the dc voltage is measured .
this difference between back and forth motion of the vortex lattice , due to a small dc tilt , results in a discrete stepping procedure up to the sixth order . and @xmath53 dependence of the dc voltage at a magnetic field value of 1.195 mt and a rf modulation of 150 mhz , scaledwidth=45.0% ] fig .
[ iac_idc_dependence ] shows the dependence of the dc voltage , normalized to @xmath79 , as a function of the ac drive and the dc tilt . at small dc tilts ( below @xmath95 100 @xmath11a )
the behavior is exactly the same as shown in fig .
[ signle_ac_sweep ] . in this region ,
the applied dc current creates a small asymmetry and the vortex lattice , depending on the exact value of the ac drive , can either move one step extra in the positive direction ( maxima in dc voltage ) or an equal amount of steps in both directions ( minima in dc voltage ) . around 100
@xmath11a the oscillations vanish ( white region ) and the dc voltage readout amounts @xmath79 .
the reason for this behavior is a subtle balance between the ac drive and the dc tilt .
exactly at the point were the ac amplitude is strong enough to move the vortex lattice in the hard direction ( i.e. against the dc tilt ) it is also sufficiently strong to move the vortex lattice an additional step in the ` easy ' direction .
therefore the overall motion of the vortex lattice will be independent on the applied ac drive and equal to 1 .
this subtle balance is destroyed in favor of ` easy ' flow if the applied dc current is increased above 100 @xmath11a . again the dc voltage readout ( @xmath96 ) indicates that the vortex lattice moves two steps ( maxima in dc voltage ) or one step ( minima in dc voltage ) extra in the positive direction . by increasing the dc tilt
even further we could even see a difference of 4 unit cells between the motion in the positive and negative direction .
we have demonstrated that the particular geometry of the two - fold symmetric array of magnetic triangles can induce an order - disorder transition as the system switches from the as - grown state to the magnetized state .
the disordered phase is characterized by a random distribution of two opposite chirality magnetic buckle states , whereas in the order state all triangles are in a well defined unique magnetic buckle state .
the latter phase exhibits unidimensional lines of strong stray fields which represent channels for easy flux motion resulting in a clear anisotropy of the critical current .
a quantitative estimation shows that the current needed to induce inter - channels hopping is about 5 times larger than that needed for setting intra - channel motion .
it is important to stress that even though we have presented clear evidence of the influence of the ordering of magnetic dipoles induced by the magnetic templating on the vortex dynamics , it is still uncertain whether the ultimate source of disorder in the as - grown state is the random distribution of magnetic poles in the buckle state or the presence of triangles in the magnetic vortex state along one line of connecting triangles .
in addition , we showed that the high mobility of the vortex lattice is a consequence of the generation of vortex - antivortex pairs by the magnetic template .
this picture was confirmed experimentally by high frequency transport measurements and low frequency ratchet experiments . due to the self organized commensurability of these channels a pronounced mode locking effect is observed .
these features were used to probe and tune the discrete motion of the composite lattice of vortices and antivortices .
this combination of techniques represent a powerful tool for understanding the motion of vortices at microscopic scale .
this work was supported by the k.u.leuven research fund goa/2004/02 program , nes esf program , the belgian iap , the fund for scientific research flanders ( f.w.o.vlaanderen ) .
v.m . acknowledge funding support from u.s .
nsf , grant eccs-0823813 ( vm ) .
a.v.s . and j.v.d.v .
are grateful for the support from the fwo - vlaanderen .
finally the authors also would like to acknowledge help of scientec and nanotec with the mfm images .
silhanek , w. gillijns , v.v .
moshchalkov , v. metlushko , b. ilic , appl .
. lett . * 89 * , 182505 ( 2006 ) ; n. verellen , a. v. silhanek , w. gillijns , v.v .
moshchalkov , v. metlushko , f. gozzini , and b. ilic , appl .
. lett . * 93 * , 022507 ( 2008 ) ; a. hoffmann , l. fumagalli , n. jahedi , j. c. sautner , j. e. pearson , g. mihajlovic , v. metlushko , phys .
b * 77 * , 060506 ( 2008 ) ; j. e. villegas , k. d. smith , lei huang , yimei zhu , r. morales , i. k. schuller , phys .
b * 77 * , 134510 ( 2008 ) ; v. vlasko - vlasov , u. welp , g. karapetrov , v. novosad , d. rosenmann , m. iavarone , a. belkin , and w .- k .
kwok , phys .
b * 77 * , 134518 ( 2008 ) ; a. belkin , v. novosad , m. iavarone , j. pearson , and g. karapetrov , phys . rev .
b * 77 * , 180506(r ) ( 2008 ) .
d. g. gheorghe , r. j. wijngaarden , w. gillijns , a. v. silhanek , and v. v. moshchalkov , phys .
b * 77 * , 054502 ( 2008 ) ; w. gillijns , a.v .
silhanek , v.v .
moshchalkov , phys .
b * 74 * , 220509(r ) ( 2006 ) . | the vortex dynamics in a low temperature superconductor deposited on top of a rectangular array of micrometer size permalloy triangles is investigated experimentally .
the rectangular unit cell is such that neighboring triangles physically touch each other along one direction .
this design stabilizes remanent states which differ from the magnetic vortex state typical of individual non - interacting triangles .
magnetic force microscopy images have revealed that the magnetic landscape of the template can be switched to an ordered configuration after magnetizing the sample with an in - plane field .
the ordered phase exhibits a broad flux flow regime with relatively low critical current and a highly anisotropic response .
this behavior is caused by the spontaneous formation of two separated rows of vortices and antivortices along each line of connected triangles .
the existence of a clear flux flow regime even for zero external field supports this interpretation .
the density of induced vortex - antivortex pairs is directly obtained using a high frequency measurement technique which allows us to resolve the discrete motion of vortices .
strikingly , the presence of vortex - antivortex rows gives rise to a self organized synchronized motion of vortices which manifests itself as field independent shapiro steps in the current - voltage characteristics . |
high temperature superconductivity ( htsc ) is one of the most intriguing phenomena in modern solid state physics .
experimentally , htsc is observed in layered cuprate compounds .
the undoped cuprates are antiferromagnetically ordered insulators which develop the characteristic superconducting `` dome '' upon doping with charge carriers .
htsc is interesting not only for promising technological applications , but also from a theoretical point of view .
the relevant ingredients for htsc are believed to be the following . *
low dimensionality ( 2d ) .
* strong short - range repulsion between electrons .
* doped mott insulator . taking these 3 ingredients ,
htsc is modeled in the tight - binding description by large-@xmath2 hubbard models or @xmath0-@xmath1 models on the square lattice : @xmath3 where @xmath4 , @xmath5 and @xmath6 are the pauli matrices .
the gutzwiller projector @xmath7 prevents electrons from occupying the same lattice site .
the non - perturbative nature of the @xmath0-@xmath1 model makes it an outstanding problem to solve in dimensions larger than one .
analytical techniques ( renormalized or slave - boson mean - field theories @xcite ) are very crude and numerical techniques ( e.g. exact diagonalization or cluster dmft ) are restricted to very small clusters or infinite dimensions , or they fail on the sign problem ( qmc ) .
an alternative approach was suggested by anderson shortly after the experimental discovery of htsc , when he proposed a gutzwiller - projected bcs wave function as superconducting ground state for cuprates @xcite . following this conjecture ,
many variational studies have been performed on the basis of what is called anderson s ( long range ) rvb state .
this state turned out to have very low variational energy , close to exact ground state energies , as well as high overlap with the true ground states of small @xmath0-@xmath1 clusters @xcite . on the other hand , many experimental facts about cuprate superconductors can be reproduced and are consistent with the variational results : e.g. clearly favored @xmath8-wave pairing symmetry , doping dependency of the nodal fermi velocity and the nodal quasiparticle weight .
many of these successful efforts following anderson s proposal are summarized in the `` plain vanilla rvb theory '' of htsc , recently reviewed in @xcite . with help of the relatively recent technique of angle - resolved photoemission spectroscopy ( arpes ) , experimentalists can probe the electronic structure of low - lying excitations inside the copper planes .
the intensity measured in arpes is proportional to the one particle electronic spectral function : @xmath9 @xcite .
it is therefore interesting to explore spectral properties within the framework of gutzwiller - projected variational quasiparticle ( qp ) excitations . in this contribution we will discuss some of our results reported in @xcite . for more details , in particular for more reference to experimental studies , we invite the reader to consult that paper .
anderson s rvb state is given by @xmath10 we further define projected bcs quasiparticle excitations in a similar way , @xmath11 the unprojected states in eqs . and are the usual ingredients of the bcs theory , @xmath12 , @xmath13 , @xmath14 , @xmath15 , @xmath16 - \mu$ ] , @xmath17 $ ] .
@xmath18 is the gutzwiller projector and @xmath19 projects on the subspace with @xmath20 holes . @xmath21 and @xmath22 are normalized to one .
the wave functions have two variational parameters , @xmath23 and @xmath24 , which we adjusted to minimize the energy of the @xmath0-@xmath1 hamiltonian for the experimentally relevant value @xmath25 and every doping level .
note that in the rvb theory , @xmath23 and @xmath24 are variational parameters without direct physical significance ; physical quantities like excitation gap , superconducting order , or chemical potential must be calculated explicitly .
the spectral weights of the coherent low - lying quasiparticles can be written as [ eq : weights]@xmath26 these weights are measured in arpes experiments as the residues of the spectral function @xmath27 @xcite . note that in conventional bcs - theory , @xmath28 and @xmath29 .
the variational monte carlo technique ( vmc ) allows to evaluate fermionic expectation values of the form @xmath30 for a given state @xmath31 . in order to calculate the spectral weights by vmc
, the following exact relations can be used .
[ eq : z]@xmath32 where @xmath33 is hole doping @xcite .
we use vmc to calculate the superconducting order parameter @xmath34 , as well as diagonal matrix elements in the optimized @xmath0-@xmath1 ground state . using relations
, we can then derive the spectral weights .
the disadvantage of this procedure is large errorbars around the center of the brillouin zone where both @xmath35 and @xmath36 are small .
in fig . [ fig : cc_x ] , we plot the nearest - neighbor superconducting orderparameter @xmath37 as a function of doping .
the curve shows close quantitative agreement with the result of ref .
@xcite , where the authors extracted the same quantity from the long - range asymptotics of the nearest - neighbor pairing correlator , @xmath38 . with the method employed here
, we find the same qualitative and quantitative conclusions of previous authors @xcite : vanishing of superconductivity at half filling , @xmath39 , and at the superconducting transition on the overdoped side , @xmath40 .
the optimal doping is near @xmath41 . in the same plot
we also show the commonly used gutzwiller approximation where the bcs orderparameter is renormalized by the factor @xmath42 @xcite .
the gutzwiller approximation underestimates the exact value by approximately @xmath43 .
( calculated in the 14@xmath44 system ) .
the errorbars are smaller than the symbol size . the same quantity calculated in the gutzwiller approximation is also shown for comparison .
the variational parameter @xmath23 is shown with the scale on the right.[fig : cc_x ] ] in fig .
[ fig : cut_z_196 ] , we plot the spectral weights @xmath45 , @xmath46 , and @xmath47 along the contour @xmath48 in the brillouin zone for different doping levels .
figure [ fig : nfs ] shows the contour plots of @xmath47 in the region of the brillouin zone where our method produces small errorbars . from these data
, we can make the following observations .
* in the case of an unprojected bcs wave function , the total spectral weight is constant and unity over the brillouin zone . introducing the projection operator
, we see that for low doping ( @xmath49 ) , the spectral weight is reduced by a factor up to @xmath50 .
the renormalization is asymmetric in the sense that the electronic spectral weight @xmath35 is more reduced than the hole spectral weight @xmath46 . for higher doping ( @xmath51 ) ,
the spectral weight reduction is much smaller and the electron - hole asymmetry decreases . * since there is no electron - hole mixing along the zone diagonal ( @xmath8-wave ) , the spectral weights @xmath35 and @xmath46 have a discontinuity at the nodal point .
our data shows that the total spectral weight is continuous across the nodal point .
strong correlation does not affect these features of uncorrelated bcs - theory . ) , 22 holes ( lower left plot , @xmath52 ) , 34 holes ( upper right plot , @xmath53 ) , and 46 holes ( lower right plot , @xmath54 ) on 196 sites .
the spectral weights are plotted along the contour @xmath48 ( shown in inset ) . plus
signs ( @xmath55 ) denote the spectral weight @xmath56 , crosses ( @xmath57 ) denote @xmath58 , errorbars are shown .
solid dots denote their sum , the total spectral weight @xmath59 , errorbars not shown . on the horizontal axis ,
the star ( @xmath60 ) denotes the intersection with the unprojected fermi surface along the @xmath61 direction ; the thick dot is the nodal point . both @xmath56 and @xmath58
jump at the nodal point , while @xmath59 is continuous .
the intersection with the effective fermi surface ( see text ) is marked by an arrowhead . on the diagonal ( last segment ) , @xmath62 is given in units of @xmath63 .
[ fig : cut_z_196 ] ] ) , 22 holes ( lower left plot , @xmath52 ) , 34 holes ( upper right plot , @xmath53 ) , and 46 holes ( lower right plot , @xmath54 ) on 196 sites .
the spectral weights are plotted along the contour @xmath48 ( shown in inset ) . plus
signs ( @xmath55 ) denote the spectral weight @xmath56 , crosses ( @xmath57 ) denote @xmath58 , errorbars are shown .
solid dots denote their sum , the total spectral weight @xmath59 , errorbars not shown . on the horizontal axis ,
the star ( @xmath60 ) denotes the intersection with the unprojected fermi surface along the @xmath61 direction ; the thick dot is the nodal point . both @xmath56 and @xmath58
jump at the nodal point , while @xmath59 is continuous .
the intersection with the effective fermi surface ( see text ) is marked by an arrowhead . on the diagonal ( last segment ) , @xmath62 is given in units of @xmath63 .
[ fig : cut_z_196 ] ] . the effective fs ( full line ) and unprojected fs ( dashed line ) are also shown .
the doping levels are @xmath64 , @xmath65 , @xmath66 , and @xmath67 ( from left to right ) .
the @xmath55 signs indicate points where the values are known within small errorbars ( see section _ method : vmc_).[fig : nfs ] ] in strongly interacting fermi systems , the notion of a fermi surface ( fs ) is not clear at all .
there are , however , several experimental definitions of the fs .
most commonly , @xmath68 is determined in arpes experiments as the maximum of @xmath69 or the locus of minimal gap along some cut in the @xmath70-plane .
the theoretically better defined locus of @xmath71 is also sometimes used .
the various definitions of the fs usually agree within the experimental uncertainties @xcite .
the different definitions of the fs in htsc were recently analyzed theoretically in refs .
@xcite . here
, we propose an alternative definition of the fermi surface based on the ground state equal - time green s functions . in the unprojected bcs state , the underlying fs is determined by the condition @xmath72
. we will refer to this as the _ unprojected fs_. since @xmath73 and @xmath74 are the residues of the qp poles in the bcs theory , it is natural to replace them in the interacting case by @xmath35 and @xmath46 , respectively .
we will therefore define the _ effective fs _ as the locus @xmath75 . in fig .
[ fig : nfs ] , we plot the unprojected and the effective fs which we obtained from vmc calculations .
the contour plot of the total qp weight is also shown .
it is interesting to note the following points .
* in the underdoped region , the effective fs is open and bent outwards ( hole - like fs ) . in the overdoped region ,
the effective fs closes and embraces more and more the unprojected one as doping is increased ( electron - like fs ) . *
luttinger s rule @xcite is clearly violated in the underdoped region , i.e. the area enclosed by the effective fs is not conserved by the interaction ; it is larger than that of the unprojected fs .
* in the optimally doped and overdoped region , the total spectral weight is approximately constant along the effective fs within errorbars . in the highly underdoped region
, we observe a small concentration of the spectral weight around the nodal point ( @xmath76 ) .
large `` hole - like '' fs in underdoped cuprates has also been reported in arpes experiments by several groups @xcite .
it should be noted that a negative next - nearest hopping @xmath77 would lead to a similar fs curvature as we find in the underdoped region .
we would like to emphasize that our original @xmath0-@xmath1 hamiltonian as well as the variational states do not contain any @xmath77 .
our results show that the outward curvature of the fs is due to strong repulsion , without need of @xmath77 .
the next - nearest hopping terms in the microscopic description of the cuprates may not be necessary to explain the fs topology found in arpes experiments .
remarkably , if the next - nearest hopping @xmath77 is included in the variational ansatz ( and not in the original @xmath0-@xmath1 hamiltonian ) , a finite and negative @xmath77 is generated , as it was shown in ref .
apparently , in this case the unprojected fs has the tendency to adjust to the effective fs .
a similar bending of the fs was also reported in the recent analysis of the current carried by gutzwiller - projected qps @xcite .
a high - temperature expansion of the momentum distribution function @xmath78 of the @xmath0-@xmath1 model was done in ref .
@xcite where the authors find a violation of luttinger s rule and a negative curvature of the fs .
our findings provide further evidence in this direction .
a natural question is the role of superconductivity in the unconventional bending of the fs . in the limit @xmath79 , the variational states are gutzwiller - projected excitations of the fermi sea and the spectral weights are step - functions at the ( unprojected ) fs . in a recent paper @xcite
, it was shown that @xmath80 for the projected fermi - sea , which means that the unprojected and the effective fs coincide in that case .
this suggests that the `` hole - like '' fs results from a non - trivial interplay between strong correlation and superconductivity . at the moment
, we lack a qualitative explanation of this effect .
however , it may be a consequence of the proximity of the system to the non - superconducting `` staggered - flux '' state @xcite or to antiferromagnetism near half - filling @xcite .
sb would like to thank the organizers of the _ eleventh training course in the physics of correlated electron systems and htsc _ in salerno , italy , where this work was presented .
we would like to thank george jackeli for many illuminating discussions and continuous support .
we also thank claudius gros , patrick lee , and seiji yunoki for interesting discussions .
this work was supported by the swiss national science foundation .
a. damascelli , z. hussain , and z .- x .
shen , rev .
phys . , 473 ( 2003 ) ; j. c. campuzano , m. r. norman , and m. randeria , in _ physics of superconductors _ , edited by k. h. bennemann and j. b. ketterson ( springer , berlin , 2004 ) , vol .
ii , pp . 167 - 273 ; cond - mat/0209476 .
t. yoshida , x. j. zhou , t. sasagawa , w. l. yang , p. v.bogdanov , a. lanzara , z. hussain , t. mizokawa , a. fujimori , h. eisaki , z .- x .
shen , t. kakeshita , and s. uchida , phys .
lett . * 91 * , 027001 ( 2003 ) . | we analyze the electronic green s functions in the superconducting ground state of the @xmath0-@xmath1 model using gutzwiller - projected wave functions , and compare them to the conventional bcs form .
some of the properties of the bcs state are preserved by the projection : the total spectral weight is continuous around the quasiparticle node and approximately constant along the fermi surface . on the other hand ,
the overall spectral weight is reduced by the projection with a momentum - dependent renormalization , and the projection produces electron - hole asymmetry in renormalization of the electron and hole spectral weights .
the latter asymmetry leads to the bending of the effective fermi surface which we define as the locus of equal electron and hole spectral weight .
address = institute of theoretical physics , ecole polytechnique fdrale de lausanne ( epfl ) , ch-1015 lausanne , switzerland |
we thank werner porod , debottam das and daniel e. lopez - fogliani for fruitful discussions . a.v .
acknowledges support from the anr project cpv - lfv - lhc nt09 - 508531 .
this work was supported by the spanish micinn under grants fpa2008 - 00319/fpa and fpa2011 - 22975 by the multidark consolider csd2009 - 00064 , by the generalitat valenciana grant prometeo/2009/091 and by the eu grant unilhc pitn - ga-2009 - 237920 .
y. fukuda _ et al .
_ [ super - kamiokande collaboration ] , phys .
lett . * 81 * , 1562 ( 1998 ) sno , q. r. ahmad _ et al .
* 89 * , 011301 ( 2002 ) , [ nucl - ex/0204008 ] . s. abe _ et al . _
[ kamland collaboration ] , phys .
lett . * 100 * , 221803 ( 2008 ) [ arxiv:0801.4589 [ hep - ex ] ] . y. ashie _ et al . _
[ super - kamiokande collaboration ] , phys .
lett . * 93 * , 101801 ( 2004 ) [ hep - ex/0404034 ] .
t. schwetz , m. tortola and j. w. f. valle , new j. phys .
* 13 * , 109401 ( 2011 ) [ arxiv:1108.1376 [ hep - ph ] ] .
j. adam _ et al . _
[ meg collaboration ] , phys .
* 107 * , 171801 ( 2011 ) [ arxiv:1107.5547 [ hep - ex ] ] .
k. nakamura _ et al .
_ [ particle data group collaboration ] , j. phys .
g g * 37 * , 075021 ( 2010 ) .
y. kuno , y. okada , rev .
* 73 * ( 2001 ) 151 [ arxiv : hep - ph/9909265 ] .
f. borzumati and a. masiero , phys .
lett . * 57 * , 961 ( 1986 ) .
k. s. babu and c. kolda , phys .
lett . *
89 * , 241802 ( 2002 ) [ hep - ph/0206310 ] .
e. arganda and m. j. herrero , phys .
d * 73 * , 055003 ( 2006 ) [ hep - ph/0510405 ] .
a. faessler , t. s. kosmas , s. kovalenko and j. d. vergados , nucl .
b * 587 * ( 2000 ) 25 .
t. goto , y. okada and y. yamamoto , phys .
d * 83 * ( 2011 ) 053011 [ arxiv:1012.4385 [ hep - ph ] ] .
j. hisano , t. moroi , k. tobe , m. yamaguchi , phys . rev .
* d53 * ( 1996 ) 2442 [ hep - ph/9510309 ] .
e. lunghi , a. masiero , i. scimemi and l. silvestrini , nucl .
b * 568 * ( 2000 ) 120 [ hep - ph/9906286 ] .
w. porod , comput .
commun .
* 153 * ( 2003 ) 275 w. porod and f. staub , arxiv:1104.1573 [ hep - ph ] .
r. n. mohapatra and j. w. f. valle , phys .
d * 34 * , 1642 ( 1986 ) . v. de romeri , m. hirsch and m. malinsky , phys .
d * 84 * ( 2011 ) 053012 [ arxiv:1107.3412 [ hep - ph ] ] .
a. abada , d. das and c. weiland , arxiv:1111.5836 [ hep - ph ] .
b. c. allanach , a. dedes and h. k. dreiner , phys .
d * 69 * ( 2004 ) 115002 [ erratum - ibid .
d * 72 * ( 2005 ) 079902 ] [ hep - ph/0309196 ] .
h. k. dreiner , m. hanussek and s. grab , phys .
d * 82 * ( 2010 ) 055027 [ arxiv:1005.3309 [ hep - ph ] ] .
r. barbier , c. berat , m. besancon , m. chemtob , a. deandrea , e. dudas , p. fayet and s. lavignac _ et al .
_ , phys .
rept . * 420 * , 1 ( 2005 ) [ hep - ph/0406039 ] . | lepton flavor violation ( lfv ) has been observed in neutrino oscillations . for charged lepton
fv decays only upper limits are known , but sizable branching ratios are expected in many neutrino mass models .
high scale models , such as the classical supersymmetric seesaw , usually predict that decays @xmath0 are roughly a factor @xmath1 maller than the corresponding decays @xmath2 . here
we demonstrate that the @xmath3-penguin diagram can give an enhancement for decays @xmath0 in many extensions of the minimal supersymmetric standard model ( mssm ) .
we first discuss why the @xmath3-penguin is not dominant in the mssm with seesaw and show that much larger contributions from the @xmath3-penguin are expected in general .
we then demonstrate the effect numerically in two example models , namely , the supersymmetric inverse seesaw and r - parity violating supersymmetry .
_ introduction : _ neutrino oscillation experiments @xcite have firmly established that lepton flavor is violated in the neutrino sector , with two of the three measurable mixing angles being surprisingly large .
observation of the characteristic `` neutrino dip '' leaves no doubt that neutrinos have mass @xcite and quite accurate values for the mass squared differences are known now @xcite . in the charged lepton sector , however , only upper limits on lfv branching ratios , such as @xmath4 @xcite or @xmath5 @xcite , exist .
extending the standard model ( sm ) only by neutrino masses does not automatically lead to measurable charged lfv ( clfv ) , but sizable branching ratios are expected in many models .
in fact , on quite general grounds one expects large clfv , if physics beyond the sm exists at the tev scale . a prime example for this observation is supersymmetry ( susy ) . here
, the mass matrices of the new scalar particles need not ( and in general will not ) be aligned with those of the sm fermions .
clfv decays will occur and one can estimate roughly the branching ratios for radiative decays as @xcite @xmath6 where @xmath7 parameterizes the dominant off - diagonal elements of the soft susy breaking slepton mass matrices and @xmath8 is the typical mass of the susy particles , expected to be in the ballpark of @xmath9 tev .
rather small off - diagonal elements are required to satisfy experimental bounds @xcite . in the constrained minimal supersymmetric extension of the sm ( cmssm ) ,
on the other hand , clfv is zero , just as in the sm , simply because neutrinos are assumed to be massless . extending the cmssm to include neutrino masses ( and mixings ) , for example by a seesaw mechanism
, then leads to clfv decays , because the flavor violation necessarily present in the yukawa couplings is transmitted to the slepton mass matrices in the rge ( renormalization group equation ) running @xcite . in such high - scale neutrino mass models , with only mssm particle content at the electroweak scale , it has been shown that the photonic penguin diagram gives the dominant contribution to @xmath0 decays in large regions of parameter space . in the limit of large @xmath10 @xcite . ] in this case a simple relation can be derived @xcite @xmath11 thus , usually it is concluded that the decays @xmath12 are more constraining than the decays @xmath0 .
apart from the photonic penguin , there are also box diagrams , higgs- and @xmath3-penguin contributing to the decays @xmath0 .
the latter diagram is not _ per se _ smaller than the photonic penguin .
rather , as we will show below , in models with only the mssm particle content and couplings , the @xmath3-penguin is suppressed by a subtle cancellation among different terms in the amplitude .
such a cancellation , however , can be easily spoiled if there are ( a ) new couplings and/or ( b ) a larger particle content than in the mssm .
then , as we will discuss , the @xmath3-penguin can easily give the dominant contribution to @xmath0 .
we will demonstrate this fact numerically with two typical example models : ( i ) a supersymmetric inverse seesaw and ( ii ) r - parity violating susy .
the former is an example of a model with extended particle content , while the latter is an example of a model with the mssm particle content but new interactions .
as we will show , in such models @xmath0 can be more constraining than @xmath12 .
this is the main result of the present paper . finally , we emphasize that the @xmath3-penguin can be dominant also in other observables and for other theoretical models , although this fact has not , in general , been discussed before .
for example , @xmath3-dominance can be found in @xmath13 conversion in nuclei in supersymmetric models with r - parity violation , as can be seen from the numerical results of reference @xcite , although the authors do not discuss it .
similarly , in the little higgs model of @xcite one finds numerical results with parameter points where @xmath14 , despite the authors concluding that both are correlated .
_ analytical discussion : _ the total width of the @xmath0 decay contains contributions from the photon penguin , the higgs penguin , the @xmath3-penguin and boxes . considering only contributions from photon and @xmath3-penguin , which are the ones of interest to us and numerically the most important ones , the total width @xmath15 can be written as @xcite : @xmath16 \label{decay}\end{aligned}\ ] ] here , terms denoted @xmath17 ( @xmath18 ) are due to photon ( @xmath3 ) exchange and @xmath19 denotes their interference terms , irrelevant for the following discussion .
both , photon and @xmath3 penguins have chargino - sneutrino and neutralino - slepton contributions .
exact definitions can be found in @xcite .
we will focus on the chargino loops for brevity here , since the effects we are interested in are most pronounced in these loops .
the photon contributions are @xmath20 whereas the @xmath21-contributions read @xmath22 with @xmath23 . in these expressions
@xmath24 denote combinations of rotation matrices and coupling constants and @xmath25 and @xmath26 are short - hands for the passarino - veltman loop functions which depend on @xmath27 . for precise definitions
see @xcite .
the scaling @xmath28 and @xmath29 can be understood , in principle , from simple dimensional analysis .
the width of the decay is proportional to @xmath30 , so both @xmath17 and @xmath18 must be @xmath31 . in this case
it is the smallest mass term in the loop which sets the scale , which in @xmath18 is @xmath32 . due to the masslessness of the photon in case of @xmath17 the smallest mass scale in the loop is @xmath33 .
with @xmath34 the @xmath3 penguin can , in principle , be even more important than the photonic one .
numerically , however , it has been found in case of the mssm that the photonic penguin is dominant @xcite .
this can be understood as follows . to simplify the discussion ,
we neglect first @xmath35 , since it is always proportional to the charged lepton yukawa couplings .
consider then @xmath36 generations of sneutrinos and neglect the chargino mixing .
in this simplified scenario only the wino contributes to @xmath37 and it can be written as @xmath38 with @xmath39 summing over the index @xmath40 is implied .
the terms in the sum come from different types of diagrams : wave function diagrams ( @xmath41 ) , penguins with the @xmath3-boson attached to the chargino line ( @xmath42 ) or the sneutrino line ( @xmath43 ) .
moreover , @xmath44 , @xmath45 , @xmath46 is a @xmath47 unitary matrix that diagonalizes the mass matrix of the sneutrinos and we used the abbreviations @xmath48 . the sum in eq .
vanishes exactly as can be seen by grouping the different terms @xmath49 with @xmath50 . using the exact expressions for the loop functions @xcite
one finds that the masses cancel out and these combinations become just numerical constants : @xmath51 and @xmath52 .
therefore , one is left with @xmath53 , which vanishes due to unitarity of the @xmath46 matrix in supersymmetry , the z - penguin contributions are found to be sub - dominant due to the same type of cancellation that is found in our work . ] . this cancellation can be spoiled by two effects , either ( i ) the sneutrinos mix with other particles which are not @xmath54 doublets so that the factorization no longer holds , or ( ii ) the charginos are not pure wino and higgsino states .
the last effect is of course present in the mssm and therefore this cancellation is not exact . nevertheless , the @xmath3-contributions are suppressed due to their proportionality to the square of the chargino mixing angle ( two wino - higgsino insertions are necessary since there is no @xmath55 coupling ) .
we neglected so far higgsino interactions because in many models these couplings are very small in comparison to the gauge interactions ( for example , a susy scale type - i seesaw model would have @xmath56 ) .
however , in models where the higgsino can have much larger yukawa interactions , a large enhancement of the @xmath3-contributions can be expected .
this will be addressed numerically in the next section . before turning to a numerical discussion
, we consider for simplicity a toy model consisting of two generations of left - handed sneutrinos which can mix with one generation of right - handed sneutrinos .
a @xmath57 rotation matrix is in general parametrized by 3 angles , but we will assume here for simplification that two of them vanish and call the third one @xmath58 .
in addition , we introduce a new interaction for the higgsinos @xmath59 .
we give in fig .
[ fig : approx ] the computed @xmath60 for arbitrarily chosen sneutrino and chargino masses as a function of @xmath58 for different values of @xmath61 .
the red dotted line shows the case for @xmath62 . as clearly seen , @xmath60 depends on the left - right mixing already for small values of @xmath61 .
however , increasing @xmath61 , @xmath60 becomes totally dominated by the new @xmath61 interactions and enhances @xmath63 .
for our toy model as a function of the sneutrino left - right mixing angle @xmath58 and for different values of @xmath61 : @xmath64 ( blue ) , @xmath65 ( black dashed ) , 0.1 ( blue dashed ) and 1.0 ( black ) . ] _ numerical examples : _ we turn to the full fledged numerical study of two examples : an inverse seesaw model and the mssm with @xmath66-parity violation . for this purpose ,
we have created for both models spheno modules @xcite using the mathematica package sarah @xcite .
these modules calculate the low - energy observables exactly including all possible diagrams @xcite .
_ inverse seesaw : _ in inverse seesaw the mssm particle content is extended by three generations of right - handed neutrino superfields @xmath67 and of gauge singlets @xmath68 which carry lepton number @xcite .
the superpotential reads @xmath69 after electroweak symmetry breaking ( ewsb ) the effective mass matrix for the light neutrinos is approximately @xmath70 . since @xmath71 can be of @xmath72 kev or even smaller while @xmath73 is of @xmath74 , the neutrino yukawa couplings have to be much larger than for a standard weak - scale seesaw to explain neutrino data . due to the extended particle content
, new contributions for @xmath75 are expected in the inverse seesaw .
for example , the higgs mediated contributions were recently studied in @xcite . in fig .
[ fig : invseesaw ] ( top ) we show the different contributions to @xmath76 for a variation of the susy masses . to disentangle rge effects we have calculated once the spectrum for a cmssm input ( @xmath77 gev , @xmath78 tev , @xmath79 , @xmath80 gev ) and rescaled all dimensionful parameters at the susy scale .
this changes the sfermion masses but not the mixing matrices .
@xmath81 has been chosen to explain neutrino data for @xmath82 kev and @xmath83 tev .
clearly , the @xmath3-penguins dominate and are nearly independent of the susy scale . only in the limit
@xmath84 the other contributions can compete . in fig .
[ fig : invseesaw ] ( bottom ) the branching ratios for @xmath4 and @xmath5 and the current experimental bounds of @xmath85 and @xmath86 are depicted @xcite . while @xmath87 would be in conflict with experiment only for @xmath88 tev
, @xmath76 rules out the entire range . as function of the lightest sneutrino
mass : @xmath3-penguins ( red dotted ) , photonic penguins ( blue dashed ) , combined higgs - penguins / box diagrams ( green ) .
bottom : @xmath76 ( black ) and @xmath89 ( blue ) and the current experimental bounds ( dashed lines ) .
the dips are an effect of a mass crossing between charginos and sneutrinos .
, title="fig : " ] + as function of the lightest sneutrino mass : @xmath3-penguins ( red dotted ) , photonic penguins ( blue dashed ) , combined higgs - penguins / box diagrams ( green ) .
bottom : @xmath76 ( black ) and @xmath89 ( blue ) and the current experimental bounds ( dashed lines ) .
the dips are an effect of a mass crossing between charginos and sneutrinos .
, title="fig : " ] in this example , we have assumed @xmath71 and @xmath73 to be diagonal and all flavor violation comes from @xmath81 , as is usually done in literature . however , neutrino oscillation data could equally well be fitted with the flavor violation coming from @xmath71 and @xmath73 . in that case
clfv observables would be much smaller and consistent with experimental data .
however , the relative order between 2- and 3-body decays wo nt change , i.e. @xmath90 will be most likely observed before @xmath91 if inverse seesaw is realized in nature . _
@xmath66-parity violation : _ as second example , we take the mssm particle content but extend the superpotential by the lepton number violating terms @xcite @xmath92 while the @xmath93-parameters are highly constrained by neutrino data @xcite , the bounds for the tri - linear couplings are much weaker and some entries can be of @xmath94 @xcite . in the following ,
all entries of @xmath95 and @xmath96 are set to zero but @xmath97 and @xmath98 .
we give in fig .
[ fig : trpv ] the results for @xmath76 and @xmath87 for a mixed bi- and tri - linear as well as for the pure tri - linear scenario varying @xmath99 . in the mixed case @xmath100 and the vacuum expectation values of the sneutrinos , @xmath101 , have been chosen to be consistent with neutrino data and a moderate flavor violation in the sneutrino sector has been induced by @xmath102 .
it can be seen that in the mixed case @xmath103 holds when @xmath104 crosses @xmath105 , while for the pure tri - linear case without any flavor violation at tree level in the sneutrino sector the three body decays dominate even for much smaller values . in both cases
we get an upper limit for @xmath106 of @xmath107 from the bounds on @xmath108 for sneutrino masses of 730 gev .
so far , in the literature just the limits for @xmath109 gev from the photonic penguins @xcite have been published .
these are much weaker , after rescaling the bound @xmath110 .
( black ) and @xmath87 ( blue ) varying @xmath104 in a mixed bi- and trilinear ( solid lines ) and a pure tri - linear ( dotted lines ) @xmath66pv scenario .
the dashed lines show the experimental limits.,title="fig : " ] + _ summary : _ we have shown in this letter that the @xmath3-penguin can give the dominant contribution in lepton flavor violating three body decays in many models . the importance of the @xmath3-penguin increases with increasing susy particles masses . as numerical examples ,
we have briefly discussed the supersymmetric inverse seesaw and the mssm with @xmath66-parity violation . |
let @xmath1 be an interval of the real line and @xmath2 be a finite collection of points that split @xmath1 into subintervals @xmath3 , @xmath4 .
a transformation @xmath5 of the interval @xmath1 that rearranges the subintervals by translation is called an _ interval exchange transformation _ ( see figure [ fig1 ] ) . to be precise , the restriction of @xmath5 to any @xmath3 is a translation , the translated subinterval remains within @xmath1 and does not overlap with the other translated subintervals .
the definition is still incomplete as values of @xmath5 at the points @xmath6 are not specified .
the standard way to do this , which we adopt , is to require that @xmath5 be right continuous .
that is , we consider the half - closed interval @xmath7 partitioned into smaller half - closed intervals @xmath8 .
the interval exchange transformation @xmath5 is to translate each @xmath9 so that the images @xmath10 form another partition of @xmath11 .
let @xmath12 be a @xmath13-dimensional vector whose coordinates are lengths of the intervals @xmath14 .
let @xmath15 be a permutation on @xmath16 that tells how the intervals are rearranged by @xmath5 .
namely , @xmath17 is the position of @xmath18 when the the intervals @xmath19 are ordered from left to right . for the example in figure [ fig1 ] , @xmath20 .
we refer to the pair @xmath21 as a combinatorial description of @xmath5 .
given an integer @xmath22 , a @xmath13-dimensional vector @xmath12 with positive coordinates that add up to the length of @xmath11 , and a permutation @xmath15 on @xmath16 , the pair @xmath21 determines a unique interval exchange transformation of @xmath11 .
the converse is not true .
any partition of @xmath11 into subintervals that are translated by @xmath5 gives rise to a distinct combinatorial description .
clearly , such a partition is not unique .
however there is a unique partition with the smallest possible number of subintervals .
the interval exchange transformations have been popular objects of study in ergodic theory .
first of all , the exchange of two intervals is equivalent to a rotation of the circle ( it becomes one when we identify the endpoints of the interval @xmath11 thus producing a circle ) .
the exchanges of three or more intervals were first considered by katok and stepin @xcite .
the systematic study started since the paper by keane @xcite who coined the term .
for an account of the results , see the survey by viana @xcite .
all interval exchange transformations of a fixed interval @xmath7 form a transformation group @xmath23 .
changing the interval , one obtains an isomorphic group .
indeed , let @xmath24 be another interval and @xmath25 be an affine map of @xmath11 onto @xmath26 .
then @xmath27 if and only if @xmath28 .
we refer to any of the groups @xmath29 as the _ group of interval exchange transformations_. the present notes are concerned with group - theoretic properties of @xmath29 .
an important tool here is the _ scissors congruence invariant _ or the _ sah - arnoux - fathi ( saf ) invariant _ of @xmath30 introduced independently by sah @xcite and arnoux and fathi @xcite .
the invariant can be informally defined by @xmath31 ( the integral is actually a finite sum ) .
the importance stems from the fact that @xmath32 is a homomorphism of the group @xmath29 onto @xmath33 . as a consequence ,
the invariant vanishes on the commutator group , which is a subgroup of @xmath29 generated by commutators @xmath34 , where @xmath5 and @xmath35 run over @xmath29 . in this paper , we establish the following properties of the commutator group of the group of interval exchange transformations @xmath29 .
[ main1 ] the following four groups are the same : * the group of interval exchange transformations with zero saf invariant , * the commutator group of the group of interval exchange transformations , * the group generated by interval exchange transformations of order @xmath0 , * the group generated by interval exchange transformations of finite order .
[ main2 ] the quotient of the group of interval exchange transformations by its commutator group is isomorphic to @xmath33 .
[ main3 ] the commutator group of the group of interval exchange transformations is simple .
it has to be noted that most of these results are already known . a theorem by sah reproduced in veech s paper @xcite contains theorems [ main2 ] , [ main3 ] , and part of theorem [ main1 ] .
unfortunately , the preprint of sah @xcite was never published .
hence we include complete proofs .
a new result of the present paper is that the commutator group of @xmath23 is generated by elements of order @xmath0 .
this is the central result of the paper as our proofs of the theorems are based on the study of elements of order @xmath0 .
the paper is organized as follows .
section [ elem ] contains some elementary constructions that will be used in the proofs of the theorems .
the scissors congruence invariant is considered in section [ inv ] .
section [ comm ] is devoted to the proof of theorem [ main1 ] .
theorem [ main2 ] is proved in the same section .
section [ simp ] is devoted to the proof of theorem [ main3 ] .
the author is grateful to michael boshernitzan for useful and inspiring discussions .
let us choose an arbitrary interval @xmath7 . in what follows
, all interval exchange transformations are assumed to be defined on @xmath11 .
also , all subintervals of @xmath11 are assumed to be half - closed intervals of the form @xmath36 .
the proofs of theorems [ main1 ] and [ main3 ] are based on several elementary constructions described in this section .
first of all , we introduce two basic types of transformations used in those constructions .
an _ interval swap map of type @xmath37 _ is an interval exchange transformation that interchanges two nonoverlapping intervals of length @xmath37 by translation while fixing the rest of the interval @xmath11 .
a _ restricted rotation of type @xmath38 _ is an interval exchange transformation that exchanges two neighboring intervals of lengths @xmath37 and @xmath39 ( the interval of length @xmath37 must be to the left of the interval of length @xmath39 ) while fixing the rest of @xmath11 .
the type of an interval swap map is determined uniquely , and so is the type of a restricted rotation .
clearly , any interval swap map is an involution .
the inverse of a restricted rotation of type @xmath38 is a restricted rotation of type @xmath40 .
any restricted rotation of type @xmath41 is also an interval swap map of type @xmath37 .
[ elem1 ] any interval exchange transformation @xmath5 is a product of restricted rotations
. moreover , if @xmath5 exchanges at least two intervals and @xmath42 is the set of their lengths , it is enough to use restricted rotations of types @xmath38 such that @xmath43 .
the exchange of one interval is the identity . in this case , take any restricted rotation @xmath25 .
then @xmath44 , which is a product of restricted rotations .
now assume that @xmath5 exchanges @xmath45 intervals .
let @xmath14 be the intervals , ordered from left to right , and @xmath21 be the corresponding combinatorial description of @xmath5 .
then @xmath42 is the set of coordinates of the vector @xmath12 . for any permutation @xmath46 on @xmath16 , let @xmath47 denote a unique interval exchange transformation with the combinatorial description @xmath48 . given two permutations @xmath46 and @xmath49 on @xmath16 , let @xmath50 . for any @xmath51 , @xmath52 , the transformation @xmath53 translates the interval @xmath54 onto @xmath55 .
it follows that @xmath53 has combinatorial description @xmath56 , where @xmath57 for @xmath52 .
now suppose that @xmath15 is expanded into a product of permutations @xmath58 .
for any @xmath59 , @xmath60 , let @xmath61 .
then @xmath62 , where @xmath63 and @xmath64 for @xmath65 . by
the above each @xmath66 has combinatorial description @xmath67 , where the vector @xmath68 is obtained from @xmath12 by permutation of its coordinates . in the case
@xmath69 is a transposition of neighboring numbers , @xmath70 , the transformation @xmath66 is a restricted rotation of type @xmath71 .
notice that @xmath72 .
it remains to observe that any permutation @xmath15 on @xmath16 is a product of transpositions of neighboring numbers .
indeed , we can represent @xmath15 as a product of cycles .
further , any cycle @xmath73 of length @xmath74 is a product of @xmath75 transpositions : @xmath76 .
a cycle of length @xmath77 is the identity , hence it equals @xmath78 .
finally , any transposition @xmath79 , @xmath80 , is expanded into a product of transpositions of neighboring numbers : @xmath81 , where @xmath82 .
notice that the set @xmath42 in lemma [ elem1 ] depends on the combinatorial description of the interval exchange transformation @xmath5 .
the lemma holds for every version of this set .
[ elem2 ] any interval exchange transformation of finite order is a product of interval swap maps .
suppose @xmath83 are nonoverlapping intervals of the same length @xmath37 contained in an interval @xmath11 .
let @xmath35 be an interval exchange transformation of @xmath11 that translates @xmath84 onto @xmath85 for @xmath86 , translates @xmath87 onto @xmath88 , and fixes the rest of @xmath11 . if @xmath45 then @xmath35 is the product of @xmath89 interval swap maps of type @xmath37 .
namely , @xmath90 , where each @xmath91 interchanges @xmath84 with @xmath85 by translation while fixing the rest of @xmath11 . in the case
@xmath92 , @xmath35 is the identity .
then @xmath93 for any interval swap map @xmath94 on @xmath11 .
let @xmath5 be an interval exchange transformation of @xmath11 that has finite order . since there are only finitely many distinct powers of @xmath5 , there are only finitely many points in @xmath11 at which one of the powers has a discontinuity .
let @xmath95 be a partition of @xmath11 into subintervals created by all such points .
by construction , the restriction of @xmath5 to any @xmath9 is a translation and , moreover , the translated interval @xmath18 is contained in another element of the partition . since the same applies to the inverse @xmath96 , it follows that @xmath18 actually coincides with some element of the partition .
hence @xmath5 permutes the intervals @xmath97 by translation .
therefore these intervals can be relabeled as @xmath98 , @xmath99 , @xmath100 ( @xmath101 and @xmath102 are some positive integers ) , so that @xmath5 translates each @xmath98 onto @xmath103 if @xmath104 and onto @xmath105 if @xmath106 . for any @xmath107 let @xmath108 be an interval exchange transformation that coincides with @xmath5 on the union of intervals @xmath98 , @xmath100 , and fixes the rest of @xmath11 .
it is easy to observe that the transformations @xmath109 commute and @xmath110 . by the above each @xmath108
can be represented as a product of interval swap maps .
hence @xmath5 is a product of interval swap maps as well .
[ elem3 ] any interval swap map is a commutator of two interval exchange transformations of order @xmath0 .
let @xmath5 be an interval swap map of type @xmath37 .
let @xmath111 and @xmath112 be nonoverlapping intervals interchanged by @xmath5 .
we split the interval @xmath113 into two subintervals @xmath114 and @xmath115 .
similarly , @xmath116 is divided into @xmath117 and @xmath118 .
now we introduce three interval swap maps of type @xmath119 : @xmath120 interchanges @xmath121 with @xmath122 , @xmath123 interchanges @xmath124 with @xmath125 , and @xmath126 interchanges @xmath121 with @xmath124 .
the maps @xmath127 permute the intervals @xmath128 by translation and fix the rest of the interval @xmath11 .
it is easy to see that @xmath129 .
hence @xmath130 is an element of order @xmath0 .
further , we check that @xmath131 maps @xmath121 onto @xmath122 , @xmath122 onto @xmath124 , @xmath124 onto @xmath125 , and @xmath125 onto @xmath121
. therefore the second iteration @xmath132 interchanges @xmath121 with @xmath124 and @xmath122 with @xmath125 , which is exactly how @xmath5 acts .
thus @xmath133 . for the next two constructions
, we need another definition .
the _ support _ of an interval exchange transformation @xmath5 is the set of all points in @xmath11 moved by @xmath5 .
it is the union of a finite number of ( half - closed ) intervals .
for instance , the support of a restricted rotation of type @xmath38 is a single interval of length @xmath134 .
the support of an interval swap map of type @xmath37 is the union of two nonoverlapping intervals of length @xmath37 .
note that any interval swap map is uniquely determined by its type and support .
the same holds true for any restricted rotation .
[ elem4 ] let @xmath135 and @xmath136 be interval swap maps of the same type .
if the supports of @xmath135 and @xmath136 do not overlap then there exists an interval exchange transformation @xmath35 of order @xmath0 such that @xmath137 .
let @xmath37 be the type of @xmath135 and @xmath136 .
let @xmath113 and @xmath138 be nonoverlapping intervals of length @xmath37 interchanged by @xmath135 .
let @xmath116 and @xmath139 be nonoverlapping intervals of length @xmath37 interchanged by @xmath136 .
assume that the supports of @xmath135 and @xmath136 do not overlap , i.e. , the intervals @xmath140 do not overlap with each other .
let us introduce two more interval swap maps of type @xmath37 : @xmath120 interchanges @xmath113 with @xmath116 and @xmath123 interchanges @xmath138 with @xmath139 . since the supports of @xmath120 and @xmath123 do not overlap , the transformations commute .
hence the product @xmath130 is an element of order @xmath0 .
the maps @xmath135 , @xmath136 , and @xmath35 permute the intervals @xmath140 by translation and fix the rest of the interval @xmath11 .
one easily checks that @xmath137 .
[ elem5 ] let @xmath135 and @xmath136 be restricted rotations of the same type .
if the supports of @xmath135 and @xmath136 do not overlap then @xmath141 is the product of three interval swap maps .
let @xmath38 be the type of @xmath135 and @xmath136 .
let @xmath142 be the support of @xmath135 and @xmath143 be the support of @xmath136 .
the transformation @xmath136 translates the interval @xmath144 by @xmath39 and the interval @xmath145 by @xmath146 .
the inverse @xmath147 is a restricted rotation of type @xmath40 with the same support as @xmath135 .
it translates the interval @xmath148 by @xmath37 and the interval @xmath149 by @xmath150 .
assume that the supports @xmath113 and @xmath116 do not overlap .
let @xmath120 be the interval swap map of type @xmath37 that interchanges the intervals @xmath122 and @xmath124 , let @xmath123 be the interval swap map of type @xmath39 that interchanges @xmath121 and @xmath125 , and let @xmath126 be the interval swap map of type @xmath134 that interchanges @xmath113 and @xmath116 .
it is easy to check that @xmath151 ( see figure [ fig3 ] ) .
[ elem6 ] suppose @xmath5 is a restricted rotation of type @xmath38 , where @xmath152 .
then there exist interval swap maps @xmath120 and @xmath123 such that @xmath153 is a restricted rotation of type @xmath154 . .
] let @xmath155 be the support of @xmath5 .
we define three more transformations with supports inside @xmath156 : @xmath120 is an interval swap map of type @xmath39 that interchanges the intervals @xmath157 and @xmath158 , @xmath123 is an interval swap map of type @xmath39 that interchanges @xmath159 and @xmath158 , and @xmath25 is a restricted rotation of type @xmath154 with support @xmath160 .
let us check that @xmath161 .
since @xmath152 , the points @xmath162 and @xmath163 divide @xmath156 into three subintervals @xmath164 , @xmath165 , and @xmath166 .
the map @xmath25 translates @xmath113 by @xmath39 , @xmath116 by @xmath167 , and fixes @xmath168 .
then the map @xmath120 translates @xmath168 by @xmath146 , @xmath169 by @xmath37 , and fixes @xmath170 .
therefore the product @xmath171 translates @xmath113 by @xmath39 , @xmath116 by @xmath39 , and @xmath168 by @xmath146 .
this is exactly how @xmath5 acts .
similarly , we check that @xmath172 .
it remains to notice that @xmath173 .
[ elem7 ] let @xmath5 be a nontrivial interval exchange transformation .
then there exist @xmath174 and , for any @xmath175 , interval swap maps @xmath176 such that @xmath177 is an interval swap map of type @xmath178 .
since @xmath5 is not the identity , we can find an interval @xmath179 such that @xmath5 translates @xmath26 by some @xmath180 .
let @xmath181 . given any @xmath178 , @xmath175
, we introduce two intervals @xmath182 and @xmath183 . by construction , @xmath156 and @xmath113 do not overlap .
besides , @xmath5 translates @xmath156 onto @xmath113 .
let @xmath184 be an interval swap map of type @xmath185 that interchanges two halves @xmath186 and @xmath187 of the interval @xmath156 .
let @xmath120 be an interval swap map of type @xmath185 that interchanges two halves @xmath188 and @xmath189 of @xmath113 .
since @xmath5 translates @xmath156 onto @xmath113 , it follows that @xmath190 .
further , let @xmath123 be an interval swap map of type @xmath185 that interchanges @xmath191 with @xmath121 .
the maps @xmath192 permute the nonoverlapping intervals @xmath193 by translation and fix the rest of the interval @xmath11 .
it is easy to check that @xmath194 is an interval swap map of type @xmath178 that interchanges @xmath156 with @xmath113 .
let us recall the construction of the tensor product .
suppose @xmath195 and @xmath196 are vector spaces over a field @xmath197 .
let @xmath198 be a vector space over @xmath197 with basis @xmath199\}_{(v , w)\in v\times w}$ ] .
let @xmath200 denote the subspace of @xmath198 spanned by all vectors of the form @xmath201-z[v_1,w]-z[v_2,w]$ ] , @xmath202-z[v , w_1]-z[v , w_2]$ ] , @xmath203-\alpha z[v , w]$ ] , and @xmath204-\alpha z[v , w]$ ] , where @xmath205 , @xmath206 , and @xmath207 .
the _ tensor product _ of the spaces @xmath195 and @xmath196 over the field @xmath197 , denoted @xmath208 , is the quotient of the vector space @xmath198 by @xmath200 . for any @xmath209 and @xmath210 the coset @xmath211+y(v , w)$ ] is denoted @xmath212 . by construction
, @xmath213 is a bilinear mapping on @xmath214 . in the case
@xmath215 , for any vectors @xmath216 we define the _ wedge product _ @xmath217 .
the subspace of @xmath218 spanned by all wedge products is denoted @xmath219 . by construction
, @xmath220 is a bilinear , skew - symmetric mapping on @xmath221 .
[ inv1 ] suppose @xmath195 is a vector space over a field @xmath197 and @xmath222 are linearly independent vectors .
then the wedge products @xmath223 , @xmath224 , are linearly independent in @xmath219 . for
any bilinear function @xmath225 let @xmath226 denote a unique linear function on @xmath227 such that @xmath228)=\om(v , w)$ ] for all @xmath216 . since @xmath229 is bilinear , the function @xmath226 vanishes on the subspace @xmath230 .
hence it gives rise to a linear function @xmath231 . by construction , @xmath232 for all @xmath216 .
let us extend the set @xmath233 to a basis @xmath234 for the vector space @xmath195 .
for any @xmath235 we denote by @xmath236 a unique bilinear function on @xmath221 such that @xmath237 if @xmath238 and @xmath239 for any other pair @xmath240 . the function @xmath236 gives rise to a linear function @xmath241 on @xmath218 as described above . for any @xmath242 , @xmath243 , we have @xmath244 if @xmath245 and @xmath246 , @xmath247 if @xmath248 and @xmath249 , and @xmath250 otherwise .
consider an arbitrary linear combination @xmath251 with coefficients @xmath252 from @xmath197 .
it is easy to observe that @xmath253 for any @xmath254 .
therefore @xmath255 unless all @xmath252 are zeros .
thus the wedge products @xmath223 , @xmath224 , are linearly independent over @xmath197 .
let @xmath5 be an interval exchange transformation of an interval @xmath7 .
consider an arbitrary partition of @xmath11 into subintervals , @xmath256 , such that the restriction of @xmath5 to any @xmath9 is a translation by some @xmath257 .
let @xmath258 be the length of @xmath9 , @xmath259 .
scissors congruence invariant _ , also known as the _ sah - arnoux - fathi ( saf ) invariant _ , of @xmath5 is @xmath260 regarded as an element of the tensor product @xmath261 .
one can easily check that @xmath262 for any restricted rotation @xmath5 of type @xmath38 and @xmath263 for any interval swap map @xmath35 .
the term ` scissors congruence invariant ' is partially explained by the following lemma .
[ inv2 ] the scissors congruence invariant @xmath264 of an interval exchange transformation @xmath5 does not depend on the combinatorial description of @xmath5 .
let @xmath265 be a partition of the interval @xmath11 into subintervals such that the restriction of @xmath5 to any @xmath9 is a translation by some @xmath257 .
let @xmath266 be another partition into subintervals such that the restriction of @xmath5 to any @xmath267 is a translation by some @xmath268 .
let @xmath258 denote the length of @xmath9 ( @xmath269 ) and @xmath270 denote the length of @xmath267 ( @xmath60 ) .
we have to show that @xmath271 coincides with @xmath272 in @xmath261 . for any @xmath52 and @xmath60 the intersection @xmath273
is either an interval or the empty set .
we let @xmath274 be the length of the interval in the former case and @xmath275 otherwise .
further , let @xmath276 clearly , @xmath277 whenever @xmath273 is an interval . otherwise @xmath275 and @xmath278 . in any case , @xmath279 . therefore @xmath280 .
for any @xmath281 , nonempty intersections @xmath273 , @xmath282 , form a partition of the interval @xmath9 into subintervals .
hence @xmath283 .
it follows that @xmath284 .
similarly , we obtain that @xmath285 .
thus @xmath286 . in view of lemma [ inv2 ] ,
for any interval @xmath7 we can consider the invariant @xmath32 as a function on @xmath23 , the set of all interval exchange transformations of @xmath11 .
[ inv3 ] the scissors congruence invariant @xmath32 is a homomorphism of the group @xmath23 to @xmath261 .
consider arbitrary interval exchange transformations @xmath5 and @xmath35 of the interval @xmath11 .
we have to show that @xmath287 .
let @xmath288 be a partition of @xmath11 into subintervals such that the restrictions of both @xmath35 and @xmath289 to any @xmath9 are translations by some @xmath257 and @xmath290 , respectively .
let @xmath258 be the length of @xmath9 , @xmath52 .
then @xmath291 it is easy to see that for any @xmath52 the image @xmath292 is an interval of length @xmath258 and the restriction of @xmath5 to @xmath292 is the translation by @xmath293 . besides , the intervals @xmath294 form another partition of @xmath11 .
it follows that @xmath295 since @xmath296 for all @xmath52 , we obtain that @xmath287 . in the remainder of this section
we show that @xmath32 is actually a homomorphism of @xmath23 onto @xmath33 .
[ inv4 ] for any @xmath297 there exist pairs of positive numbers @xmath298 , @xmath299 such that * @xmath300 , * @xmath301 or @xmath302 for @xmath303 , * @xmath304 or @xmath305 .
we define a finite or infinite sequence of pairs inductively . first of all , @xmath300 .
further , assume that the pair @xmath306 is defined for some positive integer @xmath51 . if @xmath307 then this is the last pair in the sequence .
otherwise we let @xmath301 if @xmath308 and @xmath302 if @xmath309 .
since @xmath310 , it follows by induction that @xmath311 for all @xmath51 .
if the sequence @xmath312 is finite and contains @xmath313 pairs , then @xmath305 and we are done .
if the sequence is infinite , it is enough to show that @xmath314 as @xmath315 . for any positive integer @xmath313 let @xmath316 . since @xmath317 and @xmath318 are nonincreasing sequences of positive numbers ,
so is the sequence @xmath319 . by construction , @xmath320 for all @xmath51 .
it follows that the series @xmath321 is convergent .
in particular , @xmath322 as @xmath315 . note that if @xmath323 for some @xmath51 , then @xmath324 so that @xmath325 .
this implies @xmath314 as @xmath315 .
[ inv5 ] for any @xmath326 and @xmath327 there exist @xmath328 , @xmath329 , such that @xmath330 in @xmath33 .
note that @xmath331 for all @xmath332 .
therefore in the case @xmath333 it is enough to take @xmath334 , where @xmath335 .
now assume that @xmath336 . clearly , in this case @xmath37 and @xmath39 are nonzero .
since @xmath337 and @xmath338 for all @xmath326 , it is no loss to assume that @xmath37 and @xmath39 are positive . by lemma [ inv4 ] , there exist pairs of positive numbers @xmath300 , @xmath299 such that @xmath301 or @xmath302 for @xmath303 , and also @xmath304 or @xmath305 . since @xmath339 and @xmath340 for all @xmath341 , it follows by induction that @xmath342 , @xmath343 .
then @xmath344 as @xmath345
. thus @xmath304 .
[ inv6 ] an element @xmath346 is the scissors congruence invariant of some interval exchange transformation in @xmath23 if and only if @xmath347 .
as already mentioned before , the saf invariant of a restricted rotation of type @xmath38 is @xmath348 . by lemma [ elem1
] , any @xmath349 is a product of restricted rotations . since @xmath32 is a homomorphism of the group @xmath23 due to lemma [ inv3 ] , we obtain that @xmath264 is a finite sum of wedge products .
hence @xmath350 .
let @xmath101 denote the length of the interval @xmath11 .
by lemma [ inv5 ] , for any @xmath326 one can find @xmath328 , @xmath351 , such that @xmath352 . by the choice of @xmath353 and @xmath354
, the group @xmath23 contains a restricted rotation of type @xmath355 .
it follows that any wedge product in @xmath33 is the saf invariant of some interval exchange transformation in @xmath23 .
since @xmath32 is a homomorphism of @xmath23 , any sum of wedge products is also the saf invariant of some @xmath349 .
any @xmath347 is a linear combination of wedge products with rational coefficients . since @xmath356 for all @xmath326 and @xmath357 , the element @xmath358 can also be represented as a sum of wedge products . by the above , @xmath359 for some @xmath349
we begin this section with a technical lemma that will be used in the proof of the principal lemma [ comm2 ] below .
[ comm1 ] suppose @xmath360 are positive numbers . then there exist positive numbers @xmath361 linearly independent over @xmath362 such that each @xmath363 is a linear combination of @xmath361 with nonnegative integer coefficients .
the proof is by induction on the number @xmath13 of the reals @xmath360 .
the case @xmath92 is trivial .
now assume that @xmath364 and the lemma holds for the numbers @xmath365 .
that is , there exist positive numbers @xmath361 linearly independent over @xmath362 such that each @xmath363 , @xmath86 is a linear combination of @xmath361 with nonnegative integer coefficients . if the reals @xmath366 and @xmath367 are linearly independent over @xmath362 , then we are done .
otherwise @xmath367 is a linear combination of @xmath366 with rational coefficients .
let us separate positive and negative terms in this linear combination : @xmath368 , where @xmath369 are positive rationals and the indices @xmath370 are all distinct .
it is possible that there is no negative term at all .
since @xmath366 and @xmath367 are positive numbers , we can find positive rationals @xmath371 such that @xmath372 and @xmath373 is positive for @xmath374 .
let @xmath375 for any
@xmath376 different from @xmath377 . then @xmath378 are positive numbers linearly independent over @xmath362 . by construction , @xmath379 and @xmath380 for @xmath374
. therefore each of the numbers @xmath366 and @xmath367 is a linear combination of @xmath378 with nonnegative rational coefficients .
it follows that each of the numbers @xmath360 is also a linear combination of @xmath378 with nonnegative rational coefficients .
then there exists a positive integer @xmath381 such that each @xmath363 is a linear combination of @xmath382 with nonnegative integer coefficients .
let us call a product of restricted rotations _ balanced _ if for any @xmath310 the number of factors of type @xmath38 in this product matches the number of factors of type @xmath40 .
[ comm2 ] any interval exchange transformation with zero saf invariant can be represented as a balanced product of restricted rotations .
consider an arbitrary interval exchange transformation @xmath5 of an interval @xmath11 .
if @xmath5 is the identity , then for any restricted rotation @xmath25 on @xmath11 we have @xmath44 , which is a balanced product of restricted rotations .
now assume @xmath5 is not the identity .
let @xmath265 be a partition of @xmath11 into subintervals such that the restriction of @xmath5 to any @xmath9 is a translation .
note that @xmath45 .
let @xmath360 be lengths of the intervals @xmath14 .
by lemma [ comm1 ] , one can find positive numbers @xmath361 linearly independent over @xmath362 such that each @xmath363 is a linear combination of @xmath361 with nonnegative integer coefficients .
then each @xmath9 can be partitioned into smaller intervals with lengths in the set @xmath383 . clearly , the restriction of @xmath5 to any of the smaller intervals is a translation , hence lemma [ elem1 ] applies here .
we obtain that @xmath5 can be represented as a product of restricted rotations , @xmath384 , such that the type @xmath38 of any factor satisfies @xmath43 . for any @xmath385 let @xmath386 denote the number of factors of type @xmath387 in this product .
then @xmath388 since the numbers @xmath366 are linearly independent over @xmath362 , it follows from lemma [ inv1 ] that the wedge products @xmath389 , @xmath390 , are linearly independent in @xmath33 .
therefore @xmath391 only if @xmath392 for all @xmath393 , @xmath394
. then @xmath392 for all @xmath385 , which means that the product @xmath395 is balanced .
the next lemma is an extension of lemma [ elem6 ] that will be used in the proofs of lemmas [ comm4 ] and [ comm5 ] below .
[ comm3 ] given @xmath297 , there exist @xmath328 , @xmath329 , such that any restricted rotation @xmath5 of type @xmath38 can be represented as @xmath396 , where @xmath25 is a restricted rotation of type @xmath355 and @xmath35 is a product of interval swap maps .
consider an arbitrary restricted rotation @xmath5 of type @xmath397 , where @xmath398 .
if @xmath399 then lemma [ elem6 ] implies that @xmath396 , where @xmath25 is a restricted rotation of type @xmath400 and @xmath35 is an interval swap map . in the case
@xmath401 , we observe that the inverse map @xmath96 is a restricted rotation of type @xmath402 .
the same lemma [ elem6 ] implies that @xmath403 , where @xmath404 is a restricted rotation of type @xmath405 and @xmath406 is an interval swap map .
note that @xmath407 and @xmath408 is a restricted rotation of type @xmath409 . by lemma [ inv4 ] , there exist pairs of positive numbers @xmath410 such that * @xmath300 , * @xmath301 or @xmath302 for @xmath303 , * @xmath304 or @xmath305 . clearly , @xmath411 for @xmath412 . by induction
, it follows from the above that there exist interval exchange transformations @xmath413 and @xmath414 such that @xmath415 is a restricted rotation of type @xmath306 , @xmath108 is an interval swap map , and @xmath416 for @xmath417 .
we have @xmath418 , where @xmath419 is a product of interval swap maps .
if @xmath304 then we are done .
otherwise @xmath305 so that @xmath420 itself is an interval swap map , hence @xmath5 is a product of interval swap maps . in this case , take an arbitrary restricted rotation @xmath25 of type @xmath355 , where @xmath421 . since @xmath25 is also an interval swap map , we obtain @xmath422 , where @xmath423 is a product of interval swap maps .
[ comm4 ] let @xmath135 and @xmath136 be restricted rotations of the same type .
then @xmath141 is a product of interval swap maps .
the lemma has already been proved in one case .
if the supports of @xmath135 and @xmath136 do not overlap then @xmath141 is the product of three interval swap maps due to lemma [ elem5 ] .
we are going to reduce the general case to this particular one .
let @xmath38 be the type of the restricted rotations @xmath135 and @xmath136 .
first we assume there exists an interval @xmath424 of length @xmath134 that does not overlap with supports of @xmath135 and @xmath136 .
let @xmath425 denote a unique restricted rotation of type @xmath38 with support @xmath156 .
by lemma [ elem5 ] , both @xmath426 and @xmath427 are products of three interval swap maps .
hence @xmath428 is the product of six interval swap maps .
the above assumption always holds in the case when @xmath429 , where @xmath101 is the length of @xmath11 .
indeed , let us divide the interval @xmath11 into @xmath430 pieces of length @xmath431 .
then the support of @xmath135 , which is an interval of length @xmath134 , overlaps with at most two pieces .
the same is true for the support of @xmath136 .
therefore we have at least one piece with interior disjoint from both supports .
this piece clearly contains an interval of length @xmath134 .
now consider the general case .
it follows from lemma [ comm3 ] that @xmath432 and @xmath433 , where @xmath176 are products of interval swap maps while @xmath434 are restricted rotations of the same type @xmath355 such that @xmath435 .
note that @xmath436 and @xmath437 are also products of interval swap maps . by
the above @xmath438 is the product of six interval swap maps .
then @xmath439 is a product of interval swap maps as well .
[ comm5 ] let @xmath5 be a restricted rotation and @xmath35 be an arbitrary interval exchange transformation .
then the commutator @xmath34 is a product of interval swap maps .
let @xmath38 be the type of the restricted rotation @xmath5 and @xmath26 be the support of @xmath5 .
first assume that the restriction of the transformation @xmath440 to @xmath26 is a translation .
then @xmath441 is also a restricted rotation of type @xmath38 , with support @xmath442 .
therefore @xmath34 is a product of interval swap maps due to lemma [ comm4 ] . in the general case
, we choose an interval @xmath424 such that @xmath440 is a translation when restricted to @xmath156 .
let @xmath178 denote the length of @xmath156 .
according to lemma [ comm3 ] , we have @xmath443 , where @xmath184 is a product of interval swap maps and @xmath425 is a restricted rotation of some type @xmath355 such that @xmath329 .
obviously , @xmath444 is also a product of interval swap maps . since @xmath329
, there exists a restricted rotation @xmath135 of type @xmath355 with support contained in @xmath156 . by the above the commutator @xmath445 is a product of swap maps . by lemma
[ comm4 ] , @xmath446 and @xmath426 are also products of interval swap maps .
note that @xmath447 therefore @xmath448 , where @xmath120 and @xmath123 are products of interval swap maps .
consider an arbitrary factorization @xmath449 such that each @xmath91 is an interval swap map .
then @xmath450 .
clearly , each @xmath451 is an interval exchange transformation of order @xmath0 and hence a product of interval swap maps due to lemma [ elem2 ] .
it follows that @xmath34 can also be represented as a product of interval swap maps .
[ comm6 ] any balanced product of restricted rotations is also a product of interval swap maps .
the proof is by strong induction on the number @xmath313 of factors in a balanced product .
let @xmath384 be a balanced product of @xmath313 restricted rotations and assume that the lemma holds for any balanced product of less than @xmath313 factors .
let @xmath38 be the type of @xmath135 .
first consider the case @xmath452 . in this case
, @xmath135 is an interval swap map . if @xmath453 then we are done .
otherwise @xmath454 , where @xmath455 is a balanced product of @xmath456 restricted rotations . by the inductive assumption ,
@xmath35 is a product of interval swap maps , and so is @xmath5 .
now consider the case @xmath457 . in this case
, there is also a factor @xmath458 of type @xmath40 .
let @xmath120 be the identity if @xmath459 and @xmath460 otherwise .
let @xmath123 be the identity if @xmath461 and @xmath462 otherwise .
we have @xmath463 since @xmath147 is a restricted rotation of type @xmath40 , it follows from lemma [ comm4 ] that @xmath464 is a product of interval swap maps .
since @xmath465 is the commutator of the restricted rotation @xmath458 and the interval exchange transformation @xmath436 , it is a product of interval swap maps due to lemma [ comm5 ] .
if @xmath466 then @xmath467 is the identity and we are done .
otherwise we observe that @xmath467 is a balanced product of @xmath468 restricted rotations . by the inductive assumption , @xmath467 is a product of interval swap maps , and so is @xmath5 .
theorem [ main1 ] let @xmath469 be the group of interval exchange transformations of an arbitrary interval @xmath7 .
let @xmath470 be the set of all elements in @xmath471 with zero saf invariant .
@xmath470 is a normal subgroup of @xmath471 as it is the kernel of the homomorphism @xmath32 ( see lemma [ inv3 ] ) .
let @xmath472 denote the commutator group of @xmath471 , i.e. , the subgroup of @xmath471 generated by commutators @xmath34 , where @xmath473 . also , let @xmath474 be the subgroup of @xmath471 generated by all elements of order @xmath0 and @xmath475 be the subgroup generated by all elements of finite order .
we have to prove that the groups @xmath470 , @xmath472 , @xmath474 , and @xmath475 coincide .
since the scissors congruence invariant @xmath32 is a homomorphism of @xmath471 to an abelian group , it vanishes on every commutator .
it follows that @xmath476 .
lemmas [ comm2 ] and [ comm6 ] imply that any element of @xmath470 is a product of interval swap maps , which are elements of order @xmath0 .
therefore @xmath477 .
the inclusion @xmath478 is trivial . by lemma [ elem2 ]
, any element of @xmath475 is a product of interval swap maps , which are commutators due to lemma [ elem3 ] .
hence @xmath479 .
we conclude that @xmath480 .
theorem [ main2 ] according to lemma [ inv3 ] , the saf invariant @xmath32 , regarded as a function on the group @xmath23 of interval exchange transformations of an interval @xmath11 , is a homomorphism to @xmath261 .
therefore the quotient of @xmath23 by the kernel of this homomorphism is isomorphic to its image . by lemma [ inv6 ] , the image of the homomorphism is @xmath33 . by theorem [ main1 ] ,
the kernel is the commutator group of @xmath23 .
let @xmath469 be the group of interval exchange transformations of an arbitrary interval @xmath7 . in this section we show that the commutator group @xmath481 $ ] of @xmath471 is simple .
let @xmath5 be an arbitrary interval swap map in @xmath471 . denote by @xmath37 the type of @xmath5 .
let @xmath160 and @xmath482 be the nonoverlapping intervals interchanged by @xmath5 .
we choose a sufficiently large positive integer @xmath381 such that @xmath483 . for any @xmath484 let @xmath415 denote the interval exchange transformation that interchanges intervals @xmath485 and @xmath486 by translation while fixing the rest of the interval @xmath11 .
it is easy to see that @xmath487 .
note that each @xmath415 is an interval swap map of type @xmath483 .
let @xmath488 be the subgroup of @xmath471 generated by all interval swap maps of types less than @xmath178 .
by the above the group @xmath488 contains all interval swap maps in @xmath471 . in view of lemma [ elem2 ] ,
@xmath488 coincides with the subgroup of @xmath471 generated by all elements of finite order . by theorem [ main1 ] , @xmath489 $ ] .
let @xmath101 be the length of the interval @xmath11 .
consider arbitrary interval swap maps @xmath491 of the same type @xmath492 .
let us divide the interval @xmath11 into @xmath493 pieces of length @xmath494 .
the support of @xmath135 is the union of two intervals of length @xmath37 . since @xmath492 , each interval of length @xmath37 overlaps with at most two of the ten pieces .
hence the support of @xmath135 overlaps with at most @xmath495 pieces .
the same is true for the support of @xmath136 .
therefore we have at least two pieces with interior disjoint from both supports .
clearly , one can find two nonoverlapping intervals @xmath113 and @xmath116 of length @xmath37 in these pieces .
let @xmath425 be the interval swap map of type @xmath37 that interchanges @xmath113 and @xmath116 by translation and fixes the rest of @xmath11 . by construction , the support of @xmath425 does not overlap with the supports of @xmath135 and @xmath136 .
it follows from lemma [ elem4 ] that @xmath496 and @xmath497 for some elements @xmath498 of order @xmath0 . by theorem [ main1 ]
, the commutator group @xmath481 $ ] contains all elements of order @xmath0 in @xmath471 .
in particular , it contains @xmath135 , @xmath136 , @xmath120 , and @xmath123 . then @xmath499 $ ] as well . since @xmath500 , the elements @xmath135 and @xmath136 are conjugated in @xmath481 $ ] .
theorem [ main3 ] suppose @xmath501 is a nontrivial normal subgroup of @xmath481 $ ] .
let @xmath5 be an arbitrary element of @xmath501 different from the identity .
by lemma [ elem7 ] , there exist @xmath502 and , for any @xmath503 , interval swap maps @xmath498 such that @xmath177 is an interval swap map of type @xmath178 .
the interval swap maps @xmath120 and @xmath123 are involutions .
they belong to @xmath481 $ ] due to lemma [ elem3 ] .
since @xmath501 is a normal subgroup of @xmath481 $ ] that contains @xmath5 , it also contains the interval exchange transformations @xmath96 , @xmath504 , @xmath505 , and @xmath506 .
we obtain that for any @xmath503 the subgroup @xmath501 contains an interval swap map of type @xmath178 . by lemma [ simp2 ] ,
there exists @xmath507 such that any two interval swap maps in @xmath471 of the same type @xmath508 are conjugated in @xmath481 $ ] .
it follows that all interval swap maps in @xmath471 of types less than @xmath509 are also in @xmath501 . according to lemma [ simp1 ] ,
the commutator group of @xmath471 is generated by these maps .
hence @xmath510 $ ] .
thus the only nontrivial normal subgroup of @xmath481 $ ] is @xmath481 $ ] itself .
that is , @xmath481 $ ] is a simple group . | we study the group of interval exchange transformations and obtain several characterizations of its commutator group . in particular
, it turns out that the commutator group is generated by elements of order @xmath0 . |
in these lectures we explore the subject of nonequilibrium dynamics . before getting into any kind of detail
let us first establish what we mean by a nonequilibrium system .
this is best done by taking stock of our understanding of an equilibrium system .
consider the canonical ( boltzmann ) distribution for a systems with configurations labelled @xmath1 each with energy @xmath2 : @xmath3 where @xmath4 .
the task is to calculate the partition function @xmath5 from which all thermodynamic properties , in principle , can be computed .
the distribution ( [ pcan ] ) applies to systems in thermal equilibrium _
i.e. _ free to exchange energy with an environment at temperature @xmath6 .
it can easily be generalised to systems free to exchange particles , volume etc but it always relies on the concept of the system being at equilibrium with its environment .
if one were interested in dynamics , for example to simulate the model on a computer , one might choose transition rates between configurations to satisfy @xmath7 where @xmath8 is the transition rate from configuration @xmath9 to @xmath1 .
condition ( [ db ] ) is known as detailed balance and guarantees ( under the assumption of ergodicity to be discussed below ) that starting from some nonequilibrium initial condition the system will eventually reach the steady state of thermal equilibrium given by ( [ pcan ] ) .
we will discuss further this dynamical relaxation process and the properties of the steady state endowed by the detailed balance condition in section [ sec : theory ] . for the moment we note
that a system relaxing to thermal equilibrium is one realisation of a nonequilibrium system . in recent years
such relaxation dynamics have been of special interest , for example , in the study of glassy dynamics whereby , on timescales realisable in experiment ( or simulation ) , the system never reaches the equilibrium state and it is a very slowly evolving nonequilibrium state that is observed .
this is sometimes referred to as ` off - equilibrium ' dynamics .
also let us mention the field of domain growth whereby an initially disordered state is quenched ( reduced to a temperature below the critical temperature for the ordered phase ) and relaxes to an ordered state through a process of coarsening of domains .
the interesting physics lies in the scaling regime of the coarsening process which is observed before the equilibrium ( ordered ) state is reached .
the other meaning of nonequilibrium refers to a system that reaches a steady state , but not a steady state of thermal equilibrium .
examples of such nonequilibrium steady states are given by driven systems with open boundaries where a mass current is driven through the system .
thus the system is driven by its environment rather being in thermal equilibrium with its environment .
a pragmatic definition of a nonequilibrium system that encompasses all of the scenarios above is as a model defined by its dynamics rather than any energy function _
i.e. _ the configurations of the model are sampled through a local stochastic dynamics which _ a priori _ does not have to obey detailed balance .
these notes broadly follow the four lectures given at the summer school .
in addition a tutorial class was held to explore points left as exercises in the lectures . in the present
notes these exercises are included in a self - contained form that should allow the reader to work through them without getting stuck or else leave them for another time and continue with the main text .
the notes are structured as follows : in section 2 we give an overview of two simple models that we are mainly concerned with in these lectures . in section
3 we then set out the general theory of the type of stochastic model we are interested in and point out the technical difficulties in calculating dynamical or even steady - state properties .
section 4 is an interlude in which we introduce , in a self - contained way , a mathematical tool the @xmath0-deformed harmonic oscillator that will prove itself of use in the final two sections . in section 5
we present the solution of the partially asymmetric exclusion process and amongst other things how the phase diagram ( figure [ fig : taseppd ] ) is generalised . in section
[ sec : sbac ] we discuss the exact solution of a stochastic ballistic annihilation and coalescence model .
in this work we will focus on two exemplars of nonequilibrium systems : the partially asymmetric exclusion process and a particle reaction model .
these models have been well studied over the years and a large body of knowledge has been built up @xcite .
we introduce the models at this point but will come back to these models in more detail in sections [ sec : pasep ] and [ sec : sbac ] in which we summarise some recent analytical progress .
the asymmetric simple exclusion process ( asep ) is a very simple driven lattice gas with a hard core exclusion interaction @xcite .
consider @xmath10 particles on a one - dimensional lattice of length @xmath11 say .
at each site of the lattice there is either one particle or an empty site ( to be referred to as a vacancy or hole)there is no multiple occupancy .
the dynamics are defined as follows : during each time interval @xmath12 each particle has probability @xmath13 of attempting a jump to its right and probability @xmath14 of attempting a jump to its left ; a jump can only succeed if the target site is empty . in this work will be concerned with the case @xmath15 so that we obtain continuous time dynamics with particles attempting hops to the right with rate 1 and hops to the left with rate @xmath0 . in this limit
no two particles will jump at the same time .
the other limit of @xmath16 would correspond to fully parallel dynamics where all particles attempt hops at the same time .
this type of dynamics is employed ( for @xmath17 ) in traffic flow modelling ( see schadschneider this volume ) . to complete the specification of the dynamics we have to fix the boundary conditions .
it turns out these are of great significance .
the following types of boundary conditions have been considered : periodic : : in this case we identify site @xmath18 with site @xmath19 .
the particles then hop around a ring and the number of particles is conserved .
open boundaries : : in this case particles attempt to hop into site @xmath19 with rate @xmath20 .
a particle at site @xmath11 leaves the lattice with rate @xmath21 .
thus the number of particles is not conserved at the boundaries .
one can also understand these boundary conditions in terms of a site 0 being a reservoir of particles with fixed density @xmath20 and site @xmath18 being a reservoir with fixed density @xmath22
. we shall primarily be concerned with these boundary conditions because , as first pointed out by krug @xcite , they can induce phase transitions .
the dynamics is illustrated in figure [ fig : pasep ] .
closed boundaries : : in this case the boundaries act as reflecting walls and particles can not enter or leave the lattice . thus one has a zero current of particles in the steady state and the system obeys detailed balance
. we shall not be interested in this case .
infinite system : : finally we could consider our lattice of size @xmath11 as a finite segment of an infinite system .
the macroscopic properties of the asep in which we are interested are the steady - state current ( flux of particles across the lattice ) and density profile ( average occupancy of a lattice site ) .
these can be expressed in terms of the binary variables @xmath23 , where @xmath24 if site @xmath25 is occupied by a particle and @xmath26 if site @xmath25 is empty , and which together completely specify a microscopic configuration of the system .
then , the density at site @xmath25 is defined as @xmath27 where the angular brackets denote an average over all histories of the stochastic dynamics .
one can think of this average as being an average over an ensemble of systems all starting from the same initial configuration at time 0 .
let us consider for the moment the totally asymmetric dynamics @xmath17 ( _ i.e. _ no backward particle hops are permitted ) .
one can use the ` indicator ' variables @xmath23 intuitively to write down an equation for the evolution of the density .
for example , if we choose a non - boundary site ( so that we can avoid prescribing any particular boundary conditions ) we obtain @xmath28 note that @xmath29 gives the probability that site @xmath30 is occupied and site @xmath25 is empty .
thus , since particles hop forward with rate 1 , first term on the right hand side gives the rate at which particles enter site @xmath25 and the second term gives the rate at which they leave site @xmath25 .
if one needs more convincing of the argument used to obtain ( [ ev1 ] ) consider what happens at site @xmath25 in an infinitesimal interval @xmath31 : @xmath32 \tau_i(t ) + \left [ 1-\tau_i(t ) \right ] \tau_{i-1}(t ) & { \ensuremath{\mathrm{d}}}t \\[0.2ex ] \tau_{i}(t)\tau_{i+1}(t ) & { \ensuremath{\mathrm{d}}}t \end{array } \right .
\;.\ ] ] the first equation comes from the fact that with probability @xmath33 ( dropping terms of @xmath34 ) , neither of the sites @xmath30 or @xmath25 is updated and therefore @xmath35 remains unchanged . the second and
the third equations correspond to updating sites @xmath30 and @xmath25 respectively . if one averages ( [ rule ] ) over the events which may occur between @xmath36 and @xmath37 and all histories up to time @xmath36 one obtains ( [ ev1 ] ) .
the same kind of reasoning allows one to write down straight away an equation for the evolution of @xmath38 .
@xmath39 or for any other correlation function @xmath40 .
these equations are exact and give in principle the time evolution of any correlation function .
however , the evolution ( [ ev1 ] ) of @xmath41 requires the knowledge of @xmath42 which itself ( [ ev2 ] ) requires the knowledge of @xmath43 and @xmath44 so that the problem is intrinsically an @xmath11-body problem in the sense that the calculation of any correlation function requires the knowledge of all the others .
this is a situation quite common in equilibrium statistical mechanics where , although one can write relationships between different correlation functions , there is an infinite hierarchy of equations which in general makes the problem intractable .
for the case of open boundary conditions the number of particles in the system is not conserved .
the evolution equations ( [ ev1]-[ev2 ] ) are then valid everywhere in the bulk but they are modified at the boundary sites . for example ( [ ev1 ] ) becomes @xmath45 @xmath46 in the steady state , where the time derivatives of correlation functions are zero , ( [ ev1],[ev3],[ev4 ] ) can be rewritten as a conserved current @xmath47 @xmath48 these equations simply express that for the density to be stationary one must have the current into any site equal to the current out of that site .
for example , @xmath49 is the probability that the leftmost site is empty multiplied by the rate at which particles are inserted onto it , and hence yields the current of particles entering the system .
similarly @xmath50 is the rate at which particles leave the system . to motivate the study of this model
let us consider a few applications .
first we may consider the asep as a very simplistic model of traffic flow with no overtaking ( more realistic generalisations are discussed by schadschneider , this volume ) .
joining many open boundary systems together into a network forms a very simple model for city traffic .
such a network appears to represent faithfully real traffic phenomena such as jamming @xcite .
a major motivation for the study of the asep is its connection to interface dynamics and hence to the kpz equation ( noisy burgers equation ) .
the kpz equation is a stochastic non - linear partial differential equation that is central to much of modern statistical physics . in these lectures
we do not have time to develop the theory of the kpz equation , rather we refer the reader to @xcite . here
we just make clear the connection to a particular interface growth model . the asymmetric exclusion model may be mapped exactly onto a model of a growing interface in @xmath51 dimensions @xcite as shown in figure [ fig : singlestep ] .
the mapping is obtained by associating to each configuration @xmath52 of the particles a configuration of an interface : a particle at a site corresponds to a downwards step of the interface height of one unit whereas a hole corresponds to an upward step of one unit .
the heights of the interface are thus defined by @xmath53 the dynamics of the asymmetric exclusion process in which a particle at site @xmath30 may interchange position with a neighbouring hole at site @xmath25 , corresponds to a growth at a site @xmath25 which is a minimum of the interface height _
i.e. _ if @xmath54 then @xmath55 a growth event turns a minimum of the surface height into a maximum thus the asep maps onto what is known as a single - step growth model @xcite meaning that the difference in heights of two neighbouring positions on the interface is always of magnitude one unit .
since whenever a particle hops forward the interface height increases by two units , the velocity @xmath56 at which the interface grows is related to @xmath47 the current in the asymmetric exclusion process by @xmath57 periodic boundary conditions for the particle problem with @xmath10 particles and @xmath58 holes correspond to an interface satisfying @xmath59 , _ i.e. _ to helical boundary conditions with an average slope @xmath60 .
the case of open boundary conditions corresponds to special growth rules at the boundaries . because of this equivalence
, several results obtained for the asymmetric exclusion process can be translated into exactly computable properties of the growing interface .
one can also go on to list further applications @xcite . indeed early applications concerned biophysical problems such as single - filing constraint in transport across membranes @xcite and the kinetics of biopolymerisation @xcite .
let us now discuss some interesting results that have been derived over the last decade for the open boundary system @xcite .
for the moment we present them without proof , although we shall recover some of them in the analysis of later sections . in figure
[ fig : taseppd ] we present the phase diagram ( corresponding to the limit @xmath61 ) for the totally asymmetric exclusion process _
i.e. _ when @xmath17 as predicted within a mean - field approximation @xcite .
it should be noted that we consider the steady state in the limit @xmath62 , thus it is implicit that we have taken the limit @xmath63 first .
note there are three phases by phase it is meant a region in the phase diagram where the current has the same analytic form ; phase boundaries divide regions where the current @xmath47 and bulk density @xmath64 ( defined as the mean occupancy of a site near the centre of the lattice as @xmath61 ) have different forms : for @xmath65 and @xmath66 , @xmath67 for @xmath68 and @xmath69 , @xmath70 for @xmath71 and @xmath72 , @xmath73 consider the high density phase . in this phase
the exit rate is low and the bulk density is controlled by the right boundary .
one can think of this as queue of traffic formed at a traffic light that does nt let many cars through at a time .
the low density phase is the opposite scenario where the left boundary controls the density through a low input rate of particles .
one can think of this as cars being let onto an empty road only a few at a time .
finally in the maximal current phase the current of particles is saturated _
i.e. _ increasing @xmath20 or @xmath21 any further does not increase the current or change the bulk density . in this phase the density of particles decays from @xmath20 its value at the left boundary to @xmath74 its value in the bulk as @xmath75 where @xmath76 is the distance from the left boundary .
this phase diagram of the model has many interesting features .
firstly we have both continuous ( from high density or low density to maximal current ) and discontinuous ( from low density to high density ) phase transitions even though it is a one - dimensional system . in the former transitions there are discontinuities in the second derivative of the current . in the latter
although the current is continuous across the phase transition its first derivative and the bulk density jump . along the transition line @xmath77 one has coexistence between a region of the high density phase adjacent to the right boundary and a region of the low density phase adjacent to the left boundary .
these results are very suggestive that the current @xmath47 can be thought of as some free energy .
we shall see later through an exact solution how this can be quantified
. also note that throughout the maximal current phase one has long - range , power - law correlations functions _
i.e. _ long - range correlations are generic .
this contrasts with usual behaviour in equilibrium systems where power - law correlations are seen at non - generic , critical points ( an exception is the kosterlitz - thouless phase that may emerge at low temperature in two - dimensional equilibrium systems ) .
we now have a look at reaction - diffusion systems wherein particles can react with each other to form some by - product or more simply annihilate each other . well - studied reaction systems are @xmath78 the first reaction signifies that when two @xmath79 particles meet they react and annihilate each other , the second that the two reagents coalesce . these systems model molecular reactions and the dynamics of laser induced quasiparticles known as excitons @xcite .
however the particles in this type of system can also represent composite objects such as aggregating traffic jams or , as we shall see , domain walls in phase - ordering kinetics .
consider first the @xmath80 reaction .
clearly the steady state of such a system is the uninteresting dead state where all particles ( except perhaps one ) have been annihilated .
the true interest lies in the late time regime wherein the system may enter a _ scaling _ regime . by this
it is meant that the system is statistically invariant under rescaling by a typical length scale that depends on time . in this case
the length scale can be taken as the typical distance between particles . for diffusing @xmath79 particles ,
known results are that for @xmath81 the typical distance between particles grows as @xmath82 or equivalently the density of a particles decreases as @xmath83 .
we shall recover this result in the next section .
the result differs from a naive mean - field description obtained by assuming @xmath84 which would predict @xmath85 .
a similar density decay law holds for the @xmath86 process and it has been shown that the two processes are basically equivalent @xcite .
in addition to having a range of applications , diffusive reaction systems have served as prototypes for the development of a variety of theoretical tools including field - theoretic techniques @xcite and the renormalisation group @xcite and exact methods in one dimension @xcite .
one can also consider _ ballistic _ particle motion . in this class of models ,
particles move deterministically with constant velocity and on meeting have some probability of coalescing into one particle or annihilating .
a seminal paper by elskens and frisch @xcite that considered the one - dimensional deterministic case where right moving and left moving particles always annihilate on contact , showed the decay depends on the initial densities of the two species .
in particular , @xmath83 only if there are equal initial densities of the right - moving and left - moving particles .
ballistic models apply to chemical reactions when the inter - reactant distance is less than what would be the mean free path of particles .
such models may also be applied to various domain growth problems @xcite and as we shall describe in section 6 , the smoothing of a growing interface @xcite .
having now given a rough overview of some of the interest in low dimensional nonequilibrium systems , the rest of these lectures will be aimed at a given a deeper understanding of how these phenomena come about in some of the simple models we have discussed . in this section we give a brief introduction to stochastic processes focussing on particular aspects that will be relevant to the models to be studied in detail in later sections .
we consider a system that can be in a finite number of microscopic configurations and whose configuration changes according to transition rates .
the simplest example is of a particle performing a random walk in continuous time on a one - dimensional lattice : the particle hops to the right with rate @xmath87 and to the left with rate @xmath0 .
( the meaning of a ` rate ' @xmath88 is that in an infinitesimal time interval @xmath31 an event happens with probability @xmath89 . ) since probability is conserved we can write a continuity equation for the probability which is referred to as ` the master equation ' @xmath90 the first two terms on the right hand side represent ` rates in ' from other configurations ( _ i.e. _ different positions of the particle ) ; the last term represents the rate out of the given configuration .
note that the master equation is linear in the probabilities . for the case of the random walker ,
the master equation ( [ rw ] ) can be solved analytically : it is a diffusion equation in discrete space but continuous time .
if we consider a periodic lattice of @xmath11 sites ( site @xmath18 is identified with site 1 ) the preferred method is to calculate the fourier modes .
defining @xmath91 yields from ( [ rw ] ) @xmath92 thus @xmath93 the steady state corresponds to the mode @xmath94 ( @xmath95 ) .
the eigenvalues @xmath96 yield decay times @xmath97 of each mode . for @xmath98 finite and @xmath11 large @xmath99
thus the equilibration time goes like @xmath100 where the dynamic exponent @xmath101 as expected of a diffusion process .
random walker on the 1d lattice review the solution of equation ( [ rw ] ) using fourier modes described above . show that if the initial condition is @xmath102 the probability distribution at time @xmath36 is given by @xmath103 where @xmath96 is given by ( [ lambdak ] ) if you are unfamiliar with the discrete fourier transform , you should first show that for @xmath104 @xmath105 and hence @xmath106 where @xmath107 is as given by equation ( [ pfourier ] ) .
consider now the same random walk problem but on an infinite rather than periodic lattice .
show from equation ( [ rw ] ) that the generating function @xmath108 obeys @xmath109 f(z , t)\ ] ] and hence @xmath110 \exp[\sqrt{pq}(z+z^{-1})t]\;. \label{ex : rwgen}\ ] ] given that the generating function of modified bessel functions of the first kind @xmath111 is defined as @xmath112\;,\ ] ] show by expanding ( [ ex : rwgen ] ) that the general solution for the random walk problem is @xmath113 \sum_{\ell = -\infty}^{\infty } \left ( \frac{p}{q } \right)^{\!\frac{x-\ell}{2 } } p(\ell , 0 ) i_{x-\ell}(2 \sqrt{pq}\ , t ) \;.\ ] ] some of the properties of the modified bessel functions @xmath111 can be gleaned by considering the special case of a symmetric random walker that begins at the origin , _ i.e. _ the case @xmath114 and @xmath115 .
then , @xmath116 .
show then that the property @xmath117 immediately follows , as does the fact that @xmath118 .
also , show from ( [ rw ] ) that the bessel functions satisfy the differential equation @xmath119 \;.\ ] ] more generally we consider ` many - body ' problems .
for example a system of many particles on a lattice obeying stochastic dynamical rules . for concreteness
we consider the simple example of the one - dimensional ising model in which at each site @xmath25 there is a spin @xmath120 and periodic boundary conditions are imposed . as mentioned in the introduction the dynamics of an equilibrium model
is usually chosen to satisfy detailed balance . in the case of the ising model under spin - flip dynamics , whereby each spin can flip with a certain rate according to the directions of the neighbouring spins , the detailed balance condition on the spin - flip rates of spin @xmath121 at site @xmath25 reads @xmath122 \label{isingdb}\ ] ] ( we have taken @xmath123 ) . in _ glauber
_ dynamics ( [ isingdb ] ) is satisfied by choosing @xmath124 } { 2}\ ] ] as is easily verified .
then , in the @xmath125 limit the spin flips for a down - spin ( @xmath126 ) at site @xmath25 occur with the rates indicated in figure [ isingt=0 ] and similarly for up - spins at @xmath127 .
@xmath128 note the final transition where the spin flips against both its neighbours is forbidden .
thus domains of aligned spins can not break up .
the domain walls between aligned domains _
i.e. _ neighbouring pairs of spins @xmath129 or @xmath130 make random walks on the bonds of the original lattice and annihilate when they meet . as noted above ,
this process of diffusing ` particles ' ( here the domain walls ) that annihilate on meeting is sometimes denoted @xmath131 .
the solution for the mean domain length has been known for some time @xcite basically it can be solved by considering the random walk performed by the length of a domain ( distance between neighbouring particles ) . to see the solution for the mean domain size within the kinetic ising model @xcite consider the master equation which can be written @xmath132 where @xmath1 is a configuration of the ising spins and @xmath133 is the same configuration with spin @xmath121 flipped .
the number of domain walls is given by @xmath134 so to study the coarsening process , in which domain walls are eliminated and aligned domains increase in size , one considers the equal time correlation function @xmath135 where the average indicates an average over initial conditions and possible histories of the stochastic dynamics .
it can be shown that @xmath136 obeys a discrete diffusion equation @xmath137 with boundary condition @xmath138 .
the fact that we obtain a closed set of equations for two point correlation functions is a very happy property that allows one to solve easily for these correlation functions basically ( [ ceq ] ) is an equation of the same form as ( [ rw ] ) but with a boundary condition corresponding to a source of walkers at site 0 : see @xcite for details .
note that the method generalises to finite temperature but not to higher dimensions .
actually higher point correlation functions are more difficult to calculate ; nevertheless it has been shown that the one - dimensional glauber model can be turned into a free fermion problem which then implies , in principle at least , that any correlation function can be calculated @xcite .
remarkably , the full domain size probability distribution has been explicitly calculated for the one - dimensional @xmath0-state potts model , which includes the ising model as the case @xmath139 @xcite .
( note that a calculation such as exercise 2 only calculates the first moment of the domain sizes ) .
the dynamics of domain walls in the @xmath0-state potts model , represented as @xmath79 particles , corresponds to diffusion and , on meeting , reaction according to @xmath140{ll } a \quad \mbox { with probability}\quad ( q-2)/(q-1)\\ \emptyset \quad \mbox{with probability}\quad 1/(q-1 ) \end{array } \right.\ ] ] to understand this equation note that when a domain is elimated each of the neighbouring domains may be in one of @xmath141 states so that they have probability @xmath142 of being in the same state and therefore coalescing .
thus for @xmath139 we recover @xmath80 and for @xmath143 ( infinite state potts model ) we recover @xmath86 .
time dependence of the 1d ising model the mean magnetisation @xmath144 in the 1d ising model is defined as @xmath145 where @xmath11 is the number of spins on the lattice , @xmath146 is the set of all @xmath147 possible spin combinations and @xmath148 is the value of the spin at site @xmath25 , in configuration @xmath1 .
show for general single spin - flip dynamics as described by ( [ kineticisingmaster ] ) that the magnetisation obeys the differential equation @xmath149 under glauber dynamics defined by equation ( [ glauberrates ] ) show that @xmath150 and hence that @xmath151t)$ ] .
also comment on the finite- and zero - temperature behaviour ( recalling that @xmath21 is inverse temperature ) .
hint : you will need to argue that the relation @xmath152 holds for any ( periodic ) combination of spins .
show that the mean density of domain walls @xmath153 in the 1d ising model can be expressed using the spin correlation function @xmath154 as @xmath155 $ ] . under zero - temperature glauber dynamics ,
the evolution of @xmath154 is described by the set of difference equations ( [ ceq ] ) which has the stationary solution @xmath156 .
verify this solution , and explain its physical meaning . to obtain the time - dependence of @xmath154
we introduce a function @xmath157 through @xmath158 .
the time - evolution is governed by equation ( [ rw ] ) with @xmath159 , @xmath160 and subject to the boundary condition @xmath161 .
this boundary condition can be satisfied through an appropriate choice of the initial conditions @xmath162 in the ` unphysical ' region @xmath163 ( this is equivalent to using the method of images @xcite ) . using the general solution ( [ ex : rwsoln ] ) and
the properties of the modified bessel functions derived in exercise 1 , show that @xmath164 if @xmath165 .
hence show that @xmath166\ ] ] is the solution for the spin correlation function that has @xmath167 and with the initial spin orientations uncorrelated , _
i.e. _ @xmath168 .
show that this implies that for these initial conditions , @xmath169 $ ] . finally , using the asymptotic form of the modified bessel functions @xmath170\ ] ]
show that the long - time behaviour of @xmath153 is @xmath171 . in order to formalise stochastic processes in general
we consider a master equation @xmath172 where @xmath173 is the rate of transition from configuration @xmath1 to @xmath9 .
it is useful to combine the transition rates into a single matrix @xmath174 so that we can write @xmath175 now let us introduce a ` bra - ket ' notation .
let the vector of probabilities at time @xmath36 ( one component for each configuration ) be @xmath176 .
thus the probability of the system being in configuration @xmath1 at time @xmath36 is @xmath177 and we can write the master equation as @xmath178 thus the master equation looks something like a schrdinger equation with ` imaginary time ' and @xmath179 replaced by @xmath180 ( this is why @xmath10 is sometimes referred to as a hamiltonian )
. however it should be recalled that in quantum mechanics it is the modulus squared of the components @xmath176 that are probabilities whereas here the components of @xmath176 are probabilities .
see @xcite , for example , for more details .
the formal solution of the master equation ( [ meq ] ) is @xmath181 where @xmath182 gives the initial conditions .
however , a more useful way of dealing with the problem is to consider the eigenvectors of the transition matrix @xmath10 .
the first thing to note is that since @xmath10 is not hermitian , in general , it has different left and right eigenvectors @xmath183 these satisfy ` bi - orthogonality ' @xmath184 and in principle ( assuming @xmath10 can be diagonalised ) we may write @xmath185 then we have using ( [ formal ] ) , ( [ complete ] ) and ( [ rev ] ) @xmath186 note that the real parts of all eigenvalues @xmath187 are negative except for one eigenvalue which is zero . the reason for this
is often referred to as the perron - frobenius theorem @xcite .
this theorem holds for a matrix with non - negative entries where a sufficiently large power of the matrix has all positive entries .
it states that the maximum eigenvalue is non - degenerate and the corresponding left and right eigenvectors have positive entries .
in our context one can relate the spectrum of @xmath10 to that of non - negative matrix @xmath188 obtained by adding some suitable multiple ( say minus the diagonal element with the largest magnitude ) of the identity matrix to @xmath10 .
the assumption that a sufficiently large power of @xmath188 has all positive entries is equivalent to the assumption that the ( finite ) system is ergodic _
i.e. _ any configuration can be reached from any other after sufficient time . in that case
the system will have a unique steady state so that the zero eigenvalue can not be degenerate and the steady state eigenvector has positive elements which correspond to probabilities . in principle , all probabilities at all times can be found if one can diagonalise the matrix @xmath10 .
unfortunately this is only possible in a few isolated cases .
generally this corresponds to many - particle systems where the particles do not interact or certain ` integrable ' systems that can be solved via the bethe ansatz .
the bethe ansatz was originally used in the context of a spin - chain model for magnetism @xcite and is suitable for a one - dimensional lattice system in which the number of particles is conserved .
it is basically a generalisation of the fourier transform .
let us briefly discuss it for the case of the asep .
the idea is to guess the following form for eigenvectors @xmath189 @xmath190 in which @xmath191 is the position of particle @xmath25 on a lattice , and @xmath192 with @xmath11 the lattice size once again .
the sum is over all permutations @xmath193 of @xmath194 and @xmath195 is some amplitude .
one uses the master equation to determine relations that must be satisfied by @xmath196 in its unphysical regions , _
e.g. _ where two particles occupy the same site and @xmath197 for some @xmath25 .
after some manipulation , one arrives at a set of nonlinear equations that must be simultaneously satisfied by all the ` momenta ' @xmath198 .
an early account of this method in the context of stochastic processes is given in @xcite .
the technique is also described in @xcite .
however even for the so - called ` integrable ' systems where the bethe ansatz works it is actually very difficult to extract the explicit eigenvectors . of course , it may not be possible to solve a model exactly , and in this situation one often turns to approximation schemes .
for example , one can try a ` short time ' expansion we simply truncate the power series defining the exponential of @xmath10 in ( [ formal ] ) after @xmath199 terms @xmath200 the finite number of terms @xmath199 on the right hand side can be calculated . from the series one can try to estimate , for example , the dynamical exponent @xcite .
a project less ambitious than solving the full dynamics is to calculate the steady state probabilities ( the eigenvector with eigenvalue 0 ) .
we denote the steady state by dropping the subscript @xmath36 so that @xmath201 is the eigenvector with eigenvalue zero @xmath202 and its components @xmath203 are the steady state probabilities .
thus one has the balance conditions @xmath204 let us review some simple ways to satisfy ( [ sscon ] ) .
first of all one has the case of detailed balance where @xmath205 however to check this condition _
i.e. _ to check whether detailed balance is satisfied we first need to know the steady state
! however there is an equivalent set of conditions called ` kolmogorov criteria ' which state that detailed balance is obeyed if and only if the transition rates satisfy @xmath206 for _ every _ possible cycle in configuration space @xmath207 . that ( [ meqn : db1 ] ) implies ( [ meqn : db2 ] ) is very simple to show . to show that the converse is also true
is most simply achieved @xcite by using ( [ meqn : db1 ] ) to determine @xmath203 .
then , one finds that @xmath203 is unique if and only if ( [ meqn : db2 ] ) is satisfied .
let us consider some of the implications of the detailed balance condition .
first note that when one has detailed balance then in the steady state there is no net flow of probability between any two configurations .
since there is no flow of probability there is nothing to distinguish the forwards direction in time from backwards direction .
therefore running the systems backwards in time will not change , for example , any two time correlation functions .
thus the system is said to be _
reversible_. now consider the simple problem of a single random walker on a periodic one - dimensional lattice , studied at the beginning of this section .
the steady state is given by equal probability of the walker occupying each site .
however , one does not have detailed balance as defined above unless the hop rates are symmetric @xmath208 .
but one knows that the only effect of @xmath209 is to impose a drift . clearly with a drift
the system is not reversible .
but the system is reversible under simultaneous reversal of time and parity ( _ i.e. _ direction ) : the parity operation results in defining the image of position @xmath76 as @xmath210 .
then trivially @xmath211 since in the steady state @xmath212 is constant in space and both transition rates are equal to @xmath87 .
more generally an _ extended _ detailed balance condition can pertain as defined in @xcite : @xmath213 where @xmath214 is the image configuration of @xmath1 under a reversible mapping .
one requires @xmath215 and @xmath216 again one can construct an equivalent condition as follows @xcite @xmath217 for _ every _ possible cycle in configuration space @xmath218 along with the additional condition ( [ addcon ] ) .
the condition ( [ edb ] ) is known in the probabilistic literature as _ dynamic reversibility _ @xcite .
more recently it has appeared in the physics literature under the name _ pairwise balance _ @xcite .
it holds for example in the asep on a ring where the image configuration is obtained by the parity operation . in this subsection
our aim is to derive the general solution for the steady state ( [ meqn : solveme ] ) for an arbitrary transition rate matrix @xmath10 .
although the solution is too general to be of any practical use it is instructive to see how a quantity analogous to a partition function naturally arises .
as it stands , the set of linear equations ( [ meqn : solveme ] ) is underdetermined because the determinant of @xmath10 is zero ( one of its eigenvalues is zero ) . however , we know that a probability distribution must be properly normalised , so consider instead the equation @xmath219 in which @xmath220 is the matrix obtained from @xmath180 ( the minus sign is introduced for convenience below ) by replacing the @xmath221 row with all ones , _
i.e. _ the vector @xmath222 .
the ket vector @xmath223 is the basis vector corresponding to configuration @xmath25 .
this set of equations can now be solved and we do so using cramer s rule .
this states that the probability of the system being in configuration @xmath224 in the steady state is @xmath225 in which the matrix @xmath226 is obtained from @xmath220 by replacing column @xmath227 with a vector with a 1 at position @xmath227 and zeros everywhere else . for simplicity
we choose @xmath228 , and to be clear we write out @xmath229 explicitly .
@xmath230 \;.\ ] ] we note that the determinant @xmath231 is equal to the determinant of the matrix obtained from @xmath232 by removing row @xmath227 and column @xmath227 . this determinant is called a cofactor of @xmath188 and will be written @xmath233 . furthermore , we observe that the determinant in the denominator of ( [ meqn : cramer ] ) is the sum of all these cofactors .
that is @xmath234 so that @xmath235 we thus identify the cofactor @xmath236 with the _ steady - state weight _ of configuration @xmath224 and @xmath237 as a _ normalisation _ for the stochastic process . in an equilibrium system
, @xmath237 is related to the partition function @xmath238 .
however , the normalisation @xmath237 obtained using the above procedure is not guaranteed to be identical to @xmath239 due to the possibility of common factors present in all the weights @xmath236 and hence @xmath237 . with this understood , ( [ zneq ] )
extends the notion of a partition function to nonequilibrium systems .
we now note an interesting and , as far as we are aware , little - known result that relates the normalisation @xmath237 ( [ zneq ] ) to the eigenvalues of the transition matrix .
the relation reads @xmath240 in which the product is over the eigenvalues of the transition rate @xmath10 except the zero eigenvalue .
this result is obtained by expanding the characteristic polynomial @xmath241 in powers of @xmath187 .
one finds @xcite @xmath242 with @xmath236 as defined above . given that @xmath243 and @xmath244 we find that @xmath245 where the last step follows because one , and only one , eigenvalue of @xmath10 is zero ( at least if the process is ergodic ) .
we now consider the existence of _ phase transitions _ in a stochastic process . as in equilibrium systems
, we identify phase transitions through discontinuities in macroscopic quantities or their derivatives . furthermore , at a first - order ( discontinuous ) transition we expect phase coexistence and at a continuous transition divergences in length and time scales . in terms of eigenvalues of the transition matrix @xmath10
, a phase transition would be indicated by the clustering of one or more eigenvalues towards zero .
for example as a discontinuous transition is approached one would expect long - lived ` excited ' states since the decay time of an eigenstate of @xmath10 is inversely proportional to the real part of the corresponding eigenvalue .
these long - lived states can be thought of as metastable states and the approach of an eigenvalue to zero in the thermodynamic ( infinite system - size ) limit suggests phase coexistence . near a continuous transition one
has diverging time scales ( as manifested by _ e.g. _ a power - law decay of an autocorrelation function ) so one expects a continuous spectrum of eigenvalues whose real parts are close to zero . bearing this in mind , we note that equation ( [ meqn : zevals ] ) implies that the steady - state normalisation @xmath237 also approaches zero at a phase transition in the thermodynamic limit .
such a scenario is analogous to yang - lee theory of equilibrium phase transitions , in which it has been rigorously shown that zeros of the grand canonical partition function in the complex - fugacity plane converge towards a point on the positive real axis that indicates the value of the fugacity at which a phase transition occurs .
similar properties are also displayed by the zeros of the canonical partition function in the complex - temperature plane ( these are also known as fisher zeros ) .
hence , it would seem likely from ( [ meqn : zevals ] ) and the above considerations that zeros of a steady - state nonequilibrium normalisation in the complex plane of some control parameter would also converge to the real axis at a nonequilibrium phase transition .
this would also imply nonanalyticities in a macroscopic variable averaged with respect to the steady - state probability distribution function .
this idea has been pursued in the context of an asymmetric exclusion process @xcite . in this case , the steady - state normalisation was solved for its zeros in the complex - fugacity plane and indeed they were found to approach the real axis for values of the model parameters that corresponded to a phase transition .
additionally , a similar approach has been tried in the context of directed percolation @xcite , a further lattice model that exhibits a nonequilibrium phase transition ( see _ e.g. _
@xcite for a review ) .
this concludes our overview of stochastic processes .
we will shortly move on to the specific models we are going to study . as mentioned in the introduction
the ` @xmath0-deformed harmonic oscillator ' plays a central role in the solution of both these models .
first let us recall the quantum harmonic oscillator well known from undergraduate physics .
the hamiltonian is @xmath246 where we have lazily set all constants @xmath247 etc to unity and where @xmath248 denotes an operator .
one way to treat the problem is to introduce two new operators @xmath249 through @xmath250 then inserting these definitions into the commutation relation @xmath251 = i{\mbox{$1\!\!1$}}$ ] yields @xmath252 also the hamiltonian ( [ ho ] ) becomes @xmath253 the operators @xmath249 are , of course , raising and lowering operators .
more generally they are bosonic operators , the defining property of which is the commutation relation ( [ boson ] ) . in the number ( or energy )
basis they obey @xmath254 thus in this basis @xmath255 and the hamiltonian is diagonal .
the number @xmath256 labels the energy eigenstates .
the quantised energy excitations obey bose statistics and the operators ( [ energybasis ] ) create and annihilate bosons respectively .
the hermite polynomials come into play when we consider the projection of an energy eigenstate onto the position eigenbasis defined by @xmath257 then @xmath258 where @xmath259 is the hermite polynomial of degree @xmath256 .
we now consider a ` deformation ' of ( [ boson ] ) by introducing a parameter @xmath0 @xcite @xmath260 observe that taking @xmath261 recovers bosonic operators whereas @xmath262 gives fermionic operators .
thus it was originally hoped that the @xmath0-deformed operators might describe some new excitations interpolating between fermions and bosons .
the operators @xmath263 and @xmath264 now operate on basis vectors @xmath265 ( with @xmath266 ) as follows : @xmath267 thus @xmath268 , which is the analogue of the energy operator , is diagonal in this basis @xmath269 the eigenstates of the ` position operator ' @xmath270 are given by @xmath271 finally , we identify the @xmath0-hermite polynomials as @xmath272 using ( [ adqaction],[aqaction ] ) and ( [ qherm ] ) it is straightforward to derive a recurrence relation @xmath273 this can be compared to the usual recurrence relation for hermite polynomials @xmath274 to make the connection between @xmath0-hermite and ordinary hermite polynomials we would like the recurrence relations to coincide in the limit @xmath275 .
the prescription is @xmath276 in which both the hermite polynomial @xmath277 and the independent variable @xmath76 are transformed .
the factors of @xmath278 that appear can be traced back to the fact that we set @xmath279 when doing the quantum mechanics , but we take @xmath270 in the @xmath0-deformed case . explicit formul for @xmath280 can be found using a generating function technique , the details of which differ slightly depending on whether @xmath281 or @xmath282 . here
we discuss in detail the case @xmath281 .
first we define a generating function @xmath283 for the @xmath0-hermite polynomials @xmath284 where have introduced a ` @xmath0-factorial ' notation defined through @xmath285 we later encounter products of these factorials for which we use a standard shorthand @xcite : @xmath286 the @xmath0-factorial in ( [ eqn : gdef ] ) is introduced for convenience as will now become apparent .
we obtain a functional relation for @xmath283 by multiplying both sides of equation ( [ qhreccur ] ) by @xmath287 and performing the required summations : @xmath288 it is convenient to parameterise the ` position ' eigenstates by an angle @xmath289 where @xmath290 and @xmath291 , and to replace @xmath292 by @xmath293 ( a function of @xmath289 ) . then ( [ eqn : gfuncrel ] ) becomes @xmath294 since @xmath281 we can iterate to obtain @xmath295 where we have used the @xmath0-factorial notation ( [ qfacgen ] ) .
note that @xmath296 is just @xmath297 which we are free to set to @xmath19 .
the infinite product @xmath298 has a well - known and easy to verify series representation @xcite valid for @xmath299 , @xmath281 @xmath300 from which , with a little effort , we may extract the form of @xmath301 . expanding both sides of ( [ eqn : gnice ] ) in @xmath187 and comparing coefficients we find @xmath302}{0pt}{}{\,n\,}{\,k\,}_{\!q } } } e^{i(n-2k)\theta}\ ] ] where @xmath303}{0pt}{}{\,n\,}{\,k\,}_{\!q}}}$ ] is the @xmath0-deformed binomial defined as @xmath304}{0pt}{}{\,n\,}{\,k\,}_{\!q}}}= \frac{(q;q)_n}{(q;q)_{n - k } ( q;q)_k}\ ] ] when @xmath305 and zero otherwise . in the limit
@xmath275 the @xmath0-binomial coefficient is equal to the conventional version @xmath306 familiar from combinatorics .
important properties of the @xmath0-hermite polynomials are their orthogonality and completeness
. it can be shown @xcite that the set of @xmath0-hermite polynomials are orthogonal with respect to a weight function @xmath307 .
that is @xmath308 where the weight function is given by @xmath309 completeness implies we can form a representation of the identity matrix : @xmath310 it is less well known that , in a similar manner to the above exposition , one can obtain explicit forms for @xmath0-hermite polynomials when @xmath282 @xcite .
it turns out that they can be obtained from ( [ eqn : hp : q<1 ] ) by making the substitution @xmath311 _ i.e. _ one replaces @xmath312 by @xmath313 and the parameter @xmath314 runs from @xmath315 to @xmath316 .
the weight function that orthogonalises these polynomials is now @xmath317 coherent states , the exponential function and their @xmath0-deformations for the ` traditional ' quantum mechanical raising and lowering operators that satisfy the commutation relation ( [ boson ] ) construct the vector @xmath318 .
show that this is a _ coherent state _ , _
i.e. _ an eigenvector of the lowering operator @xmath263 with eigenvalue @xmath319 .
now consider a @xmath0-coherent state @xmath320 satisfying the eigenvalue equation @xmath321 with @xmath263 and @xmath264 the @xmath0-deformed raising and lowering operators that obey ( [ qboson ] ) . by solving this eigenvalue equation ,
show that the @xmath0-coherent state @xmath320 may be expressed as @xmath322^{1/2 } \mu \hat{a}^\dagger ) { \ensuremath{| 0 \rangle}}\ ] ] where the @xmath0-deformation of the exponential function is defined through @xmath323 show that this series converges always for @xmath282 , for @xmath299 when @xmath281 and that @xmath324x ) = \exp(x)$ ] .
verify also that when @xmath325 and @xmath326 converges , it can also be written as @xmath327 hint : this is most easily achieved by checking that both the series and product representations of @xmath326 satisfy the functional relation @xmath328 and that they are numerically equal for a special value of @xmath329 .
finally , show that @xmath330 where @xmath331 is the generating function of @xmath0-hermite polynomials ( [ eqn : gdef],[eqn : gnice ] ) .
we now consider in more detail the partially asymmetric exclusion process ( pasep ) introduced in section [ sec : asepintro ] . recall that we discussed some of the properties of the special case where the reverse hop rate @xmath0 was set to zero and in particular we presented a phase diagram for the model ( figure [ fig : taseppd ] ) .
we now describe how one obtains such a phase diagram exactly for general @xmath0 .
we do this by relating the steady - state probability distribution of the model to products of matrices which can be mapped onto the @xmath0-deformed harmonic oscillator ladder operators . in this section
we review the matrix approach to finding the steady state of the model .
the approach has been the subject of lectures at a previous summer school in this series @xcite and the reader is referred there for further details . here
the bare essentials of the method are outlined .
consider first a configuration of particles @xmath332 and its steady - state probability @xmath333 .
we use as an ansatz for @xmath333 an ordered product of matrices @xmath334 where @xmath335 if site @xmath25 is occupied and @xmath336 if it is empty . to obtain a probability ( a scalar value ) from this matrix product ,
we employ two vectors @xmath337 and @xmath338 in the following way : @xmath339 note that here the matrices , bra and ket vectors are in an auxiliary space and are not related to the space of configurations and the probability ket vector we considered earlier .
the factor @xmath340 is included to ensure that @xmath333 is properly normalised .
this latter quantity , analogous to a partition function as was discussed in section [ sec : theory ] , has the following simple matrix expression through which a new matrix @xmath341 is defined : @xmath342 note that if @xmath343 and @xmath344 do not commute @xmath333 is a function of both the number and position of particles on the lattice , as expected for a non - trivial steady state .
the algebraic properties of the matrices can be deduced from the master equation for the process @xcite .
it can be shown that sufficient conditions for equation ( [ eqn : pansatz ] ) to hold are @xmath345 in this formulation steady state correlation functions are easily expressed .
for example , the mean occupation number ( density ) of site @xmath25 may be written as @xmath346 to get a feel for why the conditions ( [ decom][dv ] ) give the correct steady state consider the current @xmath47 of particles between sites @xmath25 and @xmath347 .
@xmath348 in the matrix product formulation the current becomes @xmath349 now using ( [ decom ] ) and the fact that @xmath350 we find @xmath351 we see that , as is required in the steady state , the current is independent of the bond chosen . to see that the current into site one reduces to the same expression we use ( [ ew ] )
@xmath352 also the current out of site @xmath11 reduces to the same expression using ( [ dv ] ) @xmath353 note that the fact that all the above expressions for the current are equivalent is a necessary condition for the matrix formulation to be correct .
a sufficient condition is to check that the master equation ( [ meqn : solveme ] ) is satisfied for all @xmath147 configurations .
the algebra ( [ decom][dv ] ) allows one to do this systematically @xcite .
our task now is to evaluate the matrix products in the above expressions for @xmath340 , @xmath47 and @xmath35 by applying the rules ( [ decom][dv ] ) . in @xcite the case @xmath17
was treated by using ( [ decom ] ) repeatedly to ` normal - order ' matrix products : that is , to obtain an equivalent sum of products in which all @xmath344 matrices appear to the left of any @xmath343 matrices .
for example , consider powers of @xmath354 @xmath355 then finding a scalar value would be straightforward using ( [ ew ] ) and ( [ dv ] ) : @xmath356 the difficulty with this approach lies in the combinatorial problem of finding the coefficients @xmath357 .
this was solved for @xmath17 in @xcite . however , the solution ( and the problem of actually calculating the current and correlation functions ) for arbitrary @xmath0 remained open for some time although the phase diagram was conjectured @xcite . recently the problem has been overcome by making the connection with the @xmath0-hermite polynomials discussed in section [ sec : qdhp ] @xcite . to make the connection with the @xmath0-deformed harmonic oscillator ,
let us define @xmath358 one finds that the algebraic rules ( [ decom][dv ] ) reduce to @xmath359 thus @xmath343 and @xmath344 are related to the @xmath0-bosonic operators discussed in section [ sec : qdhp ] ; such a relationship was first pointed out in @xcite . also we see that @xmath337 and @xmath338 are eigenvectors of the @xmath0-bosonic operators .
such eigenvectors are known as coherent states ( see exercise 3 ) . in the oscillator s `` energy '' eigenbasis @xmath360 ( [ energybasis ] ) one can check that they have components @xmath361 for the moment we consider @xmath362 , @xmath363 so that the vectors ( [ eqn : vwofn ] ) are normalisable .
we introduced earlier a matrix @xmath354 which appears in the expressions for the mean particle density and current .
we now see that this matrix can be written as a linear combination of the identity @xmath364 and the `` position '' operator @xmath270 : @xmath365 as we saw in section [ sec : qdhp ] the eigenstates of the oscillator in the co - ordinate representation are continuous @xmath0-hermite polynomials
. clearly , the eigenvectors of @xmath354 are the same as those for @xmath366 and therefore knowledge of them permits diagonalisation of @xmath354 .
let us illustrate how to apply the procedure to obtain an expression for the normalisation @xmath340 when @xmath281 .
first we insert a complete set of states into the expression for the normalisation ( [ eqn : zdef ] ) : @xmath367 by design , the matrix @xmath354 is acting on its eigenvectors , so using ( [ eqn : cofx ] ) and @xmath368 we obtain @xmath369 now we know the form of @xmath337 and @xmath338 in the @xmath370 basis ( [ eqn : vwofn ] ) .
therefore we insert a complete set of the basis vectors in ( [ eqn : zint ] ) to find @xmath371 observe that the final sum in this equation is nothing but ( [ eqn : gdef ] ) , the generating function of the @xmath0-hermite polynomials ! thus ,
when @xmath372 and @xmath373 , we may write @xmath374 where @xmath375 is given by expression ( [ eqn : gnice ] ) . putting all this together , we arrive at an exact integral form for the normalisation @xmath376^n g(\theta , w ) g(\theta , v)\ ] ] which , written out more fully and using the notation ( [ qfacgen ] ) , reads @xmath377^n \frac{(e^{2i\theta},e^{-2i\theta};q)_\infty } { ( ve^{i\theta},ve^{-i\theta},we^{i\theta},we^{-i\theta};q)_\infty } \;.\ ] ] when @xmath378 or @xmath379 equation ( [ eqn : zint : q<1:short ] ) is not well - defined because @xmath375 does not converge when @xmath380 .
rather than finding a representation of the quadratic algebra ( [ decom][dv ] ) that does not suffer from this problem , one can simply analytically continue the integral ( [ eqn : zint : q<1:long ] ) to obtain @xmath340 when @xmath381 or @xmath382 takes on a value greater than one .
since a contour integral is defined by the residues it contains , to analytically continue the integral one simply has to follow the poles of the integrand as they move in and out of the original integration contour ( see exercise 4 ) .
one can also apply the procedure of the previous section to find an integral representation of the normalisation for the case of @xmath282 .
the only difference is that we must use the @xmath282 polynomials .
it turns out that this amounts to substituting @xmath312 by @xmath313 with the limits on @xmath314 now running from @xmath315 to @xmath383 and using the weight function ( [ eqn : nu : q>1 ] ) @xmath384^n g(u , w ) g(u , v ) \;.\ ] ] it is possible to derive an alternative expression for @xmath340 which takes the form of a finite sum rather than an integral and is valid for _ all _ values of the model parameters .
such an expression implies the solution of the combinatorial problem of reordering operators @xmath385 under the rule ( [ decom ] ) .
in @xcite the formula ( eq .
55 in that paper ) was derived by explicitly evaluating the integral ( [ eqn : zint : q<1:short ] ) using further properties of @xmath0-hermite polynomials and their generating functions . the phase diagram for the partially asymmetric exclusion process
is now obtained by calculating the current through the relation @xmath386 and the above exact expressions for the normalisation @xmath340 .
.[table : forward]the @xmath61 forms of the particle current in the forward biased phases . [ cols="^,^,^ " , ] for the case of _ forward bias _ ( @xmath281 ) in which particle hops are biased from the left boundary ( where particles are inserted ) to the right boundary ( where they are removed ) one expects a nonvanishing current in the thermodynamic limit @xmath387 .
this is calculated by applying the saddle - point method to the integral ( [ eqn : zint : q<1:long])see exercise 4 .
it turns out that for large @xmath11 @xmath388 then in the thermodynamic limit @xmath389 .
phase transitions arise from different asymptotic forms of the current due to the analytic continuation ( see exercise 4 ) for the integral ( [ eqn : zint : q<1:long ] ) .
ultimately one finds the expressions presented in table [ table : forward ] and thence the phase diagram shown in figure [ fig : paseppd ] .
note that the current as given by ( [ jz ] ) is a ratio of two partition functions of size @xmath390 and @xmath11 . in equilibrium
statistical mechanics ( in the large @xmath11 limit ) this ratio is equivalent to the fugacity @xmath391 ( within the grand canonical ensemble ) . recalling that @xmath392 and the chemical potential @xmath319 is equivalent to the gibbs free energy per particle , gives credence to our claim in section 2.1.4 that the current @xmath47 acts as a free energy for the system .
the phase diagram for the current in the forward - bias regime of the pasep . ]
we note that this phase diagram has a very similar form to that of figure [ fig : taseppd ] which applied for the special case of total asymmetry @xmath17 .
that is , one finds the same three phases , namely a ( i ) high density , ( ii ) low density and ( iii ) maximal current phase . as discussed in section [ sec : asepintro ] the transition between the low and high density phases ( i ) and ( ii ) is first order and those to the maximal current phase ( iii ) are second order .
note from table [ table : forward ] that , for example , the first derivative of the current is discontinuous at the transition between the low and high density phases .
similar methods can be used to find the density profiles in the pasep : the details of the calculations and results are presented in @xcite .
it turns out that the low- and high - density phases subdivide into three subphases for nonzero @xmath0 less than one @xcite . in each of the low density subphases ,
the bulk density is @xmath393 ; however what distinguishes the subphases is that the decay of the density profile from the right boundary takes a different form in each .
the same is true of high - density phase , except that @xmath394 and that the density decay is at the left boundary .
evaluation of the current in the pasep with @xmath281 show that the integral representation of the normalisation for @xmath281 , equation ( [ eqn : zint : q<1:long ] ) , can be recast as a contour integral @xmath395 where @xmath354 is the unit circle centred on the origin , @xmath396 the large number of singularities in the integrand at the origin makes this a difficult integral to evaluate exactly using , _
e.g. _ , the residue theorem .
instead , we shall use this contour integral to obtain the currents in the pasep for large system size .
first , we will apply the saddle - point method to the above integral which is valid for the range @xmath397 .
the idea is to deform the contour of integration such that it passes through a saddle point in @xmath398 $ ] along the path of steepest descent ( this is also the line where @xmath399 = \mbox{const}$ ] ) .
show that this saddle point is at @xmath400 and that the path of steepest descent ( at least near @xmath401 ) is parallel to the imaginary axis . also , show that @xmath402 .
insert taylor expansions of @xmath403 and @xmath404 ( to second order ) about the saddle point @xmath401 into ( [ ex : oint ] ) to obtain @xmath405 you should now convince yourself that the deformation of the integration contour to pass through @xmath401 along the path of steepest descent in @xmath398 $ ] is equivalent to making the substitution @xmath406 with @xmath76 running from @xmath315 to @xmath383 .
hence show that the resulting gaussian integral will give rise to an expression of the form @xmath407 note : to save time , do not evaluate the constant @xmath408 ( although you may like to check that it is real and positive ) .
now show that the current in the pasep for @xmath372 , @xmath373 and large @xmath11 behaves as @xmath409 there now remains the question of what to do when one of the control parameters , say @xmath56 , has a magnitude greater than @xmath19 .
when @xmath372 , the contour @xmath354 encloses the poles of @xmath404 at @xmath410 but excludes those at @xmath411 . as @xmath56 is increased , the pole at @xmath412 moves outside @xmath354 and that at @xmath413 moves inside
. to obtain the analytic continuation of @xmath340
, one must add to the saddle - point expression for the integral over @xmath354 above the residue at @xmath412 and subtract that at @xmath413 .
show that this yields @xmath414 again try to avoid explicitly calculating the positive constant @xmath415 .
verify that @xmath416 and thus that , for sufficiently large @xmath11 , @xmath417 .
hence show that for @xmath418 @xmath419 now convince yourself that as @xmath56 is increased yet further , the contribution from the additional poles that have entered / exited the unit circle is smaller than that from those at @xmath420 .
consider also the effect of increasing @xmath421 to a value less than @xmath56 and show that when @xmath422 @xmath423 finally , use the information learned from this analysis to construct the phase diagram ( figure 5 ) for the current in the @xmath20-@xmath21 plane
. for @xmath282 one has a reverse bias regime @xcite where the boundary conditions force a current through the system against the direction of the reverse bias .
this gives rise to a new phase in which the current decreases exponentially with the length of the system as @xmath424 .
this phase is of interest in the context of ` backbend dynamics ' @xcite where , for example , a fluid in a permeable medium has to traverse a pore oriented against the direction of gravity .
one can predict the form of the density profile in the reverse bias regime as follows .
we expect a high - density region at the left edge of the lattice , towards which the particles are biased but can not escape , and a low - density region at the right edge of the lattice .
thus the density profile is a fermi - like function as illustrated in figure [ fig : sigmoid ] . in order
that the profile is stationary the typical time for a particle to diffuse ( against the bias ) from the edge of the high - density region to the right boundary must be equal to the typical time of a hole to diffuse from the edge of the low - density region to the left boundary .
since the diffusion rates of both particles and holes are the same , the boundary between the high- and low - density regions must be in the centre of the lattice , as illustrated in figure [ fig : sigmoid ] . proposed form of the density profile in the reverse - bias regime of the pasep . ] here we give a brief overview of other related exclusion processes .
firstly the features of the phase diagram for @xmath281 ( figure 5 ) appear robust for stochastic one - dimensional driven systems and can be predicted from considerations of domain wall dynamics @xcite .
this has been confirmed through the exact solution of the open boundary system with fully parallel dynamics @xcite .
however when the bulk dynamics become deterministic the maximal current phase is lost . for the case of symmetric exclusion @xmath261 ( no bulk drive ) one does not have phase transitions ; the current is easy to calculate and decays linearly with system size @xcite .
intuitively one can understand this as diffusion between two particle reservoirs of different density
. the diffusion current will be proportional to the concentration gradient which decreases as @xmath425 .
thus although the bulk dynamics is symmetric the boundary conditions force a weak current of particle through the system . for @xmath261 a recent work
@xcite has determined the probability of observing a particular coarse - grained density profile @xmath426 . in an equilibrium system , this probability is given by @xmath427)$ ] in which @xmath428 $ ] is a local functional that expresses the free energy difference between the equilibrium profile and @xmath64 . from the matrix product approach a form for a functional that plays a similar role in the symmetric exclusion process
was found in @xcite and is shown to be _
nonlocal_. for the particular boundary conditions where the dynamics satisfy detailed balance , @xmath428 $ ] reduces to the free energy functional known from equilibrium statistical physics and so the nonlocal functional would seem to generalise a well - known equilibrium concept .
an interesting kind of boundary - induced phase transition , manifesting spontaneous symmetry breaking , is found when the asymmetric exclusion process is generalised to two oppositely moving species of particle : one species is injected at the left , moves rightwards and exits at the right ; the other species is injected at the right , moves leftwards and exits at the left @xcite .
intuitively one can picture the system as a narrow road bridge : cars moving in opposite directions can pass each other but can not occupy the same space .
the model has a left - right symmetry when the injection rates and exit rates for the two species of particles are symmetric .
however for low exit rates ( @xmath21 ) this symmetry is broken and the lattice is dominated by one of the species at any given time .
this implies that the short time averages of currents and bulk densities of the two species of particles are no longer equal . over longer times
the system flips between the two symmetry - related states . in the @xmath429
limit the mean flip time between the two states has been calculated analytically and shown to diverge exponentially with system size @xcite .
thus the ` bridge ' model provided a first example of spontaneous symmetry breaking in a one dimensional system . in the models discussed so far the open boundaries
can be thought of as inhomogeneities where the order parameter ( particle density ) is not conserved .
inhomogeneities which conserve the order parameter can be considered on a periodic system .
indeed a single defect bond on the lattice ( through which particles hop more slowly ) is sufficient to cause the system to separate into two macroscopic regions of different densities @xcite : a high density region which can be thought of as a traffic jam behind the defect and a low density region in front of the defect . here
the presence of the drive appears necessary for the defect to induce the phase separation .
moving defects ( _ i.e. _ particles with dynamics different from that of the others ) have also been considered and exact solutions obtained @xcite . for the simple case of a slowly moving particle the phenomenon of a queue of particles forming behind it has been shown to be analogous to bose condensation @xcite .
in section [ sec : sbacintro ] we introduced two reaction systems , namely annihilation ( @xmath430 ) and coalescence ( @xmath431 ) .
as noted in section [ sec : sbacintro ] , it is well known that annihilation and coalescence reactions are equivalent when the reactant motion is diffusive @xcite and hence lead to the same @xmath432 density decay in one dimension .
we now consider in detail the distinct case of ballistic reactant motion by which it is meant that particles move with some fixed velocity .
we study a class of models that comprise an arbitrary combination of annihilation and coalescence of particles with two ( conserved ) velocities and stochastic reaction dynamics and which was solved in @xcite . as we shall see , an exact solution is possible by virtue of a matrix product method in which the matrices involved can be written in terms of the raising and lowering operators of the @xmath0-deformed harmonic oscillator .
a related , but distinct , ballistic reaction model ( not contained within the class discussed here ) had also revealed a connection between ballistic reaction systems and @xmath0-deformed operator algebras @xcite .
we now define the class of models to be considered . at time
@xmath433 reactants are placed randomly on a line .
each particle is assigned a velocity @xmath434 ( right - moving ) or @xmath435 ( left - moving ) with probability @xmath436 and @xmath437 respectively .
particles move ballistically until two collide , at which point one of four outcomes follows , see figure [ fig : outcomes ] : the particles pass through each other with probability @xmath0 ; the particles coalesce into a left ( right ) moving particle with probability @xmath438 ( @xmath439 ) ; the particles annihilate with probability @xmath440 . here
@xmath441 is the probability that some reaction occurs .
as an example of an application of this model consider the identification of right- and left - moving particles with the edges of terraces as illustrated in figure [ fig : rsos ] .
if new particles are added to the system in such a way that they only stick to the sides , and the rate of particle addition is taken to infinity , one obtains ballistic motion of terrace edges .
when two edges meet , a terrace is completed which corresponds to annihilation of approaching particles .
hence the relevant parameters of the general annihilation coalescence model are @xmath442 . the scattering reaction , which occurs with rate @xmath0 , corresponds to the possibility of a new terrace being formed when two edges meet .
the case of deterministic annihilation @xmath443 was studied by elskens and frisch @xcite . here
particles always annihilate on contact .
this case was found to exhibit a density decay that depends on time as @xmath432 , but only if the initial densities of the two particle species ( left- and right - moving ) are the same . in the following analysis of the more general class of reaction systems
, we find that such a result persists if two ` effective ' initial densities ( to be defined below ) are equal .
furthermore , we will see that the introduction of stochasticity into the reaction dynamics _
i.e. _ the parameter @xmath0 , induces a transition in the density decay form .
our aim is to calculate the density decay .
to do this we shall consider without loss of generality a right moving test particle as illustrated in figure [ fig : traj ] .
we define @xmath444 as the probability that the test particle survives up to a time @xmath36 respectively . from figure
[ fig : traj ] one can see that the initial spacing of the particles on the line does not affect the _ sequence _ of possible reactions for any given particle , in particular for the test particle ( we return to this point later ) .
also note that after a given time @xmath36 , the test particle may only have interacted with the @xmath11 particles initially placed within a distance @xmath445 ( and to the right ) of the chosen particle .
these two facts imply that the survival probability can be expressed in terms of two independent functions .
the first is @xmath446 , the probability that initially there were exactly @xmath11 particles in a region of size @xmath445 .
the second is @xmath447 , the probability that the test particle survives reactions with the @xmath11 particles initially to its right , and depends only on the sequence of the @xmath11 particles .
explicitly , @xmath448 thus the problem is reduced to two separate combinatorial problems of calculating @xmath446 and @xmath447 .
note that the choice of @xmath446 allows one to consider a model defined on a lattice or on the real line . in the present work we assume particles
are initially placed on a line with nearest - neighbour distances chosen independently from an exponential distribution with unit mean .
then we can use a well - known result @xcite @xmath449 we now show that the second problem ( calculation of @xmath447 ) can be formulated within a matrix product approach . as
an example consider a test particle encountering the string of reactants depicted in figure [ fig : traj ] .
we claim that the probability of the test particle surviving through this string may be written as @xmath450 where @xmath451 are matrices and @xmath452 , @xmath453 vectors with scalar product @xmath454 .
thus we write , in order , a matrix @xmath455 ( @xmath456 ) for each right ( left ) moving particle in the initial string .
the conditions for an expression such as ( [ eqn : fex ] ) to hold for an arbitrary string are , in fact , rather intuitive @xmath457 ) \\ \label{eqn : wdef } \langle w| l & = & \langle w| ( q+p\eta_r ) \\
r |v\rangle & = & |v\rangle\;. \label{eqn : vdef}\end{aligned}\ ] ] the condition ( [ eqn : rl ] ) just echoes the reactions that occur in figure [ fig : outcomes ] _ i.e. _ after an interaction between a right - moving and left - moving particle there are four possible outcomes ( see figure [ fig : outcomes ] ) corresponding to the four terms on the right hand side of ( [ eqn : rl ] ) with probabilities given by the respective coefficients . using ( [ eqn : rl ] ) , any initial matrix product such as ( [ eqn : fex ] ) can be reduced to a sum of terms of the form @xmath458 corresponding to all possible final states ensuing from the initial string and with coefficients equal to the probabilities of each final state .
these final states give the possible sequences of particles that the right - moving test particle will encounter .
the test particle will survive such a final state and pass through the @xmath459 left - moving particles with probability @xmath460 . the condition ( [ eqn : vdef ] )
ensures that this probability is obtained for each possible final state _
i.e._@xmath461 finally the condition ( [ eqn : wdef ] ) , along with @xmath462 , ensures that once a right - moving particle passes through all the left - moving particles in the string it no longer plays a role .
thus ( [ left ] ) becomes @xmath463 note that the reason for a matrix product formulation is different from the pasep .
there the steady state probability of any configuration , for arbitrary system size , could be written as a matrix product . here
the probability of a particle surviving some arbitrary sequence of particles can be written as a matrix product .
the above approach relies on an important property of the system which is invariance of a reaction sequence with respect to changes of initial particle spacings . to understand this , consider again figure [ fig : traj ] . by altering the initial spacings of the particles , the absolute times at which trajectories intersect and reactions may occur ( if the reactants have survived ) may be altered .
for example , by increasing the spacing between the fifth and sixth particles , the trajectories of the third and fourth particles can be made to intersect first .
however as we have already seen , for any particle , the _ order _ of intersections it encounters does not change and so the final states and probabilities are invariant .
this invariance is manifested in the matrix product by the fact that the order in which we use the reduction rule ( [ eqn : rl ] ) is unimportant _
i.e. _ matrix multiplication is associative .
thus , it is the invariance with respect to initial spacings that allows the system to be solved by using a product of matrices .
we now proceed to evaluate @xmath447 . averaging ( [ eqn : fex ] ) over all initial strings of length @xmath11 , recalling that @xmath436 and @xmath464 are the probabilities that a particle is assigned velocity @xmath465 respectively , yields @xmath466 to evaluate @xmath447 we first write @xmath467 in these equations
we have introduced two new parameters @xmath468 which we call effective initial particle densities and whose ratio @xmath469 turns out to be an important parameter of the model
. one can verify from ( [ eqn : rl][eqn : vdef ] ) that @xmath470 satisfy a @xmath0-deformed harmonic oscillator algebra @xmath471 thus ( [ eqn : f ] ) can be written as @xmath472^n | v \rangle\;. \label{eqn : fmat}\ ] ] the vectors @xmath337 and @xmath338 are eigenvectors of @xmath473 , and as in section [ sec : pasep ] , they can be explicitly calculated . at this point
one can see that we have precisely the same mathematical structure as we did when solving for the steady state of the pasep compare ( [ eqn : ho][eqn : coherent ] ) with ( [ pasepqho1][pasepqho3 ] ) . we just need to carry out the same procedure of diagonalising a matrix that is linear combination of the identity operator and the position operator of a @xmath0-deformed harmonic oscillator . rather than repeat the details here we go directly to a discussion of the phase diagram .
deterministic ballistic annihilation and random walks the elskens - frisch model @xcite of deterministic ballistic annihilation is a special case of the stochastic model described above that has the parameters @xmath474 . for the case
where particles are initially placed on all sites of a lattice with equal probability of right- and left - moving particles , write down all possible initial conditions that would allow a right - moving test particle to survive through one , two and three sites .
hence relate the survival probability of the test particle after @xmath11 sites to the probability of a random walker not returning to the origin after @xmath18 steps .
hint : devise a criterion for the survival of the test particle that involves the relative number of left- and right - moving particles for a given initial condition .
construct the matrices @xmath455 and @xmath456 and the vectors @xmath337 and @xmath338 and explain how the resulting expression for @xmath447 ( with @xmath475 ) is related to the random walk problem .
we consider the long - time density decay @xmath476 which is obtained from the survival probabilities @xmath444 , @xmath477 for right and left moving test particles through @xmath478 ( @xmath477 can be deduced from @xmath444 through the left - right symmetry of the model . ) the matrix product calculation for @xmath447 implies that the density decays as @xmath479 in this equation @xmath480 is the fraction of particles remaining once no more reactions are possible ( _ i.e. _ all particles moving in the same direction ) .
we now list the results the residual density @xmath481 when @xmath482 .
( results for @xmath483 one be deduced using the left - right symmetry of the model . ) for @xmath484 @xmath485\ ] ] for @xmath486 @xmath487 } { ( ct)^{1/2}}\ ] ] for @xmath488 @xmath489}{(ct)^{3/2}}$$\nonumber\end{gathered}\ ] ] for @xmath490 @xmath491 we now consider the phase diagram figure [ fig : sbacpd ] .
this is spanned by the values of @xmath492 ( the ratio of effective densities @xmath493 ) and @xmath0 ( the stochasticity parameter ) : @xmath492 parameterises all information concerning the asymmetry between the right and left moving particles whereas @xmath0 parameterises the level of stochasticity .
thus there is universality of ballistic annihilation and coalescence : for a generic choice of @xmath494 , @xmath495 defining a particular annihilation - coalescence model , the same four decay regimes are found by varying the initial densities or stochasticity parameter @xmath0 . in this way the universality can be considered as a ` law of corresponding states ' .
the line @xmath496 ( equal effective densities ) is non - generic since a single , power law , decay regime is found .
the decay does not depend on the stochasticity @xmath0 .
this special line was found by elskens and frisch @xcite for the case of deterministic annihilation ( @xmath497 ) and we have shown that this special line is present for all combinations of annihilation and coalescence .
this phase can be understood through the picture of @xcite .
density fluctuations in the initial conditions lead to trains of left- and right - moving particles : in a length @xmath498 the excess particle number is @xmath499 which yields the @xmath432 density decay . at long times ,
the train size is large and so a particle in one train encounters many particles in the other and will eventually react making the parameter @xmath0 irrelevant .
the second phase found in @xcite ( labelled the elskens - frisch phase in the diagram ) is seen to persist for nonzero @xmath0 and in this phase the two particle species decay at equal rates .
the two new phases that arise in the full model for @xmath500 ( _ i.e. _ as a consequence of randomness in the reaction dynamics ) have the contrasting property that two particle species decay at _ unequal _ rates leaving a non - zero population of left - moving particles .
a simple example of non - equal decays is the case @xmath501 ( @xmath502 )
. then left - moving particles do not decay but simply absorb the right - moving particles with probability @xmath503 giving @xmath504 .
our results show that , in general , increasing @xmath0 leads to a non - trivial transition at @xmath505 to a regime where the two species have different decay forms .
in these lectures we have tried to give a flavour of the collective phenomena exhibited by simple low - dimensional systems with stochastic dynamics .
specifically we have seen that the partially asymmetric exclusion process exhibits a number of phase transitions ( of both first and second order ) and long - range correlations even in one dimension .
we have also shown how similar mathematical techniques can be applied in the distinct context of a particle reaction system to obtain an exact solution .
we hope that the reader with a background in equilibrium statistical physics is pleasantly surprised by the diversity of phenomena that emerge in low - dimensional nonequilibrium systems .
we have also endeavoured to illustrate how novel analytical techniques are being developed for these systems . through the inclusion of background material and the exercises , we hope also to have inspired confidence in the reader to approach the literature and make an active contribution to this fertile field . *
acknowledgments * : we would like to thank our collaborators f. colaiori , f. essler and y. kafri with whom the work of section 5 and section 6 was carried out and also m. j. e. richardson who contributed to the early development of that work
. g. m. schtz _ exactly solvable models for many - body systems far from equilibrium _ in _ phase transitions and critical phenomena _ vol .
* 19 * , c. domb and j. l. lebowitz ( eds . ) , ( academic press , london , 2000 ) . | in these lectures we give an overview of nonequilibrium stochastic systems .
in particular we discuss in detail two models , the asymmetric exclusion process and a ballistic reaction model , that illustrate many general features of nonequilibrium dynamics : for example coarsening dynamics and nonequilibrium phase transitions . as a secondary theme
we shall show how a common mathematical structure , the @xmath0-deformed harmonic oscillator algebra , serves to furnish exact results for both systems .
thus the lectures also serve as a gentle introduction to things @xmath0-deformed . and nonequilibrium dynamics , stochastic processes , phase transition , asymmetric exclusion process , reaction kinetics 02.50.-r , 05.40.-a , 05.70.fh |
as pointed out in a recent paper by irvine and bouwmeester @xcite , electromagnetic knots are exact solutions of the classical maxwell equations of electromagnetism in vacuum .
they appeared firstly in a paper by raada @xcite in 1989 .
raada himself @xcite has used these solutions as the basic elements of a topological model of electromagnetism which is locally equivalent to maxwell s standard theory but implies furthermore some topological quantization conditions with interesting physical meaning @xcite .
electromagnetic knots are defined through two fundamental complex scalar fields @xmath0 whose level curves coincide with the magnetic and electric lines respectively , each one of these lines being labelled by the constant value of the corresponding scalar .
both scalars are assumed to have only one value at infinity , which is equivalent to compactify the physical three - space to the sphere @xmath1 .
moreover , the complex plane is compactified to the sphere @xmath2 via stereographic projection . as a result of such compactifications ,
the scalars @xmath3 and @xmath4 can be interpreted , at any time , as maps @xmath5 , which can be classified in homotopy classes , characterized by the value of the hopf index @xmath6 @xcite .
it can be shown that the two scalars have the same hopf index and that the magnetic and the electric lines are generically linked with the same gauss linking number @xmath7 .
if @xmath8 is the multiplicity of the level curves ( i.e. the number of different magnetic or electric lines that have the same label @xmath3 or @xmath4 ) , then the hopf index of both scalars is @xmath9 .
the hopf index can thus be interpreted as a generalized linking number if we define a line as a level curve with @xmath8 disjoint components .
an important feature of the model is that the faraday 2-form @xmath10 and its dual @xmath11 are proportional to the two pull - backs of @xmath12 , the area 2-form in @xmath2 , by @xmath3 and @xmath4 , @xmath13 where @xmath14 is a constant introduced so that the magnetic and electric fields have correct dimensions and @xmath15 is the velocity of light in vacuum . in the international system of units , @xmath14 can be expressed as a pure number times the planck constant @xmath16 times the light velocity @xmath15 times the vacuum permeability @xmath17 . as a consequence of the definitions ( [ pullback ] ) ,
the maps @xmath3 and @xmath4 are dual to one another , @xmath18 , where * is the hodge or duality operator
. this duality condition guarantees that both @xmath19 and @xmath20 obey the maxwell equations in empty space without the need of any other requirement .
the electromagnetic fields obtained in this way are called electromagnetic knots .
they are radiation fields as they verify the condition @xmath21 .
it can be proved ( see @xcite for the details ) that any radiation field in vacuum is locally equivalent to an electromagnetic knot . moreover , because of the darboux theorem , any electromagnetic field in empty space can be expressed locally as the sum of two radiation fields .
consequently , a model of electromagnetism based on these electromagnetic knots is locally equivalent to maxwell standard theory .
however , its difference from the global point of view has interesting consequences , as are the following topological quantizations : ( i ) the electric charge of any point particle must necessarily be equal to an integer multiple of the fundamental value @xmath22 ( see @xcite ) .
( ii ) the electromagnetic helicity @xmath23 is also quantized @xcite , where @xmath24 and @xmath25 are the classical expressions of the number of right- and left - handed photons contained in the field ( i.e. @xmath26 , @xmath27 being fourier transforms of the vector potential @xmath28 in the classical theory , but creation and annihilation operator in the quantum version ) .
in fact , for any electromagnetic knot , @xmath29 , which is a remarkable relation between the hopf index ( i.e. the generalized linking number ) of the classical field and the classical case of the difference @xmath30 .
( iii ) the topology of the model implies also the quantization of the energy of the electromagnetic field in a cavity @xcite .
( iv ) electromagnetic knots are compatible with the quantization of the magnetic flux of a superconducting ring @xcite , which in standard theory is always an integer multiple of @xmath31 , @xmath32 being the value of the magnetic monopole . in this work
we study a classical charged particle in a knotted electromagnetic field .
we find that the particles can accelerate to light velocity .
we first revise the construction and some physical propeties of the electromagnetic field built from the hopf map between the compactified three - space ( the sphere @xmath1 ) and the compactified complex plane ( the sphere @xmath2 ) .
in particular , we pay attention to the electromagnetic energy density and how it evolves with time and introduce a mean quadratic radius of the energy density .
then we consider the relativistic motion of electrons in this electromagnetic field .
we study the trajectories of the electrons and their velocities , showing that they become ultrarelativistic for a wide range of the electromagnetic energy of the knot .
finally we give some conclusions and prospects of future work .
a method to find explicitly some electromagnetic knots can be found in @xcite .
let @xmath33 , @xmath34 , two complex scalar fields such that can be considered as maps @xmath35 after identifying the physical space @xmath36 with @xmath1 and the complex plane with @xmath2 .
they have to satisfy the following two conditions : \1 .
the level curves of @xmath37 must be orthogonal , in each point , to the level curves of @xmath38 , since we know that electromagnetic knots are radiation fields ( @xmath39 ) . \2 .
the hopf index of @xmath37 and of @xmath38 are equal , @xmath40 .
this is necessary to ensure that the condition @xmath21 is maintained during time evolution .
given @xmath37 and @xmath38 with this two conditions , we can build the magnetic and electric fields at @xmath41 as @xmath42 it is convenient to work with dimensionless coordinates in the mathematical spacetime @xmath43 , and in @xmath2 . in order to do that
, we define the dimensionless coordinates @xmath44 , related to the physical ones @xmath45 ( in the si of units that we will use in this work ) by @xmath46 and @xmath47 , where @xmath48 is a constant with length dimensions .
now , let us consider the hopf map , @xmath49 whose fibres are been used as a basis for a case of knotted entanglement in reaction - diffusion models , in particular for a fitzhugh - nagumo model @xcite .
we also consider the map corresponding to the change @xmath50 in ( [ 11.19 ] ) , @xmath51 because of their construction , it is obvious that both maps ( [ 11.19 ] ) and ( [ 11.20 ] ) have the same hopf index .
in fact , these maps have hopf index @xmath52 and their fibrations are mutually orthogonal at each point .
consequently , we can build an electromagnetic knot from these maps .
the cauchy data for the magnetic and electric fields are @xmath53 from ( [ 11.27 ] ) , two vector potentials @xmath54 and @xmath55 can be computed , such that @xmath56 , @xmath57 , with the results @xmath58 the magnetic and electric helicities of this knot are defined to be @xmath59 where @xmath60 is the vacuum permittivity . taking into account the cauchy data ( [ 11.27 ] ) and the potentials ( [ 11.28 ] ) , the electromagnetic helicity yields @xmath61 to find the electromagnetic knot at any time from the cauchy data ( [ 11.27 ] ) , we use fourier analysis .
the fields turn out to be @xcite , @xmath62 where the quantities @xmath63 , @xmath64 , @xmath65 are defined by @xmath66 and the vectors @xmath67 and @xmath68 are @xmath69 this solution fulfills now @xmath70 and @xmath71 at any time .
it is possible to obtain directly the electromagnetic field ( [ 11.35 ] ) from the time - dependent expressions ( [ pullback ] ) . in terms of the magnetic and the electric fields
, we have @xmath72 where the time - dependent expressions of the maps @xmath3 and @xmath4 are ( see @xcite ) @xmath73 note that the level curves of both maps @xmath3 and @xmath4 remain linked with a gauss linking number equal to 1 .
the evolution of the curved lines of these maps gives the evolution of the force lines of the magnetic and electric fields given by the expressions ( [ timeeb ] ) .
the energy density of the electromagnetic field ( [ 11.35 ] ) is given by , @xmath74 the maximum of the energy density is located at @xmath75 during time evolution .
the function @xmath76 is symmetric in the coordinates @xmath77 and @xmath78 . in figure
[ fig1 ] , we show some isosurfaces of the energy density @xmath76 for times @xmath79 . the energy density levels represented
are @xmath80 , @xmath81 and @xmath82 in @xmath83 units .
it can be seen how the isosurfaces spread as the energy density goes to zero . for time
@xmath84 , the @xmath82 and @xmath81 levels have dissapeared .
note that the total electromagnetic energy of the knot @xmath85 remains constant . at @xmath86 ,
the energy density has spherical symmetry and its maximum is located at the origin . as time increases , the symmetry is broken and the maximum is located at @xmath75 and @xmath87 close to @xmath88 .
approximately , the position of the maximum of the energy density can be found up to @xmath89 at @xmath75 , @xmath90 . , @xmath91 , @xmath89 , @xmath84 .
coordinates @xmath92 are dimensionless and are related to the physical coordinates @xmath93 through @xmath94 .
the levels correspond to values @xmath80 , @xmath81 and @xmath82 of the density energy in @xmath95 units ( see eq .
( [ 11.38 ] ) ) .
the density energy levels displayed are @xmath80 , @xmath81 and @xmath82 in increasing order of colour intensity . at @xmath84 , the levels @xmath81 and @xmath82 are not present.,title="fig:",scaledwidth=45.0% ] , @xmath91 , @xmath89 , @xmath84
. coordinates @xmath92 are dimensionless and are related to the physical coordinates @xmath93 through @xmath94 .
the levels correspond to values @xmath80 , @xmath81 and @xmath82 of the density energy in @xmath95 units ( see eq .
( [ 11.38 ] ) ) .
the density energy levels displayed are @xmath80 , @xmath81 and @xmath82 in increasing order of colour intensity . at @xmath84 , the levels @xmath81 and @xmath82 are not present.,title="fig:",scaledwidth=45.0% ] , @xmath91 , @xmath89 , @xmath84 .
coordinates @xmath92 are dimensionless and are related to the physical coordinates @xmath93 through @xmath94 .
the levels correspond to values @xmath80 , @xmath81 and @xmath82 of the density energy in @xmath95 units ( see eq .
( [ 11.38 ] ) ) .
the density energy levels displayed are @xmath80 , @xmath81 and @xmath82 in increasing order of colour intensity . at @xmath84 ,
the levels @xmath81 and @xmath82 are not present.,title="fig:",scaledwidth=45.0% ] , @xmath91 , @xmath89 , @xmath84 .
coordinates @xmath92 are dimensionless and are related to the physical coordinates @xmath93 through @xmath94 .
the levels correspond to values @xmath80 , @xmath81 and @xmath82 of the density energy in @xmath95 units ( see eq .
( [ 11.38 ] ) ) .
the density energy levels displayed are @xmath80 , @xmath81 and @xmath82 in increasing order of colour intensity . at @xmath84 , the levels @xmath81 and @xmath82 are not present.,title="fig:",scaledwidth=45.0% ] the energy density of the knot extends to infinity , but we can define a mean quadratic radius of the energy distribution as @xmath96 where @xmath97 is the position of the maximum of the distribution . at @xmath86 ,
the distribution has spherical symmetry and the mean quadratic radius of the distribution is given by @xmath98 .
the maximum has a value of the ( dimensionless ) energy density equal to @xmath99 and at a distance equal to the mean quadratic value , the dimensionless energy density is @xmath100 .
this means that , initially , more than @xmath101 of the energy is locallized inside a sphere of radius @xmath48 centered at the origin .
as time evolves the mean quadratic radius of the distribution spreads out ( its value at @xmath102 or @xmath89 is about @xmath103 ) and the position of the center is at @xmath104 .
note that the distribution is not well characterized as a sphere as time evolves .
one can also compute the poynting vector @xmath105 of the electromagnetic field , obtaining @xmath106 as can be seen , it has a single contribution along the @xmath107-axis .
this explains why the maximum of the energy density moves along this axis for the electromagnetic knot studied in this paper .
now we apply the electromagnetic knot studied in the previous section to the following situation .
suppose that this knot has been created in certain region of the space , so we have at @xmath41 a knot initially centered at the origin , that moves with time along the @xmath107-axis as we discussed in the previous section .
let us assume there are free electrons with negligible initial velocities in units of the light velocity @xmath15 .
we consider the evolution of these electrons under the electromagnetic knot field .
the velocity of the electrons increases by the action of the electromagnetic field , so we will solve the relativistic equation for the motion of single electrons @xcite , considered as test particles which do not affect the value of the electromagnetic field obtained from the hopf fibration .
the equation to be considered is @xmath108 where @xmath109 c is the electron charge , and @xmath110 kg is its rest mass . using dimensionless quantities and the expressions ( [ 11.35 ] ) for the electromagnetic knot of the hopf fibration
, we find @xmath111 different possible physical situations can be studied with equation ( [ relmotion2 ] ) by changing the value of the dimensionless prefactor @xmath112 . for electrons , taking into account that the total energy of the field is @xmath113 and that the size of the knot is characterized initially by @xmath48 , the prefactor in equation ( [ relmotion2 ] ) can be written as @xmath114 where @xmath115 is measured in joules and @xmath48 in meters . .
from left to right and top to bottom , @xmath91 , @xmath89 , @xmath84 .
the energy levels shown correspond to values @xmath80 , @xmath81 and @xmath82 of the density energy in @xmath95 for the same times ( see figure [ fig1 ] ) .
the initial velocity of the electrons is @xmath116 .
the electron initial positions considered are given by @xmath117 , with @xmath118 , by @xmath119 , and by @xmath120 .
after @xmath84 these electrons moves almost freely .
the final values of the velocities at @xmath84 run from @xmath121 to @xmath122 in units of the light velocity @xmath15.,title="fig:",scaledwidth=45.0% ] . from left to right and top to bottom , @xmath91 , @xmath89 , @xmath84 .
the energy levels shown correspond to values @xmath80 , @xmath81 and @xmath82 of the density energy in @xmath95 for the same times ( see figure [ fig1 ] ) .
the initial velocity of the electrons is @xmath116 .
the electron initial positions considered are given by @xmath117 , with @xmath118 , by @xmath119 , and by @xmath120 .
after @xmath84 these electrons moves almost freely .
the final values of the velocities at @xmath84 run from @xmath121 to @xmath122 in units of the light velocity @xmath15.,title="fig:",scaledwidth=45.0% ] . from left to right and top to bottom , @xmath91 , @xmath89 , @xmath84 .
the energy levels shown correspond to values @xmath80 , @xmath81 and @xmath82 of the density energy in @xmath95 for the same times ( see figure [ fig1 ] ) . the initial velocity of the electrons is @xmath116 .
the electron initial positions considered are given by @xmath117 , with @xmath118 , by @xmath119 , and by @xmath120 .
after @xmath84 these electrons moves almost freely .
the final values of the velocities at @xmath84 run from @xmath121 to @xmath122 in units of the light velocity @xmath15.,title="fig:",scaledwidth=45.0% ] in figure [ fig2 ] , the evolution of the position of electrons can be seen when the prefactor is @xmath123 .
the initial velocity of the electrons is @xmath116 .
we have taken for these electrons initial positions along the @xmath124- , @xmath107- and @xmath125-axis . in each axis
, the dimensionless position is @xmath126 . in the figure ,
the electron trajectories are plotted from @xmath86 up to @xmath91 ( in the first pannel ) , up to @xmath89 ( in the second pannel ) , and up to @xmath84 ( in the third pannel ) .
we also include some energy density levels of the electromagnetic knot at the time considered , corresponding to values @xmath80 , @xmath81 and @xmath82 in @xmath95 units ( see the caption of figure [ fig1 ] for more details ) . by doing so , in figure [ fig2 ]
we see not only the evolution of the position of some free electrons , but also the region in which the interaction with the electromagnetic knot ( whose center moves along the @xmath107-axis with a velocity close to @xmath15 ) is more important .
this fact explains some features of the behaviour of the electron trajectories . at the beginning , up to @xmath91 , the region in which the electrons are is influenced by high values of the knot energy density , so that the electrons are accelerated and their trajectories are curved by the influence of the electromagnetic force . from @xmath91
up to @xmath89 , the center of the knot has moved along the @xmath107-axis and many electrons are now in a region in which the electromagnetic density is smaller
. however , the electromagnetic force is still important so that many electrons tend to follow the motion of the center of the knot ( a small fraction of the electrons seem to be curved towards the opposite direction ) .
this is also the situation from @xmath89 up to @xmath84 , during which electrons reach high values of the velocity , in the range from @xmath127 to @xmath128 in units of the light velocity @xmath15 . after this time , the influence of the electromagnetic knot on the electron trajectories is smaller , and they can be considered as almost free . with a value @xmath129 . from left to right and top to bottom , @xmath91 , @xmath89 , @xmath84 .
the energy levels shown correspond to values @xmath80 , @xmath81 and @xmath82 of the density energy in @xmath95 for the same times ( see figure [ fig1 ] ) .
the initial velocity of the electrons is @xmath116 .
the electron initial positions considered are given by @xmath117 , with @xmath118 , by @xmath119 , and by @xmath120 .
the final values of the velocities at @xmath84 run from @xmath130 to @xmath131 in units of the light velocity @xmath15.,title="fig:",scaledwidth=45.0% ] with a value @xmath129 . from left to right and top to bottom , @xmath91 , @xmath89 , @xmath84 .
the energy levels shown correspond to values @xmath80 , @xmath81 and @xmath82 of the density energy in @xmath95 for the same times ( see figure [ fig1 ] ) .
the initial velocity of the electrons is @xmath116 .
the electron initial positions considered are given by @xmath117 , with @xmath118 , by @xmath119 , and by @xmath120 .
the final values of the velocities at @xmath84 run from @xmath130 to @xmath131 in units of the light velocity @xmath15.,title="fig:",scaledwidth=45.0% ] with a value @xmath129 . from left to right and top to bottom , @xmath91 , @xmath89 , @xmath84 .
the energy levels shown correspond to values @xmath80 , @xmath81 and @xmath82 of the density energy in @xmath95 for the same times ( see figure [ fig1 ] ) .
the initial velocity of the electrons is @xmath116 .
the electron initial positions considered are given by @xmath117 , with @xmath118 , by @xmath119 , and by @xmath120 .
the final values of the velocities at @xmath84 run from @xmath130 to @xmath131 in units of the light velocity @xmath15.,title="fig:",scaledwidth=45.0% ] in figure [ fig3 ] , the evolution of the position of electrons is studied when the prefactor is @xmath129 .
the initial velocity of the electrons is @xmath116 , and the initial positions in each axis are given by @xmath126 in units of the knot size @xmath48 as before . as in figure
[ fig2 ] , the electron trajectories are plotted in figure [ fig3 ] from @xmath86 up to @xmath91 ( in the first pannel ) , up to @xmath89 ( in the second pannel ) , and up to @xmath84 ( in the third pannel ) , and we include some energy density levels of the electromagnetic knot at the time considered , corresponding to values @xmath80 , @xmath81 and @xmath82 in @xmath95 units . up to @xmath91 , the electrons get higher acceleration and their trajectories are more curved by the influence of the electromagnetic force than in the case of @xmath123 ( figure [ fig2 ] ) . from @xmath91
up to @xmath89 , all the electrons tend to follow the motion of the center of the knot , a situation much clearer from @xmath89 up to @xmath84 .
the final velocities of these electrons are in the range from @xmath132 to @xmath133 in units of the light velocity @xmath15 . and [ fig3 ] with a value @xmath134 .
from left to right and top to bottom , @xmath91 , @xmath89 , @xmath84 .
the energy levels shown correspond to values @xmath80 , @xmath81 and @xmath82 of the density energy in @xmath95 for the same times .
the initial velocity of the electrons is @xmath116 .
the electron initial positions considered are given by @xmath117 , with @xmath118 , by @xmath119 , and by @xmath120 .
the final values of the velocities at @xmath84 run from @xmath135 to @xmath136 in units of the light velocity @xmath15.,title="fig:",scaledwidth=45.0% ] and [ fig3 ] with a value @xmath134 . from left to right and top to bottom , @xmath91 , @xmath89 , @xmath84 .
the energy levels shown correspond to values @xmath80 , @xmath81 and @xmath82 of the density energy in @xmath95 for the same times .
the initial velocity of the electrons is @xmath116 .
the electron initial positions considered are given by @xmath117 , with @xmath118 , by @xmath119 , and by @xmath120 .
the final values of the velocities at @xmath84 run from @xmath135 to @xmath136 in units of the light velocity @xmath15.,title="fig:",scaledwidth=40.0% ] and [ fig3 ] with a value @xmath134 . from left to right and top to bottom , @xmath91 , @xmath89 , @xmath84 .
the energy levels shown correspond to values @xmath80 , @xmath81 and @xmath82 of the density energy in @xmath95 for the same times .
the initial velocity of the electrons is @xmath116 .
the electron initial positions considered are given by @xmath117 , with @xmath118 , by @xmath119 , and by @xmath120 .
the final values of the velocities at @xmath84 run from @xmath135 to @xmath136 in units of the light velocity @xmath15.,title="fig:",scaledwidth=50.0% ] in figure [ fig4 ] , the evolution of the position of electrons can be seen when the prefactor is @xmath134 .
the initial velocity and initial position of the electrons are the same as in the cases @xmath123 ( figure [ fig2 ] ) and @xmath129 ( figure [ fig3 ] ) . as in previous cases ,
electron trajectories are plotted in figure [ fig4 ] from @xmath86 up to @xmath91 ( in the left - hand side pannel ) , up to @xmath89 ( in the center pannel ) , and up to @xmath84 ( in the right - hand side pannel ) , and we include some energy density levels of the electromagnetic knot at the time considered , corresponding to values @xmath80 , @xmath81 and @xmath82 in @xmath95 units .
the global behaviour of the electron trajectories is very similar to the case in which @xmath129 , but now the electrons are even more accelerated and their trajectories are even more curved by the influence of the electromagnetic force .
all the electrons clearly tend to follow the motion of the center of the knot along the @xmath107-axis , with final velocities in the range from @xmath137 to @xmath138 in units of the light velocity @xmath15 .
in this paper we have considered the classical relativistic motion of charged particles in a knotted electromagnetic field .
we have seen how to construct electromagnetic knots from maps between the three - sphere and the two - sphere . in the particular case of the hopf map ,
whose fibres are mutually linked , we have written the expression for the electromagnetic field .
this is a solution of the maxwell equations in vacuum such that any pair of electric lines is a link and any pair of magnetic lines is a link .
we have considered some properties of this field , in particular the electromagnetic energy density .
we have seen that the most part of the energy density of the knot is , at @xmath41 , localized into a sphere whose radius is the mean quadratic radius of the energy density distribution . as time evolves , the spherical symmetry is broken and the mean quadratic radius of the distributios spreads out .
we have considered the relativistic motion of electrons ( considered as point particles ) in the electromagnetic field of the hopf map .
we have computed the trajectories of the electrons starting with zero initial velocities , and we have seen that these electrons are strongly accelerated by the electromagnetic force , becoming ultrarelativistic in a period of time that depends on the knot size . finally we consider that a deeper understanding of the interaction between electromagnetic knots and test particles could be useful to design experiments to produce knots in the laboratory .
the authors thank support from the spanish ministerio de educacin y ciencia under project aya2009 - 14027-c07 - 04 . | in this paper we consider the classical relativistic motion of charged particles in a knotted electromagnetic field . after reviewing how to construct electromagnetic knots from maps between the three - sphere and the two - sphere ,
we introduce a mean quadratic radius of the energy density distribution in order to study some properties of this field .
we study the classical relativistic motion of electrons in the electromagnetic field of the hopf map , and compute their trajectories .
it is observed that these electrons initially at rest are strongly accelerated by the electromagnetic force , becoming ultrarelativistic in a period of time that depends on the knot energy and size . |
for several reasons there is particular interest in the low - lying eigenmodes of the dirac operator in qcd . from phenomenological models
it is expected that for sufficiently light quark masses physics , e.g. hadron correlators , is dominated by these modes .
calculations on the lattice enable us to study the relevance of low - lying eigenmodes for various observables directly . furthermore , the eigenmodes of the dirac operator carry information about the topological content of the background gauge field .
the topological properties can also be probed using gluonic methods .
these however typically suffer from large fluctuations on very short scales and therefore require some sort of smoothing procedures .
usually the quantities of interest are affected by the applied filtering method .
results from fermionic methods on the other hand are expected to be sensitive to the effects caused by explicitly breaking chiral symmetry when using wilson fermions .
although the symmetry is expected to be restored in the continuum limit , these effects are potentially large on coarse lattices .
however , these can be reduced by applying symanzik s improvement program . removing discretization effects of @xmath1 , where @xmath2 is the lattice spacing ,
is achieved by adding a single counterterm to the action .
the improved wilson - dirac operator then reads : @xmath3 , \label{eq : dirac}\ ] ] where @xmath4 is the hopping term of the usual wilson fermion action . for the calculations presented here we used o(20 - 50 ) dynamical configurations with @xmath5 flavours of degenerate quarks for six different combinations of @xmath6 and @xmath7 .
we used lattices of size @xmath8 with a lattice spacing @xmath2 that varies between 0.104(1 ) and 0.095(1 ) fm ( using the force scale @xmath9 fm ) .
for the lightest and heaviest sea quarks the ratio @xmath10 was 0.60 and 0.83 , respectively . for comparison
we performed the same calculations using quenched configurations at @xmath11 and @xmath12 fm .
o(100 ) smallest eigenvalues of the massless dirac operator calculated on 20 gauge dynamical ( left ) and quenched ( right ) configurations . ]
the wilson - dirac operator @xmath13 is a non - hermitian operator .
the chirality @xmath14 of an eigenmode @xmath15 can be calculated using the corresponding right eigenvector @xmath16 : @xmath17 we can distinguish between two types of eigenmodes .
firstly , complex modes with non - vanishing imaginary part and @xmath18 and , secondly , real modes with vanishing imaginary part and @xmath19 . in principle , it is possible to find the real modes by determining the zero modes of the hermitian operator @xmath20 .
if @xmath21 is a real mode of @xmath13 , then @xmath20 will have a zero mode for @xmath22 .
in practice it might however be difficult to detect several real modes which lie close to each other .
we therefore decided to calculate the 70 - 100 smallest eigenvalues of the non - hermitian operator @xmath13 using the arnoldi algorithm @xcite .
results for the massless dirac operator @xmath23 are plotted in fig .
[ fig : spectrum ] .
for dynamical fermions one would expect eigenmodes around zero to be suppressed . as can be seen from fig .
[ fig : spectrum ] , this seems indeed to be the case .
this effect is more pronounced if one looks at the density of the real modes ( fig .
[ fig : densreal ] ) .
density of the real modes as a function of the real part .
the area under the curve is normalized to one .
the graphs are ordered from heaviest , i.e. infinite , ( top ) to smallest ( bottom ) sea quark mass . ] real modes close to ( or below ) zero , which correspond to zero modes of @xmath20 at @xmath24 , are believed to be the origin of so - called exceptional configurations , which are encountered frequently in the quenched approximation when approaching the chiral regime .
although these modes are expected to be suppressed for dynamical fermions , there is a non - zero probability for this kind of configurations to occur ( see fig .
[ fig : densreal ] ) .
the correlation of the chirality @xmath25 shown in fig .
[ fig : corr ] was found to be narrow , confirming the observations made in @xcite .
typically the chirality of small real modes is concentrated in a few time slices .
chirality per timeslice @xmath26 for real modes ( upper picture ) and the normalized correlation function of the chirality ( lower picture ) @xmath27 on a configuration at @xmath28 , @xmath29 .
the real mode with the smallest eigenvalue is plotted with a solid line . ]
the topological susceptibility normalised by the volume is defined by @xmath30 the topological charge @xmath31 can be obtained via the atiyah - singer index theorem , which is approximately true on the lattice .
it establishes a relation between the difference in the number of real modes with negative ( @xmath32 ) and positive ( @xmath33 ) chirality and the topological charge : @xmath34 for a subset of configurations we compared the results for @xmath31 with those obtained from gluonic operators using the boulder smoothing technique .
the maximum difference found was @xmath35 . for a more detailed comparison see @xcite .
the results for the topological susceptibility are plotted in fig .
[ fig : suscept ] .
the discrepancy between the results from the fermionic method applied in this work and from the gluonic method used by hart and teper is likely to be explained by a lack of statistics .
another source of uncertainty arises from the fact that we only calculated the lowest 70 - 100 eigenvalues .
although the probability for real modes is expected to decrease when moving away from zero , it does not become zero . to check for this possible error
, we looked at the topological susceptibility as a function of the largest real mode taken into account ( see fig . [
fig : plateau ] ) . for all data sets we found a reasonable plateau .
topological susceptibility as a function of the pseudoscalar meson mass .
the solid symbols show the results from this work , while ukqcd results using gluonic methods @xcite are plotted with open symbols . ]
topological susceptibility versus the cut - off of used real modes . ]
comparing real modes of quenched and dynamical configurations we found evidence for an expected depopulation of the region around zero .
the results for @xmath31 obtained by using the atiya - singer - theorem were in reasonable agreement with numbers obtained from gluonic methods .
our preliminary results for the topological susceptibility were found to be similar to other calculations using the same configurations but gluonic methods .
the numerical calculations were performed on the hitachi sr8000 at lrz ( munich ) , the cray t3e at epcc ( edinburgh ) and zib ( berlin ) .
we wish to thank all institutions for their support
. 9 r.b .
lehoucq , d.c .
sorensen , c. yang , `` arpack users guide '' , 1997 ; k. maschhoff , d. sorensen , 1996 .
t. degrand , a. hasenfratz , phys . rev .
d * 64 * ( 2001 ) 034512 .
a. hart , this conference .
a. hart , m. teper [ hep - lat/0108022 ] . | we present preliminary results for the topological charge and susceptibility determined from the low - lying eigenmodes of the wilson - dirac operator .
these modes have been computed on dynamical configurations with @xmath0 non - perturbatively improved wilson fermions .
we compare our results with the eigenmodes of fermions in the quenched approximation . |
the advances in the nanotechnology make it possible to produce small quantum dots with desired controllable properties .
the analogy between such a system and an atom has proved to be quite close @xcite . since however
no natural restrictions for choosing the control parameters exist , the quantum dots may show new properties that we can not obtain from studies on atoms .
for example , fano resonances have been observed experimentally in quantum dots @xcite , and the fano parameters may be complex @xcite .
theoretically , the fano parameter expresses the interference between the resonant part of the transmission and a smooth ( direct ) nonresonant part .
it is real when the resonance part is caused by the existence of an isolated resonance state @xcite as it is the case usually in atoms .
when however the resonant part itself results from the interference of , e.g. , two neighboring resonance states , the fano parameter becomes complex @xcite . the same may appear in a cavity due to the absorption at the walls @xcite .
further , it has been stated @xcite that the line shape of a fano resonance may be affected by some dephasing @xcite that may be caused by intrinsic sources ( e.g. , from electron - electron interactions ) as well as by extrinsic sources ( e.g. , radiation , magnetic impurities ) @xcite . for a quantitative study , the phase rigidity @xcite @xmath5 has been introduced that characterizes the degree to which the wave function @xmath6 is really complex ( the phase @xmath7 arises from a transformation so that re@xmath8 and im@xmath8 are orthogonal ) .
an experimental and theoretical study @xcite showed that the two different mechanisms , dephasing and dissipation , are equivalent in terms of their effect onto the evolution of fano resonance line shapes .
another interesting observation in the experimental results on quantum dots @xcite is that , as a function of the gate voltage that controls the transparencies of the point contacts , the widths of the observed resonances behave non - monotonic .
the conductance peaks start as narrow breit - wigner resonances when the quantum dot is pinched off , then widen as the contacts are opened into resonances exhibiting the kondo effect @xcite .
as the contacts are opened further , the resonances become more narrow and have the fano form with some background conductance .
a similar result is obtained in a study on a tunable microwave scattering device @xcite . the explanation given in ref .
@xcite is that diffraction at the contacts to the quantum dots is strongest at intermediate point contact transparencies , leading to large sticking probabilities .
the non - monotonic increase of resonance widths as a function of the degree of opening the system is however a typical feature of open quantum systems @xcite .
it was found first in theoretical nuclear reaction studies by using the formalism of the effective non - hermitian hamiltonian @xmath9 @xcite , then in atoms @xcite and in microwave cavities @xcite where it is proven experimentally @xcite .
also the transmission through a double quantum dot is studied as a function of the coupling strength between dot and attached leads @xcite . in the theoretical studies ,
the non - monotonic behavior of the resonance widths is caused by the width bifurcation that may appear at the avoided crossings of resonance states in the complex energy plane .
further studies of open quantum systems showed that the real and imaginary parts of the eigenfunctions of @xmath9 evolve more or less independently from one another in the avoided level crossing scenario @xcite and that long - range correlations occur between the different states of the system @xcite .
the effective hamiltonian used in the random matrix interpolation between the standard ensembles with real and complex matrix elements , is @xmath10 where @xmath11 and @xmath12 are real and complex random hermitian matrices , respectively , and @xmath13 is a crossover parameter @xcite .
for such an ensemble , the eigenvector elements acquire correlations between the elements of the same eigenvector @xcite and between different eigenvectors @xcite . for individual systems , such a crossover
may be observed already in a billiard with only two attached waveguides @xcite .
the wave functions in the cross - over regime show long - range correlations @xcite like the eigenvectors of the hamiltonian @xmath14 .
these long - range correlations of the wave functions in the real - to - complex crossover have recently been measured in an open microwave billiard @xcite .
thus , the results obtained experimentally and theoretically for the profile of the transmission peaks , for the intensity fluctuations and also for the long - range correlations of the wave functions point to the fact that the system through which the transmission occurs may be essentially different from the original closed quantum system without attached leads .
the differences arise from the interaction of neighboring resonance states via the continuum of scattering states . in quantum dots
, these differences can be traced experimentally by varying the gate voltage that controls the transparencies of the point contacts . in the following
, we will study the phases of the eigenfunctions of the effective hamiltonian in the avoided level crossing scenario in detail .
we will show that they are not fixed but vary due to the mutial influence of neighboring resonance states . in this spirit
, a neighboring resonance state causes some `` perturbation '' for the considered state which is similar to that caused by an impurity .
there is however an important difference between these two cases .
in contrast to the perturbation by an impurity , the mutial interaction between neighboring states can not be avoided .
it is an internal property of an open quantum system in the regime of overlapping resonances . in sect .
ii , we repeat some general properties of the interplay between the internal and external interaction in an open quantum system .
while the internal interaction is of standard type , the external interaction occurs additionally via the common continuum of scattering wave functions .
the amplitude for the transmission through a quantum dot will also be given .
it is resonant in relation to the effective hamiltonian @xmath9 of the open quantum system .
then we define the phase rigidity @xmath2 of the eigenfunctions @xmath0 of @xmath9 .
further , we consider the scattering wave functions @xmath4 being solutions of @xmath15 in the total function space .
in the interior of the system , the @xmath4 can be represented by the @xmath0 and , as a consequence , the phase rigidity @xmath3 of the scattering wave functions @xmath4 is related to that of the eigenfunctions @xmath0 . in sect .
iii , we provide numerical results for the relation between transmission and phase rigidity @xmath3 of the scattering wave functions for three special cases .
the results are summarized and some conclusions are drawn in the last section .
the relation between a closed and the corresponding open quantum system is as follows .
a closed quantum system is described by a hermitian hamilton operator @xmath16 the eigenvalues and eigenfunctions of which contain the internal interaction @xmath17 of the discrete states .
when embedded into the common continuum of scattering states , the discrete eigenstates of the closed system turn over into resonance states with a finite lifetime .
the effective hamiltonian @xmath9 of the open quantum system contains @xmath16 as well as an additional term @xcite that describes the coupling of the resonance states to the common environment , @xmath18 here @xmath19 stand for the coupling matrix elements between the _ eigenstates _ of @xmath16 and the environment that may consist of different continua @xmath20 , e.g. the scattering waves propagating in the left ( @xmath21 ) and right ( @xmath22 ) leads attached to a quantum dot @xcite .
they are described by the hamiltonian @xmath23 .
@xmath9 is non - hermitian , its eigenvalues @xmath24 and eigenfunctions @xmath0 contain the external interaction @xmath25 of the resonance states via the continuum ( @xmath25 is used here instead of the concrete values @xmath26 and @xmath27 ) . while @xmath17 causes level repulsion in energy ,
@xmath25 is responsible for the bifurcation of the widths of the resonance states .
as long as @xmath28 , the spectroscopic properties of the open system are similar to those of the corresponding closed system .
the lifetimes ( widths ) of the resonance states are , as a rule , comparable in value , i.e. the states exist at the same time and can influence one another .
usually , the states avoid crossing in the complex energy plane in a similar manner as it is well known from the avoided crossings of the discrete states of a closed system . when however @xmath29 , the resonance states do no longer exist at the same time due to their different lifetimes owing to widths bifurcation @xcite .
they do not cross therefore in the complex energy plane .
most interesting is the situation @xmath30 .
the interplay between @xmath17 and @xmath25 may cause unexpected and even counterintuitive results such as the non - monotonic dependence of the resonance widths on the degree of opening the system @xcite . since the effective hamiltonian @xmath9 depends explicitely on the energy @xmath31 , so do its eigenvalues @xmath24 .
the energy dependence is weak , as a rule , in an energy interval that is determined by the width of the resonance state .
the solutions of the fixed - point equations @xmath32 and of @xmath33 are numbers that coincide approximately with the poles of the @xmath34 matrix .
the width @xmath35 determines the time scale characteristic of the resonance state @xmath36 .
the amplitude for the transmission through a quantum dot is @xcite @xmath37 where the scattering wave functions in the leads are denoted by @xmath38 . according to ( [ trheff ] ) ,
the transmission is _ resonant _ in relation to @xmath9 . in ( [ trheff ] ) ,
the eigenvalues @xmath24 with their full energy dependence are involved .
the eigenfunctions @xmath0 of @xmath9 are complex and biorthogonal , @xmath39 the value @xmath40 is a measure for the biorthogonality of the eigenfunctions @xmath0 of @xmath9 and for their phase rigidity , @xmath41 .
it is large for an isolated resonance state at the energy @xmath42 , where the transmission probability has a peak . in the regime of overlapping resonance states however , where avoided level crossings appear @xcite , @xmath2 is usually small .
approaching a branch point in the complex energy plane where two eigenvalues @xmath43 and @xmath44 coalesce , we have @xcite @xmath45 and @xmath46 , and further @xmath47 here , the widths bifurcate : with further increasing @xmath25 , one of the states aligns with a scattering state ( channel wave function @xmath48 or @xmath49 in the one - channel transmission ) and becomes short - lived while the other one becomes long - lived .
for large @xmath25 , we have therefore only one long - lived resonance state and , as for non - overlapping resonance states , @xmath50 at the energy @xmath51 of the long - lived state .
the scattering wave function @xmath52 is solution of the schrdinger equation @xmath53 in the total function space that consists of the discrete states of the quantum dot and of the scattering states @xmath38 in the attached leads .
@xmath54 is hermitian .
the solution reads @xcite @xmath55 where @xmath56\phi_\lambda ( r ) $ ] is the wave function of the resonance state @xmath36 . according to ( [ total ] )
, the eigenfunctions @xmath57 of the effective hamiltonian @xmath9 give the main contribution to the scattering wave function @xmath58 in the interior of the cavity .
here , @xmath59 according to the relation ( [ total1 ] ) , the phase rigidity @xmath3 of the scattering wave function , eq . ( [ rig1 ] ) , is generally determined to the values of the phase rigidity @xmath2 of the individual resonance states @xmath36 . while @xmath2 characterizes the phase rigidity of the special resonance state @xmath36 , @xmath3 contains the @xmath2 from the different states that contribute to the scattering wave function @xmath4 in the considered energy region .
this difference between @xmath3 and @xmath2 is illustrated best by the following example .
approaching a branch point in the complex energy plane , the contributions to ( [ total1 ] ) from the two states @xmath36 and @xmath60 with the relation ( [ nl5 ] ) of their wave functions to each other , cancel each other .
the phase rigidity @xmath3 of the scattering wave function is determined therefore by the contributions of other states that , as a rule , are relatively far from one another .
that means , in spite of @xmath61 , @xmath62 might be large at the branch point .
we mention that the scattering wave function ( [ total1 ] ) has a similar structure as the transmission amplitude ( [ trheff ] ) .
both expressions consist of a sum over overlapping resonance states with the weight factor @xmath63 .
we expect therefore some correlation between both values .
this relation , being trivial for isolated resonances , is of special interest in the regime of overlapping resonances .
the transmission may be large not only at the positions @xmath64 of the resonance states @xmath36 as in the case of isolated resonances .
it may be enhanced in a larger energy region when @xmath65 , i.e. when some states @xmath36 are ( partly ) aligned with the scattering states @xmath38 so that @xmath66 for them .
in such a case , the matrix elements @xmath67 may be large not only at @xmath64 but in a larger energy region . for illustration
, we will provide in the following section some numerical results for @xmath68 and @xmath62 for two model quantum billiards with a small number of states the transmission through which is studied earlier , and for a realistic chaotic quantum billiard .
the correlation between @xmath68 and @xmath69 in the overlapping regime can clearly be seen in all cases .
for illustration , let us first consider the relation between the degree of resonance overlapping and the phase rigidity in the transmission through a 1d - system , see fig .
the widths of the states in the middle @xmath70 of the spectrum are larger than those near to the thresholds @xmath71 as can be seen immediately from the profile of the transmission peaks .
therefore , also the degree of resonance overlapping increases towards the middle of the spectrum .
the results show that the transmission increases with increasing resonance overlapping and , correspondingly , the phase rigidity decreases .
that means , the transmission increases when the resonance states align with the scattering wave function which is expressed by the decreasing of @xmath3 .
( full lines ) and phase rigidity @xmath62 ( dashed lines ) for a chain of 6 sites as a function of energy with v=0.5 ( left ) and v=0.7 ( right ) . for
the model see ref .
, title="fig:",width=151,height=113 ] ( full lines ) and phase rigidity @xmath62 ( dashed lines ) for a chain of 6 sites as a function of energy with v=0.5 ( left ) and v=0.7 ( right ) . for the model see ref .
, title="fig:",width=151,height=113 ] let us now consider a double quantum dot consisting of two identical single dots that are connected by an internal wire .
suppose both dots and the wire have one state each . at @xmath72 , the eigenvalues @xmath24 and @xmath44 of the effective hamiltonian coalesce and a branch point in the complex energy plane appears @xcite .
when @xmath73 , the states repel each other in energy while widths bifurcation starts beyond the branch point where @xmath74 . in fig .
[ fig2 ] , we show the transmission @xmath68 and the landscape of @xmath62 over energy @xmath31 and coupling strength @xmath25 ( for fixed @xmath17 ) . the smallest value of @xmath62 ( @xmath75 ) is reached when the transmission is maximal with a plateau @xmath76 ( for the plateau compare fig . 4 in @xcite ) .
at @xmath72 , @xmath62 is relatively large due to the fact that the contributions of the two states @xmath36 and @xmath60 cancel each other .
beyond @xmath77 , the phase rigidity increases further up to its maximal value 1 .
( top ) and the landscape of the phase rigidity @xmath62 ( bottom , thin lines ) for a double quantum dot over energy @xmath31 and coupling strength @xmath25 .
the distance between the contour lines is @xmath78 .
the minimal value @xmath79 is surrounded by a high density of contour lines .
the highest shown contour line corresponds to @xmath80 .
the re@xmath81 ) of the three eigenstates ( thick lines in both panels of the figure ) are calculated at @xmath70 .
the branch point is at @xmath82 .
@xmath83 . around @xmath84 ,
the phase rigidity is minimal and the transmission maximal with a plateau @xmath76 ( for the plateau compare fig . 4 in ref .
@xcite ) . , title="fig:",width=234,height=211 ] ( top ) and the landscape of the phase rigidity @xmath62 ( bottom , thin lines ) for a double quantum dot over energy @xmath31 and coupling strength @xmath25 .
the distance between the contour lines is @xmath78 .
the minimal value @xmath79 is surrounded by a high density of contour lines .
the highest shown contour line corresponds to @xmath80 .
the re@xmath81 ) of the three eigenstates ( thick lines in both panels of the figure ) are calculated at @xmath70 .
the branch point is at @xmath82 .
@xmath83 . around @xmath84 ,
the phase rigidity is minimal and the transmission maximal with a plateau @xmath76 ( for the plateau compare fig . 4 in ref .
@xcite ) . , title="fig:",width=234,height=211 ] in fig .
[ fig3 ] , we show @xmath68 and @xmath62 calculated for a realistic chaotic system with many states and many avoided level crossings . the high correlation between the two values @xmath68 and @xmath69 can be seen also under these conditions . in some energy intervals , @xmath68 is near to @xmath85 . here
@xmath62 is small , in agreement with the results shown in fig .
[ fig2 ] .
( top ) and phase rigidity @xmath62 ( bottom ) for a sinai billiard over energy @xmath31 and radius @xmath86 of the circular disk ( size of the billiard : @xmath87 in units of the width of the leads ) .
the calculations are performed in the tight - binding lattice model @xcite .
, title="fig:",width=283,height=245 ] + ( top ) and phase rigidity @xmath62 ( bottom ) for a sinai billiard over energy @xmath31 and radius @xmath86 of the circular disk ( size of the billiard : @xmath87 in units of the width of the leads ) .
the calculations are performed in the tight - binding lattice model @xcite .
, title="fig:",width=283,height=245 ]
from the results obtained , we conclude that @xmath88 characterizes the scenario of overlapping resonances in which the resonance states interact with one another and avoided ( and true ) crossings appear , generally . in this scenario ,
the system can not be described by a hermitian hamilton operator that provides rigid phases of its eigenfunctions ( @xmath89 ) .
rather , @xmath9 is non - hermitian , its eigenfunctions are biorthogonal and @xmath2 varies with energy and coupling strength . since the scattering wave function and the eigenfunctions of the effective hamiltonian
are related to one another according to ( [ total1 ] ) , also @xmath3 varies between 1 and 0 in the regime of overlapping resonances .
that means , the reduction of the phase rigidity @xmath3 of the scattering wave function is caused , at least partly , by the mutial interaction of neighboring resonances .
it is an internal property of an open quantum system in the overlapping regime and can not be avoided .
the reduction of the phase rigidity @xmath3 is , in the one - channel transmission , an expression for the ( partial ) alignment of two of the eigenfunctions @xmath90 of the effective hamiltonian @xmath9 with the two scattering wave functions @xmath91 in the overlapping regime . due to this alignment
( described by @xmath92 ) , the transmission through the system is enlarged in a certain energy region around the energies @xmath93 of the two resonance states . as can be seen from fig .
[ fig2 ] and the corresponding discussion in @xcite , the profile of the transmission is , when @xmath94 , completely different from that through two isolated resonance states .
it is plateau - like .
this result explains the correlation between @xmath68 and @xmath69 which can be seen in all our numerical studies ( figs .
[ fig1 ] , [ fig2 ] and [ fig3 ] ) .
both values are , according to eqs .
( [ trheff ] ) and ( [ total1 ] ) , characterized by the contributions from a sum of overlapping resonance states . since the variation of @xmath2 and @xmath62 with energy follows from the interaction of the resonance states via the common continuum of scattering wave functions , it is related , generally , to the profile of fano resonances .
the relation is however unique only with respect to @xmath2 .
for example , the mutial influence of neighboring resonances is , according to the results presented in @xcite , maximum when two eigenvalues of the effective hamiltonian coincide .
the line profile of two completely overlapping resonance states is , in the one - channel case and up to the background term , given by @xcite @xmath95 where @xmath96 and @xmath97 . according to this equation ,
the line profile of the two resonances differs strongly from that of isolated resonances @xcite ( determined by the rigid value @xmath89 ) . in correspondence to this result , we have @xmath98 in the one - channel case when two resonance states completely overlap .
however @xmath62 might be large in this case as for weakly overlapping resonance states , as discussed above .
this result underlines once more the difference between the value @xmath2 characteristic of a special resonance state @xmath36 of the system , and @xmath3 characteristic of the behavior of a sum of resonance states in the overlapping regime .
thus , the avoided crossings of resonance states represent one of the sources for the dephasing @xcite observed in experimental data .
the variation of @xmath99 and @xmath100 between 0 and 1 is characteristic of an open quantum system with overlapping resonance states , as stated above .
therefore , dephasing and dissipation should be equivalent in terms of their effect onto the fano profile .
such a result is observed experimentally , indeed @xcite . as to averaged values
, we mention the following results obtained earlier . the distributions of @xmath62 with energy and ensemble averaging calculated for large chaotic cavities , do not depend on the concrete shape of the cavity @xcite since the averaging smears the different contributions to @xmath62 . as a result
, the distribution of @xmath62 is characterized , in such a case , only by the number of channels @xcite .
we underline that the interaction of neighboring resonance states via the common continuum which is considered here as source for the dephasing @xcite , is fundamentally different from , e.g. , the electron - electron interaction or the interaction due to some radiation .
it is rather of nonlinear geometrical origin , related to branch points in the complex energy plane , as can be seen in the following manner . on the one hand ,
the branch points determine the physical properties of an open quantum system in the overlapping regime as discussed in @xcite .
they are , generally , related to the avoided crossings of resonance states which demonstrate the mutual influence of resonance states . on the other hand ,
as shown in the present paper , the mutual interaction of neighboring resonance states is accompanied by phase changes of the eigenfunctions @xmath0 of @xmath9 as well as of the scattering wave functions @xmath4 inside the cavity . in any case
, the interaction of neighboring resonance states via the common continuum and the resulting dephasing can not be neglected in an open quantum system in the regime of overlapping resonances .
a. yacoby , m. heiblum , d. mahalu , and h. shtrikman , phys .
lett . * 74 * , 4047 ( 1995 ) ; e. buks , r. schuster , m. heiblum , d. mahalu , v. umansky , and h. shtrikman , phys .
lett . * 77 * , 4664 ( 1996 ) ; r. schuster , e. buks , m. heiblum , d. mahalu , v. umansky , and h. shtrikman , nature ( london ) * 385 * , 417 ( 1997 ) .
e. persson , k. pichugin , i. rotter , and p. eba , phys .
e * 58 * , 8001 ( 1998 ) ; p. eba , i. rotter , m. mller , e. persson , and k. pichugin , phys . rev .
e*61 * , 66 ( 2000 ) ; i. rotter , e. persson , k. pichugin , and p. eba , phys .
e * 62 * , 450 ( 2000 ) . | we consider the effective hamiltonian of an open quantum system , its biorthogonal eigenfunctions @xmath0 and define the value @xmath1 that characterizes the phase rigidity of the eigenfunctions @xmath0 . in the scenario with avoided level crossings
, @xmath2 varies between 1 and 0 due to the mutual influence of neighboring resonances .
the variation of @xmath2 may be considered as an internal property of an _ open _ quantum system . in the literature ,
the phase rigidity @xmath3 of the scattering wave function @xmath4 is considered .
since @xmath4 can be represented in the interior of the system by the @xmath0 , the phase rigidity @xmath3 of the @xmath4 is related to the @xmath2 and therefore also to the mutual influence of neighboring resonances . as a consequence ,
the reduction of the phase rigidity @xmath3 to values smaller than 1 should be considered , at least partly , as an internal property of an open quantum system in the overlapping regime .
the relation to measurable values such as the transmission through a quantum dot , follows from the fact that the transmission is , in any case , resonant with respect to the effective hamiltonian .
we illustrate the relation between phase rigidity @xmath3 and transmission numerically for small open cavities . |
at cern s large hadron collider ( lhc ) , it is widely anticipated that signals of new physics , for example supersymmetry , may manifest themselves as large excesses of data compared to expected qcd and electroweak backgrounds at high momentum scales @xcite .
the estimation of these backgrounds will be one of the elements in ascertaining the presence of any new physics from such signals .
consequently , considerable effort is being invested across the particle physics community in the development of methods to understand and predict backgrounds ( some of the issues involved are described nicely in ref .
@xcite ) . given the qcd methods that are available today , some of the best prospects for obtaining systematic , accurate predictions of backgrounds involve next - to - leading order ( nlo ) qcd calculations . by carrying out a systematic expansion in the strong coupling and obtaining the first two terms ( leading order ( lo ) and nlo ) for a given process , one often obtains predictions that are accurate to @xmath2 .
the importance of nlo predictions in the lhc programme has motivated a large calculational effort destined to extend the range of processes known at nlo ( for reviews , see refs .
@xcite ) .
while the majority of nlo calculations show some degree of convergence relative to the lo results , several groups have commented in recent years on the appearance of @xmath0 factors , ratios of nlo to lo results , that grow dramatically towards high transverse momenta @xcite ( similar behaviour is visible also in @xcite ) .
the problem generally occurs for hadronic observables ( jet transverse momenta , etc . ) in processes that involve heavy vector bosons or heavy quarks , at scales far above the boson or quark mass . of the hardest jet ( middle ) , and the scalar sum of the transverse momenta of all the jets , @xmath3 ( right ) .
the bands correspond to the uncertainty from a simultaneous variation of @xmath4 by a factor of two either side of a default @xmath5{p_{t , j1}^2 + m_{\mathrm{z}}^2}}$ ] .
the jet algorithm is anti-@xmath6 @xcite with @xmath7 and only events whose hardest jet passes a cut @xmath8 are accepted .
the cross sections include the branching ratio @xmath9 . ,
scaledwidth=95.0% ] fig .
[ fig : lo_nlo_dist ] illustrates this for the @xmath10z+jet process at lhc ( @xmath11 ) energies .
it shows the distributions of three observables that are non - zero for configurations involving a z - boson and one or more partons : the transverse - momentum of the z - boson ( @xmath12 ) , the transverse - momentum of the highest-@xmath1 jet ( @xmath13 ) and the effective mass ( scalar sum of the transverse momenta ) of all jets ( @xmath3 ) . at lo ,
all three distributions are identical . at nlo ,
the @xmath12 observable is rather typical of a qcd observable : its distribution has a nlo @xmath0-factor of about @xmath14 , fairly independently of @xmath12 , and its scale dependence is reduced with respect to lo .
the @xmath13 distribution is more unusual : at high @xmath1 it has a @xmath0-factor that grows noticeably with @xmath13 , reaching values of about @xmath15 , which is anomalously large for a qcd correction .
the @xmath3 observable is even more striking , with @xmath0-factors approaching @xmath16 . + ( a ) + ( b ) + ( c ) given that fig .
[ fig : lo_nlo_dist ] involves momentum scales where @xmath17 , one is driven to ask how it is that such `` giant '' @xmath0-factors can arise . as touched on in @xcite , and discussed in more detail in @xcite for the @xmath13 case , the answer lies in the appearance of diagrams with new kinematic topologies at nlo .
this is illustrated in fig .
[ fig : z+jet - diags ] : at lo the only event topology ( a ) is that of a z - boson recoiling against a quark or gluon jet .
one type of nlo diagram involves gluon radiation from this basic topology , giving modest corrections to all our observables .
however , there are also nlo diagrams ( b , c ) whose topology is that of a dijet event , in which a soft or collinear z - boson is radiated from outgoing or incoming legs .
these diagrams do not contribute significantly to the @xmath12 distribution , because the z - boson carries only a moderate fraction of the total @xmath1 .
however when examining @xmath13 , it is irrelevant whether the z boson is soft or not .
contributions b and c then lead to a result that is of order @xmath18 , where the double logarithm comes from the integration over soft and collinear divergences for z emission .
the ratio of the nlo to lo results is therefore @xmath19 , , and are usually much smaller .
examples do exist of `` giant '' ew effects when tagging flavour @xcite .
] rather than just @xmath20 , hence the @xmath0-factor that grows large with increasing @xmath1 . also comes from the fact that one can have @xmath21 scattering that emits a z , whereas the @xmath22 partonic channel does not contribute at lo .
] for the @xmath3 observable the enhancement is even bigger because the dijet topology leads to @xmath23 instead of @xmath24 at lo .
while it is reassuring that we can understand the physical origins of the large @xmath0-factors in fig .
[ fig : lo_nlo_dist ] , we are still left with doubts as to the accuracy of the nlo z+jet predictions for @xmath13 and @xmath3 , since they are dominated by the lo result for the z+2-parton topologies .
one way forward would be to calculate the full nnlo corrections for the z+jet process .
however , while work is progressing on nnlo calculations of @xmath25 processes with qcd final states ( see e.g. @xcite and references therein ) , results are not yet available ; nor are they likely to become available any time soon for some of the more complex processes where giant @xmath0-factors have been observed ( e.g. some observables in @xmath26 @xcite ) . alternatively one could simply try to avoid observables like @xmath13 and @xmath3 in inclusive event samples . for example , with additional cuts on the vector - boson momentum or a second jet , refs .
@xcite showed that the @xmath0-factors are significantly reduced .
however , given the many analyses that are foreseen at the lhc , it is likely that at least a few will end up probing regions where giant @xmath0-factors are present . to understand how else one might address the problem of giant @xmath0-factors , one can observe that in our @xmath27 example , the bottleneck in obtaining a nnlo prediction is the inclusion of the two - loop @xmath28 contributions and proper cancellation of all infrared and collinear divergences . yet
the two - loop ( and squared one - loop ) contribution will have the topology of diagram a in fig .
[ fig : z+jet - diags ] and should not be responsible for the dominant part of the nnlo correction , which will instead come from diagrams with the topology of b and c , with either an extra qcd emission or a loop .
so if one includes tree - level @xmath29 and 1-loop @xmath30 diagrams ( i.e. @xmath31 at nlo ) and supplements them with even a crude approximation to the two - loop @xmath32 result , one that suffices merely to cancel all divergences , then one should have a good approximation to the full nnlo result ( a related observation has been exploited to obtain approximate nnlo results for high-@xmath1 j/@xmath33 production in @xcite ) .
the purpose of this article is to develop a general method for obtaining such rough estimates of missing loop corrections .
our approach , called loopsim , will be based on unitarity . after explaining how it works in section [ sec : loopsim - section ] , and outlining a secondary `` reference - observable '' approach for control purposes
, we will test the method by comparing its results to full nnlo predictions for lepton-@xmath1 spectra in drell - yan production in section [ sec : dy - comparison ] , apply it to our z+jet observables in section [ sec : results ] and finally , in section [ sec : qcd ] , examine whether it can be of use even in the absence of giant @xmath0-factors , specifically for a number of dijet observables .
the main ingredient of the loopsim method is a procedure for taking a tree - level event with @xmath34 final state particles and supplementing it with a series of events with @xmath35 particles ( approximate 1-loop events ) , @xmath36 particles ( approximate 2-loop events ) , etc . , such that the sum of the weights of the full set of events is zero .
this `` unitarity '' property will ensure that all the soft and collinear divergences of the tree - level matrix elements will cancel against identical divergences in the simulated loop contributions . structure .
, scaledwidth=80.0% ] an outline of the procedure is given in fig .
[ fig : loopsim - diag ] . given a tree - level input event ( a )
, the first step is to interpret it as a sequence of emissions ( as if it had been produced by a parton shower ) , so that for example ( diagram b ) one can view particle 2 as having been emitted from particle 1 , and particle 4 as emitted from the beam .
the attribution of an emission sequence can be performed with the help of a suitable sequential - recombination jet algorithm and will be most meaningful in the limit that emissions are strongly ordered in angle and energy . the next stage is to decide which particles reflect the underlying hard structure of the event . if the event structure at the lowest possible order is that of a @xmath37 scattering , then one should identify two outgoing `` born '' particles .
the born particles will remain present in all the approximate `` loop '' events that are generated .
they are represented as thick red lines in diagram ( c ) .
again this step is most meaningful when all non - born emissions are soft and collinear .
one then generates a set of simulated `` 1-loop '' events by finding all ways of recombining one emitted particle with its emitter , diagrams ( d , e ) .
each such `` 1-loop '' event comes with a relative weight of @xmath38 compared to the tree - level diagram .
similarly the set of simulated `` 2-loop '' events is obtained by finding all ways of recombining two emitted particles with their emitter(s ) ( diagram f ) , each with relative weight @xmath39 ; and so forth down to events where only born particles remain ( in fig .
[ fig : loopsim - diag ] this is already reached at the two - loop level ) .
note that the loop - diagrams drawn in fig .
[ fig : loopsim - diag ] are not intended to represent the actual feynman diagrams that would be relevant at 1 and 2-loop level .
instead they indicate the way in which we have approximated the loop divergences , as the unitarising counterparts of the divergences that appear for each emission in the soft and collinear limits . given the above procedure for unitarising tree - level events
, we shall see that it is then straightforward to extend it to event sets that also include exact loop diagrams .
we start by examining the loopsim procedure in the simple case of purely gluonic tree - level events
. this will suffice to introduce most of the relevant concepts .
section [ sec : loopsim - flavour ] will then discuss some of the additional issues that arise for events with quarks and vector bosons , while the handling of events sets that include exact loop diagrams will be left to section [ sec : loopsim - nlo ] .
it is helpful to introduce some notation : firstly , @xmath40 is the number of final - state particles present in the lowest relevant order ( i.e. the number of final - state `` born '' particles ) .
for instance @xmath41 if considering higher - order corrections to dijet events , as in fig .
[ fig : loopsim - diag ] .
@xmath42 represents a generic event with @xmath34 final state particles .
so the starting event of fig .
[ fig : loopsim - diag ] would be labelled @xmath43 .
finally , @xmath44 will be an operator that acts on an event @xmath42 and returns all the events at @xmath45 loops obtained from @xmath42 using the loopsim method .
for instance , fig .
[ fig : loopsim - diag]d , e represents the action of @xmath46 on the input @xmath43 event ( a ) .
the central part of the loopsim method involves the construction of the operator @xmath47 acting on @xmath42 for all @xmath48 ( @xmath49 because the number of real final state particles can not be smaller than that of the lowest order event ) .
recall that the primary function of the loopsim method is to cancel the divergences that appear in the soft and collinear limits . in these limits
, events can be interpreted as stemming from a sequence of probabilistic ( parton - shower ) type @xmath50 splittings of some original hard born particles .
the knowledge of the splitting structure will help us generate loop events to cancel the divergences .
the attribution of a branching sequence is most easily performed using a sequential recombination jet algorithm ( and is inspired by the ckkw matching procedure @xcite ) .
we will use the cambridge / aachen ( c / a ) algorithm @xcite , which has important advantages over the @xmath6 algorithm when dealing with nested collinear divergences ( avoiding `` junk '' jets @xcite ) . as a first step , to each of the @xmath51 particles in the event @xmath42 , we assign a unique `` identity '' index @xmath52 .
we then run the cambridge / aachen ( c / a ) algorithm on the event .
it repeatedly clusters the pair of particles that are closest in angle , i.e. with smallest @xmath53 where @xmath54 is the usual squared angular distance in the @xmath55 plane , and @xmath56 is a free parameter , the loopsim radius .
the c / a algorithm continues until all the @xmath57 , at which point the remaining particles are deemed to cluster with the beam .
an @xmath58 clustering in the c / a algorithm can be reinterpreted as a @xmath59 splitting .
the c / a algorithm does not distinguish in any way between @xmath60 and @xmath61 .
however , in the soft limit , say @xmath62 , rather than viewing @xmath63 as splitting to @xmath60 and @xmath61 , it is a better reflection of the divergent structure of the amplitude to view @xmath63 as having emitted a soft gluon @xmath61 .
then @xmath60 is nothing other than particle @xmath63 with some small fraction of its energy removed . to account for this , in an @xmath58 clustering , if @xmath64 , then we declare that the `` identity '' @xmath65 of particle @xmath63 should be the same as that of particle @xmath60 , @xmath66 .
also we record @xmath67 as being a `` secondary emitter '' and remember that the object with identity @xmath68 has been emitted from the object with identity @xmath67 .
( exchange @xmath69 if @xmath70 ) .
this is represented in fig . [
fig : loopsim - diag]b by the fact that particle @xmath71 is a straight line , off which particle @xmath72 has been emitted ; the identity of the @xmath73 combination is @xmath74 . for an @xmath75 clustering , we record @xmath67 as having been emitted from the beam .
the next step in attributing structure to the event is to decide which event particles should be viewed as the born particles , i.e. which particles are responsible for the hard structure in the event .
inspired by the original formulation of the cambridge algorithm @xcite , for every @xmath76 recombination we assign a @xmath6 algorithm type hardness measure @xmath77 @xcite .
, which however is identical for our default choice of @xmath78 . ] for every beam recombination
, we assign a hardness @xmath79 .
we then work through the recombinations in order of decreasing hardness .
for an @xmath76 recombination ( or @xmath80 splitting ) , assuming @xmath60 is harder than @xmath61 , we mark @xmath81 as a born particle .
if fewer than @xmath40 particles have already been marked as born particle , we also mark @xmath68 as a born particle . for an @xmath75 recombination
, we mark @xmath67 as a born particle .
this is repeated until @xmath40 particles have been marked as born ( a particle may be marked more than once ; in such a case its marking counts only once ) .
as an example , in fig . [
fig : loopsim - diag ] , the hardest recombination will be between particle @xmath82 and the beam , so particle @xmath82 is marked as a born particle .
the next hardest recombination will that of @xmath83 with the beam
. therefore we mark @xmath84 as a born particle .
this exhausts the number ( @xmath41 ) of born particles that need to be marked . at the end of the above procedure
, every particle will have been marked as emitted either from the beam or from another particle , and some particles will also have been marked as secondary emitters and/or born particles .
thus in figure [ fig : loopsim - diag ] , particle @xmath71 is labelled as having been emitted from the beam , it is a secondary emitter and a born particle ; particle @xmath72 is labelled as having been emitted from particle @xmath71 ; particle @xmath82 is a born particle , emitted from the beam ; and particle @xmath85 was emitted from the beam .
the structure that we attribute is of course physically unambiguous only in the presence of strong ordering of emission angles and energies .
however , as we shall argue in section [ sec : loopsim - expected - accuracy ] , the mistakes that we make for non - ordered configurations should have a small impact for observables with giant @xmath0-factors .
once every particle is labelled in an event @xmath42 , one can compute the result of @xmath86 , which is a set of events @xmath87 .
for an event @xmath42 with respectively @xmath40 born particles and @xmath88 non - born secondary emitters , we define @xmath89 to be the maximum number of particles that will be allowed to become virtual in a given event .
it is obvious that born particles will not become virtual .
additionally , secondary emitters will also not become virtual . to understand why ,
consider the event @xmath90{figs / loopsim - secondary - explanation.eps } \end{varwidth}\ ] ] in which particle @xmath82 is a secondary emitter , since it emitted @xmath85 .
there is a divergence for @xmath85 to be collinear to @xmath82 only if @xmath82 is a final - state particle .
if instead @xmath82 is made virtual , then the divergence for emitting @xmath85 no longer exists ( there is no divergence for emission from internal lines in a diagram ) , so that the weight for the diagram in which @xmath82 is virtual would be the weight of the tree - level diagram times a small coefficient @xmath91 .
this simplest way of accounting for this is to approximate @xmath92 and thus not generate events in which secondary emitters are made virtual ( a more detailed discussion is given in appendix [ sec : secondary - emitters ] ) .
having understood which particles can be made virtual , the operator @xmath93 , when applied on an event @xmath42 , generates all the @xmath94 diagrams in which @xmath45 particles become virtual . for the virtual events to cancel the infrared and collinear divergences that appear in the tree - level diagram
, we need an infrared and collinear ( irc ) safe procedure to make particles virtual . for instance
, the divergent weight of an event with two collinear partons @xmath60 and @xmath61 has to be cancelled by that of corresponding virtual event ( @xmath61 makes a loop over @xmath60 ) when computing the distribution of any irc safe observable ; and two collinear partons , if not virtualised , have to remain collinear when another particle becomes virtual .
there are two ways for a particle @xmath61 to make a loop : * if it is labelled as clustering with particle @xmath60 , then one has to spread the momentum of particle @xmath61 over @xmath60 and all the particles that are labelled as clustering with it but which were emitted after @xmath61 according to the c / a clustering sequence ( i.e. at smaller angle ) .
the exact procedure is explained in detail in appendix [ app : recoil_procedure ] , and is designed to ensure that the recombination maintains any collinearity properties of non - looped particles and is invariant under longitudinal boosts .
when @xmath61 is the only particle that clusters with @xmath60 , then the procedure becomes equivalent to adding the momenta of particles @xmath60 and @xmath61 , @xmath95 , and then rescaling the momentum @xmath96 such that its mass is set to @xmath97 , while leaving its transverse components @xmath98 , @xmath99 and its rapidity unchanged . *
if particle @xmath61 is labelled as clustering with the beam , then when it is `` looped '' it is simply removed from the event . note that looping particles with the beam is less trivial than it may seem at first sight , because of an interplay with factorisation and the pdfs .
nevertheless it can be shown , appendix [ sec : incoming - partons ] , that for particle types that are included in the pdfs it does make sense to loop them .
a @xmath1 imbalance will result from the looping of particles with the beam , and so after all loops have been made , we apply a transverse boosts to all remaining event particles , conserving their rapidities , so as to bring the total transverse momentum to zero ( again , see appendix [ app : recoil_procedure ] ) .
there is some arbitrariness to our procedures for producing physical kinematics in the looped events .
one avenue for future work would be to examine the impact of making different choices . the operator @xmath100 has the following properties @xmath101 if @xmath102 is the weight of event @xmath42 , then each of the events generated by the @xmath103 operator has a weight @xmath104 once all the @xmath103 have been calculated for @xmath105 , one has to combine them in order to subtract all the soft and collinear divergences that appear in the calculation of @xmath42 and the virtual diagrams generated from it .
this is done by the operator @xmath106 , which is defined as @xmath107 it generates all the necessary looped configurations that have the same order in @xmath108 as the original tree - level diagram .
it is straightforward to see that the total weight of the diagrams obtained from the @xmath106 operator is @xmath97 . indeed ,
if we apply it to an event @xmath42 whose maximum number of virtual particles is @xmath109 , we get @xmath110 for @xmath111 .
we note that the above procedure for approximating loop diagrams does not generate the finite terms needed to cancel the scale - dependence of lower - order diagrams .
while it would be straightforward to include such terms , we believe that in the absence of full loop calculations , not including them helps ensure that the standard procedure of variation of renormalisation and factorisation scales is more likely to provide some form of reasonable estimate of the uncertainties on our results . in order to illustrate the action of the operator @xmath93 , we give below some simple examples in the pure glue case . in each of these examples
, only the born particles are labelled with numbers [ eq : u - b - l - series ] @xmath112{figs / ex_b201.eps } \end{varwidth } \right ) & = & -\ ; \begin{varwidth}{\textwidth }
\includegraphics[scale=0.35]{figs / ex_b202.eps } \end{varwidth } \;- \ ; \begin{varwidth}{\textwidth } \includegraphics[scale=0.35]{figs / ex_b203.eps } \end{varwidth } \ , , \\ \label{u_4_2_2 } u_{l=2}^{b=2 } \left ( \begin{varwidth}{\textwidth } \includegraphics[scale=0.35]{figs / ex_b201.eps } \end{varwidth } \right ) & = & \begin{varwidth}{\textwidth } \includegraphics[scale=0.35]{figs / ex_b204.eps } \end{varwidth } \ , , \\ \label{u_5_2_2_1 } u_{l=2}^{b=2 } \left ( \begin{varwidth}{\textwidth } \includegraphics[scale=0.35]{figs / ex_b2n501.eps } \end{varwidth } \right ) & = & \begin{varwidth}{\textwidth } \includegraphics[scale=0.35]{figs / ex_b2n502.eps } \end{varwidth } \;+ \ ; \begin{varwidth}{\textwidth } \includegraphics[scale=0.35]{figs / ex_b2n503.eps } \end{varwidth } \;+ \ ; \begin{varwidth}{\textwidth } \includegraphics[scale=0.35]{figs / ex_b2n504.eps } \end{varwidth } \ , , \\ \label{u_5_2_3 } u_{l=2}^{b=3 } \left ( \begin{varwidth}{\textwidth } \includegraphics[scale=0.35]{figs / ex_b301.eps } \end{varwidth } \right ) & = & \begin{varwidth}{\textwidth } \includegraphics[scale=0.35]{figs / ex_b302.eps } \end{varwidth } \ , , \\ \label{u_5_2_2_2 } u_{l=2}^{b=2 } \left ( \begin{varwidth}{\textwidth } \includegraphics[scale=0.35]{figs / ex_b205.eps } \end{varwidth } \right ) & = & \begin{varwidth}{\textwidth } \includegraphics[scale=0.35]{figs / ex_b208.eps } \end{varwidth } \ , , \\ \label{u_5_3_2 } u_{l=3}^{b=2 } \left ( \begin{varwidth}{\textwidth } \includegraphics[scale=0.35]{figs / ex_b205.eps } \end{varwidth } \right ) & = & \begin{varwidth}{\textwidth } 0 \end{varwidth } \ , .
\end{aligned}\ ] ] eq .
( [ u_4_1_2 ] ) gives an example of singly - looped configurations ( `` 1-loop diagrams '' ) generated by loopsim when studying the @xmath113 contributions to qcd dijet production .
( [ u_4_2_2 ] ) shows the `` 2-loop diagrams '' generated from the same event .
the next equation shows what happens if we add one more particle to the final state .
if , eq .
( [ u_5_2_3 ] ) , we now set the number of born particles for the same event to be @xmath82 , we obtain only one @xmath72-loop diagram instead of three , as represented in eq .
( [ u_5_2_3 ] ) . ) where the particle labelled @xmath82 in eq .
( [ u_5_2_3 ] ) is virtual , the event will resemble a two - jet event and so not pass the 3-jet cuts .
however if one studies the 3-jet cross section in a kinematic region where the cuts allow one of the jets to be much softer than the others , then to obtain sensible results it becomes necessary to use @xmath41 and include all diagrams on the right - hand side of eq .
( [ u_5_2_2_1 ] ) . ] finally , the last two examples of eq .
( [ eq : u - b - l - series ] ) give a case with a splitting : the emitter is not looped , even if it is not a born particle .
we also give a few examples of the action of the @xmath106 operator : @xmath114{figs / ex_b201.eps } \end{varwidth } \right ) & = & \begin{varwidth}{\textwidth } \includegraphics[scale=0.35]{figs / ex_b201.eps } \end{varwidth } -\ ; \begin{varwidth}{\textwidth } \includegraphics[scale=0.35]{figs / ex_b202.eps } \end{varwidth } \;- \ ; \begin{varwidth}{\textwidth } \includegraphics[scale=0.35]{figs / ex_b203.eps } \end{varwidth } \;+ \ ; \begin{varwidth}{\textwidth } \includegraphics[scale=0.35]{figs / ex_b204.eps } \end{varwidth } \,,\\
u_{\forall}^{2 } \left ( \begin{varwidth}{\textwidth } \includegraphics[scale=0.35]{figs / ex_b205.eps } \end{varwidth } \right ) & = & \begin{varwidth}{\textwidth } \includegraphics[scale=0.35]{figs / ex_b205.eps } \end{varwidth } -\ ; \begin{varwidth}{\textwidth } \includegraphics[scale=0.35]{figs / ex_b206.eps } \end{varwidth } \;- \ ; \begin{varwidth}{\textwidth } \includegraphics[scale=0.35]{figs / ex_b207.eps } \end{varwidth } \;+ \ ; \begin{varwidth}{\textwidth } \includegraphics[scale=0.35]{figs / ex_b208.eps } \end{varwidth } \,,\;\ ; \\
u_{\forall}^{2 } \left ( \begin{varwidth}{\textwidth } \includegraphics[scale=0.35]{figs / ex_b2secem01.eps } \end{varwidth } \right ) & = & \begin{varwidth}{\textwidth } \includegraphics[scale=0.35]{figs / ex_b2secem01.eps } \end{varwidth } -\ ; \begin{varwidth}{\textwidth } \includegraphics[scale=0.35]{figs / ex_b2secem02.eps } \end{varwidth } \ , .
\end{aligned}\ ] ] in the last case , only one particle can become virtual because there are two secondary emitters which can not be looped .
let us now examine some of the issues that arise if we are to extend the loopsim method to processes with quarks and vector bosons . and
@xmath115 may get recombined by the c / a algorithm . ]
we start with quarks and consider the situation depicted in fig .
[ fig : quark_quark_recombination ] . in this case , applying the c / a algorithm as in the previous section will lead to the recombination of the two quarks @xmath116 and @xmath115 , which is clearly not physical .
if flavour information is available for events , then one can veto on such a clustering , for instance by defining the clustering distance @xmath117 between two quarks to be infinite . as discussed in @xcite such a modification alone
is not sufficient to systematically guarantee sensible treatment of flavour in jet clustering .
@xcite have both discussed the further modifications needed in the case of the @xmath6 algorithm .
a proper handling of flavour within loopsim might seek to extend those modifications to the c / a algorithm .
however , neither of the nlo programs that we use , mcfm and nlojet`++ ` , provide information on particle flavours , so we defer such modifications to future work and just maintain the @xmath118 distance for all partons . for observables that are not flavour - sensitive this should not be a major drawback , given the observation @xcite that divergences associated with the mistreatment of flavour are strongly subleading .
were we to be interested in heavy tagged quarks , more careful treatment might well be needed .
note that there are also subtleties related to flavour and pdfs , discussed in appendix [ sec : incoming - partons ] .
what about non - qcd particles , specifically vector bosons ?
let us examine the case of z bosons .
a z can be emitted from quarks or antiquarks and we would like this to be reflected when establishing the approximate emission sequence , because if the z has been emitted from a quark , then that quark is a secondary emitter and should not be looped .
in other cases a z boson may be the hardest isolated object in an event .
then it is to be considered a born particle . on the other hand we wo nt necessarily wish to consider diagrams where a z boson is looped , because they would represent electroweak corrections , not qcd corrections .
one issue in dealing with electroweak particles is that they are not emitted from gluons .
if one could distinguish between quarks and gluons , then this could be accounted for during the c / a clustering , by defining the distance @xmath119 between a z and a gluon to be infinite .
since we will not know which partons are quarks or gluons , we adapt frixione s isolation procedure @xcite to decide if a z boson relatively close in angle to a parton @xmath60 is likely to have been emitted from @xmath60 .
more precisely , if @xmath120 then we define @xmath121 , otherwise @xmath122 .
when recombining @xmath60 and z into a particle @xmath63 , then the identity index @xmath65 is set equal to @xmath67 ( a quark and a z give a quark ) .
our procedure means that a z that is very collinear to a parton is always considered to be emitted from that parton this makes sense because such configurations are much more likely to occur when the parton is a quark .
in contrast a soft parton in the general vicinity of a z is not clustered with the z , which is sensible given that most soft partons tend to be gluons .
finally , for a recombination between a parton @xmath60 and z , we define the hardness of the branching @xmath123 as . ]
@xmath124 while a recombination of a z with the beam has a hardness @xmath125 the latter means that for an event with just a parton and a recoiling z boson , the parton s beam hardness will always be lower than the z s , implying that for @xmath126 it is the z - boson that will be the single born particle , as should be the case , at least when the parton has @xmath127 , i.e. in the kinematic regime that dominates the total cross section for the z. once a structure has been assigned to an event with a z boson , the next question is that of the looping procedure .
when looping partons it remains identical to before , with just a small extension of the recoil procedure in order to deal with decay products of the z boson ( see appendix [ app : recoil_procedure ] ) . in the situations where the z is
not a born particle ( it is never an emitter ) , straightforwardly following the procedure of section [ sec : looping - particles ] , one would deduce that one should loop the z as well : @xmath128{figs / loopsim - zg - looped01.eps } \end{varwidth } \right ) = -\ ; \begin{varwidth}{\textwidth } \includegraphics[scale=0.35]{figs / loopsim - zg - looped02.eps } \end{varwidth } \;- \ ; \begin{varwidth}{\textwidth } \includegraphics[scale=0.35]{figs / loopsim - zg - looped03.eps } \end{varwidth}\ ] ] ( straight lines are partons , either quarks or gluons ) . the rightmost diagram , with the looped z , is not , however , a qcd loop diagram : it is an electroweak loop correction to a multijet event .
the loopsim procedure does not aim to reproduce electroweak loop corrections ( though in this case it might be a reasonable approximation ) .
furthermore , in any analysis that tags on z bosons , such a diagram would not be tagged and so would not contribute .
thus , although the loopsim procedure naturally generates events with looped z bosons , events like the rightmost diagram of eq .
( [ eq : looping - z ] ) are simply to be discarded . for events with @xmath129 bosons ,
the same procedure can be used as for z s .
note , however , that while the `` looped '' z events may give a reasonable approximation to actual electroweak loop diagrams , looped @xmath129 events will not .
this is because @xmath130-boson emission changes quark flavour : consider a tree - level diagram @xmath131 , with the @xmath132 emitted collinearly off the incoming @xmath133 , converting it into a @xmath134 .
the loopsim procedure would give a `` loop '' diagram @xmath135 , with the @xmath132 looped .
however no such loop diagram exists and the correct loop diagram would instead involve @xmath136 .
, with the incoming gluon splitting collinearly to give @xmath137 , and the outgoing collinear @xmath138 being looped
. here , however , we are saved by the interplay between loopsim and pdfs , as discussed in appendix [ sec : incoming - partons ] .
it is crucial in this respect that all flavours that get looped are included in the pdfs . ]
this is closely related to the phenomenon of bloch - nordsieck violation @xcite that is found when considering electroweak double logarithms .
since we in any case discard events in which electroweak bosons are looped , this should not be a problem for the practical use of loopsim in events with @xmath129 bosons . before explaining how we merge exact higher orders calculations ,
let us mention how we use the loopsim method in practice on tree - level events at several different orders .
we introduce the notation x@@xmath139@xmath140lo to denote an approximation to the n@xmath140lo cross section for producing x , with all loop terms estimated through the loopsim procedure .
it is obtained by applying the @xmath106 operator to all tree - level diagrams that can contribute up to n@xmath140lo .
for instance , one can write [ eq : all - nlo - res ] @xmath141 notice that @xmath142 and @xmath143 .
the terms @xmath144 and @xmath145 simulate up to one - loop corrections , and @xmath146 simulates up to two - loop corrections .
now let us see how things work beyond tree - level accuracy .
we define @xmath147 to be a generic event at @xmath45 loops ( exactly calculated ) with @xmath34 particles in the final state .
we first consider the case where only one - loop corrections are computed exactly , so that we have tree - level events @xmath148 and exact one - loop events @xmath149 .
as before we can apply the unitarisation operator to the tree - level events , @xmath150 .
however , since we now include exact 1-loop contributions , @xmath149 , we must remove the approximate 1-loop contributions @xmath151 that are contained in @xmath150 .
this alone is not sufficient , because among the extra contributions from the exact 1-loop terms , there will be pieces that are finite for a given @xmath152-parton configuration , but that can lead to divergences when integrated over the @xmath152-parton phase space . to cancel these extra divergences , we should introduce additional approximate higher - loop contributions , which can be obtained by applying the unitarisation operator @xmath106 to the difference between the exact and approximate one - loop terms .
so , rather than including just events @xmath149 and subtracting @xmath151 , we include events @xmath153 and subtract @xmath154 .
it is convenient to express this through a new operator @xmath155 such that @xmath156 where the extra subscript @xmath71 on the @xmath155 indicates that it is the form to use when the exact 1-loop result is to be included .
the action of @xmath157 depends on the number of loops already included in the event on which it operates : we subtract the one - loop contribution returned by loopsim only in tree - level events . with this notation
, one can compute the higher order corrections to eqs .
( [ eq : all - nlo - res ] ) to one - loop accuracy , [ eq : nnlo - formulations ] @xmath158 where the `` only '' subscript on z+@xmath34j@nlo@xmath159 means that we take the highest order that contributes , i.e. here @xmath160 , since the lo , @xmath161 , piece of z+@xmath34j@nlo , is already taken into account in the z+@xmath152j@nlo contribution .
this implies that one should use consistent renormalisation and factorisation scale choices across all different orders of the calculation .
note that in eq .
( [ eq : nnlo - formulations ] ) we have introduced the notation @xmath139@xmath140n@xmath162lo to denote an approximation to the n@xmath163lo result in which the @xmath164 highest loop contributions have been approximated with loopsim .
the extension of the procedure beyond one - loop accuracy is simple .
for instance , at two - loop accuracy , one has to subtract the approximated two - loop contribution @xmath165 in eq .
( [ u_1_loop_accuracy_l0 ] ) , and the other approximated two - loop contribution @xmath166 in eq .
( [ u_1_loop_accuracy_l1 ] ) , giving @xmath167\,,\label{u_2_loop_accuracy_l0}\\ u_{\forall,2}^b(e_{n-1,1 } ) & = u_{\forall}^b(e_{n-1,1})-u_{\forall}^b\left(u_1^b(e_{n-1,1})\right)\,,\\ u_{\forall,2}^b(e_{n-2,2 } ) & = u_{\forall}^b(e_{n-2,2})\ , .
\end{aligned}\ ] ] therefore , once z+j@nnlo is calculated , one may compute for instance @xmath168 to be complete , let us mention the generalisation of our procedure to @xmath169-loop accuracy @xmath170 we noted at the end of section [ sec : looping - particles ] that the plain loopsim procedure does not generate the finite terms needed to cancel residual scale dependence from lower orders
. with the introduction of the exact loop contributions , those finite terms do now get included .
thus for a given number of exact plus simulated loops , as we increase the number of exact loops , we should expect to see reductions in scale uncertainties .
let us briefly explain why the loopsim method is expected to work in the presence of giant @xmath0-factors .
we consider an observable @xmath171 computed respectively at nlo and @xmath139 .
we define @xmath172 such that @xmath173 and we assume that @xmath174 .
this huge @xmath0-factor may come from logarithmic enhancements in the real nlo diagram or the appearance of new scattering channels in the perturbative expansion .
the computation of @xmath175 gives the exact real part of the nlo calculation as well as the divergent terms of the virtual correction .
therefore @xmath176 where , in writing @xmath177 , we mean that the term missing in the @xmath139calculation , the finite part of the 1-loop correction , is not especially enhanced .
this leads to @xmath178 the relative difference between the approximate and exact nlo calculations is thus suppressed by the inverse @xmath0-factor .
next , consider @xmath139accuracy .
the difference between @xmath179 and @xmath180 comes from the parts of the two - loop corrections that are finite and associated with the lo topology , so that they should be free of the enhancements that led to the large nlo @xmath0-factor .
this implies @xmath181 if we define @xmath182 such that @xmath183 , then we can write @xmath184 if @xmath174 , one can expect @xmath185 too .
given the novelty of the loopsim method , it is useful to have an alternative way of estimating the size of the nnlo contributions that we will approximate with loopsim .
here we outline such an alternative method , which , though less flexible than the loopsim approach , will provide a valuable cross - check and help us build our confidence in results of the loopsim method .
let us explain it for observables in the z+j process .
our aim is to estimate @xmath180 for some observable @xmath171 .
is the cross section for the observable @xmath171 to pass some given cuts ; it is only for brevity that we use here the somewhat inaccurate shorthand `` cross section for observable @xmath171 '' . ]
we assume that we have a reference observable which is identical to the observable @xmath171 at lo .
for instance , one might consider @xmath186 and @xmath187 .
we can write the nnlo z+j prediction for @xmath171 in terms of the nnlo prediction for the reference observable plus the nlo z+2j difference between @xmath171 and the reference cross section @xmath188 the second equality is possible because 2-loop nnlo corrections to z+j have the topology of z+j at lo
. therefore , their contributions to the observables @xmath171 and ref are identical and cancel in the difference in eq .
( [ eq : ref - meth - exact ] ) .
if we have reason to believe that the perturbative expansion for the reference observable converges well , we can conclude that @xmath189 is genuinely a small correction .
then @xmath190 i.e. we approximate the nnlo distribution for @xmath171 in terms of the nlo distribution for the @xmath191 observable and a nlo calculation for difference between the @xmath171 and @xmath191 distributions , both of which are exactly calculable . the missing part is suppressed by a relative factor @xmath192 , as for the loopsim method . for z+j ,
one can see from fig .
[ fig : lo_nlo_dist ] that @xmath12 seems to be an acceptable reference observable for @xmath193 and @xmath3 . in the sections that follow we shall
, for brevity , refer to the rhs of eq .
( [ eq : ref - nnlo - final ] ) as `` ref .
@xmath139 '' even though it does not quite adhere to our the meaning of @xmath139as set out in section 2 , i.e. in terms of the specific sets of tree - level and loop diagrams that are included exactly .
the cross section for the drell - yan process is known with exclusive final states up to nnlo accuracy @xcite . above a certain value of lepton transverse momentum ,
one finds giant corrections to the lepton @xmath1 spectra when going from lo to nlo and large ones from nlo to nnlo .
this gives us an opportunity to directly test the performance of the loopsim method by comparing its @xmath139results to exact nnlo spectra for lepton pair production . before examining @xmath139results
, it is useful to compare @xmath139with nlo . if they are in reasonable agreement for some observable , then that serves as a first indication that the loopsim estimate of missing loop corrections is sensible for that observable .
[ fig : dy - nlo ] gives the comparison of the @xmath139 , nlo and lo results for the production of an @xmath194 pair within the mass window of @xmath195 gev at a proton - proton centre of mass energy of @xmath196 tev .
the left - hand plot shows the cross section differential in the transverse momentum of the harder of the two leptons .
the right - hand plot gives the corresponding k factor with respect to lo .
the results were obtained with mcfm 5.3 @xcite , with its default set of electroweak parameters and nnlo mstw2008 parton distribution functions .
the uncertainty bands in fig .
[ fig : dy - nlo ] correspond to varying the renormalisation and factorisation scales @xmath197 by a factor of @xmath198 and @xmath72 around a default choice of @xmath199 . in the @xmath139result we fixed the value of the loopsim radius parameter to be @xmath200 , which naturally places interparticle and particle - beam clustering on the same footing ( though the @xmath139result here is actually independent of @xmath201 , because there is at most one isolated qcd parton in the final state ) .
there are three relevant regions of transverse momentum in fig .
[ fig : dy - nlo ] . for @xmath202 ( low @xmath1 )
the distribution is dominated by on - shell z - bosons and its shape is governed by the angular distribution of the z decays in their centre - of - mass frame .
the peak close to @xmath203 corresponds to z - bosons that decay in a plane at right - angles to the beam . for @xmath204 ( intermediate @xmath1 ) ,
the lo distribution comes from z - bosons that are off shell , which allows the @xmath1 of the lepton to be larger than @xmath203 .
the narrow width of the z causes the distribution to fall very steeply .
the @xmath205 upper edge of this region is a consequence of our cut on @xmath206 .
above @xmath205 ( high @xmath1 ) the lo distribution is zero . in the low @xmath1 region ,
the nlo correction is moderate and negative .
there is no strong reason to believe that the loopsim method should work here , but it turns out that the @xmath139result reproduces the structure of the correction , even if its scale dependence remains much larger than that of the nlo result ( this is because the loopsim procedure does not include the finite terms that would partially cancel the lo scale dependence ) . in the intermediate @xmath1 region , we see a `` giant '' nlo @xmath0-factor .
it comes about because initial - state radiation can give a boost to the z - boson , causing one of the leptons to shift to higher @xmath1 ( it becomes the `` max '' lepton ) .
the spectrum of qcd radiation falls much less steeply than the z - boson lineshape , so this nlo correction dominates over the lo result . in this region
the exact loop correction , proportional to the lo result , becomes almost irrelevant and we see near perfect agreement between @xmath139and nlo . in the high-@xmath1 region
only the real emission diagrams of z@nlo contribute and @xmath139becomes identical to nlo ( both correspond to the z+j@lo result ) .
similar results hold for the @xmath207 distribution , while the @xmath208 lacks the giant @xmath0-factor in the intermediate region . a similar comparison between @xmath139and nnlo spectra is shown in fig . [ fig : dy - nnlo ] .
the nnlo results were obtained with dynnlo 1.0 @xcite , used with a set of electroweak parameters compatible with that of mcfm . and @xmath209 contributions , dynnlo includes among its parameters a cut on the @xmath1 of the z boson .
the cut is applied to both real and virtual terms and its impact should vanish as it is taken towards zero .
it is , however , required to be non - zero for the numerical stability of the mcfm z+j nlo calculation that is among the components of dynnlo .
we set the cut equal to @xmath210 in the @xmath20 term and to @xmath211 in the @xmath209 term .
a related @xmath211 cut was placed on the @xmath209 piece of the @xmath139result ( while none was used at @xmath20 ) .
the impact of the @xmath211 cut is small but not entirely negligible close to the peak ( where , physically , nnlo should in any case be supplemented with a resummation ) . ] in the low-@xmath1 region we find quite good agreement between the @xmath139and nnlo results ( with somewhat larger uncertainty bands for @xmath139 ) . such a result was not guaranteed a priori , even if it is not entirely surprising given the reasonable agreement that we saw between @xmath139and nlo . in the intermediate @xmath1 region , where the nnlo / nlo corrections are substantial , the agreement is excellent .
this was expected . at high @xmath1
the agreement should be exact , and does seem to be , within statistical fluctuations .
the dependence on @xmath201 ( shown in the right - hand plot ) has been estimated by varying its value from 0.5 to 1.5 .
the effects are small .
finally , we note that similar features and a similar level of agreement between @xmath139and nnlo are to be found in the @xmath208 and @xmath207 distributions .
in the previous section , we studied the z production process and showed that our procedure correctly reproduces the @xmath1 distribution of the hardest lepton at nnlo , even , unexpectedly , in regions where the @xmath0-factor is not large . in this section
we study the z+j process , whose nnlo cross - section is not known yet , but which leads to giant @xmath0-factors at nlo for some observables as explained in the introduction .
therefore , their nnlo contributions are expected to be accurately described by the loopsim method . throughout this section
we use mcfm 5.7 , including the z+2j process at nlo @xcite , with the nlo cteq6 m pdfs .
we will take three different values for the renormalisation and factorisation scales : @xmath213 , @xmath214 and @xmath215 , with @xmath216 where @xmath13 is the transverse momentum of the hardest jet . at high @xmath1
, this scale choice should be quite similar to that used in @xcite and has the same @xmath1 scaling as those in @xcite .
the @xmath201 uncertainty is measured at @xmath217 using three different values for it : @xmath218 , @xmath71 , @xmath14 .
in addition to the 3 observables shown in the introduction , @xmath12 , @xmath13 and @xmath219 , we will also consider @xmath220 we only include events for which @xmath221 .
+ as a first investigation of the performance of the loopsim method , let us examine how the @xmath139approximation compares to the full nlo result .
[ fig : nlo_nlo ] shows the @xmath0-factors for the @xmath139and nlo predictions , with uncertainty bands from scale and @xmath201 variations . in the upper - left plot
, one sees that the @xmath139prediction for the @xmath12 distribution gives a somewhat smaller @xmath0-factor than the nlo result .
we interpret this as being because certain genuine loop effects are not taken into account by the loopsim method , for example those related to threshold logarithms , which depend crucially on the factorisation scheme of the parton distribution functions .
the @xmath139result does , however , reproduce the @xmath1 dependence of the @xmath0-factor , i.e. the dip towards @xmath222 .
this dip arises because of the requirement in our event selection that there should be at least one jet with @xmath8 . at lo
this induces a step - function in the @xmath12 distribution at @xmath223 . at nlo ,
soft and collinear emissions smoothen out that threshold and the @xmath139calculation correctly reproduces the resulting interplay between real and virtual terms . in the three remaining plots of fig .
[ fig : nlo_nlo ] , for @xmath13 , @xmath3 and @xmath224 , all of which have giant @xmath0-factors , one sees good agreement between the @xmath139and nlo results .
this is because the dominant nlo contribution comes from events in the b and c - type configurations of fig .
[ fig : z+jet - diags ] , for which there is no corresponding qcd loop correction .
the loopsim method merely serves to cancel the divergences that arise from soft and collinear emissions off @xmath171-type configurations and these are not dominant overall .
the @xmath201 dependence , also shown on these four plots , only comes from @xmath71-loop events generated by loopsim .
therefore , for an observable @xmath171 studied in z+j@@xmath139with two different values @xmath225 and @xmath226 for @xmath201 , one can write : @xmath227 as long as @xmath171 coincides with @xmath12 at lo ( it does for each of @xmath13 , @xmath3 and @xmath228 ) .
this means that the _ absolute _ uncertainty due to @xmath201 is the same for @xmath171 and @xmath12 .
therefore , the _ relative _ uncertainty due to @xmath201 is expected to be roughly inversely proportional to the @xmath0-factor for @xmath171 , in analogy with the discussion of sec .
[ sec : loopsim - expected - accuracy ] .
this explains why the @xmath201 dependence ( solid cyan band ) looks significantly smaller for @xmath13 , @xmath3 and @xmath224 than it does for @xmath12 plot .
+ results at @xmath139are given in fig .
[ fig : nnlo_nlo ] . in the case of @xmath12
the result is similar to the nlo result , and the scale uncertainties remain largely unchanged .
in other words , since z+2j topologies do not dominate the high-@xmath12 distribution , adding nlo corrections to them ( i.e. @xmath139z+j ) makes no difference either to the result or to the uncertainties .
we have also shown the dependence on the choice of @xmath229 in the loopsim procedure .
it is smaller than the scale dependence .
the @xmath13 distribution gets a correction that is just within the nlo uncertainty band , with @xmath139uncertainties that are about half the size of the nlo band .
adding in the @xmath139term has made a real difference .
this is precisely what we expect : the observable is dominated by z+2-parton configurations , and these were only present at tree - level in the nlo z+j calculation .
our use of @xmath139provides the additional 1-loop z+2-parton and tree - level z+3-parton configurations that come with nlo z+2j accuracy . given the improvement in scale uncertainty , we need to ask whether the uncertainty due to @xmath201 variation might somehow eliminate part of this benefit .
it is , however , small .
the reasons are similar to those given around eq .
( [ eq : rls - uncert ] ) .
the @xmath3 and @xmath224 distributions get significant @xmath139corrections , with @xmath139/nlo @xmath0-factors of about @xmath230 .
absolute scale uncertainties increase slightly compared to nlo , but because of the large @xmath0-factor , relative scale uncertainties diminish . at first sight , it is somewhat disturbing that the @xmath139and nlo uncertainty bands do nt overlap .
given the novelty of the loopsim method , one should therefore ask whether this is reasonable and whether there is any way of cross - checking the result .
a first observation is that since @xmath139z+j is really nlo of the dominant z+2j component , the large @xmath139corrections that we see are comparable to an @xmath231 @xmath0-factor for going from lo to nlo in the z+2j prediction .
there are many contexts where nlo and lo results are not compatible within scale uncertainties , and so it is not unreasonable that the same should be seen here .
still , we would like to have some more quantitative cross checks that our results are sensible .
one option is to consider the alternative `` reference - observable '' method presented in section [ sec : alternative_method ] , which only makes use of standard nlo calculations to compute the approximate nnlo corrections . the comparison between the two methods is shown in fig .
[ fig : z+j_nnlo_alt ] for @xmath3 and @xmath13 , where we have taken @xmath12 as the reference observable .
one notices near perfect agreement for @xmath3 and very good agreement for @xmath13 .
this gives us some degree of confidence that the @xmath139loopsim results provide an accurate description of the nnlo behaviour for these observables .
a second option for cross - checking the large @xmath139effects for @xmath3 and @xmath224 , is to examine whether @xmath232 type observables might generally be `` difficult '' .
to do so we look at them in the case of qcd jet events .
we have seen that the @xmath139@xmath0-factors for the two effective - mass variables , @xmath224 and @xmath3 , in @xmath212jet(s ) events are about a factor of two above the nlo @xmath0-factor . since nlo is the first order at which we see the dominant `` dijet '' topology for the @xmath233 variables in z+jet(s ) , fig .
[ fig : z+jet - diags]b , c , it might be instructive to establish a correspondence with a simpler process , qcd dijet production . having a nlo z+j prediction is analogous to a lo dijet prediction ; and the @xmath139z+j predictions should be analogous to nlo dijet predictions .
nlo cross sections for dijet observables can be calculated exactly and therefore we can check whether nlo @xmath0-factors of order @xmath72 appear for effective - mass observables in pure qcd events .
we will consider several effective - mass observables : an @xmath234 variable , which sums over the @xmath34 hardest jets above some threshold ( @xmath235 ; such a cut is often imposed experimentally we did not apply this kind of cut on the @xmath232 variables ; one purpose in applying it here is to ascertain whether the large higher - order effects persist even with it . ] ) @xmath236 where @xmath237 is the transverse momentum of the @xmath238 hardest jet .
upper limits on the number of jets included in the effective mass are common in susy searches @xcite .
we also define an effective mass for all jets above the @xmath239 threshold , @xmath240 which is similar to the @xmath3 and @xmath224 observables of section [ sec : results ] .
finally , for completeness we will consider the distributions of @xmath13 , @xmath241 and the inclusive jet spectrum .
all our results in this section will be for a centre - of - mass energy of @xmath242 , to allow comparison to results in the current run of the lhc . at lo ,
the distributions of @xmath243 ( @xmath244 ) , @xmath245 , @xmath13 , and @xmath241 will all be identical .
the inclusive jet spectrum will have a distribution that is twice as large ( because each of the two jets contributes ) .
note that we do not impose any rapidity acceptance limits on the jets : though such a cut would have been trivial to include in the loopsim procedure , it would have complicated somewhat the reference - observable approach that we will consider at the end of the section .
loopsim results with a rapidity cuts of @xmath246 on the jets are available from the authors on request . and
@xmath247 observables , at lo , where they are identical , and at nlo where they have substantially different @xmath0-factors .
right : the nlo @xmath0-factors for the @xmath248 bin for each choice of variable @xmath249 among the following : the inclusive jet spectrum , the @xmath1 distribution of the hardest ( @xmath13 ) and second hardest ( @xmath241 ) jets , ( half ) the effective mass of the two hardest jets ( @xmath250 ) , three hardest jets ( @xmath251 ) and of all jets above @xmath252 ( @xmath233 ) . also shown on the right are the @xmath139results for the @xmath0-factors .
the nlo and @xmath139(@xmath253 ) widths correspond to the uncertainty due to simultaneous renormalisation and factorisation scale variation by a factor of two around a central value @xmath254 .
the @xmath139(@xmath56 ) width shows the uncertainty from a variation of @xmath201 in the range @xmath255 .
, title="fig:",scaledwidth=47.0% ] and @xmath247 observables , at lo , where they are identical , and at nlo where they have substantially different @xmath0-factors .
right : the nlo @xmath0-factors for the @xmath248 bin for each choice of variable @xmath249 among the following : the inclusive jet spectrum , the @xmath1 distribution of the hardest ( @xmath13 ) and second hardest ( @xmath241 ) jets , ( half ) the effective mass of the two hardest jets ( @xmath250 ) , three hardest jets ( @xmath251 ) and of all jets above @xmath252 ( @xmath233 ) . also shown on the right are the @xmath139results for the @xmath0-factors .
the nlo and @xmath139(@xmath253 ) widths correspond to the uncertainty due to simultaneous renormalisation and factorisation scale variation by a factor of two around a central value @xmath254 .
the @xmath139(@xmath56 ) width shows the uncertainty from a variation of @xmath201 in the range @xmath255 .
, title="fig:",scaledwidth=38.0% ] figure [ fig : jet - dist+nlo - kfacts](left ) shows the distributions for two observables , @xmath247 and @xmath256 at lo ( where they are identical ) and at nlo , as determined using nlojet`++ ` @xcite with cteq6 m pdfs .
a first comment is that @xmath232 receives a nlo @xmath0-factor of order @xmath72 , just like the @xmath139enhancements in the z+j case .
this provides supporting evidence as to their legitimacy .
a second comment is that the cross sections are large : these observables will be easily accessible with a few pb@xmath257 of integrated luminosity at a @xmath242 lhc , allowing for an early experimental verification of the large @xmath0-factor for @xmath232 . the other observable in the left - hand plot of fig . [
fig : jet - dist+nlo - kfacts ] , @xmath241 , has a very different @xmath0-factor , somewhat below @xmath71 .
the right - hand plot shows the nlo @xmath0-factors for our full range of observables , focusing on a single bin of the left - hand one , from @xmath258 .
the pattern that we see here allows us to make some deductions .
firstly , the @xmath250 variable , which sums the @xmath1 s of the two leading jets , is free of large nlo enhancements .
it is the addition of the third jet in @xmath251 and @xmath233 that brings about the enhancement .
a natural interpretation is the following : it is common for a third , soft jet to be present due to initial state radiation .
this third jet shifts the @xmath233 distribution to slightly larger values , and because the distribution falls very steeply , that leads to a non - negligible enhancement .
this suggests that if , in section [ sec : results ] , we had used effective mass observables with at most two objects in the sum , then the @xmath139/nlo ratios would have been close to @xmath71 .
we have verified that this is indeed the case . the pattern for @xmath259 and @xmath256 in fig . [ fig : jet - dist+nlo - kfacts ] can also be explained in similar terms : a soft isr emission boosts the hard dijet system , breaking the degeneracy between the @xmath1 s of the two hardest jets .
it is jet @xmath71 that shifts to larger @xmath1 ( giving a @xmath0-factor @xmath260 ) , while jet @xmath72 shifts to lower @xmath1 and so it gets a @xmath0 factor below @xmath71 . for the inclusive jet spectrum , and for @xmath250 ,
this effect balances out .
in addition , final - state radiation from one of the jets can cause it to shift to lower @xmath1 ( becoming the 2nd jet ) , further reducing the @xmath0-factor for the distribution of @xmath241 . of the different variables ,
it is only the inclusive jet @xmath1 and @xmath250 for which there is a clear reduction in scale uncertainty in going from lo to nlo .
figure [ fig : jet - dist+nlo - kfacts](right ) also shows the @xmath139results ( including uncertainties both from scale variation and from the loopsim parameter @xmath201 ) . despite the fact that none of the @xmath0-factors is parametrically large ( except arguably for @xmath251 and @xmath233 ) , the @xmath139results are remarkably effective at reproducing the pattern of nlo @xmath0-factors , albeit with a small systematic shift and generally larger scale uncertainties .
one can also verify that , to within @xmath2 , the @xmath1 dependence of the nlo @xmath0-factors is reproduced at @xmath139 . given this success of @xmath139 , and the observed limited convergence of some of the observables at nlo , it is interesting to examine what happens at @xmath139 , where the additional 3j@nlo contribution that we require is again obtained using nlojet`++ ` .
results are shown in fig .
[ fig : jets - k - nnlo ] . and nlo @xmath0-factors relative to the lo predictions , as a function of @xmath1 ( or @xmath247 , etc . ) , for the collection of jet observables described in the text . ] for the inclusive jet spectrum and @xmath250 ,
which already saw large reductions in scale - dependence at nlo , the @xmath139corrections have essentially no meaningful effect : they neither significantly affect the central values , nor reduce the scale uncertainties .
for these observables , nlo already converged well , and adding a subset of the nnlo corrections without the 2-loop part can not improve the result . for the other effective mass observables ,
the situation is quite different . with @xmath251 ,
the @xmath139result is close to the nlo result and the scale uncertainty is much reduced , i.e. this observable seems to come under control at @xmath139 .
in contrast , @xmath233 is subject to quite a large further correction , with the central value at @xmath139lying outside the nlo uncertainty band , and the @xmath139uncertainty band ( dominated by scale variation ) only marginally smaller than at nlo .
why is this ?
perhaps we are seeing the effect of a second isr emission , which shifts the @xmath232 distribution to even higher values ?
given that @xmath251 converges and @xmath232 does not , such an explanation is not unattractive .
it is also consistent with the decrease in @xmath0-factor at low @xmath233 , where the @xmath252 @xmath1 cutoff on the jets contributing to the @xmath233 sum will eliminate the isr enhancement .
a definitive conclusion would however probably require further study . for the remaining two observables , @xmath259 and @xmath256
, the @xmath139contribution goes in the opposite direction from the nlo correction and at low @xmath1 it seems that the series fails to converge .
this is , we believe , closely related to observations of insufficiencies of nlo predictions for dijet cross sections in dis and photoproduction when identical @xmath1 cuts are imposed on both jets @xcite ( equivalent to integrating the @xmath261 distribution above that cut ) .
the worse convergence at low @xmath1 is probably due to the larger fraction of subprocesses that involve gluons in the underlying @xmath25 scattering , so that perturbative corrections tend to go as @xmath262 rather than as @xmath263 at higher @xmath1 . considering that we do not have giant nlo @xmath0-factors for the jet processes shown here
, one may question the validity of the information obtained from the loopsim procedure .
an important cross check comes from a comparison with the reference - observable technique . examining fig .
[ fig : jet - dist+nlo - kfacts ] ( right ) , one sees two natural reference observables : the inclusive jet spectrum and @xmath250 , both of which show `` perturbative '' @xmath0-factors and small scale dependence at nlo . here
we will use ( half ) the inclusive jet spectrum as the reference observable ( results with @xmath250 would be almost identical ) .
results with those from the reference - observable method , here using the inclusive jet @xmath1 spectrum as the reference observable . in the loopsim results ( labelled @xmath139 )
, the uncertainty bar spans the envelope of the scale and @xmath201 uncertainties .
the results are for the @xmath264 bin for each observable @xmath249 , as in fig .
[ fig : jet - dist+nlo - kfacts ] .
, scaledwidth=38.0% ] figure [ fig : jets - nnlo - ref ] provides a comparison of the loopsim @xmath139results ( showing the envelope of the scale and @xmath201 uncertainties ) with the reference - observable @xmath139results .
the comparison is given for all observables except the reference observable itself .
the agreement between the two methods is striking , with the reference - observable method giving just a small shift of the @xmath0-factors relative to the loopsim results .
the shift is identical for all the observables , as it has to be : it is simply equal to the difference between the nlo and @xmath139results for the reference observable .
insofar as we believe the scale dependence to be representative of the true nlo uncertainty on the inclusive jet spectrum , uncertainty for the inclusive jet spectrum is larger than the nlo uncertainty , it may be that our symmetric scale variation is underestimating somewhat the uncertainties present at nlo . to be conservative ,
it might have been safer to vary the renormalisation and factorisation scales independently . ]
the results for the other observables should therefore be good approximations to the full nnlo results .
several cases of lhc observables with giant nlo @xmath0-factors have come to light in recent years .
they are characterised by the presence at nlo of new partonic scattering topologies that have large enhancements over the lo topologies . in these cases , nlo calculations , while important in highlighting the presence of the large @xmath0-factors , can not on their own provide accurate predictions . in this article
we have examined how to address this problem by combining nlo results for different multiplicities , for example z+j@nlo with z+2j@nlo .
our main , most flexible method , loopsim , makes use of unitarity to cancel the infrared and collinear divergences that appear when one tries , say , to apply z+2j@nlo calculations to observables that are non - zero starting from z+1-parton .
we referred to the result as z+j@@xmath139 , where the `` @xmath139 '' indicates that the highest loop contribution to the nnlo result ( the two - loop part ) has been estimated with loopsim . in introducing a new approximate method for estimating nnlo corrections
, significant evidence needs to be provided that the method is meaningful .
firstly , we gave reasons why , in cases with giant @xmath0-factors associated with new nlo topologies , we expect @xmath139results to be a good approximation to nnlo results .
as a next step , we carried out studies comparing z/@xmath265@@xmath139(dy ) to nnlo predictions for the @xmath266 process . in comparing the dy lepton @xmath1 @xmath139distributions to nnlo
we found near - perfect agreement in a region of giant @xmath0-factors , @xmath267 .
interestingly , even in the region where the nlo @xmath0-factor was not large , @xmath268 , the @xmath139results provided a significantly better approximation to nnlo than did the plain nlo result .
this need not always be the case , but is , we believe , connected to the observation that our @xmath139results reproduced much of the structure seen at nlo ( recall , z@@xmath139means combining z@lo with z+j@lo ) . for z+j production , the first step of our validation procedure was to compare @xmath139and nlo results .
all observables with giant @xmath0-factors showed good agreement between the two ( one with a moderately large @xmath0-factor did not ) . for those observables , @xmath139always appeared to provide extra information : either suggesting a convergence of the perturbative series , with reduced scale uncertainties ( for @xmath13 ) , or an indication of substantial further higher order corrections ( for the effective - mass type observables @xmath3 and @xmath224 ) .
almost identical results were seen with our alternative `` reference - observable '' estimate of the nnlo contribution .
the large @xmath139corrections that we saw for effective mass observables led us to examine a range of effective - mass and jet observables in the simpler context of pure jet events ( with the expectation that z+j@nnlo might be similar to 2j@nlo ) .
there we saw a significant nlo @xmath0-factor for all effective mass variables except one , @xmath250 , which summed over just the two leading jets . in the z+j case
we had summed over all jets and hence it is not surprising that we should have observed substantial @xmath139/nlo ratios .
even though the observables in the pure jets case did not display giant @xmath0-factors , the pattern of nlo results was remarkably well reproduced at @xmath139 .
this encouraged us to study @xmath139predictions , which provided substantial extra information for several of the observables , with the reference - observable method again giving important cross checks .
since the cross sections for the jet observables are large , these results could easily be tested with early lhc data .
we close this article with a few lines on the relation between loopsim and other predictive methods .
there is a close connection between @xmath139(or @xmath139@xmath139 ) and ckkw and mlm @xcite matching , since they also both provide ways of combining tree - level results with different multiplicities .
of course ckkw and mlm matching provide an interface with parton showers too , which the loopsim method does not . on the other hand
it is significantly easier to include multiple loop orders into the loopsim method than it is within matrix - element / parton - showering matching procedures ( though work is ongoing in this direction see e.g. @xcite ) .
an interesting cross - check of the loopsim method will come with the completion of the nnlo calculations for the z+j and dijet processes . at that point
the method could also , for example , be used to merge z@nnlo with z+j@nnlo , so as to provide an @xmath139prediction for quantities like the drell - yan lepton @xmath1 spectrum .
the value of the loopsim method also goes hand - in - hand with progress on 1-loop calculations , especially with the prospect of automated nlo calculations now on the horizon ( for example @xcite ) .
note that the loopsim code , which will be made public in due course , can currently only deal with hadron - collider processes involving any number of light partons and up to one vector boson .
it would benefit from further work to appropriately include heavy quarks and additional bosons .
we are grateful to fabio maltoni and simon de visscher for stimulating discussions and for sharing results with us on the behaviour of mlm type matching for the z+jet processes , to giulia zanderighi for insightful comments and a careful reading of the manuscript , to massimiliano grazzini and giancarlo ferrera for assistance with dynnlo , to matteo cacciari and gregory soyez for fastjet development work that has been useful in implenting the loopsim procedure and to philippe schwemling and dirk zerwas for helpful exchanges about @xmath232 definitions .
gps wishes to thank the high energy theory group at rutgers university , the princeton university physics department and the aspen center for physics for hospitality during various stages of this work .
this work was supported by the french agence nationale de la recherche , under grant anr-09-blan-0060 .
in this appendix , we provide further details on how we perform the recoil of an event when a particle becomes virtual , including the treatment of the decay products of the z boson . we first examine the simpler case of a particle that makes a loop with the beam , then we show how to deal with a particle that makes a loop with another particle .
let us assume that particle @xmath269 makes a loop with the beam . to balance the event moment , we follow the following procedure : 1 . for each particle @xmath270 , store its rapidity @xmath271 .
2 . perform a separate longitudinal boost on each particle so as to bring its rapidity to @xmath97 ( i.e. get a purely transverse event ) .
3 . compute @xmath272 where @xmath273 is the energy of particle @xmath60 in the purely transverse event .
4 . define @xmath274 and boost all particles into the rest frame of @xmath63 ( so that the total transverse momentum balances ) .
perform a longitudinal boost on each particle so that it recovers its original rapidity @xmath271 . for the case
where two particles , @xmath269 and @xmath275 , are looped with the beam , replace @xmath270 with @xmath276 and in eq .
( [ eq : pt - balance - boost ] ) replace @xmath277 with @xmath278 , etc . in the case where the z decays , for instance into @xmath72 leptons ,
the procedure is identical except that we apply to the leptons the same longitudinal boosts as for the z ( the rapidity of the leptons is thus not necessarily @xmath97 when we apply the transverse boost ) .
this conserves the property that the sum of the leptons momenta is still the z momentum in the `` looped '' event .
the logic of the above procedure is that if we had attempted to apply a transverse boost without stages 2 and 5 , we would have found that our choice of transverse boost , and the corresponding mapping of high-@xmath1 particles momenta , would be affected by the presence of energetic particles collinear to the beam .
this would have made the procedure collinear unsafe .
is the last to be clustered with the quark ( which roughly corresponds to an early time emission ) and gluon @xmath85 is the first to be clustered . in the event where gluon @xmath72 makes a loop over the quark
, we spread the gluon @xmath72 s momentum over the quark s momentum and the momenta of gluons that were emitted _ after _ it , i.e. gluons @xmath82 and @xmath85 ( an earlier time emission like gluon @xmath71 can not be affected ) . ]
let us consider the situation depicted in fig .
[ fig : pp - recombination ] : four gluons are emitted from the same quark , but at different angles : @xmath279 and gluon @xmath72 becomes virtual .
the virtualisation of gluon @xmath72 over the quark can not have an impact on gluon @xmath71 , which was emitted earlier in an angular - ordered picture .
but it has an impact on gluons @xmath82 and @xmath85 .
more precisely , let the @xmath280 be the momenta in the original event and @xmath281 the momenta in the event where gluon @xmath72 is virtual .
we define @xmath282 and then set the @xmath283 as follows : @xmath284 subsequently each particle s @xmath283 momentum is adjusted such that its mass is @xmath97 ( or @xmath199 if gluon @xmath82 is a z boson rather than a gluon ) , keeping its transverse components @xmath98 , @xmath99 and its rapidity unchanged .
this can be easily generalised to any number of particles recombining with the same hard one : for each recombined particle @xmath60 , we spread the looped particle over the hard particle @xmath285 and over any non - looped emissions from @xmath285 that are at smaller angle ( i.e. earlier in the c / a clustering sequence ) than @xmath60 . in eqs .
( [ eq : reshuffle - looped - momenta - tot],[eq : reshuffle - looped - momenta ] ) it is always the original particle momenta that are used to determine the @xmath286 ratio , so that the result is independent of the order in which we perform the recombinations .
this procedure is designed to ensure collinear safety : if , for instance , gluon @xmath85 is collinear to the quark in the original event , then it remains collinear in the looped event . and
if it is gluon @xmath85 ( emitted after all the others ) that is looped , only the quark momentum is rescaled and its direction barely changes , so that angles between the quark and the other gluons stays the same . in the case where the z decays into @xmath72 leptons , one applies the following procedure to each of the leptons : 1 . perform a longitudinal boost of the z boson respectively in the original event and the looped event such that it has @xmath97 rapidity in each case .
call the momenta obtained @xmath287 and @xmath288 respectively .
2 . perform a longitudinal boost of the lepton from the original event into the frame where the initial z has @xmath97 rapidity .
3 . define a purely transverse vector @xmath63 such that @xmath289 is transformed to @xmath290 if it is boosted into @xmath63 s rest frame : @xmath291 with @xmath292 4 .
boost the lepton s momentum into @xmath63 s rest frame .
apply to the lepton the longitudinal boost that brings @xmath290 to its true rapidity in the looped event .
we are aware of the cumbersome nature of these procedures .
a simplification of them that retained the relevant collinear - safety properties would certainly be of interest .
without going into a full proof , we shall here illustrate why the loopsim method is sensible even in the presence of incoming hadrons , by considering what happens at @xmath139 .
we start with a lo cross section for a process producing @xmath34 hard objects @xmath293 for compactness of notation , we have dropped the @xmath294 dependence in the differential tree - level partonic cross section @xmath295 .
we have also not yet specified our choice for the factorisation scale @xmath296 .
we assume that @xmath297 contains the necessary constraints to relate the incoming partonic momenta to the outgoing momenta .
we further integrate over the phase - space @xmath298 , and include a function @xmath299 , which is @xmath71 if the momenta pass our cuts and @xmath97 otherwise .
we now imagine that there is some transverse - momentum scale @xmath300 below which no radiation is emitted . to @xmath20
, the pdfs @xmath301 can be written in terms of pdfs at scale @xmath300 : @xmath302 where we sum implicitly over repeated indices .
we also define an unregularised splitting function @xmath303 such that @xmath304 , with @xmath305 embodying the virtual parts of the splitting function ( it is zero for @xmath306 ) .
next , we write the lo cross section in terms of a pdf for proton @xmath307 that has been evaluated at scale @xmath300 : @xmath308\,.\end{gathered}\ ] ] note that the first term in round brackets on the second line corresponds to real emission of a parton .
however that parton is not taken into account in the @xmath309 factor .
next we examine the structure of the @xmath139contribution , @xmath310\,,\end{gathered}\ ] ] where the @xmath311 represent the momenta when the loopsim procedure has looped @xmath312 . in the limit
in which @xmath312 is collinear to incoming parton @xmath60 , with momentum @xmath313 , the @xmath314-parton differential cross section and phase - space simplify @xmath315 where @xmath316 . by `` collinear '' we will mean @xmath317 where @xmath318 is the momentum transfer in the hard process .
in this limit we also have that @xmath319 ( for @xmath320 ) .
so , still working within the collinear limit , we can now rewrite eq .
( [ eq : pdf - nlo - start ] ) as @xmath321 f_{i / a}(x_a'/z , { \mu_\text{\sc f}}^2)\,.\end{gathered}\ ] ] next , we exchange @xmath322 , replace @xmath323 and change the scale @xmath324 in @xmath325 to be @xmath326 , which is allowed because it corresponds to an @xmath209 change ( while here we consider only @xmath20 ) : @xmath327 f_{i / a}(x_a / z , q_0 ^ 2)\,.\end{gathered}\ ] ] note now that if we take @xmath328 in eq .
( [ eq : pdf - lo - q0 ] ) , then the second term in square brackets in eq .
( [ eq : pdf - nlo - end ] ) cancels the first term in round brackets in the second line of eq .
( [ eq : pdf - lo - q0 ] ) . in other words for initial - state radiation ,
the action of loopsim is not so much to provide virtual corrections as to cancel the real - emission terms already included implicitly through the pdfs in the leading order cross section .
in contrast , the true virtual terms are already included through the pdfs themselves , i.e. through the second term in round brackets in eq .
( [ eq : pdf - lo - q0 ] ) . as an example , consider @xmath329 . at @xmath330
we will have events such as @xmath331 , where the outgoing quark comes from collinear initial - state splitting @xmath332 , with an underlying hard subprocess @xmath333 . from these events
loopsim will generate a configuration in which the outgoing quark is `` looped '' .
this will come in with a pdf weight that is the product of a gluon distribution and a quark distribution , so it appears that we have a ( negative ) @xmath334 contribution , which would be unphysical .
however in the lo cross section with a factorisation scale @xmath335 , when we write @xmath333 , part of the @xmath336 pdf comes from @xmath337 splitting . if we were just to add the real @xmath331 diagram to the lo cross section alone , then in the collinear limit we would be double counting the part already included in the pdf . with the negative `` @xmath334 '' loopsim contribution ,
what happens is that we simply remove the @xmath336 pdf component , generated from @xmath338 splitting , that was implicitly included at lo with an incorrect final state ( i.e. lacking an outgoing quark ) , since we are now putting it in with the correct final state through the real @xmath339 diagram .
note that we have not yet worked out the full extension of this discussion to higher orders .
the details would depend on the precise higher orders that we have in mind , for example @xmath139@xmath139versus @xmath139 .
however , regardless of these details , the fundamentally unitary nature of the loopsim procedure is important in ensuring that the simulated `` loops '' simply bring about an overall consistent set of final states while maintaining the total cross section as calculated with a sensible factorisation scale choice .
in section [ sec : looping - particles ] we discussed the special treatment needed for `` secondary emitters '' , i.e. non - born particles that have emitted something . in our procedure
, secondary emitters do not get looped : when particle @xmath61 makes a loop over @xmath60 , this is justified by the collinear enhancement of the matrix element due to @xmath60 and @xmath61 being close in angle . but the emission of the same @xmath61 from the configuration where @xmath60 is virtual does not have such a collinear enhancement , so one must not take it into account .
another way to understand it is to consider the example of @xmath72-gluon emission from a @xmath340 dipole .
the squared matrix element for the emission of @xmath72 real energy - ordered ( @xmath341 ) gluons , @xmath342 , @xmath343 , can be expressed as @xcite @xmath344 with since the @xmath346 term diverges when @xmath343 is collinear to @xmath347 or @xmath348 ( unlike the @xmath349 term ) , it becomes relevant when @xmath343 is considered to have been emitted from @xmath347 or @xmath336 independently of @xmath342 .
the @xmath349 term diverges when @xmath343 is collinear to @xmath342 ( unlike @xmath346 ) , so it becomes relevant when @xmath343 is considered to have been emitted from @xmath342 .
this is depicted in fig .
[ fig : two_gluons_emission ] , which also shows the virtual corrections ( cf .
one notices that the @xmath349 term only appears when @xmath342 is real .
the diagrams where @xmath342 is virtual are taken into account when @xmath343 is emitted from @xmath347 or @xmath336 .
therefore , @xmath342 can not become virtual when @xmath343 makes a loop over it .
dipole : ( a ) gluons @xmath71 and @xmath72 real ; ( b ) gluon @xmath71 real and @xmath72 virtual ; ( c ) gluon @xmath71 virtual and @xmath72 real ; ( d ) gluons @xmath71 and @xmath72 virtual . in each case , when needed , we use the decomposition into @xmath346 and @xmath349 pieces to separate what can be seen as the emission of gluon @xmath72 from gluon @xmath71 , and what can be seen as the emission of gluon @xmath72 directly from @xmath340 . ]
i. hinchliffe , f. e. paige , m. d. shapiro , j. soderqvist and w. yao , phys .
d * 55 * , 5520 ( 1997 ) [ arxiv : hep - ph/9610544 ] . s. abdullin _ et al . _ [ cms collaboration ] , j. phys .
g * 28 * , 469 ( 2002 ) [ arxiv : hep - ph/9806366 ] .
g. aad _ et al . _
[ the atlas collaboration ] , arxiv:0901.0512 [ hep - ex ] .
g. l. bayatian _ et al . _
[ cms collaboration ] , j. phys .
g * 34 * , 995 ( 2007 ) .
m. spiropulu , arxiv:0801.0318 [ hep - ex ] .
t. yamazaki [ atlas collaboration and cms collaboration ] , arxiv:0805.3883 [ hep - ex ] .
s. yamamoto [ atlas collaboration ] , arxiv:0710.3953 [ hep - ex ] .
m. l. mangano , eur .
j. c * 59 * ( 2009 ) 373 [ arxiv:0809.1567 [ hep - ph ] ] .
z. bern _ et al . _
[ nlo multileg working group ] , arxiv:0803.0494 [ hep - ph ] .
j. r. andersen _ et al . _ [ sm and nlo multileg working group ] , arxiv:1003.1241 [ hep - ph ] .
j. m. campbell , f. maltoni and f. tramontano , phys .
lett . * 98 * ( 2007 ) 252002 [ arxiv : hep - ph/0703113 ] .
a. banfi , g. p. salam and g. zanderighi , jhep * 0707 * ( 2007 ) 026 [ arxiv:0704.2999 [ hep - ph ] ] .
j. m. butterworth , a. r. davison , m. rubin and g. p. salam , phys .
* 100 * ( 2008 ) 242001 [ arxiv:0802.2470 [ hep - ph ] ] . c. w. bauer and b. o. lange , arxiv:0905.4739 [ hep - ph ] .
a. denner , s. dittmaier , t. kasprzik and a. muck , arxiv:0906.1656 [ hep - ph ] . c. oleari , presentation at the mc4lhc readiness workshop , cern , geneva , march 2010 [ http://indico.cern.ch/getfile.py/access?contribid=46&sessionid=3&resid=0&materialid=slides&confid=74601 ] .
f. febres cordero , l. reina and d. wackeroth , phys .
d * 80 * ( 2009 ) 034015 [ arxiv:0906.1923 [ hep - ph ] ] .
t. binoth , t. gleisberg , s. karg , n. kauer and g. sanguinetti , phys .
b * 683 * ( 2010 ) 154 [ arxiv:0911.3181 [ unknown ] ] .
j. m. campbell and r. k. ellis , phys .
d * 60 * ( 1999 ) 113006 [ arxiv : hep - ph/9905386 ] ; + j. campbell , r.k .
ellis , et al .
m. cacciari , g. p. salam and g. soyez , jhep * 0804 * ( 2008 ) 063 [ arxiv:0802.1189 [ hep - ph ] ] .
p. ciafaloni and d. comelli , jhep * 0609 * ( 2006 ) 055 [ arxiv : hep - ph/0604070 ] .
e. w. n. glover and j. pires , arxiv:1003.2824 [ hep - ph ] .
p. artoisenet , j. m. campbell , j. p. lansberg , f. maltoni and f. tramontano , phys .
* 101 * , 152001 ( 2008 ) [ arxiv:0806.3282 [ hep - ph ] ] .
s. catani , f. krauss , r. kuhn and b. r. webber , jhep * 0111 * ( 2001 ) 063 [ arxiv : hep - ph/0109231 ] .
s. catani , y. l. dokshitzer , m. olsson , g. turnock and b. r. webber , phys .
b * 269 * , 432 ( 1991 ) ; s. d. ellis and d. e. soper , `` successive combination jet algorithm for hadron collisions , '' phys . rev .
d * 48 * , 3160 ( 1993 ) [ hep - ph/9305266 ] .
a. banfi , g. p. salam and g. zanderighi , eur .
j. c * 47 * , 113 ( 2006 ) [ arxiv : hep - ph/0601139 ] .
s. hoeche , f. krauss , s. schumann and f. siegert , jhep * 0905 * ( 2009 ) 053 [ arxiv:0903.1219 [ hep - ph ] ] .
s. frixione , phys .
b * 429 * ( 1998 ) 369 [ arxiv : hep - ph/9801442 ] .
m. ciafaloni , p. ciafaloni and d. comelli , phys .
* 84 * ( 2000 ) 4810 [ arxiv : hep - ph/0001142 ] .
k. melnikov and f. petriello , phys .
d * 74 * ( 2006 ) 114017 [ arxiv : hep - ph/0609070 ] . s. catani , l. cieri , g. ferrera , d. de florian and m. grazzini , phys .
* 103 * ( 2009 ) 082001 [ arxiv:0903.2120 [ hep - ph ] ] . j. m. campbell and r. k. ellis , phys .
d * 62 * ( 2000 ) 114012 [ arxiv : hep - ph/0006304 ] .
s. catani and m. grazzini , phys .
* 98 * ( 2007 ) 222002 [ arxiv : hep - ph/0703012 ] .
m. grazzini , http://theory.fi.infn.it/grazzini/dy.html s. frixione and g. ridolfi , nucl .
b * 507 * ( 1997 ) 315 [ hep - ph/9707345 ] .
s. chekanov _ et al . _
[ zeus collaboration ] , eur .
j. c * 23 * ( 2002 ) 13 [ arxiv : hep - ex/0109029 ] .
_ et al . _
[ h1 collaboration ] , eur .
j. c * 33 * ( 2004 ) 477 [ arxiv : hep - ex/0310019 ] .
a. banfi and m. dasgupta , jhep * 0401 * ( 2004 )
027 [ arxiv : hep - ph/0312108 ] .
j. alwall _ et al .
_ , eur .
j. c * 53 * ( 2008 ) 473 [ arxiv:0706.2569 [ hep - ph ] ] . n. lavesson and l. lonnblad , jhep * 0812 * ( 2008 ) 070 [ arxiv:0811.2912 [ hep - ph ] ] .
r. k. ellis , k. melnikov and g. zanderighi , arxiv:0906.1445 [ hep - ph ] . c. f. berger _
_ , arxiv:0907.1984 [ hep - ph ]
. g. bevilacqua , m. czakon , c. g. papadopoulos and m. worek , arxiv:1002.4009 [ hep - ph ] .
a. bassetto , m. ciafaloni and g. marchesini , phys .
rept . * 100 * , 201 ( 1983 ) .
f. fiorani , g. marchesini and l. reina , nucl .
b * 309 * , 439 ( 1988 ) .
y. l. dokshitzer , g. marchesini and g. oriani , nucl .
b * 387 * ( 1992 ) 675 . | hadronic observables in z+jet events can be subject to large nlo corrections at tev scales , with @xmath0-factors that even reach values of order 50 in some cases .
we develop a method , loopsim , by which approximate nnlo predictions can be obtained for such observables , supplementing nlo z+jet and nlo z+2-jet results with a unitarity - based approximation for missing higher loop terms .
we first test the method against known nnlo results for drell - yan lepton @xmath1 spectra .
we then show our approximate nnlo results for the z+jet observables .
finally we examine whether the loopsim method can provide useful information even in cases without giant @xmath0-factors , with results for observables in dijet events that can be compared to early lhc data . |
, the time at which the signal is to be estimated.,scaledwidth=48.0% ] estimation theory is the science of determining the state of a system , such as a dice , an aircraft , or the weather in boston , from noisy observations @xcite . as shown in fig .
[ classes ] , estimation problems can be classified into four classes , namely , prediction , filtering , retrodiction , and smoothing . for applications that do not require real - time data , such as sensing and communication , smoothing is the most accurate estimation technique .
i have recently proposed a time - symmetric quantum theory of smoothing , which allows one to optimally estimate classical diffusive markov random processes , such as gravitational waves or magnetic fields , coupled to a quantum system , such as a quantum mechanical oscillator or an atomic spin ensemble , under continuous measurements @xcite . in this paper ,
i shall demonstrate in more detail the derivation of this theory using a discrete - time approach , and how it closely parallels the classical time - symmetric smoothing theory proposed by pardoux @xcite .
i shall apply the theory to the design of homodyne phase - locked loops ( pll ) for narrowband squeezed optical beams , as previously considered by berry and wiseman @xcite .
i shall show that their approach can be regarded as a special case of my theory , and discuss how their results can be generalized and improved .
i shall also discuss the weak value theory proposed by aharonov _
et al . _
@xcite in relation with the smoothing theory , and how their theory may be regarded as a smoothing theory for quantum degrees of freedom . in particular , the smoothing quasiprobability distribution proposed in ref .
@xcite is shown to naturally arise from the statistics of weak position and momentum measurements .
this paper is organized as follows : in sec .
[ classical ] , pardoux s classical time - symmetric smoothing theory is derived using a discrete - time approach , which is then generalized to the quantum regime for hybrid classical - quantum smoothing in sec . [ hybrid ] .
application of the hybrid classical - quantum smoothing theory to pll design is studied in sec .
[ adaptive ] .
the relation between the smoothing theory and aharonov _ et al . _
s weak value theory is then discussed in sec .
[ conclusion ] concludes the paper and points out some possible extensions of the proposed theory .
consider the classical smoothing problem depicted in fig . [ classical_smoothing ] .
let @xmath0}\end{aligned}\ ] ] be a vectoral diffusive markov random process that satisfies the system it differential equation @xcite @xmath1 where @xmath2 is a vectoral wiener increment with mean and covariance matrix given by @xmath3 the superscript @xmath4 denotes the transpose .
the vectoral observation process @xmath5 satisfies the observation it equation @xmath6 where @xmath7 is another vectoral wiener increment with mean and covariance matrix given by @xmath8 for generality and later purpose , @xmath2 and @xmath7 are assumed to be correlated , with covariance @xmath9 define the observation record in the time interval @xmath10 as @xmath11 the goal of smoothing is to calculate the conditional probability density of @xmath12 , given the observation record @xmath13 in the time interval @xmath14 .
it is more intuitive to consider the problem in discrete time first .
the discrete - time system and observation equations ( [ system ] ) and ( [ observ ] ) are @xmath15 the observation record @xmath16 } & \equiv { \left\{\delta y_{t_0},\delta y_{t_0+\delta t},\dots,\delta y_{t-\delta t}\right\}}\end{aligned}\ ] ] also becomes discrete .
the covariance matrices for the increments are @xmath17 and the increments at different times are independent of one another . because @xmath18 and @xmath19 are proportional to @xmath20 , one should keep all linear _ and _ quadratic terms of the wiener increments in an equation according to it calculus when taking the continuous time limit . with correlated @xmath18 and @xmath19 , it is preferable , for technical reasons , to rewrite the system equation ( [ discrete_system ] ) as @xcite @xmath21},\end{aligned}\ ] ] where @xmath22 can be arbitrarily set because the expression in square brackets is zero .
the system equation becomes @xmath23}\delta t+ d(x_t , t)\delta y_t \nonumber\\&\quad + b(x_t , t)\delta w_t - d(x_t , t)\delta v_t.\end{aligned}\ ] ] the new system noise is @xmath24\delta t.\end{aligned}\ ] ] the covariance between the new system noise @xmath25 and the observation noise @xmath19 is @xmath26}\delta t,\end{aligned}\ ] ] and can be made to vanish if one lets @xmath27 the new equivalent system and observation model is then @xmath28 } \nonumber\\&\quad + b(x_t , t)\delta u_t , \label{new_system}\\ \delta y_t & = c(x_t , t)\delta t + \delta v_t , \label{new_observ}\end{aligned}\ ] ] with covariances @xmath29}\delta t , \\ { \left\langle\delta v_t\delta v_t^t\right\rangle } & = r(t)\delta t , \\ { \left\langle\delta u_t\delta v_t^t\right\rangle } & = 0.\end{aligned}\ ] ] the new system and observation noises are now independent , but note that @xmath30 becomes dependent on @xmath31 . according to the bayes theorem ,
the smoothing probability density for @xmath12 can be expressed as @xmath32 } ) & = \frac{p(\delta y_{[t_0,t-\delta t]}|x_\tau ) p(x_\tau ) } { p(\delta y_{[t_0,t-\delta t ] } ) } , \label{bayes } \\
p(\delta y_{[t_0,t-\delta t ] } ) & = \int dx_\tau p(\delta y_{[t_0,t-\delta t]}|x_\tau ) p(x_\tau),\end{aligned}\ ] ] where @xmath33 and @xmath34 is the _ a priori _ probability density , which represents one s knowledge of @xmath12 absent any observation
. functions of @xmath12 are assumed to also depend implicitly on @xmath35 . splitting @xmath36}$ ] into the past record @xmath37}\end{aligned}\ ] ] and the future record @xmath38}\end{aligned}\ ] ] relative to time @xmath35 , @xmath39 in eq .
( [ bayes ] ) can be rewritten as @xmath40}|x_\tau ) & = p(\delta y_{\textrm{past}},\delta y_{\textrm{future}}|x_\tau ) \nonumber\\ & = p(\delta y_{\textrm{future}}|\delta y_{\textrm{past}},x_\tau ) p(\delta y_{\textrm{past}}|x_\tau).\end{aligned}\ ] ] because @xmath19 are independent increments , the future record is independent of the past record given @xmath12 , and @xmath40}|x_\tau ) & = p(\delta y_{\textrm{future}}|x_\tau)p(\delta y_{\textrm{past}}|x_\tau).\end{aligned}\ ] ] equation ( [ bayes ] ) becomes @xmath32 } ) & = \frac{p(\delta y_{\textrm{future}}|x_\tau)p(\delta y_{\textrm{past}}|x_\tau)p(x_\tau ) } { \int dx_\tau ( \textrm{numerator } ) } \nonumber\\ & = \frac{p(\delta y_{\textrm{future}}|x_\tau)p(x_\tau|\delta y_{\textrm{past } } ) } { \int dx_\tau ( \textrm{numerator})}. \label{twofilter}\end{aligned}\ ] ] thus , the smoothing density can be obtained by combining the filtering probability density @xmath41 and a retrodictive likelihood function @xmath42 . to derive an equation for the filtering probability density @xmath41 , first express @xmath43})$ ] in terms of @xmath44})$ ] as @xmath45 } ) & = \int dx_t p(x_{t+\delta t},x_t|\delta y_{[t_0,t ] } ) \nonumber\\ & = \int dx_t p(x_{t+\delta t}|x_t,\delta y_{[t_0,t ] } ) p(x_t|\delta y_{[t_0,t]}).\end{aligned}\ ] ] @xmath46})= p(x_{t+\delta
t}|x_t,\delta y_t,\delta y_{[t_0,t-\delta t]})$ ] can be determined from the system equation ( [ new_system ] ) and is equal to @xmath47 , due to the markovian nature of the system process .
so @xmath45 } ) & = \int dx_t p(x_{t+\delta t}|x_t,\delta y_{t } ) p(x_t|\delta y_{[t_0,t ] } ) , \label{ck}\end{aligned}\ ] ] which is a generalized chapman - kolmogorov equation @xcite .
@xmath48 is @xmath49}^{-1 } \delta z_t\bigg\ } , \label{gauss_system}\end{aligned}\ ] ] where @xmath50}.\end{aligned}\ ] ] next , write @xmath51})$ ] in terms of @xmath52})$ ] using the bayes theorem as @xmath53 } ) & = p(x_t|\delta y_{[t_0,t-\delta t]},\delta y_t ) \nonumber\\ & = \frac{p(\delta y_t|x_t,\delta y_{[t_0,t-\delta t]})p(x_t|\delta y_{[t_0,t-\delta t ] } ) } { \int dx_t ( \textrm{numerator } ) } \nonumber\\ & = \frac{p(\delta y_t|x_t)p(x_t|\delta y_{[t_0,t-\delta t ] } ) } { \int dx_t ( \textrm{numerator } ) } , \label{bayes2}\end{aligned}\ ] ] where @xmath54 } ) = p(\delta y_t|x_t)$ ] due to the markovian property of the observation process .
@xmath55 is determined by the observation equation ( [ new_observ ] ) and given by @xmath56}^t { \left[r(t)\delta t\right]}^{-1}\nonumber\\&\quad\times { \left[\delta y_t - c(x_t , t)\delta t\right]}\bigg\}. \label{gauss_observ}\end{aligned}\ ] ] hence , starting with the _ a priori _ probability density @xmath57 , one can solve for @xmath41 by iterating the formula @xmath45 } ) & = \int dx_t p(x_{t+\delta t}|x_t,\delta y_t ) \nonumber\\&\quad\times \frac{p(\delta y_t|x_t)p(x_t|\delta y_{[t_0,t-\delta t ] } ) } { \int dx_t ( \textrm{numerator})}. \label{iterate}\end{aligned}\ ] ] to obtain a stochastic differential equation for the filtering probability density , defined as @xmath58 in the continuous time limit , one should expand eq .
( [ iterate ] ) to first order with respect to @xmath59 and second order with respect to @xmath31 in a taylor series , then apply the rules of it calculus .
the result is the kushner - stratonovich ( ks ) equation @xcite , generalized for correlated system and observation noises by fujisaki _
et al . _
@xcite , given by @xmath60 } \nonumber\\&\quad + { \left(c-{\langlec\rangle}_f\right)}^tr^{-1}d\eta_t f \nonumber\\&\quad -\sum_{\mu}{\frac{\partial } { \partial x_\mu } } { \left[{\left(bsr^{-1}d\eta_t\right)}_\mu f\right ] } , \label{kushner}\end{aligned}\ ] ] where @xmath61 the initial condition is @xmath62 @xmath63 is called the innovation process and is also a wiener increment with covariance matrix @xmath64 @xcite . a linear stochastic equation for an unnormalized @xmath65
is called the duncan - mortensen - zakai ( dmz ) equation @xcite , given by @xmath66 } \nonumber\\&\quad + c^tr^{-1}dy_t f -\sum_{\mu}{\frac{\partial } { \partial x_\mu } } { \left[{\left(bsr^{-1}dy_t\right)}_\mu f\right ] } , \label{zakai}\end{aligned}\ ] ] where the normalization is @xmath67 to solve for the retrodictive likelihood function @xmath42 , note that @xmath68 but @xmath69 can also be expressed in terms of the multitime probability density as @xmath70 } ) & = \int dx_{[\tau , t ] } p(x_{[\tau , t]},\delta y_{[\tau , t-\delta t]}),\end{aligned}\ ] ] where @xmath71 } & \equiv { \left\{x_\tau , x_{\tau+\delta t},\dots , x_t\right\ } } , \\ \int dx_{[\tau , t ] } & \equiv \int dx_\tau \int dx_{\tau+\delta t}\dots\int dx_t.\end{aligned}\ ] ] the multitime density can be rewritten as @xmath72},\delta y_{[\tau , t-\delta t ] } ) & = p(x_t|x_{[\tau , t-\delta t]},\delta y_{[\tau , t-\delta t ] } ) \nonumber\\&\quad\times p(x_{[\tau , t-\delta t]},\delta y_{[\tau , t-\delta t ] } ) . \label{multitime}\end{aligned}\ ] ] again using the markovian property of the system process , @xmath73},\delta y_{[\tau , t-\delta t ] } ) = p(x_t|x_{t-\delta t},\delta y_{t-\delta t } ) , \label{markov_system}\end{aligned}\ ] ] which can be determined from the system equation ( [ new_system ] ) and is given by eq .
( [ gauss_system ] ) .
furthermore , @xmath74},\delta y_{[\tau , t-\delta t]})$ ] in eq .
( [ multitime ] ) can be expressed as @xmath75},\delta y_{[\tau , t-\delta t ] } ) & = p(\delta y_{t-\delta t}|x_{[\tau , t-\delta t]},\delta y_{[\tau , t-2\delta t ] } ) \nonumber\\&\quad\times p(x_{[\tau , t-\delta t]},\delta y_{[\tau , t-2\delta t ] } ) . \label{multitime2}\end{aligned}\ ] ] using the markovian property of the observation process , @xmath76},\delta y_{[\tau , t-2\delta t ] } ) = p(\delta y_{t-\delta t}|x_{t-\delta t } ) , \label{markov_observe}\end{aligned}\ ] ] which can be determined from the observation equation ( [ new_observ ] ) and is given by eq .
( [ gauss_observ ] ) . applying eqs .
( [ multitime ] ) , ( [ markov_system ] ) , ( [ multitime2 ] ) , and ( [ markov_observe ] ) repeatedly , one obtains @xmath70 } ) & = \int dx_t \int dx_{t-\delta t } p(x_t|x_{t-\delta t},\delta y_{t-\delta t } ) \nonumber\\&\quad\times p(\delta y_{t-\delta t}|x_{t-\delta t } ) \nonumber\\&\quad\times \int dx_{t-2\delta t } p(x_{t-\delta t}|x_{t-2\delta t},\delta y_{t-2\delta t } ) \nonumber\\&\quad\times p(\delta y_{t-2\delta t}|x_{t-2\delta t})\dots
\nonumber\\&\quad\times \int dx_\tau p(x_{\tau+\delta t}|x_\tau,\delta y_\tau ) \nonumber\\&\quad\times p(\delta y_{\tau}|x_\tau)p(x_\tau ) . \label{expand}\end{aligned}\ ] ] comparing this equation with eq .
( [ retro ] ) , @xmath42 can be expressed as @xmath77 defining the unnormalized retrodictive likelihood function at time @xmath78 as @xmath79 one can derive a linear backward stochastic differential equation for @xmath80 by applying it calculus backward in time to eq .
( [ retro2 ] ) .
the result is @xcite @xmath81 which is the adjoint equation of the forward dmz equation ( [ zakai ] ) , to be solved backward in time in the backward it sense , defined by @xmath82 with the final condition @xmath83 the adjoint equation with respect to a linear differential equation @xmath84 is defined as @xmath85 where @xmath86 is a linear operator and @xmath87 is the adjoint of @xmath86 , defined by @xmath88 with respect to the inner product @xmath89 after solving eq .
( [ zakai ] ) for @xmath90 and eq .
( [ backward_zakai ] ) for @xmath91 , the smoothing probability density is @xmath92 since @xmath90 and @xmath91 are solutions of adjoint equations , their inner product , which appears as the denominator of eq .
( [ smooth ] ) , is constant in time @xcite .
the denominator also ensures that @xmath93 is normalized , and @xmath90 and @xmath91 need not be normalized separately .
the estimation errors depend crucially on the statistics of @xmath94 .
if any component of @xmath94 , say @xmath95 , is constant in time , then filtering of that particular component is as accurate as smoothing , for the simple reason that @xmath96 must be the same for any @xmath35 , and one can simply estimate @xmath97 at the end of the observation interval ( @xmath98 ) using filtering alone .
this also means that smoothing is not needed when one only needs to detect the presence of a signal in detection problems @xcite , since the presence can be regarded as a constant binary parameter within a certain time interval . in general , however , smoothing can be significantly more accurate than filtering for the estimation of a fluctuating random process in the middle of the observation interval .
another reason for modeling unknown signals as random processes is robustness , as introducing fictitious system noise can improve the estimation accuracy when there are modeling errors @xcite .
if @xmath99 , @xmath100 , and @xmath101 are gaussian , one can just solve for their means and covariance matrices , which completely determine the probability densities .
this is the case when the _ a priori _ probability density @xmath57 is gaussian , and @xmath102 the means and covariance matrices of @xmath99 , @xmath100 , and @xmath101 can then be solved using the linear mayne - fraser - potter ( mfp ) smoother @xcite .
the smoother first solves for the mean @xmath103 and covariance matrix @xmath104 of @xmath99 using the kalman filter @xcite , given by @xmath105 with the initial conditions at @xmath106 determined from @xmath57 .
the mean @xmath107 and covariance matrix @xmath108 of @xmath100 are then solved using a backward kalman filter , @xmath109 with the final condition @xmath110 and @xmath111 . in practice
, the information filter formalism should be used to solve the backward filter , in order to avoid dealing with the infinite covariance matrix at @xmath112 @xcite .
finally , the smoothing mean @xmath113 and covariance matrix @xmath114 are @xmath115 note that @xmath107 and @xmath108 are the mean and covariance matrix of a likelihood function @xmath116 and not those of a conditional probability density @xmath117 , so to perform optimal retrodiction ( @xmath118 ) one should still combine @xmath107 and @xmath108 with the _ a priori _ values @xcite .
consider the problem of waveform estimation in a hybrid classical - quantum system depicted in fig .
[ hybrid_smooth ] .
the classical system produces a vectoral classical diffusive markov random process @xmath94 , which obeys eq .
( [ system ] ) and is coupled to the quantum system . the goal is to estimate @xmath12 via continuous measurements of both systems .
this setup is slightly more general than that considered in @xcite ; here the observations can also depend on @xmath94 .
this allows one to apply the theory to pll design for squeezed beams , as considered by berry and wiseman @xcite , and potentially to other quantum estimation problems as well @xcite .
the statistics of @xmath94 are assumed to be unperturbed by the coupling to the quantum system , in order to avoid the nontrivial issue of quantum backaction on classical systems @xcite . for simplicity ,
in this section we neglect the possibility that the system noise driving the classical system is correlated with the observation noise , although the noise driving the quantum system can still be correlated with the observation noise due to quantum measurement backaction .
just as in the classical smoothing problem , the hybrid smoothing problem is solved by calculating the smoothing probability density @xmath119 . because a quantum system is involved , one may be tempted to use a hybrid density operator @xcite to represent one s knowledge about the hybrid classical - quantum system .
the hybrid density operator @xmath120 describes the joint classical and quantum statistics of a hybrid system , with the marginal classical probability density for @xmath12 and the marginal density operator for the quantum system given by @xmath121 } , \\
\hat\rho(\tau ) & = \int dx_\tau \hat\rho(x_\tau),\end{aligned}\ ] ] respectively .
the hybrid operator can also be regarded as a special case of the quantum density operator , when certain degrees of freedom are approximated as classical .
unfortunately , the density operator in conventional predictive quantum theory can only be conditioned upon past observations and not future ones , so it can not be used as a quantum version of the smoothing probability density . the classical time - symmetric smoothing theory , as a combination of prediction and retrodiction , offers an important clue to how one can circumvent the difficulty of defining the smoothing quantum state . again casting the problem in discrete time , and defining a hybrid effect operator as @xmath122 , which can be used to determine the statistics of future observations given a density operator at @xmath35 , @xmath123},\end{aligned}\ ] ]
one may write , in analogy with eq .
( [ twofilter ] ) @xcite , @xmath32 } ) & = \frac{p(x_\tau,\delta y_{\textrm{future}}|\delta y_{\textrm{past } } ) } { p(\delta y_{\textrm{future}}|\delta y_{\textrm{past } } ) } \nonumber\\ & = \frac{{\operatorname{tr}}[\hat e(\delta y_{\textrm{future}}|x_\tau ) \hat\rho(x_\tau|\delta y_{\textrm{past } } ) ] } { \int dx_\tau { \operatorname{tr}}[\hat e(\delta y_{\textrm{future}}|x_\tau ) \hat\rho(x_\tau|\delta y_{\textrm{past } } ) ] } , \label{two_qfilter}\end{aligned}\ ] ] where @xmath124 is the analog of the filtering probability density @xmath41 and @xmath122 is the analog of the retrodictive likelihood function @xmath42 .
one can then solve for the density and effect operators separately , before combining them to form the classical smoothing probability density .
since the hybrid density operator can be regarded as a special case of the density operator , the same tools in quantum measurement theory can be used to derive a filtering equation for the hybrid operator .
first , write @xmath125})$ ] in terms of @xmath126})$ ] as @xmath127 } ) & = \int dx_t \mathcal k(x_{t+\delta t}|x_t)\hat\rho(x_t|\delta y_{[t_0,t ] } ) , \label{qck}\end{aligned}\ ] ] where @xmath128 is a completely positive map that governs the markovian evolution of the hybrid state independent of the measurement process . equation ( [ qck ] ) may be regarded as a quantum version of the classical chapman - kolmogorov equation . for infinitesimal @xmath129 , @xmath130}_{x= x_{t+\delta t}}.\end{aligned}\ ] ]
the hybrid superoperator @xmath131 can be expressed as @xmath132 } \nonumber\\&\quad + \frac{1}{2}\sum_{\mu,\nu}{\frac{\partial ^2}{\partial x_\mu\partial x_\nu } } { \left[{\left(bqb^t\right)}_{\mu\nu}\hat\rho(x)\right ] } , \label{super_l}\end{aligned}\ ] ] where @xmath133 governs the evolution of the quantum system , @xmath134 governs the coupling of @xmath94 to the quantum system , via an interaction hamiltonian for example , and the last two terms governs the classical evolution of @xmath94 .
next , write @xmath126})$ ] in terms of @xmath135})$ ] using the quantum bayes theorem @xcite as @xmath136 } ) & = \hat\rho(x_t|\delta y_{[t_0,t-\delta t]},\delta y_t ) \nonumber\\ & = \frac{\mathcal j(\delta y_t|x_t)\hat\rho(x_t|\delta y_{[t_0,t-\delta t ] } ) } { \int dx_t{\operatorname{tr}}(\textrm{numerator})}. \label{qbayes}\end{aligned}\ ] ] the measurement superoperator @xmath137 , a quantum version of @xmath55 , is defined as @xmath138 for infinitesimal @xmath59 and measurements with gaussian noise , the measurement operator @xmath139 can be approximated as @xcite @xmath140,\end{aligned}\ ] ] where @xmath141 is a vectoral observation process , @xmath142 is a vector of hybrid operators , generalized from the purely quantum @xmath143 operators in ref .
@xcite so that the observations may also depend directly on the classical degrees of freedom , and @xmath144 is assumed to be positive . to cast the theory in a form similar to the classical one , perform unitary transformations on @xmath141 and @xmath143 , @xmath145 where @xmath146 is a unitary matrix , and rewrite the measurement operator as @xmath147 @xmath148 is a generalization of @xmath149 in the classical case , and @xmath150 is again a positive - definite matrix that characterizes the observation uncertainties and is real and symmetric with eigenvalues @xmath151 .
note that @xmath152 is defined as the adjoint of each vector element , and @xmath4 is defined as the matrix transpose of the vector . for example , @xmath153 the evolution of @xmath135})$ ] can thus be calculated by iterating the formula @xmath154 } ) \nonumber\\ & = \int dx_t \mathcal k(x_{t+\delta t}|x_t ) \frac{\mathcal j(\delta y_t|x_t)\hat\rho(x_t|\delta y_{[t_0,t-\delta t ] } ) } { \int dx_t{\operatorname{tr}}(\textrm{numerator})}.\end{aligned}\ ] ] taking the continuous time limit via it calculus and defining the conditional hybrid density operator at time @xmath78 as @xmath155 one obtains @xcite @xmath156 } , \label{qks}\end{aligned}\ ] ] where @xmath157 } , \\
d\eta_t & \equiv dy_t - \frac{dt}{2}{\langle\hat c+\hat c^\dagger\rangle}_{\hat f}\end{aligned}\ ] ] is a wiener increment with covariance matrix @xmath64 @xcite , h.c .
denotes the hermitian conjugate , and the initial condition is the _ a priori _ hybrid density operator @xmath158 .
equation ( [ qks ] ) is a quantum version of the ks equation ( [ kushner ] ) and can be regarded as a special case of the belavkin quantum filtering equation @xcite .
a linear version of the ks equation for an unnormalized @xmath159 is @xmath160 and the normalization is @xmath161}.\end{aligned}\ ] ] equation ( [ qzakai ] ) is a quantum generalization of the dmz equation ( [ zakai ] ) . taking a similar approach to the one in sec .
[ retro_smooth ] and using the quantum regression theorem , one can express the future observation statistics as @xcite @xmath162 } \label{qretro}\\ & = \int dx_t { \operatorname{tr}}\bigg[\int dx_{t-\delta t } \mathcal k(x_t|x_{t-\delta t } ) \nonumber\\&\quad\cdot \mathcal j(\delta y_{t-\delta t}|x_{t-\delta t } ) \nonumber\\&\quad\cdot \int dx_{t-2\delta t } \mathcal k(x_{t-\delta t}|x_{t-2\delta t } ) \nonumber\\&\quad\cdot \mathcal j(\delta y_{t-2\delta t}|x_{t-2\delta t})\dots
\nonumber\\&\quad\cdot \int dx_\tau \mathcal k(x_{\tau+\delta t}|x_\tau ) \mathcal j(\delta y_{\tau}|x_\tau)\hat\rho(x_\tau)\bigg ] , \label{qexpand}\end{aligned}\ ] ] which are analogous to eq .
( [ retro ] ) and eq .
( [ expand ] ) , respectively . comparing eq .
( [ qretro ] ) with eq .
( [ qexpand ] ) , and defining the adjoint of a superoperator @xmath163 as @xmath164 , such that @xmath165 } & = { \operatorname{tr}}{\left\{{\left[\mathcal o^*\hat e(x)\right]}\hat\rho(x)\right\}},\end{aligned}\ ] ] the hybrid effect operator can be written as @xmath166 the operation @xmath167 may also be regarded as a hybrid superoperator on a hybrid operator , and is the adjoint of @xmath168 , defined by @xmath169 with respect to the hilbert - schmidt inner product @xmath170}. \label{inner}\end{aligned}\ ] ] one can then rewrite eqs .
( [ qretro ] ) , ( [ qexpand ] ) , and ( [ hybrid_effect ] ) more elegantly as @xmath171 in the continuous time limit , a linear stochastic differential equation for the unnormalized effect operator @xmath172 can be derived .
the result is @xcite @xmath173 to be solved backward in time in the backward it sense , with the final condition @xmath174 equation ( [ backward_qzakai ] ) is the adjoint equation of the forward quantum dmz equation ( [ qzakai ] ) with respect to the inner product defined by eq .
( [ inner ] ) .
it is a generalization of the classical backward dmz equation ( [ backward_zakai ] ) . finally , after solving eq .
( [ qzakai ] ) for @xmath175 and eq .
( [ backward_qzakai ] ) for @xmath176 , the smoothing probability density is @xmath177 } { \int dx{\operatorname{tr}}[\hat g(x,\tau)\hat f(x,\tau)]}. \label{qsmooth}\end{aligned}\ ] ] the denominator of eq .
( [ qsmooth ] ) ensures that @xmath93 is normalized , so @xmath175 and @xmath176 need not be normalized separately .
table [ analogs ] lists some important quantities in classical smoothing with their generalizations in hybrid smoothing for comparison .
[ cols="<,<,<,<",options="header " , ] to solve eqs .
( [ qzakai ] ) , ( [ backward_qzakai ] ) , and ( [ qsmooth ] ) , one way is to convert them to equations for quasiprobability distributions @xcite .
the wigner distribution is especially useful for quantum systems with continuous degrees of freedom .
it is defined as @xcite @xmath178 where @xmath179 and @xmath180 are normalized position and momentum vectors .
it has the desirable property @xmath181 which is unique among generalized quasiprobability distributions @xcite . the smoothing probability density given by eq .
( [ qsmooth ] ) can then be rewritten as @xmath182 where @xmath183 and @xmath184 are the wigner distributions of @xmath185 and @xmath186 , respectively .
equation ( [ wigner_smooth ] ) resembles the classical expression ( [ smooth ] ) with the quantum degrees of freedom @xmath179 and @xmath180 marginalized .
if @xmath187 is nonnegative and the stochastic equations for @xmath188 and @xmath189 converted from eqs .
( [ qzakai ] ) and ( [ backward_qzakai ] ) have the same form as the classical dmz equations given by eqs .
( [ zakai ] ) and ( [ backward_zakai ] ) , the hybrid smoothing problem becomes equivalent to a classical one and can be solved using well known classical smoothers .
for example , if @xmath188 and @xmath189 are gaussian , @xmath93 is also gaussian , and their means and covariances can be solved using the linear mfp smoother described in sec .
consider the pll setup depicted in fig . [ pll ] .
the optical parametric oscillator ( opo ) produces a squeezed vacuum with a squeezed @xmath180 quadrature and an antisqueezed @xmath179 quadrature .
the squeezed vacuum is then displaced by a real constant @xmath190 to produce a phase - squeezed beam , the phase of which is modulated by @xmath191 , an element of the vectoral random process @xmath94 described by the system it equation ( [ system ] ) .
the output beam is measured continuously by a homodyne pll , and the local - oscillator phase @xmath192 is continuously updated according to the real - time measurement record .
the use of pll for phase estimation in the presence of quantum noise has been mentioned as far back as 1971 by personick @xcite .
wiseman suggested an adaptive homodyne scheme to measure a constant phase @xcite , which was then experimentally demonstrated by armen _ et al . _
for the optical coherent state @xcite .
berry and wiseman @xcite and pope _ et al . _
@xcite studied the problem with @xmath193 being a wiener process .
berry and wiseman later generalized the theory to account for narrowband squeezed beams @xcite .
_ et al.__also studied the problem for the case of @xmath94 being a gaussian process @xcite , but the squeezing model considered in refs .
@xcite is not realistic . using the hybrid smoothing theory developed in sec .
[ hybrid ] , one can now generalize these earlier results to the case of an arbitrary diffusive markov process and a realistic squeezing model .
let @xmath194 be the hybrid density operator for the combined quantum - opo - classical - modulator system .
the evolution of the opo below threshold in the interaction picture is governed by @xmath195 } , \\
\hat h_0 & = -i\frac{\hbar\chi}{2 } { \left(\hat a\hat a-\hat a^\dagger\hat a^\dagger\right ) } \\ & = \frac{\hbar\chi}{2}{\left(\hat q\hat p + \hat p \hat q\right)},\end{aligned}\ ] ] where @xmath196 is the annihilation operator for the cavity optical mode , and @xmath197 and @xmath198 are the antisqueezed and squeezed quadrature operators , respectively , defined as @xmath199 with the commutation relation @xmath200 & = i.\end{aligned}\ ] ] the classical phase modulator does not influence the evolution of the opo , so @xmath201 but it modulates the opo output .
@xmath148 in this case is @xmath202 where @xmath203 is the transmission coefficient of the partially reflecting opo output mirror , @xmath204 , and the symbol and sign conventions here roughly follows those of refs .
@xcite . to ensure the correct unconditional quantum dynamics ,
the hamiltonian should be changed to ( ref .
@xcite , sec .
11.4.3 ) @xmath205},\end{aligned}\ ] ] in order to eliminate the spurious effect of the displacement term in @xmath206 on the opo .
after some algebra , the forward stochastic equation for the wigner distribution @xmath188 becomes @xmath207 } \nonumber\\&\quad -{\left[{\left(\chi -\frac{\gamma}{2}\right)}{\frac{\partial } { \partial q}}{\left ( qf\right)}+{\left(-\chi -\frac{\gamma}{2}\right)}{\frac{\partial } { \partial p}}{\left(pf\right)}\right ] } \nonumber\\&\quad + \frac{\gamma}{4}{\left({\frac{\partial^2 f}{\partial q^2}}+{\frac{\partial^2
f}{\partial p^2}}\right)}\bigg\ } \nonumber\\&\quad + dy_t \bigg[\sin(\phi-\phi_t ' ) { \left(2b+\sqrt{2\gamma}q + \sqrt{\frac{\gamma}{2}}{\frac{\partial } { \partial q}}\right ) } \nonumber\\&\quad + \cos(\phi-\phi_t ' ) { \left(\sqrt{2\gamma}p+\sqrt{\frac{\gamma}{2}}{\frac{\partial } { \partial p}}\right)}\bigg]f . \label{wigner_zakai}\end{aligned}\ ] ] this is precisely the classical dmz equation ( [ zakai ] ) with correlated system and observation noises .
the equivalent classical system equations are then @xmath208 and the equivalent observation equation is @xmath209 where @xmath210 and @xmath211 are independent wiener increments with covariance @xmath212 . @xmath210 and @xmath211 , which appear in both the system equation and the observation equation , are simply quadratures of the vacuum field , coupled to both the cavity mode and the output field via the opo output mirror
. equations ( [ equiv_system ] ) and ( [ equiv_observ ] ) coincide with the model of berry and wiseman in ref .
@xcite when @xmath94 is a wiener process , and eq .
( [ wigner_zakai ] ) is the continuous limit of their approach to phase estimation .
this approach can also be regarded as an example of the general method of accounting for colored observation noise by modeling the noise as part of the system @xcite .
if @xmath213 , @xmath214 is an additive white gaussian noise , and the model is reduced to that studied in refs .
@xcite . in that case , it is desirable to make @xmath192 follow @xmath193 as closely as possible , so that @xmath5 can be approximated as @xmath215 and the kalman filter can be used if @xmath94 is gaussian @xcite .
provided that eq .
( [ approx ] ) is valid , one should make @xmath192 the conditional expectation of @xmath216 , given by @xmath217 for phase - squeezed beams , it also seems desirable to make @xmath192 close to @xmath193 in order to minimize the magnitude of @xmath218 .
equation ( [ lo_phase ] ) may not provide the optimal @xmath192 in general , however , as it does not necessarily minimize the magnitude of @xmath218 or the estimation errors .
the optimal control law for @xmath192 should be studied in the context of control theory . while @xmath192 needs to be updated in real time and must be calculated via filtering
, the estimation accuracy can be improved by smoothing .
the backward dmz equation for @xmath189 is the adjoint equation with respect to eq .
( [ wigner_zakai ] ) , given by @xmath219 } \nonumber\\&\quad + \frac{\gamma}{4}{\left({\frac{\partial^2 g}{\partial q^2}}+{\frac{\partial^2 g}{\partial p^2}}\right)}\bigg\ } \nonumber\\&\quad + dy_t \bigg[\sin(\phi-\phi_t ' ) { \left(2b+\sqrt{2\gamma}q - \sqrt{\frac{\gamma}{2}}{\frac{\partial } { \partial q}}\right ) } \nonumber\\&\quad + \cos(\phi-\phi_t ' ) { \left(\sqrt{2\gamma}p-\sqrt{\frac{\gamma}{2}}{\frac{\partial } { \partial p}}\right)}\bigg]g , \label{wigner_bzakai}\end{aligned}\ ] ] and the smoothing probability density @xmath93 is given by eq .
( [ wigner_smooth ] ) .
the use of linear smoothing for the case of @xmath94 being a gaussian process and @xmath214 being a white gaussian noise has been studied in refs .
practical strategies of solving eqs .
( [ wigner_zakai ] ) and ( [ wigner_bzakai ] ) in general are beyond the scope of this paper , but classical nonlinear filtering and smoothing techniques should help @xcite .
one can also use the hybrid smoothing theory to study the general problem of force estimation via a squeezed probe beam and a homodyne pll , by modeling the phase modulator as a quantum mechanical oscillator instead and combining the problem studied in this section with the force estimation problem studied in ref .
previous sections focus on the estimation of classical signals , but there is no reason why one can not apply smoothing to quantum degrees of freedom as well , as shown in fig . [ quantum_smoothing ] .
first consider the predicted density operator at time @xmath35 conditioned upon past observations , given by @xmath220},\end{aligned}\ ] ] where the classical degrees of freedom are neglected for simplicity .
the predicted expectation of an observable , such as the position of a quantum mechanical oscillator , is @xmath221 } = \frac{{\operatorname{tr}}[\hat o\hat f(\tau)]}{{\operatorname{tr}}[\hat f(\tau)]}. \label{predict}\end{aligned}\ ] ] one may also use retrodiction , after some measurements of a quantum system have been made , to estimate its initial quantum state before the measurements @xcite , using the retrodictive density operator defined as @xmath222}.\end{aligned}\ ] ] the retrodicted expectation of an observable is @xmath223 } = \frac{{\operatorname{tr}}[\hat g(\tau)\hat o]}{{\operatorname{tr}}[\hat g(\tau)]}. \label{retrodict}\end{aligned}\ ] ] causality prevents one from going back in time to verify the retrodicted expectation , but if the degree of freedom with respect to @xmath224 at time @xmath35 is entangled with another `` probe '' system , then one can verify the retrodicted expectation by measuring the probe and inferring @xmath224 @xcite .
the idea of verifying retrodiction by entangling the system at time @xmath35 with a probe can also be extended to the case of smoothing , as proposed by aharonov _
_ @xcite . in the middle of a sequence of measurements ,
if one weakly couples the system to a probe for a short time , so that the system is weakly entangled with the probe , and the probe is subsequently measured , the measurement outcome on average can be characterized by the so - called weak value of an observable , defined as @xcite @xmath225 } { { \operatorname{tr}}[\hat g(\tau)\hat f(\tau)]}.\end{aligned}\ ] ] the weak value becomes a prediction given by eq .
( [ predict ] ) when future observations are neglected , such that @xmath226 , and becomes a retrodiction given by eq .
( [ retrodict ] ) when past observations are neglected and there is no _ a priori _ information about the quantum system at time @xmath35 , such that @xmath227 .
when @xmath228 and @xmath229 are incoherent mixtures of @xmath224 eigenstates , @xmath230 the weak value becomes @xmath231 and is consistent with the classical time - symmetric smoothing theory described in sec .
[ classical ] .
hence , the weak value can be regarded as a quantum generalization of the smoothing estimate , conditioned upon past and future observations .
one can also establish a correspondence between a classical theory and a quantum theory via quasiprobability distributions .
given the smoothing probability density in terms of the wigner distributions in eq .
( [ wigner_smooth ] ) , one may be tempted to undo the marginalizations over the quantum degrees of freedom and define a smoothing quasiprobability distribution as @xmath232 where @xmath233 and @xmath234 are the wigner distributions of @xmath228 and @xmath229 , respectively .
intriguingly , @xmath235 , being the product of two wigner distributions , can exhibit quantum position and momentum uncertainties that violate the heisenberg uncertainty principle .
this has been shown in ref .
@xcite , when the position of a quantum mechanical oscillator is monitored via continuous measurements and smoothing is applied to the observations . from the perspective of classical estimation theory , it is perhaps not surprising that smoothing can improve upon an uncertainty relation based on a predictive theory .
the important question is whether the sub - heisenberg uncertainties can be verified experimentally .
@xcite argues that it can be done only by bayesian estimation , but in the following i shall propose another method based on weak measurements .
it can be shown that the expectation of @xmath179 using @xmath235 is @xmath236 } { { \operatorname{tr}}[\hat g(\tau)\hat f(\tau ) ] } = { \operatorname{re } } { } _ { \hat g}{\langle\hat q\rangle}_{\hat f},\end{aligned}\ ] ] which is the real part of the weak value , and likewise for @xmath237 , so the smoothing position and momentum estimates are closely related to their weak values .
more generally , consider the joint probability density for a quantum position measurement followed by a quantum momentum measurement , conditioned upon past and future observations : @xmath238 } , \label{joint_measurement } \\
\mathcal c & \equiv \int dy_q dy_p { \operatorname{tr}}\big[\hat g(\tau)\hat m_p(y_p)\hat m_q(y_q ) \nonumber\\&\quad\times \hat f(\tau)\hat m_q^\dagger(y_q)\hat m_p^\dagger(y_p)\big],\end{aligned}\ ] ] where the measurement operators @xmath239}{|q\rangle}{\langleq| } , \\ \hat m_p(y_p ) & = \int dp { \left(\frac{\epsilon_p}{2\pi}\right)}^{\frac{1}{4 } } \exp{\left[-\frac{\epsilon_p}{4}(y_p - p)^2\right]}{|p\rangle}{\langlep|}\end{aligned}\ ] ] are assumed to be gaussian and backaction evading .
after some algebra , @xmath240}\tilde p(q , p ) , \label{convolution}\\ \tilde p(q , p ) & \equiv \frac{1}{2\pi\mathcal c } \int du dv\exp{\left(-\frac{\epsilon_q u^2+\epsilon_p v^2}{8}\right ) } \nonumber\\&\quad\times { \left\langlep+\frac{v}{2}\right|}\hat g(\tau){\left|p-\frac{v}{2}\right\rangle}\exp(ivq ) \nonumber\\&\quad\times { \left\langleq-\frac{u}{2}\right|}\hat f(\tau){\left|q+\frac{u}{2}\right\rangle}\exp(ipu).\end{aligned}\ ] ] from the perspective of classical probability theory , eq .
( [ convolution ] ) can be interpreted as the probability density of noisy position and momentum measurements with noise variances @xmath241 and @xmath242 , when the measured object has a classical phase - space density given by @xmath243 . in the limit of infinitesimally weak measurements , @xmath244 , and @xmath245
thus , @xmath235 can be obtained approximately from an experiment with small @xmath246 and @xmath247 by measuring @xmath248 for the same @xmath186 and @xmath185 and deconvolving eq .
( [ convolution ] ) . in practice ,
@xmath246 and @xmath247 only need to be small enough such that @xmath249 .
this allows one , at least in principle , to experimentally demonstrate the sub - heisenberg uncertainties predicted in ref .
@xcite in a frequentist way , not just by bayesian estimation as described in ref .
note , however , that @xmath235 can still go negative , so it can not always be regarded as a classical probability density .
this underlines the wave nature of a quantum object and may be related to the negative probabilities encountered in the use of weak values to explain hardy s paradox @xcite .
in conclusion , i have used a discrete - time approach to derive the classical and quantum theories of time - symmetric smoothing .
the hybrid smoothing theory is applied to the design of pll , and the relation between the proposed theory and aharonov _ et al . _
s weak value theory is discussed .
possible generalizations of the theory include taking jumps into account for the classical random process @xcite and adding quantum measurements with poisson statistics , such as photon counting @xcite .
potential applications not discussed in this paper include cavity quantum electrodynamics @xcite , photodetection theory @xcite , atomic magnetometry @xcite , and quantum information processing in general . on a more fundamental level
, it might also be interesting to generalize the weak value theory and the smoothing quasiprobability distribution to other kinds of quantum degrees of freedom in addition to position and momentum , such as spin , photon number , and phase .
a general quantum smoothing theory would complete the correspondence between classical and quantum estimation theories .
discussions with seth lloyd and jeffrey shapiro are gratefully acknowledged . this work is financially supported by the keck foundation center for extreme quantum information theory . h. l. van trees , _ detection , estimation , and modulation theory , part i _ ( wiley , new york , 2001 ) ; _ detection , estimation , and modulation theory , part ii : nonlinear modulation theory _
( wiley , new york , 2002 ) ; _ detection , estimation , and modulation theory , part iii : radar - sonar processing and gaussian signals in noise _ ( wiley , new york , 2001 ) .
p. warszawski , h. m. wiseman , and h. mabuchi , * 65 * , 023802 ( 2002 ) ; p. warszawski and h. m. wiseman , j. opt .
b : quant .
semiclass .
opt . * 5 * , 1 ( 2003 ) ; * 5 * , 15 ( 2003 ) ; n. p. oxtoby , p. warszawski , h. m. wiseman , he - bi sun and r. e. s. polkinghorne , * 71 * , 165317 ( 2005 ) .
v. p. belavkin , radiotech .
elektron . * 25 * , 1445 ( 1980 ) ; v. p. belavkin , in _ information complexity and control in quantum physics _ , edited by a. blaquire , s. diner , and g. lochak ( springer , vienna , 1987 ) , p. 311 ; v. p. belavkin , in _
stochastic methods in mathematics and physics _ , edited by r. gielerak and w. karwowski ( world scientific , singapore , 1989 ) , p. 310 ; v. p. belavkin , in _ modeling and control of systems in engineering , quantum mechanics , economics , and biosciences _ , edited by a. blaquire ( springer , berlin , 1989 ) , p. 245 .
a. barchielli , l. lanz , and g. m. prosperi , nuovo cimento , * 72b * , 79 ( 1982 ) ; found .
* 13 * , 779 ( 1983 ) ; h. carmichael , _ an open systems approach to quantum optics _ ( springer - verlag , berlin , 1993 ) .
s. m. barnett , d. t. pegg , j. jeffers , o. jedrkiewicz , and r. loudon , * 62 * , 022313 ( 2000 ) ; s. m. barnett , d. t. pegg , j. jeffers , and o. jedrkiewicz , * 86 * , 2455 ( 2001 ) ; d. t. pegg , s. m. barnett , and j. jeffers , * 66 * , 022106 ( 2002 ) . | classical and quantum theories of time - symmetric smoothing , which can be used to optimally estimate waveforms in classical and quantum systems , are derived using a discrete - time approach , and the similarities between the two theories are emphasized .
application of the quantum theory to homodyne phase - locked loop design for phase estimation with narrowband squeezed optical beams is studied .
the relation between the proposed theory and aharonov _ et al . _ s weak value theory is also explored . |
this paper is devoted to the application of the semidefinite programming method to estimates of the size of the largest possible two - distance set on the sphere @xmath3 .
a spherical two - distance set is a finite collection @xmath4 of unit vectors in @xmath0 such that the set of distances between any two distinct vectors in @xmath4 has cardinality two . estimating the maximum size @xmath5 of such a set is a classical problem in distance geometry that has been studied for several decades .
we begin with an overview of known results .
a lower bound on @xmath5 is obtained as follows .
let @xmath6 be the standard basis in @xmath7 .
the points @xmath8 form a spherical two - distance set in the plane @xmath9 ( after scaling ) , and therefore @xmath10 the first major result for upper bounds was obtained by delsarte , goethals , and seidel @xcite .
they proved that , irrespective of the actual values of the distances , the following `` harmonic '' bound holds true : @xmath11 they also showed that this bound is tight for dimensions @xmath12 where it is attained by sets of equiangular lines .
moreover , the results of @xcite , bannai et al .
@xcite , and nebe and venkov @xcite imply that @xmath5 can attain the harmonic bound only if @xmath13 with the exception of an infinite sequence of values of @xmath14 that begins with @xmath15 . therefore , unless @xmath16 is of the above form , @xmath17 these results are proved using the link between 2-distance sets and tight spherical 4-designs established in @xcite .
another advance in estimating the function @xmath5 was made by musin @xcite .
let @xmath18 and suppose that @xmath19 where @xmath20 are the values of the squared distances between the points .
musin proved that @xmath21 he then used delsarte s linear programming method to prove that @xmath22 if @xmath23 here we make another step for spherical two - distance sets , extending the range of dimensions in which the bound is tight .
the state of the art for @xmath5 can be summarized as follows .
we have @xmath24 @xmath25 and @xmath26 if @xmath27 then @xmath28 as detailed in the remarks after eq . above .
the part of this theorem that is established in the present paper relates to dimensions @xmath1 and @xmath29 our results are computational in nature and are obtained using the semidefinite programming method .
the other parts of this theorem follow from the results in @xcite .
as far as actual constructions of spherical two - distance sets are concerned , rather little is known beyond the set of midpoints of the edges of a regular simplex mentioned above .
another way of constructing such sets is to start with a set of equiangular lines in @xmath0 @xcite .
if the angle between each pair of lines is @xmath30 then taking one point from each pair of points on @xmath31 defined by the line , we obtain a two - distance set with @xmath32 the largest possible number of equiangular lines in @xmath0 is @xmath33 ( this result is due to gerzon , see @xcite ) .
this bound is attained for @xmath34 for instance , for @xmath35 the set of 6 lines is obtained from 6 diagonals of the icosahedron , which gives many ways of constructing inequivalent spherical two - distance sets of cardinality 6 .
the only three instances in which the known spherical two - distance sets are of cardinality greater than @xmath33 occur in dimensions @xmath36 and @xmath37
a semidefinite program is an optimization problem of the form @xmath38 where @xmath39 is an @xmath40 variable matrix , @xmath41 and @xmath42 are given hermitian matrices , @xmath43 is a given vector and @xmath44 is the inner product of two matrices .
semidefinite programming is an extension of linear programming that has found a range of applications in combinatorial optimization , control theory , distance geometry , and coding theory .
general introduction to semidefinite programming is given , for instance , in @xcite .
the main problem addressed by the sdp method in distance geometry is related to deriving bounds on the cardinality of point sets in a metric space @xmath45 with a given set of properties such as a given minimum separation between distinct points in the set .
the sdp method has its roots in harmonic analysis of the isometry group of the metric space in question .
it is broadly applicable in both finite and compact infinite spaces .
examples of the former include the hamming and johnson spaces , their @xmath46-analogs , other metric spaces on the set of @xmath16-strings over a finite alphabet , as well as the finite projective space .
the main example in the infinite case is given by real and complex spheres , although the sdp method is also applicable in other compact homogeneous spaces .
working out the details in each example is a nontrivial task that includes analysis of irreducible modules in the space of functions @xmath47 under the action of the isometry group @xmath48 of @xmath45 .
the zonal matrices that arise in this analysis initially have large size that can be reduced relying on symmetries arising from the group action .
this gives rise to an sdp optimization problem that is solved by computer for a given set of dimensions ( the numerical part is also not straightforward and rather time - consuming ) .
foundations and analysis of particular cases have been the subject of a considerable number of research and overview publications in the last decade ; see in particular recent surveys @xcite and references therein .
the origins of the sdp method and the discussed applications can be traced back to the work of delsarte @xcite which introduced the machinery of association schemes in the analysis of point configurations ( codes ) in finite spaces .
delsarte derived linear programming ( lp ) bounds on the cardinality of a set of points in the space under the condition on the minimum separation of distinct points in the set .
delsarte s results were linked to harmonic analysis and group representations in the works of delsarte , goethals and seidel @xcite ( for the case @xmath31 ) and kabatyansky and levenshtein @xcite ( for general compact symmetric spaces ) . from now on we focus on the case @xmath49
let @xmath50 denote the gegenbauer polynomials of degree @xmath51 .
they are defined recursively as follows : @xmath52 and @xmath53 delsarte et al .
@xcite showed that for any finite set of points @xmath54 @xmath55 the proof of this inequality in @xcite used the addition formula for spherical harmonics .
an earlier , geometric proof of had been given by schoenberg @xcite , although his work was not known to researchers in the area discussed until at least the 1990s .
positivity conditions give rise to the lp bound on the cardinality of spherical two - distance sets .
[ thm : dgs ] ( delsarte et al .
@xcite ) let @xmath54 be a finite set and suppose that @xmath56 for any @xmath57 then @xmath58 in this theorem @xmath59 are the optimization variables that refer to the number of ordered pairs of points in @xmath4 with inner product @xmath60 and @xmath61 , respectively .
for instance , @xmath62 this theorem is a specialization of a more general lp bound on spherical codes of @xcite .
applications of semidefinite programming in coding theory and distance geometry gained momentum after the pioneering work of schrijver @xcite that derived sdp bounds on codes in the hamming and johnson spaces .
schijver s approach was based on the so - called terwilliger algebra of the association scheme and formed a far - reaching generalization of the work of delsarte @xcite .
elements of the groundwork for sdp bounds in the hamming space were laid by dunkl @xcite , although this connection was also made somewhat later @xcite .
we refer to @xcite for a detailed general survey of the approach via association schemes and further references .
sdp bounds for the real sphere were derived by bachoc and vallentin @xcite in the context of the kissing number problem .
the kissing number @xmath63 is the maximum number of unit spheres that can touch a unit sphere without overlapping , i.e. the maximum number of points on the sphere such that the angular separation between any pair of them is at least @xmath64 following @xcite , define a @xmath65 matrix @xmath66 by setting @xmath67 where @xmath68 is a positive integer , and a matrix @xmath69 by setting @xmath70 where the sum is over all permutations on 3 elements .
note that @xmath71 for all @xmath72 and all @xmath73 one of the main results of @xcite is that for any finite set of points @xmath54 @xmath74 the matrices @xmath75 play the role of the constraints @xmath76 in the general sdp problem .
positivity constraints give rise to a general sdp bound on the cardinality of point sets obtained in @xcite , where it was used to improve upper bounds on @xmath63 in small dimensions . in the next section we state a specialization of this bound for the case of 2-distance sets . as a final remark
, we note that constraints arise from the unrestricted action of @xmath48 on @xmath77 constraints are obtained by considering only actions that fix an arbitrary given point on the sphere .
further sdp bounds can be obtained by considering zonal matrices that arise from actions that fix any given number of points ; however even for two points , actual evaluation of the bounds requires significant computational effort @xcite .
the general sdp bound on spherical codes of @xcite specializes to our case as follows .
[ thm : sdp ] let @xmath4 be a spherical two - distance set with inner products @xmath60 and @xmath61 .
let @xmath68 be a positive integer .
the cardinality @xmath78 is bounded above by the solution of the following semidefinite programming problem : @xmath79 in this theorem the variables @xmath86 refer to the number of ordered pairs of vectors in @xmath4 with inner product @xmath60 and @xmath61 respectively ; namely we have @xmath87 we note that the sdp problem seeks to optimize the same linear form as the lp problem , but adds more constraints on the configuration . because of this , theorem [ thm : sdp ] usually gives tighter bounds than theorem [ thm : dgs ] .
this fact is evident from the table below and is also known from the calculation of kissing numbers in @xcite .
several remarks are in order .
first , implementation of sdp for two - distance sets differs from earlier computations in @xcite in that in our case there are no limits on the minimum separation of the points .
next , we restrict our calculations to the case @xmath88 as no improvement is observed for larger values .
finally , by a result larman et al .
@xcite , if @xmath89 then the inner products @xmath90 are related by @xmath91 where @xmath92 is an integer .
thus we obtain a family of sdp bounds parametrized by @xmath93 since @xmath94 we get that @xmath95 moreover , if @xmath96 then @xmath78 can not be large by the rankin bounds @xcite , and if @xmath97 then @xmath78 is bounded by .
we conclude as follows .
for instance , for @xmath101 we obtain that @xmath102 . partitioning @xmath103 into a number of small segments , we plot the value @xmath104 as a function of @xmath60 evaluated at the nodes of the partition .
the result is shown in fig .
[ fig : subfig1 ] .
a part of the segment around the maximum appears in fig .
[ fig : subfig2 ] .
this computation gives an indication of the answer , but in principle the value @xmath104 could oscillate between the nodes of the partition .
ruling this out requires perturbation analysis of the sdp problem which is not immediate .
@xmath106 @xmath107 @xmath108 @xmath109 we need to estimate from above the maximum value of this problem over @xmath110.$ ] accounting for a continuous value set of the parameter in sdp problems is a challenging task .
we approach it by employing the sum - of - squares method .
constraints - impose positivity conditions on some univariate polynomials of @xmath60 for @xmath111 the following sequence of steps transforms the constraints to semidefinite conditions .
observe that a polynomial @xmath112 of degree at most @xmath14 satisfies @xmath113 for @xmath114 if and only if the polynomial of degree at most @xmath115 @xmath116 for all @xmath117 next , a polynomial nonnegative on the entire real axis can be written as a sum of squares , @xmath118 where the @xmath119 are polynomials .
further , by a result of nesterov @xcite , a polynomial @xmath120 of degree @xmath115 is a sum of squares if and only if there exists a positive semidefinite matrix @xmath121 such that @xmath122 where @xmath123 thus , constraints - can be transformed to semidefinite conditions . as a result , we obtain an sdp problem that can be solved by computer . we solved the resulting problem for @xmath124 using the matlab toolbox sostools @xcite in the yalmip environment @xcite .
an advantage in using sostools is that it accepts @xmath60 as an sdp variable , thereby accounting for all the values of @xmath60 in the segment .
thus , we obtain the value @xmath125 however , this may impose excessive constraints on the value of the sdp problem because all the conditions for different values of @xmath60 are involved at the same time . to work around this accumulation ,
we use a sub - partitioning of the segment @xmath103 into smaller segments . for each of them , sostools outputs the largest value of the minimum of the sdp problem over all @xmath60 in the segment .
it turns out that , in many cases , the maximum of these solutions is smaller than @xmath126 computed directly by the package .
the estimates of the answer computed from the primal problem serve as a guidance of the needed step length of the partition .
the solution of the sum - of - squares sdp optimization problem provides a rigorous proof for the estimates obtained by discretizing the primal problem - . for instance , for @xmath1 we partition @xmath127 into 20 subsegments , finding 276.5 as the maximum value of the dual sdp problem for @xmath128 etc .
the results of the calculation are summarized in the table below .
the part of the table for @xmath129 except for the values of the sdp bound , is from @xcite .
the improvement provided by theorem [ thm : sdp ] over the lp bound is quite substantial even for relatively small dimensions .
the lp bound is above @xmath33 for @xmath130 and is not included starting with @xmath131 the cases @xmath132 and @xmath133 are not resolved by sdp , although for @xmath134 we still obtain an improvement over the harmonic bound .
the value of @xmath51 shown in the table accounts for the largest value of the sdp problem among the possible choices of @xmath51 .
this guarantees that the value sdp@xmath98 is equal to or smaller than the number in the table for all the possible values of the inner products @xmath90 in the point set .
notice that for @xmath132 the sdp bound coincides with the bound .
for @xmath1 the results of @xcite leave two possibilities , @xmath135 and 277 .
the sdp bound resolves this for the former , establishing the corresponding part of the claim in . as is seen from fig .
[ fig : subfig2 ] , the largest value of sdp@xmath98 is attained for @xmath136 and is equal to 276 .
this case corresponds to 276 equiangular lines in @xmath137 with angle @xmath138 , which can be constructed either using strongly regular graphs or the leech lattice ( see @xcite for details ) . c. bachoc , d.c .
gijswijt , a. schrijver , and f. vallentin , _ invariant semidefinite programs _
, handbook on semidefinite , conic and polynomial optimization , internat .
management sci .
166 , springer , new york , ny , 2012 , pp .
219269 .
w. j. martin and h. tanaka , _ commutative association schemes _ , european j. combin . * 30 * ( 2009 ) , no . 6 , 14971525 .
h. d. mittelman and f. vallentin , _ high accuracy semidefinite programming bounds for kissing numbers _ , experimental mathematics * 19 * ( 2010 ) , 174178 .
a. schrijver , _ new code upper bounds from the terwilliger algebra and semidefinite programming _ , ieee trans .
theory * 51 * ( 2005 ) , no . 8 , 28592866 . f. vallentin , _ symmetry in semidefinite programs _ , linear algebra appl .
* 430 * ( 2009 ) , no . 1 , 360369 . | a spherical two - distance set is a finite collection of unit vectors in @xmath0 such that the distances between any two distinct vectors assume only two values .
we use the semidefinite programming method to compute improved estimates of the maximum size of spherical two - distance sets .
exact answers are found for dimensions @xmath1 and @xmath2 where previous results gave divergent bounds .
dept . of electrical and computer engineering and institute for systems research ,
university of maryland , college park , md 20742 , and institute for information transmission problems , russian academy of sciences , moscow , russia .
email : [email protected] . ] dept . of mathematics and
institute for systems research , university of maryland , college park , md 20742 .
email : [email protected] ] |
superradiance , the collective emission of a sample of atoms , is one of the most celebrated examples of cooperative quantum phenomena , while synchronization is a widespread and paradigmatic emergent behavior in complex systems ( see @xcite and @xcite , respectively , for comprehensive reviews ) . in the most idealized form of superradiance , as first described by dicke in his seminal paper @xcite , the atoms identically interact with the electromagnetic field , a situation encountered , e.g. , when these are confined in a region of space smaller than the wavelength of the resonant modes .
when the initial atomic state is a highly entangled symmetric state , the emission is cooperative and takes place in a rapid burst at a rate proportional to the square of the number of atoms .
this is a clear signature of the presence of constructive quantum interference . if the emission starts from the state in which all the atoms are in their excited state , the collective atomic state acts as an antenna amplifying at a macroscopic level the quantum field fluctuation which trigger the emission . during the last years @xcite , superradiance in the presence of a small number of emitters
was observed by employing trapped ions @xcite and superconducting qubits @xcite .
superradiant emission has been shown to be a useful resource towards the realization of single - photon sources @xcite , quantum memories @xcite , and laser cooling @xcite .
in particular , one - dimensional quantum electrodynamics systems are very promising to observe superradiant emission , as the interaction can be maintained over large distances @xcite .
furthermore , such a flexible setup makes it possible to explore different coupling regimes , and then to study the transition from the case of independent emitters to the collective - decay limit .
the counterpart of superradiance , that is , subradiance , was also experimentally observed in the case of a pair of emitters @xcite , and more recently , in a large system of resonant scatterers in an extended and dilute cold - atom sample @xcite .
the aim of this paper is to discuss the connections of superradiance with spontaneous synchronization , a widespread phenomenon of adjustment of the dynamical rhythms of _ different _ interacting systems @xcite .
this emergent phenomenon has been largely studied in classical regimes and multidisciplinary contexts .
recently it has been predicted into the quantum regime , in several platforms including harmonic networks @xcite , mechanical resonators @xcite , atomic and ions setups @xcite .
different approaches have been considered to frame quantum synchronization both referring to local or global -e.g quantum correlation- measures ( see @xcite and references therein ) . for the purpose of this work ,
we address the collective quantum phenomenon of super and subradiance to establish a connection with spontaneous synchronization between the local dynamics of the atoms ; we therefore consider spontaneous synchronization as the adjustment to a common rhythm of local observables as in refs .
@xcite .
we mention that both synchronization and superradiance have also been considered in the presence of external driving . driven synchronization , or entrainment , in the quantum regime
have been considered for instance in refs .
@xcite . on the other hand , superradiance in the presence of driving has been identified as lasing with an extremely narrow linewidth @xcite , with a collective emission with an extremely high degree of phase coherence @xcite and quantum phase - locking @xcite .
it can in general be observed in incoherently pumped arrays of strongly interacting atomic quantum dipoles @xcite . indeed , the emission takes place in rapid bursts with large fluctuations in the emission delay time among the various experimental observations @xcite .
further features in the superradiant emissions that are a signature of synchronization take place when the emission is due to a sample of two different , detuned atomic species @xcite .
the aim of this work is to consider both superradiance and spontaneous synchronization taking place during dynamical relaxation into a common environment , in the absence of external driving , and in the presence of detuning between the atoms . in order to get a full analytic treatment , we consider within the liouville formalism the case of two detuned two - level systems in the presence of dissipation induced by a thermal environment , displaying both superradiant and subradiant emission and dynamical synchronization .
we note that , although collective effects are more prominent in the presence of large numbers of emitters , most features are already present in the decay of just two atoms coupled via dipole interactions @xcite .
we will show that , in the presence of pure dissipation , the two phenomena are different manifestations of a sole property of the model .
indeed , dynamical synchronization is due to the presence of coherence between the ground state and the subradiant state .
thus , this kind of synchronization has a pure quantum origin , as it directly relies on quantum interference .
the paper is organized as follows . in sec .
[ model ] , we introduce the model and the master equation used to solve the dynamics . in sec .
[ dmev ] , the evolution of the system is studied employing the so - called liouville representation for the density matrix . in sec . [
sec : synchronization ] , we introduce quantum synchronization and quantify it in our model .
super and subradiance are studied in sec .
[ subr ] and compared to synchronization .
the case of purely dephasing noise is the object of sec .
[ depnoise ] .
finally , conclusions are given in sec .
[ conclusions ] .
our system @xmath0 consists of two atoms that interact with a common bosonic environment @xmath1 .
we will show that this model , already considered in several works @xcite , allows for a fully analytical description of the origin of the connection between the phenomena of synchronization and of super and subradiance .
the two atoms ( described by qubits @xmath2 and @xmath3 ) have different frequencies , @xmath4 and @xmath5 , and their free hamiltonian is ( @xmath6 ) @xmath7 where @xmath8 and @xmath9 and @xmath10 indicate respectively the ground and the excited state of each atom .
the eigenstates of the free hamiltonian @xmath11 form the so - called decoupled basis @xmath12 with corresponding energies @xmath13 , where @xmath14 is the average frequency and @xmath15 is the detuning between atoms .
the total hamiltonian of the atoms interacting with the environment is @xmath16 , where @xmath17 corresponds to the radiation environment and @xmath18 is the interaction hamiltonian between the two atoms ( @xmath19 , with @xmath20 , is the dipole operator of atom @xmath21-th ) and the environmental field @xmath22 .
the interaction hamiltonian can be thus put under the form @xcite @xmath23 where @xmath24 . in the limit of weak system - bath coupling and for zero temperature , the dynamics of the reduced density matrix of the two atoms @xmath25 can be studied performing the born - markov and the secular approximation , taking care about the fact that , in the limit of small detuning @xmath26 , also terms oscillating with frequency @xmath27 must be kept .
the master equation reads @xcite @xmath28 + \sum_{ij } \gamma_{ij } \left ( \sigma_i^- \rho \sigma_j^+ -\frac{1}{2 } \left\ { \sigma_j^+ \sigma_i^-,\rho \right\ } \right),\ ] ] where @xmath29 , @xmath30 . in the following we use the notation @xmath31 and define @xmath32
while @xmath33 is assumed to be real ( its maximal value is given by @xmath34 ) .
the lamb - shift hamiltonian is @xmath35 the local shifts @xmath36 lead to irrelevant renormalization of the natural frequencies @xmath4 and @xmath5 . the different parameters appearing in the master equation are directly connected to the correlations functions of the environment , and thus to its spectral density .
henceforth , in order to explore different regimes , we assume to be able to tune the values of these parameters freely .
the dynamics of the system can be studied by adopting the so - called liouville representation of the density matrix . in this representation , the master equation ( [ masterequation ] )
is mapped into a set if linearly coupled differential equations for the elements of the reduced density matrix of the two atoms : @xmath37 , where @xmath38 is a vector in the hilbert - schmidt space , @xmath39 : @xmath40 where @xmath41 . the inner product in the hilbert - schmidt space is defined by @xmath42 .
as shown in appendix [ appendixl ] , the total liouvillian @xmath43 turns out to be the direct sum of 5 terms , @xmath44 , @xmath45 .
thus , @xmath46 can be decomposed into 5 independent subspaces , @xmath47 , whose dimensions are @xmath48 , @xmath49 , @xmath49 , @xmath2 and @xmath2 .
the form of each @xmath50 and the corresponding right ( @xmath51 ) and left ( @xmath52 ) eigenvectors , together with the eigenvalues , is explicitly reported in appendix [ appendixl ] .
we stress that these eigenvectors are not constrained to represent valid states .
in fact , as we will see later , the only physical state that is also an eigenvector of @xmath43 is the ground state , while the totally excited state is the only physical eigenvector of @xmath53 .
this implies that , for instance , a generic @xmath54 can not be taken as initial state . on the other hand ,
the dynamics of any density matrix can always be described as a superposition of such nonphysical objects , whose linear combination is constrained to be a density matrix . within this representation , a key role is played by @xmath55 being @xmath56 generated by the projectors of the four states forming the decoupled basis and by @xmath57 , @xmath58 . as shown in sec . [ eigenvectors and eigenvalues ] , the right ( r ) eigenvectors of @xmath55 ( eq .
( [ eigensysteml0 ] ) ) are simply expressed in terms of two nonorthogonal states , @xmath59 and @xmath60 , given by @xmath61 this parameter @xmath62 characterizes all the decay rates of the system ( we use the notation @xmath63 $ ] and @xmath64 $ ] ) appearing in the eigenvalues of @xmath55 ( eq .
( [ eigensysteml0 ] ) ) . in the symmetric configuration ,
when @xmath65 and @xmath66 , @xmath62 simply reduces to @xmath67 . on the other hand , the left ( l ) eigenvectors of @xmath55 ( eq
( [ leftl0 ] ) ) are given in terms of the states @xmath68 the property @xmath69 implies that @xmath70 .
( energy equal to @xmath71 ) to the ground state @xmath72 ( energy equal to @xmath73 ) ( see eq .
( [ rate equation ] ) ) . the channel passing through the state @xmath74 ( whose oscillations are associated to @xmath75 , see eq .
( [ eigensystemlx1 ] ) ) is characterized by a decay rate towards the ground state equal to @xmath76 ( @xmath77 ) .
the channel passing through the state @xmath78 ( whose oscillations are associated to @xmath79 ) is characterized by a decay rate towards the ground state equal to @xmath80 . for simplicity , the channels connecting @xmath81 and @xmath82 through the coherences between @xmath83 and @xmath84 are not shown.,scaledwidth=47.0% ] the states @xmath59 and @xmath83 , and @xmath60 and @xmath84 , appear to be a natural extension of the symmetric , @xmath85 , and antisymmetric , @xmath86 , states , and they reduce to them in the symmetric configuration ( @xmath65 and @xmath66 ) .
their relevance in the two - qubit dynamics is twofold .
on the one hand , @xmath59 and @xmath60 enter in the expression of the right eigenvectors @xmath87 , so that they can naturally be used to get simple expressions for the evolution of the density matrix operator ( see eq . ( [ rhot ] ) ) .
on the other hand , @xmath83 and @xmath84 enter in the expression of the left eigenvectors @xmath88 ( see eq .
( [ eig ] ) ) so that the rate equations for the evolution of the populations of the density matrix have the simple form reported in eq .
( [ rate equation ] ) .
this is schematically represented in fig .
[ figtrans ] , which shows that the states @xmath83 and @xmath84 connect the excited state to the ground state through two independent channels .
in particular , the decay rates from @xmath83 and @xmath84 to the ground state are given , respectively , by @xmath89 and @xmath80 , natural extension of the decay rates in the symmetric configuration , @xmath90 and @xmath91 .
the states @xmath83 and @xmath84 play indeed a role analogous to the one played by the superradiant state @xmath92 and by the subradiant state @xmath93 in the symmetric case .
we now show that the evolution of the density matrix operator is simply expressed in terms of the right eigenvectors of @xmath43 . by using the decomposition for the identity operator in the hilbert - schmidt space reported in eq .
( [ identity ] ) and @xmath37 , one can show that , given an arbitrary initial state @xmath94 , the evolved density matrix can be written ( apart from the case of degeneracy in the spectrum of the liouvillian , as in the case of the dicke limit when @xmath95 and @xmath66 ) as @xmath96 where @xmath97 runs over the five subspaces and @xmath21 between 1 and @xmath98 . the above equation is the main tool we are going to use in the next sections to make a connection between the phenomena of synchronization and of super and subradiance . before doing that
, we consider a couple of examples that will be useful in the following of this paper .
under particular initial conditions the dynamics is governed by only one decay rate ( different from zero ) .
if we start from @xmath99 ( i.e. , we prepare the system in the pure state @xmath60 ) , one can see from eq .
( [ eigensysteml0 ] ) that only two right eigenvectors of @xmath100 contribute to its dynamics , @xmath101 .
it follows that , in eq .
( [ rhot ] ) , @xmath102 ( the others @xmath103 are equal to zero ) so that at time @xmath104 we have @xmath105 + { \vert a_r a_r { \rangle \hspace{-0.2em } \rangle } } \mathrm{e}^{-\left(\gamma_0- v_\mathrm{r}\right)t}\ , \nonumber .\end{aligned}\ ] ] the above equation clearly shows that in this case the vector @xmath106 is dynamically coupled only to @xmath107 , its decay rate being given by @xmath108 .
its role is in this sense analogous to the one played by the subradiant antisymmetric state when @xmath65 and @xmath66 .
analogous considerations hold for the vector @xmath109 .
if we prepare the system in @xmath110 , this implies , using eq .
( [ eigensysteml0 ] ) , that @xmath111 .
it follows that , @xmath112 , so that the evolution takes the form @xmath113 + { \vert s_r s_r { \rangle \hspace{-0.2em } \rangle } } \mathrm{e}^{-\left(\gamma_0 + v_\mathrm{r}\right)t}\ , .
\nonumber\end{aligned}\ ] ] the role of the vector @xmath114 is thus analogous to the one played by the superradiant symmetric state in the case when @xmath65 and @xmath66 .
this section is devoted to the quantum synchronization phenomenon in our two - qubit system .
as we deal with a system relaxing towards a thermal state(at zero temperature , i.e. the ground state ) , the emergence of synchronization is assessed in a transient ( preasymptotic ) regime @xcite .
we focus on the case of spontaneous synchronization , emerging during the dynamics in the absence of pumping sources at the atoms locations .
quantum spontaneous synchronization can be quantified through different measures , and local and global indicators and as well as correlations have been used to this purpose , as reviewed in ref .
here we are interested in comparing the collective phenomenon of superradiance with the synchronous evolutions of the atoms ( local ) dynamics . to this aim , we compute the evolution of the average of arbitrary single atom operators @xmath115 ( @xmath116 ) , looking at their respective oscillatory ( generally irregular ) dynamics .
we consider the expectation values @xmath117 and @xmath118 where @xmath119 indicate the average of an arbitrary operator @xmath120 at time @xmath104 .
synchronization between these local observables is quantified through the pearson s correlation coefficient , a widely used measure of the degree of linear dependence between two variables . given two time - dependent variables @xmath121 and @xmath122 , the pearson s factor @xmath123 is @xmath124 where @xmath125 and @xmath126 are expectation values of quantum operators .
this synchronization measure is evaluated in time , considering a sliding window of length @xmath127 . as a consequence of the definition
, @xmath123 gives a value between @xmath128 and @xmath129 , where @xmath128 ( @xmath129 ) indicates the presence of in phase ( antiphase ) synchronization and @xmath130 the absence of any synchronization .
this measure can be generalized to the case of certain time delay between the local dynamics allowing for a phase shift , i.e. embodying a time delay @xmath131 in one of the two signals within the definition of @xmath123 as discussed in ref .
@xcite . in this way
, @xmath132 would indicate perfect time - delayed synchronization with delay @xmath131 .
the dynamical study conducted using the liouville representation allows us to establish how the time scale of the phenomenon of synchronization depends analytically on the physical parameters appearing in the master equation .
any pair of arbitrary local operators @xmath115 , @xmath116 can be written in the single atom basis ( bloch representation ) @xmath133 : @xmath134 by looking at the decomposition of the liouvillian operator ( see sec .
[ ldecomp ] ) , we find that only the right eigenvectors of @xmath135 and @xmath136 contribute to the average of @xmath137 and @xmath138 , while only the ones of @xmath139 contribute to the average of @xmath140 .
it follows that , using eq .
( [ rhot ] ) , we have @xmath141,\nonumber \\
\langle \sigma^y_k \rangle_t & = & \sum_{i } 2 \left| p_{0\ , i}^{b}\ , \langle \tau^b_i \rangle_{yk } \right| \ , \mathrm{e}^{\mathrm{re}(\lambda_i^b ) t } \ , \cos \left[\mathrm{im}(\lambda_i^b ) t+ \varphi_{i , yk}^b \right],\nonumber \\
\langle \sigma^z_k \rangle_t & = & \sum_{i } p_{0\ , i}^{a}\ , \langle \tau^a_i \rangle_{zk } \ , \mathrm{e}^{\lambda_i^a t},\end{aligned}\ ] ] where @xmath142 , @xmath143 , being @xmath144 , and where we used @xmath145 , @xmath146 and @xmath147 . here
, @xmath148 is the matrix obtained mapping the eigenstate @xmath149 back onto the hilbert space ( @xmath150 ) .
we remind that @xmath151 is not necessarily a density matrix , while any @xmath152 is well defined at any time .
( [ eigensysteml0 ] ) implies that only one frequency , equal to @xmath153 , enters in @xmath154 .
the term oscillating with this frequency decays with a rate @xmath155 . as for the averages of @xmath156 and @xmath157 ,
each of them contains four oscillating terms . in the worst - case scenario ,
when all the four terms contribute to these averages , the ratios between the instantaneous amplitudes of these oscillating terms are proportional to @xmath158t}$ ] . by using eq .
( [ eigensystemlx1 ] ) one sees that the terms with smallest decay rates are the one with @xmath159 and @xmath160 .
the long lasting ratio between the various terms and the one with smallest decay @xmath161 , decays as @xmath162t}=\mathrm{e}^{-v_\mathrm{r}t } $ ] . in a time scale given by some units of @xmath163
we expect that only one term still contributes to the averages of @xmath164 and @xmath165 .
this is the time scale separation enabling the emergence of spontaneous synchronization @xcite . on the other hand ,
the decay rate of the oscillating term in @xmath154 is always larger than @xmath166 .
the time scale for the occurrence of synchronization is thus characterized by a parameter @xmath167 equal to the real part of @xmath62 , while the frequency of the long lasting term , giving the frequency of the synchronized oscillations , is related to the imaginary part of @xmath62 : @xmath168 the real and imaginary part of @xmath62 obey @xmath169 .
the spontaneous emergence of synchronization in a time window is then due to the long lifetime of the eigenmode @xmath170 . in other word
the synchronization between averages of arbitrary local operators is due to the presence of long standing coherences between the ground state and the state @xmath60 , whose role in the dynamics has been already identified as analogous to the subradiant antisymmetric state .
this finding clearly points out in an analytical way the connection between synchronization and subradiant emission , which will be further detailed in the next section .
+ the above considerations allow us to determine the analytical dependence of synchronization on the physical parameters appearing in the master equation .
figure [ figsp ] shows the inverse time of synchronization emergence @xmath167 depending on @xmath171 and on @xmath172 , in the case @xmath173 . in particular , when @xmath174 , @xmath62 reduces to @xmath175 so that for @xmath176 , @xmath167 vanishes and no time scale separation ( synchronization ) is found @xcite . decreasing the atoms detuning a sharp transition to synchronization is predicted for @xmath177 .
this transition to a synchronous dynamics when decreasing the detuning is smoothed for @xmath178 as shown in fig .
[ figsp ] . in fig .
[ figsync1 ] the emergence of spontaneous synchronization is shown in the case when the initial state is @xmath179 .
the plot shows that , during the first stage of the evolution , the dynamics of the averages @xmath180 and @xmath181 are characterized by several frequencies while , after @xmath182 , only one term oscillating with the frequency @xmath183 survives . in the inset ,
the time evolution of the synchronization factor @xmath184 is plotted as a function of time .
the dashed curve shows that the two averages @xmath180 and @xmath181 become almost antisynchronized after some units of @xmath185 .
this means that while the two signals have become almost monochromatic , their dephasing is close ( but not equal ) to @xmath186 . the time delayed synchronization @xmath187 is represented by the solid line .
the delay @xmath131 is the one maximizing the value of @xmath123 . + as discussed at the end of sec .
[ dmev ] , there are particular choices of the initial state for which only a frequency is involved in the dynamics giving rise to synchronous dynamics from the initial time , as we show in fig .
[ figsync2 ] .
figure [ figsync2](a ) and fig .
[ figsync2](a ) refer , respectively , to @xmath188 and to @xmath189 . in both cases , by using eqs .
( [ eigensysteml0 ] , [ eigensystemlx1_b ] ) and ( [ mediaa ] ) it is easy to see that only one term is present in the averages of @xmath190 and @xmath181 . in fig .
[ figsync2](a ) the only frequency present is related to @xmath191 , while in fig . [
figsync2](b ) only @xmath192 contributes .
the two signals are synchronized from the beginning , being different both their decay rates and the relative phase between the two averages ( time delay ) .
in particular , starting from the initial state @xmath188 , the two averages result to be equal to @xmath193 , \nonumber \\ \langle \sigma^x_2 \rangle_t&= & \frac{\mathrm{e}^{-\frac{1}{2 } \left ( \gamma_0 -v_{\mathrm{r } } \right ) t } } { \sqrt{1+|\alpha_a|^2 } } \ , \cos \left[\nu_s t \right].\end{aligned}\ ] ] the relative delay of the two synchronized signals is then equal to @xmath194 ( in the plot a , this is @xmath195 ) .
+ analogously , when the the initial state is @xmath189 , the two averages are given by @xmath196 , \nonumber \\
\langle \sigma^x_2 \rangle_t&= & \frac{\mathrm{e}^{-\frac{1}{2 } \left ( \gamma_0 - v_{\mathrm{r } } \right ) t}}{\sqrt{1+|\alpha_s|^2 } } \ , \cos \left[\left(\omega_0 + \frac{1}{2}v_{\mathrm{i } } \right)t \right].\end{aligned}\ ] ] the relative delay of the two averages is now equal to @xmath197 ( in fig .
[ figsync2](b ) , this is @xmath198 ) .
the value of the coefficient @xmath199 can be taken as a measure of the transition between the common - bath ( @xmath200 ) and the separate - bath ( @xmath201 ) cases .
the transition between these two cases is discussed in ref . @xcite . in the limit of completely independent atoms
, we also have @xmath174 . in the case of independent atoms ( @xmath202 ) ,
@xmath190 and @xmath181 contain just one term decaying with rates respectively equal to @xmath203 and @xmath204 and oscillating with frequencies equal to @xmath4 and @xmath5 .
no synchronization is then possible in this limit case .
it is the presence of a collective environment which is responsible of the spontaneous synchronization described in this section .
in order to study the occurrence of super and/or subradiance we compute the evolution of the total radiation rate ( given in photons per second ) @xcite , @xmath205 given by the average of the operator @xmath206 .
all the matrix elements associated to the operators @xmath207 live in the liouvillan subspace @xmath139 .
it follows that @xmath208 is given by @xmath209 where @xmath210 .
in particular @xmath211 ( the ground state does not contribute to @xmath208 ) .
in the two cases when the initial states are chosen , respectively , as @xmath212 and @xmath114 , by using eqs .
( [ rhoat ] ) and ( [ rhost ] ) , one finds @xmath213 + the above equations describe how the two - atom system emits when initially prepared either in the state @xmath60 or in @xmath59 .
both emissions are characterized by only one decay rate , @xmath214 and @xmath215 respectively , displaying the known different radiances for nonvanishing @xmath216 .
this feature is a first quantitative link with the synchronization parameter @xmath217 defined in eq .
( [ sincronizzazione ] ) . when @xmath167 increases towards its maximum given by @xmath155 ( faster synchronization ) , the decay rates in eq .
( [ averageias ] ) go , respectively , to @xmath130 ( subradiance ) and to @xmath218 ( superradiance ) .
figure [ figsr1 ] shows the evolution of @xmath208 ( log scale ) for several initial conditions .
one sees that , by starting from @xmath59 or @xmath60 , the decay of @xmath208 is governed by only one decay rate ( see eq .
( [ averageias ] ) ) .
the same holds for the case of independent atoms ( case i. a. : @xmath219 ) having individual decay rates equal to @xmath155 . for any other initial states ,
the evolution of @xmath208 is not an exponential as it depends on more than one decay rate .
starting from the symmetric , @xmath92 , and antisymmetric , @xmath93 , states , the total radiation rates are initially the larger and the smaller among all the possible one - excitation states . however , during the dynamics , part of the population remains trapped in @xmath60 , whose decay rate ( the decay rate of the eigenvector @xmath220 ) governs the final part of the evolution of @xmath208 .
similar considerations holds for the initial states @xmath221 and @xmath222 ( the states that diagonalize the dressed hamiltonian @xmath11 + @xmath223 of eq .
( [ masterequation ] ) ) . also the case in which initially both atoms
are excited , the condition typically considered in the studies of super and subradiance , is plotted .
+ in general ( i.e. for most atoms initial states ) , the large - time behavior of @xmath224 is governed by the long - lasting eigenvector : @xmath225 , that is the subradiant contribution of @xmath226 weighted by the overlap between this eigenmode and the considered initial state .
furthermore , the ratios between the other contributions and the one with the slowest decay , decay all faster than @xmath227-\lambda_6^a ) t}=\mathrm{e}^{(\mathrm{re}[\lambda_4^a]-\lambda_6^a ) t}=\mathrm{e}^{- v_\mathrm{r } t}$ ] , that is with a characteristic time which is equal to the synchronization time ( see eq .
( [ sincronizzazione ] ) ) , @xmath228 .
this clearly shows how the time scales of synchronization and of subradiant emission are intimately related .
this analysis establishes analytically that the two phenomena under study , that is , synchronization and sub and subradiance , occur at the same time scale .
this is displayed in fig .
[ figcomparison ] representing the delayed synchronization measure @xmath229 defined in eq .
[ eq : pears ] and a subradiance parameter @xmath230 , defined as the minimum between @xmath231 and its inverse @xmath232 .
@xmath230 is such that it goes to 1 when @xmath208 approaches its large time value @xmath233 ( from above or below ) .
@xmath229 is computed using in eq .
( [ eq : pears ] ) @xmath234 as local operators and allowing for a time delay between the two signals .
the comparison is done by varying @xmath199 from @xmath130 to its maximum @xmath155 ( @xmath173 in the figure ) .
the above parameters are computed at a time @xmath235 such that for low values of @xmath199 , @xmath190 and @xmath236 are not yet synchronized and @xmath208 contains still more than one contribution from the eigenstates of @xmath139 . by increasing @xmath199
, @xmath216 increases and we may observe at the chosen time @xmath237 the occurrence of synchronization and @xmath208 approaching its subradiant contribution @xmath238 .
so far we have discussed the case of a fully dissipating environment and found a perfect connection between the time scales of emergence of synchronization and superradiance . as discussed in ref .
@xcite , the presence of decohering noise even if into a common environment does not favor synchronization . here
, we also show that it alters the interplay with superradiance .
although the presence of decoherence can be treated exactly @xcite , in the weak - coupling limit dephasing effects can be included by adding a phenomenological term to the master equation of eq .
( [ masterequation ] ) @xcite : @xmath239 where @xmath240 is assumed to be real and @xmath241 .
one can verify that this extra term does not modify the decomposition into the direct sum of sectors of the new lindbladian .
in particular , looking at @xmath242 , its new eigenvalues @xmath243 are all equal to the old ones apart from the common term @xmath244 that must be subtracted to all of them .
it follows that , analogously to what found is sec .
[ sec : synchronization ] the time scales for the emergence of quantum synchronization ( survival of one term in the average of arbitrary single atom operators , with respect to the other contributions ) remains the same , given by @xmath228 .
however , the amplitude of this survival term now is lowered at any time @xmath104 by a factor @xmath245 . as for @xmath246 ,
the situation is different and we can distinguish two main cases . in the case
when @xmath247 , @xmath248 is not modified and our previous analysis remains the same .
when this condition is not satisfied , in the general case we do not have analytical expressions for the eigenvalues @xmath55 any more .
however , a numerical analysis shows that the ratio between the exponentials governed by the two smallest ( in modulus ) decay rates now depends on the value of @xmath249 .
this means that in general the time scales of spontaneous synchronization and of the emergence of subradiant emission are not equal any more .
superradiance is a genuinely collective phenomenon , as it can not be explained by taking only individual constituents of the whole atomic system .
the main question addressed in this paper concerns the possible evidence of such a global effect on the dynamics and synchronization of local observables .
in fact , we have found a strict analytical relationship between subradiance and the dynamical synchronization of local trajectories .
considering two detuned qubits coupled to a dissipative bath , by means of the liouville formalism , we have studied the microscopic mechanisms responsible for the two phenomena and found that , actually , synchronization and super and subradiance are two sides of the same coin in our model . as a matter of fact ,
the synchronization between averages of arbitrary single atom operators is due to the presence of long standing coherences between the ground state and a state that can be identified as the responsible for subradiant emission , whose experimental detection is known to be more demanding @xcite .
the fact that the decay rate due to such a subradiant state is twice the decay rate of the coherence responsible for synchronization clearly shows that the two phenomena are intimately related and occur on the same timescale .
the case of two atoms we have discussed here offers the advantage that a fully analytical treatment is available which makes patent the liaison between subradiance and synchronization .
the comparison in the case on ensemble of @xmath250 atoms can not be treated analytically and requires numerical or approximated descriptions , like the mean field adopted for instance in ref.@xcite to assess the steady - state in presence of driving . on the other hand , by adding the effects of a dephasing noise , the correspondence between synchronization and collective super and subradiance effects can be altered .
the occurrence of synchronization in classical systems as well as its definition in a broad spectrum of regimes is well established @xcite . when it comes to the quantum realm , it becomes interesting to identify specific quantum signatures of this effect .
recently , there have been many attempts to quantify the appearance of synchronization also through quantum indicators of ( global ) correlations @xcite .
our work represents an original contribution also in this context given that it offers an explanation of spontaneous synchronization of local observables as the manifestation of collective quantum interference .
in other words , the present analytical description through local generic observables dynamics allows to establish how synchronization witnesses a collective quantum phenomenon of subradiance , representing then a genuine quantum feature of synchronization , intimately different from its classical counterpart .
thanks d. viennot for useful discussions .
fundings from eu project quprocs ( grant agreement no .
641277 ) , mineco ( grant no .
fis2014 - 60343-p ) , `` vicerectorat dinvestigaci
i postgrau '' through the visiting professors program of the uib and from `` soutien aux enseignants - chercheurs et chercheurs - nouveaux arrivants 2016 '' of the ufc are acknowledged .
to treat the dynamics of the system , we adopted in sec .
[ dmev ] the so - called liouville representation of the density matrix , defined by @xmath251 where @xmath41 .
the inner product in the hilbert - schmidt space is defined by @xmath42 .
it follows that for any operator of the system , @xmath252 , we have @xmath253 where @xmath254 denotes the matrix transposition of @xmath252 and @xmath255 the four dimension identity matrix . by using these rules , we obtain the liouville representation of the master equation of eq .
( [ masterequation ] ) , @xmath256 , where @xmath257 is defined by : @xmath258 , \nonumber\end{aligned}\ ] ] where @xmath259 .
the liouville representation of the master equation is thus a schrdinger equation governed by a nonself - adjoint generator @xmath260 . by using eq .
( [ costruzionel ] ) , one finds that the total liouvillian can be decomposed in five jordan blocks .
the hilbert - schmidt space @xmath261 can be thus decomposed into five independent subspaces : @xmath262 , @xmath45 , with @xmath263 , @xmath264 and @xmath265 , such that @xmath266 ( @xmath267 being an operator of @xmath268 ) .
@xmath269 is generated by the basis @xmath270 , @xmath271 , @xmath57 , @xmath58 , @xmath272 , @xmath273 , @xmath274 by @xmath275 , @xmath276 , @xmath277 , @xmath278 , @xmath279 by their complex conjugate counterparts @xmath280 , @xmath281 , @xmath282 , @xmath283 , @xmath284 by @xmath285 and @xmath286 by its conjugate @xmath287 .
let @xmath296 , @xmath297 and @xmath298 be the eigenvalues , the right , and the left instantaneous eigenvectors of @xmath43 .
they are defined by @xmath299 we can write the identity in the hilbert - schmidt space , @xmath300 , by means of the right and the left instantaneous eigenvectors of @xmath43 as : @xmath301 where @xmath303 and @xmath304 are the projectors of the kets @xmath60 and @xmath59 , defined in eq .
( [ stilde e atilde ] ) , and where @xmath305 , @xmath306 , @xmath307 and @xmath308 have a complicate dependence on the various parameters of the model , not reported here .
the left eigenvectors in the sector @xmath55 are ( we report the right eigenvectors of @xmath311 ) @xmath312 where @xmath313 is the vector associated to the identity in the four dimension space ( in the decoupled basis it is @xmath314 ) and where @xmath315 [ [ mathcall_b - mathcall_c - mathcall_d - and - mathcall_c ] ] @xmath292 , @xmath291 , @xmath293 and @xmath291 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ the eigenvalues of @xmath292 are @xmath316-i \omega_0 , \\
\lambda_2^b=-\frac{1}{2 } \left [ 3\gamma_0 -v^ * \right]-i \omega_0 , \\
\lambda_3^b=-\frac{1}{2 } \left ( \gamma_0+v\right)-i \omega_0 , \\\lambda_4^b=-\frac{1}{2 } \left(\gamma_0 -v \right)-i \omega_0 .
\end{cases}\end{aligned}\ ] ] concerning the right and the left eigenvectors of @xmath292 we found the analytical expressions for @xmath317 and @xmath318 ( we report the right eigenvectors of @xmath319 ) : @xmath320 and @xmath321 the expressions for the other eigenvectors are particularly cumbersome and are then not reported here . here
we report the equation of motions for the density matrix elements belonging to the sector @xmath55 . in the decoupled basis
the rate equations for the populations and the coherences between @xmath328 and @xmath329 can be immediately derived using the form of @xmath55 of eq .
( [ l0 ] ) ( we use the notation @xmath330 ) . in this basis , the populations of @xmath328 and @xmath329 are coupled with their coherences .
the special role played by the states @xmath331 and @xmath332 emerges clearly when we derive the equation of motions for their populations , together with the equation of motions for their coherences , and for the ground state .
these can be obtained by recombining the rate equations in the decoupled basis , exploiting eq .
( [ stilde e atilde ] ) , or , equivalently by using the left eigenvectors of eq .
( [ leftl0 ] ) and the relation , for an arbitrary operator @xmath252 , @xmath333^ * $ ] : @xmath334 @xmath335 \rho_{e , e } \nonumber \\ & & - \left ( \gamma_0 + v_{\mathrm{r}}\right ) \rho_{s_l , s_l } , \nonumber \\
\frac{\mathrm{d } \rho_{s_l , a_l } } { \mathrm{d}t } & = & |\alpha_a| \frac{\gamma_1 + \gamma_2 \alpha_s
\alpha_a^*+\gamma_{12}(\alpha_s+\alpha_a^*)}{(1+|\alpha_a|^2 ) } \rho_{e , e } \nonumber \\ & & - \left ( \gamma_0 + i v_{\mathrm{i}}\right ) \rho_{s_l , a_l } , \nonumber \\ \frac{\mathrm{d } \rho_{a_l , s_l } } { \mathrm{d}t } & = & |\alpha_a| \frac{\gamma_1 + \gamma_2 \alpha_a \alpha_s^*+\gamma_{12}(\alpha_a+\alpha_s^*)}{(1+|\alpha_a|^2 ) } \rho_{e , e } \nonumber \\ & & - \left ( \gamma_0 - i v_{\mathrm{i}}\right ) \rho_{a_l , s_l } , \nonumber \\ \frac{\mathrm{d } \rho_{a_l , a_l } } { \mathrm{d}t } & = & \left[\gamma_0-v_\mathrm{r}+(\gamma_1-\gamma_2)\frac{1-|\alpha_a|^2 } { 1+|\alpha_a|^2}\right ] \rho_{e , e } \nonumber \\ & & - \left ( \gamma_0 -v_{\mathrm{r}}\right ) \rho_{a_l , a_l } , \nonumber \\ \nonumber \frac{\mathrm{d \rho_{g , g } } } { \mathrm{d}t } & = & \frac{1+|\alpha_a|^2}{|1+\alpha_a^2|^2 } \big [ ( 1+|\alpha_a|^2 ) \left(\gamma_0+v_\mathrm{r } \right)\nonumber \\ & & \!\!\!\!\!\ ! \!\!\!\!\!\ ! \!\!\!\!\!\ ! \!\!\ ! \times \rho_{s_l , s_l } + \big(- \gamma_1 |\alpha_a| + \gamma_2 \frac{\alpha_a^{*\ , 2 } } { |\alpha_a| } + \gamma_{12 } \nonumber \\ & & \!\!\!\!\!\ ! \!\!\!\!\!\ ! \!\!\!\!\!\ ! \!\!\
! \times \big(\alpha_a|\alpha_a| -\frac{\alpha_a}{|\alpha_a| } \big ) \big ) \rho_{s_l , a_l } + \big(- \gamma_1 |\alpha_a| \nonumber \\ & & \!\!\!\!\!\ ! \!\!\!\!\!\ ! \!\!\!\!\!\ ! \!\!\ !
+ \gamma_2 \frac{\alpha_a^ { 2 } } { |\alpha_a| } + \gamma_{12 } \big(\alpha_a^*|\alpha_a| -\frac{\alpha_a^*}{|\alpha_a| } \big ) \big ) \rho_{a_l , s_l } \nonumber \\ & & \!\!\!\!\!\ ! \!\!\!\!\!\ ! \!\!\!\!\!\ ! \!\!\ ! + ( 1+|\alpha_a|^2)\left(\gamma_0 -v_\mathrm{r } \right ) \rho_{a_l , a_l } \big].\end{aligned}\ ] ]
duan , m. d. lukin , j. i. cirac , and p. zoller , nature ( london ) * 414 * , 413 ( 2001 ) ; r. reimann , w. alt , t. kampschulte , t. macha , l. ratschbacher , n. thau , s. yoon , and d. meschede , phys .
lett . * 114 * , 023601 ( 2015 ) .
g. heinrich , m. ludwig , j. qian , b. kubala , and f. marquardt , phys .
lett . * 107 * , 043603 ( 2011 ) ; m. ludwig and f. marquardt , phys .
lett . * 111 * , 073603 ( 2013 ) ; a. mari , a. farace , n. didier , v. giovannetti , and r. fazio , phys . rev
. lett . * 111 * , 103605 ( 2013 ) .
i. goychuk , j. casado - pascual , m. morillo , j. lehmann , and p. hnggi , phys .
97 * , 210601 ( 2006 ) ; o. v. zhirov and d. l. shepelyansky , phys . rev . lett . * 100 * , 014101 ( 2008 ) ; s. b. shim , m. imboden , and p. mohanty , science * 316 * , 95 ( 2007 ) .
g. m. palma , a. vaglica , c. leonardi , f. a. m. deoliveira , and p. l. knight , opt .
79 * , 337 ( 1990 ) .
g. m. palma , a. vaglica , c. leonardi , and a. k. ekert , opt.comm . * 86 * , 192 ( 1991 ) . c. leonardi and a. vaglica , j. phys .
b : at . mol .
* 14 * l307 ( 1981 ) ; opt . commun . * 53 * , 340 ( 1985 ) .
z. ficek , r. tana , and s. kielich , physica a * 146 * , 452 ( 1987 )
. g. m. palma and p. l. knight , phys . rev.a * 39 * , 1962 ( 1989 ) ; a. k. ekert , g. m. palma , s. m. barnett , and p. l. knight , phys .
a * 39 * , 6026 ( 1989 ) .
z. ficek and r. tana , phys . rep . *
372 * , 369 ( 2002 ) .
g. s. agarwal , in _ quantum statistical theories of spontaneous emission and their relation to other approaches , edited by g. hhler _ ( springer tracts in modern physics vol .
70 , berlin : springer , pp 1 - 128 , 1974 ) . | we study the relationship between the collective phenomena of super and subradiance and spontaneous synchronization of quantum systems . to this aim
we revisit the case of two detuned qubits interacting through a pure dissipative bosonic environment , which contains the minimal ingredients for our analysis . by using the liouville formalism ,
we are able to find analytically the ultimate connection between these phenomena .
we find that dynamical synchronization is due to the presence of long standing coherence between the ground state of the system and the subradiant state .
we finally show that , under pure dissipation , the emergence of spontaneous synchronization and of subradiant emission occur on the same time scale .
this reciprocity is broken in the presence of dephasing noise . |
one of the most remarkable and fundamentally important result of the field quantization is the casimir effect which is a force arising from the change of the zero point energy caused by imposing the boundary conditions ( bc ) @xcite .
this force is the macroscopic aspect of the quantum electrodynamics that provides a direct line between quantum field theory and the macroscopic world .
the original calculation of the casimir force between two perfectly conducting parallel plates immersed in the quantum electromagnetic vacuum is based on the definition of the casimir energy in the presence and the absence of boundary surfaces @xcite that leads to an attractive observable force @xmath0 between the plates , where @xmath1 is the planck constant , @xmath2 is the speed of light and @xmath3 is the distance between the plates .
in other way one can consider this effect by evaluating the radiation pressure on macroscopic objects @xcite . as the magnitude of the casimir force is substantial at @xmath4
this effect is relevant in nano - technology @xcite and should be take into account to design and actuate microelectromechanical ( mems ) systems . moreover the possibility of transducing the energy from the vacuum is investigated by means of mems @xcite .
many attempts have been focused on observing the casimir force and performing high - precision measurements during last few years @xcite .
all these experiments are in agree with the prediction of the casimir @xcite within a few percents . these deviation from the ideal force
may due to temperature , roughness of surfaces and finite dielectric constants that have been covered in a very recent book entitled by advanced in the casimir effect @xcite .
in 90th decade , golestanian and kardar developed a path integral approach to investigate the dynamic casimir effect in the system of two corrugated conducting plates surrounded by the quantum vacuum @xcite .
emig and his colleagues also used the path integral formalism to obtain normal and lateral casimir force between two sinusoidal corrugated perfect conductor surfaces @xcite . later on , the exact mechanical response of the quantum vacuum to the dynamic deformations of a cavity and the rate of dissipation have been calculated by using path integral scheme @xcite .
this motivated us to investigate the casimir effect in the presence of a magnetodielectric medium by quantizing the electromagnetic ( em ) field using path integral formalism .
our system contains of a magnetodielectric medium with permitivity @xmath5 and permeability @xmath6 , enclosed by two semi - infinite ideal metals ( @xmath7 for the ideal metal in left - hand side and @xmath8 for right - hand side one ) as depicted in fig .
( [ schematic - fig ] ) .
we model the magnetodielectric medium by a continuum of harmonic oscillators ( hopfield model ) @xcite . )
is enclosed between two perfect parallel conductors ( @xmath9 and @xmath10 ) .
the distance between conductors is @xmath3 , and the @xmath11 direction is perpendicular to the surfaces of the media . ]
the outline of this paper is as follows : in sec.[path - scalar ] , in order to introduce the scheme , we first quantize the simple case of a scaler klein - gordon field in the presence of a medium . sec .
[ force - scalar ] is devoted to obtain the casimir force for the scalar filed in the presence of a medium for different kinds of boundary conditions . in sec.[path - electromagnetic ] , we develop our formalism to the case of em field in the presence of a magnetodielectric medium .
sec.[force - electromagnetic ] gives the casimir force for the case of em field .
finally , the conclusions and outlooks are in the sec .
[ conclusion ] .
to illustrate the method and also for later convenience , before considering the em field in the presence of a medium , we consider the simplest case i.e. a scaler massless field . in the next section
we will show that the klein - gordon field can be corresponded to each polarization of the em field .
therefore , let us consider the following lagrangian for the total system @xmath12 where @xmath13 is the lagrangian density of the massless klein@xmath14gordon filed .
the medium is modeled by a continuum of harmonic oscillators as @xcite @xmath15 where @xmath16 is an oscillator s field , @xmath17 is the density of matter field and the interaction between the system and its medium is defined by @xmath18 where @xmath19 in the next section we will show that the quantity @xmath20 is in fact the polarization field corresponding to the medium and the interaction ( [ 5 ] ) will become the electric - dipole interaction . for further use
we define @xmath42 the generating function of the interacting fields can be written in terms of the free generating function as @xcite @xmath43 & = & z^ { - 1 } [ 0]e^{\imath \int d^{n+1 } z{\cal{l}}_{int } ( \frac{\delta } { \delta j_\varphi ( z)},\frac{\delta } { \delta j_p(z ) } ) } z_0 [ j_{\varphi } , j_\omega ] \nonumber\\ & & \hspace{-1.8 cm } = z^ { - 1 } [ 0 ] \sum\limits_{n = 0}^\infty \frac{1}{n!}\bigg [ \imath\int d^{n+1}z\frac{\delta } { { \delta j_\varphi(z ) } } \cdot \frac{\partial } { { \partial z_0}}\frac{\delta } { { \delta j_{\varphi } ( z ) } } \bigg]^n
\times \nonumber\\ & & \hspace{+0.6 cm } \times z_0 [ j_{\varphi } , j_\omega ] , \end{aligned}\ ] ] where @xmath44 $ ] is the partition function of the free space .
thus the green s function of klein - gordon filed can be obtained via @xmath45}}{{\delta j_{\varphi } ( x)\delta j_{\varphi } ( y)}}\big |_{j_{\varphi},j_{\omega } = 0}.\ ] ] combining the generating function ( [ 18 ] ) , the green s function ( [ 19 ] ) and the definition of @xmath46 ( [ 5 ] ) yield the following series for the green s function @xmath47 it is appropriate to use a general green s function in @xmath28-dimensional fourier space that is @xmath48 therefore the green s functions of the free fields @xmath49 and @xmath16 in the fourier space might be given by @xmath50 and @xmath51 respectively . since we are interested in retarded green s functions we have added @xmath52 to the denominator of the equation ( [ 23 ] ) . since the reservoir field is assumed to be homogeneous , the green s function of the reservoir does not depend on @xmath53 in the above equation . using the equations ( [ 22 ] ) and ( [ 23 ] ) , @xmath54 can be written in the fourier space as @xmath55^n\}\nonumber\\ & & \hspace{-1.8 cm } = \frac{g_0 ( { \bf k},\omega ) } { 1 - \int d\omega'\omega^2\nu ^2 ( \omega ' ) g_{\omega'}(\omega)g_0 ( { \bf k},\omega ) } \nonumber\\ & & \hspace{-1.8 cm } = \frac{1}{{{\bf k}^2-\omega ^2- \frac{1}{\rho}\int d\omega'\frac{\nu ^2 ( \omega')\omega^2 } { \omega ' ^2 -\omega ^2 + \imath 0^+ } } } .\end{aligned}\ ] ] this green s function can also be obtained directly from the heisenberg equations of motion . by direct substitution
we can show that the green s function @xmath56 satisfies the equation @xmath57 which is the motion equation of dissipation field with susceptibility of the medium @xmath58 , with the following fourier transform @xmath59 from the equations ( [ 24 ] ) and ( [ 26 ] ) it is clear that the modified green s function @xmath60 can be obtained from the free field green s function @xmath61 in eq .
( [ 15 ] ) simply by replacing @xmath62 with @xmath63 , where @xmath64 .
it can be easily shown that this susceptibility satisfies the kramers - kronig relations as expected . by the same technique we can obtain correlation between the polarization field and the klein - gordon field . to this end we define @xmath65 as @xmath66}{\delta j_\varphi \delta j_p},\ ] ] that can be obtained via direct calculation similar to the procedure that ended to @xmath60 as @xmath67 where @xmath68 if we again write @xmath69 in the fourier space we find @xmath70^n \big\}\nonumber\\ & = & \frac{g_{0p } ( { \bf k},\omega ) g_0 ( { \bf k},\omega ) } { 1 - \int d\omega'\omega^2\nu ^2 ( \omega ' ) g_{\omega'}(\omega)g_0 ( { \bf k},\omega ) } \nonumber\\ & = & \frac{g_{0p } ( { \bf k},\omega ) } { { { \bf k}^2-\omega ^2- \frac{1}{\rho}\int d\omega'\frac{\nu ^2 ( \omega')\omega^2 } { \omega ' ^2 -\omega ^2 + \imath 0^+ } } } .\end{aligned}\ ] ] comparing eqs .
( [ 16 ] ) and ( [ 29 ] ) , yields @xmath71 , consequently @xmath72 can be rewritten as @xmath73 the other important correlation function is @xmath74 which is defined via generating function @xmath75 $ ] as @xmath76}{\delta j_p\delta j_p}.\ ] ] by straightforward calculations one can obtain @xmath77 as @xmath78 the imaginary part of the response function can be read from eq .
( [ 26 ] ) as @xmath79 . here to illustrate the validity of our results , i.e. eq .
( [ 31]-[33 ] ) , we compare them with the results of the other methods of field quantization .
in other conventional methods of phenomenological field quantization @xcite , the fields can be divided into positive ( @xmath80 ) and negative ( @xmath14 ) frequencies parts which satisfy the constitutive relation @xmath81 after quantization of the fields , where by @xmath82 we mean operator and the operator with positive frequency is the hermitian conjugate of the negative one .
@xmath83 is the noise part of the polarization field that according to the fluctuation - dissipation theorem @xcite satisfies @xmath84= \pi im\chi(\omega)\delta({\bf x}-{\bf x}'),\ ] ] which @xmath85 $ ] denotes the commutator of two operators .
if we use the eqs .
( [ 34 ] ) and ( [ 35 ] ) to obtain the green s functions , we achieve the same results as the eqs .
( [ 31]-[33 ] ) .
this shows the validity of our path integral quantization . according to the definition of the green s functions , after integrating over gaussian fields one can read the generating function ( [ 18 ] ) as @xmath86&&=exp\bigg\ { \imath \int dx \int dx ' \big [ j_\varphi g_{\varphi,\varphi}(x - x')j_\varphi \nonumber\\ & & + j_\varphi g_{p,\varphi}(x - x')j_p+j_pg_{p , p}(x - x')j_p \big ] \bigg\}. \nonumber\\\end{aligned}\ ] ]
in this section we briefly review the path integral technique to calculate the casimir force .
let us consider two conducting plates faced each other at the distance @xmath3 and embedded in an arbitrary medium .
the field @xmath49 satisfies the dirichlet @xmath87 or neumann @xmath88 boundary conditions on surface , where @xmath89 is an arbitrary point on the @xmath90th conducting plate . to obtain the partition function from the lagrangian we use the wick s rotation , @xmath91 and change the signature of the space - time from minkowski to euclidean .
the diriclet or neumann boundary conditions can be taken into account using the auxiliary fields @xmath92 @xcite @xmath93e^{\imath\int d x_\alpha \psi ( x_\alpha ) \varphi ( x_\alpha ) } .\ ] ] and @xmath94e^{- \imath\int d x_\alpha { \partial_n}\psi ( x_\alpha ) \varphi ( x_\alpha ) } .\ ] ] after wick s rotation the dirichlet and neumann partition functions can be cast into the form @xmath95 \prod\limits_{^{a = 1 } } ^2 { \cal d}[\psi_\alpha ( x_\alpha ) ] ) e^{s_d[\varphi ] } , \ ] ] and @xmath96 \prod\limits_{^{a = 1 } } ^2 { \cal d}[\psi_\alpha ( x_\alpha ) ] ) e^{s_n[\varphi ] } \ ] ] respectively , where @xmath97 is the partition function of the free space , and @xmath98 = \int d^{(n+1 ) } x \nonumber\\ & & \times \big\ { { \cal l } \big ( \varphi(x ) \big)+ \varphi ( x)\!\ ! \sum\limits_{\alpha = 1}^2 \ ! \int \!\ !
d^{(n ) } x\delta ( x \!\ ! - \!\ !
x_\alpha ) \psi _
\alpha ( x ) \big\}.\end{aligned}\ ] ] and @xmath99 = \int
d^{(n+1 ) } x & & \nonumber\\ & & \hspace{-3.8 cm } \times \{{\cal l } ( \varphi(x))+ \varphi ( x)\sum\limits_{\alpha = 1}^2 \int d^{(n ) } x\delta ( x - x_\alpha ) \partial_n\psi _ \alpha ( x)\}.\end{aligned}\ ] ] using the same procedure of ref .
( @xcite ) the parttion functions for the dirichlet and neumann bc can be read @xmath100 and @xmath101 where @xmath102,\ ] ] and @xmath103,\ ] ] where @xmath104 is the green s function of the fields after wick rotation .
we define the effective action as @xmath105 where @xmath106 can be either for the dirichlet or neumann bc , in order to calculate the casimir force by applying derivative with respect to the distance between the plates @xmath107 it is easy to show that the contribution of the dirichlet is the same as that of the neumann bc to the casimir energy and hence the casimir force in the presence of a isotropic and homogenous medium like @xcite . so that in the next sections we treat only the dirichlet bc . in this section
we would like to obtain the casimir force in the presence of an absorptive medium .
before we obtain the casimir force for interacting fields , we investigate the possibility of the existence of the casimir force due to the matter field alone . using the lagrangian ( [ 4 ] ) and expression for the effective action ( [ 56 ] )
we find the @xmath108 tensor as @xmath109.\ ] ] but since @xmath110 for this situation , the noninteracting matter field alone , does not lead to any modified casimir force .
this result is clear since we model the matter field by the hopfield model * * * * * * [ hopfield s reference]******. this model is based on an independent set of harmonic oscillators and imposing any condition on one of these oscillators does not affect the others , and hence we do not expect any casimir effect . for a dissipative field @xmath49 ( [ 25 ] ) we may consider three different boundary conditions , * i * ) imposing the boundary condition on the klein - gordon field : for this case the @xmath111 tensor can be read @xmath112.\ ] ] since @xmath113 is diagonal in the fourier space , to obtain the casimir force we proceed in this space .
the fourier transformation of @xmath114 is @xmath115 where @xmath116 , @xmath117 is a vector parallel to the conductor , @xmath118 the temporal component of the @xmath119 , @xmath120 and @xmath121 .
thus for the case * i * the casimir force is @xmath122,\ ] ] where @xmath123^{1/2}$ ] . in the absence of the medium between the conductors , @xmath124
, we recover the original casimir force between two plates immersed in the quantum vacuum of a scalar field @xmath125 and for a non absorptive medium with the susceptibility @xmath126 , we find the modified casimir force as @xmath127 the above relation is fully in agree with the result of the lifshitz theory of fluctuation - induced force bewteen media @xcite,@xcite .
the equation ( [ 61 ] ) is interesting since it is the reminiscent of the bose - einstein distribution .
in fact , @xmath128 can be interpreted as the force density due to the bosons in the state @xmath119 .
* ii * ) imposing the boundary condition on the polarization field : in this case @xmath129 tensor is @xmath130.\ ] ] here the casimir force is @xmath131,\ ] ] where @xmath132 .
although the noise operators do not have any spatial correlation but their presence on the surface can affect the casimir force and decrease it due to polarization .
* iii * ) imposing the boundary condition on the both of polarization and klein - gordon fields : we can easily show that @xmath133 is a @xmath134 tensor with the form of @xmath135 , \nonumber\\\end{aligned}\ ] ] where @xmath136 is the @xmath137 unit matrix multiplied by @xmath138 .
it can be easily shown in this case that the casimir force will be the same as that of imposing the boundary condition only on the klein - gordon field i.e. case * i*. we can interpret this situation with the aid of eq .
( [ 34 ] ) .
according to this relation if the bc is imposed only on the klein - gordon field then @xmath139 is not zero , and if the bcs are imposed on the both of polarization and klein - gordon fields then the bc will be imposed on @xmath139 automatically which according to the begining of the sec .
[ casimir - different - bc - scalar ] does not lead to aany casimir effect .
the physical interpretation of the first and third condition is obvious .
these conditions arise when we want to calculate the casimir force between two perfect conductors that enclose a medium .
but the second bc can happen when we want to consider the system of containing two dielectric slabs .
this kind of boundary condition leads to the casimir force between two dielectric slabs which is under consideration .
here we only consider the boundary conditions in cases * i * and * iii * , and we shall treat the case * ii * in elsewhere @xcite .
in this section we develop the formalism to em field in the presence of a magnetodielectric medium @xcite .
the lagrangian of em field in the presence of a medium can be written as @xmath140 where @xmath141 is @xmath142 where @xmath143 and @xmath144 are the electric and magnetic fields .
they can be written in terms of scalar and vector potentials @xmath145 and @xmath146 respectively as @xmath147 and @xmath148 . in this work
we use the coulomb gauge @xmath149 , i.e. @xmath146 is a transverse field . @xmath150 and @xmath151 reffer to the polarization and magnetization of the medium respectively and can be written as @xmath152 where @xmath153 is an oscillator s vector field .
the interaction part of the lagrangian is @xmath154 where @xmath155 and @xmath156 are polarization and magnetization of the medium defined by @xmath157 the euler - lagrange equations lead us to the fact that @xmath158 is not an independent dynamical variable . to obtain @xmath145 in terms of the other independent dynamical variables ,
it is appropriate to write the fields in fourier space where the longitudinal and transverse parts of a field can be separated using the unit vectors @xmath159 and @xmath160 ( @xmath161 ) respectively . @xmath162 and @xmath163 are perpendicular to each other and @xmath164 . using these unit
vectors the scalar potential can be written in terms of the longitudinal part of the matter field as ( in what follows we indicate the fields in fourier space by a @xmath165 over them . ) @xmath166 by applying this recent relation to the eq .
( [ 67 ] ) , we can easily show that the longitudinal part of the electromagnetic field only changes the longitudinal component of @xmath150 as @xmath167 where @xmath168 and @xmath169 .
it is worth noting that only the transverse parts of lagrangians have contribution to the casimir effect and the longitudinal part does not lead to any casimir force .
therefore we just consider the transverse part of the lagrangian which in the fourier space is @xmath170 where @xmath171 @xmath172 and @xmath173 and @xmath174 refer to the interaction of the polarization and magnetization fields with the electromagnetic field respectively which are @xmath175 and @xmath176 where @xmath177 is the antisymmetric tensor . to obtain the casimir force between two plates with axial symmetry due to the transverse part of the lagrangian
, the filed modes can be divided into tm and te modes @xcite .
it can be shown that for our case as we consider the electromagnetic field in the presence of a homogenous , isotropic and flat magnetodielectric medium , similar to @xcite again the field can be divided into tm and te modes which refer to the dirichlet and neumann bc respectively .
as for the situation under study here for tm and te modes the casimir force is the same , therefore here we consider only the tm mode .
according to the above discussion the full lagrangian can be rewritten as @xmath178 where @xmath179 is the lagrangian density of a massless klein@xmath14gordon filed and @xmath180 the interaction term is defined by @xmath181 where @xmath182 and @xmath183 .
it can be seen that the lagrangian ( [ 79 ] ) is similar to the lagrangian ( [ 2 ] ) .
in fact these lagrangians are the same if we take @xmath184 .
this is the reason why we called @xmath20 as a polarization field and the results obtained in the sec .
[ path - scalar ] can be used for a polarizable medium . by mixing ( [ 18 ] ) and (
[ 82 ] ) , and use the same procedure in the sec .
[ path - scalar ] , after some manipulations we obtain the green s function as @xmath185 which is the green s function of em field in the presence of magnetodielectric medium with the electric and magnetic susceptibilities @xmath186 and @xmath187 respectively .
these susceptibilities satisfy the kramers -kronig relations as expected .
the other green s functions or correlation functions are @xmath188 @xmath189 @xmath190 @xmath191 the first terms in the right hand side of eqs .
( [ 88 ] ) and ( [ 89 ] ) are related to the noise operators of the system which satisfy the fluctuation - dissipation theorem , like ( [ 35 ] ) .
consequently the generating function is @xmath192&=&\exp\bigg\{\imath \int dx \int dx ' \bigg [ j_\varphi(x ) g_{\varphi,\varphi}(x - x')j_\varphi(x)\nonumber\\ & & \hspace{-2 cm } + j_\varphi(x ) g_{p,\varphi}(x - x')j_p(x)+j_\varphi(x ) g_{p,\varphi}(x - x')j_p(x)\bigg]\bigg\}. \nonumber\\\end{aligned}\ ] ] this relation can be used to obtain the casimir force in the next section .
here again we consider three different boundary conditions similar to the sec .
[ force - scalar ] varies kinds of the fields . *
i * ) imposing the boundary condition on em field : if we impose the dirichlet bc on em field the partition function becomes @xmath193 where @xmath194,\ ] ] and @xmath195 is the green s function ( [ 83 ] ) with imaginary time . to calculate @xmath196
, we invoke the fourier space where @xmath197 is diagonal and its elements have the form @xmath198 where here @xmath199 and @xmath200 .
finally the casimir force is obtained as @xmath201,\ ] ] where @xmath202 .
this equation is like the equation ( [ 61 ] ) the only difference is in the definition of @xmath203 .
if we impose the neumann bc on the em field we will achieve the same result as that of the dirichlet bc .
* ii * ) we can impose the bc on polarization and magnetization fields . as we mentioned after the equation ( [ 65 ] ) , this situation does not appear in our problem since we consider a magnetodielectric medium surrounded by two perfect conductors which lead to the bc on em field .
but for two magnetodielectric slabs this kind of bc may appear which may be dealt with elsewhere @xcite .
* iii * ) we can impose the boundary condition on both matter and em fields . in this case if we redo the calculations , we conclude that this situation leads to the same result as case * i*. this shows that for a magnetodielectric medium the conditions * i * and * iii * lead to the same result for the casimir force .
in this article we quantized the electromagnetic field in the presence of a magnetodielectric medium in the frame work of path integrals . for a medium with a given susceptibility ,
the modified casimir force is obtained for different boundary conditions .
the present approach can be generalized to the case of rough perfect conductors in the presence of a general medium straightforwardly @xcite . | in this article we investigate the casimir effect in the presence of a medium by quantizing the electromagnetic ( em ) field in the presence of a magnetodielectric medium by using the path integral formalism . for a given medium with definite electric and magnetic susceptibilities , explicit expressions for the casimir force
are obtained which are in agree with the original casimir force between two conducting parallel plates immersed in the quantum electromagnetic vacuum . |
the study of gross properties of bulk self - gravitating objects using thomas - fermi model has been discussed in a very lucid manner in the text book on statistical physics by landau and lifshitz @xcite . in this book
the model has also been extended for the bulk system with ultra - relativistic electrons as one of the constituents .
analogous to the conventional white dwarf model , these electrons in both non - relativistic and ultra - relativistic cases provide degeneracy pressure to make the bulk system stable against gravitational collapse .
the results are therefore an alternative to lane - emden equation or chandrasekhar equation @xcite .
the mathematical formalism along with the numerical estimates of various parameters , e.g. , mass , radius , etc . for white dwarf stars
may be taught as standard astrophysical problems in the master of science level classes for the students having astrophysics and cosmology as spacial paper .
the standard version of thomas - fermi model has also been taught in the m.sc level atomic physics and statistical mechanics general classes .
more than two decades ago , bhaduri et.al . @xcite
have developed a formalism for thomas - fermi model for a two dimensional atom .
the problem can also be given to the advanced level m.sc .
students in the quantum mechanics classes . the numerical evaluation of various quantities associated with the two dimensional atoms
are also found to be useful for the students to learn numerical techniques , computer programing along with the physics of the problem . the work of bhaduri et.al .
is an extension of standard thomas - fermi model for heavy atoms into two - dimensional scenario . a two - dimensional version of thomas - fermi model has also been used to study the stability and some of the gross properties of two - dimensional star cluster @xcite . in our opinion
this is the first attempt to apply thomas - fermi model to two - dimensional gravitating object .
however , to the best of our knowledge the two - dimensional generalization of thomas - fermi model to study gross properties of bulk self - gravitating objects , e.g. , white dwarfs has not been reported earlier .
this problem can also be treated as standard m.sc level class problem for the advanced level students with astrophysics and cosmology as special paper . in this article
we shall therefore develop a formalism for two dimensional version of thomas - fermi model to investigate some of the gross properties of two - dimensional hypothetical white dwarf stars .
the work is essentially an extension of the standard three - dimensional problem which is discussed in the statistical physics book by landau and lifshitz @xcite .
the motivation of this work is to study newtonian gravity in two - dimension .
analogous coulomb problem with logarithmic type potential has been investigated in an extensive manner . however , the identical problem for gravitating objects has not been thoroughly studied ( except in @xcite ) .
one can use this two - dimensional gravitational picture as a model calculation to study the stability of giant molecular cloud during star formation and also in galaxy formation .
the article is arranged in the following manner . in the next section
we have developed the basic mathematical formalism for a two - dimensional hypothetical white dwarf star . in section - iii , we have investigated the gross properties of white dwarf stars in two - dimension . in section - iv ,
the stability of two - dimensional white dwarfs with ultra - relativistic electrons as one of the constituents have been studied . finally in the last section we have given the conclusion of this work .
in this section we start with the conventional form of poisson s equation , given by @xcite ( see also @xcite ) @xmath0 where @xmath1 is the gravitational potential , @xmath2 is the surface density of matter and for the two - dimensional scenario @xmath3 has to be expressed in its two - dimensional form .
let us assume that the bulk object in two dimensional geometry is an hypothetical white dwarf star for which the inward gravitational pressure is balanced by the outward degeneracy pressure of the two dimensional electron gas .
the mass of the object is coming from the heavy nuclei distributed in two - dimensional bounded geometry . now the fermi energy for a two dimensional electron gas is given by @xmath4 where @xmath5 and @xmath6 are respectively the fermi momentum and the mass of the electrons .
further , the number of electrons per unit surface area is given by @xmath7 hence the electron fermi energy is @xmath8 let @xmath9 is the baryon mass per electron @xcite , then it can very easily be shown that @xmath10 where @xmath2 is the mass density ( mass per unit area ) of the matter . then following @xcite ,
the thomas - fermi condition is given by @xmath11 since @xmath1 is a function of radial coordinate @xmath12 , the chemical potential @xmath13 and the matter density @xmath2 also depend on the radial coordinate @xmath12 . then replacing the gravitational potential @xmath1 with the electron fermi energy @xmath13 from eqn.(6 ) , the poisson s equation ( eqn.(1 ) )
can be written as @xmath14 assuming circular symmetry i.e. , independent of @xmath15 coordinate , replacing @xmath16 by @xmath17 from eqn.(5 ) and expressing @xmath3 in two - dimensional polar form , we have from eqn.(7 ) @xmath18 substituting @xmath19 , where @xmath20 may be called the scaled radial coordinate with the constant @xmath21 we can write down the poisson s equation in the following form @xmath22 it is quite obvious that in this case the solution is given by @xcite @xmath23 where @xmath24 is the ordinary bessel function of order zero .
the surface of the two - dimensional bulk object is obtained from the first zero @xmath25(say ) of the bessel function @xmath24 , i.e. , @xmath26 hence the radius of the object is given by @xmath27 .
further , at the centre , i.e. , for @xmath28 , @xmath29 therefore the solution can also be written in the form @xmath30 hence the central density is given by @xmath31 therefore we can also write @xmath32 gives the variation of matter density with @xmath33
with circular symmetry , the poisson s equation in polar coordinate can also be written in the following form @xmath34 let us now integrate this differential equation with respect to r from @xmath35 , which is the centre to @xmath36 , the surface of the bulk two dimensional stellar object
. then we have @xmath37 where @xmath38 the mass of the object . now expressing @xmath1 in terms of @xmath13 and using the solution for @xmath13 ( eqn.(14 ) )
, we finally get @xmath39 where we have used the standard relation @xmath40 @xcite .
therefore this transcendental equation ( eqn.(20 ) ) can be solved numerically to get mass - radius relation for the two - dimensional stellar objects in thomas - fermi model in the non - relativistic scenario . in fig.(1 ) , we have shown graphically the form of mass radius relation for such objects .
for the sake of illustration we have chosen the central density @xmath41 in such a manner that the maximum mass of the object is @xmath42 .
the average density can also be obtained from the definition @xmath43 which can be expressed in terms of central density , and is given by @xmath44
in this section following @xcite we have considered a two - dimensional hypothetical white dwarf star composed of massive ions and degenerate ultra - relativistic electron gas in two - dimensional geometrical configuration .
the stability of the object is governed by electron degeneracy pressure . in the ultra - relativistic scenario
the energy of an electron is given by the usual expression @xmath45 where @xmath46 is the velocity of light and @xmath47 is the electron momentum .
the fermi energy for this degenerate electron gas is then given by @xmath48 with @xmath49 , the electron fermi momentum . hence the number density ( the surface value ) for ultra - relativistic electron gas can be written in the following form @xmath50 therefore the electron fermi energy @xmath51 where @xmath52 and @xmath2 is the mass density for such ultra - relativistic matter . then following eqn.(7 ) , we have the thomas - fermi equation in two - dimension in the ultra - relativistic electron gas scenario @xmath53 where @xmath20 is the scaled radial coordinate and the constant @xmath54 unfortunately the non - linear differential equation given by eqn.(27 ) can not be solved analytically . to obtain its numerical solution
we have used the standard four point runge - kutta numerical technique and a code is written in fortran 77 to solve eqn.(27 ) using the initial conditions ( i ) @xmath55 , at @xmath28 ( which indicates the origin of the two - dimensional white dwarf ) , which is the maximum value of @xmath13 , then as a consequence ( ii ) @xmath56 , the other initial condition . the surface of this bulk two - dimensional object
is then obtained from the boundary condition @xmath57 , with @xmath25 the scaled radius parameter . from the numerical solution of eqn.(27 ) with the initial conditions ( i ) and ( ii ) , the actual value of the radius can be obtained . as a cross check ,
we have used mathematica in linux platform and got almost same result . in our formalism , instead of solving for the gravitational potential in two - dimension , we have solved numerically for the electron chemical potential with the initial conditions ( i ) and ( ii ) . the initial condition ( i ) depends on the central density @xmath41 , which we have supplied by hand as input . in the non - relativistic case by trial and error method
we have fixed the value @xmath58 g @xmath59 to get @xmath60 .
whereas in the ultra - relativistic electron gas scenario , we have started from @xmath61 g @xmath59 and gone upto @xmath62 g @xmath63 .
since we are solving numerically for @xmath13 and the central density is supplied by hand , the singularity at the origin ( @xmath28 ) does not appear in our analysis .
singularity at the origin also do not appear when lane - emden equation ( newtonian picture ) and tov equation ( general relativistic scenario ) are solved numerically by supplying the central density by hand . in this case
the mass - radius relation can also be derived from eqn.(17 ) with @xmath1 replaced by @xmath13 from eqn.(6 ) , the thoma - fermi condition , and is given by @xmath64 unlike the non - relativistic scenario , in this case the derivative term at the surface has to be obtained from the numerical solution of thomas - fermi equation ( eqn.(27 ) ) . in fig.(2 ) we have shown the graphical form of mass radius relation for the ultra - relativistic case .
the qualitative nature of fig.(1 ) and fig.(2 ) are completely different . unlike the non - relativistic situation , where the mass radius relation is obtained only for a particular central density , in this case each point on the m - r curve corresponds a particular central density .
this is obvious from the initial condition ( i ) , where @xmath41 is the given central density . since the central density has a fixed value in the non - relativistic scenario , the mass of the object increases with the increase of its radius . to make these two figures more understandable , in fig.(3 ) and
fig.(4 ) we have plotted the variations of mass and radius respectively for the object against the central density in the ultra - relativistic scenario . in fig.(3 ) we have plotted the mass of the object expressed in terms of solar mass , with the central density of the object expressed in terms of normal nuclear density . in fig.(4 ) we have shown the variation of radius of the object with central density . from the study of density profiles of the object in the ultra - relativistic scenario ,
we have noticed that for the low value of the central density , the numerical solution of eqn.(27 ) shows that the matter density vanishes or becomes negligibly small for quite large radius value , whereas for very high central density , the matter density vanishes very quickly .
therefore in this model the stellar objects with very low central density are large in size and since the matter density is low enough , the mass is also quite small . on the other hand for the large values of central density , the objects are small in size , i.e. , they are quite compact in size but massive enough .
this study explains the nature of variations of mass and radius with central density for the ultra - relativistic case .
further in ultra - relativistic scenario the average density of matter inside the bulk two - dimensional stellar object is given by @xmath65 which exactly like the non - relativistic case also depends on the central density and the radius of the object .
however , unlike the non - relativistic situation , here the average density depends on the surface value of the gradient of electron fermi energy , which has to be obtained numerically from the solution of eqn.(27 ) .
in conclusion we would like to comment that although the formalism developed here is for a hypothetical stellar object , which is basically an extension of standard three dimensional problem discussed in the book by landau & lifshitz , we strongly believe that it may be considered as interesting post - graduate level problem for the physics students , including its numerical part .
this problem can also be treated as a part of dissertation project for post graduate physics students .
the study of thomas - fermi model in two - dimension for self - gravitating objects may be used for model calculations of star formation and galaxy formation from giant gaseous cloud .
finally , we believe that the problem solved here has some academic interest . l.d .
landau and e.m lifshitz , statistical physics , part - i , butterworth / heinemann , oxford , ( 1998 ) pp .
shapiro and s.a .
teukolsky , black holes , white dwarfs and neutron stars , john wiley and sons , new york , ( 1983 ) pp .
61 and 188 .
h.q . huang and k , n , yu , stellar astrophysics , springer , ( 1998 ) , pp . 337 .
r.k bhaduri , s das gupta and s.j lee , am .
j. phys . ,
* 58 * ( 1990 ) 983 .
sanudo , a.f .
pacheco , revista mexicana de fisica , * e53 * , ( 2006 ) 82 .
sanu engineer , k. srinivasan and t. padmanabhan , astrophys .
* 512 * , ( 1999 ) 1 .
m. abramowitz and i.a stegun ( ed . ) , handbook of mathematical functions , dovar publication , ny , ( 1972 ) pp . | in this article we have solved an hypothetical problem related to the stability and gross properties of two dimensional self - gravitating stellar objects using thomas - fermi model .
the formalism presented here is an extension of the standard three - dimensional problem discussed in the book on statistical physics , part - i by landau and lifshitz .
further , the formalism presented in this article may be considered as class problem for post - graduate level students of physics or may be assigned as a part of their dissertation project . |
with the maturing of the quasi - phase - matching ( qpm ) technique , in particular by electric - field poling of ferro - electric materials , such as linbo@xmath0 @xcite , by poling of polymers @xcite and quantum - well disordering in semiconductors @xcite , the number of applications of quadratic nonlinear ( or @xmath1 ) materials has increased significantly .
even complicated qpm grating structures are now commercially available in periodically poled linbo@xmath0 ( ppln ) .
it is therefore more important than ever to have complete knowledge of the effects a qpm grating has on the properties of @xmath1 materials .
the most fundamental effect of a qpm grating , with a certain spectrum of spatial wave vectors , is to allow noncritical phase matching at all wavelengths for which the wave - vector mismatch @xmath2 matches a component of the grating spectrum .
thus qpm gratings allow for efficient multiple wavelength second harmonic generation ( shg ) @xcite , which may , for example , be used for multiple - channel wavelength conversion @xcite .
in addition to providing phase matching , qpm gratings have the basic effect that they induce an asymmetric cubic nonlinearity ( acn ) in the equations for the average field in the form of self- and cross - phase modulation terms @xcite .
this acn appears in linear and/or nonlinear periodic qpm gratings of arbitrary shape @xcite , it can be focusing or defocusing , depending on the sign of the phase mismatch @xcite , and its strength can be increased ( e.g. , dominating the kerr nonlinearity ) by modulation of the grating @xcite . in continuous - wave operation
the acn induces an intensity - dependent phase mismatch , just as the inherent kerr nonlinearity , with potential use in switching applications @xcite . the acn further explains a wide spectrum of novel fundamental properties of solitons @xcite and modulational instability ( mi ) @xcite .
importantly the acn is a general effect of non - phase - matched wave interaction and as such appear also in homogeneous @xmath1 materials in the cascading limit .
in fact , in this case the asymmetric signature of the acn may be measured as the difference between the properties in upconversion and downconversion , since there is no effective @xmath1 nonlinearity competing with the acn as in qpm gratings .
such an experiment was recently reported @xcite and thus the acn has now been confirmed both numerically and experimentally . in this paper , we present the results of a complete study into the modulational instability of beams in 1 + 1d qpm waveguides .
we find that in general , the mi gain spectrum has a multipeaked structure of up to three fundamental bands with accompanying overtones .
our previous work@xcite concentrated on the structure of the fundamental ( long - wave instability ) bands at low transverse wavenumbers .
these bands are due to mi of the averaged field , and are predicted by averaged equations , provided the effective acns are taken into account for the most accurate description .
for example , for qpm with a simultaneous linear and nonlinear grating and/or with a nonlinear grating with a dc - value ( as with qpm in polymers @xcite and semiconductors @xcite ) , the acns can suppress these fundamental bands , making plane waves modulationally stable over hundreds of diffraction lengths @xcite .
exact floquet calculations and numerical simulations confirmed the predictions for the low - frequency bands , but also revealed high - frequency bands not predicted by the averaged theory .
these gain bands were surmised to be related to the inherent instability in homogeneous ( non - phase - matched ) @xmath1 media @xcite .
however , an accurate _ analytic _ description of the first overtone bands was not derived and the higher overtone bands were not discussed at all .
here we concentrate on ppln slab waveguides , which are @xmath1 media with a purely nonlinear qpm grating with no dc - component .
although more - general kinds of grating can be analysed with our method , we choose to focus on the simpler case for clarity .
for example , with this simple qpm grating we do not need to take into account the acn for an accurate description of mi @xcite , however our analytic method does allow for such cubic nonlinearities to be included .
we present the first analytical and numerical description of the complete mi spectrum with all overtones .
we show explicitly that the overtone series are caused by mi in the rapidly varying components of the propagating fields , which are in turn induced by the grating .
we derive approximate analytic expressions for the positions of the gain bands and compare them with exact floquet results and direct numerical simulation , to find good agreement .
we consider a linearly polarized electric field propagating in a lossless @xmath1 slab waveguide with a qpm grating under conditions for ( the most efficient ) first - order qpm and type i shg .
the normalized dynamical equations for the slowly varying envelopes of the fundamental @xmath3 and second harmonic ( sh ) @xmath4 take the form @xmath5 where the asterisk means complex conjugate .
the @xmath6 and @xmath7 scales have been normalised by @xmath8 and @xmath9=@xmath10 , respectively , and @xmath11=@xmath12 is the normalized wave - vector mismatch .
the nonlinearity is periodic in @xmath7 with expansion @xmath13 where @xmath14=@xmath15 ( @xmath16 is real ) and the grating wave number @xmath17 is defined to be positive .
the grating will force the same periodicity in the propagating fields .
we therefore expand these also : @xmath18 where @xmath19=@xmath20 is the residual mismatch , with @xmath21=@xmath22 for first - order qpm . substituting all expansions into eqs .
( [ field_eqns1]-[field_eqns2 ] ) gives @xmath23 where @xmath24 .
this set of equations has plane - wave solutions of the form @xmath25 , @xmath26 @xcite . to study mi
we consider small perturbations of the plane - wave solutions : @xmath27\exp{(i\lambda z ) } , \\ & & \nonumber\\ v_n(x , z)&=&\left[\bar v_n + \mu_n(x , z)\right]\exp{(i2\lambda z)}. \end{aligned}\ ] ] substitution into eqs .
( [ coupled1]-[coupled2 ] ) gives the linearized equations @xmath28 where @xmath29 .
writing the perturbations in the form @xmath30 one obtains a linear matrix equation for the perturbation @xmath31=@xmath32 , which couples all fourier components . to derive a simple result we consider nearly phase - matched interaction with @xmath33@xmath34@xmath35@xmath361 in the typical ( for ppln ) square qpm grating , for which @xmath14=0 for @xmath37 even and @xmath14=@xmath38 for @xmath37 odd .
the requirement @xmath39 allows us to assume that the amplitudes @xmath40 and @xmath41 vary slowly compared to @xmath42 and that the higher harmonics are much smaller than the dc - fields , @xmath43 , @xmath44 , @xmath45 . in this case
the structure of the matrix equation means that the evolution of the perturbation in each component is to a good approximation decoupled from the other components , @xmath46\vec{\delta}_n \equiv m_n \vec{\delta}_n , \label{mi_matrix}\end{aligned}\ ] ] where @xmath47=@xmath48 , @xmath49=@xmath50 , @xmath51=@xmath52 , and @xmath53=@xmath54 .
the eigenvalues of @xmath55 are @xmath56 where @xmath57 any positive real part of an eigenvalue corresponds to mi with the gain @xmath58 .
for the gratings that we consider here , @xmath59 and @xmath60 .
analysing the gain versus @xmath47 and @xmath53 reveals three gain bands , with extrema remaining close to @xmath61 ( diagonal branch db ) and @xmath62 ( hyperbolic branches hb@xmath63 , hb@xmath64 ) , as illustrated in fig .
[ branches ] for @xmath65 and @xmath66 .
the extrema @xmath61 of db correspond to @xmath67 while the extrema @xmath62 of hb@xmath63 and hb@xmath64 correspond to @xmath68 for @xmath69 the hb@xmath63 bands appear at @xmath70 and the hb@xmath64 bands at @xmath71 .
thus we have the structure of up to 3 gain bands in the average field ( @xmath37=0 ) , each with a set of equally spaced ( in @xmath72 ) weak overtone gain bands ( @xmath73 ) .
however , the gain is large only near @xmath47=0 ( see fig . [ branches ] ) . for @xmath74
, @xmath75 can be satisfied without violating the assumptions only on hb@xmath64 , and thus the overtones of the other branches will be small .
consider a system at exact phase - matching @xmath76 , with dimensionless parameters @xmath77 and @xmath78 .
for a linbo@xmath0 qpm grating with period @xmath79 designed to phase - match at wavelengths of around @xmath80 nm , these dimensionless parameters correspond to having an input intensity of @xmath81gw@xmath82m@xmath83 , if one assumes a waveguide depth of @xmath84 m .
the distance scaling parameters for this case are @xmath85 m and @xmath86 mm . for fixed distance
scaling , different values of the magnitude of @xmath87 correspond to different input intensities , and different signs of @xmath87 correspond to different ratios of the fundamental to second harmonic@xcite . on the other hand , keeping @xmath88 fixed while varying the input intensity corresponds to changing the distance scaling : to a good approximation , @xmath89 , @xmath90 , and hence @xmath91 . for @xmath87=1
our analytic results predict only one fundamental @xmath37=0 gain band ( hb@xmath63 ) and three gain bands in all overtones .
however the overtones of db and hb@xmath63 are too small to be seen .
thus the visible gain bands have maxima at @xmath92=1.18 ( hb@xmath63 ) , @xmath93=36.64 , @xmath94=51.84 , etc .
( all hb@xmath64 ) .
in physical units these are @xmath95m@xmath96 , @xmath97m@xmath96 , and @xmath98m@xmath96 , respectively .
the analytical results agree with exact floquet calculations @xcite , as seen in fig .
[ gtheory ] . for @xmath87=@xmath99 two fundamental @xmath37=0 bands are predicted ( db and hb@xmath63 ) , with one set of visible overtones from hb@xmath64 .
the maxima are at @xmath92=2 ( db ) , @xmath92=3.38 ( hb@xmath63 ) , @xmath93=36.70 , @xmath94=51.88 , etc .
( all hb@xmath64 ) .
in physical units these are @xmath100m@xmath96 , @xmath101m@xmath96 , @xmath102m@xmath96 , and @xmath103m@xmath96 , respectively .
this is again confirmed by the exact floquet calculations seen in fig .
[ gtheory ] .
figure [ gtheory ] also shows the gain profiles predicted by eq .
( [ evalues ] ) , which agree well with the exact floquet results .
direct numerical simulation @xcite of eqs .
( [ field_eqns1]-[field_eqns2 ] ) with mi seeded by noise also confirmed our results ( see fig .
[ gnumerical ] ) .
we see from eqs .
( [ nudia]-[nuhyp ] ) that the dimensionless spectral positions of the higher - order bands depends on @xmath17 , whereas the fundamental bands do not .
the different dependence implies a different scaling with input intensity , since for fixed @xmath88 , both the distance scalings @xmath8 , @xmath9 and the dimensionless wave vector @xmath17 depend on the intensity .
the result is that the spatial frequencies of the higher - harmonic instabilities remain the same , in physical units , whereas the frequencies of the fundamental - band instabilities generally increase with the quartic root of the intensity .
in conclusion , we have presented a simple theory able to accurately predict the complete mi spectrum in general qpm gratings in @xmath1 materials . in particular ,
we have predicted and verified that overtone gain bands originate from mi in the higher - order fourier components of the field .
this research is supported by the danish technical research council ( grant no .
26 - 00 - 0355 ) and the australian research council .
v. ricci , g.i .
stegeman , and k.p .
chan , poling of multilayer polymer films for modal dispersion phase matching of second - harmonic generation : effects in glass - transition temperature matching in different layers " , j. opt .
b. * 17 * , 13491353 ( 2000 ) .
a. saber helmy , d.c .
hutchings , t.c .
kleckner , j.h .
marsh , a.c .
bryce , j.m .
arnold , c.r .
stanley , j.s .
aitchison , c.t.a .
brown , k. moutzouris , and m. ebrahimzadeh , quasi - phase - matching in gaas - alas superlattice waveguides via bandgap tuning using quantum well intermixing " , , 13701372 ( 2000 ) .
p. baldi , c.g .
trevino - palacios , g.i .
stegeman , m.p . de micheli , d.b .
ostrowsky , d. delacourt , and m. papuchon , simultaneous generation of red , green and blue light in room temperature periodically poled lithium niobate waveguides using single source " , electron . lett . * 31 * , 13501351 ( 1995 ) .
p. di trapani , a. bramati , s. minardi , w. chinaglia , c. conti , s. trillo , j. kilius , and g. valiulis , focusing versus defocusing nonlinearities due to parametric wave mixing " , 183902 - 1183902 - 4 ( 2001 ) . | we consider plane waves propagating in quadratic nonlinear slab waveguides with nonlinear quasi - phase - matching gratings .
we predict analytically and verify numerically the complete gain spectrum for transverse modulational instability , including hitherto undescribed higher order gain bands . |
let @xmath0 be an infinite sequence of iid random variables with mean @xmath1 and variance @xmath2 .
two key results in probability theory are the central limit theorem and the law of the iterated logarithm .
the _ central limit theorem _ ( clt ) states that for @xmath6 , one has @xmath7 where @xmath8 and @xmath9 denotes the standard gaussian distribution . the _ law of the iterated logarithm _
( lil ) , proved by khinchin @xcite and kolmogorov @xcite , asserts that @xmath10 = 1 .\ ] ] the @xmath11 term reveals a subtle correlation between the @xmath12 s , especially those with indices close to each other .
the theory of random graphs ( hypergraphs ) contains several central limit theorems , some of which are among the most well known results in the field .
it is natural to wonder if the lil also holds .
the goal of this paper is to initiate this investigation and provide the first few rigorous results . to our surprise , this natural problem has not been studied before and we hope this paper will motivate further activity .
let @xmath13 be a fixed constant in @xmath14 .
we consider the infinite random hypergraph @xmath15 on the vertex set @xmath16 where we add every @xmath5-subset @xmath17 as an edge with probability @xmath13 independently .
this gives rise to a nested sequence of random hypergraphs where @xmath18 is defined by restriction to the first @xmath19 vertices @xmath20:=\{1 , \dots , n \}$ ] .
the atom iid variables are @xmath21 which represent the edges ( @xmath22 if @xmath23 forms an edge and @xmath1 otherwise ) . in the case of graphs ( that is @xmath24 )
we denote @xmath25 by @xmath26 and @xmath18 by @xmath27 . in this way
, we obtain the usual binomial random graph model .
we also consider the infinite random bipartite graph @xmath28 on vertex set @xmath29 , where @xmath30 and @xmath31 are two disjoint copies of @xmath16 , and every pair @xmath32 forms an edge with probability @xmath13 , independently .
let @xmath33 be obtained from @xmath28 by restricting @xmath30 and @xmath31 to their first @xmath19 elements .
many clt s in the theory of random graphs involve some sort of counting functions .
for instance , counting the number of copies of a fixed graph ( such as triangles or @xmath34 s ) is a classical problem ; see @xcite and the references therein ( the interested reader can also find a detailed discussion in @xcite , chapter 6 ) . in this case , the question of when the clt holds is well understood .
fix a nonempty graph @xmath35 , and let @xmath36 count the number of copies of @xmath35 in @xmath27 .
let @xmath37 .
if @xmath38 is such that @xmath39 and @xmath40 , then @xmath41)/\sqrt{var(x_n)}$ ] tends in distribution to @xmath9 .
it is more challenging to count global objects . in @xcite janson
considered the numbers of spanning trees , perfect matchings and hamilton cycles in random graphs .
he showed these counting functions are log - normal for @xmath27 in certain ranges of density .
results of a similar flavor ( and shorter proofs ) were also obtained later by gao @xcite .
let @xmath36 be the random variable that counts number of spanning trees , perfect matchings , or hamilton cycles in @xmath27 .
fix a constant @xmath42 .
let @xmath43 .
if @xmath44 , then @xmath45)+\frac{1-p(n)}{c p(n)}\right)\rightarrow n\left(0,\frac{2(1-p)}{c}\right)\]]where @xmath46 in the case of spanning trees and hamilton cycles , and @xmath47 in the case of perfect matchings . throughout this paper , we use @xmath36 to denote a parameter of the random model under consideration ( that is , @xmath18 or @xmath33 ) , with mean @xmath48 and variance @xmath49 , which may vary in each occasion .
first , we consider the case @xmath36 is the number of copies of a fixed graph @xmath4 in @xmath27 and prove [ main1 ] for a fixed graph @xmath4 , let @xmath36 denote the number of copies of @xmath4 in @xmath27 .
the sequence @xmath36 satisfies the lil , namely @xmath50=1.\ ] ] the key ingredient in the proof of theorem [ main1 ] is to overcome the fact that the terms in @xmath36 are not completely independent .
second , we consider the case where @xmath36 is the number of perfect matchings in @xmath33 . in this case , we obtain a lil for the random variable @xmath51 .
[ thm : main]let @xmath36 be the number of perfect matchings in @xmath33 and set @xmath52 . then the sequence @xmath53 satisfies the lil , namely @xmath54=1\end{aligned}\ ] ] third , we consider the number of hamilton cycles in @xmath27 and prove [ thm : main2 ] let @xmath36 be the number of hamiltonian cycles in @xmath27 and set @xmath52 .
the sequence @xmath53 satisfies the lil , namely @xmath55=1.\ ] ] the proofs of the last two theorems are more involved .
our new key ingredient is a large deviation bound on @xmath36 ( the number of perfect matchings or hamiltonian cycles , respectively ) , which appears to be new and could be of independent interest .
note that we did not write @xmath56)/\sqrt{var(y_n)}$ ] in theorems [ thm : main ] and [ thm : main2 ] , as the expected value and variance of @xmath53 are unknown .
we conjecture that constants used in the theorem are good approximations of these quantities .
next , we consider the case of @xmath5-uniform random hypergraphs . in this
setting , the clt and the lil for the number of copies of a fixed subhypergraph can be obtained in a similar way to the graph case . therefore , we focus on global structures , hamiltonian cycles in particular . to start ,
there are many ways to define a cycle in a hypergraph .
we work with the following : an @xmath57-overlapping hamilton cycle is a cyclic ordering of the vertices @xmath58 for which the edges consisting of @xmath5 consecutive vertices and two consecutive edges overlap in exactly @xmath57 vertices .
the case @xmath59 is known as a
loose hamilton cycle " and the case @xmath60 is known as
tight hamilton cycle " ( note that the case @xmath61 corresponds to a perfect matchings ) .
our next result works for all @xmath57 , but for the sake of presentation we state it for loose hamilton cycles ( which from now on will be referred to as hamilton cycles ) .
let @xmath62 denote the number of hamilton cycles in @xmath18 with means @xmath63 and variance @xmath64 .
we have found out , somewhat surprisingly , that for @xmath65 , @xmath66 themselves satisfy the clt , as opposed to the case @xmath24 where @xmath67 satisfies the clt .
the reason lies in the fact that unlike the case @xmath24 , for @xmath68 , if we choose a few hamilton cycles at random , it is very unlikely for them to have common edges and therefore the variance of the counting function is much smaller compared to @xmath69 .
a similar observation has been used by dudek and frieze in @xcite and @xcite where they determined the threshold behavior of @xmath57 hamilton cycles . [
main : clt ] for any @xmath70 , the sequence @xmath71 satisfies the clt , namely @xmath72 finally , we show that for @xmath73 , the sequence @xmath74 satisfies a lil .
[ main : ill ] for @xmath73 , the sequence @xmath74 satisfies the lil , namely @xmath75=1.\ ] ] we conclude this section with a few remarks .
first , there are many other clts in the random graphs / hypergraphs literature , and it is natural to raise the validity of the lil in each situation . we hope that this paper will motivate further research in this direction . as far as the new results are concerned , we prove them under the condition that @xmath13 is a fixed constant in @xmath14 . since we work with a random infinite graph , letting @xmath13
depend on @xmath19 ( as one usually does for @xmath27 ) does not make sense
. however , one can still consider the sparse case by modifying the definition .
for instance , one can say that the edge @xmath76 appears with probability @xmath77 , independently , where @xmath78 is a sequence of positive numbers tending to @xmath1 with @xmath5 .
it is an interesting question to determine those ranges of densities for which lil holds .
for a technical reason , the proof of theorem [ main : ill ] requires @xmath73 .
we leave the case @xmath79 as an open problem .
* notation .
* throughout the paper , we assume that @xmath19 is sufficiently large , whenever needed .
all asymptotic notation is used under the assumption that @xmath80 .
we denote by @xmath81 the complete graph on the vertex set @xmath20=\{1,\ldots , n\}$ ] and set @xmath82 . for a graph @xmath4
we define @xmath83 to be the set of all ( labeled ) copies of @xmath4 in the infinite complete graph on vertex set @xmath84 . for each @xmath85 , we define @xmath86 to be the subset of @xmath83 , consisting of all copies of @xmath4 in @xmath81 .
given a copy @xmath87 , we denote by @xmath88 and @xmath89 its vertex set and edge set , respectively . in the special case where @xmath4 is a _ triangle _
( that is , a graph on @xmath90 vertices @xmath91 where all the three possible edges @xmath92 appear ) , we replace @xmath83 with @xmath93 in all of the previous notation .
moreover , we assume that an enumeration @xmath94 is fixed so that for every @xmath85 we have @xmath95 , where @xmath57 is the number of labeled copies of @xmath4 in @xmath81 .
note that such an enumeration can be easily obtained by an induction on @xmath19 .
to each copy @xmath87 , we associate an indicator random variable @xmath96 . for a collection @xmath97 of copies of @xmath98
we have @xmath99 . finally ,
since we make extensive use of the standard gaussian @xmath9 , we recall its distribution function @xmath100 = \frac{1}{\sqrt{2\pi}}\int_{-\infty}^xe^{-t^2/2}dt.\ ] ] * organization of the paper . *
the rest of the paper is organized as follows . in section [ sec : tools ] we collect the tools which are used for the proof of our main results . in sections
[ sec : upper ] and [ sec : lower ] we prove the upper and lower bounds for theorem [ main1 ] , and in section [ sec : calculations ] we explain some inequalities we use during the proof . in section [ perfect matchings ]
we prove theorem [ thm : main ] , and in section [ sec : main2 ] we prove theorem [ thm : main2 ] .
both of these sections are split into two subsections containing the proof of the upper bound and the lower bound , respectively .
section [ sec : clt hypergraphs ] contains the proof of [ main : clt ] , and section [ sec : ill hypergraph ] contains the proof of [ main : ill ] .
section [ upper tail estimates ] contains the new large deviation estimates we need on perfect matchings and hamilton cycles .
the appendix contains some rather routine , but tedious , calculations and approximations that we use throughout the paper .
in this section we introduce the main tools to be used in the proofs of our results . as a first tool , we present janson s inequality ( see e.g.@xcite , theorem 2.14 ) , which will be used in order to get lower tail estimates for the number of copies of a fixed graph @xmath4 in certain random graphs . for the convenience of the reader , we state it tailored for our use later ( with respect to the @xmath101 s which were previously defined ) . before doing so , we need some notation .
let @xmath102 be two positive integers , and let @xmath103 .
consider the random variable @xmath104 , let @xmath105 be its expectation , and let @xmath106 $ ] . with this notation in hand we may state the theorem .
[ janson ] for a fixed graph @xmath4 and for every @xmath107 we have @xmath108\leq e^{-\frac{t^2}{2\delta}}.\ ] ] [ remjanson ] for the special case where @xmath4 is a triangle , it is easy to show ( by fixing the intersection edge ) that @xmath109 .
we make use of this later .
another tool to be used in our proofs is the following well known lemma due to borel and cantelli .
let @xmath110 be a sequence of events
. then 1 . if @xmath111<\infty$ ] , then @xmath112=0.\ ] ] 2 .
if @xmath113=\infty$ ] and in addition all the @xmath114 s are independent , then @xmath112=1.\ ] ] the following is a theorem due to rinott @xcite which shows that the sum of dependent random variables satisfies clt , and measures the error term based on the dependencies between the variables . before stating the theorem we need the following definition .
[ def : dependencygraph ] let @xmath115 be a collection of random variables .
a graph @xmath116 on a vertex set @xmath117 is called a _ dependency graph _ for the collection if for any pair of disjoint subsets @xmath118 for which there are no edges of @xmath116 between @xmath119 and @xmath120 , the random variables @xmath121 and @xmath122 are independent . with this definition in hand , we can now state the result from @xcite .
[ thm : rinott ] let @xmath123 be a collection of random variables .
let @xmath124 and @xmath125 and @xmath126 .
let @xmath116 be a dependency graph for the collection and suppose that @xmath127 a.a.s .
for every @xmath128 and that @xmath129 .
then @xmath130-\phi(x)\right|\leq \frac { bc}\sigma \left(\sqrt{\frac 1{2\pi}}+16 \left(\frac{n}{\sigma^2}\right)^{1/2}c^{1/2}b+10\left(\frac{n}{\sigma^2}\right ) cb^2\right).\ ] ] [ rem1 ] note that whenever @xmath131 the expression in the right hand side of the inequality in theorem [ thm : rinott ] is @xmath132 . assuming this , since @xmath133 , for large enough @xmath134 it follows by theorem [ thm : rinott ] that @xmath135-\frac { 1}{x\sqrt{2\pi } } e^{-x^2/2}\right|= o\left(\frac{bc}{\sigma}\right).\ ] ] the key tools in the proofs of theorems [ thm : main ] and [ thm : main2 ] are the following concentration bounds , which may be of independent interest .
we postpone their proofs to section [ upper tail estimates ] . [ conc0]let @xmath136 be the number of perfect matchings in @xmath33 .
let @xmath137 be a constant .
there is a constant @xmath138 , depending on @xmath139 , such that for any @xmath140 , and @xmath141 , we have @xmath142\leq c^k(\mathbb{e}[x_{n , m}])^k\ ] ] markov s bound implies that for @xmath143 one has : @xmath144\geq k \mathbb{e}[x_{n , m}]]\leq ( c / k)^k\]]by taking @xmath145 , @xmath146 , and @xmath147 , we have the following corollary [ concentration0]let @xmath148 be a constant .
there is a constant @xmath149 ( depending on @xmath13 ) such that for any @xmath150 one has @xmath151\leq n^{-4}\ ] ] the concentration bounds for hamilton cycles are as follows [ conc1 ] let @xmath136 be the number of hamilton cycles in @xmath152 .
let @xmath137 be a constant .
there is a constant @xmath138 , depending on @xmath139 , such that for any @xmath140 , and @xmath153 we have : @xmath142\leq c^k(\mathbb{e}[x_{n , m}])^k\ ] ] again , markov s bound implies that for @xmath143 one has : @xmath144\geq k \mathbb{e}[x_{n , m}]]\leq ( c / k)^k\]]by taking @xmath145 , @xmath154 , and @xmath155 , we have the following corollary [ concentration]let @xmath148 be a constant .
there is a constant @xmath149 ( depending on @xmath13 ) such that for any @xmath156 one has @xmath151\leq n^{-4}\ ] ] the last lemma is an approximation to the lower factorial that we will use throughout .
[ lower factorial]let @xmath157 be integers such that @xmath158 .
then , @xmath159 in the proof of the upper - tail estimate for perfect matchings , we will need bregman s theorem , which allows us to bound the number of perfect matchings given the degree sequence : [ bregman]let @xmath35 be a bipartite graph with two color classes @xmath160 and @xmath161 .
denote by @xmath162 the number of perfect matchings and @xmath163 the degree of @xmath164
. then @xmath165
let @xmath4 be a graph on @xmath57 vertices , where @xmath57 is a fixed constant .
for the sake of simplicity of notation , throughout the whole proof we omit the up - script @xmath35 from the random variables . in order to prove theorem [ main1 ] we aim to show that for every @xmath166 we have both the upper bound @xmath167=0,\ ] ] and the lower bound @xmath168=1.\ ] ] since throughout the proof we make use of theorem [ thm : rinott ] for estimating the upper tails of random variables of the form @xmath169 , it will be convenient to introduce some notation .
for every @xmath170 let @xmath171 , where @xmath172 .
let us define a dependency graph for @xmath173 in the following manner .
the vertex set of @xmath174 is @xmath173 , and the edge set consists of all pairs @xmath175 for which @xmath176 ( that is , pairs of copies of @xmath4 which share at least one edge ) .
note that it trivially follows from the enumeration of @xmath83 that @xmath177 is the number of copies of @xmath4 with at least one vertex taken from @xmath178 .
in addition , it is easy to see that @xmath179 where @xmath180 is the maximum number of automorphisms of @xmath4 preserving some edge .
now , let us denote by @xmath181 and let @xmath182 and @xmath183 be its expectation and variance , respectively . trivially , we have @xmath184 and @xmath185 for every @xmath186
therefore , while applying theorem [ thm : rinott ] for a large @xmath134 with @xmath187 and @xmath188 , using remark [ rem1 ] we obtain @xmath189=\frac 1{x\sqrt{2\pi } } e^{-x^2/2}+o\left(\frac{n^{\ell-2}}{\sigma_{n , m}}\right ) \textrm { ( * ) } .\ ] ] note that whenever we use @xmath190 , one should verify that the error term is negligible compared to the first summand on the right hand side .
most of the times it will be quite easy to check and therefore we omit the calculations .
for some relevant estimates on the variances that we use in the proof , the reader should consult section [ sec : calculations ] .
let us start with proving the upper bound .
let @xmath166 be some positive constant and let @xmath191 .
by @xmath190 we have @xmath192&=\frac 1{x\sqrt{2\pi } } e^{-x^2/2}+o\left(\frac{1}{n}\right)\\ & = o\left((\log n)^{-(1+\varepsilon/4)^2}\right).\end{aligned}\ ] ] using this estimate for every @xmath19 of the form @xmath193 ( where @xmath194 ) , we obtain that @xmath195=\sum_{k=1}^\infty o\left(k^{-(1+\varepsilon/4)^2}\right)<\infty , \ ] ] and therefore , it follows from the borel - cantelli lemma that for some @xmath196 we have @xmath197=1.\ ] ] in order to complete the proof ( of the upper bound ) , we need to close the gaps " .
that is , we need to show that there exists @xmath198 such that @xmath199=1.\ ] ] to this end we act in the following way .
fix @xmath194 which is close enough to @xmath2
( to be determined later ) , and we show that @xmath200 , where @xmath201.\ ] ] therefore , using the borel - cantelli lemma we conclude that there exists @xmath202 for which @xmath203=1.\ ] ] next , recall that @xmath204=1,\ ] ] and set @xmath205 .
all in all , we obtain that with probability 1 , for every @xmath206 and for every @xmath207 we have @xmath208 + ( x_{a^k}-\mu_{a^k})\\&=(x_{n , a^k}-\mu_{n , a^k})+(x_{a^k}-\mu_{a^k})\\ & < \frac{\varepsilon}2\sigma_n\sqrt{2\log\log { n}}+(1+\varepsilon/4)\sigma_{a^k}\sqrt{2\log\log n}\\ & < ( 1+\varepsilon)\sigma_n\sqrt{2\log\log n},\end{aligned}\ ] ] as desired . in order to complete our argument
, we need to estimate @xmath209 and to show that indeed @xmath210 .
this is done in the following claim , which is a modification of levy s inequality to our special case of dependent random variable .
[ claim1 ] @xmath209 is such that @xmath210 .
fix @xmath211 . for each @xmath102 and for each @xmath212 , let @xmath213 denote the event @xmath214 .
let @xmath215 , @xmath216 , and for every @xmath217 define @xmath218 now , let @xmath219 and note that @xmath220 and that @xmath221 $ ] .
we start with evaluating the following probability : @xmath222=\sum_{j = a^k}^{n}\pr\left[a_j\cap \mathcal e_{n , a^k,\tau/2}\right].\end{aligned}\ ] ] note that if @xmath223 , then in particular we have @xmath224 therefore , we conclude that @xmath225 which is equivalent to @xmath226 moreover , a moment s thought reveals that for every @xmath227 , the events @xmath228 and @xmath229 are negatively correlated , and therefore , one can upper bound by @xmath230&\leq \sum_{j = a^k}^{n}\pr\left[a_j\right]\pr\left[x_{n , j}<\mu_{n , j}-\tau/2\right].\end{aligned}\ ] ] now , since clearly @xmath231 , and since @xmath232 , it follows by theorem [ janson ] that for every @xmath233 we have @xmath234&=\pr\left[\frac{x_{n , j}-\mu_{n , j}}{\sigma_{n , j}}<\frac{\tau}{2\sigma_{n , j}}\right]\nonumber \\ & = \exp\left(-\omega(1)\right)=o(1).\end{aligned}\ ] ] combining with and , we obtain @xmath235= \delta_k \cdot o(1).\end{aligned}\ ] ] as a penultimate step , we need to estimate @xmath236 $ ] . in order to do
so we first observe that since we choose @xmath194 to be very close to @xmath2 , it is easy to verify that in this case we have @xmath237 ( while @xmath238 ) . now , together with @xmath190 , these estimates imply that for some small constant @xmath239 we have @xmath240&=\pr\left[\frac{x_{n , a^k}-\mu_{n , a^k}}{\sigma_{n , a^k}}\geq \tau/(2\sigma_{n , a^k})\right ] \nonumber \\ & \leq \exp\left ( -\frac{\tau^2}{8\sigma^2_{n , a^k}}\right ) \nonumber\\ & = \exp \left(-\frac{ca^{4k}\log\log a^{k+1}}{a^{4k}(a-1)}\right),\end{aligned}\ ] ] and by choosing @xmath241 , one can upper bound with @xmath242 ( for large @xmath5 ) .
all in all , we obtain @xmath243&=\pr\left[m_n\cap \mathcal e_{n , a^k,\tau/2}\right]+\pr\left[m_n\cap \neg\mathcal e_{n , a^k,\tau/2}\right]\\ & \leq \delta_k\cdot o(1)+k^{-2},\end{aligned}\ ] ] and therefore , @xmath244 and @xmath210 as desired .
this completes the proof of the claim , and therefore the proof of the upper bound as well . before we proceed to the lower bound ,
let us make a few observations which can be obtained in a similar way as the above proof .
we make use of those in the next subsection . 1 . for every @xmath166 we have @xmath245=0 $ ] .
2 . for @xmath211 ,
let @xmath246 be the random variable counting the number of copies of @xmath4 with vertices from both @xmath247 and @xmath248 $ ] .
let us also denote by @xmath249 and @xmath250 its expectation and variance , respectively .
then , for every @xmath166 we have @xmath251=0.\ ] ] let @xmath166 be some fixed positive constant , we aim to show that @xmath252=1.\ ] ] to this end , we focus on integers @xmath253 of the form @xmath193 , where @xmath194 is a large enough constant to be determined later . for a fixed @xmath211 , let @xmath254 be the random variable that counts the number of copies of @xmath4 which
are fully contained in @xmath255 .
note that the set @xmath256 is clearly independent , and that the random variables @xmath254 are distributed the same as @xmath257 ( and therefore , @xmath258 for every @xmath5 ) .
therefore , one can easily check that for large @xmath259 and @xmath5 we have @xmath260 ( this can be verified using the simple observation that @xmath261 and the estimate given in section [ sec : calculations ] ) .
now , letting @xmath262 it follows by @xmath190 that for some @xmath263 we have @xmath264=\omega\left(k^{-1+\gamma}\right),\ ] ] and therefore , @xmath265=\infty.\ ] ] using the borel - cantelli lemma it thus follows that @xmath266=1.\ ] ] now , let us choose @xmath194 to be a fixed large enough constant so that for sufficiently large @xmath5 the following inequalities hold ( the existence of such @xmath259 for which all these inequalities hold follows immediately from the relevant estimates in section [ sec : calculations ] ) : 1 .
@xmath267 and 2 .
@xmath268 and 3 .
@xmath269 all in all , combining the above mentioned estimates and @xmath270-@xmath271 we conclude @xmath272 as desired .
this completes the proof . in this section
we verify , @xmath273 and @xmath271 , by estimating the relevant variances . before doing so , recall that @xmath274 where @xmath275 .
moreover , note that whenever @xmath276 and @xmath277 are independent , then @xmath278 . therefore , given a subset @xmath279
, it follows that @xmath280 in addition , recall that each of the @xmath281 s is an indicator random variable for an appearance of a certain copy of @xmath4 ( where @xmath282 and @xmath283 ) , and therefore we have @xmath284 next , recall that @xmath13 and @xmath285 are fixed constants and that we always assume @xmath259 and @xmath5 to be large enough . in particular , it easy to see that the ( asymptotically ) largest element in the right hand side of is the case @xmath286 .
now we can give some easy estimates .
* estimating @xmath287 : * recall that @xmath36 is a sum of indicator random variables for all the ( labeled ) copies of @xmath4 in @xmath81 .
therefore , there exists a constant @xmath138 ( which depend of the number of automorphisms which preserve some edge ) such that the number of pairs @xmath288 of copies of @xmath4 which intersect in exactly one edge is roughly @xmath289 .
therefore , running over all possible intersection edges we obtain that @xmath290 now , note that since @xmath291 by taking @xmath259 to be sufficiently large we obtain that @xmath292 which is of the same order of magnitude as @xmath293 .
this verifies . in order to verify @xmath273
all we need is to note that the quantity @xmath294 is a function that tends to infinity whenever @xmath259 does .
finally , in order to verify @xmath271 let us first estimate @xmath295 .
* estimating @xmath296 : * let @xmath297 and @xmath298 .
recall that @xmath246 counts the number of copies of @xmath4 with vertices from both @xmath299 and @xmath248 $ ] .
in this case , assuming @xmath259 goes to infinity , it is easy to see that the largest summand in [ sumvar ] is obtained whenever the intersection edge is between @xmath248 $ ] and @xmath255 .
therefore , for some constant @xmath300 ( which does not depend on @xmath259 ) we obtain @xmath301 note that by and it follows that @xmath302 , and therefore , by taking @xmath259 to be sufficiently large , @xmath271 trivially holds . throughout the next section we are going to let @xmath136 be the number of perfect matchings in @xmath303 ,
@xmath36 the number of perfect matchings cycles in @xmath33 and @xmath304 .
we aim to prove : @xmath305=1\]]it will be enough to show that for @xmath166 we have both the upper bound @xmath306=0,\nonumber\end{aligned}\ ] ] and the lower bound @xmath307=1.\nonumber\end{aligned}\]]note that in the equations above we have @xmath308 , but those can be replaced by @xmath309 since the two quantities are asymptotically equal .
we need to prove that for any fixed @xmath166 @xmath310=0.\ ] ] by corollary [ concentration0 ] , there is a constant @xmath149 such that for all @xmath311 @xmath312\ ] ] with probability at least @xmath313 .
taking @xmath314 , we conclude that with the same probability @xmath315 + \log k.\ ] ] we use the following approximation of the expected value , @xmath316=n!p_m^n\exp\left(-\frac{1-p_m}{2p_m}+o(1/n)\right)\ ] ] the calculation for which can be found in the appendix .
this yields , @xmath317 = \log ( n !
p_m^n ) - \frac{1-p_m}{2p_m } + o(1),\ ] ] with @xmath318 .
the rhs can be written as @xmath319 let @xmath320 be the random variable that counts the number of edges in @xmath33 . by conditioning on @xmath321 and using the union bound ( over the range @xmath322 ) , we can conclude that with probability at least @xmath323 @xmath324 where @xmath36 denotes the number of perfect matchings in @xmath33 , and @xmath325 is the indicator of the event @xmath326 that @xmath33 has at least @xmath327 and at most @xmath328 edges . by chernoff s bound ,
@xmath329 with probability at least @xmath323 . by the union bound @xmath330 with probability at least @xmath331 .
let @xmath332)/\sqrt{var ( e_n)}$ ] , we have @xmath333}{n^2}\right)\nonumber\\ & = \log\left(\left(\frac{p(1-p)}{n^2}\right)^{1/2}e_n^*+p\right)\nonumber\\ & = \log\left(p\left(\frac{1-p}{p}\right)^{1/2}\frac{e_n^*}{n}+p\right)\nonumber\\ & = \log p+\log \left(1+\left(\frac{1-p}{p}\right)^{1/2}\frac{e_n^*}{n}\right)\nonumber\\ & = \log p+\left(\frac{1-p}{p}\right)^{1/2}\frac{e_n^*}{n}+o(1/n^2 ) . \nonumber\end{aligned}\ ] ] plugging the last estimate into we obtain , with the same probability @xmath334 note that with probability at least @xmath335 we have @xmath336 , in which case @xmath337 becomes @xmath338 .
thus , with probability at least @xmath339 we obtain @xmath340 since @xmath341 , we have , by the borell - cantelli lemma that the event in holds with probability 1 for all sufficiently large @xmath19 . on the other hand , by the kolmogorov - khinchin theorem
, @xmath342 satisfies lil and thus @xmath343happens with probability @xmath2 for all sufficiently large @xmath19 .
for all sufficiently large @xmath19 , @xmath344 is larger than the error term @xmath345 , and we have @xmath346 proving equation ( [ upper pm ] ) .
for the lower bound we need to show that there exists a sequence @xmath347 of indices such that with probability 1 , @xmath348 holds for infinitely many @xmath5 .
let @xmath349 be a constant . by the proof of (
* theorem 15 ) , we know @xmath350happens with probability @xmath351 , and @xmath342 is as in the last section . from the standard proof of lil for the sum of iid random variables @xcite , we see that there is a sequence @xmath352 ( where @xmath353 is an integer larger than 1 ) for which we have : @xmath354 happens infinitely often with probability one .
restricting ourselves to this subsequence and denoting by @xmath114 the event that @xmath355 holds for @xmath253 , we have @xmath356=o(1/c^k)\]]so in particular we have @xmath357 < \infty\ ] ] by borel - cantelli lemma , we have that with probability equal to @xmath2 , for all large @xmath5 : @xmath358 let @xmath5 be large enough so that @xmath359 .
then , with probability equal to @xmath2 we have that for infinite many @xmath5 : @xmath360just as desired . throughout the next section we are going to let @xmath136 be the number of hamilton cycles in @xmath152 ,
@xmath36 the number of hamilton cycles in @xmath27 and @xmath304 .
the structure of the proof is identical to the one done for theorem , so we omit some of the calculations .
we aim to prove : @xmath361+\frac{1-p}{p}}{\sqrt{\frac{2(1-p)}{p}}\sqrt{2\log \log { n}}}=1\right]=1\]]it will be enough to show that @xmath166 we have both the upper bound @xmath362+\frac{1-p}{p}}{\sqrt{\frac{2(1-p)}{p } } } \geq ( 1+\varepsilon)\sqrt{2\log\log { n\choose 2 } } \textrm { for infinitely many } n\right]=0,\nonumber\end{aligned}\ ] ] and the lower bound @xmath362+\frac{1-p}{p}}{{\sqrt{\frac{2(1-p)}{p } } } } \geq ( 1-\varepsilon)\sqrt{2\log\log { n\choose 2 } } \textrm { for infinitely many } n\right]=1.\nonumber\end{aligned}\]]note that in the equations above we have @xmath363 , but those can be replaced by @xmath309 since the two quantities are asymptotically equal .
let @xmath166 , and let @xmath364 be the number of hamilton cycles in the complete graph @xmath81 . with this notation
one has , @xmath365=np_m^n\exp\left(-\frac{n^2}{2m}(1-p_m)+o(1)\right).\end{aligned}\]]for a proof of , the reader can check the appendix . by using corollary [ concentration ]
, we have @xmath366\ ] ] with probability at least @xmath367 . applying the log function and using estimate ( [ ey ] )
we obtain @xmath368let @xmath320 be the random variable which counts the number of edges in @xmath369 , by conditioning on @xmath321 and using union bound ( over the range @xmath156 ) , with probability @xmath335 we have @xmath370where now we use @xmath36 ( number of hamilton cycles in @xmath371 ) and @xmath372 is the indicator random variable that the number of edges in @xmath27 is in the range @xmath373 $ ] . by chernoff s bound ,
@xmath374 with probability at least @xmath335 .
hence , by the union bound we have @xmath375with probability at least @xmath376 . by a similar calculation to the one done in section , we get @xmath377 plugging it into , we obtain that @xmath378+\left(\frac{2(1-p)}{p}\right)^{1/2}e_n^*-\frac{n^2}{2}\left(\frac{1}{e_n}-\frac{1}{{n\choose 2}p}-\frac{p-1}{{n\choose 2}p}\right)+o(1).\nonumber\end{aligned}\ ] ] note that since with probability @xmath379 we have that ( say ) @xmath380 , it follows that @xmath381 without affecting the error probability .
all in all , with probability @xmath382 we have @xmath383 + \frac{1-p}{p}}{\sqrt{\frac{2(1-p)}{p}}}\leq e_n^*+o(1).\end{aligned}\ ] ] by the borel - cantelli lemma , we see that for large @xmath19 , with probability one , equation @xmath384 holds .
since @xmath342 satisfies lil , we can upper bound the rhs of of @xmath384 by @xmath385 for large @xmath19 with probability one . all in all , @xmath386 + \frac{1-p}{p}}{\sqrt{\frac{2(1-p)}{p}}}\leq(1+\varepsilon)\sqrt{2\log \log { n\choose 2}}\]]holds for all large @xmath19 with probability one , which proves the upper bound .
recall that in order to prove the lower bound one needs to show that for every @xmath166 we have @xmath387+\frac{1-p}{p}}{{\sqrt{\frac{2(1-p)}{p}}}}\geq ( 1-\varepsilon)\sqrt{2\log \log { n\choose 2 } } \text { for infinite many } n\right]=1\end{aligned}\]]by the proof of [ @xcite , theorem 1 ] we have that for any fixed constant @xmath349 : @xmath388+\frac{1-p}{p } } { \sqrt{\frac{2(1-p)}{p}}}>c\right]=o(1/n)\ ] ] by repeating the idea of the lower bound on theorem , we obtain @xmath389+\frac{1-p}{p}}{\sqrt{\frac{2(1-p)}{p}}}\geq ( 1-\varepsilon)\sqrt{2\log \log { n\choose 2}}\]]holds for infinite many @xmath19 with probability @xmath2 , which proves .
in this section we will be working with loose hamilton cycles in random hypergraphs @xmath18 .
note that we require that @xmath390 is an integer ( which shall denote the number of edges of a hamilton cycle ) .
thus , we will assume the divisibility condition @xmath391 throughout the rest of the section .
let @xmath83 be the set of all hamilton cycles in the complete @xmath5-uniform hypergraph on @xmath19 vertices . then , @xmath392 ways to label the vertices consecutively ( and the edges are determined trivially , including the one edge which goes back to the beginning of the labeling ) . in each of the @xmath393 edges , for the non - overlapping " vertices ( there are @xmath394 such vertices ) ,
the order is not important .
therefore , one should divide by @xmath395 .
finally , note that each hamilton cycle can be obtained in @xmath396 ways ( @xmath393 overlapping vertices " to be placed as vertex number @xmath2 , and two isomorphic ways to label the vertices consecutively ) .
now we are ready to prove theorem [ main : clt ] .
let @xmath320 denote the number of edges of @xmath18 , and @xmath397 be the number of hamilton cycles of @xmath18 .
the idea of the proof is to compare @xmath36 to @xmath320 .
specifically , we want to show that @xmath398\end{aligned}\]]goes to zero ( recall that for a r.v @xmath277 , we let @xmath399}{\sqrt{var(y)}}$ ] ) .
since clearly @xmath342 converges to @xmath9 , the theorem will follow . to this end
we will show that @xmath400 and @xmath342 are almost perfectly linearly correlated . meaning that @xmath401 . recall that @xmath402-\mathbb{e}[x_n]\mathbb{e}[e_n]}{\sqrt{var(x_n)var(e_n)}}.\end{aligned}\ ] ] let @xmath403 be the event
@xmath4 appears in @xmath18 " .
hence , @xmath404=p^m\ ] ] let @xmath405 ( that is , @xmath406 ) , and by linearity of expectation , we have : @xmath407=np^{m}\ ] ] also , since @xmath408 we have @xmath409 and @xmath410={n\choose
we compute the missing quantities .
denote by @xmath326 the set of edges in the complete @xmath5-uniform hypergraph , and denote by @xmath411 the event the edge @xmath412 appears in @xmath18 " .
then , @xmath413=\sum_{h\in\mathcal h , e\in\mathcal e } \mathbb{e}[x_h\cdot e_e]\]]by symmetry , by fixing one hamilton cycle @xmath414 , we have : @xmath413=n\left(\sum_{e\in\mathcal e } p^{|h\cup e|}\right)=n\left(\left ( { n\choose k } - m\right ) p^{m+1}+mp^{m}\right)\]]hence , @xmath415=\mathbb{e}[x_n](\mathbb{e}[e_n]+m(1-p))$ ] , and we get @xmath416(m(1-p))$ ] .
lastly , we compute the variance of @xmath36 .
@xmath417=\sum_{h_1,h_2\in\mathcal h } p^{|h_1\cup h_2|}\]]again , by fixing an arbitrary hamilton cycle @xmath4 , we get @xmath417=n\left(\sum_{h_1}p^{|h\cup h_1|}\right)\]]let @xmath418 be the number of hamilton cycles that intersect @xmath4 in exactly @xmath259 edges . with this notation , @xmath417=n\left(\sum_{a= 0}^{m } n(a ) p^{2m - a}\right)\]]let @xmath419 .
then , @xmath417=n^2p^{2m}\left(\sum_{a=0}^{m } \alpha_a p^{-a}\right)\]]hence , @xmath420)^2\left(-1+\left(\sum_{a= 0}^{m } \alpha_a p^{-a}\right)\right):=(\mathbb{e}[x_n])^2f(n)\]]plugging back into : @xmath421(m)(1-p)}{\sqrt{(\mathbb{e}[x_n])^2f(n){n\choose k}p(1-p)}}=\frac{m(1-p)}{\sqrt{{n\choose k}p(1-p)f(n)}}\end{aligned}\]]writing out @xmath422 : @xmath423hence , @xmath424we are going to show that the sum is negligible compared to the first summand .
first of all , note that @xmath425 by a simple union bound .
in general , to bound @xmath426 , we pick the @xmath427 edges from @xmath4 we are going to intersect . there are @xmath428 ways to do so .
next , collapse each one of those edges into a single vertex .
thus , we now have @xmath429 vertices .
note that the number of vertices is still divisible by @xmath430 , as it should be the case .
next , we form a hamilton cycle on these vertices
. there are @xmath431ways to do so . in order to see this , just note that we replace @xmath19 by @xmath429 and @xmath393 by @xmath432 in equation ( [ size ] ) .
lastly , once the hamilton cycle has been formed , we can uncollapse each one of the @xmath427 edges , so we obtain an extra factor of @xmath433 . hence , @xmath434for a constant @xmath138 depending on @xmath5 . plugging back on @xmath422 we get : @xmath435to handle the summation , we are going to split it into two sums : @xmath436note that in the range @xmath437 , we have by lemma [ lower factorial ] : * @xmath438 , * @xmath439 , and * @xmath440 . hence , @xmath441for @xmath442 , we can upper bound @xmath443 , and @xmath444 to obtain : @xmath445using this in the definition of @xmath446 we obtain : @xmath447thus , @xmath448where the second inequality is just from cauchy schwarz
. then we have that the lower bound is : @xmath449which we can re - write using a taylor expansion as : @xmath450hence , expanding and using we have : @xmath451=\mathbb{e}[x_n^*]+\mathbb{e}[e_n^*]-2cov(x_n^*,e_n^*)=2 - 2(1-o(n^{2-k})=o(n^{2-k})\ ] ] hence , when @xmath68 , we have that the above tends to zero .
this completes the proof of theorem [ main : clt ] .
now we are going to use theorem [ main : clt ] to derive lil for @xmath400 .
first we note that since @xmath452 is the summation of @xmath453 i.i.d .
random variables , then we have that @xmath342 obeys the lil . that is , @xmath454with probability @xmath2 for large enough @xmath19 and with probability 1 we also have @xmath455 infinitely often .
note that we write @xmath309 instead of @xmath456 , which holds because they are asymptotically equal ( as @xmath5 is fixed ) .
furthermore , @xmath457}{t^2}=o\left(\frac{1}{t^2n^{k-2}}\right)\]]let @xmath458 .
we obtain : @xmath459if @xmath460 , then we have : @xmath461and by the borel - cantelli lemma we have that with probability 1 , only finite many of those events can happen . that is , with probability 1 we have @xmath462 for all @xmath19 sufficiently large .
hence , with probability one , for infinitely many @xmath19 we have : @xmath463hence , we obtain the law of iterated logarithm for hamilton cycles provided that @xmath460 .
in this section we present new upper - tail estimates needed in the proofs of theorems [ thm : main ] and [ thm : main2 ] .
we denote by @xmath464 the complete bipartite graph and let @xmath465 denote the set of all perfect matchings in @xmath464
. clearly , we have @xmath466 for each @xmath467 , let @xmath468 to denote the indicator random variable for the event
@xmath469 appears in @xmath470 " .
it is easy to see that @xmath471 = \frac{(m)_n}{(n^2)_n},\ ] ] and @xmath472 = n !
\frac { ( m)_n } { ( n^2 ) _ n } = n!\left ( - \frac{1 -p_m}{p_m } + o(1/n ) \right)\ ] ] where @xmath473 . for the calculation of equation , see the appendix . in general , for any fixed bipartite graph @xmath4 with @xmath474 edges , the probability that @xmath303 contains @xmath4 is precisely @xmath475 thinking of @xmath4 as the ( simple ) graph formed by the union of perfect matchings @xmath476 , observing that @xmath477 , we obtain that @xmath478 = \sum_{p_1, ... ,p_k\in \mathcal p } \mathbb e [ x_{p_1 } \dots x_{p_k } ] = \sum_{a=0}^{(k-1)n}m(a)\frac{(m)_{kn - a}}{{(n^2)}_{kn - a}},\ ] ] where @xmath479 is the number of ( ordered ) @xmath5-tuples @xmath480 , whose union contains exactly @xmath481 edges .
our main task is to bound @xmath479 from above .
fix @xmath259 and let @xmath482 be the set of all sequences @xmath483 of non - negative integers where @xmath484 for each sequence @xmath485 , let @xmath486 be the number of @xmath5-tuples @xmath487 such that for every @xmath488 , we have @xmath489 .
clearly , we have @xmath490 we construct a @xmath5-tuple in @xmath491 according to the following algorithm : * let @xmath492 be an arbitrary perfect matching . *
suppose that @xmath493 are given , our aim is to construct @xmath494 .
pick @xmath495 edges to be in @xmath496 as follows : first , pick a subset @xmath497 of @xmath495 vertices from the first color class ( say @xmath498 ) .
next , from each vertex pick an edge which appears in @xmath499 so that the chosen edges form a matching .
let us denote the obtained partial matching by @xmath500 , and observe that @xmath501 , and that @xmath502 is a set of size @xmath495 ( where @xmath503 denotes the second color class ) .
* find a perfect matching @xmath504 between @xmath505 and @xmath506 which has an empty intersection with @xmath507 , and set @xmath508 .
next , we wish to analyze the algorithm
. there are @xmath509 ways to choose @xmath492 . having chosen @xmath493
, there are @xmath510 ways to choose @xmath497 .
each vertex in @xmath497 has at most @xmath511 different edges in @xmath499 .
thus , the number of ways to choose @xmath500 is at most @xmath512 .
moreover , once @xmath497 and @xmath513 are defined , the number of ways to choose @xmath504 is at most @xmath514 .
this way , we obtain @xmath515 by the multinomial identity and the definition of the set @xmath516 , @xmath517 therefore @xmath518 this estimate is sufficient in the case when @xmath259 is relatively large .
however , it is too generous in the case when @xmath259 is small ( the main contribution in lhs of comes from this case ) . in order to sharpen the bound ,
we refine the estimate on the number of possible @xmath504 s that one can choose in the last step of the algorithm , call this number @xmath519 ( clearly , @xmath519 also depends on the @xmath520s and we estimate a worse case scenario )
. let @xmath521 be the bipartite graph between @xmath505 and @xmath522 formed by the edges which are not in @xmath499 .
for each @xmath523 , let @xmath524 be its degree in @xmath521 . by the bregman - minc inequality ( see theorem [ bregman ] )
@xmath525 it is clear from the definition that for each @xmath526 @xmath527 call a vertex @xmath526 _ good _ if @xmath528 and _ bad _ otherwise .
it is easy to see that @xmath526 is good if and only if it has exactly @xmath511 different edges in @xmath507 and none of these edges hits @xmath513 .
it follows that the number of good vertices is at least @xmath529 since @xmath530 is monotone increasing , it follows that @xmath531 comparing to the previous bound of @xmath532 , we gain a factor of @xmath533 ^{n - ka}.\ ] ] a routine calculation ( see appendix ) shows that whenever @xmath534 , the rhs is @xmath535 thus , for such values of @xmath259 , we have @xmath536 where the constant 2 can be replaced by any constant larger than 1 .
now we are ready to bound @xmath537 .
recall ( [ rearranged ] ) @xmath538 we split the rhs as @xmath539 where @xmath540 .
the assumption @xmath541 of the lemma guarantees that @xmath542 .
let @xmath543 . by and lemma [ lower factorial ] and a routine calculation , we have @xmath544 on the other hand , @xmath545 so @xmath546 where @xmath547 is a constant depending on @xmath13 .
( in fact we can replace the constant @xmath548 by any constant larger than 1 in the definition of @xmath547 ; see the remark following ) .
to bound @xmath442 , we use and lemma [ lower factorial ] to obtain @xmath549 notice that we no longer have the term @xmath550 .
however , as @xmath259 is large , there is a much better way to bound @xmath551 stirling s approximation yields @xmath552 it follows that @xmath553 and thus is negligible for our needs
. therefore , @xmath554= s_1 + s_2 \le c_1^k ( n ! ) p_m^{nk}.\ ] ] finally , note that implies @xmath555)^k = ( n!)^k p_m^{nk } \exp \left(\frac{k(1-p_m ) } { p_m } + o(k / n)\right ) \ge c_2(n!)^k p_m^{nk},\ ] ] for an appropiate constant @xmath556 .
thus , we get @xmath557/(\mathbb{e}[x_{n , m}]^k)\leq c^k$ ] by setting @xmath558 .
let @xmath81 be the complete graph of @xmath19 vertices and denote by @xmath83 the set of hamilton cycles in @xmath81 .
clearly , @xmath559 for each @xmath414 , let @xmath403 denote the indicator random variable for the event
@xmath4 appears in @xmath152 " .
it is easy to see that @xmath404=\frac{(m)_n}{{n\choose 2}_n}\]]thus , @xmath560=n\frac{(m)_n}{{n\choose 2}_n}\end{aligned}\]]where above and henceforth we let @xmath364 . by lemma [ lower factorial ] , @xmath561hence , calculating the @xmath5-th moment
we obtain : @xmath562=\sum_{h_1,\ldots , h_k\in\mathcal h}\mathbb{e}[x_{h_1}\ldots x_{h_k}]=\sum_{a=0}^{(k-1)n}m(a)\frac{(m)_{kn - a}}{{n\choose 2}_{kn - a}}\end{aligned}\]]where @xmath479 is the number of ( ordered ) @xmath5-tuples @xmath563 .
the following lemma gives us bounds for @xmath479 , and it is true for @xmath153 . [
upper bound ma]for @xmath479 defined above , if @xmath564 we have : @xmath565and for @xmath566 we have the following weaker bound : @xmath567 splitting the sum in , @xmath568&=\sum_{a=0}^{\log^3 n}m(a)\frac{(m)_{kn - a}}{{n\choose 2}_{kn - a}}+\sum_{a=\log^3 n+1}^{(k-1)n}m(a)\frac{(m)_{kn - a}}{{n\choose 2}_{kn - a}}=s_1+s_2\end{aligned}\]]allows us to use lemma .
we bound the two sums separately : @xmath569on the other hand , @xmath570so @xmath571for some appropriate constant @xmath547 ( which depends on @xmath572 ) . to bound @xmath442 : @xmath573however for this case , it is enough to bound the summation using stirling s approximation , and use @xmath574 : @xmath575it follows that @xmath576and is thus totally negligible for our needs .
therefore , @xmath142=s_1+s_2\leq c_1^kn^kp_m^{nk}\]]finally , raising equation ( [ exp x ] ) to the @xmath5-th power yields : @xmath577)^k = n^kp_m^{nk}\exp\left(-\frac{(1-p_m)k}{p_m}+o(1)\right)\geq c_2^kn^kp_m^{nk}\]]for some constant @xmath556 .
hence , @xmath578}{(\mathbb{e}[x_{n , m}])^k}\leq ( c_1/c_2)^k\]]and setting @xmath558 finishes the proof .
fix @xmath579 , and let @xmath482 be the set of all the sequences @xmath580 of non - negative integers
where @xmath581for each @xmath582 , let @xmath491 be the number of @xmath5-tuples @xmath583 such that for @xmath584 we have @xmath585 .
clearly we have , @xmath586we know describe how to construct @xmath5-tuples in @xmath491 . 1 .
pick an arbitrary @xmath587 .
assume we are given @xmath588 .
construct a set @xmath500 of edges , of size @xmath495 such that @xmath589 .
3 . complete @xmath500 into a hamilton cycle .
next we analyze the algorithm .
clearly there are @xmath590 ways to perform the first step . for the moment , assume that the number of ways to perform step 2 and 3 ( for a fixed @xmath427 ) is given by : @xmath591then , for fixed @xmath592 we would have the following upper bound on @xmath491 : @xmath593by the multinomial identity and the definition of the set @xmath516 we have , @xmath594so we obtain the upper bound on @xmath479 , @xmath595as claimed .
hence to finish we need to upper bound steps 2 - 3 of the algorithm .
+ + * upper bound on steps 2 and 3 .
* assume we are given @xmath596 .
for each vertex @xmath526 , consider the set @xmath597 defined as follows : @xmath598which we shall refer to as the list of bad vertices of @xmath526 .
note that for each @xmath526 , we have @xmath599 . pick a subset @xmath600 of size @xmath495 , say @xmath601 .
we can do so in @xmath602 ways .
then , for each @xmath603 , we select an element , @xmath604 , on its list @xmath605 , and we match them through an edge . hence , we have at most @xmath606number of ways to construct @xmath500 .
now our task is to upper bound the number of ways we can complete @xmath500 into a hamilton cycle without using any edges in @xmath607 .
+ + first , we are going to collapse the edges in @xmath500 into vertices , and identify them by @xmath604 .
hence , we now have @xmath608 as vertex set ( that is , @xmath609 vertices ) . we are going to upper bound a bigger quantity : the number of * oriented * hamilton cycles , such that for no vertex @xmath526 , we have @xmath610 for some @xmath611 , which henceforth we shall refer to as @xmath526 is bad " .
+ + let @xmath612 be the quantity we wish to upper bound ( that is , the number of oriented hamilton cycles with no bad vertices ) . hence ,
@xmath613where @xmath614 .
we now give upper and lower bounds on @xmath615 , and we also argue why it is enough to consider the terms up to @xmath616 : + + * upper bound on @xmath617 * : first we choose the @xmath128 vertices that will be bad .
there are @xmath618 ways to do so .
say we chose @xmath619 .
then there are at most @xmath620 many ways to make each vertex bad , hence a total of at most @xmath621 ways to make @xmath622 bad ( @xmath623 ) .
hence , we have @xmath624 for some @xmath625 in its set @xmath626 .
collapse @xmath622 and @xmath625 onto a single vertex ( for @xmath623 ) , so now we have @xmath627 vertices
. then form any oriented hamilton cycle on these vertices , so we have @xmath628 ways to do so ( then uncollapse them to obtain an oriented hamilton cycles on @xmath609 vertices ) . hence , @xmath629but since we are considering @xmath630 and @xmath631 we have : @xmath632 + + * truncation : * we show that @xmath633 is small . indeed , @xmath634where the second to last inequality holds since the summands are in decreasing order ( as @xmath427 is at most @xmath635 ) .
+ + * lower bound on @xmath636 * : for this bound , we are only going to consider @xmath619 such that their lists are disjoint . intuitively , almost all @xmath618 options are good since the sizes of the lists are of order @xmath427 ( which will be logarithmic ) .
let @xmath637 be the number of @xmath619 such that @xmath638 for @xmath639 and @xmath640 .
hence , @xmath641now , we compute @xmath637 : first we choose @xmath642 so that @xmath643 .
there are @xmath644 options for @xmath642 .
then , choose @xmath645 so that @xmath646 and @xmath647 is empty .
there are at most @xmath648 many vertices , @xmath649 , such that @xmath650 is not empty ( to see this note that @xmath651 has size @xmath652 and each member of @xmath651 is in at most @xmath652 many lists ) .
hence , the number of ways to pick @xmath645 is at least @xmath653 .
continue in the manner to obtain ( after dividing by the @xmath654 that comes from double counting ) the following lower bound : @xmath655where the last inequality uses @xmath656 and @xmath657 .
we compare with @xmath618 as follows : @xmath658where above we use @xmath659 which is valid for @xmath630 .
hence , putting everything together we arrive at the lower bound : @xmath660hence , we have that for all @xmath630 the following bounds on @xmath615 : @xmath661which implies : @xmath662where the last equality uses the fact that @xmath663 .
putting everything together we have : @xmath664thus , the number of ways to complete @xmath500 into a hamilton cycles is upper bounded by : @xmath665putting it together with the upper bound on the number of ways to construct @xmath500 we obtain that the upper bound on steps 2 and 3 of our algorithm is given by : @xmath666
* proof of lemma [ lower factorial ] : * let @xmath667 be such that @xmath158 .
then , @xmath668as claimed .
+ + * approximation of expected value ( perfect matchings ) : * for a subgraph @xmath4 of @xmath464 with exactly @xmath474 edges , the probability that @xmath4 appears in @xmath303 is exactly : @xmath669let @xmath4 be a perfect matching on @xmath464 , then @xmath670 , so we can apply lemma [ lower factorial ] to obtain : @xmath671where in the last equality we used @xmath672 . since there are a total of @xmath509 perfect matchings , we obtain by lineartiy : @xmath316=n!p_m^n\exp\left(-\frac{1-p_m}{2p_m}+o(1/n)\right)\ ] ] * approximation of expected value ( hamilton cycles ) : * just like above , let @xmath4 be a hamilton cycle in @xmath81 . then the probability that @xmath4 appears in @xmath152 is given by : @xmath673by linearity , one obtains the desired approximation .
+ + * computation of equation * : we are going to use the following upper and lower bounds for the factorial : @xmath674hence , @xmath675 ^{n - ka}&\leq\left[\frac{(\sqrt{2\pi d}(d / e)^de^{1/12d})^{1/d}}{(\sqrt{2\pi d}(d / e)^d)^{1/d}}\right]^{n - ka}\nonumber\\ & = \left[(1+o(n^{-2}))\frac{d(2\pi d)^{1/2d}}{d(2\pi d)^{1/2d}}\right]^{n - ka}\nonumber\\ & = ( 1+o(n^{-1})\left[\frac{(2\pi d)^{1/2d}}{(2\pi d)^{1/2d}}\right]^{n - ka}\left[\frac{d}{d}\right]^{n - ka}\nonumber\\ & = ( 1+o(1))\left[1-\frac{t-1}{n-\ell_t}\right]^{n - ka}\nonumber\\ & = ( 1+o(1))e^{t-1}\nonumber \end{aligned}\]]as desired .
( here we use the assumption that @xmath676 . )
we would like to thank kyle luh for his useful comments during the draft of this paper .
a. barbour , m. karoski , and a. ruciski .
_ a central limit theorem for decomposable random variables with applications to random graphs_. journal of combinatorial theory , series b , 47(2 ) ( 1989 ) , 125-145 . | a milestone in probability theory is the _ law of the iterated logarithm _ ( lil ) , proved by khinchin and independently by kolmogorov in the 1920s , which asserts that for iid random variables @xmath0 with mean @xmath1 and variance @xmath2 @xmath3 = 1 .\ ] ] in this paper we prove that lil holds for various functionals of random graphs and hypergraphs models .
we first prove lil for the number of copies of a fixed subgraph @xmath4 .
two harder results concern the number of global objects : perfect matchings and hamiltonian cycles .
the main new ingredient in these results is a large deviation bound , which may be of independent interest . for random @xmath5-uniform hypergraphs
, we obtain the central limit theorem ( clt ) and lil for the number of hamilton cycles . |
ultra - luminous x - ray sources ( ulxs ) are generally defined as extra - nuclear galactic x - ray sources with x - ray luminosities @xmath4 erg s@xmath1 , a luminosity that is seldom reached by any galactic ( milky way ) x - ray binary ( xrb ) .
if ulxs emit their x - rays isotropically , by eddington arguments , the central object could be a black hole ( bh ) with mass m @xmath5 5 m@xmath6 , and for eddington ratios of @xmath710@xmath8@xmath910@xmath1 , the implied mass of the central object is @xmath750@xmath9500 m@xmath6 .
therefore , some ulxs could be accreting intermediate - mass black holes ( imbhs , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) . the majority of ulxs , however , are likely accreting _ stellar _ mass black holes ( e.g. , @xmath1010
m@xmath6 ) as there appears to be a strong association between ulx sources and star formation in galaxies @xcite .
one popular model claims that ulxs are merely stellar - mass bhs in high - mass xrbs , and mild beaming of their x - ray luminosities produces the observational illusion of extremely high x - ray luminosities ( e.g. , * ? ? ?
* ; * ? ? ?
hundreds of ulxs are known at the present time ( e.g. , roberts & warwick 2000 , colbert & ptak 2002 , and chandra ulx catalogs by ptak et al .
, swartz et al . 2004 , and liu & bregman 2005 ) .
accreting binaries such as these could be precursors to eventual gravitational radiation events .
further , ulxs are important as they often dominate the total observed x - ray flux in starburst galaxies ( see * ? ? ?
their relationship with star formation also appears to extend to cosmologically interesting ( @xmath11 ) look - back times @xcite and may provide a cosmological tool .
unfortunately , all but the very brightest ulx sources are too faint for detailed x - ray spectroscopy studies ( the best cases must rely upon resolving power of r@xmath12 at 6.4 kev , available with x - ray ccds ; * ? ? ?
there is perhaps hope for ascertaining the nature of ulx sources by studying their interactions with their surroundings , such as gas , dust and stars ; such information may be captured using infrared spectral diagnostics . @xcite were the first to show that ionization - sensitive indices based on mid - infrared ratios were helpful for identifying the nature of heavily obscured nuclear sources @xcite .
we have thus taken advantage of the resolving power ( r @xmath13 ) of the _ infrared spectrograph _ ( irs ) on board the _ spitzer space telescope _ ( sst ) to calculate highly accurate ionization diagnostic ratios .
the nearby galaxy pair ngc 4485/90 provides an excellent opportunity for studying ulx sources ( see * ? ? ?
there are six ulxs in ngc 4485/90 ( d @xmath14 7.8 mpc , * ? ? ?
* ) , far more than is typical ( @xmath15 2 in most cases , and 0.2 , on average ; * ? ? ?
* ; * ? ? ?
* ) . the unprecedented sensitivity and angular resolution afforded by the _ spitzer _
space telescope s infrared spectrograph ( irs ) coupled with the proximity of ngc 4485/90 allow us to carry out this important experiment .
we have performed spectral mapping of the system ngc 4485/90 by using the mapping mode of the irs @xcite on board the sst .
rectangular regions of @xmath16 and @xmath17 ( i.e. zones of @xmath18 kpc and @xmath19 kpc at the system ) were mapped using both the short - high ( sh , @xmath20 @xmath3 m ) and long - high ( lh , @xmath21 @xmath3 m ) slits , respectively . six rectangular regions centered at the positions of the ulxs ( table [ tbl-1 ] )
were covered using a @xmath22 and @xmath23 grid array for the long and and short slits of the irs .
in addition two positions in ngc 4490 that are _ not _ near ulxs were observed with the same configuration as the one used for the ulxs .
the integration time was 240 s at each position , and the step size is half the slit width perpendicular to the slit .
this configuration allows net exposures of @xmath24 s for rectangular sizes @xmath25 and @xmath26 for long and short high slits , respectively
. sets of 8 exposures were taken for each position with the irs .
the pre - reduced basic calibrated data ( _ bcd _ ) files were obtained from the _ spitzer science center _ ( ssc ) data base . in order to extract the spectra for the sh and lh modes
, we used smart ( spectroscopy modeling analysis and reduction , * ? ? ? * ) which is publicly available software provided by the ssc .
we have used line identifications from @xcite , @xcite and nist @xcite .
we have combined both modes by re - scaling the lh mode spectra to the sh mode spectra by adding or subtracting a constant and then re - binning to a resolution of 0.0287 @xmath3m / pixel . in fig .
[ fig1 ] we show the combined spectra from all the positions . as we can see from fig .
[ fig1 ] , some of the line features are present or stronger in the ulx region than in the comparison region spectra .
the most striking examples are features such as [ s iv]10.51 , [ ne ii]12.81 , [ na i]14.6 , [ ne iii]15.55 , [ s iii]18.71 , and [ s iii]33.48 @xmath3 m that seem to be significantly stronger in the comparison regions and ulx-1(which is thought to be a snr rather than an accreting binary ) , than in the rest of the sources . in the opposite way , the [ si ii]34.82 @xmath3 m line is stronger in ulx-2 to ulx-6 . the purpose of this work is to generate some emission line diagnostic diagrams similar to commonly used optical emission line diagrams , which have proven useful in separating accretion - dominated sources from those dominated by stellar sources ( e.g. , * ? ? ?
* ; * ? ? ?
* and references therein ) .
a popular mid - infrared diagnostic first put forth in an iso spectroscopic survey of ultra - luminous infrared galaxies ( ulirgs ) by @xcite , and later explored by @xcite , plots the emission line ratio [ o iv]25.91 @xmath3 m @xmath27 [ ne ii]12.81 @xmath3 m versus the strength of a mid - infrared _ poly aromatic hydrocarbons _ ( pah ) features .
these two lines however , are relatively weak and difficult to detect in lower luminosity systems . taking advantage of the strong low ionization line emission of [ si ii]34.82 @xmath3 m , @xcite suggested alternative diagrams to distinguish agns from sf regions such as [ ne iii]15.55@xmath3m@xmath27[ne ii]12.81@xmath3 m and [ s iii]33.48@xmath3m@xmath27[si ii]34.82@xmath3 m .
these ratios proved to be successful in separating accretion - powered vs. sf powered regions in the sings galaxy sample .
we have calculated these ratios for our eight spectra , shown in fig .
we have also plotted the data from @xcite that include nuclear and extra - nuclear regions from sings , the smc / lmc and galactic h ii regions .
errors in the emission line fluxes calculation are around 10% . the boundaries in fig .
[ fig2 ] are defined with the same curves as in @xcite . as we can see from fig .
[ fig2 ] , ulx-2 , ulx-3 , ulx-4 , ulx-5 , and ulx-6 are in zone i where seyfert and liners are observed , while ulx-1 , comp - nw , and comp - cent are located in zone iii as star forming systems . for ulx-2 and ulx-6 the line ratio [ s iii]33.48 @xmath3m@xmath27[si ii]34.82 @xmath3
m is @xmath28 , which is extremely low compared with extragalactic objects in fig .
the sings group has recently developed models ( _ pahfit _ , * ? ? ?
* ) to characterize pah features .
the models include starlight continuum , featureless thermal dust continuum , pure rotational lines of h@xmath29 , fine - structure lines , dust emission features , and dust extinction .
the starlight continuum is represented by a blackbody emission was fixed at @xmath30 k. nine thermal dust continuum components are represented by blackbodies at fixed temperatures @xmath31 k. the program _
pahfit _ performs a fit through two hundred iterations , each iteration determines a @xmath32 value for all nine temperatures , until a minimum value for @xmath32 is found .
dust temperatures are based on the shape of the continuum spectrum .
we show the corresponding fits in table [ tbl-1 ] , where we show ulx-2 and ulx-6 have hotter components , but the relative intensity of those components is low for ulx-1 . the presence of a hotter dust component will flatten the spectra for our range of wavelengths .
this effect is seen in fig [ fig1 ] for ulx-2 and ulx-6.the rest of the parameters modeled will be subject of a forthcoming paper ( vzquez , et al . 2007b , _ in preparation _ ) .
finally , we have found a strong linear anti - correlation between the observed @xmath33 and @xmath34([s iii]33.48 @xmath3m@xmath27[si ii]34.82 @xmath3 m ) .
the regression and intercept coefficients are @xmath35 and @xmath36 , respectively with a rms = 0.08 .
for the first time the ir continuum and lines ( @xmath37 @xmath3 m ) have been observed in ulxs by using the power of the _ spitzer _ irs .
it is likely that this ulxs are irradiating nearby star - forming clouds .
although the geometry is probably thus quite complex , we do find some hints of information on the underlaying x - ray radiation .
the most striking result from the data analyzed in this work is the one in fig .
[ fig2 ] , which shows that five of the six ulxs reported by @xcite fall in the `` agn zone '' determined by the infrared diagnostic emission line ratios defined by the emission lines of ne , s , and si @xcite .
both of the comparison regions used here , which do not harbor ulx sources , fall in the zone of sf .
the sf zone is also occupied by extra - nuclear star - forming regions in the sings galaxy sample and smc / lmc and galactic h ii regions
. the emission line at [ n i]10.33 and other unidentified features also behave like [ si ii]34.82 @xmath3 m , showing a stronger contribution in ulx-2 to 6 than in ulx-1 and both of the comparison regions .
the presence of a hotter dust component in our pah / continuum modeling ( see table [ tbl-1 ] ) , suggests that at least for ulx-2 and ulx-6 the [ si ii ] lines could be originating from an accreting source .
a dilution factor could be affecting the rest of the ulxs possibly due to their location within their host galaxy . all of this is consistent with the anti - correlation we have found between the line emission ratio of [ s iii]33.48@xmath3m@xmath27[si ii]34.82@xmath3 m and the observed x - ray luminosity of the six ulxs , the more luminous the sources the lower the ratio of these emission lines .
different scenarios for the strong [ si ii ] line have been suggested in the literature . in these scenarios
the increasing density in h ii , photo - dissociation ( pdr ) , and x - ray dissociation regions ( xdr ) , respectively , increases the intensity of the [ si ii]34.82 @xmath3 m line ( see * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
a third scenario suggests that heavy elements such as si , mg , and fe may be returned to the gas phase by dust destruction ( e.g. , sputtering ) in regions subject to strong shocks caused by stellar winds , starbursts , and agn activity @xcite .
details of the geometry and ionization balance in the region require more detailed fitting , and also spatial mapping of spectral parameters as ulxs are thought to have drifted from their birthplaces .
the origin of these lines could thus be disentangled with detailed irs mapping observations and irac imaging , which are the subject of a forthcoming paper ( vzquez et al .
2007b ) . with the exception of a weak detection in ulx-1
, we have not detected the emission line [ o iv]25.91 @xmath3 m , this line is a strong feature in snr ( e.g. , * ? ? ?
* ; * ? ? ?
* ) , supporting previous suggestions that ulx-1 is a snr ( e.g. , based on radio and x - ray constraints ; * ? ? ?
the mid - infrared emission line diagnostics used to uncover the nature of agns and sf galaxies has been shown here to work well separating sfrs and young sn from probably true accreting ulxs ( according to x - ray observations ; * ? ? ?
but this is a small sample of ulxs to claim a general rule that could be improved observing a higher number of ulxs .
the present work shows that a new diagnostic window has been opened on ulxs and their immediate environment via important mid - infrared features like [ si ii]34.82 @xmath3 m , made observationally available by the extraordinary performance of the irs on board the _ spitzer _ space telescope . armus , l. , et al .
2006 , , 640 , 204 baldwin , j. a. , phillips , m. m. , & terlevich , r. 1981 , , 93 , 5 colbert , e. , et al .
2004 , , 602 , 231 colbert , e. , & mushotzky , r. 1999 , , 519 , 89 colbert , e. , & ptak , a. 2002 , , 143 , 25 dale , d. a. et al .
2006 , , 646 , 161 genzel , r. et al .
1998 , , 498 , 579 higdon , s. j. u. , et al .
2004 , , 116 , 975 hornschemeier , a. e. , et al .
2004 , , 600 , 147 houck , j. r. , et al .
2004 , , 154 , 18 kaufman , m. j. , wolfire , m. g. , & hollenbach , d. j. 2006 , , 644 , 283 kauffmann , g. , heckman , t. m. , de lucia , g. , brinchmann , j. , charlot , s. , tremonti , c. , white , s. d. m. , & brinkmann , j. 2006 , , 367 , 1394 king , a. r. , et al .
2001 , , 552 , 109 king , a. r. 2006 , in compact stellar x - ray sources . edited by walter lewin & michiel van der klis .
cambridge astrophysics series , no .
cambridge , uk : cambridge university press , p. 507
laurent , o. , et al .
2000 , , 359 , 887 lehmer , b. d. , brandt , w. n. , hornschemeier , a. e. , alexander , d. m. , bauer , f. e. , koekemoer , a. m. , schneider , d. p. , & steffen , a. t. 2006 . , 131 , 2394 liu , ji - feng , bregman , j. n. 2005 , , 157 , 59 maloney , p. r. , hollenbach , d. j. , & tielens , a. g. g. m. 1996 , , 466 , 561 martin , w. c. , sugar , j. , musgrove , a. , dalton , g. r. , wiese , w. l. , & fuhr , j. r. 1995 , nist database for atomic spectroscopy , version 2.0 , nist standard reference database 61 ( gaithersburg : nist ) meijerink , r. , & spaans , m. 2005 , , 436 , 397 miller , c. m. , & colbert , e. j. m. 2004 , int . j. mod .
d , vol . 13 , no .
( astro - ph/0308402 ) miller , j. m. , zezas , a. , fabbiano , g. , & schweizer , f. 2004 , , 609 , 728 morris , p. w. , et al .
2006 , , 640 , l179 peeters , e. , spoon , h. w. w. , & tielens , a. g. g. m. 2004b , , 613 , 986 ptak , a. f. , & colbert , e. j. m. 2004 , , 606 , 291 ptak , a. , & griffiths , r. 1999 , , 517 , 85 roberts , t. , et al .
2002 , , 337 , 677 roberts , t. , & warwick , r. 2000 , , 315 , 98 smith , j. d. t. , et al .
2004 , 154 , 159 smith , j. d. t. , et al .
2007 , accepted sturm , e. , et al .
2002 , , 393 , 821 swartz , d. a. , ghosh , k. k. , tennant , a. f. , & wu , k. 2004 , , 154 , 519 tully , r. b. 1988 , catalogue of nearby galaxies , cambridge and new york , cambridge university press , 1988 , p 221 williams , r. m. , chu , y. h. , gruendl , r. 2006 , , 132 , 1877 zezas , a. , & fabbiano , g. 2002b , , 577 , 726 1 & 12:30:29.5 & + 41:39:27 & ngc 4490 radio src .
possible snr & 1.0 ( 4.9 ) & 50 , 75 , & 72000 , 13000 & 3310 + & & & & & 200 , 300 & 0.1 , 1.1 & + 2 & 12:30:30.6 & + 41:41:42 & ngc 4485 x-1 & 4.0 ( 4.6 ) & 135 , 200 & 56 , 2.3 & 4810 + 3 & 12:30:30.8 & + 41:39:11 & ngc 4490 transient x - ray src . & 2.9 ( 4.6 ) & 90 , 110 & 740 , 240 & 3.93 + 4 & 12:30:32.3 & + 41:39:18 & ngc 4490 x-1 & 1.9 ( 2.6 ) & 75 , 90 & 2500 , 960 & 1.75 + 5 & 12:30:36.3 & + 41:38:37 & ngc 4490 x-2 & 1.9 ( 2.6 ) & 75 , 90 & 8800 , 16 & 7.94 + 6 & 12:30:43.2 & + 41:38:18 & ngc 4490 x-4 & 3.1 ( 4.4 ) & 75 , 90 & 360 , 580 & 2990 + & & & & & 110 , 135 & 190 , 10 & + | constraining the astrophysical nature of ultra - luminous x - ray ( ulx ) sources , which have x - ray luminosities exceeding 10@xmath0 erg s@xmath1 , has been elusive due to the optical faintness of any counterparts . with high spectral resolution observations in the @xmath230@xmath3 m wavelength range
we have conducted an experiment to study six ulx sources in the ngc 4485/90 galaxy pair .
we have found that five of the six ulxs , based on mid - infrared spectral diagnostics , show the characteristic higher ionization features that are found in agn .
the sixth source , ulx-1 , is consistent with being a supernova remnant .
the chief infrared spectral diagnostics used are the ratios of [ s iii]/[si ii ] _ vs _ [ ne iii]/[ne ii ] . in two instances
fits to the continuum and _ poly aromatic hydrocarbons _ ( pah ) features also indicate higher dust temperatures that are characteristic of accreting sources .
overall , however , we find the continuum is dominated by stellar processes , and the best diagnostic features are the emission lines .
high spectral resolution studies in the mid - infrared thus appear to show great promise for determining the astrophysical nature of ulxs . |
there are four gamma - ray bursts ( hereafter grbs ) displaying the fe k@xmath0 emission lines in their afterglows ( @xcite,@xcite,@xcite,@xcite ) .
what are the implications of these observations ?
a fireball model @xcite , which explains the behaviors of grbs well , needs a central engine to give an initial energy input to the fireballs .
what is a central engine ?
this has been a long - term mystery and a controversial problem . however , with lines in the afterglows
, we have an important clue for understanding the environments around grbs . to construct a model which explains the line emission , the burst and the afterglow at the same time are of wide interest .
thus , many models which had been proposed so far before the first line detection in 1997 should be modified to explain the line emission . amongst them , we categorize them into two : binary - neutron - star merger models @xcite,@xcite and massive - star - related models ( e.g. , @xcite,@xcite,@xcite ) . between them , we take the massive - star - related models .
this is because binary - neutron - star merger models , which produce grbs during coalescence , take a time of almost the order of @xmath2 yr to merge and it happens far away from the star - forming regions .
this contradicts with the observational facts that grbs have often been detected in the star - forming region ( e.g. , @xcite,@xcite ) . on the contrary ,
the following two facts strongly motivate us to consider the massive - star - related models : ( 1 ) that the possible association of grb 980425 with the type ic supernova 1998bw was observed ( @xcite ) , in spite of the chance probability for a spatial and temporal coincidence of grb with the type ic supernova being less than @xmath3 , and ( 2 ) that the detections of grbs in the star - forming regions have been counted near twenty events , as stated above ( e.g.,@xcite,@xcite ) . now
, which model should we take amongst massive - star - related models ?
two well - known scenarios are supranova model ( @xcite,@xcite ) and hypernova model ( @xcite,@xcite,@xcite ) . in the scenario of supranova model ,
at first , the usual supernova explosion creates a rapidly spinning neutron star .
secondly , several months or years later , the neutron star , spinning down by emitting gravitational and electro - magnetic waves , collapses into a kerr black hole .
thirdly , the black hole s rotational energy is converted to power the grb @xcite .
they assumed the ring remnants , produced during the supernova explosion .
however it may not be probable , since the usual supernovae remnants have shell structures .
in addition , if the morphology of the remnant around grbs is shell - like , a fireball will be scattered inevitably by the supernova remnants to decelerate it within a day and will probably show detectable x - ray absorption or emission lines @xcite .
this contradicts the observation of x - ray afterglow of grb970508 which lasted for a year and showed no detectable x - ray absorption or emission lines . in their latest paper @xcite
, they assume a plerionic remnant , but it may be too speculative because of its special morphology . among hypernova models ( @xcite,@xcite,@xcite ) , @xcite considered a situation , in which an extended magnetically - dominated wind from a grb impacts the expanding envelope of a massive progenitor to produce the fe line , and most of the continuum could still be explained from the standard decelerating fireball .
also @xcite successfully reproduced the line and the continuum at the same time . here , we suggest an alternative and natural model .
we assumed a ring - like remnant , such that observed by hst around sn 1987a ( e.g. , @xcite,@xcite ) . in @xcite , they discussed a strongly anisotropic grb environment ; that is , a ring .
they calculated excellently the time evolution of the interaction between the ring and the fireball , as well as various radiative processes , photoionization , fluorescence , recombination , electron - impact ionization , compton scattering , bremsstrahlung and coulomb scattering . as a result
, they partially succeeded in reproducing the spectrum of grb 970508 .
according to their model , however , the continuum spectrum is dominated by the thermal bremsstrahlung emitted in the shocked region between the incoming fireball and the ring , which contradicts the power - law continuum suggested from the four above - mentioned observations with fe lines . also , in @xcite , matter amounting to @xmath4 with a velocity of @xmath5 cm s@xmath6 created by hypernova model @xcite is ejected along the equatorial plane to hit the ring created during the merger of the progenitors evolved to helium stars .
however thermal bremsstrahlung was also dominant over the spectrum .
if the systems of both the line - emission region and the continuum flux region are the same , the above problem seems to still be inevitable .
we present a model to reproduce the observation of grb 000214 , in which we consider two systems for the line emission and the continuum .
that is to say , the fe k@xmath0 line is produced by the fluorescence of the fe atoms in the equatorial ring illuminated by the x - ray afterglow .
on the other hand , the continuum power - law flux is obtained by the afterglow emitted from the fireball toward the observer ( see .
figure 1 ) .
moreover , we considered the interaction between the ring and the fireball to investigate whether the thermal bremsstrahlung emission from the shocked region , which was dominant for the continuum in @xcite , is real or not . as a result
, we find that the thermal bremsstrahlung emission from the shocked region is negligible by elaborately estimating the temperature behind the shock wave . in this paper
, we propose a more clear picture to explain the observational facts ; i.e. , flux level , spectrum shaping and duration of the fe line .
observational evidences of of grb 000214 are described in section 2 . in section 3 , a physical picture for our model is stated .
discussion are presented in section 4 .
we give here the observation of grb 000214 @xcite .
grb 000214 had a fluence of @xmath7 erg @xmath8 and a duration of @xmath9 s in the energy band of the gamma - ray burst monitor on board the bepposax satellite ( 40 700 kev ) .
the fluence of the prompt x - ray was @xmath10 erg cm@xmath8 ( 2 10 kev ) , which was detected by the wide field cameras , also on board the bepposax satellite .
a follow - up observation with the bepposax narrow field instruments began about 12 hr after the corresponding grb , and lasted for 104 ks .
the effective exposure time was 51000s on source - time for bepposax medium - energy concentrator/ spectrometer ( mecs ) and 15000s in the low - energy concentrator/ spectrometer ( lecs ) ; the energy band for each was 1.6 10 kev and 0.1 4.0 kev . a spectral analysis using the data from mecs and lecs showed that the energy spectrum had a power - law @xmath12 photon index of @xmath13 and @xmath14 ( 2 10kev ) = @xmath15 erg cm@xmath8 s@xmath6 .
line emission was identified with the fe k@xmath0 ( for hydrogen - like iron ) with a cosmological redshift of @xmath16 = 0.47 .
for the fe line fitting in the spectrum , a narrow gaussian was the best fit , which had a line - centroid energy of @xmath17 = 4.7 @xmath18 0.2 kev and @xmath19 = @xmath20 erg cm@xmath8 s@xmath6 .
the gaussian fit for the emission line was too narrow to find the intrinsic velocity of the matter . for grb 000214 , from its narrowness
, it can only be inferred that the intrinsic velocity was at most sub - relativistic .
we focus on the fact that the continuum flux decreased faster than the line flux during the observation of the x - ray afterglow .
this motivates us to think about independent systems for the line - emitting region and the continuum - emitting region .
what is the motivation for our model ?
as written at the end of section 2 , the continuum flux decreased faster than the line flux during the observation of the x - ray afterglow .
also the observed continuum spectrum was power - law .
if both the line emitting - region and the region for the continuum spectrum were supposed to be the same place , as assumed by @xcite , the line would disappear within about one day .
in addition , thermal bremsstrahlung emission is dominant . both of them are not in the cases for the observation of the afterglow of grb 000214 . here , we suggest an idea in which the systems for the line - emitting region and the continuum - emitting region are different . that is to say ,
non - thermal continuum spectrum of grb 000214 @xcite can be explained dominantly by the afterglow toward the observer , and the emission of the fe k@xmath0 line can be explained by the fluorescence of the fe atoms in the ring illuminated by the x - ray afterglow . in our model
, we needed an anisotropic environment around grb following @xcite ( see also figure.1 ) .
in addition , we assumed an anisotropic energy deposition of the burst .
an anisotropic energy deposition means that the energy per unit solid angle is emitted much stronger along the jet axis toward the observer than other regions .
here , we should clarify the configuration of the energy deposition .
basically , the fireball is emitted in every direction . to characterize the anisotropic energy distribution of a burst , we introduce two angles , @xmath21 in radian and @xmath22 in steradian , which are explained below . radiation from relativistically moving matter
is beamed in the direction of the motion to within an angle of @xmath23 radian , where @xmath24 is the lorentz factor of the relativistically moving matter .
also let @xmath25 be the angular size ( in steradian ) of the relativistically moving matter that emits a burst . to keep the total energy minimum , we set @xmath25 to be equal to @xmath26 .
we define @xmath27 , which is the energy emitted within @xmath28 str in the cone - like region , as stated above ( see figure.1 ) . as for the ring ,
@xmath29 is emitted within @xmath30 str , which is the covering angle ( in steradian ) of the ring region .
also , for all regions except the cone - like and ring regions , there is no way to specify the outflow of energy .
this is because it can not be observed by us due to the relativistic beaming effect stated above . for the region except for the cone - like and ring regions , @xmath31 , which is assumed to be @xmath32
is emitted when we calculate the total energy of the system ( see section 4 ) .
we note both @xmath27 and @xmath29 represent the fireball s kinetic energy . for clarity
, we divide the physics involved here into two parts to explain the observation of grb 000214 .
the fireball evolution toward us , which well explains the continuum spectrum , is in subsection 3.2 , and the shock interaction of the fireball with the ring is in subsubsection 3.3.1 , stating that the thermal bremsstrahlung emission from the shocked region is negligible . in subsubsection 3.3.2 , the line emission mechanism is stated .
we analyze the afterglow of grb 000214 following sari , piran and narayan @xcite .
seven free parameters required to determine the spectrum in their model are determined below .
@xmath33 is given as @xmath34 ( see figure.1 ) , which is the kinetic energy of the fireball , estimated as if the explosion were spherical .
in fact , if the explosion is jet - like , the intrinsic energy , @xmath27 , is much smaller .
also , @xmath35 is a parameter , which measures the ratio of the magnetic field energy density to the total thermal energy , @xmath36 is a parameter , which measures the fraction of the total thermal energy which goes into the electrons in the thermal motions , @xmath37 is the number density for the interstellar matter , @xmath38 is days from the burst , @xmath39 is the initial fireball s lorentz factor and @xmath40 is the distance from the grb center to the observer . actually , from section 2 , @xmath38 = 0.5 , @xmath41 cm , while @xmath37 is set to 1 @xmath42 .
then , the number of free parameters are reduced to four .
we fix @xmath43 ergs , @xmath44 and @xmath45 .
we notice here that about one tenth of @xmath33 is emitted as gamma - ray ( e.g.,@xcite ) .
the radiation of x - ray afterglow of grb000214 is specified as adiabatic and fast cooling with the above parameters ( e.g. , @xcite ) .
we explain it shortly below .
slow cooling & arbitrary @xmath36 & impossible + fast cooling & @xmath46 & @xmath47 + table1 : classification of the x - ray afterglow emission ( e.g.,@xcite ) .
+ the radiation of grb afterglows is generally divided into three types ( see table 1 ) . at first , we determine whether the radiation is fast or slow cooling . to do this , we are required to prepare the two quantities , @xmath48 , and @xmath49 .
both of them are given like below : @xmath50 if a single electron with lorentz factor : @xmath48 loses its all kinetic energy by the synchrotron radiation , it means that the electrons with lorentz factor greater than @xmath48 are rapidly cooled . on the other hand , as in @xcite , @xmath51 above @xmath49 , the number density of electrons,@xmath52,accelerated behind the shock obeys a power - law as @xmath53 .
if @xmath54 , it is fast cooling , because the large amount of the electrons are rapidly cooled down by the synchrotron radiation . on the contrary , if @xmath55 , it is a slow cooling , only small fraction of the electrons can be cooled . in our case
, it is the fast cooling with the above parameters .
next , we determine whether the radiation is radiative or adiabatic .
to do this , we need the quantity @xmath36 , which is the fraction of the total thermal energy @xmath56 which goes into the energy , @xmath57 , of the electrons in the random motions , and given as @xmath58 as the exact value of @xmath36 can not be determined from the observation , it is a free parameter .
we set @xmath59 to fit the observation . in the end
, we conclude the radiation is adiabatic and fast cooling . for an adiabatic and fast cooling blast wave ( from @xcite ) , @xmath60 where @xmath61 and @xmath62 correspond to the relevant lorenz factors , @xmath48 and @xmath49 , for the observed energy range of 2 10 kev .
the flux of the afterglow , @xmath14 , can be estimated as @xmath63 , where we take the value of p to 4 to fit the observation . in figure 2
, we can fit the observed continuum spectrum well .
near 1 kev , although it may seem that the spectrum is different , it is due to the absorption ( murakami , private communication ) .
thus , essentially , the spectrum is a good reproduction of the observation .
we divide this subsection into two to explain the fe line from the shocked ring . at first , we state the thermal history of the ring , and secondly the line - emission mechanism .
we parameterize the ring .
we assume a ring mass , @xmath64 of @xmath65 , an inner radius , @xmath66 of @xmath67 , the tenfold iron overabundance with respect to the solar abundance , @xmath68 of @xmath69 ; the energy shedding towards the ring , @xmath70 is assumed to be @xmath71 erg as the fireball s kinetic energy , the covering angle of @xmath72 ( see figure1 ) , and the width of the ring @xmath73r to be @xmath74 cm , which is from the duration of the observation lasting about 100 ks .
we notice here that about one tenth of @xmath70 is emitted as gamma - ray ( e.g. , @xcite ) and about one thousandth of @xmath70 is emitted as x - ray , it is inferred by comparing the fluence of the gamma - ray with that of x - ray in the observation of grb 000214 ( e.g. , @xcite ) . from the observation
, @xmath66 is determined by the time lag by about one day time lag between the occurrence of grb and the appearance of the fe line , @xmath75 the fireball emits grb at @xmath76 cm and the afterglow at @xmath77 cm , provided that one hundredth of the number density of the ring prevails to @xmath78 cm ( e.g. , @xcite ) . after the half day since the ignition of the fireball , powered by the explosion energy @xmath70 , the fireball with the initially loaded mass of @xmath79 hits the ring . if we define @xmath80 , where the fireball sweeps the amount of mass ( @xmath81 ) , then @xmath82 after the interaction the blast wave will be decelerated to sub - relativistic speed ( e.g. , @xcite ) : @xmath83 and @xmath84 it shows that before @xmath85 , the ultra - relativistic behavior of the shocked fluids can be described by the blandford - mckee solution @xcite , and after @xmath85 it can be described by the sedov - taylor solution .
we should estimate the temperature in the two regions , that is to say , that of the ultra - relativistic region ( hereafter urr ) , and of the newtonian region ( hereafter nr ) . at first , for urr , the energy density and the number density of the shocked region are estimated analytically ( e.g. , @xcite ) @xmath86 @xmath87 where @xmath88 is the distance of the shock wave measured from @xmath66 , @xmath89 is the distance of the matter behind the shock wave measured from @xmath66 , and @xmath90 is the lorentz factor of the shock wave in the rest frame of the unshocked ring . both of them can be estimated from @xmath70 and @xmath91 , as follows : ( e.g. , @xcite ) @xmath92 @xmath93 on the other hand , the energy density of the nucleus in the shocked region at a given temperature is @xmath94 equating equations ( 13 ) with ( 17 ) , we can estimate the upper limit of the temperature in the shocked region by setting @xmath95 . for the uur , @xmath96
also , for the nr , @xmath97 in @xcite , the temperature of the shocked region was estimated to be @xmath98 , assuming that the total kinetic energy of the incoming fireball was roughly equal to the total thermal energy in the shocked region . however , for the explicit description of the thermal history of the ring , it is inevitable to deal with the shock . therefore , we have solved the shock dynamics analytically to estimate the temperatures of shocked regions , which are classified into two cites , that is to say , urr and nr . for the thermal bremsstrahlung emission in the shocked region ,
the intrinsic emissivity is given as @xcite @xmath99 , where the summation are taken from j=1 ( hydrogen ) to j=26 ( iron ) , and @xmath100 represent electron number , mass fraction , the number density of the j species of nucleus , and the gaunt factor is excluded in the equation ( 20 ) , which is on the order of unity ( e.g. , @xcite , page 161 ) .
we remark that both @xmath101 , and @xmath102 in equation ( 20 ) are the quantities estimated in the nr region , because after @xmath103 , the dynamics of the shock wave can be described by the sedov - taylor solution . for the temperature in the equations ( 18 ) and ( 19 ) ,
the thermal bremsstrahlung emission from the shocked region is negligible . it is because under the relevant temperature ,
the energetic electrons can not be bent by the coulomb interaction with the nucleus , leading the emissivity of the thermal bremsstrahlung lower . and the fluorescent line emission will be suppressed , since all the nucleon will be ionized under the temperature for the production the fluorescent line . therefore if the line - emitting region and the continuum - emitting region are the same , as in @xcite , neither the continuum nor the line can be explained . in our model
, the problem can be resolved to take the two systems for the line - emitting region and the - continuum emitting region into consideration . in our model , the continuum can be determined only by the afterglow pointing toward us ( see figure 2 ) . for fe k@xmath0 emission ,
three mechanisms are proposed by @xcite , namely , fluorescence in an optically thin ring , thermal emission from the ring and reflection ( see figure1 in @xcite ) . in our model ,
the fe k@xmath0 line is produced by the fluorescence scenario , because the others have problems . in a model of thermal emission from the ring ,
if the ring is thermalized by the shocked region to produce line emission , it contradicts with the observation as stated in subsection 3.1 . in the reflection model , the line
is explained by the fluorescent fe line near the surface of the ring , if the ring is optically thick to the photons of the x - ray afterglow .
however , just after the arrival time of the burst photon , the fireball will hit the ring and will alter the non - thermal plasma for the fluorescent line into the thermal plasma .
it seems difficult to explain the line by the reflection model .
for grb 000214 , to avoid this problem , we considered slightly ionized fe atoms in an optically thin plasma .
we emphasize here that the fluorescence comes from an optically thin region .
as stated in subsection 3.3.1 , after @xmath104 s the shock wave becomes non - relativistic and the dynamics of the shock can be described by the sedov taylor solution .
therefore , the radii of the shocked wave , which can reach after one day from the burst , can be estimated as , @xmath105 thus , the effective volume , @xmath106 for the fluorescence is about half of the total volume of the ring , on the other hand , the volume of the shocked region is @xmath107 .
the shock wave has passed half of the total volume of the ring , in which the plasma is thermalized so as not to emit the line by fluorescence .
the time evolution of the fluorescence is stated roughly in the following . at first , photons from the x - ray afterglow are absorbed in the ring to render the electrons in the neutral fe atoms ionized .
secondly , the fluorescent k@xmath0 line is emitted by the transitions of the electrons .
the process of making the fluorescence line is effective at the time when the ring is illuminated not by a prompt burst , but by the x - ray afterglow .
this is because the recombination time - scale for explaining the observed line should be less than @xmath108 if the line is produced by the prompt gamma - ray ; on the other hand ; the recombination time - scale during the prompt gamma - ray is less than @xmath109s ( e.g.,@xcite ) .
we thus find that the line can not be produced by the prompt gamma - ray .
as we state in the following in detail , the recombination time required for the line is not very short for the x - ray afterglow . to explain the line flux by fluorescence
, we demand two conditions .
the first one is that the plasma in the ring should not be fully ionized , because if the region is fully ionized , line should not be emitted .
the second one is that the optical depth for the line should be on the order of unity .
this is because if the region is optically thick to line photons , we can not observe the line . in the following ,
we refer to the above two conditions .
the first one is that the plasma in the ring illuminated by the x - ray afterglow should not fully ionize the plasma . for determining the ionization state of the plasma , which is illuminated by the incoming photons
, we have only to compare the ionization time scale , @xmath110 , with @xmath111 . for this purpose
, we should specify the temperature of the ring after illumination by the x - ray afterglow .
the heating rate , @xmath112 , can be estimated as ( nakayama , masai in preparation ) , @xmath113 , where @xmath114 is the approximated photoionization cross section @xcite , @xmath115 , @xmath116 is the edge energy of fe k of 7.5 kev , and @xmath117 is the luminosity of the x - ray afterglow ( from 1 kev ) , where we assume the spectrum obeys power law , like that of the x - ray afterglow pointing toward us .
the quantity of @xmath118 , which is the illumination time by the x - ray afterglow of the ring , can be estimated as , @xmath119 the temperature of the illuminated ring of @xmath120 can be estimated as , @xmath121 and , @xmath122 we can estimate the recombination time - scale for the fe k@xmath0 line as @xcite , @xmath123 , where @xmath124 and where @xmath125 is given as , @xmath126 on the other hand , @xmath127 can be estimated as ( nakayama , masai in preparation ) @xmath128 with the above two time - scale , we can find that the time - scales for the ionization is comparable with that of recombination .
in other words , the plasma is not fully ionized by the incident photons of the x - ray afterglow .
we conclude that the fluorescence line is emitted from the non - thermal plasma .
secondly , we demand both the thomson optical depth , @xmath129 , and the optical depth of the bound free transition @xmath130 in the plasma be on the order unity .
this is because the large optical depth broadens and vanishes the line feature .
the optical depth of the thomson scattering can be estimated as , @xmath131 the optical depth for the bound free transition of a single fe atom can estimated as ( e.g. , @xcite ) , @xmath132 line photons are optically thin to the thomson scattering and the bound
free transition in the illuminated ring and can escape from the ring without smearing too much the iron line .
hence , we can roughly estimate the line emission by fluorescence . from @xcite , @xmath133 , where @xmath134 is the fraction of the total x - ray afterglow fluence absorbed by the ring and reprocessed into the line
; @xmath134 can be estimated as ( e.g. , @xcite ) @xmath135 where @xmath136 is the maximum energy of the x - ray afterglow , which is on the order of several ten kev and @xmath137 is the edge energy , which is 7.5 kev . as a result , we reproduced the line ( see figure.2 ) .
we discuss the significance of this study concerning the mechanism that produces a grb .
no simulations , challenging to explain the stellar collapses and the birth of the grb at the same time , have succeeded . in numerical simulations of collapsars ( e.g. , @xcite ) ,
a very steep collimation of the jet whose opening angle of about 1@xmath138 is generated .
on the contrary , from the observation of sn1998bw , only 1% of the polarization in the optical band was detected , which showed that the explosion may not be so collimated . in our model
, the fireball should spread out in every direction for illuminating the fe atoms in the equatorial ring to emit the fe k@xmath0 line by fluorescence , which may be consistent with the observation .
this picture will strongly constrain the grb models .
we also assumed in this study the existence of a ring which has a tenfold fe overabundance with respect to the solar abundance .
you may have thought that this value is relatively high for an object whose redshift ( z ) is 0.5 . however , because the grb are born in the star - forming region , the composition of a slightly high metalicity of fe might be justified .
it is noted that the optical flash of grb 990123 @xcite may be explained by the black body emission from the shocked region of the ring .
we will discuss this problem in the forthcoming paper .
we refer to the amount of the total energy emitted in our system as the kinetic energy of a fireball . if we take the initial lorentz factor as 200 , then from equation ( 1 ) , @xmath139 erg within @xmath140 str , and we can roughly estimate @xmath141 erg for the region except for the cone region . as a result ,
the total amount of kinetic energy is nearly @xmath142 erg .
this energy is compatible with the usual grbs .
you may wonder why there are only four gamma - ray bursts displaying fe line in the x - ray afterglow and the others are not .
we present two explanations for this question . at first
, this may be because of the variations of distance from an origin of gamma - ray burst to a ring .
if the distance is too far away , the flux emitted from the fireball will be so small that the resulting intensity of the fe line is too weak to be observed . secondly , this may be the variations of the opening angle of the jet .
if this is so collimated as not to illuminate the ring , the line will not be emitted .
it is concluded in our model that the angular size of the relativistically moving matter that emits the burst should be very large ( at least , the opening angle should be larger than several degrees ) in order to explain the intensity of the iron line .
thus , there should be correlation between the opening angle of the burst and existence of the iron line emission in the x - ray afterglow .
we hope that this tendency will be confirmed by the observations in the near future .
the duration of the line emission in the x - ray afterglow is nearly one day for grb 970508 @xcite , on the other hand , is more than one day for grb 000214 @xcite .
why are there variations of the duration of the fe line emission ?
we consider this may be because of the variations of the width of a ring .
if it is narrow , the line will disappear soon , on the other hand , if it is wide , the line will survive longer .
we think the former is the case of grb 970508 , and the latter is that of grb 000214 .
it is a difficult problem to determine observationally whether the line photons originate from helium - like ( 6.4 kev ) or hydrogen - like ( 6.95 kev ) fe atoms . for grb 970828
, it was determined that the latter is correct by detecting the emission lines from a possible host galaxy ( yoshida et al . in preparation ) , although there is no observation such like that in the case of grb 000214 .
the ionization parameter in our model can be estimated as in @xcite , @xmath143 the fe atoms are ionized for this value of the ionization parameter ( e.g. , @xcite ) .
this indicates the possibility that the line originates from hydrogen - like fe atoms in our model . to discuss whether the hydrogen - like or helium - like lines are prominent , we are supposed to require more detailed analysis on the emission - line features .
therefore , as a future work , we will perform a series of precise calculations on the ionization states in a ring illuminated by the x - ray afterglow and investigate the effects of fluorescence on the line emission in detail .
the authors are grateful to professor k. masai , professor h. murakami and dr .
d. yonetoku for useful discussions . | we present a model for explaining the fe k@xmath0 line and the continuum in the afterglow of grb 000214 . in this paper , we pose the importance to seek the physically natural environment around grb 000214 . for reproducing the observation , we need a ring - like remnant around the progenitor , like that of sn 1987a produced by the mass - loss of the progenitor and the fireball spread over all directions .
the observation of grb 000214 , in which the continuum power - law spectrum decreased faster than the line , motivated us to consider two independent systems for the line emission and the continuum spectrum . at first
, the continuum spectrum can be fitted by the afterglow emission of the fireball pointing toward the observer , which does not collide with the ring because the emission of grb and the afterglow are highly collimated to the observer by the relativistic beaming effect .
secondly , the line can be fitted by the fluorescence of the fe atoms in the ring illuminated by the x - ray afterglow .
the significance of this study is that our model may strongly constrain the grb models . @xmath1
gamma - rays : bursts x - rays : general ism : supernova remnants line : formation gamma - rays : individual ( grb 000214 ) |
it is not possible to measure a mass composition of primary cosmic rays ( pcr ) in the energy range @xmath0 ev using the direct method .
nothing remains , but to resort to the indirect methods when for the similar estimates the measurement of different components of extensive air shower ( eas ) are used .
that can be characteristics of longitudinal or lateral development of shower in the air .
and usually it is connected with the analysis of the most sensitive to the composition of shower components which differ from each other by the character of formatting and absorbing in the atmosphere , for example , of the charged particle flux ( electron , muon ) , the flux of erenkov or ionization radiation .
fig.[fig1 ] presents the eas yakutsk array experimental data and calculations by the model with the decelerated and moderate dissipation of the energy into the electromagnetic eas component : quasiscaling models ( solid line ) and qgsjet ( dashed line ) @xcite . from fig.[fig1 ] it is seen both the agreement of experimental data and calculations by qgsjet model ( proton ) in the region @xmath1 ev , and disagreement at @xmath2 ev .
the scaling model gives a noticeably greater value of @xmath3 in relation to the experimental data that is doubtlessly also connected with the break of scaling function in the region of ultra
high energies . the experimental data in fig.[fig1 ]
is well approximated by the expression of a form : @xmath4 the relation ( [ eq1 ] ) is primarily important for the comparison of estimations of @xmath5 obtained at the yakutsk and fly s eye arrays @xcite .
the calculations in @xcite ( see fig.[fig1 ] ) have been carried out by the qgsjet model in the case of primary proton and iron nucleus . a good agreement of our calculations in the case of the primary proton is observed .
the comparison of experimental data with calculations for the proton and iron nucleus indicates to the fact that the mass composition of particles of cosmic radiation in the energy region of @xmath6 ev and above @xmath7 ev must differ . at @xmath1
ev the mass composition is most likely close to the proton one .
the showers with @xmath9 ev and zenith angles @xmath10 have been chosen .
it is required so that the shower axis is within the array and not less than three muon detectors ( one of them is at a distance of 1000 m from the shower axis ) operate during the shower .
the density of muons flux @xmath11 is like a median between the densities adjusted to @xmath12 m , according to a mean lateral muon distribution function . in this case
the detectors which have been operated in the shower ( they have given zero indications ) , i.e. the detector threshold were used in analyze too .
the muon flux densities at have distances of 300 and 600 m from the shower axis have been calculated in analogous way .
results are in fig.[fig2]a and [ fig2]b
. calculations @xcite have shown that @xmath13 weekly depends on a zenith angle in the interval of @xmath14 .
so for the comparison of calculations and experimental data we take a mean angle @xmath15 . in order to take into account both physical fluctuations in measurement of @xmath11 and methodical ones ( apparatus errors and accuracy of the locate of the shower axis ) we have considered the judicial interval in one @xmath16 . in fig.[fig2]a and [ fig2]b the calculation result
is shown by gev ( % ) . @xmath17 and @xmath18 .
b ) @xmath11 vs. @xmath5 relations for observed events @xmath19 ev ( square ) , @xmath20 ev ( points ) and @xmath21 ev ( triangles ) . expected @xmath22 bounds for the distributions are indicated for proton , iron and gamma ray primary by different curve as in the legend.,title="fig:",width=283 ] gev ( % ) . @xmath17 and @xmath18 .
b ) @xmath11 vs. @xmath5 relations for observed events @xmath19 ev ( square ) , @xmath20 ev ( points ) and @xmath21 ev ( triangles ) .
expected @xmath22 bounds for the distributions are indicated for proton , iron and gamma ray primary by different curve as in the legend.,title="fig:",width=283 ] dotted line in the case of primary proton and by a dashed line for the iron nuclei .
a solid line shows the upper and lower limits for the case if eas with @xmath23 ev would generate by a primary @xmath24quantum .
the calculation for the primary @xmath24quantum has been taken from @xcite .
dots for @xmath25 and circles for @xmath26 the experimental data show in fig.[fig2]a . in fig.[fig2]b the showers in the energy range of @xmath19 ev are shown by squares , the showers with @xmath23 ev are shown by dots and the showers of maximum energy are shown by triangles .
the comparison of experimental data presented in fig.2a with calculations by the qgsjet model carried out for the case of the primary proton and iron nuclei confirms the hypothesis on the fact that the considerable portion of ultimate energy eas have been formed by protons .
their portion decreases below the energy @xmath27 ev .
it is seen from fig.2b , the basic mass of points ( showers with @xmath23 ev ) falls into the interval for the proton .
23 points fall into a zone of superposition of a proton and iron nuclei and 19 showers from 116 fall into a zone of upper boundary of calculation for the primary @xmath24quantum .
it testifies to the fact that even taking into account there exists a probability that showers of such energies can be generated by neutral particles and , in particular , by a primary @xmath24quantum .
then it is justified to use the analysis of directions of arrival of showers with @xmath23 ev for the search of the sources of highest energy cosmic rays . in this case
the determination accuracy of arrival angles of such showers must be not worse than @xmath28 .
a large number of erenkov detectors operating in the individual eas events and also the use of a new version of the qgsjet model allow to obtain quantitative estimations of mass composition of pcr .
for this aim we have compared experimental data ( see fig.[fig3 ] ) and theoretical predictions by the qgsjet model for different primary nucleus with the use of @xmath30 criterion .
the value of @xmath30 has been determined by the equality @xmath31 ^ 2}{{n_{\mathrm{theor}}}({x_{\mathrm{max } } } ) } \label{eq2}\ ] ] where @xmath32 is the experimental number of showers in the @xmath33 interval .
@xmath34 is the analogous number of showers calculated under the assumption that the mass number of the nucleus is equal to @xmath35 , and @xmath36 is the probability of the fact that the shower with the energy @xmath5 is formed by a primary particle @xmath35 .
then : @xmath37 the analysis of form of the experimental distribution of @xmath29 at optimal value of @xmath30 with a definite portion of probability does nt contradict to the following relationship for five nuclei component : @xmath38 ev : : : p : @xmath39% , @xmath40 : @xmath41% , m : @xmath42% , h : @xmath43% , fe : @xmath44% ; @xmath45 ev : : : p : @xmath46% , @xmath40 : @xmath47% , m : @xmath48% , h : @xmath49% , fe : @xmath50% ; @xmath51 ev : : : p : @xmath52% , @xmath40 : @xmath53% , m : @xmath54% , h : @xmath44% , fe : @xmath55% .
thus in the framework of the qgsjet model one can suppose that the mass composition of pcr changes transferring from the energy range @xmath56 ev to energy range @xmath57 ev .
at @xmath1 ev the primary cosmic radiation consists from @xmath58% protons and helium nuclei , a portion of the rest nuclei in the range where there is the second irregularity in the energetic spectrum type `` ankle '' does nt exceed @xmath59% .
a high content of proton and helium nuclei in pcr in the region of formation of `` ankle '' is most likely connected with an appreciable contribution into the overall flux of cosmic radiation in the earth s vicinity of radiation coming from beyond our galaxy limits .
this work has been financially supported by rfbr , grant 020216380 , grant 030217160 and grant intas 03 - 51 - 5112 . 9 knurenko s.p .
26th icrc .
salt lake sity .
song c. , cao z. , dawson b.r .
astroparticle phys .
14 ( 2000 ) 713 .
sleptsova v.r .
b ( proc . suppl . ) . 2003.v.122 .
k. shinozaki et al . //
28th icrc . | in the work the data of the yakutsk complex eas array and their comparison with calculation in the case of primary nuclei of different chemical elements are presented .
the calculation by qgsjet model have been used interpreting experimental data . |
particles , magnetic fields , and radiation all contribute to the propulsion of relativistic jets , though the relative contribution of each is still an open matter of debate . in certain situations ,
however , the mechanism responsible for launching the jet operates simultaneously with the release of a large amount of energy in the form of radiation , making it likely that photons dominate the bulk energetics .
this scenario occurs , for example , during the super - eddington phase of jetted tidal disruption events ( tdes ) , such as _ swift _ j1644 + 57 @xcite and _ swift _ j2058 + 05 @xcite .
radiation - dominated jets should also be present in the collapsar picture of long gamma - ray bursts ( grbs ; @xcite ) , where the energy released in the form of gamma - rays is ultimately derived from accretion onto a black hole , the associated accretion luminosity exceeding the eddington limit of the hole by more than ten orders of magnitude . in both of these cases ,
the propagation of the radiation - dominated jet is modulated by the presence of a radiation pressure - supported environment ; for super - eddington tdes , this environment is in the form of a highly inflated , quasi - spherical torus of fallback debris @xcite , while a `` cocoon '' of shocked jet material @xcite and the overlying stellar envelope itself @xcite serve as the confining medium for grbs . in a companion paper ( @xcite , hereafter paper i )
, we presented a model that describes the viscous interaction of a radiation - dominated , relativistic jet with its surrounding medium .
in that analysis we treated the jet and its surroundings as two separate fluids , interacting with one another via small anisotropies in the comoving radiation field that are explicitly accounted for in the equations of radiation hydrodynamics in the viscous limit .
this two - stream approximation , in agreement with the non - relativistic analysis of @xcite , demonstrates the manner in which the shear between the two fluids carves out a region of low density material within the boundary layer between them .
we also deduced the dependence of the boundary layer thickness on the asymptotic properties of the jet and the ambient medium .
these models also show , however , that the contact discontinuity separating the jet and its surroundings , necessary for maintaining their respective identities , results in the likely unphysical vanishing of the density of scatterers along that surface of separation .
the contact discontinuity also prevents the jet from entraining ambient material ; since the jet in the two - stream model we considered had an infinite amount of momentum , any entrainment or lack thereof is formally inconsequential to the evolution of the system .
however , realistic jets those with finite extent will almost certainly engulf more material as they expand into their surroundings ; because the total amount of momentum in the system , which is realistically finite , must be conserved , that entrainment will then result in an overall deceleration of the outflow that can not be captured in the two - stream model . in view of these unphysical properties of the two - stream treatment the vanishing of the mass density of scatterers along the contact discontinuity and the lack of entrainment we present here an alternative boundary layer scenario to describe the interaction of a relativistic , radiation - dominated jet with its surroundings .
this `` free - streaming jet '' model , which has a well - known counterpart in the non - relativistic , incompressible limit ( see chapter 10 , section 12 of @xcite ) , assumes that the jet is injected through a narrow opening into a static , homogenous medium , and that far enough from that opening the entire system can be modeled as a single fluid that is independent of the details at the injection point . by considering the jet and the ambient medium as one fluid
we obviate the need for a contact discontinuity , which we demonstrate allows the density of scatterers to remain finite throughout the flow and for the jet to entrain material .
in section 2 of this paper we present the equations of radiation hydrodynamics in the viscous limit .
section 3 uses those equations to analyze the free - streaming jet model , and we demonstrate the existence of approximate self - similar solutions in the limit that the interaction between the jet and the ambient medium is concentrated in a thin boundary layer . in section 4
we discuss the implications of our model and make comparisons to the two - stream scenario , and in section 5 we conclude and consider the application of the free - streaming jet model to super - eddington tdes , grbs , and other astronomical sources .
when changes in fluid quantities over the mean free path of a photon are small , radiation behaves like an effective viscosity and transfers momentum and energy between neighboring fluid elements .
the precise form of the viscosity can be determined by investigating the general relativistic boltzmann equation , which was recently done by @xcite for the case where thomson scattering dominates the interactions between the photons and scatterers in the fluid rest frame . in this limit
, they found that the relativistic equations of radiation hydrodynamics for a cold gas ( gas pressure much less than the gas rest mass density and radiation pressure ) are ( see their equation ( 49 ) ; see also paper i ) : @xmath0}+\frac{1}{3}g^{\mu\nu}\partial_{\mu}e ' \\
-\frac{8}{27}\nabla_{\mu}\bigg{[}\frac{e'}{\rho'\kappa}\pi^{\mu\sigma}\pi^{\nu\beta}\bigg{(}\nabla_{\sigma}u_{\beta}+\nabla_{\beta}u_{\sigma}+g_{\beta\sigma}\nabla_{\alpha}u^{\alpha}\bigg{)}\bigg { ] }
\\ -\frac{1}{3}\nabla_{\mu}\bigg{[}\frac{e'}{\rho'\kappa}\bigg{(}\pi^{\mu\sigma}u^{\nu}+\pi^{\nu\sigma}u^{\mu}\bigg{)}\bigg{(}4u^{\beta}\nabla_{\beta}u_{\sigma}+\partial_{\sigma}\ln{}e'\bigg{)}\bigg { ] } = 0 \label{radhydroco}.\end{gathered}\ ] ] here the speed of light has been set to one , greek indices range from 0 3 , @xmath1 is the fluid rest frame mass density of scatterers , @xmath2 is the fluid rest frame radiation energy density , @xmath3 is the scattering opacity ( in units of @xmath4 g@xmath5 ) , @xmath6 is the metric of the spacetime , @xmath7 is the covariant derivative , @xmath8 is the four - velocity of the flow , and @xmath9 is the projection tensor . the einstein summation convention has been adopted here , meaning that repeated upper and lower indices imply summation .
this equation also shows , in agreement with previous findings @xcite , that the coefficient of dynamic viscosity , @xmath10 , for an optically - thick , radiation - dominated gas is @xmath11 we will also write down the gas energy equation , obtained by contracting equation with the four - velocity @xmath12 , which gives ( see equation ( 50 ) of @xcite ) @xmath13 } \\
+ \frac{8}{27}\frac{e'}{\rho'\kappa}\bigg{(}\nabla_{\sigma}u_{\beta}+\nabla_{\beta}u_{\sigma}+g_{\sigma\beta}\nabla_{\alpha}u^{\alpha}\bigg{)}\pi^{\mu\sigma}\nabla_{\mu}u^{\beta } \\ + \frac{1}{3}\pi^{\mu\sigma}\nabla_{\mu}\bigg{[}\frac{e'}{\rho'\kappa}\bigg{(}4u^{\beta}\nabla_{\beta}u_{\sigma}+\partial_{\sigma}\ln{e'}\bigg{)}\bigg { ] } \\
+ \frac{1}{3}\frac{e'}{\rho'\kappa}\bigg{(}4u^{\beta}\nabla_{\beta}u_{\sigma}+\partial_{\sigma}\ln{e'}\bigg{)}\bigg{(}2u^{\mu}\nabla_{\mu}u^{\sigma}+u^{\sigma}\nabla_{\mu}u^{\mu}\bigg { ) } \label{gasenergyco}.\end{gathered}\ ] ] to close the system , we require that the normalization of the four - velocity be upheld and that particle flux be conserved : @xmath14 @xmath15 } = 0 \label{masscont}.\ ] ] equations and constitute six linearly independent equations for the six unknowns @xmath8 , @xmath2 , and @xmath1 .
in addition to the energy density of the radiation , @xmath2 , one can also calculate the number density of photons by requiring that the number flux , @xmath16 , be conserved .
one can show @xcite that the equation @xmath17 becomes , in the viscous limit , @xmath18 } \\
= \nabla_{\mu}\bigg{[}\frac{1}{\rho'\kappa}\bigg{(}\frac{10}{9}n'u^{\mu}\nabla_{\sigma}u^{\sigma}+n'u^{\alpha}\nabla_{\alpha}u^{\mu}+\frac{1}{3}\pi^{\mu\sigma}\nabla_{\sigma}n'\bigg{)}\bigg { ] } \label{fluxeq},\end{gathered}\ ] ] where @xmath19 is the rest - frame number density of photons .
once we solve the equations of radiation hydrodynamics for the four - velocity of the fluid and the mass density of scatterers , we can solve equation for the number flux of photons .
the goal of the next two sections is to apply equations and to the boundary layers established between fast - moving jets and their ambient media . for a more thorough discussion of the nature of the equations of radiation hydrodynamics in the viscous limit , we refer the reader to @xcite .
in this section we consider the problem where a narrow stream of material is continuously injected into a plane - parallel , static , ambient medium , known as the free - streaming jet problem .
if the reynolds number of the outflow is high , the transition between the stream of material and the external environment will be confined to a thin layer , permitting the use of a boundary layer approximation .
the basic setup is similar to that of the two - stream problem ( see paper i ) , with the motion of the 2-d , plane - parallel injected stream predominantly along the @xmath20-direction , the majority of the variation along @xmath21 , no variation or velocity in @xmath22 , and the point of injection at @xmath23 . now , however , there is no contact discontinuity between the stream and the ambient environment , meaning that the entire system is considered a single fluid .
we therefore have less freedom in prescribing the asymptotic characteristics of the jet and the environment ; however , this configuration permits mixing between the two media , allowing the outflow to entrain material , which almost certainly occurs in realistic jets .
we can reduce the complexity of equations and by assuming that the interaction between the jet and the ambient medium takes place over a thin boundary layer of thickness @xmath24 , thin in the sense that @xmath25 is a small number when @xmath26 is a typical length along the jet . in this case
, @xmath27 scales in an identical fashion to that derived for the two - stream boundary layer , which can be determined by comparing leading - order terms in the boundary layer thickness to the inviscid terms in the gas energy equation . setting the left - hand side of equation to @xmath28
, we find , as in paper i , @xmath29 here @xmath30 and @xmath31 are the jet velocity and lorentz factor , respectively , measured at some characteristic length along the jet axis @xmath32 , and @xmath1 is a characteristic density of scatterers throughout the outflow . due to the fact that the boundary layer thickness is the same , the boundary layer equations governing the outflow are identical to those found in paper i , which can be compactly written as @xmath33}+\frac{1}{3}g^{\mu\nu}\partial_{\mu}e ' = \frac{8}{27}\frac{\partial}{\partial{y}}\bigg{[}\frac{e'}{\rho'\kappa}\frac{\partial{u^{\nu}}}{\partial{y}}\bigg { ] } \label{bleqs},\ ] ] @xmath15 } = 0.\ ] ] in paper i we dealt with the @xmath34 and @xmath35 components of equation and its contraction with @xmath12 the gas energy equation .
for the free - streaming jet problem , however , it will be more convenient to deal with the @xmath34 component of equation , the gas energy equation , and the contraction of equation with the projection tensor @xmath36 , which , as we will see , is the relativistic , viscous counterpart of the bernoulli equation .
performing a few manipulations , we find that these equations become , respectively , @xmath37 @xmath38}+\frac{1}{3}e'\nabla_{\mu}u^{\mu } = \frac{8}{27}\frac{e'}{\rho'\kappa}\bigg{(}\frac{\partial{s}}{\partial{y}}\bigg{)}^2 \label{gasen1},\ ] ] @xmath39 } \label{zmom1},\ ] ] @xmath15 } = 0 \label{masscont1}.\ ] ] where @xmath40 the first of these shows that the radiation energy density , and consequently the pressure , is constant across the boundary layer . in the inviscid limit , equation can be transformed to give @xmath41 } = 0,\ ] ] which , as we mentioned , is the relativistic generalization of the bernoulli equation .
we will also assume that the ambient energy density is independent of @xmath20 , i.e. , that @xmath42 , with @xmath43 a constant .
because it will be convenient , we will change variables from @xmath21 to @xmath44 , where @xmath44 is given by @xmath45 which is related to the optical depth across the boundary layer as measured from the axis . in terms of this variable , equations and
become , respectively , @xmath46 @xmath47 where we have used the assumption that @xmath42 .
it should also be noted that the @xmath21-component of the convective derivative is now with respect to @xmath44 , not @xmath21 , i.e. , @xmath48 in the ensuing section we will seek self - similar solutions to these equations .
we can immediately solve the continuity equation by introducing the stream function @xmath49 via @xmath50 @xmath51 where the factor of @xmath3 ensures that @xmath49 remains dimensionless .
in paper i we showed that there exist self - similar solutions for @xmath49 , the velocity , the comoving density of scatterers and comoving density of photons in terms of the variable @xmath52 , where @xmath53 .
one difference between the two - stream problem and this type of outflow , however , is that we expect the jet to expand into the ambient medium , entraining material in the process .
therefore , as we look farther along the @xmath20-direction , the amount of inertia predominantly in the form of radiation for the systems that we are considering contained in the flow will increase .
owing to the conservation of momentum , the @xmath20-component of the velocity should thus be a decreasing function of @xmath20 . in light of this observation
, we will assume that the @xmath20-component of the four - velocity scales as @xmath54 where @xmath32 is a characteristic length scale in the @xmath20-direction , @xmath55 is a positive constant , for brevity we defined @xmath56 , and @xmath57 is a function of the self - similar variable @xmath58 ; @xmath59 is the characteristic boundary layer thickness in terms of @xmath44 , which , from equations and in the previous subsection , is given by @xmath60 , or @xmath61 we are most interested in the behavior of the properties of the outflow when the velocities are relativistic , as this is the limit that is most applicable to sources of astronomical interest , and in order for the self - similar nature of our solutions to be upheld , the dependence of our self - similar functions on the bulk properties of the outflow , such as the @xmath20-dependence of the jet lorentz factor along the axis , should be minimal .
a fully self - similar solution is likely impossible here , as the speed of light plays a role in setting a finite scale factor for our solutions and becomes problematic when we try to connect the ultra- and non - relativistic regions of the outflow .
we will show , however , that the non - self - similarity of our solutions only affects a small region of the outflow .
in paper i , we showed that the self - similarity of the comoving density of scatterers , @xmath1 , was approximately satisfied , i.e. , we could find solutions with @xmath62 .
for the case at hand , then , this assumption implies that @xmath63 is independent of @xmath64 .
this means , however , that the observer - frame density of scatterers , given by @xmath65 , can be made arbitrarily large ( as @xmath63 is independent of @xmath64 in the self - similar limit ) , which is a nonsensical result . motivated by this reasoning
, we conclude that the comoving density of scatterers is unlikely to be independent of @xmath64 . on the contrary
, a more reasonable approximation is that the _ observer_-frame density of scatterers varies self - similarly .
investigating equations and , we see that the @xmath1-dependent quantity that enters both is the combination @xmath66 .
our expectation that the observer - frame density varies self - similarly then prompts the assumption @xmath67 where @xmath68 , @xmath69 being the density of scatterers in the ambient medium .
( note that we could have simply let @xmath70 , but equation , which is merely a change of variables from the initial assignment @xmath70 , will allow equations and to be written in a more compact form . )
we can determine the value of @xmath55 , which controls how rapidly the flow decelerates due to entrainment , by integrating the momentum flux , @xmath71 for a radiation - dominated system , over the entire boundary layer , and requiring that the result be independent of @xmath20 .
this is equivalent to requiring that the total momentum contained in the outflow be conserved . integrating @xmath72 from @xmath73 to @xmath74 , using equations and , taking the ultrarelativistic limit and changing variables from @xmath21 to @xmath75 , we find @xmath76 since the integral in this equation is a constant that is greater than zero , we find that @xmath77 if the momentum flux is conserved
. with this value of @xmath55 , the @xmath20-component of the four - velocity scales as @xmath78 and the self - similar variable @xmath75 is given by @xmath79 we then find from equation that the stream function must satisfy @xmath80 equation then gives the @xmath21-component of the four - velocity : @xmath81 } \label{vysim},\ ] ] where , both here and in future equations , a subscripted @xmath75 on the functions @xmath57 and @xmath63 denotes differentiation with respect to @xmath75 and the number of subscripts indicates the number of derivatives , i.e. , @xmath82 , @xmath83 , etc . inserting equations , and into equations and ,
changing variables from @xmath44 to @xmath75 and performing a bit of algebra , we find that they become , respectively , @xmath84 @xmath85 now , the solutions we seek for @xmath57 and @xmath63 should depend only on @xmath75 , as otherwise our assumption of the self - similarity of those functions breaks down .
because of the complicated dependence of @xmath86 on @xmath87 , we see that the solutions will not be self - similar for arbitrary @xmath88 .
however , in the ultrarelativistic limit , for which @xmath89 , we see that @xmath90 , and it is apparent that the solutions are indeed self - similar , i.e. , the @xmath64 dependence of equations and drops out . as these equations are third order in @xmath57 and first order in @xmath63 , we require four boundary conditions to solve them numerically . since @xmath87 is given by equation and it is assumed that @xmath91 along the axis , our first boundary condition is given by @xmath92 . for the second , we note that the jet axis is the streamline along which the @xmath21-component of the velocity is zero . from equation ,
then , we find @xmath93 ( we will see that @xmath94 at @xmath95 ) . the third boundary condition is obtained by noting that the @xmath20-component of the velocity should approach zero as we proceed into the ambient medium , so that , from equation , we find @xmath96 . finally , the density of scatterers should approach that of the ambient medium far from the jet center .
equation then gives @xmath97 .
as we noted , the truly self - similar limit of equations and is obtained by setting @xmath98 , but we encounter an issue with this scaling when we consider the boundary conditions on our flow in the ambient medium .
in particular , we expect the outflow velocity to approach zero for @xmath99 ; when the velocity becomes subrelativistic , however , @xmath100 , when in actuality the @xmath101 limit of equation is @xmath102 .
thus , we see that taking the ultrarelativistic limit of equations and will not result in the solutions matching the correct boundary conditions far from the jet center . to correct this problem , we will allow the function @xmath86 to take on its full form and treat @xmath64 as a constant , which will break the self - similarity of our solutions .
however , since our boundary conditions do not depend on the value of @xmath64 , the functions @xmath57 and @xmath63 themselves should be largely independent of that parameter .
we therefore expect the assumption of self - similarity to be upheld in the ultrarelativistic ( @xmath103 ) and the non - relativistic ( @xmath99 ) limits of our solutions , with small deviations from self - similarity in the trans - relativistic regime of the outflow . , which is the normalized @xmath20-component of the four - velocity , with @xmath75 for @xmath104 ( blue , solid curve ) , @xmath105 ( purple , dot - dashed curve ) , and @xmath106 ( red , dashed curve ) .
as expected , the curves are all coincident when @xmath103 and @xmath99 , with a non - self - similar transition ( one that depends on @xmath64 ) in between those two limits.,width=336 ] , which is approximately the inverse of the lab - frame density , plotted with respect to @xmath75 for the same set of @xmath64 chosen in figure [ fig : fpsims ] . as was true for @xmath107 , @xmath63 is approximately self - similar close to and far from the jet , with the deviation from self - similarity in the trans - relativistic region being apparent but small.,width=336 ] to exemplify this point , figures [ fig : fpsims ] and [ fig : gsims ] show , respectively , the variation of @xmath107 the normalized @xmath20-component of the four - velocity and @xmath63 approximately the inverse of the observer - frame density in terms of the parameter @xmath75 for @xmath108 and 25 .
as we anticipated , the functions for different @xmath64 are indistinguishable for @xmath103 and @xmath99 , meaning that the self - similarity is nearly exact in those regions . in between those limits , where the flow is trans - relativistic , the deviations from self - similarity are apparent , though they remain small . keeping the full form of @xmath86 in equations and therefore preserves well the self - similarity of our solutions and provides a reasonable interpolation between the relativistic and non - relativistic regions of the flow .
the solution for @xmath63 approaches infinity as we near the axis , which can be seen by investigating the small-@xmath75 behavior of equation .
specifically , letting @xmath109 , @xmath110 , @xmath111 , and @xmath112 , equation can be written as @xmath113 where @xmath114 is the second derivative of @xmath57 evaluated at @xmath95 .
this equation can be integrated , and we find @xmath115 @xmath116 being a constant of integration .
since it is roughly proportional to @xmath117 , the value of @xmath63 can not be negative , meaning that @xmath116 must be greater than zero .
therefore , the asymptotic , @xmath103 behavior of @xmath63 is @xmath118 .
in addition to the mass density of scatterers and the velocity , we can also calculate the number density of photons , @xmath19 , throughout the boundary layer .
as we showed in paper i , the equation of photon number conservation becomes , to lowest order in the boundary layer thickness , @xmath119 performing a few manipulations , this equation becomes @xmath120 as was true for the density of scatterers , we expect the observer frame number density of photons to vary approximately self - similarly , which will be true if we let @xmath121 where @xmath122 is the number density of photons in the ambient medium and @xmath123 is a dimensionless function .
inserting this ansatz into equation gives @xmath124 for the boundary conditions on @xmath123 , we first require that the number density of photons approach that of the ambient medium in the @xmath99 limit .
equation then gives @xmath125 .
for the second condition , return to equation , integrate both sides from @xmath73 to @xmath74 , and require that the derivative of @xmath126 vanish in both of those limits .
doing so yields @xmath127 the right - hand side can be determined by returning to the continuity equation , integrating from @xmath73 to @xmath74 and performing a few manipulations to show that @xmath128 , where @xmath129 is the function @xmath57 evaluated at infinity . using the definition of @xmath19 in terms of @xmath123
, we find that equation becomes @xmath130 which serves as our second boundary condition on @xmath123 .
this integral states that the increase in the number flux of photons occurs at a rate provided by the influx of material at infinity .
the equations derived in this section were all written in terms of the variable @xmath75 , which is itself a function of @xmath63 via equations and . to write the solutions in terms of the physical parameter @xmath21
, we can return to equation , differentiate both sides with respect to @xmath21 , rearrange the resulting equation and integrate to yield @xmath131 where @xmath132 is a dummy variable of integration .
once we calculate the functions @xmath133 and @xmath134 , this relation can be integrated and solved numerically to yield @xmath135 .
this expression also shows that @xmath136 which we can use in equation to give @xmath137 equation also confirms that @xmath94 when @xmath95 , which we used in order to determine the boundary condition @xmath93 .
in this section we plot solutions for the outflow velocity , the density of scatterers and the density of photons for various values of @xmath138 and @xmath64 .
as was mentioned in the previous subsection , the physical self - similar variable against which we would like to plot our solutions is given by @xmath139 .
however , as is apparent from equation , the definition of @xmath139 depends on @xmath64
. therefore , if we are comparing , for example , the outflow velocity of two systems with differing @xmath64 , we must incorporate the @xmath64 dependence in @xmath139 so that the range of physical space that we consider for each solution is the same .
for this reason , in this section we will plot our solutions as functions of the variable @xmath140 -component of the four - velocity ( @xmath107 ) for @xmath141 and @xmath104 , 10 , and 25 ( the solid , blue curve , the dot - dashed , purple curve , and the dashed , red curve , respectively ) , which , for @xmath142 , correspond to @xmath143 and 25 .
we see that the width of the boundary layer is nearly unchanged as we alter the value of @xmath64 .
, width=336 ] . for all solutions
the number density of scatterers approaches zero as we near the center of the jet .
we see that the average number density of scatterers within the boundary layer is lower for larger lorentz factors.,width=336 ] used in figure [ fig : fpplotsgamjet ] .
the photon number density closely follows that of the scatterers.,width=336 ] -component of the three - velocity normalized by @xmath144 ( see equation ) for the same set of parameters used in figure [ fig : fpplotsgamjet ] . for positive @xmath145 ,
each solution is initially positive and reaches a relative maximum before approaching a negative constant , which shows that the flow expands outwards near the center of the jet and entrains ambient material far from the axis .
, width=336 ] figure [ fig : fpplotsgamjet ] shows the solution for @xmath107 , the normalized @xmath20-component of the four - velocity , for @xmath104 , 10 , and 25 . since @xmath146 , these values of @xmath64 scale approximately linearly with the lorentz factor until @xmath147 .
the outflow velocity is maximized at the origin and decays as we move farther into the ambient medium .
we see that the width of the boundary layer , loosely defined as the value of @xmath145 at which @xmath107 is some fraction of its central value , is nearly unchanged as we modify @xmath64 .
the average value of the normalized @xmath20-component of the velocity is also slightly larger for larger @xmath64 .
figure [ fig : rhoplotsgamjet ] demonstrates how the normalized , fluid - frame mass density of scatterers varies as we traverse the boundary layer for the same set of parameters chosen in figure [ fig : fpplotsgamjet ] . since @xmath63 approaches infinity as we near the origin , the mass density of scatterers , related to @xmath63 by equation , equals zero at the origin for all of the solutions , meaning that the center of the jet is evacuated of massive particles .
the average comoving density of scatterers across the boundary layer is also lower for larger @xmath64 , in accordance with equation . in figure
[ fig : hplotsgamjet ] we plot the normalized number density of photons for the same set of @xmath64 .
it is evident that the density of photons closely follows the density of scatterers throughout the boundary layer .
we see , however , that the density of photons stays above and below the density of scatterers as we move toward and away from the center of the jet , respectively .
the photon density also remains finite at the center of the jet .
the @xmath21-component of the three - velocity normalized by @xmath148 is illustrated in figure [ fig : vyplotsgamjet ] for the same set of @xmath64 used in figure [ fig : fpplotsgamjet ] .
for @xmath149 , each solution initially has a positive @xmath150 , which shows that the jet material expands away from the axis .
the @xmath21-component of the three - velocity then reaches a relative maximum , one which increases slightly for larger @xmath88 , before approaching a negative , constant value .
this behavior is then inverted for negative @xmath75 . because the transverse velocity approaches a negative constant for @xmath151 and a positive constant for @xmath152
, we see that the jet entrains material from the ambient medium . , which is the normalized @xmath20-component of the four - velocity , for @xmath105 and @xmath153 , 1 , and 10 , which correspond to the blue , solid curve , the purple , dot - dashed curve , and the red , dashed curve , respectively .
increasing the value of @xmath138 , we see , has little effect on the solution , while decreasing @xmath138 drastically widens the boundary layer.,width=336 ] chosen in figure [ fig : fpplotsmujet ] .
the mean value of the density decreases as @xmath138 decreases.,width=336 ] chosen in figure [ fig : fpplotsmujet ] .
this figure demonstrates , as we saw in figure [ fig : hplotsgamjet ] , that the density of photons tracks that of the scatterers.,width=336 ] -component of the three - velocity , normalized by @xmath148 , for the same set of @xmath138 chosen in figure [ fig : fpplotsmujet ] .
we see that the relative maximum increases for smaller @xmath138.,width=336 ] figures [ fig : fpplotsmujet ] [ fig : vyplotsmujet ] illustrate how our solutions depend on @xmath138 .
as is apparent , changing the value of @xmath138 does not drastically alter the qualitative aspects of the functions .
we do see , however , that decreasing @xmath138 from 1 to 0.1 results in a large increase in the boundary layer thickness ; on the contrary , changing @xmath138 from 1 to 10 results in only a slight narrowing of its thickness .
it is also evident that a smaller @xmath138 compared to 1 results in a lower average value of the density throughout the boundary layer and a larger peak in the transverse velocity @xmath150 .
figures [ fig : fpplotsgamjet ] [ fig : vyplotsgamjet ] demonstrate that the width of the boundary layer is nearly independent of @xmath64 , which is the result of the scaling of our boundary layer thickness @xmath27 .
specifically , note from equation that @xmath27 is given by @xmath154 . since our ansatz posited that @xmath155 , with @xmath63 a dimensionless function of order unity , and @xmath156 for @xmath157 , the boundary layer thickness is roughly independent of @xmath64 .
figures [ fig : fpplotsmujet ] [ fig : vyplotsmujet ] illustrate that a value of @xmath158 causes the boundary layer thickness to increase dramatically compared to @xmath141 , while setting @xmath159 causes only a slight narrowing of the width compared to @xmath141 .
this dependence is due to the fact that the viscous heating , which increases the specific entropy by decreasing the density of scatterers ( at fixed pressure ) , is most efficient when the flow is compressible , as is evident from the gas energy equation . the fluid only becomes compressible , however , when the flow is supersonic , and we can show @xcite that the sound speed of a radiation - dominated gas is @xmath160 when @xmath161 , the sound speed reduces to @xmath162 , and the location at which the flow becomes transsonic extends farther into the ambient medium , widening the boundary layer .
conversely , when @xmath163 , the sound speed approaches a constant @xmath164 , which results in only a slight narrowing of the boundary layer .
equation shows that the energy density of the radiation , and hence the pressure , is constant across the jet , which is a statement of the causal connectedness of the boundary layer . in order for this equation to remain valid , then , we require that the transverse sound crossing time over the boundary layer thickness @xmath24 be less than the time it takes the fluid to traverse the distance @xmath26 . since the transverse sound speed is given @xmath165 , where @xmath166 is given by equation , this requirement yields the inequality @xmath167 .
once this inequality is no longer satisfied , equation does not hold , and we must include more terms in equations to account for the gradients in the energy density of the radiation . as we noted in section 3 ,
the density approaches zero as we near the axis of the jet .
physically , this effect arises from the fact that , for @xmath168 , the lorentz factor grows unbounded and the boundary layer thickness goes to zero .
therefore , the center of the jet originates along a curve of infinite shear and , consequently , infinite entropy . since the specific entropy scales as @xmath169 and @xmath2 is constant across the boundary layer , we see that the density of scatterers must equal zero at the center of the jet . from figures [ fig : vyplotsgamjet ] and [ fig :
vyplotsmujet ] , we see that the @xmath21-component of the velocity initially causes the outflow to expand into its surroundings .
when @xmath170 becomes large , however , the directionality of @xmath150 reverses toward the jet .
the outflow therefore entrains material from the ambient medium not only by expanding in the transverse direction , but also by dragging material in from the environment .
interestingly , the @xmath21-component of the velocity approaches a non - zero , constant value as we move into the external medium .
this behavior arises from the fact that the jet is removing material from the system at a rate @xmath171 , showing that amount of mass excavated from the envelope increases as we move along @xmath20 .
therefore , in order to maintain a steady - state , we require a constant influx of material at infinity that can resupply the amount lost due to the jet .
this interpretation is substantiated by integrating the continuity equation from @xmath73 to @xmath172 , which shows that the @xmath21-component of the velocity at infinity scales as @xmath173 .
the factor of @xmath174 arises from the fact that the mass loss rate scales as @xmath175 ; therefore , if one could decrease the mass loss rate to one that was constant in @xmath20 , then the @xmath21-component of the velocity would vanish at infinity .
likewise , if one could create a scenario in which @xmath176 scaled as a _ negative _ power of @xmath20 , then the value of @xmath177 would maintain an efflux of material at infinity to keep the system from amassing inertia towards the center of the jet ( thus violating the steady - state assumption ) .
the free - streaming jet solutions analyzed in this paper are only valid when the lorentz factor is relativistic , i.e. , when @xmath178 .
when @xmath179 , @xmath180 , @xmath181 , and we can show that conservation of momentum along the @xmath20-axis results in the scaling @xmath182 , @xmath183 , which agrees with the results of the incompressible , non - relativistic theory ( in chapter 10 , section 12 of @xcite , see their discussion at the top of page 383 ) .
therefore , once @xmath184 , the outflow will undergo a non - self - similar transition from the solution presented here to its non - relativistic counterpart .
the free - streaming jet model has observational consequences .
for example , the decrease in the density along the jet axis means that observers looking down the barrel of the jet see farther into the outflow . because @xmath185 and @xmath186 , where @xmath187 is the lab - frame radiation energy density , this means that those observers see a higher energy density of photons .
also , those same observers , using the fact that the lab - frame radiation number density is @xmath188 , see an energy per photon of @xmath189 .
therefore , not only do the on - axis observers see a more lorentz - boosted spectrum because they can see deeper into the outflow , their spectrum is also hardened from the fact that @xmath19 is minimized along the axis of the jet ( where @xmath190 in the figures of section 3 ) .
paper i analyzed how the presence of a radiation pressure - supported envelope affected the propagation of a radiation - dominated , relativistic jet through the two - stream approximation .
this approximation treats the jet and its surroundings as semi - infinite , separate fluids , a contact discontinuity serving as the surface of separation between the two .
in contrast , the free - streaming jet solution presented here treats the entire system the jet and its surroundings as a single fluid .
there is thus no formal distinction between `` jet material '' and `` ambient material , '' meaning that no contact discontinuity exists in the system . by comparing figure 1 of paper i and
figure [ fig : fpplotsgamjet ] of the previous section , we see that the @xmath20-component of the four - velocity at a fixed @xmath20 behaves similarly between the two models .
namely , the solution starts at some `` jet '' velocity , which corresponds to @xmath172 in the two - stream model and to @xmath191 in the jet model , and smoothly transitions to a velocity of approximately zero over the extent of a few boundary layer thicknesses
. however , the full spatial dependence of the velocity , one that includes variation in the @xmath20-direction , differs drastically between the two models : while the two - stream jet maintains a constant @xmath88 along @xmath20 , we found here that the @xmath20-component of the four - velocity scales as @xmath192 , implying that the overall velocity of the jet slows as we look farther down @xmath20 .
this behavior arises from the fact that the free - streaming jet can entrain material , this entrainment causing an increase in the inertia contained in the outflow and a resultant decrease in its velocity .
the general behavior of the number densities of scatterers and photons between the models is also similar , which can be understood by comparing figures 2 and 3 of paper i to figures [ fig : rhoplotsgamjet ] and [ fig : hplotsgamjet ] of section 3 , respectively .
in particular , both densities decrease within the boundary layer separating the outflow and its surroundings , asymptotically approaching their jet values as @xmath193 in the two - stream model and as @xmath194 in the free - streaming jet model . likewise , each approaches its ambient value as @xmath195 in the two - stream model and as @xmath196 in the free - streaming jet model .
although the gross properties of both are similar , one striking difference arises , however , in the behavior of the scatterer density : in the two - stream model , the existence of the contact discontinuity causes the function @xmath63 , and hence the density @xmath1 , to vanish within the boundary layer ( it vanishes specifically at @xmath191 the location of the contact discontinuity ) .
conversely , @xmath1 remains non - zero throughout the boundary layer that connects the jet and its surroundings in the model presented here , and only as we near the center of the jet does the density of scatterers go to zero .
the presence of a contact discontinuity thus results in the likely non - physical vanishing of the massive particles within the boundary layer .
finally , the scaling of the boundary layer thickness itself differs between the two models . in the two - stream case
, we found that @xmath197 ( see equation ( 15 ) of paper i ) .
therefore , as one looks farther down the @xmath20-axis , the boundary layer that develops between the jet and the ambient environment extends into both media at a rate proportional to @xmath198 .
contrarily , equation shows that the free - streaming jet boundary layer expands into its surroundings as @xmath199 ( though this transitions to @xmath200 in the non - relativistic limit ; see the discussion at the top of page 383 in chapter 10 , section 12 of @xcite ) .
the free - streaming jet boundary layer thus expands more rapidly than does the two - stream solution .
employing the equations of radiation hydrodynamics in the viscous limit , which are applicable as long as changes in fluid quantities are small over the mean free path of a photon , we analyzed the dynamics of a relativistic , free - streaming jet under the boundary layer approximation .
this approximation , which should be upheld in jets with transverse optical depths substantially greater than one , states that variations in the properties of the outflow are confined to a thin layer of width @xmath27 , and it allowed us to transform the full set of equations into a set of greatly simplified boundary layer equations .
perhaps the biggest difference between the two - stream solutions , presented and analyzed in paper i , and the free - streaming jet solutions presented here is in the distinction between the jet and the ambient medium . in the former ,
the two are considered as distinct , interacting entities , which allows one to specify separately their asymptotic properties .
the latter approach , on the other hand , considers the whole configuration as a single fluid . because one has more freedom in specifying the properties of the outflow , the two - stream solution has the added benefit of being able to treat scenarios in which the properties of the jet and the ambient medium differ significantly .
however , maintaining the distinction between the two media necessitates the existence of a surface of contact between the two across which no fluid can flow , meaning that the jet can not entrain ambient material .
furthermore , as was demonstrated in paper i , this boundary condition results in the density formally vanishing at the interface , which is likely non - physical . by treating the outflow and the environment as one fluid , we demonstrated that entrainment does occur in the radiation - viscous , free - streaming jet solution , which causes the @xmath20-component of the four - velocity of the jet to slow as @xmath201 ( this power - law , however , may differ if the adopted symmetry is azimuthal as opposed to planar ; see below ) .
these solutions also show that the comoving densities of scatterers and photons decrease dramatically within the boundary layer , with the density of scatterers approaching zero as one nears the center of the jet . therefore , because observers that look `` down the barrel of the jet '' can see farther into the outflow , they see a more lorentz - boosted energy density than those that view the outflow off - axis .
they also observe a higher energy per photon , given by @xmath202 , both because @xmath19 is minimized along the axis and because @xmath203 , the observed lorentz factor , is larger .
such features are in qualitative agreement with the event _ swift _ j1644 + 57 , where such an observer orientation is invoked to explain the x - ray emission @xcite . in addition to super - eddington tdes and long grbs ( as well as the relatively new class of `` ultra - long '' grbs ; @xcite ) the two applications considered in the introduction the free - streaming jet model could be applied to other sources .
as mentioned in paper i , this model may also be relevant to microquasars @xcite ( particularly those that fall in the class of ulx s ; @xcite ) such as the object ss 433 ( @xcite ; see @xcite for an application of the two - stream model to this source ) . additionally , a jetted quasi - star
a protogalactic gas cloud supported by a supercritically accreting black hole @xcite provides another situation in which a radiation - dominated jet propagates alongside a radiation pressure - supported envelope .
the free - streaming jet solution presented here is limited to describing plane - parallel , two - dimensional systems .
one consequence of this assumption is that the entrainment of material , which slows down the jet as @xmath185 , happens effectively in one dimension .
if , on the other hand , one imposed azimuthal symmetry and described the system in terms of spherical coordinates @xmath204 , which is likely more relevant for realistic jetted systems , the entrainment would occur in two dimensions .
other conditions being equal , the slowing of the jet along the axis would then be more pronounced , i.e. , the lorentz factor would scale as @xmath205 with @xmath206 .
the other main assumption of the free - streaming jet model presented here is that the energy density of the ambient medium is independent of @xmath20 , i.e. , @xmath42 .
if a pressure gradient were present , this force would tend to accelerate ( or , in principle , decelerate ) the jet material , offsetting the power - law scaling @xmath185 .
in fact , one can imagine that if the pressure gradient were strong enough , it could indeed reverse the overall slowing of the jet and cause the outflow to accelerate .
if the energy density scaled as @xmath207 , which is relevant for super - eddington tdes as long as the constant @xmath208 satisfies @xmath209 @xcite , this situation could be actualized near the launch point of the jet . in an ensuing paper ,
we plan to compare more quantitatively the predictions made by the free - streaming jet model presented here and the observations of _ swift _ j1644 + 57 .
we will also extend our analysis to incorporate a spherical geometry as azimuthal symmetry is almost certainly more relevant for this and other systems than the planar symmetry adopted here as well as radially - dependent ambient pressure and density profiles .
arav n. , begelman m.c .
, 1992 , apj , 401 , 125 arav n. , begelman m.c .
, 1993 , apj , 413 , 700 begelman m.c . , king a.r . , pringle j.e . ,
2006a , mnras , 370 , 399 begelman m.c . , rossi e.m . , armitage p.j .
, 2008 , mnras , 387 , 1649 begelman m.c . , volonteri m. , rees m.j .
, 2006b , mnras , 370 , 289 blandford r.d . , jaroszyski m. , kumar s. , 1985 , mnras , 215 , 667 burrows d.n .
, kennea j.a . ,
ghisellini g. , et al . , 2011 , nature , 476 , 421 cenko s.b .
, krimm h.a .
, horesh a. , et al . , 2012 ,
apj , 753 , 77 coughlin e.r . ,
begelman m.c .
, 2014a , apj , 781 , 82 coughlin e.r . , begelman m.c . , 2014b , apj , 797 , 103 coughlin e.r . , begelman m.c . , 2015 , apj , accepted ( paper i ) czerny b. , janiuk a. , sikora m. , et al . , apj , 2012 , 755 , l15 fabrika s. , 2004 , asprv , 12 , 1 fender r.p . , belloni t.m . , gallo e. , 2004 , mnras , 355 , 1105 king a.r . , davies m.b . , ward m.j . , et al . , 2001 , apj , 552 , l109 kundu p.k . , cohen i.m . , 2008 , _ fluid mechanics _ , academic press levan a.j . , tanvir n.r
, fruchter a.s .
, et al . , 2014 , apj , 781 , 13 loeb a. , laor a. , 1992 , apj , 384 , 115 lpez - cmara d. , morsony b.j . ,
begelman m.c . , lazzati d. , 2013 , apj , 767 , 19 macfadyen a.i . , woosley s.e . , 1999 ,
apj , 524 , 262 matzner c.d . , 2003 , mnras , 345 , 575 mszros p. , rees m.j . , 1993 , apj , 405 , 278 morsony b.j . , lazzati d. , begelman m.c . , 2007 , apj , 665 , 569 piran t. , 2004 , rvmp , 76 , 1143 rees m.j . , mszros p. , 1994
, apj , 430 , l93 woosley s.e . , 1993 ,
apj , 405 , 273 woosley s.e .
, heger a. , 2006 , apj , 637 , 914 zauderer b.a . ,
berger e. , soderberg a.m. , et al . , 2011 ,
nature , 476 , 425 | we analyze the interaction of a radiation - dominated jet and its surroundings using the equations of radiation hydrodynamics in the viscous limit . in a previous paper we considered the two - stream scenario , which treats the jet and its surroundings as distinct media interacting through radiation viscous forces . here
we present an alternative boundary layer model , known as the free - streaming jet model where a narrow stream of fluid is injected into a static medium and
present solutions where the flow is ultrarelativistic and the boundary layer is dominated by radiation .
it is shown that these jets entrain material from their surroundings and that their cores have a lower density of scatterers and a harder spectrum of photons , leading to observational consequences for lines of sight that look `` down the barrel of the jet . ''
these jetted outflow models may be applicable to the jets produced during long gamma - ray bursts and super - eddington phases of tidal disruption events . |
the high mass x - ray binary ( hmxb ) was discovered in outburst by the swift burst alert telescope ( bat ) on 18 december 2005 @xcite .
it was soon recognized to be an x - ray pulsar with a @xmath415s spin period @xcite and its companion was classified as a b0ve star at a distance of @xmath410kpc @xcite . shortly after the discovery , the rossi x - ray timing explorer ( _ rxte _ ) began monitoring the source and continued to do so for nearly five years .
the categorization of swiftj1626.6@xmath05156 as a be / x - ray ( bex ) binary places it in a group that comprises most of the hmxb population , so we will describe it in terms associated with that group of objects .
its 220kev pca light curve is shown in figure [ fig1 ] , spanning the beginning of _ rxte _ s monitoring in 2006 january through 2009 january , when the source had returned to quiescence . 5156 , taken from heasarc s _ rxte _ mission - long data archive ( ftp://legacy.gsfc.nasa.gov/xte/data/archive/missionlongdata ) .
it shows the background subtracted 220kev count rate as recorded in ` standard2 ` mode ( 16s time resolution ) , averaged over each observation and normalized to a single pcu .
the three vertical lines indicate when the longest observations were taken in january 2006 .
lo1 is the blue dashed line , lo2 the green dot - dashed line , and lo3 the red dotted line .
the shaded regions represent the data sections , with ds1 in yellow , ds2 in orange , and ds3 in teal .
[ fig1 ] ] cccccc lo1 & 91081 - 13 - 01 & 2006 jan 1 & 3736.05 & 26.4 & 8.4 + lo2 & 91081 - 13 - 02 & 2006 jan 6 & 3741.15 & 23.6 & 7.4 + ds1 & 91082 - 01 - 01 91082 - 01 - 27 & 2006
jan 1230 & 3747.733765.10 & 23.3 & 7.1 + lo3 & 91081 - 13 - 04 & 2006 jan 31 & 3766.59 & 27.6 & 7.7 + ds2 & 91082 - 01 - 32 91082 - 01 - 93 & 2006 feb 1mar 13 & 3767.643807.52 & 51.2 &
+ ds3 & 92412 - 01 - 27 93402 - 01 - 48 & 2006 jun 262008 oct 10 & 3912.214750.82 & 78.8 &
swiftj1626.6@xmath05156 was discovered during a type ii outburst , which is thought to occur when the be star s circumstellar disk expands and temporarily engulfs the neutron star , leading to enhanced accretion and therefore a large increase in x - ray emission ( * ? ? ?
* and references therein ) .
@xcite describe the exponential decay of the outburst and the flaring behavior that occurred in the weeks immediately following the discovery .
additional flaring occurred following jd 2453800 ( 2006 march 5 ) , after which the source transitioned into an epoch of long - term , quasi - periodic x - ray oscillations @xcite .
a lomb - scargle periodogram @xcite of the light curve first revealed an oscillation period from @xmath4jd2454000jd2454350 ( 2006 september2007 september ) of @xmath447days and from @xmath4jd24543502454790 ( 2007 september2008 november ) of @xmath43/2 of this , or @xmath472.5days @xcite .
recent work by @xcite found the long - term variability timescale to range from @xmath445 to 95days .
additionally , from pulse timing analysis , they find the binary orbit to be nearly circular ( eccentricity 0.08 ) with a period of 132.9days , meaning the oscillations occur on timescales of @xmath41/2 and @xmath42/3 of the orbital period . the oscillations therefore can not be explained by bursts of accretion during periastron passages . on jd2454791 ( 2008
november 11 ) the oscillatory behavior ceased and the 220kev pca count rate from swiftj1626.6@xmath05156 dropped to @xmath4@xmath6 .
many bex and other hmxb systems harbor neutron stars with very strong magnetic fields , typical strengths being @xmath7 g , and some of these systems exhibit cyclotron resonance scattering features ( crsfs or cyclotron lines ) .
observing cyclotron lines in the spectra of these systems is the only direct way to measure the magnetic field strengths of neutron stars . at the time of writing
there are sixteen neutron star x - ray binaries with confirmed crsfs ( * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* and references therein ) , eight of which are transients ( swiftj1626.6@xmath05156 being the ninth ) . of the transients four systems
are bex binaries ( swiftj1626.6@xmath05156 being the fifth ) and four are oex binaries .
the persistent sources are hmxbs with an o- or b - star accompanying the neutron star ( with exception of her x-1 and 4u1626@xmath067 , where the companion is of lower mass ) .
the fundamental cyclotron line energy is related to the magnetic field strength of the neutron star @xmath8kev where @xmath9 and @xmath10 is the surface gravitational redshift . assuming a reasonable value of @xmath11
, corresponding to a typical neutron star mass of @xmath12 , allows the direct calculation of @xmath13 in the region local to the magnetic pole where the cyclotron lines are produced . in this paper
we discuss the discovery of a crsf in the spectrum of swiftj1626.6@xmath05156 . in
[ sec : observations ] we describe our data reduction and analysis techniques .
[ sec : spectra ] presents results from modeling the broad band x - ray spectra , including crsfs and other spectral properties of the source .
we summarize our findings and compare swiftj1626.6@xmath05156 with other cyclotron line sources in [ sec : discussion ] and conclude in [ sec : conclusion ] .
for this analysis , we considered data from the proportional counter array ( pca ; * ? ? ? * ) and the high energy x - ray timing experiment ( hexte ; * ? ? ? * ) onboard _ rxte _ that were taken between december 2005 and april 2009 . from hexte , we used spectra from only the b cluster , since the a cluster had stopped rocking at the time of these observations , i.e. , no background measurements for cluster a were available anymore .
we extracted spectra in ` standard2 ` mode from the top layer of each pca proportional counter unit ( pcu ) separately .
we then combined all spectra from a given observation into one , following the recipe in the _ rxte _ cookbook and excluding data taken within 10min of the south atlantic anomaly . for
a more detailed description of the data reduction and screening procedure see @xcite .
we considered the pca energy range 422kev and the hexte range 1860kev when modeling the phase averaged spectra .
three long observations with pca exposure time @xmath120ks are available in the _ rxte _ archive , along with hundreds of shorter monitoring observations .
we combined the individual pointings of each observation as described in the _ rxte _ cookbook .
the three `` long observations '' , which we refer to as lo1 , lo2 , and lo3 , were each analyzed separately , while the data between and beyond these observations were combined into `` data sections '' , referred to as ds1 , ds2 , and ds3 .
data prior to lo1 were excluded in order to avoid spectral contamination from flaring events ( for a description of the flares , see * ? ? ? * ) .
for ds3 we combined only observations from the oscillating phase that had an average pca 220kev count rate of 10@xmath14 or higher .
the observation identification numbers and other information are given in table [ tab1 ] .
the locations in time of lo13 are shown by the colored lines and of ds13 by the shaded regions in figure [ fig1 ] .
the hexte data have low count rates , so we rebinned the counts at high energies using the ` ftool ` ` grppha ` .
for lo1 and lo2 we rebinned by three energy channels above 50kev and for lo3 by four channels above 48kev .
we did not rebin any counts for ds1 .
the hexte data of ds2 and ds3 were so noisy we decided not to use them as they might skew our spectral fits .
we used ` xspec12 ` models @xcite for all spectral fitting .
we modeled the combined pca and hexte ( where present ) phase averaged spectra of each lo and ds with a power law modified by a smooth exponential rollover ( ` cutoffpl ` in ` xspec ` ) including photoelectric absorption ( ` phabs ` in ` xspec ` ) and a gaussian 6.4kev fe k@xmath15 emission line ( ` gauss ` in ` xspec ` ) : @xmath16 } \right \ } \label{eq1}\end{aligned}\ ] ] where @xmath17 and @xmath18 are the absorption model components and are respectively the hydrogen column density per h atom for material of cosmic abundance ( @xmath19 ) and the bound - free photoelectric absorption cross section ( @xmath20 ) .
we used the abundances of @xcite and the cross sections of @xcite .
the first term in eq . [ eq1 ] is the cutoff power law , where @xmath21 is the power law normalization ( @xmath22 at 1kev ) , @xmath23 is the power law photon index , and @xmath24 is the e - folding energy of the exponential rollover .
the second term is the gaussian emission line , in which @xmath25 is its normalization ( @xmath26 ) , @xmath27 its energy , and @xmath28 its width .
this model was applied concurrently to the pca and hexte data , differing only by a constant flux cross - calibration factor where the constant was fixed at 1 for the pca and fit to @xmath40.8 for hexte . for each spectral
fit we also fit the background strength adjusting it on a level of @xmath40.1% using the command ` recornorm ` in ` xspec ` .
460kev spectrum and best - fit model for observation lo3 are shown in the main panel .
pca data are shown in red , hexte data in blue . _
a ) _ the @xmath29 residuals left from fitting the spectrum with a cutoff power law .
the spectrum peaks near @xmath30kev and dips near @xmath31kev . _
b ) _ residuals from adding a gaussian emission line at @xmath32kev to the cutoff power law , completing our continuum plus fe line fit from equation [ eq1 ] .
the depression at @xmath33kev remains .
_ c ) _ including a gaussian optical depth absorption line near @xmath33kev removes the absorption - like feature and greatly improves the model , giving our best fit ( equation [ eq2 ] ) .
there may be a second feature near @xmath34kev . _
d ) _ a second gaussian optical depth absorption feature included near @xmath34kev marginally improves the fit .
the remaining residual near @xmath35kev is most likely caused by an instrumental effect ( [ sec : iodine]).[fig2 ] ] 10kev in lo1 ( blue ) , lo2 ( green ) , and lo3 ( red ) .
a decrease of the central energy between lo1lo2 ( close in time ) and lo3 is apparent , as well as an increase in line depth .
[ fig3 ] ] residuals for a continuum fit using the model from eq .
[ eq1 ] without and with the iron emission line for lo3 are shown in figs [ fig2]a and [ fig2]b , respectively .
a depression in the residuals at @xmath410kev is apparent . in order to account for this feature , we modified the continuum plus fe line model with a multiplicative absorption - like cyclotron line with a gaussian optical depth profile ( ` gabs ` in ` xspec ` ) : @xmath36 \right \ } \label{eq2}\ ] ] here @xmath37 , @xmath38 , and @xmath39 are respectively the optical depth , line width , and line energy of the cyclotron line .
the residuals of a fit to lo3 using the full model @xmath40 of eq .
[ eq2 ] , i.e. , including the cyclotron line , are displayed in figure [ fig2]c . in the following sections we describe the modeling and interpretation of this line as a crsf including several tests that led to the adoption of the model described above as the best - fit model for the phase averaged spectra ( ` constant`@xmath41`phabs`@xmath41(`cutoffpl ` + ` gauss`)@xmath41`gabs ` in ` xspec ` ) .
ccccccccccccccc lo1 & @xmath42 & @xmath43 & @xmath44 & @xmath45 & @xmath46 & @xmath47 & @xmath48 & @xmath49 & @xmath50 & @xmath51 & @xmath52 & @xmath53 + [ 1 mm ] lo2 & @xmath54 & @xmath55 & @xmath56 & @xmath57 & @xmath58 & @xmath59 & @xmath60 & @xmath61 & @xmath62 & @xmath63 & @xmath64 & @xmath65 + [ 1 mm ] ds1 & @xmath66 & @xmath67 & @xmath68 & @xmath69 & @xmath70 & @xmath71 & @xmath72 & @xmath73 & @xmath74 & @xmath75 & @xmath76 & @xmath77 + [ 1 mm ] lo3 & @xmath78 & @xmath79 & @xmath80 & @xmath81 & @xmath82 & @xmath83 & @xmath84 & @xmath85 & @xmath86 & @xmath87 & @xmath88 & @xmath89 + [ 1 mm ] ds2 & @xmath90 & @xmath91 & @xmath92 & @xmath93 & @xmath94 & @xmath95 & @xmath96 & @xmath97 & @xmath98 & @xmath99 & @xmath100 & @xmath101 + [ 1 mm ] ds3@xmath102 & & @xmath103 & @xmath104 & @xmath105 & @xmath106 & @xmath95 & @xmath107 & @xmath108 & @xmath109 & @xmath99 & @xmath110 & @xmath111 we used lo3 , the longest observation of swiftj1626.6@xmath05156 , to illustrate the results obtained with the best - fit model described in the previous section .
this observation was taken on 2006 january 31 , during the outburst decay ( figure [ fig1 ] ) .
the first pca pointing is @xmath421ks long and is followed immediately by a shorter pointing of @xmath46ks .
we combined them to form a single spectrum with pca integration time @xmath427ks .
the corresponding hexte observation has @xmath47.6ks of live time .
the best - fit parameters for the lo3 spectrum using the model of eq .
[ eq2 ] , along with the iron line equivalent width and other characteristics , can be found in the fourth row of table [ tab2 ] .
the phase averaged spectrum of lo3 and best - fit model are shown in the main panel of figure [ fig2 ] . inspecting the residuals obtained after fitting the continuum model and iron line ( equation [ eq1 ] ) shown in figure
[ fig2]b more closely reveals the absorption - like feature near 10kev and/or an emission - like feature near 14kev .
here we treated the feature as absorption - like only ( equation [ eq2 ] ) and obtained a very good fit . see [ sec : crab][sec : modified ] for testing possible alternative descriptions .
fitting the feature with a gaussian optical depth absorption line flattened the residuals in figure [ fig2]c , yielding @xmath112 for 56 degrees of freedom .
the common @xmath29 test available in ` xspec ` could not be applied in this case , as it is not applicable to multiplicative model components @xcite .
we instead used the more appropriate ratio of variances @xmath113-test @xcite to calculate the probability that including this absorption - like feature in the model improves the fit by chance .
as discussed by @xcite , this @xmath113-test is not available in ` xspec ` ; throughout this paper , we use the idl routine ` mpftest ` to calculate the probability of chance improvement ( pci ) from one model to another .
if one model is not a significant improvement over another , then the pci will be large .
the pci of the model that includes the gaussian - depth absorption feature ( figure [ fig2]c ) compared to that without ( figure [ fig2]b ) is @xmath114 .
the improvement to the fit due to use of the absorption - like model component is therefore very statistically significant .
the best - fit parameters of this model are given in table [ tab2 ] .
the crsf is detected in the earlier two long outburst observations , lo1 and lo2 , as well .
figure [ fig3 ] shows the line profiles for lo1lo3 .
we also detected a strong iron line in each of the spectra .
see table [ tab2 ] for the best - fit results .
one question we may ask is whether or not the cyclotron line persists throughout and beyond the primary outburst decay . to answer this we combined sections of data into single spectra to search for the line .
the data selection is described in 2 and summarized in table [ tab1 ] .
we found that both the iron line and the crsf were present throughout the active phase of swiftj1626.6@xmath05156 , as the features are detected in each of the data sections ds1ds3 as well as in the long observations .
interestingly , we found that the spectrum from the oscillatory stage of the light curve ( ds3 ) is best fit by the addition of a gaussian emission line at @xmath415kev , a component that was not necessary in any of the spectra taken during the outburst decay ( see [ sec : modified ] ) but that is commonly used to fit hmxb spectra @xcite .
all fit parameters are given in table [ tab2 ] .
the values of @xmath39 are suggestive of an evolution of the cyclotron line energy with luminosity , which is discussed in
[ sec : luminosity ] . note that the parameters might be affected by additional uncertainties due to the range of fluxes covered for ds1ds3 and missing hexte data for ds2ds3 , though .
`` bumps '' or `` wiggles '' in the spectra of accreting x - ray pulsars , especially near 10kev , are common and have been discussed at some length by @xcite .
they find residuals in the data / model ratio at the @xmath40.8% level in all the accreting systems they examine .
these residuals could easily be interpreted as cyclotron features .
there have been marginal detections of absorption - like features , possibly cyclotron lines , near 10kev in other sources .
one example is the recent work on by @xcite .
these authors see what may be an absorption - like feature at 10kev , but the count rate is too low to claim a significant detection .
one must thus take good care to ensure that the feature we see is truly a cyclotron line and not due to calibration uncertainties , additional emission ( e.g. , from the galactic ridge ) , or incorrect modeling of the continuum .
we note that the residuals we see at @xmath410kev are on the 310% level ( figure [ fig3 ] ) , i.e. , comparatively strong .
nevertheless we address these potential sources of uncertainties in [ sec : crab ] , [ sec : ridge ] , and
[ sec : cont][sec : modified ] , respectively .
we also note that @xcite state that they do not detect a cyclotron line in the _ rxte _ outburst data .
however , these authors include a 9kev edge in their model apparently masking the crsf .
this model choice misses the typical crsf properties of the observed residual that we present in this paper ( e.g. , its luminosity or pulse phase dependence ) .
5156 ( observation lo3 ) to crab counts according to the procedure described by @xcite .
the depression at @xmath410kev is present in this model- and calibration - independent representation of the spectrum of swiftj1626@xmath05156 as well .
[ fig4 ] ] one test we can perform in order to validate the existence of the cyclotron line in the phase averaged spectrum is to compare the spectrum of swiftj1626.6@xmath05156 with that of a non - cyclotron line source like the crab pulsar . to be as consistent as possible in our comparison
, we used swiftj1626.6@xmath05156 and crab data only from pcu 2 , which is the best calibrated of the five pcus @xcite and which is most often turned on during observations .
the crab data included both the pulsar and the nebula , and were taken between november 2005 and march 2006 , the same time period during which the swiftj1626.6@xmath05156 observations lo13 were taken .
we first searched for an absorption - like feature near 10kev in the crab spectral residuals .
the crab has no crsfs , so if such a feature were seen in the crab spectrum , we could assume that the swiftj1626.6@xmath05156 feature was instrumental in origin .
we modeled the crab spectrum with an absorbed power law , freezing @xmath17 at @xmath115 , the value determined for the crab by @xcite .
the power law parameters were left free , and we found that @xmath116 with a normalization of 10.343@xmath22 at 1kev gave the best fit .
the data / model ratio is @xmath40.99 near 10kev , which is consistent with expected calibration uncertainties @xcite .
no absorption - like feature is found at 10kev in the crab spectrum ( and no rollover ) .
we repeated the procedure with pcu 2 data from swiftj1626.6@xmath05156 during lo3 , using a cutoff power law and fixing @xmath17 and @xmath23 to the values in table [ tab2 ] . in this case
the ratio near 10kev is @xmath40.91 , a much larger residual than consistent with calibration errors .
cccccc ` cutoffpl ` & @xmath82 & @xmath85 & @xmath117 & @xmath118 & + [ 1 mm ] ` power`@xmath41`highecut ` & @xmath82 & @xmath119 & @xmath120 & @xmath121 & @xmath122 + [ 1 mm ] ` power`@xmath41`fdcut ` & @xmath82 & @xmath123 & @xmath124 & @xmath121 & @xmath125 + [ 1 mm ] ` npex ` & @xmath126 & @xmath127 & @xmath128 & @xmath129 & @xmath130 we next took a model and calibration independent approach by comparing the raw counts of swiftj1626.6@xmath05156 with those of the crab in each spectral energy bin following the procedure of @xcite .
we first subtracted the background counts from each raw spectrum , making sure to account for the difference in exposure time between the two sources .
we then divided the swiftj1626.6@xmath05156 count rates by those of the crab , resulting in the ratio shown in figure [ fig4 ] . at @xmath1316kev
the j1626/crab ratio reflects the steeper slope of the crab and at @xmath13214kev the missing crab rollover . between these energies
an enhancement at 6.4kev due to the iron line in swiftj1626.6@xmath05156 and a clear depression at @xmath410kev , rather than a smooth transition between the iron line and the power law tail , are apparent .
the same kind of crab ratio is shown , e.g. , for the @xmath455kev crsf of in the upper panel of figure 3 of @xcite .
this plot shows similar cyclotron line and rollover features as the j1626/crab ratio ( it differs at lower energies due to the overall different energy range ) .
independently of the response function or spectral model the j1626/crab ratio therefore supports that the absorption - like feature does not result from a calibration error .
we also investigated the possibility that cyclotron features in the spectrum arise from the galactic ridge emission by analyzing data from when the source was quiescent .
while swiftj1626.6@xmath02126 remained fairly bright and highly variable for about 1000days after the initial outburst , it then quickly faded to quiescence ( figure [ fig1 ] ) and stayed reliably low for the last two years of the _ rxte _ mission . during this time
the source continued to be monitored regularly by the pca .
we combined these late observations to obtain a spectrum with a total of 42ks exposure , and compared this with our outburst observations , in order to put limits on the contribution of the flux during the outburst observations from the underlying galactic ridge emission . in the 10kev region surrounding the fundamental cyclotron line
, the ridge emission has a flux of less than 1% of the count rate of swift1626.6@xmath05156 during our observations ( lo3 ) . in the energy range around the first harmonic ( @xmath418kev ,
see [ sec : harmonic ] and [ sec : phaseresolved ] ) no emission from the ridge is detected .
therefore galactic ridge emission can not explain the deviations that we measure . in order to test the continuum model dependence of our results we in turn replaced the cutoff power law model by three other standard empirical pulsar continuum models and repeated the fit for lo3 .
the first of these was the power law with a high energy cutoff ( ` power`@xmath41`highecut ` in ` xspec ` where , in contrast to the smooth cutoff power law , the rollover in the spectrum only starts at a certain energy , @xmath133 @xcite .
while this model is often used , it can create an artificial line - like residual near @xmath133 due to the discontinuity at that energy ( * ? ? ? * and references therein ) .
next we fit the spectrum with the fermi - dirac cutoff from @xcite ( ` power`@xmath41`fdcut ` in ` xspec ` ) .
this model has a smooth , continuous rollover described by @xmath134 and @xmath24 , which are , however , not directly comparable to the values obtained with ` cutoffpl ` or ` power`@xmath41`highecut ` due to differences in the continua .
the final model we used is the negative positive exponential model from @xcite ( ` npex ` in ` xspec ` ) , which consists of two power laws with a smooth exponential cutoff described by @xmath24 .
each of these models was fit to lo3 modified by absorption and with the added 6.4kev gaussian emission line and cyclotron line .
the fit values of the gaussian emission line and the crsf energy found for each model are given in table [ tab : new ] .
these alternative continuum models were found to produce fits of similar quality when compared with each other and with the ` cutoffpl ` model .
they also produced identical results , within errors , for all comparable parameters , most notably those of the 10kev crsf .
the @xmath29 test can not be used to compare different continuum models @xcite . to quantify whether or not the ` cutoffpl ` model , which has the lowest @xmath135 of the four continuum models tested ,
actually improves the continuum fit when compared to the alternative models , we calculated the pci ( as in [ sec : best - fit ] ) of the ` cutoffpl ` model . @xcite
discuss the appropriateness of this @xmath113-test for comparison between different models , and the statistic is applied in this way by @xcite .
table [ tab : new ] gives the reduced @xmath135 , number of degrees of freedom , and pci ( compared to ` cutoffpl ` ) for each alternative model .
we find that the pci is between @xmath42550% , meaning that the ` cutoffpl ` improves the continuum fit marginally at best . because none of the canonical models clearly gave a best fit to the continuum , nor adjusted the energies of the 6.4kev line or the 10kev feature , and since ` cutoffpl ` has the smallest number of parameters , we defaulted to this model , our original choice , as a result of this test .
the ` cutoffpl ` model is also well suited for comparisons with many earlier results for similar sources @xcite .
also , in order to further check that the @xmath31kev feature is robustly described as a typical crsf , we compared the gaussian optical depth absorption model with a lorentzian profile model ( ` cyclabs ` in ` xspec ` ) @xcite . taking the known systematic shift into account we found similar centroid energies for the absorption - like feature using either model .
we chose to use ` gabs ` in our analysis because its parametrization provides an easier to interpret and compare characteristic energy parameter ( in contrast to ` gabs ` the energy parameter of ` cyclabs ` does not directly reflect the energy where the absorption - like feature in the spectrum is deepest ) .
many accreting pulsar spectra are either not adequately or at least not uniquely described by the standard empirical continuum models above , especially in the 1020kev range @xcite .
incorrect modeling of the exact shape of the rollover can create residuals in addition to crsf features or , in the worst case , mimic crsfs .
there are many examples for the application of continuum models with slightly modified cutoff characterization .
@xcite noted the likely presence of a two step change of the spectral slope for ( as well as a potential cyclotron line residual near 36kev ) , the first change occurring at 1020kev and the other one at higher energies , however , they were able to model this behavior taking advantage of the discontinuity and the two characteristic energies of the ` highecut ` model mentioned above .
other authors used an additional polynomial component in order to describe a more complex rollover shape , i.e. , @xcite for ( an established cyclotron line source ) and @xcite for ( noting a potential cyclotron line residual near 63kev ) . the correct continuum and therefore the presence of a 25kev crsf in ( in addition to the established @xmath455kev one )
has long been debated @xcite . in the most recent analysis
its presence was confirmed in a _ suzaku _
observation @xcite .
another widely used modification is obtained by adding a broad gaussian component , most often in emission , to a standard or already modified model @xcite . since our swiftj1626.6@xmath05156 observations show a positive residual at @xmath414kev before modeling the cyclotron line ( figure [ fig2]b ) we applied this modified model to test whether our cyclotron line detection is robust against the choice of a more complex rollover description . ) for observation lo3 ( residuals shown in figure [ fig2]b ) , we further smoothed the continuum in this alternative fit by first including a broad gaussian emission feature centered at 13.7kev . while this flattens the residuals substantially , the absorption - like feature at 10kev still remains .
_ b ) _ final fit for this mixed emission / absorption approach after also including the 9.7kev cyclotron line .
the remaining residual near @xmath35kev is most likely caused by an instrumental effect ( [ sec : iodine ] ) .
[ fig5 ] ] we therefore altered our fit procedure by first adding a broad emission line near @xmath136kev to the initial continuum plus fe line fit for lo3 ( figure [ fig2]b , equation [ eq1 ] ) .
the residuals of this fit are shown in figure [ fig5]a .
the negative residual near 10kev is still clearly visible , giving a @xmath137 of 2.46 for 56 dof .
thus an absorption - like feature was still required for an adequate description .
multiplying by a gaussian optical depth absorption feature and refitting we obtained @xmath138kev , @xmath139kev , @xmath140 for the cyclotron line and @xmath141kev , @xmath142kev , @xmath143photons@xmath144s@xmath145 for the emission - like line .
this fit with both lines yielded @xmath146 @xmath147 ( figure [ fig5]b ) , as opposed to @xmath148 @xmath147 for the original best - fit ( figure [ fig2]c , table [ tab2 ] ) .
the fits are equally good ( pci@xmath14950% ) . while the depth of the cyclotron line was smaller in the modified fit , the line was still detected at the same energy as before ( table [ tab2 ] ) .
the parameters of the modified fit characterizing the continuum shape were @xmath150 and @xmath151kev , consistent within errors with the original fit ( table [ tab2 ] ) .
the same fe line parameters were obtained for both fit procedures .
this implies that , while the underlying continuum might be slightly different from a standard cutoff power law , such a difference can not be significantly detected in this observation .
the absorption - like feature at 10kev on the other hand is significant .
it appears at 9.7kev regardless of the presence or absence of a broad emission feature and its energy is stable .
based on these results and since the model without the broad emission feature is the simpler one , it is the one we chose as our best - fit model . the same overall behavior has been seen for all the datasets with the exception of ds3 , for which a better fit was obtained including a broad emission feature , see notes of table [ tab2 ] for further details .
this can be understood since the ds3 spectrum was obtained averaging over many datasets , which might contain some evolution of the continuum .
returning to figure [ fig2]c we point out a small depression near 18kev . in figure
[ fig2]d we show the obtained residuals after fitting this feature with another gaussian optical depth absorption line with centroid energy @xmath152kev , @xmath153 , and @xmath154 , which could not be constrained , fixed at @xmath40.02 .
this additional component reduced @xmath155 slightly but did not significantly improve the fit , as its pci@xmath440% from the ratio of variances @xmath113-test @xcite .
we therefore did not include this feature in our best - fit model of the phase averaged spectrum , i.e. , table [ tab2 ] .
while we would not normally consider this a detection , we also found that the addition of this 18kev feature greatly improved the pulse phase resolved spectral fits ( [ sec : phaseresolved ] ) during the pulse peak .
it is marginally present in the phase averaged spectrum of lo2 as well .
it is thus likely the first harmonic of the fundamental crsf , which is significantly detected in only a few profile phase bins ( see [ sec : phaseresolved ] ) . even after fitting the harmonic feature there remained an even more obvious line - like residual near 36kev ( figures [ fig2]d and [ fig5]b ) .
it is tempting to interpret this as a further harmonic .
there might even be another weak residual at @xmath13130kev .
however , the effective area of hexte shows a sharp drop at 33kev due to the k - edge of iodine with a smooth recovery until about 45kev(see figure 5 of * ? ? ?
so there is a good chance that ( at least part of ) any residuals in this energy range ( is ) are due to imperfect calibration @xcite .
in addition the strongest internal hexte background line , which is due to k - lines from the tellurium daughters of various iodine decays , sits at 30kev @xcite .
we therefore decided to not further pursue the potential existence of harmonics beyond the first one at 18kev .
if swiftj1626.5@xmath05156 shows other outbursts in the future the detection of these harmonics might be possible with higher sensitivity with _ suzaku _ , or better , _ nustar _ or _ astro - h_. 5156 ( dashed line ) as well as the best - fit power law index ( crosses ) .
the shaded region indicates the four phase bins surrounding the pulse peak , in which a second crsf line is necessary to produce an acceptable fit .
[ fig6 ] ] we again used lo3 , which shows the strongest cyclotron line , for a phase resolved spectral analysis .
we extracted good xenon event data with a time resolution of 3.9ms and corrected the event times to the solar system barycenter .
a period search on these data resulted in a pulse period of @xmath156s , consistent with the spin down evolution presented by @xcite .
we folded the 3.525kev events on this period in order to obtain the pulse profile , see dashed line in the upper panel of figure [ fig6 ] . as presented by @xcite the pulse profile shows a single broad peak with a roughly constant maximum over four phase bins .
the falling edge seems to be steeper than the rising edge , which may be an artifact from the period folding .
we did not include a binary correction since effects of the @xmath4133d orbit @xcite are negligible for our @xmath13130ks observation .
we divided the pulse profile into 16 phase bins for spectral analysis and used ` ikfasebin ` to extract spectra for each phase .
each spectrum was fit using eq .
[ eq2 ] with an additional edge component at @xmath45kev to account for the xenon l - edge calibration feature @xcite .
still we found it necessary to remove one spectral bin at @xmath44.5kev ( spectral bin 12 ) from each phase s spectrum to improve the overall fit .
we assume that the systematic residual in this spectral bin is due to calibration uncertainties at lower energies and has no physical significance for the model .
the column density and the iron emission energy were frozen at @xmath157 and 6.4kev the values from the phase averaged fit , respectively . in order to better constrain the spectral index ,
the folding energy of the cutoff power law was also frozen at the phase averaged value of 10.98kev .
ccccccccccccccc @xmath158 & @xmath159 & @xmath160 & @xmath161 & @xmath162 & @xmath163 & @xmath164 & @xmath165 & @xmath166 & @xmath166 & @xmath166 & @xmath166 & * 1.34 * & @xmath167 & @xmath168 + @xmath169 & @xmath170 & @xmath171 & @xmath172 & @xmath173 & @xmath174 & @xmath175 & @xmath176 & @xmath166 & @xmath166 & @xmath166 & @xmath166 & * 0.98 * & @xmath177 & @xmath178 + @xmath179 & @xmath180 & @xmath181 & @xmath182 & @xmath183 & @xmath184 & @xmath185 & @xmath186 & @xmath166 & @xmath166 & @xmath166 & @xmath166 & * 0.93 * & @xmath187 & @xmath188 + @xmath189 & @xmath190 & @xmath191 & @xmath192 & @xmath193 & @xmath194 & @xmath195 & @xmath196 & @xmath166 & @xmath166 & @xmath166 & @xmath166 & * 0.68 * & @xmath197 & @xmath198 + @xmath199 & @xmath200 & @xmath201 & @xmath202 & @xmath203 & @xmath204 & @xmath205 & @xmath206 & @xmath166 & @xmath166 & @xmath166 & @xmath166 & * 0.96 * & @xmath207 & @xmath208 + @xmath209 & @xmath210 & @xmath211 & @xmath205 & @xmath212 & @xmath213 & @xmath165 & @xmath214 & @xmath166 & @xmath166 & @xmath166 & @xmath166 & * 1.69 * & @xmath215 & @xmath216 + @xmath217 & @xmath218 & @xmath219 & @xmath220 & @xmath221 & @xmath222 & @xmath182 & @xmath223 & @xmath166 & @xmath166 & @xmath166 & @xmath166 & * 1.41 * & @xmath224 & @xmath225 + @xmath226 & @xmath227 & @xmath228 & @xmath229 & @xmath230 & @xmath231 & @xmath232 & @xmath233 & @xmath234 & @xmath235 & @xmath236 & * 1.24 * & @xmath237 & @xmath238 & @xmath239 + @xmath240 & @xmath241 & @xmath242 & @xmath243 & @xmath244 & @xmath245 & @xmath246 & @xmath247 & @xmath248 & @xmath235 & @xmath249 & * 1.05 * & @xmath250 & @xmath251 & @xmath188 + @xmath33 & @xmath252 & @xmath253 & @xmath254 & @xmath255 & @xmath256 & @xmath257 & @xmath258 & @xmath259 & @xmath235 & @xmath260 & * 1.33 * & @xmath261 & @xmath262 & @xmath263 + @xmath264 & @xmath265 & @xmath266 & @xmath267 & @xmath268 & @xmath269 & @xmath270 & @xmath271 & @xmath272 & @xmath235 & @xmath273 & * 1.17 * & @xmath274 & @xmath275 & @xmath239 + @xmath276 & @xmath277 & @xmath278 & @xmath267 & @xmath279 & @xmath280 & @xmath281 & @xmath282 & @xmath166 & @xmath166 & @xmath166 & @xmath166 & * 1.43 * & @xmath283 & @xmath284 + @xmath285 & @xmath210 & @xmath286 & @xmath254 & @xmath287 & @xmath288 & @xmath289 & @xmath165 & @xmath166 & @xmath166 & @xmath166 & @xmath166 & * 0.89 * & @xmath290 & @xmath291 + @xmath292 & @xmath293 & @xmath294 & @xmath161 & @xmath295 & @xmath296 & @xmath297 & @xmath298 & @xmath166 & @xmath166 & @xmath166 & @xmath166 & * 1.03 * & @xmath299 & @xmath300 + @xmath301 & @xmath302 & @xmath303 & @xmath172 & @xmath304 & @xmath305 & @xmath306 & @xmath254 & @xmath166 & @xmath166 & @xmath166 & @xmath166 & * 1.02 * & @xmath307 & @xmath308 + @xmath309 & @xmath159 & @xmath310 & @xmath311 & @xmath312 & @xmath313 & @xmath314 & @xmath315 & @xmath166 & @xmath166 & @xmath166 & @xmath166 & * 0.99 * & @xmath316 & @xmath317 + using only the continuum plus fe line model ( eq . [ eq1 ] ) on the phase resolved spectra led to unacceptable fits throughout the whole pulse , see column @xmath318 in table [ tab3 ] . including a crsf in the form of a gaussian optical depth absorption line at @xmath319kev ( eq . [ eq2 ] ) improved the fits significantly , but did not result in a sufficiently low @xmath155 during the maximum of the main peak , see column @xmath320 in table [ tab3 ] . including a second absorption line in these four phase bins improved the fits to acceptable values , see column @xmath321 in table [ tab3 ] .
the results from these fits are shown in figure [ fig6 ] and table [ tab3 ] .
the pci @xcite in the last column was calculated for the best fit in each phase bin , i.e. , between the models with no crsf and one crsf for phase bins 17 and 1216 , and between no crsf and two crsfs for phase bins 811 .
the four phase bins requiring the second cyclotron line are highlighted by the shaded region in figure [ fig6 ] .
the width of the second line was frozen at @xmath235kev in order to better constrain the other parameters .
the centroid energy is roughly twice the energy of the fundamental line and is consistent with being its first harmonic .
the phase resolved results for the harmonic likely explain the additional feature mentioned in the description of the phase averaged spectrum ( [ sec : harmonic ] , figure [ fig2]d ) .
it has such a low significance in the phase averaged spectrum because it is not detected at most rotation phases and its energy changes with phase .
there are other crsfs that are only detected in phase resolved spectra .
for example , the second harmonic of the crsf in her x-1 is detected only in the descent of the main peak @xcite .
figure [ fig6 ] shows that the power law index hardens throughout the peak and softens in the minimum .
as can be expected the power law normalization roughly follows the pulse profile , reaching @xmath4@xmath322 in the peak .
the crsf energy shows a smaller increase during the rising edge , then first a decline and then a drastic increase of more than 20% in the peak itself , followed by a slow decline over the falling edge and minimum of the pulse profile .
the width and depth of the crsf are variable as well with a dip in the rising edge and a possible increase through the peak and a rapid decline at the beginning of the falling edge .
we found evidence for cyclotron resonance scattering features at @xmath410 and 18kev in the spectrum of the be / x - ray binary swiftj1626.6@xmath05156 . while the harmonic crsf was not seen at a significant level in the phase averaged spectrum , pulse phase resolved spectroscopy showed it is a necessary addition to the spectral fit between phases @xmath40.40.7 , i.e. , during the pulse peak .
we can use the cyclotron line energy to estimate the strength of the magnetic field local to the line production region near the neutron star s polar regions .
the fundamental and subsequent harmonic cyclotron line energies are related to the local magnetic field strength by @xcite @xmath323 where @xmath324 is the electron rest mass , @xmath325 is the integer harmonic number , @xmath326 g is the critical field strength for resonant scattering , @xmath327 is the angle between the photon direction and the magnetic field vector , @xmath10 is the gravitational redshift at the neutron star surface ( @xmath328 for a neutron star with mass @xmath329 and radius @xmath330 cm ) and @xmath331 .
a fundamental energy of @xmath410kev therefore implies a magnetic field of @xmath332 g @xmath333 g . as a consistency check we estimated the global magnetic field strength , assuming that the neutron star has reached its equilibrium spin rate through interplay between an accretion disk and the magnetic field .
we refer to @xcite and ( * ? ? ?
* chapter 15 ) for our calculation and use the pulse period of lo3 .
while the pulse period evolved over the outburst , the change was small enough to not throw the spin out of equilibrium : we refer to figure 3 of @xcite , which shows a change of only 0.2% in the pulsar s spin period over the whole activity phase .
this small change in @xmath334 had little effect on our estimate of the global magnetic field .
we estimated @xmath335 by equating the alfvn radius @xmath336 with the corotation radius at which the orbital velocity is equivalent to the surface velocity of the rotating star , thus giving @xmath337rads@xmath145 where @xmath338 here @xmath339 is the mass accretion rate and @xmath340 is the dipolar magnetic moment .
@xmath341s is our measured spin period , so @xmath342 and @xmath343 cm . to estimate the average @xmath344 over the source s lifetime
, we must first estimate the luminosity of the source outside of outburst .
@xcite estimated a source distance of @xmath345kpc .
we assume that the current low luminosity state of the source is typical outside of outburst and from the source s spectrum during this state we derived a 260kev flux of @xmath346 , giving a luminosity of @xmath347 at 10kpc for isotropic emission .
the mass accretion rate for an efficiency @xmath348 is then @xmath349 .
using equation [ ra ] we find @xmath350 and the global dipolar magnetic field strength @xmath351 g . as a rough estimate
this is reasonably close to the field strength derived from the cyclotron line ( see previous section ) .
something else to consider is the relationship between the cyclotron line energy and the source s x - ray luminosity . and are examples of systems containing an accreting magnetized neutron star whose cyclotron line energy is positively correlated with luminosity @xcite .
a counter example is , in which the cyclotron line energy increases as the source s luminosity decays and vice versa @xcite-@xmath352 correlation @xcite . in this case
the presence of the correlation , however , depends on the continuum model @xcite . ] .
it was suggested by @xcite that a positive correlation between @xmath39 and luminosity would occur for systems accreting at a sub - eddington rate , while the negative correlation would occur when the accretion is locally super - eddington .
they show that the fractional change in @xmath39 is directly proportional to the fractional change in @xmath352 .
if such a relationship could be confirmed with cyclotron line sources of known distances it would be possible to use observations of cyclotron lines as standard candles .
the 90% errors on the cyclotron line energy in table [ tab2 ] point to a possible correlation between @xmath39 and luminosity .
we calculated the @xmath353 errors on the line energy of lo1 , lo3 , and ds3 , observations which are clearly separated in time , and found that the energies are not consistent with each other within the errors .
the @xmath353 confidence range on the crsf energy is 10.1510.55kev during lo1 , 9.589.85kev during lo3 , and 9.329.56kev during ds3 .
the measured decrease in the line energy with decreasing flux ( luminosity ) is indicative of a positive correlation between @xmath39 and @xmath352 and is apparent at the 99.7% confidence level in our data .
further observations during a future outburst are necessary to confirm this correlation .
@xcite recently derived a new expression for the critical luminosity assuming a simple physical model for the accretion column . according to their estimate , and using the @xmath13 field determined in [ sec : bfield1 ] , the critical luminosity for swiftj1626.5@xmath05156 is @xmath354 @xmath355 .
the normalized 320kev luminosity for lo1 , the observation with the highest flux @xmath113 ( table [ tab2 ] ) , is @xmath356 at a distance @xmath357kpc , with all observations from ds1 and later falling below @xmath358 .
thus swiftj1626.6@xmath05156 was mainly observed in the sub - critical accretion regime and the observed positive @xmath39-@xmath352 correlation is consistent with the theoretical expectation .
the strong @xmath13220% variation of the fundamental energy with pulse phase that we measured for swiftj1626.6@xmath05156 is consistent with several other sources , namely cen x-3 @xcite , vela x-1 @xcite , 4u0115@xmath35963 @xcite , and gx 301@xmath02 @xcite . in all cases
the fundamental energy varies by about 1030% over the pulse .
this variability has been attributed to the change in viewing angle and accretion column throughout the pulse , resulting in a variable local magnetic field as a function of pulse phase .
specifically the @xmath39 versus phase profile of cen x-3 has very similar properties to the one of swiftj1626.6@xmath05156 ( though shifted in phase from the falling to the rising edge of the pulse ) : a sharp rise followed by a slower decay .
for cen x-3 these @xmath39 variations were shown to be inconsistent with a changing viewing angle on a pure dipole magnetic field but it was found that it is possible that emission from above both poles is observed @xcite .
the width and depth of the line are also generally variable over pulse phase , similar to what was observed for swiftj1626.6@xmath05156 .
we interpret the phase dependence of the crsf parameters as additional evidence that the feature is real and not an artifact of the spectral fitting .
the ratio between the fundamental and harmonic line energies for the phase averaged fit is 1.9 and ranges from 1.7 to 2.0 for the four phase bins where the harmonic line is detected .
thus the harmonic probably does not occur at strictly twice the fundamental energy .
other sources , for example v0332@xmath35953 @xcite and 4u0115@xmath35963 @xcite , have also displayed cyclotron lines at non - integer multiples of the fundamental energy .
spacing that is narrower than in the harmonic case is expected when general relativistic effects are taken into account @xcite , while a wider spacing has , e.g. , been explained as being due the emission regions of the two lines being located at different heights in the accretion column @xcite .
for swiftj1626.6@xmath05156 the relativistic effect might have been observed , similar to v0332@xmath35953 @xcite53 the situation is not clear , however , since ratios @xmath12 have also been obtained @xcite , depending on the line and continuum model . ] .
swiftj1626.6@xmath05156 is located in the direction of the galactic plane , so the diffuse galactic fe k emission must at least contribute to our measured iron line flux . from the quiescence observations we already know that this contribution is small ( [ sec : ridge ] ) .
in addition we can make the following estimate : for a typical galactic ridge location , the total fe k line flux is @xmath360 @xcite . the pca solid opening angle is 0.975deg@xmath361 and the effective area of one pcu at 6.4kev is @xmath362 , so the diffuse fe k line flux ( averaged over the area ) seen by the pca is @xmath4@xmath363 @xcite .
we calculated the 320kev fe k line flux from each of the three long observations by first finding the flux of the best - fit model and then removing the emission line and , without refitting , calculating the flux again .
we find fe k fluxes of @xmath4@xmath364 from lo1 , @xmath4@xmath365 from lo2 , and @xmath4@xmath366 from lo3 , indicating that most of the iron line emission is coming from swiftj1626.6@xmath05156 .
we further note that the flux of the iron line , @xmath367 , as well as of the absorption @xmath368 decrease over the time of the outburst ( table [ tab2 ] ) .
since @xmath369 and since @xmath344 depends on the amount of material in the system , this may reflect the diminishment of material from the neutron star s surroundings .
we have clearly detected a spectral feature at @xmath410kev in the spectrum of the be / x - ray binary swiftj1626.6@xmath05156 . every aspect of our analysis points to this feature being a cyclotron resonance scattering feature .
the parameters of the cyclotron line vary with pulse phase in a manner consistent with other accreting pulsars displaying cyclotron lines .
the second harmonic of the crsf , at an energy of @xmath418kev , is seen between pulse phases 0.40.7 .
the fundamental cyclotron line and the @xmath46.4kev iron emission line persist throughout the source s outburst and oscillatory stages .
the neutron star s magnetic field strength of @xmath4@xmath370 g , derived from the cyclotron line energy , is physically realistic and in the expected range for high - mass x - ray binaries .
interestingly it is among the lowest @xmath13 fields measured for an accreting pulsar so far , together with that of 4u0115@xmath35963 @xcite .
we find that this field strength is roughly consistent with the scenario of spin equilibration via disk accretion and we do not rule out the possibility that swiftj1626.6@xmath05156 may host an accretion disk .
comparing the @xmath371 confidence intervals of the fundamental crsf energies in lo1 , lo3 , and ds3 reveals a positive correlation between the line energy and the source luminosity .
this behavior is similar to that of her x-1 and gx 304@xmath01 , in which the luminosity and line energy are also positively correlated , and according to the model of @xcite and @xcite implies that the outburst decay of swiftj1626.6@xmath05156 was sub - eddington .
observations during future outbursts of swiftj1626.6@xmath05156 will be beneficial in confirming this relationship between its luminosity and cyclotron line energy .
we thank w. coburn for pointing out swiftj1626.6@xmath05156 as a candidate cyclotron line source .
we acknowledge the useful comments of the referee that substantially improved this paper , and thank him or her for pointing out the use of the ratio of variances @xmath113-test from which we derived the probability of chance improvement for a given model .
this work was funded under solicitation nnh07zda001n - swift4 of the swift guest investigator program and nsf grant ast0708424 .
j. wilms acknowledges partial funding from the bundesministerium fr wirtschaft und technologie through deutsches zentrum fr luft- und raumfahrt grants 50 or 0808 and 50 or 0905 .
s. suchy acknowledges the support of nasa contract nas5 - 30720 and nasa grant nnx08az82 g . | swiftj1626.6@xmath05156 is a be / x - ray binary that was in outburst from december 2005 until november 2008 .
we have examined _ rxte_/pca and hexte spectra of three long observations of this source taken early in its outburst , when the pca 220kev count rate was @xmath170counts@xmath2 , as well as several combined observations from different stages of the outburst .
the spectra are best fit with an absorbed cutoff power law with a @xmath3kev iron emission line and a gaussian optical depth absorption line at @xmath410kev .
we present strong evidence that this absorption - like feature is a cyclotron resonance scattering feature , making swiftj1626.6@xmath05156 a new candidate cyclotron line source .
the redshifted energy of @xmath410kev implies a magnetic field strength of @xmath5 g in the region of the accretion column close to the magnetic poles where the cyclotron line is produced .
analysis of phase averaged spectra spanning the duration of the outburst suggests a possible positive correlation between the fundamental cyclotron energy and source luminosity .
phase resolved spectroscopy from a long observation reveals a variable cyclotron line energy , with phase dependence similar to a variety of other pulsars , as well as the first harmonic of the fundamental cyclotron line . |
phase - separating binary fluids form complex patterns of domains after a quench into the two - phase region of the phase diagram .
the domain morphology is determined by a number of factors such as the volume fractions of the two phases , their viscosities , and any external forces applied to the system @xcite .
of particular interest here is the effect of an external shear flow applied to the fluid .
the shear flow competes with the phase - separation process , influencing the morphology and stability of the domains . besides being a fascinating problem in nonequilibrium physics
, this question is of practical significance because the final properties of industrial materials involving binary fluids often depend on the domain morphology . at late times after a quench into an unstable state , a phase - separating binary fluid consists of domains of the two phases which coarsen with time .
the presence of a shear flow dramatically alters the kinetics of the phase separation .
the effects of the deformation by the shear flow depend on the relations between the various time scales in the system .
the characteristic time scale for the shear flow is just the inverse shear rate , @xmath0 . when this time is shorter than the characteristic relaxation time of thermal concentration fluctuations @xmath1 , @xmath2 , the system is in a `` strong - shear '' regime in which the critical fluctations are modified by the flow .
conversely , in the `` weak - shear '' regime @xmath3 the critical fluctuations are not affected by the flow .
a third time scale , which will be crucial here , is associated with the domains .
clearly when the growth rate of the domains , or the growth rate of any instabilities associated with the domains , is smaller than the time associated with the flow , the shear flow will affect the morphology and stability of the domains . of interest in this paper is the competition between the shear flow and the coarsening process .
the shear flow tends to deform and elongate or fragment domains , whereas the thermodynamics favors coarsening to larger , isotropic domains .
this competition leads to the formation of a nonequilibrium , dynamic steady state in which the coarsening is stopped by the shear flow @xcite . when the viscosities of the two phases are similar and one phase forms a droplet phase , the steady state consists of somewhat deformed droplets of the order of the taylor breakup size @xmath4 , where @xmath5 is the surface tension , @xmath6 the viscosity and @xmath7 the shear rate @xcite . on the other hand , when the two phases are both percolated the morphology is an anisotropic bicontinuous structure with apparently stable domains highly elongated along the flow direction .
the anisotropy in these bincontinuous patterns is much larger than the aspect ratio of 2 or 3 seen for isolated droplets @xcite .
microscope observations have shown that the domains can be elongated into long cylinders @xcite . in weak shear , @xmath8 ,
these stringlike domains still undergo frequent breakup , reconnection , and branching , whereas in the strong - shear regime the system forms a `` string phase '' consisting of macroscopically long cylindrical domains aligned with the flow direction .
these are surprising observations , since a long fluid cylinder at rest is unstable against breaking up into spherical droplets via the rayleigh capillary instability @xcite .
in the situation under consideration here , the string is a domain of one phase immersed in the other , which we would also expect to be thermodynamically unstable since the cylinder could lower its surface energy by spheroidizing .
thus the shear flow stabilizes both the thermodynamic instability towards phase separation and the hydrodynamic instability of these highly elongated domains .
the goal of this paper is to explore the stabilization of cylindrical domains by an imposed shear flow .
it is a sequel to my previous work on the stabilization by shear flow of a two - dimensional , lamellar domain in phase - separating binary fluids @xcite . in that paper
it was shown that a lamellar domain at rest with diffuse interfaces is unstable towards a `` varicose '' instability .
this instability is essentially a coarsening effect and depends on the finite width of the interfaces ; in the limit of mathematically sharp interfaces a lamellar domain is stable .
an external shear flow stabilizes the lamellar domain by advecting the top and bottom interfaces with respect to each other so that they no longer can maintain the exact phase relation which produces the unstable varicose mode .
the instability of a quiescent cylindrical domain of one phase immersed in the other is somewhat different , owing to the different dimensionality .
there are two separate forces driving instability , one hydrodynamic and one thermodynamic .
rayleigh @xcite and later tomotika @xcite analyzed the instability of an infinitely long viscous fluid cylinder immersed in an immiscible fluid to axisymmetric varicose perturbations as shown in fig .
[ varicose ] .
when the wavelength of the varicosity is equal to or longer than the circumference of the cylinder the perturbation is unstable and grows .
this is because the higher curvature in the necks leads to a higher laplace pressure there than in the bulges , which tends to drive fluid from the necks toward the bulges .
eventually the cylinder will break up into spherical droplets , with less total surface energy than the original cylinder .
however , even in the absence of fluid motion ( e.g. consider a cylindrical domain in a solid binary alloy ) a cylindrical domain in the two - phase state is still unstable .
this is due to the gibbs - thomson effect , in which the chemical potential depends on the curvature @xcite .
the higher curvature at the necks will drive a diffusive flux towards the bulges , also leading to instability .
both of these mechanisms are present even in the limit that the interfaces are mathematically sharp . in this paper
i perform a linear stability analysis to investigate the effect of an external shear flow on the stability of a single infinitely long cylindrical domain , perfectly aligned with the flow .
i consider late times after a quench into the two - phase region , when the fluid consists of domains of the two phases close to their equilibrium concentrations , separated by well - defined interfaces .
the formulation of the problem allows in principle for diffuse interfaces with a finite width @xmath9 ( when the viscosities of the two phases are equal ) , but in practise results are much easier to obtain in the limit of mathematically sharp interfaces and the corrections due to finite widths will not affect the results qualitatively . the shear rate @xmath7 is assumed to be small enough that the system is in the weak - shear regime , @xmath3 , so that the shear flow will not influence the structure of the interfaces themselves . in this case the usual hydrodynamic equations for a phase - separating fluid are valid . as mentioned above
, the string phase itself seems to form only in the strong - shear regime @xcite .
however , here i will be concerned not with the formation of the string phase but with its stability .
the goal is to understand the mechanism by which the shear stabilizes these remarkably elongated domains .
the results may also illuminate the stability of the highly anisotropic , bicontinuous morphologies observed in weak shear .
i will neglect the ends of the string and also the possiblity that it could be inclined at a small angle to the flow direction .
this approximation seems reasonable given the extraordinarily high aspect ratio observed for the strings and the fact that long slender drops in shear have a long central portion which is cylindrical and aligned with the flow @xcite .
as well as shedding light on the stability of elongated domains in phase - separating fluids , this work encompasses the problem of the effect of shear flow on the purely hydrodynamic viscous rayleigh instability in immiscible fluids ( neglecting the thermodynamic effects ) . to my knowledge this problem has not been solved before in the particular limit examined here .
russo and steen @xcite and lowry and steen @xcite found that axial flows can suppress capillary instabilities on cylindrical interfaces .
most other studies in the fluid dynamics literature concerning the effects of externally imposed flows on long fluid cylinders have been in the context of drop breakup .
several authors have considered the linear stability of an infinitely long fluid cylinder in an elongational flow @xcite .
the flow field limits the growth of any disturbance to a finite value so that there is no true instability , and the cylinder is stabilized .
however , some disturbances have time to grow transiently to a finite amplitude comparable to the decreasing radius of the elongating cylinder , causing breakup ( even though the disturbances do not grow exponentially ) .
khakhar and ottino @xcite extended the analysis to general linear flows including shear flow , but only in the case of small asymmetry , when the shear part of the flow is small compared to the stretching . in this paper
i will explore the opposite limit in which the stretching is negligible but the asymmetry is large .
finally , hinch and acrivos studied a finite , long slender drop in shear flow @xcite .
they find steady - state solutions for the shape of the drop at all shear rates , but these equilibrium solutions are unstable above a critical shear rate .
this is essentially due to the fact that the ends of the drop are not completely aligned with the flow , so that at sufficiently high shear rates the drop can not balance the stretching of the ends and it extends transiently , becoming progressively thinner .
this does not happen for an infinitely long cylinder , as considered in this paper . in sec .
[ eqtns ] i will describe the model equations of motion used to describe the fluid . in sec .
[ stab ] the equations of motion are linearized for small perturbations about a cylindrical domain .
approximate solutions can be found by writing the stability eigenvalue equation as a matrix equation in a truncated set of `` basis '' states , corresponding to different perturbation modes of the cylinder .
the matrix elements are calculated in sec .
[ noshear ] .
we will see in sec .
[ shearres ] that the shear flow has the effect of mixing nonaxisymmetric disturbances with the axisymmetric varicose mode , leading to stabilization in some circumstances .
i will first discuss the results for the special case in which the viscosities of the two phases are equal , and then generalize to the case of arbitrary viscosity ratio .
some discussion of the relations of this work to experiment will be presented in sections [ relexp ] and [ disc ] .
i use the same equations of motion as in @xcite .
a simple binary mixture can be described by one scalar order parameter @xmath10 , the difference in concentration between the two components .
since we are interested in late times after a temperature quench when the system consists of well - defined domains , the usual ginzburg - landau form for the coarse - grained free energy of a symmetrical mixture is sufficient to describe the thermodynamics : @xmath11=\int d{\bf r } \left(\frac{1}{2 } k \left({\bf \nabla}\phi\right)^2 - \frac{1}{2}r_0\phi^2 + \frac{1}{4}g\phi^4 \right),\ ] ] where @xmath12 and @xmath13 are positive constants so that the fluid is in the two - phase region . minimizing the homogeneous part of @xmath14 leads to the values of the concentration in the two bulk phases at equilibrium : @xmath15 the fluid
is assumed to be incompressible and sufficiently viscous that inertial effects are negligible .
the equations of motion for the system are then the modified cahn - hilliard equation for @xmath10 , the stokes ( creeping flow ) equation for the velocity field * u * , and the incompressibility condition : @xmath16 here @xmath17 is a concentration - independent mobility ; @xmath18 is the viscosity ; and @xmath19 is the pressure , which in general is determined by the incompressibility condition ( [ incomp1 ] ) .
the equation for the velocity ( [ n - s ] ) is generalized to include the coupling of the order parameter to the velocity field @xcite .
this term leads to a capillary force at interfaces , where gradients in @xmath10 induce fluid flow . equations ( [ ch])([incomp1 ] ) are the same as those of `` model h '' ( without the thermal noise terms ) used to study critical binary fluids @xcite .
these equations have been used extensively to study phase separation in binary fluids @xcite . now consider a single cylindrical domain of radius @xmath20 composed of say , phase @xmath21 with viscosity @xmath22 , immersed in an infinite region of phase @xmath23 with viscosity @xmath24 as illustrated in fig .
the external shear flow is imposed along the @xmath25 direction by applying a constant shear stress @xmath26 far from the cylinder . below
i will allow for a finite width interface between the two phases only in the case that the viscosities are equal , @xmath27 ; when the viscosities between the two phases are different i will assume the interfaces are sufficiently sharp that the viscosity changes discontinuously at the interface so that eq .
( [ n - s ] ) holds in the two different phases separately .
the first step in a stability analysis of the cylinder is to derive the steady - state solutions to the equations of motion which correspond to these assumptions and to the geometry of fig .
we therefore assume that @xmath10 is a function of @xmath28 only and that the velocity is only nonzero in the @xmath25 direction , @xmath29 , and look for time - independent solutions .
the cahn - hilliard equation ( [ ch ] ) has steady state solutions satisfying @xmath30 where @xmath31 is the exchange chemical potential .
using cylindrical coordinates @xmath32 the stationary concentration profile @xmath33 therefore satisfies @xmath34 for a sufficiently large radius , @xmath35 will approach the profile for a flat interface between the two coexisting phases , @xmath36 , \quad \text{large $ { \cal r}$ } , \label{profiler}\ ] ] where the width of the interface between the two coexisting phases is the thermal correlation length @xmath37 .
i will assume throughout that @xmath38 , so that eq . ( [ profiler ] )
is reasonable .
often it will be justified to further approximate the interfacial profile by a step function so the interfaces are sharp , @xmath39 .
\label{step}\ ] ] note that for either interfacial profile there is a surface tension associated with the presence of the interface , which is just the excess free energy per unit area at the interface @xcite : @xmath40 if the viscosities of the two phases are equal , @xmath27 , applying a constant shear stress @xmath26 far from the cylinder leads to simple shear flow everywhere , with stationary velocity field @xmath41 where @xmath42 is the shear rate . more generally , for arbitrary viscosity ratio @xmath43 the stationary velocity field @xmath44 will have a different slope in the two phases . taking the interface to be mathematically sharp as in eq . ( [ step ] ) , we can solve equations ( [ n - s ] ) and ( [ incomp1 ] ) for @xmath45 inside and outside the cylinder separately and match the solutions at the interface at @xmath46 .
the velocity field must be regular at the origin and correspond to simple shear flow far from the cylinder , so that @xmath47 where the shear rate is defined in terms of the outer viscosity , @xmath48 .
solving for @xmath45 gives @xmath49 \cos \theta { \bf \hat{x } } , & r > { \cal r}. \end{array } \right . \label{us}\ ] ] it is convenient to rewrite the equations in dimensionless form by scaling lengths by the correlation length @xmath9 , the concentration by its equilibrium magnitude in the bulk phases @xmath50 , and time by the characteristic diffusion time @xmath1 .
the velocity is scaled by the correlation length over the diffusion time : @xmath51 note that the new dimensionless correlation length is @xmath52 . in dimensionless form
the equations of motion are now @xmath53
@xmath54 @xmath55 the equations are characterized by a dimensionless parameter , the rescaled viscosity @xmath6 : @xmath56 ( in the case of two different viscosites there are two dimensionless parameters , @xmath57 and @xmath58 . ) in dimensionless form the stationary solutions corresponding to the cylindrical domain in shear are @xmath59 @xmath60 \cos \theta { \bf \hat{x } } , & \bar{r } > r , \end{array } \right .
\label{us2}\ ] ] where the dimensionless radius of the cylinder is @xmath61 .
the dimensionless shear rate @xmath62 is simply the product of the shear rate and the diffusion time @xmath1 and thus represents a second dimensionless parameter that characterizes the strength of the shear flow .
in this section i present the strategy for calculating the stability of the cylindrical domain . we know that an infinite cylinder is unstable to varicose perturbations .
one could imagine other , nonaxisymmetric perturbations of the cylinder as well , such as the `` undulation '' mode shown in fig .
[ m1 ] . in sec .
[ noshear ] we will find that the cylinder is stable to all of these perturbations . the shear flow will have the effect of mixing the different possible perturbations .
consider small perturbations about the stationary solutions found above ( in the rest of the discussion i will drop the bars over the dimensionless variables for clarity ) : @xmath63 to linear order in the small perturbations @xmath64 and @xmath65 , the equations of motion are @xmath66 @xmath67 @xmath68 here @xmath6 is the appropriate viscosity for whichever region is under consideration , @xmath69 is the @xmath28 component of the perturbed velocity field * v * , and primes indicate differentiation with respect to @xmath28 .
@xmath70 is a function of the stationary concentration profile : @xmath71 the time dependence of the perturbations is determined by the concentration equation ( [ conc ] ) .
the system is translationally invariant in the @xmath25 direction , so we can write any perturbation as a sum over fourier modes in @xmath25 .
since we are interested in the growth ( or damping ) of perturbations we take @xmath72 we will be interested in long - wavelength fluctuations for which the dimensionless wave number @xmath73 ( let @xmath74 be the wave number and @xmath75 be the growth ( damping ) rate in the original variables ) . substitution into eq .
( [ conc ] ) leads to an eigenvalue equation for the growth rate @xmath76 : @xmath77 a real , positive value of @xmath78 indicates stability of the cylinder against the perturbation . since @xmath69 depends on @xmath64 through eq .
( [ hydro ] ) , this eigenvalue equation is essentially an integro - differential equation in which the expression for @xmath69 acts as an integral operator on @xmath64 .
( [ ev ] ) can not be solved exactly , so we need an approximate approach . following the calculational approach outlined in @xcite ,
first consider the cahn - hilliard part of eq .
( [ ev ] ) , without the hydrodynamic terms : @xmath79 where we have defined the operators @xmath80 @xmath81 this part of eq .
( [ ev ] ) includes the dynamics of the concentration field on the scale of the interface .
since @xmath82 ( @xmath83 ) , the shear flow acts on the scale of the domains but is not strong enough to alter the interfacial profile . as we will see shortly , this assumption allows us to find an approximate solution for @xmath64 .
let @xmath84 be the set of eigenfunctions of eq .
( [ ch2 ] ) and define a set of `` conjugate '' functions by @xmath85 then one can show that @xmath86 and @xmath87 are hermitian operators ( although their product is not ) as long as the @xmath84 and @xmath88 obey periodic boundary conditions or vanish at infinity .
the eigenvalues @xmath89 are real and the eigenfunctions and their conjugates are orthogonal : @xmath90 for any pair of trial functions @xmath91 and @xmath92 obeying the same boundary conditions , there is a variational relation which gives an upper bound on the lowest eigenvalue @xmath76 @xcite : @xmath93 here the parentheses again indicate inner products .
this variational theorem can be exploited to find solutions to eq .
( [ ch2 ] ) corresponding to various perturbations of the cylinder .
application of eq .
( [ var ] ) requires a good trial function @xmath91 .
the smallest eigenvalues of eq .
( [ ch2 ] ) will correspond to eigenfunctions describing @xmath94-dependent deformations of the cylinder , in which the interface is translated by a small amount but the interfacial width remains fixed @xcite .
higher eigenvalues correspond to other deformation modes in which the structure of the interface changes , such as breathing modes which change the width of the interface .
i will neglect all such modes here , since they are more quickly damped than the slow translational modes and are not important to the dynamics on the scale @xmath95 .
thus , we can solve the cahn - hilliard part of the eigenvalue equation ( [ ev ] ) by using the variational theorem with a trial function corresponding to the translational deformation of interest .
the translational modes can be characterized by their angular dependences .
any general perturbation of the concentration field can be expanded as a fourier series in @xmath94 : @xmath96 where @xmath97 is the function necessary to translate the interface by a small amount @xmath98 in the @xmath99 direction ( the functional form of the @xmath97 will be calculated below in sec .
[ diffusive ] ) . the cahn - hilliard part of the eigenvalue equation
can then be rewritten as @xmath100 where the operators are @xmath101 each mode @xmath102 corresponds to a different geometrical perturbation mode of the cylinder . in the absence of the external shear flow , the cylindrical domain will be unstable to the axisymmetric , @xmath103 varicose mode as discussed in the introduction ( see fig . [ varicose ] ) .
we will see below that the @xmath104 mode shown in fig . [
m1 ] is an exact solution to eq .
( [ ch2 ] ) for @xmath105 . at @xmath105
, it simply corresponds to a uniform translation of the entire cylinder and is thus marginally stable with eigenvalue @xmath106 .
we might anticipate that the cylinder will be stable to higher modes in @xmath107 as well .
note that the shear flow term in eq .
( [ ev ] ) is proportional to @xmath108 , so this term should have the effect of mixing modes with different values of @xmath107 .
now consider the full eigenvalue equation , eq .
( [ ev ] ) .
we are interested in the stability of perturbations characterized by the various @xmath94-dependent translational modes .
the functions @xmath109 are approximate eigenvectors of eq .
( [ ch2 ] ) . to solve the full equation we adopt an approximation similar to `` tight - binding '' or @xmath110 approximations used in solid state physics .
we assume the translational modes are good basis states for the full problem and write eq .
( [ ev ] ) as a matrix equation in this basis .
we can truncate the matrix to only include a finite number of states @xmath107 and then diagonalize the matrix to find the stability eigenvalues .
this is valid when the two hydrodynamic terms are small enough that they only cause mixing among the @xmath107 states included in the basis ; they must be small relative to the distance to the next higher eigenvalue not included .
also , the shear flow must satisfy @xmath82 so that it is reasonable to only consider the translational deformation modes . in the strong - shear regime @xmath111 ( @xmath112 )
the shear flow might couple to other modes that we have neglected , which alter the width or shape of the interfacial profile itself , since these modes only damp out on a time scale of roughly @xmath1 .
note that the term containing @xmath113 in eq .
( [ ev ] ) depends on @xmath64 through the hydrodynamic equation ( [ hydro ] ) .
so for each mode @xmath107 we can solve eq .
( [ hydro ] ) for @xmath113 , assuming that @xmath114 is given by the approximate basis function @xmath115 . in general the resulting velocity field
can then also be expanded as a fourier series . denoting by @xmath116 the solution for @xmath65 obtained from substituting eq .
( [ fourier ] ) and @xmath117 into eq .
( [ hydro ] ) , we can expand @xmath118 the @xmath94 dependence of @xmath65 will not necessarily be the same as that of @xmath119 so in general the coefficients @xmath120 in the sum will be nonzero even for @xmath121 . to obtain the effective matrix equation corresponding to eq .
( [ ev ] ) we write @xmath114 as a vector @xmath122 and multiply on the left in eq .
( [ ev ] ) by the corresponding conjugate vector . here
the @xmath123 are the amplitudes of the small perturbations @xmath119 .
recall the conjugate function @xmath124 is defined by @xmath125 so it satisfies the poisson equation @xmath126 we can expand @xmath124 in the same way as @xmath64 so that @xmath127 in which case @xmath128 .
we can easily solve for @xmath129 using the green s function for the laplacian in cylindrical coordinates .
the result is @xmath130 where @xmath131 ( @xmath132 ) indicates the lesser ( greater ) of @xmath28 and @xmath133 , and @xmath134 are the modified bessel functions .
then eq .
( [ ev ] ) becomes after multiplying on the left by @xmath135 and integrating over all @xmath94 , ( 3.5,.25 ) ( 0,.125)(1,0)3 @xmath136 ( 6.5,.25 ) ( 3.75,.125)(1,0)3 2 solving this equation gives approximate stability eigenvalues @xmath78 for the cylinder in the shear flow .
we have used the definition @xmath137 ; the diagonal diffusive terms are then exactly the variational expression ( [ var ] ) .
the shear terms involving the stationary velocity @xmath138 are completely off - diagonal , so they will indeed have the effect of mixing the modes .
we will find in sec .
[ vmu ] that the off - diagonal elements involving @xmath65 only mix modes that differ by @xmath139 as written in eq .
( [ matrix1 ] ) and that they are also directly proportional to the shear rate @xmath140 .
thus in the absence of the shear flow , @xmath141 , the matrix is completely diagonal and the @xmath107 modes are independent , with stability eigenvalues @xmath142 these zero - shear stability eigenvalues are the sum of two terms , one due to hydrodynamic transport in the system and the other due to diffusive transport . solving eq .
( [ matrix1 ] ) for non - zero shear rate requires truncating the matrix at some mode @xmath107 ; since we expect only the @xmath103 mode to be ( possibly ) unstable , we might anticipate that only a few of the higher modes are needed to investigate the behavior of the @xmath103 mode under shear . to summarize the results of this section
, the equations of motion were first linearized in the small perturbations @xmath64 and @xmath65 and expressed parametrically in terms of the wave number @xmath143 .
the perturbations of the cylinder of interest here , the translational modes , were characterized by their dependence on @xmath94 .
a variational expression was introduced for the diffusive part of the problem , and the eigenvalue equation for the full problem was written as a matrix equation in the basis of the @xmath107 translation modes . in the remaining sections
i calculate the various matrix elements in eq .
( [ matrix1 ] ) , which requires solving the hydrodynamic equation for the perturbed velocity field @xmath65 , and then solve the matrix equation itself and examine the results for various parameters of interest .
we begin by calculating the diffusive contribution to the matrix elements @xmath144 as defined in eq .
( [ evm ] ) .
we need to determine the @xmath28 dependence of the basis functions @xmath119 . for @xmath145 and @xmath105 , translating the entire cylindrical interface by an amount @xmath146 , @xmath147 , requires adding the function @xmath148 to the original interfacial profile . but
the translated interface , and therefore @xmath149 , should also be an exact solution of ( [ chm ] ) .
we can verify this by differentiating eq .
( [ profile ] ) for the stationary solution @xmath35 with respect to @xmath28 : @xmath150 but this is simply equivalent to @xmath151 so @xmath152 is an exact eigenfunction of @xmath153 for @xmath105 , with eigenvalue @xmath154 .
we can exploit this solution to approximate @xmath155 for general @xmath107 .
since @xmath156 ( @xmath157 ) and since @xmath158 is only significant near the interface at @xmath159 , we can in general replace the term @xmath160 by @xmath161 in the expression for @xmath162 ( [ fm ] ) .
but then we would expect all @xmath107 modes to have roughly the same radial dependence as the @xmath104 mode , so we can approximate @xmath163 for nonzero @xmath143 , @xmath164 , so this gives @xmath165 from eq .
( [ evm ] ) the diffusive part of the diagonal matrix element @xmath166 is then @xmath167 we are assuming we can approximate @xmath168 by the flat interface form eq .
( [ profiler ] ) so @xmath169 .
since all modes have the same radial dependence we set @xmath170 in the denominator as well , which gives @xmath171 @xmath172 the diffusive contribution to @xmath76 has thus been reduced to quadrature .
equations ( [ wmdf ] ) and ( [ tphimr ] ) are integrable numerically and are valid for diffuse interfaces of width @xmath9 .
however , qualitatively the results are the same for sharp interfaces , in which case the result can be expressed in closed form . the easiest way to take
the sharp interface limit is to take @xmath173 in all of the integrals , where @xmath174 is the dirac delta function .
( this is equivalent to using the step - function profile eq .
( [ step ] ) . )
the prefactors maintain the correct normalizations , so we are taking the width of the interface @xmath175 while keeping the surface tension @xmath5 constant .
the delta functions then allow easy evaluation of the necessary integrals .
the conjugate function @xmath176 is @xmath177 so the normalization integral is @xmath178 substituting into eq .
( [ wmdf ] ) we find the diffusive term in the stability eigenvalue is @xmath179 in the original variables , the dispersion relation is @xmath180 which we note can be written independently of @xmath9 as @xmath181 we immediately see that the axisymmetric @xmath103 mode is thermodynamically unstable for @xmath182 , _
i.e. _ for wavelengths longer than the circumference of the cylinder , whereas the @xmath104 translation mode is stable for all @xmath183 as predicted .
[ wdg0 ] shows the dimensionless part of the first three modes , @xmath184 where @xmath185 .
these are the diffusive contributions to the stability eigenvalues in the absence of the shear flow . fig .
[ finite ] shows the difference between using finite width interfaces in eq .
( [ wmdf ] ) and using the sharp interface expression ( [ wmds ] ) for a cylinder of radius @xmath186 , for the lowest two eigenvalues , @xmath187 and @xmath188 .
the finite width curve is at most @xmath189 more negative than the sharp interface curve for @xmath187 , and @xmath190 larger than the sharp interface curve for @xmath188 , at @xmath191 ( near the maximally unstable varicose mode ) .
the shapes of the curves are qualitatively the same in both cases .
as @xmath192 increases and the step - function approximation for the interfacial profile becomes better , the difference between the curves grows smaller .
for @xmath193 the differences are reduced to roughly @xmath194 and @xmath195 , respectively .
thus the qualitative behavior should remain unchanged for a domain with sharp interfaces , but detailed comparison with experimental or simulational data may require including the effect of having diffuse interfaces . next consider the matrix elements involving the perturbed velocity field @xmath65 .
this section will be limited to the special case in which the viscosities of the two phases are equal , @xmath196 .
then it will turn out that the velocity matrix elements are completely diagonal , even with the shear flow , and it will be possible to obtain closed form expressions for @xmath197 .
recall that the perturbed velocity field @xmath65 satisfies equations ( [ hydro ] ) and ( [ incomp ] ) : @xmath198 @xmath199 where @xmath200 . to solve for @xmath201 , we follow a general procedure from happel and brenner @xcite . taking the divergence of eq .
( [ hydro2 ] ) and applying the incompressibility condition ( [ incomp2 ] ) leads to a poisson equation for the pressure @xmath19 : @xmath202 expanding the pressure as @xmath203 we find @xmath204 satisfies @xmath205 the homogeneous part of which is simply the modified bessel equation . using a green s function and requiring @xmath204 to be finite at @xmath206 and to vanish as @xmath207 , we have @xmath208 \nonumber \\ & & \cong \int_{0}^{\infty}dr ' \ , \frac{d}{dr'}\left [ k_{m}(kr_{>})i_{m}(kr_{<})\right ] \left(\frac{1}{2 } k^{2}r ' \right .
\nonumber \\ & & \left . \quad
\quad + \frac{m^{2}-1}{2r ' } \right ) { \mathrm sech}^{4}(r'-r ) , \label{pressure}\end{aligned}\ ] ] where we have used eq .
( [ fmphim ] ) . substituting this expression for @xmath19 into eq .
( [ hydro2 ] ) gives a vector poisson equation for * v*. in cylindrical coordinates , the @xmath28 and @xmath94 components of @xmath65 are coupled : [ vcoupled ] @xmath209 we can solve for both together by writing [ vexp ] @xmath210 so that @xmath211 has a phase shift relative to @xmath69 .
( [ vcoupled ] ) then become @xmath212 adding these together we find @xmath213 the homogeneous part of this equation is the modified bessel equation , with general solutions @xmath214 and @xmath215 .
similarly subtracting gives @xmath216 so now the homogeneous part of the equation has solutions @xmath217 and @xmath218 .
we can thus construct exact green s functions for the combinations @xmath219 and @xmath220 .
the solutions to the inhomogeneous equations are therefore ( 3.5,.25 ) ( 0,.125)(1,0)3 [ vsoltn ] @xmath221 all of the functions in these integrals are known , so we have reduced the solution for @xmath222 to quadrature . for the case of equal viscosities , we could therefore again integrate numerically to find the matrix elements for finite width interfaces .
instead i will proceed with the sharp interface approximation eq .
( [ sharpint ] ) . in this limit
the velocity matrix elements in eq .
( [ matrix1 ] ) , divided by the normalization integrals , are @xmath223 thus @xmath224 is given not surprisingly by the radial velocity at the linearized position of the interface , @xmath225 evaluated at @xmath159 . from eq .
( [ pressure ] ) we find the pressure @xmath226 is given by @xmath227 substituting into eq .
( [ vsoltn ] ) , solving for @xmath69 , and performing some reductions using bessel function identities , we find @xmath228 \nonumber \\ & & \left .
+ \frac{1}{8}q^{2}[k_{m+1}(q)i_{m+1}(q ) - k_{m-1}(q)i_{m-1}(q ) ] [ k_{m-1}(q)i_{m+1}(q)-k_{m+1}(q)i_{m-1}(q ) ] \right\ } \label{vrm}\end{aligned}\ ] ] ( 6.5,.25 ) ( 3.75,.125)(1,0)3 2 where we have set @xmath229 for convenience .
these give the hydrodynamic part of the stability eigenvalues in the absence of the shear flow .
since i have taken the sharp interface limit , this part of the calculation has been completely decoupled from the dynamics of the concentration field @xmath64 ; eq . ( [ wh ] ) is equivalent to the usual kinematic condition in hydrodynamic stability calculations .
thus the result eq .
( [ vrm ] ) is an exact , rather than approximate , solution for the hydrodynamic stability of a cylinder of one fluid immersed in a second immiscible one .
it can therefore be compared directly with previous results . for the varicose mode
@xmath103 we obtain the stability eigenvalue @xmath230.\end{aligned}\ ] ] putting this back in dimensional form we have @xmath231 .
\label{tomotika}\end{aligned}\ ] ] this expression is the same as that found by stone and brenner @xcite and may be obtained from tomotika s general result for the dispersion relation of the viscous rayleigh instability , in the limit of equal viscosities between the two liquids @xcite . note each mode in eq .
( [ vrm ] ) can be written in the original variables as @xmath232 , where @xmath233 is dimensionless .
this dimensionless part of the dispersion relation is graphed in fig .
[ disp0 ] for the first three modes .
again the varicose mode is unstable for all @xmath234 , whereas the higher modes are stable for all @xmath183 . for the case of general viscosity ratio @xmath31 ,
it is much more difficult to solve the hydrodynamic equation with diffuse interfaces .
to do so requires writing the viscosity as a smooth function of @xmath64 that changes from the value inside the cylinder @xmath57 to the value outside @xmath58 in a continuous way , which introduces extra terms into the original differential equation ( [ n - s ] ) .
thus in this section i will start with sharp interfaces from the beginning .
it is then more sensible to follow a different approach to solving the hydrodynamic equation . for sharp interfaces , the term coupling the total concentration @xmath10 to the total velocity @xmath235 in eq .
( [ n - s ] ) becomes @xcite @xmath236 where @xmath237 is the curvature of the interface located at @xmath238 and @xmath239 is a unit vector normal to the interface .
this is equivalent to the usual boundary condition on the jump in the normal stress across a fluid interface . instead of including this coupling term in the hydrodynamic equation
, we can just solve the usual creeping flow equations for the perturbed velocity field @xmath240 inside and @xmath241 outside the cylinder separately , and apply the appropriate boundary conditions at the interface in the usual manner . in each region
@xmath65 satisfies @xmath242 and the pressure simply satisfies the laplace equation , @xmath243 the location of the interface of course depends on the perturbation mode @xmath119 in question ; since @xmath97 is now a step function we only need know the location of the step , which is given by @xmath244 for each mode @xmath107 , where @xmath245 is the amplitude of the small perturbation . the solutions to eqs .
( [ hydro3])([peq ] ) are straightforward so the details are given in the appendix .
the general solutions are found in terms of modified bessel functions ; the exact solutions are then found by applying the boundary conditions and solving numerically for the unknown constants .
the results are the coefficients @xmath246 in the expansion ( [ vnm ] ) . as shown in the appendix ,
the only nonzero coefficients are the ones for which @xmath247 and @xmath248 so eq . ( [ matrix1 ] ) is tridiagonal as written . as was shown in the last section in eq .
( [ wh ] ) , the velocity matrix elements are then simply given by @xmath249 in the case of zero shear flow , only the diagonal matrix elements are nonzero and the hydrodynamic part of the stability eigenvalues is therefore @xmath250 . again
the results here for @xmath141 are decoupled from @xmath64 and are therefore exact , for two immiscible liquids .
i show in the appendix that in the absence of the shear flow the velocity matrix elements and hence the eigenvalues are proportional to @xmath251 , just as in the @xmath252 case .
[ w0hmu][w2hmu ] show the dispersion relations for the dimensionless parts of the three lowest modes for different values of @xmath31 ( holding @xmath58 constant ) .
the results for the varicose @xmath103 mode are again the same as those of tomotika @xcite .
as the cylindrical domain becomes less viscous relative to the background , it becomes more unstable and the wave number of maximum instability @xmath253 becomes smaller .
note that the damping rate for the undulatory @xmath104 mode depends less strongly on @xmath31 . from fig .
[ w2hmu ] we see that the @xmath254 depends only weakly on @xmath255 .
finally , in the appendix the hydrodynamic equations are solved analytically at @xmath105 , providing a check on the numerical results .
the stability eigenvalues at @xmath105 are diagonal and independent of the shear rate @xmath140 , eq .
( [ k0exact ] ) : @xmath256 with @xmath257 .
note that since the only remaining off - diagonal elements in eq .
( [ matrix1 ] ) vanish at @xmath105 , this demonstrates that the cylinder is stable towards @xmath25-independent deformations which would result in a non - circular cross - section , even in the presence of the shear flow .
finally we evaluate the off - diagonal shear flow matrix elements @xmath258 where we have divided by the normalization integral . since we are assuming that @xmath97 has the same radial dependence for all @xmath107 , we simply have in the sharp interface limit @xmath259 thus each off - diagonal element is the same independent of @xmath107 .
with all the necessary matrix elements in hand , we can now solve for the stability eigenvalues in shear by diagonalizing eq .
( [ matrix1 ] ) . to do
so requires truncating the matrix at some point .
since the off - diagonal matrix elements are all proportional to @xmath140 , we can assume that any perturbation which damps out more quickly than the time associated with the shear flow can be ignored .
thus we only need include modes whose damping rates are less than or of the order of the applied shear rate . for comparison purposes ,
the values of the stability eigenvalues at @xmath105 give good estimates of the damping rates of high @xmath107 modes without having to calculate the full dispersion relations .
in this paper it will be sufficient to only include the first three modes , in which case the matrix equation ( [ matrix1 ] ) becomes a @xmath260 secular equation for @xmath76 : @xmath261 where i have pulled a factor of @xmath262 out of the off - diagonal velocity matrix elements , @xmath263 .
this leads to a cubic characteristic equation for the stability eigenvalues @xmath78 .
since the characteristic equation has real coefficients , the roots can be found analytically ; there will be either three real roots or one real root and a complex conjugate pair .
recall that in the absence of the shear flow , the stability eigenvalues are simply given by the sum of the diffusive and hydrodynamic terms as in eq .
( [ evm ] ) : @xmath264 i showed in section [ diffusive ] and section [ vmu ] that these two terms scale differently with the parameters of the system , with @xmath265 and @xmath266 .
the relative magnitude of these two terms depends on the dimensionless viscosity parameter @xmath267 and on the radius of the cylinder . for sufficiently viscous and/or small cylindrical domains , the diffusive term will dominate , whereas for less viscous and/or large cylindrical domains
, the hydrodynamic term dominates . in the following
i first examine the stability eigenvalues in the two extreme limits in which the hydrodynamic terms or the diffusive terms can be disregarded entirely .
i will then present some results for experimentally realistic parameter values .
first consider eq .
( [ matrix2 ] ) in the limit that the diffusive terms in the diagonal elements are negligible , so that @xmath268 .
physically this corresponds to examing the effect of the shear flow on the purely hydrodynamic rayleigh instability . since the diagonal matrix elements @xmath224
scale as @xmath251 while the off - diagonal elements do not depend on @xmath58 or @xmath95 , the stability eigenvalues @xmath76 will also scale with @xmath251 .
denote the eigenvalues found by diagonalizing eq .
( [ matrix2 ] ) with the shear flow present by @xmath269 where the index @xmath270 refers to the order of the new eigenvalues : @xmath271 . returning to the original variables ,
this relation is simply @xmath272 since @xmath273 has units of inverse time , we can measure the shear rate @xmath7 just as well in these units as in units of @xmath274 , so for the rest of this section i will take @xmath275 .
we start by considering the equal viscosity case , @xmath252 .
( [ mu1gp1])([mu1gp18 ] ) show the real part of the dimensionless dispersion relations @xmath276 for the first two stability eigenvalues in shear flow at various shear rates ( the third mode was included in the calculation but is not a particularly interesting function of @xmath140 ) .
for some values of @xmath183 the lowest two modes @xmath277 and @xmath278 are a complex conjugate pair and so they show up as a single curve in figs .
( [ mu1gp1])([mu1gp18 ] ) ; for the regions in @xmath183 for which there are two curves shown , the two eigenvalues are both completely real . we see that at small @xmath183 where there is a complex conjugate pair , the perturbations are traveling waves with an overall damping rate . at low shear rates
the mode with the lowest eigenvalue @xmath277 is still unstable , but the window of wave numbers over which @xmath279 becomes smaller as @xmath140 increases .
at some critical shear rate @xmath280 the minimum in @xmath277 crosses zero , at a critical wave number @xmath281 . above this critical shear rate ,
the instability is gone the initially unstable varicose mode has been stabilized by the applied shear flow , by being mixed with the higher modes . clearly , the critical shear rate @xmath282 must have the same dependence on @xmath24 , @xmath5 , and @xmath192 as the stability eigenvalues : @xmath283 thus the critical shear rate is a monotonically decreasing function of the radius of the cylinder and the magnitude of the outside viscosity ; the smaller the cylinder or the less viscous the fluids are overall , the faster the growth of the varicose mode and so the critical shear rate must be faster as well for stabilization to occur
. inverting eq .
( [ gammac ] ) we see that at a fixed shear rate , there is a critical radius @xmath284 above which the cylinder is stable and below which it is unstable .
thus if instead of an infinite cylinder we had a finite long cylindrical drop that was being stretched by the flow , initially small capillary disturbances on the drop would be suppressed by the flow , but as the drop thinned to a radius smaller than @xmath285 the disturbances would start to grow and the drop would break up . note that @xmath280 is considerably larger than the magnitudes of both @xmath286 and @xmath287 at all @xmath288 , so the shear rate must be faster than the rate of growth of the instability to stabilize it .
the qualitative picture here is that the shear flow advects opposite sides of the cylinder relative to each other so that the special axisymmetric , varicose perturbation no longer exists long enough to be unstable .
this picture is born out by the eigenvectors corresponding to the eigenvalues shown in figs .
[ mu1gp1][mu1gp18 ] .
[ evsgp1 ] shows a cross - section of the eigenvector corresponding to the lowest eigenvalue at @xmath289 ( where @xmath253 is the wave number of maximum instability in zero shear ) and @xmath290 , when the lowest mode is still unstable ; fig .
[ evsgp18 ] shows the same eigenvector at @xmath291 , which is above the critical shear rate and therefore stable . as the shear rate increases ,
the original varicose mode becomes more distorted as it is mixed with the other modes .
next we explore the effect of the viscosity ratio between the two phases , @xmath31 . when the viscosities of the two phases are equal , above the critical shear rate the cylinder is stable against perturbations at all wavelengths as we see in fig .
[ mu1gc ] .
this is not the case for all @xmath31 . for general @xmath31
, the shear flow does stabilize the varicose mode around the main instability at @xmath253 , but for some @xmath31 there is a residual instability left at small wave numbers ( long wavelengths ) .
an example for @xmath292 is shown in figs .
[ mup25gp14 ] and [ mup25gp16 ] .
[ mup25gp14 ] shows the lowest two stability eigenvalues at a low shear rate , when the lowest mode @xmath277 is still unstable near the original maximally unstable wave number @xmath293 .
we see that there is an additional unstable region at small @xmath183 , separate from the main instability .
this long wavelength instability remains after the main instability has been stabilized by the shear flow , as shown in fig .
[ mup25gp16 ] .
the physical significance of this residual instability will be discussed further in sec .
[ relexp ] ; it does mean that the cylinder is still unstable to very long wavelength perturbations .
no residual instability was found for @xmath294 ; as @xmath31 is either increased or decreased away from this range , a small instability appears smoothly from @xmath295 and extends over increasingly large @xmath183 as @xmath31 becomes correspondingly smaller than about @xmath296 or larger than about @xmath297 .
nevertheless we can calculate the critical shear rate required to stabilize the original maximally unstable mode , as a function of @xmath31 . the result
is shown in fig .
the graph only includes values in the range @xmath298 since for values of @xmath31 outside this range , the critical shear rate @xmath280 becomes larger than the damping rate of the @xmath299 mode ; to extend the range of @xmath31 would therefore require including the @xmath300 mode and higher as @xmath280 increased .
[ from eq .
( [ wk0 ] ) , we find @xmath301 , @xmath302 , @xmath303 , and @xmath304 so the range in fig .
[ scmu ] is reasonable . ]
@xmath305 has a minimum near @xmath306 and rises on either side , so that as the domain becomes either more or less viscous than about half the outside viscosity , it requires a higher shear rate to stabilize it .
the rather sharp bend near @xmath307 is due to the fact that the maximally unstable varicose mode with growth rate @xmath308 moves to lower wave numbers as @xmath31 decreases ( see fig .
[ w0hmu ] ) , while the @xmath104 mode in particular changes less with @xmath31 ( fig .
[ w1hmu ] ) .
the magnitudes of the two eigenvalues both increase as the inner viscosity decreases , but the growth rate of the @xmath103 mode does so more quickly and with a larger change in the dependence on @xmath309 .
thus as @xmath31 decreases the interaction between these two modes changes in such a way as to result in the fairly sharp increase in shear rate necessary for stabilization for @xmath310 . in the last section we saw that the hydrodynamic rayleigh instability can be partially or completely stabilized by the shear flow , depending on the viscosity contrast between the two phases .
the opposite limit is to consider what happens when the fluids are so viscous that the hydrodynamic terms are negligible .
then the diagonal elements in eq .
( [ matrix2 ] ) are just @xmath144 and the off - diagonal elements are the ones from the imposed flow @xmath311 .
[ wdgp1 ] and fig .
[ wdgp4 ] show the dimensionless parts of the first two modes , @xmath312 , for two different shear rates at @xmath252 . here
i have defined @xmath313 so that the trivial dependence on @xmath95 can be factored out .
again as @xmath314 is increased the window of wave numbers for which the varicose mode is unstable becomes smaller .
also , the mode of maximum instability moves to lower @xmath183 .
[ wdg2 ] shows all three modes in shear for a rather high shear rate , @xmath315 .
the varicose mode has been mostly stabilized , with a small residual instability at small wave numbers .
including the fourth mode , @xmath300 , allows us to raise the shear rate up to some fraction of the magnitude of the @xmath316 mode , @xmath317 .
( note that including the @xmath300 mode does not change the lower three modes e.g. in fig .
[ wdg2 ] at all , since it is not mixed with them at low shear rates . )
the instability at small wave numbers moves to smaller and smaller @xmath183 as @xmath314 is increased above @xmath318 towards @xmath319 , but seems pinned near @xmath295 and never quite disappears .
the changes with @xmath314 occur more slowly as @xmath314 is increased .
thus when the diffusive terms dominate the behavior , there is no well - defined critical shear rate for stabilization at all wave numbers @xmath183 .
furthermore , unlike the case of the rayleigh instability analyzed above , the mode which is maximally unstable at @xmath320 does not cross the axis at some well - defined shear rate independently of the residual instability at small wave number ; instead the maximally unstable mode just shifts with @xmath314 towards smaller @xmath183 .
we could however define a critical shear rate for stabilization at any given ( small ) wave number @xmath321 ; in this case the `` critical '' shear rate for stabilization will scale simply as @xmath322 since @xmath31 only enters in the off - diagonal shear flow terms .
once again the critical shear rate is a decreasing function of the radius @xmath95 , so for a given shear rate small cylinders will be unstable and large ones will be stable for wave numbers satisfying @xmath323 . in general the stability of the cylindrical domain will be determined by both hydrodynamic and diffusive effects , depending on the system parameters . in this section
i will first consider the parameters relevant to hashimoto _
et al_. s experimental system @xcite .
they have studied phase separation under shear flow in a pseudobinary mixture of polybutadiene ( pb ) and polystyrene ( ps ) in a common solvent of dioctylphthalate ( dop ) .
they find a correlation length @xmath324 , a surface tension on the order of @xmath325 erg , a diffusion constant on the order of @xmath326/s , the viscosity of pb / dop @xmath327 poise , and the viscosity of ps / dob @xmath328 poise @xcite . for comparison with my results i will take as an example a viscosity ratio between the two phases of @xmath329 so the cylindrical domain consists of the less viscous phase . for the possible range of values of @xmath58 in the experiment , @xmath330 , the hydrodynamic terms in the diagonal matrix elements @xmath166 are significantly larger than the diffusive terms at all reasonable values of @xmath20 .
this is not surprising ; at the large length scale of the domains , @xmath38 , we would not necessarily expect the diffusive terms to be important .
thus in this case the results of sec .
[ rstable ] apply with negligible modification .
the critical shear rate for stabilization of the cylinder at most @xmath183 , for @xmath331 and @xmath332 , is @xmath333 since the theory only applies for @xmath334 or so ( for smaller @xmath95 eq .
( [ profiler ] ) is no longer a good approximation ) , this shear rate is in the weak - shear regime , @xmath3 , and is significantly smaller than the shear rate necessary for formation of the string phase seen in the experiments .
it is thus consistent that the long cylindrical domains seen experimentally are stable , since they are seen at shear rates that are well above the shear rate required for stabilization .
[ exp ] shows the stability eigenvalues well above the critical shear rate , at @xmath335 .
although this shear rate is in the strong - shear regime @xmath112 for which the theory may not strictly be valid , it seems reasonable that the theory can be pushed into the strong shear regime at the large length scales @xmath336 considered here ( if in the strong - shear condition @xmath1 is replaced by the typical time scale for domain fluctuations , then the theory is valid here ) . for these particular parameter values ,
et al_. found that the length of the strings seen in the experiment were on the order of @xmath337 .
this length could be explained by the residual instability at small wave numbers discussed in sec .
[ rstable ] . the wavelength of maximum instability in fig .
[ exp ] is approximately @xmath338 , so the length of the strings seen experimentally may be set by the residual long wavelength varicose instability in the shear flow .
next consider the case of near - critical binary fluids . at the critical point
, @xmath252 and @xmath6 is a universal number ; for near - critical fluids it has the same order of magnitude , so we take the critical value @xmath339 . for this value of @xmath6
, the diffusive terms start to become noticeable at small radii , although they are still significantly smaller than the hydrodynamic terms at larger radii , e.g. @xmath340 or so . fig .
[ diff ] shows the varicose @xmath103 mode without the shear flow , for the diffusive and hydrodynamic terms separately at the smallest @xmath95 that is reasonable in the theory .
the diffusive terms do change the magnitude of @xmath341 and so will have a small quantitative effect on the results .
the stabilization by the shear flow is again very similar to the purely hydrodynamic case as illustrated in figs .
( [ mu1gp1])([mu1gp18 ] ) .
however , even for @xmath252 there is now once again a small residual instability at small @xmath183 due to the diffusive part of the eigenvalues that persists at high shear , for the small radius @xmath342 .
note from fig .
[ diff ] that at very small @xmath183 , @xmath343 , so it is not surprising that the stability eigenvalues in shear resemble those of fig .
[ wdg2 ] at small @xmath183 .
finally , the critical shear rate for stabilization of the main instability @xmath280 no longer scales exactly with @xmath344 at small @xmath95 due to the different scaling of the diffusive terms ( @xmath345 ) , but the difference is small .
thus , for near - critical binary fluids the effects of diffusive transport may be observable in string - like domains for sufficiently thin strings .
i have shown that shear flow can stabilize an isolated cylindrical domain in the two - phase state of a phase - separating binary fluid against varicose instabilities , by mixing the varicose mode with the other nonaxisymmetric perturbation modes of the cylinder .
essentially , the shear flow distorts the varicose mode by convecting one side of the cylindrical interface with respect to the other , eliminating the special axisymmetric , pinching character of the varicose mode that drove the instability .
both the hydrodynamic rayleigh instability and the thermodynamic , diffusive instability are suppressed by the shear flow , although there are residual instabilities at small wave numbers in the limit that the diffusive terms dominate and also for viscosity ratios @xmath31 outside the range @xmath294 .
other authors have considered the effect of flows on the rayleigh instability , but this is the first study focused on the effect of the nonaxisymmetric nature of shear flow on the rayleigh instability . comparing with the experimental results of hashimoto _
et al_. , i found the mechanism presented here for stabilization of the cylindrical domains is consistent with the stable `` string '' phase seen experimentally , and that the lengths of the strings may be set by the residual instabilty at long wavelengths .
however , it should be noted that the stability of a cylindrical domain in shear flow does not act as a criterion for the observed relationship between the shear rate and the radius of the domains seen in the string phase .
the domains in the string phase are formed through a dynamic process ; the observed radius is not a parameter of the system ( as in this calculation ) but rather is determined through the self - organization process as the shear flow competes with coarsening in the phase - separating system .
i have merely demonstrated a mechanism by which these macroscopically long , cylindrical domains may be stabilized by the shear flow .
although a few experiments have looked at the breakup of the string - like domains after complete cessation of the shear flow @xcite , it would be interesting to do a careful experimental study of the shear rate at which the strings first begin to be unstable to see if the rayleigh and/or thermodynamic varicose instabilities explored here are the main breakup mechanisms in these systems .
if so then the strings should be unstable below the critical shear rate @xmath282 found in this paper .
the results presented here may also shed light on why the shear flow can halt the phase separation and result in a dynamic steady state , even in the weak - shear regime , @xmath83 . in a concentrated phase - separating fluid when the two phases are both percolated so that the domains form a connected bicontinuous pattern , the coarsening is dominated by curvature effects .
qualitatively we can think of a piece of the interconnected structure as a cylinder of fluid immersed in the other phase .
this cylinder is susceptible to the varicose instabilities considered here , and particularly to the rayleigh instability , which leads to breakup of the cylindrical region into spheres .
siggia @xcite used this picture to explain the coarsening rates seen in concentrated binary fluids .
since the shear flow suppresses these instabilities , one might expect that it could stabilize an anisotropic , bicontinuous morphology against further coarsening . for a given shear rate , when the domains are relatively small they will be unstable and will coarsen , but once the typical length scale has grown to the critical radius for stabilization by the shear flow , @xmath346 , the parts of the bicontinuous structure that are cylindrical and aligned with the flow will no longer be unstable .
this then provides a mechanism for the creation of the nonequilibrium dynamic steady state seen in concentrated phase - separating fluids in shear flow .
i would like to thank jim langer and glenn fredrickson for many helpful discussions and support .
i thank t. hashimoto , a. onuki , and e. moses for useful discussions and correspondence , and j. goveas for references to the appropriate fluid dynamics literature .
i would also like to thank the university of california , santa barbara for financial support .
this work was supported by the mrl program of the national science foundation under award no .
dmr 96 - 32716 and by the u.s .
doe grant no .
de - fg03 - 84er45108 .
the perturbed velocity field @xmath347 corresponding to a perturbation of the interface given by @xmath348 must satisfy the hydrodynamic equations @xmath349
@xmath350 @xmath351 to solve for @xmath352 we again follow happel and brenner @xcite .
the velocity is expanded as in eq .
( [ vnm ] ) , @xmath353 we will solve for each coefficient @xmath354 separately so we take @xmath355 .
we start by solving eq .
( [ peq.2 ] ) for the pressure .
let @xmath356 ; then in general @xmath357 where @xmath358 and @xmath359 are constants .
the perturbed velocity then satisfies the inhomogeneous laplace equation ( [ hydro3.2 ] ) , so the solution will consist of a general solution to the homogeneous part plus a particular solution , @xmath360 .
writing @xmath361 , the @xmath25 component of eq .
( [ hydro3.2 ] ) is @xmath362 a particular solution is given by @xmath363 where @xmath364 and @xmath365 .
the @xmath28 and @xmath94 components of @xmath366 satisfy eq .
( [ vcoupled ] ) without the extra @xmath64 term ; assuming they are expanded as in eq .
( [ vexp ] ) the particular solutions to eq .
( [ hydro3.2 ] ) are @xmath367 @xmath368 ( the solutions to the homogeneous part of the equation will be included in the general solution below and so are not needed here ) .
the components of @xmath369 must satisfy the continuity equation ( [ incomp3.2 ] ) : @xmath370 this determines the constants @xmath371 and @xmath372 . next we need a general solution to the homogeneous part of eq .
( [ hydro3.2 ] ) .
again these are just the appropriate solutions of the laplace equation : @xmath373 @xmath374 @xmath375 and once again enforcing incompressibility gives @xmath376 and @xmath377 .
the result is a solution for @xmath378 containing six unknown constants , @xmath379 , @xmath380 , and @xmath381 : [ vg ] @xmath382 & r < r \\ -(b^{+}+b^{- } ) k_{n}(kr ) + \pi_{2}\left[k_{n}(kr ) + kr k_{n}'(kr)\right ] & r > r , \end{array}\right.\ ] ] @xmath383 & r < r \\
b^{+ } k_{n+1}(kr ) + b^{- } k_{n-1}(kr ) - \pi_{2 } kr\left[k_{n+1}'(kr ) + k_{n-1}'(kr)\right ] & r > r , \end{array } \right.\ ] ] @xmath384 & r < r \\
b^{+ } k_{n+1}(kr ) - b^{- } k_{n-1}(kr ) - \pi_{2 } kr\left[k_{n+1}'(kr ) - k_{n-1}'(kr)\right ] & r > r .
\end{array } \right.\ ] ] it now remains to apply the appropriate boundary conditions at the interface to specify the remaining constants .
these boundary conditions will apply to the total velocity and stress fields . letting @xmath245
denote the amplitude of the small perturbations , recall we have @xmath386 where @xmath387 is the steady state pressure , @xmath388 the steady state stress tensor , and @xmath5 the perturbed stress tensor . from eq .
( [ us ] ) , the components of the steady state stress tensor are @xmath389 @xmath390 ( note @xmath391 is symmetric ) .
we see that @xmath392 is continuous across the interface but @xmath393 has a jump across the interface when @xmath394 .
the difference in the steady state pressure @xmath387 across the interface is simply the laplace pressure across a cylindrical interface @xmath395 , which in our dimensionless variables ( the dimensionless surface tension is @xmath396 ) is @xmath397 the boundary conditions are @xcite : * continuity of the velocity across the interface , @xmath398 * continuity of the tangential stress across the interface , @xmath399 * jump in the normal stress across the interface due to the mean curvature @xmath400 , @xmath401 ( here @xmath400 is dimensionless ) to apply these boundary conditions we need to evaluate the appropriate components of the stress tensor on the deformed interface @xcite .
the location of the cylindrical interface for mode @xmath107 is @xmath402 so the unit normal to the interface is @xmath403 the hydrodynamic force on the perturbed surface is @xmath404 .
to lowest order in @xmath245 , the three components of @xmath87 are @xmath405 the normal stress is simply @xmath406 the two tangential components of the stress can be found from @xmath407 , giving @xmath408 finally , the mean curvature @xmath400 is @xmath409 the first term is the steady state pressure difference given by eq .
( [ laplace ] ) .
since the stationary velocity @xmath45 already satisfies the boundary conditions , the perturbed velocity @xmath65 must satisfy them separately .
denote the difference between quantities inside and outside the cylinder at @xmath159 by @xmath410 = f^{i}-f^{o}$ ] . then keeping in mind eq .
( [ laplace ] ) and that @xmath411 is continuous across the interface , from eqs .
( [ norm])([tang ] ) the six boundary conditions become : [ bcs ] @xmath412\right ] & = & \left [ \left [ v_{\theta } \right]\right ] = \left [ \left [ v_{x}\right]\right ] = 0 , \\
\left [ \left [ \sigma_{rr}\right]\right ] & = & \frac{2}{3 } \frac{k^{2}r^{2}+m^{2}-1}{r^{2 } } e^{im\theta + ikx } ,
\\ \left [ \left [ \sigma_{\theta r}\right]\right ] & = & -\left [ \left [ ik e^{im\theta + ikx } \pi^{s}_{\theta x } \right ] \right ] , \\
\left [ \left [ \sigma_{xr } \right]\right ] & = & -\left [ \left [ \frac{im}{r } e^{im\theta + ikx } \pi^{s}_{\theta x } \right]\right].\end{aligned}\ ] ] since @xmath413 depends on @xmath94 , for a given perturbation mode @xmath107 of the cylinder the solution @xmath414 given by eq .
( [ vnm ] ) will contain more than one coefficient @xmath354 .
the perturbed stress tensor @xmath415 depends only on @xmath65 so we can write @xmath416 noting that @xmath417 and using eq .
( [ pixtheta ] ) at @xmath159 we can write @xmath418\right ] = \left(-\frac{s \eta^{i}}{\mu+1 } + \frac{s \eta^{o}}{\mu+1 } \right ) \left(e^{i(m+1)\theta } - e^{i(m-1)\theta}\right).\ ] ] substituting in the full sums in eq .
( [ vnm ] ) and eq .
( [ stotal ] ) and matching terms with the same @xmath94 dependence in eqs .
( [ bcs ] ) then gives for the boundary conditions on each coefficient @xmath354 , [ bcs2 ] @xmath419\right ] & = & \left[\left [ v_{\theta , nm } \right]\right ] = \left[\left [ v_{x , nm } \right]\right ] = 0 , \\ \left[\left [ \sigma_{rr , nm } \right]\right ] & = & \frac{2}{3 } \frac{k^{2}r^{2}+m^{2}-1}{r^{2 } } \delta_{mn } , \label{normal } \\ \left[\left [ \sigma_{\theta r , nm } \right]\right ] & = & \left(\eta^{i}-\eta^{o}\right ) \frac{ks}{\mu+1 } \left(\delta_{n = m+1 } - \delta_{n = m-1}\right ) , \label{tang1 } \\ \left[\left [ \sigma_{xr , nm } \right]\right ] & = & \left(\eta^{i}-\eta^{o}\right ) \frac{ms}{r(\mu+1 ) } \left(\delta_{n = m+1 } - \delta_{n = m-1}\right ) . \label{tang2}\end{aligned}\ ] ] the right - hand side of the normal stress condition eq .
( [ normal ] ) is only nonzero for @xmath247 , and the right - hand sides of the tangential stress conditions eqs .
( [ tang1 ] ) and ( [ tang2 ] ) are only nonzero for @xmath248 .
this immediately shows that for a given perturbation mode @xmath107 of the interface , the only nonzero coefficients @xmath420 will be those for which @xmath247 , @xmath421 , and @xmath422 .
the eigenvalue equation ( [ matrix1 ] ) is thus tridiagonal . to solve for each matrix element @xmath423 ( see sec .
[ vmu ] ) , we just substitute the appropriate general solution from eqs .
( [ vg ] ) into the boundary conditions ( [ bcs2 ] ) and solve the resulting system of algebraic equations for the unknown constants .
this is best done numerically given the algebra involved and was solved using a standard algorithm @xcite .
we can however see analytically how the matrix elements depend on @xmath58 and @xmath95 .
first consider the boundary condition on the normal stress , eq .
( [ normal ] ) .
writing out the stress tensor we have @xmath424\right ] & = & \left[\left[-p_{m } + 2\eta \frac{\partial v_{r , m}}{\partial r}\right]\right ] \nonumber \\ & = & -p_{1}i_{m}(kr ) + 2\eta^{i}k v_{r , m}^{\prime , i}(kr ) + p_{2}k_{m}(kr ) - 2\eta^{o}k v_{r , m}^{\prime , o}(kr ) \nonumber \\ & = & \frac{2}{3 } \frac{k^{2}r^{2}+m^{2}-1}{r^{2 } } , \label{norm2}\end{aligned}\ ] ] using eq .
( [ pressure2 ] ) for the pressure .
the primes indicate differentiation with respect to the argument of @xmath69 , @xmath425 . dividing both sides by @xmath426 leaves
@xmath427 the left - hand side now depends only on the six integration constants and on the dimensionless parameters @xmath428 and @xmath31 ( which remain dimensionless when written in the original variables ) . for the tangential stress in eq .
( [ tang1 ] ) , @xmath429\right ] & = & \left[\left [ \eta \left(\frac{1}{r } \frac{\partial v_{r}}{\partial \theta}+ \frac{\partial v_{\theta}}{\partial r } - \frac{v_{\theta}}{r } \right ) \right]\right ] \nonumber \\ & = & \left[\left [ \eta \left(\frac{im}{r } v_{r , n}(kr ) + k v_{\theta , n}'(kr ) - \frac{1}{r } v_{theta , n}(kr ) \right ) \right]\right ] \nonumber \\ & = & \left(\eta^{i}-\eta^{o}\right ) \frac{ks}{\mu+1 } \left(\delta_{n = m+1 } - \delta_{n = m-1}\right).\end{aligned}\ ] ] dividing both sides by @xmath58 and multiplying by @xmath95 gives @xmath430 - imv_{r , n}^{o}(q ) -
qv_{\theta , n}^{\prime , o}(q ) + v_{\theta , n}^{o}(q ) = \frac{\mu - 1}{\mu + 1 } qs \left(\delta_{n = m+1 } - \delta_{n = m-1}\right ) .
\label{tang1.2}\ ] ] similarly , from eq .
( [ tang2 ] ) we have @xmath431\right ] & = & \left[\left [ \eta \left ( \frac{\partial v_{x}}{\partial r } + \frac{\partial v_{r}}{\partial x } \right ) \right]\right ] \nonumber \\ & = & \left[\left [ \eta \left(k v_{x , n}'(kr ) + ik v_{r , n}(kr ) \right)\right]\right ] \nonumber \\ & = & \left(\eta^{i}-\eta^{o}\right ) \frac{ms}{r(\mu+1 ) } \left(\delta_{n = m+1 } - \delta_{n = m-1}\right).\end{aligned}\ ] ] dividing by @xmath426 gives @xmath432 - v_{x , n}^{\prime , o}(q ) - iv_{r , n}^{o}(q ) = \frac{\mu - 1}{\mu + 1 } \frac{ms}{q } \left(\delta_{n = m+1 } - \delta_{n = m-1}\right ) . \label{tang2.2}\ ] ] both eq .
( [ tang1.2 ] ) and eq .
( [ tang2.2 ] ) depend only on the integration constants , @xmath428 , @xmath31 , and @xmath140 . in calculating the diagonal velocity matrix elements @xmath433 ,
all right - hand sides are zero except in the normal stress equation ( [ norm2 ] ) , so we see that in this case the integration constants and thus the diagonal elements as well will scale as @xmath251 . for the off - diagonal elements
the only nonzero right - hand sides are from the tangential stress conditions , so the off - diagonal elements will only depend on @xmath428 , @xmath31 and @xmath140 , and not on @xmath95 or the magnitude of the viscosities . in the absence of the shear flow
this implies that the stability eigenvalues coming from the hydrodynamic terms scale as @xmath251 .
finally , the equations ( [ bcs2 ] ) can be solved analytically at @xmath105 . in this case
the general solutions to the modified bessel equation become @xmath434 and @xmath435 , and also the left - hand sides of eqs .
( [ tang1 ] ) and ( [ tang2 ] ) are zero at @xmath105 .
this simplifies the algebra considerably .
following the same procedure as outlined above , we find that the velocity matrix elements and therefore the stability eigenvalues at @xmath105 are diagonal and independent of the shear rate @xmath140 : @xmath436 with @xmath257 .
there seems to be some debate in the literature over whether the appropriate time scale to use in dilineating weak vs. strong - shear regimes in two - phase states should be @xmath1 or the time scale associated with the growth / fluctuations of the domains .
t. hashimoto ( private communication ) ; see also @xcite , @xcite , and t. hashimoto , t. takebe , and k. fujioka , in _ dynamics and patterns in complex fluids _ , edited by a. onuki and k. kawasaki ( springer , berlin , 1990 ) , p. 86 . | the stability of a long cylindrical domain in a phase - separating binary fluid in an external shear flow is investigated by linear stability analysis . using the coupled cahn - hilliard and stokes equations ,
the stability eigenvalues are derived analytically for long wavelength perturbations , for arbitrary viscosity contrast between the two phases .
the shear flow is found to suppress and sometimes completely stabilize both the hydrodynamic rayleigh instability and the thermodynamic instability of the cylinder against varicose perturbations , by mixing with nonaxisymmetric perturbations .
the results are consistent with recent observations of a `` string phase '' in phase - separating fluids in shear . epsf 2 |
symmetry braking phase transitions in the early universe have several cosmological consequences and provide an important link between particle physics and cosmology .
in particular , different types of topological objects may have been formed by the vacuum phase transitions after planck time kibble , v - s .
these include domain walls , cosmic strings and monopoles .
a global monopole is a spherical symmetric gravitational topological defect created by a phase transition of a system comprised by self - coupling scalar field , @xmath0 , whose original global @xmath1 symmetry is spontaneously broken to @xmath2 .
the matter fields play the role of an order parameter which outside the monopole s core acquires a non - vanishing value . the global monopole was first introduced by sokolov and starobinsky soko77 .
a few years later , the gravitational effects associated with a global monopole have been considered in ref .
@xcite , where the authors have found that for points far from the monopole s center , the geometry is similar to the black - hole with a solid angle deficit . neglecting the mass term we get the point - like global monopole spacetime with the metric tensor given by the following line element @xmath3where the parameter @xmath4 is smaller than unity and depends on the energy scale where the symmetry is broken .
it is of interest to note that the effective metric produced in superfluid @xmath5 by a monopole is described by the line element ( [ mmetric ] ) with the negative angle deficit , @xmath6 , which corresponds to the negative mass of the topological object @xcite .
the quasiparticles in this model are chiral and massless fermions . in quantum field theory
the non - trivial topology of the global monopole spacetime induces non - zero vacuum expectation values for physical observables .
the quantum effects due to the point - like global monopole spacetime on the matter fields have been considered for massless scalar m - l and fermionic @xcite fields , respectively .
the influence of the non - zero temperature on these polarization effects has been discussed in ref .
@xcite for scalar and fermionic fields .
moreover , the calculation of quantum effects on massless scalar field in a higher dimensional global monopole spacetime has also been developed in ref .
the quantum effects of a scalar field induced by a composite topological defect consisting a cosmic string on a @xmath7-dimensional brane and a @xmath8-dimensional global monopole in the transverse extra dimensions are investigated in ref .
the combined vacuum polarization effects by the non - trivial geometry of a global monopole and boundary conditions imposed on the matter fields are investigated as well . in this direction ,
the total casimir energy associated with massive scalar field inside a spherical region in the global monopole background has been analyzed in refs .
@xcite by using the zeta function regularization procedure .
scalar casimir densities induced by spherical boundaries have been calculated in refs .
@xcite to higher dimensional global monopole spacetime by making use of the generalized abel - plana summation formula @xcite .
more recently , using also this formalism , a similar analysis for spinor fields obeying mit bag boundary conditions has been developed in refs .
@xcite . in general , the quantum analysis of matter fields in global monopole spacetime
consider this object as been a point - like one .
because of this fact the renormalized vacuum expectation value of the energy - momentum tensor presents singularities at the monopole s center . of course , this kind of problem can not be expected in a realistic model .
so a procedure to cure this divergence is to consider the global monopole as having a non - trivial inner structure .
in fact , in a realistic point of view , the global monopoles have a characteristic core radius determined by the symmetry braking energy scale where the symmetry of the system is spontaneously broken . a simplified model for the monopole core
is presented in ref .
@xcite . in this model
the region inside the core is described by the de sitter geometry .
the vacuum polarization effects associated with a massless scalar field in the region outside the core of this model are investigated in ref .
in particular , it has been shown that long - range effects can take place due to the non - trivial core structure .
recently the quantum analysis of a scalar field in a higher - dimensional global monopole spacetime considering a general spherically symmetric model for the core , has been considered in ref .
@xcite . in the four - dimensional version of this model the induced electrostatic self - energy and self - force for a charged particle
are investigated in ref .
continuing in this direction , in the present paper we analyze the effects of global monopole core on properties of the fermionic quantum vacuum .
the most important quantities characterizing these properties are the vacuum expectation values of the energy - momentum tensor and the fermionic condensate .
though the corresponding operators are local , due to the global nature of the vacuum , the vacuum expectation values describe the global properties of the bulk and carry an important information about the structure of the defect core .
in addition to describing the physical structure of the quantum field at a given point , the energy - momentum tensor acts as the source of gravity in the einstein equations .
it therefore plays an important role in modelling a self - consistent dynamics involving the gravitational field .
the problem under consideration is also of separate interest as an example with gravitational polarization of the fermionic vacuum , where all calculations can be performed in a closed form .
the corresponding results specify the conditions under which we can ignore the details of the interior structure and approximate the effect of the global monopole by the idealized model . the exactly solvable models of this type are useful for testing the validity of various approximations used to deal with more complicated geometries , in particular in black - hole spacetimes .
the plan of this paper is as follows .
section [ sec : eigfunc ] presents the geometry of the problem and the eigenfunctions for a massive spinor field . by using these eigenfunctions , in section [ sec : extreg ] we evaluate the vacuum expectation values of the energy - momentum tensor and the fermionic condensate in the region outside the monopole core . in section sec :
flowerpot we consider the special case of the flower - pot model for the core and the vacuum expectation values are investigated in both exterior and interior regions .
various limiting cases are considered .
the main results of the paper are summarized in section [ sec : conc ] . in appendix sec : app1 we discuss the contribution of possible bound states into the vacuum expectation values of the energy - momentum tensor and show that in the flower - pot model there are no bound states . throughout
we use the system of units with @xmath9 .
in this section we analyze the eigenfunctions for a massive fermionic field in background of the global monopole geometry considering a nontrivial inner structure to the latter .
the explicit expression for the metric tensor in the region inside the monopole core is unknown and here we consider a general model for a four - dimensional global monopole with a core of radius @xmath10 , assuming that the geometry of the spacetime is described by two distinct metric tensors in the regions inside and outside the core . adopting this model we can learn under which conditions we can ignore the specific details for its core and consider the monopole as being a point - like object . in spherical coordinates
we will consider the corresponding line element in the interior region , @xmath11 , with the form @xmath12 in the region outside , @xmath13 , the metric tensor is given by the line element ( [ mmetric ] ) , where the parameter @xmath14 codifies the presence of the global monopole .
for an idealized point - like global monopole the geometry is described by line element ( [ mmetric ] ) for all values of the radial coordinate and it has singularity at the origin . in eq .
( [ metricinside ] ) the functions @xmath15 , @xmath16 , @xmath17 are continuous at the core boundary , consequently they satisfy the conditions @xmath18if there is no surface energy - momentum tensor on the bounding surface @xmath19 , the radial derivatives of these functions are continuous as well . when the surface energy - momentum tensor is present the junctions in the radial derivatives of the components of the metric tensor are expressed in terms of the surface energy density and stresses . introducing a new radial coordinate @xmath20 with the core center at @xmath21 , the angular part of line element ( [ metricinside ] )
is written in the standard minkowskian form . however , with this choice , in general , we will obtain non - standard form of the angular part in the exterior line element ( [ mmetric ] ) . in the model under consideration
we will assume that inside the core the spacetime geometry is regular .
in particular , from the regularity of the interior geometry at the core center one has the conditions @xmath22 and @xmath23 for @xmath24 .
we are interested in quantum effects of a spinor field propagating on background of the spacetime described by line elements ( [ mmetric ] ) and ( [ metricinside ] )
. the dynamics of the massive spinor field in curved spacetime is governed by the dirac equation @xmath25with the covariant derivative operator @xmath26here @xmath27 are the dirac matrices in curved spacetime , and @xmath28 is the spin connection given in terms of the flat @xmath29 matrices by the relation @xmath30 in the equations above @xmath31 is the vierbein satisfying the condition @xmath32 . in the region inside the monopole core
we choose the basis tetrad @xmath33where the rows of the matrix are specified by the index @xmath34 and the columns by the index @xmath35 . by using eq .
( [ e ] ) , for the non - vanishing components of the spin connection we find @xmath36where the prime denotes derivative with respect to the radial coordinate , @xmath37 and @xmath38 are the standard unit vectors along the three spatial directions in spherical coordinates , and @xmath39with @xmath40 being the pauli matrices . from the obtained results , for the combination entering in the dirac equation we have @xmath41after some intermediate steps , the dirac equation reads @xmath42with @xmath43 being the standard angular momentum operator . to find the vacuum expectation value ( vev ) of the energy - momentum tensor we need the corresponding eigenfunctions .
for the problem under consideration there are two types of eigenfunctions with different parities which we will distinguish by the index @xmath44 .
these functions are specified by the total angular momentum @xmath45 and its projection @xmath46 . assuming the time dependence in the form @xmath47
, the four - component spinor fields specified by the set of quantum numbers @xmath48 with @xmath49 , can be written in terms of two - component ones as @xmath50where @xmath51and @xmath52 are the standard spinor spherical harmonics ( see , for instance , ref .
the latter are eigenfunctions of the operator @xmath53 as shown below : @xmath54note that we have the relation @xmath55 substituting the function @xmath56 into the dirac equation above , and using for the flat dirac matrices the representation given in ref .
bere82 , for the radial functions we obtain a set of two coupled linear differential equations : @xmath57 in the region @xmath13 for the radial functions we have the solutions@xmath58 , \ ; \label{fext } \\
g_{\beta } ( r ) & = & g_{\beta } ^{\mathrm{(ex)}}(r)=-\frac{1}{\sqrt{r}}\frac{% n_{\sigma } k}{\omega + m}\left [ c_{1}j_{\nu _ { \sigma } + n_{\sigma } } ( kr)+c_{2}y_{\nu _ { \sigma } + n_{\sigma } } ( kr)\right ] , \label{gext}\end{aligned}\]]where @xmath59 and @xmath60 are the bessel and neumann functions of the order @xmath61by taking into account relation ( [ relsphhar ] ) , the corresponding eigenfunctions are written in the form@xmath62 in eqs .
( [ fext ] ) and ( [ gext ] ) , the integration constants @xmath63 and @xmath64 are determined from the matching condition with the interior solution . the regular solution to eqs .
( [ f1 ] ) , ( [ g1 ] ) in the interior region , @xmath65 , we will denote by@xmath66near the core center , @xmath24 , these functions behave like @xmath67 and @xmath68 , where the radial coordinate @xmath69 is introduced in the paragraph after formula ( [ uvbound ] ) . note that the functions @xmath70 and @xmath71 are solutions of the same equation and , hence , @xmath72the interior eigenfunctions have the form@xmath73from the continuity condition of the eigenfunctions on the surface @xmath19 , for the coefficients @xmath63 and @xmath64 in eq .
( [ fext ] ) one finds@xmath74 , \label{c1 } \\
c_{2 } & = & \frac{\pi } { 2}n_{\sigma } ka^{3/2}\left [ r_{1,n_{\sigma } } ( a , k)j_{\nu _ { \sigma } + n_{\sigma } } ( ka)-r_{2,n_{\sigma } } ( a , k)j_{\nu _ { \sigma } } ( ka)\right ] .
\label{c2}\end{aligned}\]]note that from eqs .
( [ f1 ] ) , ( [ g1 ] ) , for the values of the interior radial functions on the boundary of the core we have the following relations@xmath75where @xmath76 , @xmath77 , @xmath78 , and we have introduced the notation@xmath79 substituting the expressions for the coefficients @xmath63 and @xmath64 into the formulae for the radial eigenfunctions in the exterior region , one finds@xmath80with the notation@xmath81here and in what follows , for a function @xmath82 with @xmath83 we use the tilded notation defined by the formula@xmath84 \notag \\ & = & kaf_{\nu _ { \sigma } } ^{\prime } ( ka)-\left [ a\frac{r_{1,n_{\sigma } } ^{\prime } ( a , k)}{r_{1,n_{\sigma } } ( a , k)}+\frac{1}{2}au_{a}^{\prime } + ah_{a}^{\prime } -\frac{1}{2}\right ] f_{\nu _ { \sigma } } ( ka ) . \label{fnutilde}\end{aligned}\]]in
deriving the second form we have used the relation ( [ r21rel1 ] ) and the recurrence formula for the cylindrical functions .
the eigenfunctions are normalized by the condition@xmath85where @xmath29 is the determinant for the spatial metric and @xmath86 is understood as the kronecker delta symbol for the discrete components of the collective index @xmath87 and as the dirac delta function for the continuous ones .
as the normalization integral is divergent for @xmath88 , the main contribution comes from large values @xmath89 .
we may replace the cylindrical functions with the argument @xmath90 in eq .
( [ psiint ] ) by the corresponding asymptotic expressions . in this way
one finds@xmath91 ^{-1}. \label{r1a}\]]this relation determines the normalization coefficient for the interior region .
in addition to the modes with real @xmath92 , modes with purely imaginary @xmath93 can be present .
these modes correspond to the bound states .
the eigenfunctions for the bound states and their normalization are discussed in appendix [ sec : app1 ] .
in this section we consider the vevs for the energy - momentum tensor and the fermionic condensate in the region outside the global monopole core .
we expand the field operator in terms of the complete set of eigenfunctions @xmath94 : @xmath95 , \label{operatorexp}\]]where @xmath96 is the annihilation operator for particles , and @xmath97 is the creation operator for antiparticles , @xmath98 , for @xmath99 and @xmath100 , for @xmath101 . in order to find the vev for the operator of the energy - momentum tensor , we substitute the expansion ( operatorexp ) and the analog expansion for the operator @xmath102 into the corresponding expression for spinor fields , @xmath103\ .
\label{emtform}\]]by making use of the standard anticommutation relations for the annihilation and creation operators , for the vev one finds the following mode - sum formula @xmath104where @xmath105 is the amplitude for the corresponding vacuum state . substituting the eigenfunctions ( [ psiint ] ) into the mode - sum formula ( [ modesum ] ) , using the formula @xmath106and relation ( [ r1a ] ) ,
the vev of the energy - momentum tensor is presented in the form ( no summation over @xmath35 ) @xmath107 } { \tilde{j}_{\nu _ { \sigma } } ^{2}(x)+\tilde{y}_{\nu _ { \sigma } } ^{2}(x)% } \ .
\label{qext}\]]in formula ( [ qext ] ) we use the notations @xmath108 & = & -x\left [ ( \sqrt{x^{2}+m^{2}a^{2}}-ma)g_{\nu _ { \sigma } , 0}^{2}(x , y)\right . \
\notag \\ & & \left .
+ ( \sqrt{x^{2}+m^{2}a^{2}}+ma)g_{\nu _ { \sigma } , n_{\sigma } } ^{2}(x , y)% \right ] , \label{fnueps } \\
f_{\sigma \nu _ { \sigma } } ^{(1)}\left [ x , g_{\nu
_ { \sigma } , p}(x , y)\right ] & = & % \frac{x^{3}}{\sqrt{x^{2}+m^{2}a^{2}}}\left [ g_{\nu _ { \sigma } , 0}^{2}(x , y)+g_{\nu _
{ \sigma } , n_{\sigma } } ^{2}(x , y)\right ] \
-2f_{\sigma \nu _ { \sigma } } ^{(2)}\left [ x , g_{\nu _ { \sigma } , p}(x , y)\right ] , \label{fnup } \\
f_{\sigma \nu _ { \sigma } } ^{(\mu ) } \left [ x , g_{\nu _ { \sigma } , p}(x , y)\right ] & = & \frac{x^{3}(2\nu _ { \sigma } + n_{\sigma } ) } { 2y\sqrt{x^{2}+m^{2}a^{2}}}% g_{\nu _ { \sigma } , 0}(x , y)g_{\nu _ { \sigma } , n_{\sigma } } ( x , y),\;\mu = 2,3 , \label{fnupperp}\end{aligned}\]]and the function @xmath109 is defined by eq .
( [ gnumu2 ] ) . the expression on the right of eq .
( [ qext ] ) is divergent .
it may be regularized introducing a cutoff function @xmath110 with the cutting parameter @xmath35 which makes the divergent expressions finite and satisfies the condition @xmath111 for @xmath112 . after the renormalization
the cutoff function is removed by taking the limit @xmath113 . in the discussion below
we will implicitly assume that the corresponding expressions are regularized .
the parts in the vevs induced by the non - trivial structure of the core are finite and do not depend on the regularization scheme used .
to find the part in the vev of the energy - momentum tensor induced by the non - trivial core structure , we subtract the corresponding components for the point - like monopole geometry .
the latter are given by the expressions which are obtained from eq .
( [ qext ] ) replacing the integrand by @xmath114 $ ] ( see ref .
@xcite ) . in order to evaluate the corresponding difference we use the relation @xmath115 } { \tilde{j}_{\nu _ { \sigma } } ^{2}(x)+\tilde{y}_{\nu _ { \sigma } } ^{2}(x)% } -f_{\sigma \nu _ { \sigma } } ^{(\mu ) } \left [ x , j_{\nu _ { \sigma } } ( xr / a)\right ] = -\frac{1}{2}\sum_{s=1,2}\frac{\tilde{j}_{\nu _ { \sigma } } ( x)}{\tilde{h}_{\nu _ { \sigma } } ^{(s)}(x)}f_{\sigma \nu _ { \sigma } } ^{(\mu ) } \left [ x , h_{\nu _ { \sigma } } ^{(s)}(xr / a)\right ] \ , \label{relf}\]]where @xmath116 , @xmath117 , are the hankel functions .
this allows to present the vacuum energy - momentum tensor components in the form@xmath118where @xmath119 is the vev for the point - like monopole and the part ( no summation over @xmath35 ) @xmath120 \ , \label{qrbout}\]]is induced by the non - trivial core structure . in formula ( [ qrbout ] )
, the integrand of the @xmath121 ( @xmath122 ) term exponentially decreases in the upper ( lower ) half of the complex plane @xmath123 .
consequently , we rotate the integration contour on the right of this formula by the angle @xmath124 for @xmath125 and by the angle @xmath126 for @xmath122 .
this leads to the representation@xmath127 , \label{tmunuc2}\end{aligned}\]]where @xmath128 .
first of all let us consider the part of the integral over the interval @xmath129 . by using the standard properties of the hankel functions
it can be easily seen that in this interval one has @xmath130 = e^{-i\pi \lambda } f_{\sigma \nu _ { \sigma } } ^{(\mu ) } \left [ e^{-\pi i/2}x , h_{\nu _ { \sigma } } ^{(2)}(e^{-\pi i/2}xr / a)\right ] .
\label{fnusigrel1}\]]further , from equations ( [ f1 ] ) and ( [ g1 ] ) it follows that the interior solution can be written as @xmath131 . on the base of this observation for the combination entering into the tilted notations in eq .
( [ tmunuc2 ] ) one finds@xmath132now , it can be seen that for these values @xmath123 one has@xmath133combining relations ( [ fnusigrel1 ] ) and ( [ reljh ] ) , we see that the part of the integral in eq .
( [ tmunuc2 ] ) over the interval @xmath134 vanishes . to simplify the part of the integral over the interval @xmath135
, we note that for the functions with different parities the following relation takes place:@xmath130 = -e^{-i\pi \lambda } f_{\sigma ^{\prime } \nu _ { \sigma ^{\prime } } } ^{(\mu ) } \left [ e^{-\pi i/2}x , h_{\nu _ { \sigma ^{\prime } } } ^{(2)}(e^{-\pi i/2}xr / a)\right ] , \label{frel3}\]]where @xmath136 , and @xmath137 .
further we note that the functions @xmath138 and @xmath139 satisfy the same equation and , hence , @xmath140 . by using relations ( [ r21rel1 ] ) and ( [ r21rel2 ] ) ,
now it can be seen that@xmath141from this formula , by taking into account that @xmath142 , we find the relation@xmath143combining formulae ( [ frel3 ] ) and ( [ reljh3 ] ) , we see that the different parities give the same contribution to the vev of the energy - momentum tensor . by taking into account this result and introducing the modified bessel functions , for the core - induced part in the vev we find ( no summation over @xmath144)@xmath145 .
\label{tmunu6}\]]here for a given function @xmath146 we use the notations@xmath147 & = & \left ( \frac{m^{2}}{% x^{2}}-1\right ) \left [ \left ( 1+\frac{i\eta _ { s}m}{\sqrt{x^{2}-m^{2}}}% \right ) f_{\nu } ^{2}(y)\right .
\notag \\ & & \left .
-\left ( 1-\frac{i\eta _ { s}m}{\sqrt{x^{2}-m^{2}}}\right ) f_{\nu + 1}^{2}(y)\right ] , \label{fnu0 } \\ f_{\nu } ^{(1,\eta _ { s})}\left [ x , f_{\nu } ( y)\right ] & = & f_{\nu } ^{2}(y)-f_{\nu + 1}^{2}(y)-\lambda _ { f}\frac{2\nu + 1}{y}f_{\nu } ( y)f_{\nu + 1}(y ) , \label{fnu1 } \\
f_{\nu } ^{(\mu , \eta _ { s})}\left [ x , f_{\nu } ( y)\right ] & = & \lambda _ { f}\frac{% 2\nu + 1}{2y}f_{\nu } ( y)f_{\nu + 1}(y),\;\mu = 2,3 , \label{fnu2}\end{aligned}\]]with @xmath148 ( the function @xmath149 $ ] with @xmath150 will be used below ) , and the barred notation is defined by the formula@xmath151 f(x ) .
\label{fetasbar}\]]it can be checked that the core - induced part in the vev of the energy - momentum tensor obeys the continuity equation @xmath152 , which for the geometry under consideration takes the form@xmath153 in addition , for a massless spinor field this part is traceless and the trace anomaly is contained in the point - like monopole part only
. the fermionic condensate can be found from the formula for the vev of the energy - momentum tensor by making use of the relation @xmath154 .
the condensate is presented in the form of the sum of idealized point - like monopole and core - induced parts:@xmath155by taking into account formula ( [ tmunu6 ] ) for the components of the energy - momentum tensor , for the part coming from the non - trivial core structure one finds@xmath156 .
\label{condens1}\end{aligned}\]]this formula may also be derived directly from the mode - sum formula @xmath157 with the eigenfunctions ( [ psiex ] ) . in deriving formulae ( [ tmunu6 ] ) and ( [ condens1 ] )
we have assumed that no bound states exist . in appendix
[ sec : app1 ] we show that these formulae are valid also in the case when the bound states are present . for @xmath13 the core - induced parts in the vevs of the energy - momentum tensor and the fermionic condensate , given by eqs .
( [ tmunu6 ] ) and ( condens1 ) , are finite and the renormalization is necessary for the point - like monopole parts only .
of course , we could expect this result as in the region @xmath13 the local geometry is the same in both models and , hence , the divergences are the same as well . as it has been already mentioned , if there is no surface energy - momentum tensor on the bounding surface @xmath19 , then the radial derivatives of the metric are continuous and , hence , in eqs .
( [ fnutilde ] ) and ( fetasbar ) one has @xmath158 , @xmath159 . in models with an additional infinitely thin spherical shell located at @xmath19
, these quantities are related to the components of the corresponding surface energy - momentum tensor @xmath160 . denoting by @xmath161 the normal to the shell normalized by the condition @xmath162 and assuming that it points into the bulk on both sides , from the israel matching conditions one has@xmath163 in this formula the curly brackets denote summation over each side of the shell , @xmath164 is the induced metric on the shell , @xmath165 its extrinsic curvature and @xmath166 . for the region @xmath167
one has @xmath168 and the non - zero components of the extrinsic curvature are given by the formulae@xmath169the corresponding expressions for the region @xmath170 are obtained by taking in these formulae @xmath171 , @xmath172 and changing the signs for the components of the extrinsic curvature tensor . now from the matching conditions ( [ matchcond ] ) we find@xmath173note that the combination in the square brackets in eqs .
( [ fnutilde ] ) and ( [ fetasbar ] ) is related to the surface energy - momentum tensor by the formula@xmath174where @xmath175 is the trace of the surface energy - momentum tensor .
in this section , as an application of the general results given above we consider a simple example of the core model assuming that the spacetime inside it is flat . the corresponding model for the cosmic string core was considered in refs .
@xcite and for the global monopole core in ref .
@xcite . following these papers
we will refer to this model as flower - pot model .
taking in the region inside the core @xmath171 , from the zero curvature condition one finds @xmath176 .
the value of the constant is found from the continuity condition for the function @xmath17 at the boundary which gives @xmath177 .
hence , in the flower - pot model the interior line element has the form@xmath178 ^{2}(d\theta ^{2}+\sin ^{2}\theta d\phi ^{2 } ) .
\label{intflow}\]]in terms of the radial coordinate @xmath89 the origin is located at @xmath179 . from the matching conditions ( [ matchcond2 ] ) we find the corresponding surface energy - momentum tensor with the non - zero components and the trace given by@xmath180the
corresponding surface energy density is positive for the global monopole with @xmath181 .
we will consider the vevs in the exterior and interior regions separately . in the region inside the core
the radial eigenfunctions regular at the origin are the functions@xmath182where@xmath183and @xmath184 is the standard minkowskian radial coordinate , @xmath185 .
note that for an interior minkowskian observer the radius of the core is @xmath186 .
the normalization coefficient @xmath187 is found from the condition ( [ r1a]):@xmath188with the tilted notation for the cylindrical functions@xmath189 f_{\nu _ { \sigma } } ( x ) . \label{ftildeflow}\]]note
that @xmath190 for @xmath191 . in this case for the barred notation in eq .
( [ fetasbar ] ) one has @xmath192 @xmath193 . hence , in the flower - pot model the part in the energy - momentum tensor due the non - trivial structure of the core is given by the formula @xmath194 , \label{tmunu8}\end{aligned}\]]where we have introduced the notation@xmath195 g(x ) .
\label{cfg}\]]in formula ( [ tmunu8 ] ) , @xmath196 = f_{\nu } ^{(\mu , \eta _ { s})}\left [ x , f_{\nu } ( y)\right ] $ ] for @xmath197 , and @xmath198 = \left ( \frac{% m^{2}}{x^{2}}-1\right ) \left [ f_{l/\alpha -1/2}^{2}(y)-f_{l/\alpha + 1/2}^{2}(y)\right ]
. \label{gnu0}\]]note that in terms of notation ( [ cfg ] ) one has @xmath199for @xmath191 we have @xmath200 and the vevs vanish . in the similar way , for the fermionic condensate from eq .
( [ condens1 ] ) we find@xmath201 .
\label{condfp}\end{aligned}\]]for a massless fermionic field the core - induced part in the condensate vanishes . from formulae ( [ tmunu8 ] ) and ( [ condfp ] )
it can be seen that for fixed values of @xmath89 and @xmath202 , in the limit @xmath203 the core - induced parts vanish as @xmath204 .
note that by using the recurrence relations for the modified bessel functions , the functions @xmath205 in these formulae can also be written in the form@xmath206with @xmath207 , @xmath208 and @xmath150 , @xmath148 .
in particular , from formula ( [ cif ] ) with @xmath209 it follows that @xmath210 . as we will show in appendix sec : app1 , this means that there are no bound states in the flower - pot model .
the core - induced part in the vevs of the energy - momentum tensor and the fermionic condensate given by formulae ( [ tmunu8 ] ) and ( [ condfp ] ) , are finite for @xmath13 and diverge on the core boundary .
surface divergences are well - known in quantum field theory with boundaries and are investigated for various boundary geometries and fields .
to find the corresponding asymptotic behavior for the points near the boundary , we note that in this region the main contribution comes from large values of @xmath211 . by using the uniform asymptotic expansions for the modified bessel functions for large values of the order ( see , for instance , @xcite ) , to the leading order we find@xmath212as the parts corresponding to the geometry of the point - like global monopole are finite at @xmath19 , from here we conclude that near the core the vevs are dominated by the core - induced parts . at large distances from the core , @xmath213 ,
in the case of a massless fermionic field we introduce a new integration variable @xmath214 and expand the integrand over @xmath215 .
the main contribution comes from the lowest mode @xmath216 and to the leading order one has@xmath217 .
\label{tmunulargedist}\]]note that for a massless field the integrand does not depend on @xmath89 .
the integrals in eq .
( [ tmunulargedist ] ) may be evaluated by using the formula for the integrals involving the product of the macdonald functions ( see ref .
@xcite ) . for a massive field , assuming @xmath218 , the main contribution into the integral over @xmath123 comes from the lower limit of the integration . replacing the functions @xmath219 by the corresponding asymptotic formulae for large values of the argument , to the leading order
we obtain the following estimates@xmath220as we see , in this limit the core - induced energy density and the radial stress are suppressed by the factor @xmath221 with respect to the azimuthal stress . for @xmath222
the solid angle for the exterior geometry is small and the corresponding scalar curvature is large . in this limit
we replace the modified bessel function containing in the index @xmath223 by the corresponding uniform asymptotic expansions for large values of the order and the functions containing in the argument @xmath224 by the expansions for small values of the argument . in this way
it can be seen that the main contribution into the vevs comes from the mode @xmath225 and the vevs are suppressed by the factor @xmath226 $ ] . in figure [ fig1 ]
we have plotted the dependence of the core - induced energy density ( full curves ) and radial stress ( dashed curves ) for a massless fermionic field as functions on the scaled radial coordinate @xmath227 for @xmath228 and @xmath229 .
the azimuthal stresses are found from the zero trace condition . in figure [ fig2 ] the same quantities evaluated for @xmath230
are presented as functions on the parameter @xmath14 .
now let us consider the vacuum polarization effects inside the core for the flower - pot model .
the corresponding eigenfunctions have the form given by eq .
( [ psibetin ] ) where the functions @xmath70 and @xmath231 are defined by formulae ( [ rlflow ] ) . substituting the eigenfunctions into the mode - sum formula , for the corresponding energy - momentum tensor one finds ( no summation over @xmath35)@xmath232 } { j_{\lambda _ { \sigma } } ^{2}(\alpha x)\left [ \tilde{j}_{\nu _ { \sigma } } ^{2}(x)+\tilde{y}_{\nu _ { \sigma } } ^{2}(x)\right ] } .
\label{tmunuinfl}\]]to find the renormalized vev of the energy - momentum tensor we need to evaluate the difference between the vev given by eq .
( [ tmunuinfl ] ) and the corresponding vev for the minkowski bulk:@xmath233the appropriate form for the minkowskian part is obtained from eq .
( qext ) taking @xmath191 , replacing the integrand by @xmath234 $ ] and @xmath235 . as a result for the subtracted vev one finds@xmath236 \notag \\ & & \times \left [ \frac{j_{\lambda _ { \sigma } } ^{-2}(\alpha x)/\alpha } { \tilde{j% }
_ { \nu _ { \sigma } } ^{2}(x)+\tilde{y}_{\nu _ { \sigma } } ^{2}(x)}-\frac{\pi ^{2}}{% 4}\right ] .
\label{tmunusub2}\end{aligned}\]]the integral in this formula is slowly convergent and the integrand is highly oscillatory . in order to transform the expression for the subtracted vev of the energy - momentum tensor into more convenient form
, we note that the following identity takes place@xmath237substituting eq .
( [ identflow2new ] ) into formula ( [ tmunusub2 ] ) , we rotate the integration contour in the complex plane @xmath123 by the angle @xmath124 for @xmath121 and by the angle @xmath126 for @xmath122 . under the condition @xmath238
the contributions from the semicircles with the radius tending to infinity vanish .
the integrals over the segments @xmath239 and @xmath240 of the imaginary axis cancel out and after introducing the modified bessel functions we can see that different parities give the same contribution .
consequently , the renormalized vev for the energy - momentum tensor can be presented in the form ( no summation over @xmath35)@xmath241 , \label{tmunusub3}\end{aligned}\]]where the functions @xmath196 $ ] are the same as in eq .
( [ tmunu8 ] ) with @xmath150 in eqs .
( [ fnu1 ] ) and ( [ fnu2 ] ) .
note that similar to eq .
( [ cif ] ) , the function in the numerator of the integrand is also presented in the form@xmath242for @xmath191 one has @xmath243 and , as we could expect , the vev of the energy - momentum tensor vanishes . it can be explicitly checked that the components of the tensor given by formula ( [ tmunusub3 ] ) satisfy the continuity equation ( [ conteq ] ) and this tensor is traceless for a massless field . in the way similar to that for the exterior region , for the renormalized fermionic condensate inside the core we find@xmath244 .
\label{condint}\end{aligned}\ ] ] the vevs given by formulae ( [ tmunusub3 ] ) and ( [ condint ] ) are finite for @xmath245 and diverge on the core boundary .
for the points near the boundary the main contribution comes from large @xmath211 and we replace the modified bessel functions by the corresponding uniform expansions for large values of the order . in this way it can be seen that the leading terms in the asymptotic expansion with respect to the distance from the boundary are given by the formulae@xmath246comparing these results with the corresponding formulae for the exterior region , we see that near the core boundary the energy density and azimuthal stresses have the same signs inside and outside the core , whereas the radial stresses have opposite signs .
near the core center the contribution of the summand with a given @xmath211 behaves like @xmath247 and the main contribution comes from the lowest mode @xmath225 with the leading term ( no summation over @xmath35)@xmath248where we have introduced notations@xmath249 in the case of a massless field , formula ( [ nearcenter ] ) also gives the behavior of the vacuum energy - momentum tensor in the limit when the core radius is large and @xmath69 is fixed , @xmath250 . in the same limit , for a massive field , assuming @xmath251 , we replace in eq .
( tmunusub3 ) the modified bessel functions containing in the argument @xmath252 by the corresponding asymptotic expressions for large values of the argument . by taking into account that the main contribution into the integral comes from the lower limit of the integration , to the leading order we have ( no summation over @xmath35)@xmath253 , \label{t00radlarge } \\
\langle t_{\mu } ^{\nu } \rangle _ { \mathrm{ren } } & \approx & \frac{\delta _ { \mu } ^{\nu } ( 1-\alpha ) } { 8\pi \alpha ^{4}a^{3}\tilde{r}}\sqrt{\pi \alpha am}% e^{-2\alpha am}\sum_{l=1}^{\infty } l^{2}g_{l-1/2}^{(\mu ) } \left [ m , i_{l-1/2}(m\tilde{r})\right ] , \label{tmnradlarge}\end{aligned}\]]with @xmath254 . in this case
the energy density is suppressed with respect to the vacuum stresses by an additional factor @xmath255 . for small values of the parameter @xmath14 assuming that the core radius @xmath256 for an internal minkowskian observer is fixed , we replace the functions @xmath257 in eq .
( [ tmunusub3 ] ) by the corresponding uniform asymptotic expansion for large values of the order .
the leading term is obtained by making use of the replacements@xmath258^{\prime } + \sqrt{l^{2}+y^{2}}+1/2 , \label{replace}\ ] ] in the integrand of eq .
( [ tmunusub3 ] ) with @xmath259 and @xmath260 . as a result , in this limit the vevs behave like @xmath261 . due to the factor @xmath262 in the volume element the corresponding global quantities , such as total energy vanish as @xmath14 . in figure [ fig3 ]
we have plotted the dependence of the renormalized vacuum energy density ( full curves ) and radial stress ( dashed curves ) for a massless fermionic field as functions on @xmath263 for @xmath264 and @xmath229 . in figure [ fig4 ] the same quantities evaluated for @xmath265
are presented as functions on the parameter @xmath266 .
in the present paper we have considered the polarization of the fermionic vacuum by the gravitational field of the global monopole wit a non - trivial core structure .
the previous investigations in this direction are concerned with the idealized point - like model , where the curvature has singularity at the origin . in a realistic point of view
, the global monopole has a characteristic core radius determined by the symmetry braking energy scale . for a general spherically symmetric static model of the core with finite thickness
, we have evaluated the vevs of the energy - momentum tensor and the fermionic condensate for a massive spinor field .
these quantities are among the most important characteristics of the vacuum properties , which carry an information about the core structure . in the region outside the core
we have presented the vevs as a sum of two contributions .
the first one corresponds to the geometry of a point - like global monopole and the second one is induced by the non - trivial structure of the monopole core . in the general spherically symmetric static model for the core
, we have derived closed analytic expressions for the core - induced parts given by formula ( tmunu6 ) for the energy - momentum tensor and by formula ( [ condens1 ] ) for the fermionic condensate , where the properties of the core are codified by the coefficient in the square brackets in the notation ( [ fetasbar ] ) . in appendix
[ sec : app1 ] we show that the formulae for the core - induced parts in the vevs of the energy - momentum tensor and fermionic condensate are valid also in the case when bound states are present . for points away from the core boundary these parts are finite and the renormalization is reduced to that for the point - like monopole geometry . of course
, we could expect this result as in the model under consideration the exterior geometry is the same as that for the point - like global monopole .
for the points on the boundary the vevs contain surface divergences well - known in quantum field theory with boundaries . as an example of the application of the general results , in section sec : flowerpot we have considered a simple model with a flat spacetime inside the core .
the corresponding model for the cosmic string core is known in literature as flower - pot model and here we use the same terminology for the global monopole . to have matching between the exterior and interior metrics , in this model
we need the surface energy - momentum tensor located on the boundary of the core and having components given by eq .
( surfemtflow ) . in the model with solid angle deficit ( @xmath181 )
the corresponding surface energy density is positive .
the core - induced parts of the exterior vevs in the flower - pot model are obtained from the general results taking as the interior radial functions in the eigenmodes the functions ( [ rlflow ] ) .
these parts are given by formula ( [ tmunu8 ] ) for the energy - momentum tensor and by formula ( [ condfp ] ) for the fermionic condensate .
we have investigated the core - induced parts in various asymptotic regions of the parameters . in the limit when the core radius tends to zero , @xmath203 , for fixed values @xmath89 and @xmath202 ,
these parts behave like @xmath204 . for points near the core boundary
the leading terms in the corresponding asymptotic expansions are given by formulae ( nearcore ) , ( [ nearcorecond ] ) . in this region
the total vevs are dominated by the core - induced parts . at large distances from the core
these parts tend to zero as @xmath267 for a massless field and are exponentially suppressed by the factor @xmath268 for a massive field .
we have also investigated the limit of strong gravitational fields corresponding to small values of the parameter @xmath14 . in this limit the main contribution into the vevs comes from the lowest mode @xmath225 and the vevs are suppressed by the factor @xmath226 $ ] . for the flower - pot model
we have also investigated the vevs of the energy - momentum tensor and the fermionic condensate inside the core .
though the corresponding spacetime geometry is monkowskian , the non - trivial topology of the exterior region induces vacuum polarization effects in this region as well .
the renormalization is achieved by the subtraction from the mode - sums the corresponding quantities for the minkowski spacetime . by making use of identity ( [ identflow2new ] ) , after an appropriate deformation of the integration contour , we have presented the renormalized vev of the energy - momentum tensor in the form ( [ tmunusub3 ] ) and the fermionic condensate in the form ( [ condint ] ) .
these quantities are finite for strictly interior points and diverge on the boundary of the core with the leading divergences given by formulae ( [ nearint ] ) , ( nearboundint ) . in particular , near the core boundary the energy density and azimuthal stresses have the same signs inside and outside the core , whereas the radial stresses have opposite signs . near the core center
the main contribution comes from the mode @xmath225 and the vevs tend to a finite limiting value with isotropic vacuum stresses .
although the exact behavior for the fermionic field is unknown for a realistic model of the global monopole spacetime , the flower - pot model considered here presents some expected results as , for example , finite vacuum polarization effects at the monopole s center . for large values of the core radius
the renormalized vev of the energy - momentum tensor inside the core vanishes as @xmath269 for a massless field and as @xmath270 for a massive one . in the limit @xmath271 , assuming that the core radius @xmath186 for an internal minkowskian observer is fixed , the vacuum densities in the interior region behave as @xmath261 .
note that in this paper we have considered quantum vacuum effects in a prescribed background , i.e. the gravitational back - reaction of quantum effects is not taken into account .
this back - reaction could have important effects on the dynamical evolution of the bulk model .
we do not consider this extension of the theory , but note that the results presented here constitute the starting point for such investigations .
aas was supported by pve / capes program and in part by the armenian ministry of education and science grant no . 0124 .
erbm thanks conselho nacional de desenvolvimento cientfico e tecnolgico ( cnpq ) and fapesq - pb / cnpq ( pronex ) for partial financial support .
in this appendix we consider the changes in the procedure described in the main text when bound states are present .
for these states the quantity @xmath92 is purely imaginary , @xmath272 , and the corresponding exterior eigenfunctions have the form@xmath273with @xmath274 . to have a stable ground state we will assume that @xmath275 . from the continuity of the eigenfunctions at @xmath19 one has@xmath276excluding from these relations the normalization coefficient @xmath277 we see that for possible bound states @xmath29 is a solutions of the equation@xmath278with the barred notation from eq .
( [ fetasbar ] ) .
the coefficient @xmath277 in eq .
( [ psibound ] ) is found from the normalization condition ( [ normcond ] ) . to derive a formula for the normalization integral , we rewrite equations ( [ f1 ] ) , ( [ g1 ] ) in terms of the functions @xmath279 and @xmath280 and differentiate both equations with respect to @xmath281 .
further we multiply the first equation by @xmath282 , the second one by @xmath283 and add them . combining the resulting equation with eqs .
( [ f1 ] ) , ( [ g1 ] ) , it can be seen that the following relation takes place@xmath284integrating this relation we obtain the formula@xmath285=e^{u+2h } \left [ g_{\beta } ( r)\frac{\partial f_{\beta } ( r)}{\partial \omega } -f_{\beta } ( r)\frac{\partial g_{\beta } ( r)}{\partial \omega } \right ] , \label{normint}\]]where @xmath286 is the value of the radial coordinate corresponding to the center of the core . by using this formula , in the case of bound states for the normalization integral in eq .
( [ normcond ] ) one finds@xmath287=e^{u+2h}\left [ g_{\beta } ( r)\frac{\partial f_{\beta } ( r)}{\partial \omega } -f_{\beta } ( r)\frac{\partial g_{\beta } ( r)}{\partial \omega } \right ] _ { r = a+}^{r = a-}. \label{normint}\]]the expression on the right can be further simplified by using the continuity of the eigenfunctions on the boundary of the core . in this way for the normalization coefficient one
finds the formula@xmath288 as a result , for the contribution of the bound state with @xmath272 to the vev of the energy - momentum tensor we have the formula ( no summation over @xmath35)@xmath289 , \label{tmunubound}\]]where @xmath290 and we have introduced the notations@xmath291 & = & \left ( 1-\frac{m^{2}}{\gamma ^{2}}% \right ) \left [ \left ( \frac{m}{\sqrt{m^{2}-\gamma ^{2}}}-1\right ) k_{\nu _ { \sigma } } ^{2}(y ) \right .
\nonumber \\ & & \left .
+ \left ( \frac{m}{\sqrt{m^{2}-\gamma ^{2}}}+1\right ) k_{\nu _ { \sigma } + n_{\sigma } } ^{2}(y)\right ] , \nonumber \\ b^{(1)}[\gamma , k_{\nu _ { \sigma } } ( y ) ] & = & k_{\nu _ { \sigma } } ^{2}(y)+\frac{% 2ln_{\sigma } } { \alpha y}k_{\nu _ { \sigma } } ( y)k_{\nu _ { \sigma } + n_{\sigma } } ( y)-k_{\nu _ { \sigma } + n_{\sigma } } ^{2}(y ) , \label{bmu } \\
b^{(\mu ) } [ \gamma , k_{\nu _ { \sigma } } ( y ) ] & = & -\frac{ln_{\sigma } } { \alpha y}% k_{\nu _ { \sigma } } ( y)k_{\nu _ { \sigma } + n_{\sigma } } ( y),\;\mu = 2,3 .
\nonumber\end{aligned}\]]in deriving eq .
( [ tmunubound ] ) we have used the relation @xmath292 , which directly follows from the wronskian relation for the modified bessel functions in combination with eq .
( [ appbseq ] ) . in the case when several bound states are present the sum of their separate contributions should be taken .
the vev of the energy - momentum tensor is the sum of the part coming from the modes with real @xmath92 given by eq .
( [ qrbout ] ) and of the part coming from the bound states given by eq .
( [ tmunubound ] ) . in order to transform the first part we again rotate the integration contour in eq .
( [ qrbout ] ) by the angle @xmath124 for @xmath121 and by the angle @xmath126 for @xmath122 .
but now we should take into account that the integrand has poles at @xmath293 which are zeroes of the functions @xmath294 in accordance with eq .
( [ appbseq ] ) . rotating the integration contour
we will assume that the pole @xmath295 , @xmath117 , on the imaginary axis is avoided by the semicircle @xmath296 in the right half plane with small radius @xmath297 and with the center at this pole .
the integration over these semicircles will give an additional contribution@xmath298 . \label{poles}\]]by evaluating the integrals in this formula it can be seen that this term cancels the contribution ( [ tmunubound ] ) coming from the corresponding bound state . hence , we conclude that the formulae given above for the core - induced parts in the vevs are valid in the case of the presence of bound states as well . in the flower - pot model the equation ( [ appbseq ] ) for the bound states takes the form@xmath299 in section [ sec : flowerpot ] we have shown that the function on the left of this equation is always negative and , hence , in the flower - pot model no bound states exist . | we study the vacuum polarization effects associated with a massive fermionic field in a spacetime produced by a global monopole considering a nontrivial inner structure for it . in the general case of the spherically symmetric static core with finite support we evaluate the vacuum expectation values of the energy - momentum tensor and the fermionic condensate in the region outside the core .
these quantities are presented as the sum of point - like global monopole and core - induced contributions .
the asymptotic behavior of the core - induced vacuum densities are investigated at large distances from the core , near the core and for small values of the solid angle corresponding to strong gravitational fields . as an application of general results the flower - pot model for the monopole s core
is considered and the expectation values inside the core are evaluated .
pacs number(s ) : 03.07.+k , 98.80.cq , 11.27.+d |
recall the classical linear transient 2nd - order convection - diffusion problem for an unknown scalar function @xmath1 on a bounded domain @xmath2 : @xmath3{rcll } \partial_{t}u - \varepsilon \delta u + { \boldsymbol \beta}\cdot \operatorname{\bf grad } u & = & f & \text{in } \omega\;,\\ u & = & 0 & \text{on } \partial\omega\;,\\ u(0 ) & = & u_{0}\;. \end{array}\end{gathered}\ ] ] here , and in the remainder of the paper , @xmath4 stands for a smooth vector field . for @xmath5
we encounter a singularly perturbed boundary value problem , whose _ stable _ discretization has attracted immense attention in numerical analysis , see @xcite and the many references cited therein .
the boundary value problem turns out to be a member of a larger family of 2nd - order boundary value problems ( bvp ) , which can conveniently be described using the _ calculus of differential forms_. it permits us to state the generalized transient 2nd - order convection - diffusion problems as @xmath6{rcll } \ast \partial_t \omega(t ) - \varepsilon ( -1)^{l } d\ast d \omega(t ) + \ast l_{\boldsymbol \beta } \omega ( t ) & = & \varphi \quad & \text{in } \omega\subset\mathbb{r}^{n}\;,\\ \imath^{\ast}\omega & = & 0 & \text{on } \partial\omega\;,\\ \omega(0 ) & = & \omega_{0}\;. \end{array}\end{gathered}\ ] ] these are bvps for an unknown time dependent @xmath7-form @xmath8 , @xmath9 , on the domain @xmath10 .
the symbol @xmath11 stands for a hodge operator mapping an @xmath7-form to an @xmath12-form , and @xmath13 denotes the exterior derivative .
dirichlet boundary conditions are imposed via the trace @xmath14 of the @xmath7-form @xmath15 .
we refer to ( * ? ? ?
* sect . 2 ) and ( * ? ? ?
2 ) for more details and an introduction to the calculus of differential forms in the spirit of this article . by @xmath16
we denote the lie derivative of @xmath15 for the given velocity field @xmath17 ( * ? ? ?
* ch . 5 ) .
it provides the convective part of the differential operator in .
details will be given in sect .
[ sec : extr - contr - lie ] below . in two and three dimensions
differential forms can be modelled by means of functions and vector fields through so - called vector proxies , see ( * ? ? ?
7 ) , ( * ? ? ?
* table 2.1 ) , and ( * ? ? ?
* table 2.1 ) to indicate that a form is mapped to its corresponding euclidean vector proxy ] .
thus , for @xmath18 , in the case of @xmath11 induced by the euclidean metric on @xmath19 , the following vector analytic incarnations of are obtained .
the solution @xmath7-form @xmath15 is replaced with an unknown vector field @xmath20 or function @xmath21 , _ cf . _
* sect . 2 ) .
* for @xmath22 we recover . * in the case @xmath23 we arrive at a convection - diffusion problem @xmath24{rcll } \partial_{t}{{\mathbf{u}}}+ \varepsilon{\operatorname{\bf curl}}{\operatorname{\bf curl}}{{\mathbf{u}}}- { \boldsymbol \beta}\times { \operatorname{\bf curl}}{{\mathbf{u}}}+ \operatorname{\bf grad}({{\mathbf{u}}}\cdot{\boldsymbol \beta } ) & = & { { \mathbf{f } } } & \text{in } \omega\;,\\ { { \mathbf{u}}}\times{{\mathbf{n}}}&= & 0 & \text{on } \partial\omega\;,\\ { { \mathbf{u}}}(0 ) & = & { { \mathbf{u}}}_{0}\;. \end{array } \end{gathered}\ ] ] * the corresponding boundary value problems for vector proxies of 2-forms read @xmath25{rcll } \partial_{t}{{\mathbf{u}}}-\varepsilon{\operatorname{\bf grad}}{{\mathsf{div}}}{{\mathbf{u}}}+ { \boldsymbol \beta}({{\mathsf{div}}}{{\mathbf{u } } } ) + \operatorname{\bf curl}({{\mathbf{u}}}\times{\boldsymbol \beta } ) & = & { { \mathbf{f } } } & \text{in } \omega\;,\\ { { \mathbf{u}}}\cdot{{\mathbf{n}}}&= & 0 & \text{on } \partial\omega\;,\\ { { \mathbf{u}}}(0 ) & = & { { \mathbf{u}}}_{0}\;. \end{array } \end{gathered}\ ] ] * for @xmath26 the diffusion term vanishes and we end up with pure convection problems for a scalar density @xmath21 @xmath27{rcll } \partial_{t}u + { { \mathsf{div}}}({\boldsymbol \beta}u ) & = & f & \text{in } \omega\;,\\ { { \mathbf{u}}}(\cdot,0 ) & = & { { \mathbf{u}}}_{0}\ ; , \end{array } \end{gathered}\ ] ] where we assume @xmath28 on @xmath29 though equivalent to the vector proxy formulations very much conceal the common structure inherent in - .
thus , in this article , we consistently adopt the perspective of differential forms .
thinking in terms of co - ordinate free differential forms offers considerable benefits as regards the construction of structure preserving spatial discretizations . by now this
is widely appreciated for boundary value problems for @xmath30 , where the so - called discrete exterior calculus @xcite , or , equivalently , the mimetic finite difference approach @xcite , or discrete hodge - operators @xcite have shed new light on existing discretizations and paved the way for new numerical methods .
all these methods have in common that @xmath7-forms are approximated by @xmath7-co - chains on generalized triangulations of the computational domain @xmath10 .
this preserves co - homology and , thus , plenty of algebraic properties enjoyed by the solutions of the continuous bvp carry over to the discrete setting , see , e.g. , ( * ? ? ?
moreover , simplicial and tensor product triangulations allow the extension of co - chains to discrete differential forms , which furnishes structure preserving finite element methods .
also in this paper the focus will be on galerkin discretization by means of discrete differential forms and how it can be used to deal with lie derivatives . in light of the success of discrete differential forms
, it seems worthwhile exploring their use for the more general equations .
this was pioneered in @xcite and this article continues and extends these considerations .
revealing structure is not our only motivation : as already pointed out the discretization of has been the object of intense research , but and have received scant attention , though , is relevant for numerical modelling , e.g. , in magnetohydrodynamics .
apart from standard galerkin finite element methods and @xcite , we would like to mention @xcite as attempts to devise a meaningful discretization of the transport term in .
the first paper relies on a lagrangian approach along with a divergence preserving projection , which can be interpreted as an interpolation onto co - chains .
the second article employs edge degrees of freedom ( 1-co - chains ) combined with a special transport scheme . both refrain from utilizing exterior calculus
we stress that this article is confined to the derivation and formulation of fully discrete schemes for on a single _ fixed _ triangulation of @xmath10 .
principal attention is paid to the treatment of the convective terms .
the discussion centers on structural and algorithmic aspects , whereas rigorous numerical analysis of stability and convergence analysis will be addressed in some forthcoming work , because it relies on certain technical results that must be established for each degree @xmath7 of the form @xmath15 separately .
the plan of the paper is as follows : we first review the definition of lie derivatives and material derivatives for differential forms and give a short summary on discrete forms .
next , we present a semi - lagrangian approximation procedure for material derivatives of arbitrary discrete @xmath7-forms .
afterwards we present an eulerian treatment of convection of discrete forms .
these two sections focus on the formulation of fully discrete schemes .
they are subsequently elaborated for the eddy current model in moving media as an example for non - scalar convection - diffusion , see also @xcite .
finally we present numerical experiments for convection of @xmath31-forms in @xmath32 dimensions .
they give empiric hints on convergence and stability .
we introduce a space - time domain @xmath33 $ ] where @xmath34 is a bounded curvilinear polyhedron and @xmath35 \subset \mathbb r$ ] . given a vector field @xmath36 and initial values @xmath37 the associated flow @xmath38 ^ 2 \mapsto \mathbb r^n$ ] is defined through @xmath39 \\ x(x , s , s)&=x\;. \label{eqn : charac_ode } \end{aligned}\ ] ] for fixed @xmath40 the solution @xmath41
is called the characteristic curve through @xmath40 . a unique solution of problem ( [ eqn : charac_ode ] ) exists , whenever @xmath17 is continuous in @xmath42 and lipschitz continuous in @xmath43 for fixed @xmath44 $ ] . in this case
@xmath45 hence @xmath46 is the inverse of @xmath47 : @xmath48 for convenience we abbreviate @xmath49 .
it follows directly from that the jacobian @xmath50 solves : @xmath51 we write @xmath52 for the space of differential @xmath7-forms on @xmath10 ( cf .
* chapter v , section 3 ) ) .
differential @xmath7-forms can be seen as continuous additive mappings from the set @xmath53 of compact oriented , piecewise smooth , @xmath7-dimensional sub - manifolds of @xmath10 into the real numbers .
the spaces @xmath54 can be equipped with the @xmath55-norm @xmath56 , see ( * ? ? ?
2.2 ) . for euclidean hodge operator
this agrees with the usual @xmath57-norm of the vector proxies .
although it is very common to regard the mapping @xmath58 as integration we adopt the notation @xmath59 of a duality pairing instead of @xmath60 . only for @xmath61 we will use the integral symbol
@xmath62-forms can be identified with real valued functions .
recall the definition of the directional derivative for a smooth function @xmath63 : @xmath64 the lie derivative @xmath65 of @xmath7-forms @xmath66 generalizes this .
note that the point evaluation of @xmath62-forms is replaced with evaluation on @xmath7-dimensional oriented sub - manifolds @xmath67 .
thus the lie derivative of a @xmath7-form @xmath15 is @xmath68 in terms of the pullback @xmath69 defined by @xmath70 we can also write @xmath71 following @xcite the extrusion @xmath72 is the union of flux lines emerging at @xmath73 running from @xmath74 to @xmath75 ( figure [ fig : extrusion ] ) .
the orientation of @xmath73 induces an orientation of the extrusion @xmath76 such that the boundary is : @xmath77 plugging this into the definition of the lie derivative ( [ eq : liedefinition ] ) we get by means of stokes theorem @xmath78 the contraction operator is defined as the limit of the dual of the extrusion : @xmath79 and we recover from ( [ eq : lieextrusion ] ) cartan s formula ( * ? ? ?
5.3 ) for the lie derivative : @xmath80 for @xmath62-forms the second term vanishes , for @xmath81-forms the first one .
[ t][b]@xmath76 [ t][b]@xmath73 with respect to velocity field @xmath17.,title="fig : " ] for time dependent @xmath7-forms @xmath82 the limit value : @xmath83 corresponding to is the rate of change of the action of @xmath7-forms in moving media , hence a material derivative @xcite .
we deduce : @xmath84 since the exterior derivative and the lie derivative commute we have : @xmath85 as a consequence closed forms remain closed when they are advected by the material derivative .
if @xmath86 and @xmath87 then this property is preserved : @xmath88 .
[ remark : closed ] for @xmath31-forms @xmath89 with @xmath90 with vector proxy @xmath91 formulas and give a general convective term @xmath92 while for @xmath32-forms @xmath93 with vector proxy @xmath91 the convective term reads @xmath94 for closed @xmath32-forms @xmath95 , @xmath90 with vector proxy @xmath96 , which satisfies @xmath97 , we find @xmath98 we rely on a version of discrete exterior calculus spawned by using discrete differential forms .
we equip @xmath10 with a simplicial triangulation @xmath99 .
the sets of subsimplices of dimension @xmath7 are denoted by @xmath100 and have cardinality @xmath101 . in words , @xmath102 is the set of vertices of @xmath103 , @xmath104 the set of edges , etc .
we write @xmath105 for the space of whitney @xmath7-forms defined on @xmath99 , see ( * ? ? ?
23 ) and ( * ? ? ?
* sect . 3 ) .
from the finite element point of view a whitney @xmath7-form @xmath106 is a linear combination @xmath107 of certain basis functions associated with subsimplices of @xmath103 .
the basis functions @xmath108 have compact support and vanish on all cells that have trivial intersection with subsimplex @xmath109 .
locally , e.g. in each cell @xmath110 with @xmath111 , the basis functions @xmath112 can be expressed in terms of the barycentric coordinate functions @xmath113 and whose gradients @xmath114 .
the barycentric coordinate functions @xmath115 are associated with vertices @xmath116 of cell @xmath117 , e.g. @xmath118 .
if @xmath119 denotes the index set of vertices @xmath120 of subsimplex @xmath121 then @xmath122 the construction of these spaces ensures that the traces @xmath123 on element boundaries are continuous .
the evaluation maps @xmath124 , @xmath125 are the corresponding degrees of freedoms , which give rise to the usual nodal interpolation operators @xmath126 , also called derham maps ( * ? ? ?
3.1 ) , @xmath127 discrete differential forms are available for more general triangulations ( * ? ? ?
the whitney forms feature piecewise linear vector proxies , but can be extended to schemes with higher polynomial degree ( * ? ? ?
for the sake of simplicity we are not going to discuss those .
there are two basically different strategies to discretize material derivatives .
we either approximate the difference quotient ( [ eq : matdefinition ] ) directly or use the formulation ( [ eq : material_deriv ] ) in terms of partial time derivative of forms and lie derivative .
the former policy is referred to as lagrangian and in our setting of a single fixed mesh it is known as _ semi - lagrangian_. it is explored in this section .
the latter is called _ eulerian _ and reduces to a method of lines approach based on a spatial discretization of the lie derivatives .
its investigation is postponed to sect .
[ sec : euler - appr - discr ] .
we begin the study of the semi - lagrangian method with a look at the pure transport problem : given some @xmath128 and @xmath129 find @xmath130 , such that @xmath131 here and below we assume in the following that on the boundary of @xmath10 the normal components of the vector field @xmath132 vanish , hence @xmath133 .
this is a restrictive but very common assumption for semi - lagrangian methods . only a few semi - lagrangian methods like ellam @xcite can handle a non - vanishing velocity field on the boundary .
there are two different variational formulations for the transport problem .
the _ direct _
variational formulation reads : given @xmath134 and initial data @xmath135 find @xmath136 , such that @xmath137 the _ adjoint _ variational formulation for reads : given @xmath128 and initial data @xmath138 find @xmath136 , such that @xmath139 [ lem:31 ] the direct ( [ def : direct_var ] ) and the adjoint ( [ def : weak_var ] ) variational formulations are equivalent .
let @xmath95 and @xmath140 . since @xmath141 and @xmath142 : @xmath143 and the equivalence follows .
[ cor : duallie ] under the assumption @xmath144 : @xmath145 replacing the limits in ( [ def : direct_var ] ) and ( [ def : weak_var ] ) with a finite difference quotient yields semi - discrete timestepping schemes : given @xmath128 and @xmath146 find @xmath136 , such that @xmath147 and given @xmath128 and @xmath146 find @xmath136 , such that @xmath148 restricting the semi - discrete formulation to the discrete spaces @xmath149 , we end up with the following direct and adjoint schemes : given @xmath128 and @xmath150 find @xmath151 , such that @xmath152 and find @xmath151 , such that @xmath153 a significant advantage of the semi - lagrangian approach is the preservation of closeness ( see remark [ remark : closed ] ) .
this property is important in many physical applications .
the adjoint semi - lagrangian timestepping fulfils this in a weak sense .
a form @xmath95 is weakly closed if @xmath154 .
for @xmath155 the adjoint semi - lagrangian timestepping boils down to @xmath156 as the exterior derivative and pullback commute we conclude @xmath157 with @xmath158 .
hence @xmath159 is weakly closed if @xmath160 is weakly closed .
[ remark : weakclosed ] [ remark : stab_const_cont ] another advantage of semi - lagrangian methods is the straightforward stability analysis for the homogeneous transport problem @xmath161 . the corresponding direct scheme reads : given @xmath162 , find @xmath163 @xmath164 testing with @xmath165 we immediately get the stability estimate : @xmath166 with @xmath167 . for @xmath90 and euclidean hodge the concrete constants can be deduced from vector proxy representations of the pullback @xcite .
the stability estimates then read : @xmath168 where @xmath169 is the euclidean matrix norm .
stability will depend in general on the distortion effected by @xmath170 . for @xmath171 even the usual assumption @xmath172 does not guarantee stability .
it is important to note that both the stability estimates and the preservation of weak closedness as previously stated assume exact evaluation of the bilinear forms @xmath173 and @xmath174 . any non - trivial problem will require the use of additional approximations .
we propose the following three approximation steps . 1
. first we use some numerical integrator to determine approximations @xmath175 to the exact evolutions @xmath176 of vertices @xmath177 .
note that the direct scheme requires to solve backward in time , while for the adjoint scheme we calculate forward in time .
the simplest numerical integrator for this first step is the forward euler scheme , which gives @xmath178 .
then we approximate the evolution @xmath179 of arbitrary points @xmath180 as the convex combination of the approximated evolutions of surrounding vertices : @xmath181 this yields a piecewise linear , continuous approximation @xmath182 of the exact evolution @xmath179 .
+ note that for small time steps and lipschitz continuous @xmath17 the approximating flow @xmath183 maps the simplicial triangulation @xmath99 onto a new simplicial triangulation @xmath184 , such that each subsimplex @xmath121 of @xmath99 has an image @xmath185 in @xmath186 .
finally , we apply the nodal interpolation operators of discrete forms ( * ? ? ?
3.3 ) to map the transported discrete forms @xmath187 and @xmath188 onto the space of discrete forms .
let us illustrate these three steps for the case of @xmath62-forms and @xmath31-forms : to determine now the interpolant of a discrete transported @xmath62-form @xmath189 , by linearity it is enough to consider the basis functions .
since @xmath190 for all vertices @xmath177 and @xmath62forms @xmath15 the matrix operator @xmath191 with entries @xmath192 maps the expansions coefficients of @xmath165 to the coefficients of @xmath193 .
this means that in each time step we not only need to determine the points @xmath194 but also the location within the triangulation @xmath99 . to find the element , in which @xmath195 is located , we trace the path of the trajectory from one element to the next . based on this data
the matrix entries ( [ def : p0_entries ] ) can be assembled element by element ( see fig . [
fig : transport_vertex ] for the direct scheme ) . , we move backward along the trajectory @xmath196 starting from @xmath177 and identify the crossed elements @xmath197 and @xmath198 . in this case @xmath199 and @xmath200 are the only non - zero entries in the @xmath201th row of @xmath202 . ]
the advantage of the second approximation step , the linear interpolation , is elucidated by the treatment of @xmath31-forms .
the interpolation of a transported discrete @xmath31form @xmath203 is determined by the interpolation of transported basis forms and the condition @xmath204 for all edges @xmath205 .
this defines a matrix @xmath206 , mapping the expansion coefficients of @xmath165 to those of @xmath203 .
the matrix entries @xmath207 are line integrals along the straight line from @xmath208 to @xmath209 , if @xmath210 is the edge connecting the vertices @xmath211 and @xmath212 ( see fig .
[ fig : transport_edge ] for the direct scheme ) .
( black curved line ) is approximated by a straight line @xmath213(black dashed line ) . in the case
depicted here all basis functions associated with edges @xmath214 of elements @xmath215 , @xmath216 and @xmath217 yield a non - zero entry @xmath218 . ] to determine the entries of the @xmath201-th row , we trace the straight line from @xmath219 to @xmath220 and calculate for each crossed element the line integrals for the attached basis functions
. if e.g. the line crosses an element @xmath221 with edge @xmath214 from point @xmath222 to point @xmath223 ( see fig . [
fig : sketch_edge ] for the direct scheme ) , then the element contribution to @xmath224 is : @xmath225}b_{i'}^1 = \lambda_{j'}(a)\lambda_{k'}(b)- \lambda_{k'}(a)\lambda_{j'}(b)\,,\end{aligned}\ ] ] where @xmath226 and @xmath227 are the terminal vertices of edge @xmath214 . $ ] is the intersection of the approximation of the transported edge with element @xmath221 . ]
observe that we do not need to calculate additional approximate solutions of to determine the matrix entries .
piecewise linear interpolation of the approximate evolution of vertices automatically gives approximate evolutions of all subsimplices .
although the approximate flow @xmath183 is not smooth , we can show that the fully discrete adjoint scheme preserve the weak closedness property .
is is easily established that exterior derivative and interpolation of approximate pullbacks commute ( * ? ? ?
[ lemma : commute ] let @xmath228 be the standard nodal interpolation operator and @xmath229 , where the coefficients @xmath230 are approximate solutions of the characteristic ode , then we have : @xmath231 finally we state the fully discrete semi - lagrangian timestepping schemes for the general convection - diffusion problem : given a temporal mesh @xmath232 , the discrete semi - lagrangian timestepping scheme for the direct variational formulation of convection - diffusion of differential forms is : given @xmath233 and @xmath234 , find @xmath235 such that : @xmath236 the discrete semi - lagrangian timestepping scheme for the adjoint variational formulation of convection - diffusion of differential forms is : given @xmath237 and @xmath238 , find @xmath235 such that : @xmath239 the proposed sequence of approximation steps allows for a comprehensive stability and convergence analysis of fully discrete schemes .
stability of the fully discrete schemes in @xmath240 for example follows from remark [ remark : stab_const_cont ] .
we only need to control perturbations due to some numerical integrator , the linear interpolation and interpolation of forms .
we will address this in some future work . the standard finite element approach for approximating bilinear forms like
@xmath241 would be quadrature .
but since the product of @xmath242 and @xmath243 is smooth only on intersections @xmath244 , this entails finding all intersections of an element @xmath245 of the triangulation with any other element @xmath246 of the transported triangulation , and apply some quadrature there .
we conclude that such an approach is even more complex than methods based in interpolation .
further , it is totally unclear how such an approximation affects stability and conservation of closedness .
if we use in the scalar case @xmath22 a vertex based quadrature rule to approximate the element integrals @xmath247 , the resulting scheme corresponds to a lagrange - galerkin scheme from @xcite using vertex based quadrature .
notice that in both and the unknown @xmath248 is obtained by solving a standard discrete diffusion problem arising from a symmetric positive definite bilinear form , which are amenable to fast iterative solvers @xcite . in the words of @xcite ,
the semi - lagrangian method is `` solver - friendly '' .
the eulerian approach treats the discretization in space and the discretization in time separately .
we therefore consider the stationary version of the convection - diffusion equation : @xmath249{rcll } - \varepsilon ( -1)^{l } d\ast d \omega + \ast l_{\boldsymbol \beta } \omega & = & \varphi \quad & \text{in } \omega\subset\mathbb{r}^{n}\;,\\ \imath^{\ast}\omega & = & 0 & \text{on } \partial\omega\;. \end{array}\end{aligned}\ ] ] since the approximation of the diffusion term is well - known @xcite , we will discuss only the approximation of the convective terms in .
we will pursue two different ways to obtain spatial discretization for the lie - derivative terms . the standard galerkin technique to approximate + relies on elementwise quadrature .
while evaluation of the part is standard in the finite element context , we encounter some difficulties for the term , which occurs for @xmath250 .
the term @xmath251 implicitly requires spatial derivatives that fail to be square integrable for discrete differential forms .
for instance , for @xmath31-forms @xmath95 with smooth vector proxy @xmath252 we have : @xmath253 while for discrete @xmath31-forms @xmath106 with vector proxy @xmath254 on each element the symmetric part @xmath255 vanishes . a scheme based on elementwise quadrature
will always miss @xmath256 .
stated differently , the exterior derivative @xmath257 of the contraction of a discrete form is not defined in a strong sense .
but it is defined in a weak sense : for any smooth @xmath12 form @xmath258 , and discrete differential form @xmath259 elementwise integration by parts yields : @xmath260 where @xmath261 is the restriction of @xmath13 to each element @xmath262 .
since each face inside @xmath10 occurs twice in the right sum we can rewrite this term as sum of face integrals over certain jump terms @xmath263^{l-1}$ ] .
@xmath264^{l-1 } \wedge \gamma\ , , \end{aligned}\ ] ] one is now tempted to use the right hand side of to define approximations for the convection terms .
the difficulty here are the face integrals , which are not well - defined for @xmath265 . motivated by finite volume approximations , we replace in the case @xmath266 in the face integrals @xmath267 with a flux function @xmath268 depending on the values @xmath269 and @xmath270 on adjacent elements .
we use the characterization ( [ eq : liedefinition ] ) of the lie derivative to determine such consistent flux functions .
[ def : bd ] let @xmath17 be lipschitz continuous , @xmath271 and @xmath272 then @xmath273 is the variational _ central _ finite difference , with @xmath274 we could have used one - sided finite differences as well .
next we show that the limit @xmath275 exists also for discrete differential forms .
[ lemma : standard1 ] the limit @xmath276 exists , and we have the representation @xmath277 with upwind and downwind traces @xmath278 and @xmath279 and the pointwise limit @xmath280 .
@xmath281 is the index set of faces , that are not on the boundary of @xmath10 .
it s cardinality is @xmath282 .
here we used @xmath283 let us first note that for all @xmath284 the pointwise limit @xmath285 exists .
we decompose the central difference into a sum of upwind and downwind finite differences : @xmath286 and look first at the limit for the upwind finite difference .
we split the integration over the domain @xmath10 in a sum of integrals over patches @xmath287 with smooth integrand : @xmath288 and distinguish three cases , see fig .
[ fig : limit_pic ] : * @xmath289 and @xmath290 : @xmath291 * @xmath289 and @xmath292 : this implies first that @xmath293 .
if elements @xmath262 and @xmath294 share a face @xmath295 it is clear that @xmath296 and we have @xmath297 and @xmath298 by the definition of contraction . *
@xmath299 since by assumption @xmath300 is bounded on @xmath301 we get : @xmath302 after summation over all elements we obtain : @xmath303 for the upwind finite difference .
a similar argument for the downwind finite difference shows : @xmath304 combining these two results for upwind and downwind finite difference we get the assertion . .
in the case of the upwind finite difference we have , that discrete @xmath7-forms @xmath305 are discontinuous on edges of the blue triangulation @xmath99 .
their pullbacks @xmath306 are discontinuous on edges of the red dashed triangulation @xmath307 , the image of @xmath99 . ]
the careful examination of the proof of lemma [ lemma : standard1 ] shows that the existence of the limit can be shown for quite general approximation spaces for differential forms .
we used only the continuity inside each element . for the sake of completeness
we rewrite here the standard convection bilinear form for vector proxies @xmath308 or @xmath309 in the case @xmath310 .
@xmath311 denotes the normal vector of a face .
discrete whitney @xmath62-forms are continuous . hence in this case
the face integrals in vanish and we obtain the standard bilinear form for scalar convection : @xmath312 for discrete @xmath31-forms we have both volume and face contributions : @xmath313 and similar for discrete @xmath32 forms : @xmath314 for discrete @xmath315-forms we get : @xmath316 for @xmath315-forms in 3d the limit corresponds to the non - stabilized discontinuous galerkin method for the advection problem from @xcite . in the lowest order case this is identical with the scheme for discrete @xmath315-forms .
the limit for the upwind finite difference yields the stabilized discontinuous galerkin methods with upwind numerical flux . in practice we will use local quadrature rules to evaluate the volume and face integrals in @xmath317 .
this will not destroy the consistency if the quadrature rules are first order consistent . in the definition of @xmath318 from def .
[ def : bd ] we may also appeal to lemma [ lem:31 ] and shift the difference quotient onto the test function .
this carries over to the limit and yields two different eulerian timestepping methods for problem that , in analogy to sect .
[ sec : deriv_slg ] , we may call `` direct '' and `` adjoint '' . the direct standard eulerian timestepping scheme for convection - diffusion of differential forms
is : given @xmath319 and @xmath238 , find @xmath235 such that : @xmath320 the adjoint standard eulerian timestepping is : given @xmath319 and @xmath238 , find @xmath235 such that : @xmath321 now we use the definition of lie - derivatives as limit of a difference quotient to define a proper interpolation operator for lie - derivatives of discrete forms .
let @xmath322 be a subsimplex of the triangulation .
then both limit values , the limit from the upwind direction @xmath323 and the limit from the downwind direction @xmath324 exist for lipschitz continuous @xmath17 .
the continuity of traces of discrete differential forms ensures that integrals are well - defined .
the existence of upwind and downwind limits then follows from the existence of the limits for locally linear @xmath17 , which in turn can be calculated explicitly by distinguishing different geometric situations . we skip the tedious details .
observe that in general @xmath325 e.g. for @xmath62-forms a short calculation shows @xmath326 and @xmath327 with upwind and downwind traces @xmath328 and @xmath329 . only for cases where @xmath165 is continuous in a neighbourhood of subsimplex @xmath121 upwind and downwind
limit agree . since and correspond to integral evaluation of the usual interpolation operators , we could either use the upwind or the downwind limit to define discrete lie - derivatives .
we will use the upwind limit for the direct formulation and the downwind limit for the adjoint formulation .
this is motivated by taking the cue from semi - lagrangian schemes .
let @xmath17 be lipschitz continuous and @xmath330 .
further let @xmath331 be the subsimplices corresponding to the degrees of freedom of basis functions @xmath332 then @xmath333 is the upwind interpolated discrete lie - derivative .
analogously the downwind interpolated lie - derivative is @xmath334 it follows from the cartan formula that @xmath335 with @xmath336 the exterior derivatives can be evaluated exactly for discrete differential forms .
thus , the contraction operators are the only approximations . but for locally constant discrete forms these approximations are exact and one can prove consistency @xmath337 once one has interpolation estimates for discrete differential forms .
this is outside the scope of this paper .
we finally want to stress that it is important to treat the limits and for @xmath250 as limits of co - chains rather than integrals with certain integrands that are defined as pointwise limits .
first , these pointwise limits are not well - defined since discrete forms are in general discontinuous across element boundaries .
second , even for special cases where the pointwise limit @xmath338 exist we can have @xmath339 to understand this , we may consider the 2d example depicted in fig . [ fig : limit_example ] with constant @xmath17 .
we take @xmath340 such that @xmath341 for all edges @xmath342 except the edge @xmath343 between vertices @xmath344 and @xmath345 and calculate the projection of the lie - derivative onto edge @xmath346 between vertices @xmath345 and @xmath347 .
then it is clear that @xmath348 and therefor @xmath349 on the other hand , one can easily show that : @xmath350 hence @xmath351 since an explicit calculation of the upwind limits might be expensive for arbitrary @xmath17 , we have to find consistent approximations .
the preceding discussion shows that this must be done very carefully .
replacing the integration simply with some quadrature will not yield a consistent approximation .
we combine the upwind discretization with an implicit euler method and state the implicit upwind methods for convection - diffusion for discrete differential forms .
the implicit direct upwind eulerian timestepping scheme for the variational formulation of convection - diffusion of differential forms is : given @xmath352 and @xmath238 , find @xmath235 such that : @xmath353 the adjoint upwind eulerian timestepping scheme for the adjoint variational formulation of convection - diffusion of differential forms is : given @xmath352 and @xmath238 , find @xmath235 such that : @xmath354 in contrast to the semi - lagrangian schemes and , in and we face a non - symmetric algebraic system in each time step , whose iterative solution can be challenging .
we therefor consider a second kind of upwind eulerian scheme , that treats the convection in an explicit fashion . the direct semi - implicit upwind eulerian timestepping scheme for the variational formulation of convection - diffusion of differential forms is : given @xmath352 and @xmath238 , find @xmath235 such that : @xmath355 the adjoint upwind eulerian timestepping scheme for the variational formulation of convection - diffusion of differential forms is : given @xmath352 and @xmath238 , find @xmath235 such that : @xmath356 these semi - implicit schemes do not coincide with the usual method of lines approach .
nevertheless there are two heuristic arguments that could justify the semi - implicit schemes .
note first that for discrete @xmath62-forms the semi - lagrangian and the semi - implicit eulerian scheme are identical , if we use a time step length @xmath357 and explicit euler for solving the characteristic equation .
the direct semi - lagrangian scheme for discrete @xmath62-forms was , written here for vector proxies @xmath21 and @xmath358 for @xmath165 and @xmath359 : find @xmath21 such that @xmath360 while the semi - implicit eulerian scheme was : find @xmath21 such that @xmath361 with @xmath362 on the other hand @xmath363 and by the time step condition @xmath364 : @xmath365 hence we have the identity : @xmath366 the second argument , that supports the definition of semi - implicit schemes applies to forms of any degree and follows from the identity : @xmath367 hence we can derive the semi - implicit schemes and from the semi - lagrangian schemes and in replacing e.g. @xmath368 the analysis of a semi - implicit discontinuous galerkin scheme for scalar non - linear convection - diffusion can be found in @xcite . for the linear case this scheme is modulo certain quadrature rules identical with the semi - implicit euler scheme for @xmath315-forms .
there is also a very close link between both the semi - lagrangian and semi - implicit eulerian methods and arbitrary lagrangian - eulerian methods ( ale ) @xcite .
ale methods are operator splitting methods , that treat the diffusion part in a lagrangian and the convection part in an eulerian fashion . after a certain number of lagrangian iterations
steps an eulerian iteration step maps the mesh function on the distorted mesh back to the initial mesh .
the semi - lagrangian and semi - implicit eulerian methods apply this mapping in each time step . in many problems
these remap operations should preserve certain properties of the mesh function , e.g. volume or closedness .
it is the discrete differential form approach , that naturally guaranties such properties for the methods presented here and for the ale method in @xcite .
in this section we will derive two different convection - diffusion equations for the electromagnetic part of magnetohydrodynamics ( mhd ) models .
the first one is an equation for the magnetic field @xmath369 and requires the adjoint methods while the second one for the magnetic vector potential @xmath222 can be solved via the direct methods . commonly in mhd applications ,
one neglects the displacement current .
this reduced model , called eddy current model , is a system of equations for the magnetic field @xmath370 , the electric field @xmath371 , the magnetic field density @xmath372 , the current density @xmath373 and external source @xmath374 : @xmath375 a uniformly positive conductivity @xmath376 will be assumed in the sequel . to rewrite the eddy current model in terms of a material derivative we substitute @xmath377 and add @xmath378 to faraday s law ( [ eq : fala ] ) .
this leaves the solution unchanged , since @xmath379 .
hence we end up with the system : @xmath380 next we state the two different variational formulations , _ cf . _
2.3 ) . for simplicity
we impose homogeneous electric boundary conditions on @xmath381 . testing ( [ eq : fl ] ) with @xmath382 , integration by parts yields : @xmath383 we eliminate @xmath384 using ohm s law and end up with the following variational formulation : seek @xmath385 such that : @xmath386 in order to eliminate @xmath223 as well we use corollary [ cor : duallie ] and get @xmath387 or , equivalently @xmath388 this variational formulation can be approximated using a semi - lagrangian framework yielding an algebraic system : @xmath389 where @xmath390 and @xmath391 are the mass matrices @xmath392 and @xmath393 , @xmath394 .
the evaluation of the right hand side requires the calculation of the matrix @xmath395 from .
lemma ( [ lemma : commute ] ) shows that the solution of ( [ eq : h_timestepping ] ) is weakly closed if the initial data is weakly closed and @xmath396 . alternatively , we may use one of the implicit eulerian schemes or @xmath397 where @xmath398 is the stiffness matrix for either the standard or the downwind discretization . while these schemes will not preserve the weakly closed property , the semi - implicit upwind eulerian scheme with algebraic system @xmath399 does , due to the interpolation based construction . here , for simplicity , we assume homogeneous magnetic boundary conditions on @xmath400 .
the ansatz @xmath401 and @xmath402 solves faraday s law ( [ eq : fl ] ) .
we then multiply ampere s law with a test form @xmath403 and integrate by parts : @xmath404 the material laws ( [ eq : mat_sigma ] ) and ( @xmath405 ) eliminate @xmath406 and @xmath369 and we get the @xmath222-based variational formulation : find @xmath407 such that : @xmath408 the algebraic system for the direct semi - lagrangian scheme then reads as @xmath409 while the systems for the implicit and semi - implicit eulerian schemes are : @xmath410 and @xmath411 here @xmath412 and @xmath413 are the stiffness matrices for @xmath414 and @xmath415 , @xmath416 .
@xmath417 corresponds either to the standard or the upwind eulerian discretization of the convection part .
we finally we present a few numerical examples that illustrate the performance of the methods derived here . we take @xmath418 and look at @xmath419{rcll } \partial_t \ast \omega(t ) - \varepsilon ( -1)^{l } d\ast d \omega(t ) + l_{\boldsymbol \beta } \ast\omega ( t ) & = & \varphi \quad & \text{in } \omega\subset\mathbb{r}^{2}\;,\\ \imath^{\ast}\omega & = & 0 & \text{on } \partial\omega\;,\\ \omega(0 ) & = & \omega_{0 } & \text{in } \omega \ ; \end{array}\end{gathered}\ ] ] for 1-forms @xmath89 . in vector proxy notion with @xmath252 this reads @xmath420{rcll } \partial_{t}{{\mathbf{u}}}+ \varepsilon { \operatorname{\bf curl}}\text{curl } { { \mathbf{u}}}+ { \boldsymbol \beta}({{\mathsf{div}}}{{\mathbf{u } } } ) + { \operatorname{\bf curl}}({{\mathbf{u}}}\times { \boldsymbol \beta } ) & = & { { \mathbf{f } } } & \text{in } \omega\;,\\ { { \mathbf{u}}}\cdot{{\mathbf{n}}}&= & 0 & \text{on } \partial\omega\;,\\ { { \mathbf{u}}}(0 ) & = & { { \mathbf{u}}}_{0 } & \text{in } \omega\ ; , \end{array}\end{gathered}\ ] ] with @xmath421 , @xmath422 and @xmath423 , @xmath424 .
we approximate @xmath15 by discrete @xmath31-forms @xmath165 on a triangular mesh .
we study the adjoint semi - lagrangian and eulerian methods , , and for this boundary value problem .
* experiment i. * in the first experiment we take in problem the domain @xmath425 ^ 2 $ ] , the divergence free velocity @xmath426 and @xmath427 then with @xmath428 neither the convection nor the diffusion part dominates and we expect for all three schemes first order convergence in the mesh size @xmath369 , if the time step @xmath429 is of order @xmath369 .
the convergence plots in figures [ fig : convplot1 ] and [ fig : convplot2 ] show the @xmath57-error at time @xmath430 and confirm this for different ratios @xmath431 .
* experiment iii .
* next , we examine the stability properties of the three different schemes for dominating convection , meaning that we choose @xmath435 in problem very small . while for @xmath428 all three schemes are stable for small and larger cfl - numbers ( see fig .
( [ fig : convplot1 ] ) and ( [ fig : convplot2 ] ) ) we encounter this for small @xmath435 only for the semi - lagrangian and the implicit eulerian scheme ( see fig .
( [ fig : stab1 ] ) and ( [ fig : stab2 ] ) ) . as expected in the convection dominated case
, one has to use either the semi - lagrangian or fully implicit eulerian scheme .
* experiment iv . *
another question that we address concerns the stabilization for the eulerian schemes . for the implicit timestepping schemes we need to solve in each time step a stationary convection - diffusion problem .
it is well - known that in the scalar case non - stabilized methods would produce highly oscillating solutions for convection - diffusion problems with dominating convection . to study this issue
, we consider the stationary convection diffusion problem for @xmath31-forms and @xmath418 : @xmath440{rcll } { { \mathbf{u}}}+ \varepsilon { \operatorname{\bf curl}}\text{curl } { { \mathbf{u}}}+ { \operatorname{\bf grad}}({\boldsymbol \beta } \cdot { { \mathbf{u } } } ) - { { \mathbf{r}}}\boldsymbol \beta \text{curl } { { \mathbf{u}}}&= & { { \mathbf{f } } } & \text{in } \omega=[-1,1]^2 \;,\\ { { \mathbf{u}}}\cdot{{\mathbf{n}}}&= & 0 & \text{on } \partial\omega\;,\\ { { \mathbf{u}}}(0 ) & = & { { \mathbf{u}}}_{0}\;. \end{array}\end{gathered}\ ] ] we could either use the standard approximation or the upwind interpolated discrete lie - derivative in the discretization of .
again , we choose @xmath441 and the divergence free velocity . a. bossavit .
discretization of electromagnetic problems : the `` generalized finite differences '' . in w.h.a .
schilders and w.j.w .
ter maten , editors , _ numerical methods in electromagnetics _ , volume xiii of _ handbook of numerical analysis _ , pages 443522 .
elsevier , amsterdam , 2005 .
h. de sterk .
multi - dimensional upwind constrained transport on unstructured grids for `` shallow water '' magnetohydrodynamics . in _ proceedings of the 15th aiaa cfd conference , jun 11 - 14 , anaheim , ca_. aiaa , 2001 .
hyman and m. shashkov .
mimetic finite difference methods for maxwell s equations and the equations of magnetic diffusion . in f.l .
teixeira , editor , _ geometric methods for computational electromagnetics _ , volume 32 of _ pier _ , pages 89121 . emw publishing , cambridge , ma , 2001 . | we consider generalized linear transient convection - diffusion problems for differential forms on bounded domains in @xmath0 . these involve lie derivatives with respect to a prescribed smooth vector field .
we construct both new eulerian and semi - lagrangian approaches to the discretization of the lie derivatives in the context of a galerkin approximation based on discrete differential forms .
details of implementation are discussed as well as an application to the discretization of eddy current equations in moving media . |
graphene@xcite , composed carbon atoms arranged in a two - dimensional honeycomb lattice , is the building block for all graphitic materials from the 0d fullerenes to 1d nanotubes and also the common 3d graphite .
however , following peierls and landau s arguments , the two - dimensional lattice is unstable and can not exist at any finite temperature .
therefore , graphene is often used as a toy model and viewed as an academic material until its recent discovery in laboratory@xcite . since graphene has promising application potentials in nanoscale electronic devices , it is important to study how the band structure changes with respect to the finite transverse width and also the edge topology .
in addition , we also expect the electron - electron interaction will play a crucial role for physics properties in low - energy limit .
therefore , we are motivated to investigate the ground state properties in graphene nanoribbon ( gnr ) here . with lithography techniques ,
gnr with width smaller than 20 nm has been fabricated successfully@xcite
. however , the edge roughness seems to post a tough challenge on both theoretical and experimental sides .
a breakthrough comes from chemical approach@xcite recently , that can produce gnr down to 10 nm in the controlled way .
moreover , these gnrs have remarkably smooth edges and make the fast electronics at molecular scale possible@xcite . in this work ,
we concentrate on the gnr with armchair edges since the open boundaries give rise to peculiar flat bands .
note that the edges are hydrogenated so that the dangling @xmath0 bonds are saturated and only the @xmath1 bands remain active in low energy@xcite . by combining analytic weak - coupling analysis , numerical density matrix renormalization - group ( dmrg ) method , and the first - principles calculations ,
it was shown@xcite that armchair gnr exhibits novel carrier - mediated ferromagnetism upon appropriate doping .
even though only @xmath1 bands are active in low - energy , in appropriate doping regimes , the armchair edges give rise to both itinerant bloch and localized wannier orbitals .
these localized orbitals are direct consequences of quantum interferences in armchair gnr and form flat bands with zero velocity .
the carrier - mediated ferromagnetism can thus be understood in two steps : electronic correlations in the flat band generate intrinsic magnetic moments first , then the itinerant bloch electrons mediate ferromagnetic exchange coupling among them . here
we use a variational wave function approach and try to understand how the flat - band ferromagnetism evolves with the interaction strength .
we start with the hubbard model to describe the armchair gnr , @xmath2 + u\sum_{{\bf r } } n_{\uparrow}({\bf r})n_{\downarrow}({\bf r}).\end{aligned}\ ] ] where @xmath3 is the hopping amplitude , @xmath4 is the repulsive on - site interaction , @xmath5 is the spin index , @xmath6 , and @xmath7 is taken only for nearest neighbor bonds .
rough estimates from experiments give @xmath8 ev and @xmath9-@xmath10 ev , putting the ratio @xmath11 to be of order one . before diving into the numerical details
, it is insightful to highlight the peculiar flat - band orbitals first . for the armchair gnr
, its transverse width can be labeled by an integer @xmath12 , i.e. the size of the unit cell for each sublattice .
writing down the wave function on different sublattices as the two - componenet spinor , it is easy to check that , for symmetric armchair gnr ( odd @xmath12 ) , one can construct a localized wave function at @xmath13 , @xmath14,\end{aligned}\ ] ] and show it is indeed an eigenstate of the hopping hamiltonian .
applying translational invariance to shift the wannier orbital , the flat band emerges at the end .
this is the one - dimensional analog of the landau level degeneracy for two - dimensional electrons in magnetic field .
it is interesting that the edge topology in 1d quenches the kinetic energy without the necessity to couple to external magnetic field .
furthermore , we would like to point out that the flat - band ferromagnetism in armchair gnr is not the same by mielke - tasaki@xcite mechanism since there is no direct overlap between these flat - band orbitals .
the ferromagnetism is actually mediated by other dispersive bands that couple the magnetic moments in the flat band .
on the purpose of seeing the polarization for the ground state , we introduce the following trial wave function , @xmath15 \left[\prod_{k_{s m \downarrow } } a_{s m \downarrow}^{\dag}(k_{s m \downarrow } ) \right ] valence bands ( @xmath16 ) and the conducting bands ( @xmath17 ) separately and @xmath18 is the band index .
thus , there are @xmath19 number of bands totally .
the lower bounds for @xmath20 are decided by their band structures ( which are either @xmath21 or @xmath22 ) , while the upper bounds are the fitting parameters to minimize the variational energy , reducing the problem to look for minimum in the @xmath23-dimensional ( one constraint from the total particle conservation ) parameter space . for convenience ,
we introduce the filling factor , @xmath24 for each band as the variational parameters .
since the filling factor denotes the fraction of the filled states in the band , summing @xmath25 over all bands gives the number of particles in one unit cell , i.e. @xmath26 , where @xmath27 is the average particle number per site .
meanwhile , the total spin of the ground state , @xmath28 , can also be expressed in terms of the filling factors easily .
the kinetic energy per unit cell is obtained by integrating the energy spectra to the desired fillings .
the integration is fundamental and can be expressed in terms of the elliptic integrals , @xmath29 + \sum_{s\alpha}st\nu_{s_{f\alpha}},\end{aligned}\ ] ] where @xmath30 for @xmath31 is the elliptic integral of the second kind .
the magnitude of the transverse momentum is @xmath32 and @xmath33 . for clarity
, we separate the kinetic energy from the dispersive bands and the flat bands in above .
thus , the summation of the band index @xmath34 does not include the flat bands .
it shall be clear that the kinetic energy of the flat band ( denoted by @xmath35 ) is rather trivial since it is directly proportional to filling factor multiplied by the energy @xmath36 . for dispersive bands ,
the initial momentum @xmath37 starts from either @xmath21 or @xmath1 , depending where the spectrum minimum sits in the 1d brillouin zone .
the corresponding filled momentum @xmath38 is either @xmath39 or @xmath40 .
similarly , we can also compute the interaction energy per unit cell , @xmath41 \nu_{sm\uparrow } \nu_{s'm'\downarrow},\end{aligned}\ ] ] where the effective repulsive interaction @xmath42 between the bands @xmath43 and @xmath44 is the summation inside the bracket in above .
the reason why the effective repulsion does not carry the explicit dependence on @xmath45 is due to the particle - hole symmetry in the armchair gnr .
the analytic expressions for the kinetic and interaction energies are nice but the search for minima in the global parameter space can only be done by numerical approach due to the number of all filling factors are large .
the numerical search is carried out extensively at interaction strength from @xmath46 to @xmath47 . as an illustrating example , we present our numerical results at @xmath48 and @xmath49 in fig . [
fig : s ] for @xmath50 armchair gnr . for wider gnr ,
the numerical results are qualitatively the same . meanwhile , to avoid being trapped in local minima , we randomly choose the initial searching points and select the final configurations with lowest energy .
spin polarization in each bands at different electron number per site @xmath51 for the ( a ) @xmath48 and ( b ) @xmath49 cases .
the flat - band regime ( @xmath52 ) is between the yellow dashed lines and the gray open circle denotes twice of the total spin per unit cell @xmath53 .
spin polarization from the valence bands are shown in red , green , blue open circles , corresponding to the flat band , the lower ( @xmath54 ) and the upper ( @xmath55 ) dispersive bands .
the spin polarization for the conduction bands are mirror images of these around the symmetric axis @xmath56 . for clarity of the figure ,
they are not shown here.,title="fig : " ] spin polarization in each bands at different electron number per site @xmath51 for the ( a ) @xmath48 and ( b ) @xmath49 cases .
the flat - band regime ( @xmath52 ) is between the yellow dashed lines and the gray open circle denotes twice of the total spin per unit cell @xmath53 .
spin polarization from the valence bands are shown in red , green , blue open circles , corresponding to the flat band , the lower ( @xmath54 ) and the upper ( @xmath55 ) dispersive bands .
the spin polarization for the conduction bands are mirror images of these around the symmetric axis @xmath56 . for clarity of the figure ,
they are not shown here.,title="fig : " ] for the relatively weak interaction strength @xmath48 , the influence of the band structure is still manifest .
first of all , the particle - hole symmetry is manifest in the total spin @xmath57 .
note that the finite spin polarization near @xmath58 and @xmath59 is due to the van hove singularity and is not our main focus here .
it is rather interesting that the ferromagnetic phase coincide with the flat - band regime rather nicely .
in fact , the total spin per unit cell @xmath57 mainly comes from the flat - band orbitals .
furthermore , since the upper dispersive band ( @xmath55 in light blue color ) intersects with the flat band , it is also polarized in the flat - band regime . on the other hand , the lower dispersive band ( @xmath54 in green color )
is separated by a finite gap and does not participate in the ferromagnetism .
these numerical findings agree with field - theory approach , which shows that the exchange coupling between the flat - band moments are mediated by the itinerant carriers in the intersecting dispersive bands .
as the interaction strength increases to @xmath49 , the competition between the kinetic and the interaction energies become highly non - trivial even at the mean - field level . the ferromagnetic regime
due to van hove singularity expands a little bit as expected from stoner criterion .
it is rather remarkable that the ferromagnetic phase is still confined to the flat - band regime rather well .
but , since the interaction strength is large enough , both dispersive bands contribute despite the minor gap in one of them .
the intersecting band greatly enhances the ferromagnetic correlations and accounts for roughly one - third of the total spin density .
this implies that the itinerant carriers no longer play the secondary role as messengers to line up the localized moments in the flat band .
instead , they also develop ferromagnetic correlations and contribute to the total spin density .
the role of the dispersive band with a minor gap is peculiar .
it participates the ferromagnetic phase in the first - half of the flat - band regime and drops out in the second - half .
it is not yet clear whether this asymmetrical behavior is an artifact of the simple wave function we chose , or it hints for something real in realistic armchair gnr .
we show that the armchair edges of gnr quenches the kinetic energy of the itinerant carriers and leads to the flat - band orbitals at finite doping .
this is a beautiful one - dimensional analogy of the 2d landau levels except the magnetic field is not needed here .
following the hints from the field - theory analysis , we study magnetic properties of the ground state via variational wave function approach and found the flat - band ferromagnetism in armchair gnr .
we acknowledge support from the national science council of taiwan through grants nsc-95 - 2112-m-007 - 009 and nsc-96 - 2112-m-007 - 004 and also from the national center for theoretical sciences in taiwan . | we study the electronic correlation effects in armchair graphene nanoribbons that have been recently proposed to be the building blocks of spin qubits .
the armchair edges give rise to peculiar quantum interferences and lead to quenched kinetic energy of the itinerant carriers at appropriate doping level .
this is a beautiful one - dimensional analogy of the landau - level formation in two dimensions except the magnetic field is not needed here . combining the techniques of effective field theory and variational wave function approach
, we found that the ground state exhibits a new type of flat - band ferromagnetism that hasnt been found before . at the end
, we address practical issues about realization of this novel magnetic state in experiments . |
the motivation for this project is to design a novel optical system for quasi - real time alignment of tracker detector elements used in high energy physics ( hep ) experiments .
fox - murphy _ et.al .
_ from oxford university reported their design of a frequency scanned interferometer ( fsi ) for precise alignment of the atlas inner detector @xcite .
given the demonstrated need for improvements in detector performance , we plan to design an enhanced fsi system to be used for the alignment of tracker elements in the next generation of electron positron linear collider ( ilc ) detectors .
current plans for future detectors require a spatial resolution for signals from a tracker detector , such as a silicon microstrip or silicon drift detector , to be approximately 7 - 10 @xmath1@xcite . to achieve this required spatial resolution
, the measurement precision of absolute distance changes of tracker elements in one dimension should be on the order of 1 @xmath1 .
simultaneous measurements from hundreds of interferometers will be used to determine the 3-dimensional positions of the tracker elements .
we describe here a demonstration fsi system built in the laboratory for initial feasibility studies .
the main goal was to determine the potential accuracy of absolute distance measurements ( adm s ) that could be achieved under controlled conditions .
secondary goals included estimating the effects of vibrations and studying error sources crucial to the absolute distance accuracy .
a significant amount of research on adm s using wavelength scanning interferometers already exists @xcite . in one of the most comprehensive publications on this
subject , stone _
_ describe in detail a wavelength scanning heterodyne interferometer consisting of a system built around both a reference and a measurement interferometer@xcite , the measurement precisions of absolute distance ranging from 0.3 to 5 meters are @xmath0 250 nm by averaging distance measurements from 80 independent scans .
detectors for hep experiment must usually be operated remotely for safety reasons because of intensive radiation , high voltage or strong magnetic fields .
in addition , precise tracking elements are typically surrounded by other detector components , making access difficult . for practical hep application of fsi , optical fibers for light delivery and return
are therefore necessary .
we constructed a fsi demonstration system by employing a pair of single - mode optical fibers of approximately 1 meter length each , one for transporting the laser beam to the beam splitter and retroreflector and another for receiving return beams .
a key issue for the optical fiber fsi is that the intensity of the return beams received by the optical fiber is very weak ; the natural geometrical efficiency is @xmath2 for a measurement distance of 0.5 meter . in our design
, we use a gradient index lens ( grin lens ) to collimate the output beam from the optical fiber .
we believe our work represents a significant advancement in the field of fsi in that high - precision adm s and vibration measurements are performed ( without a _ priori _ knowledge of vibration strengths and frequencies ) , using a tunable laser , an isolator , an off - the - shelf f - p , a fiber coupler , two single - mode optical fibers , an interferometer and novel fringe analysis and vibration extraction techniques .
two new multiple - distance - measurement analysis techniques are presented , to improve precision and to extract the amplitude and frequency of vibrations .
expected dispersion effects when a corner cube prism or a beamsplitter substrate lies in the interferometer beam path are confirmed , and observed results agree well with results from numerical simulation .
when present , the dispersion effect has a significant impact on the absolute distance measurement .
the limitations of our current fsi system are also discussed in the paper , and major uncertainties are estimated .
the intensity @xmath3 of any two - beam interferometer can be expressed as @xmath4 where @xmath5 and @xmath6 are the intensities of the two combined beams , @xmath7 and @xmath8 are the phases . assuming the optical path lengths of the two beams are @xmath9 and @xmath10 , the phase difference in eq .
( 1 ) is @xmath11 , where @xmath12 is the optical frequency of the laser beam , and c is the speed of light . for a fixed path interferometer , as the frequency of the laser
is continuously scanned , the optical beams will constructively and destructively interfere , causing `` fringes '' .
the number of fringes @xmath13 is @xmath14 where @xmath15 is the optical path difference between the two beams , and @xmath16 is the scanned frequency range .
the optical path difference ( opd for absolute distance between beamsplitter and retroreflector ) can be determined by counting interference fringes while scanning the laser frequency .
a schematic of the fsi system with a pair of optical fibers is shown in fig.1 .
the light source is a new focus velocity 6308 tunable laser ( 665.1 nm @xmath17 675.2 nm ) . a high - finesse ( @xmath18 ) thorlabs sa200 f
- p is used to measure the frequency range scanned by the laser . the free spectral range ( fsr ) of two adjacent f - p peaks
is 1.5 ghz , which corresponds to 0.002 nm .
a faraday isolator was used to reject light reflected back into the lasing cavity .
the laser beam was coupled into a single - mode optical fiber with a fiber coupler .
data acquisition is based on a national instruments daq card capable of simultaneously sampling 4 channels at a rate of 5 ms / s / ch with a precision of 12-bits .
omega thermistors with a tolerance of 0.02 k and a precision of 0.01 @xmath19 are used to monitor temperature .
the apparatus is supported on a damped newport optical table . in order to reduce air flow and temperature fluctuations ,
a transparent plastic box was constructed on top of the optical table .
pvc pipes were installed to shield the volume of air surrounding the laser beam . inside the pvc pipes ,
the typical standard deviation of 20 temperature measurements was about @xmath20 .
temperature fluctuations were suppressed by a factor of approximately 100 by employing the plastic box and pvc pipes .
the beam intensity coupled into the return optical fiber is very weak , requiring ultra - sensitive photodetectors for detection . considering the limited laser beam intensity and the need to split into many beams to serve a set of interferometers , it is vital to increase the geometrical efficiency . to this end
, a collimator is built by placing an optical fiber in a ferrule ( 1 mm diameter ) and gluing one end of the optical fiber to a grin lens .
the grin lens is a 0.25 pitch lens with 0.46 numerical aperture , 1 mm diameter and 2.58 mm length which is optimized for a wavelength of 630 nm .
the density of the outgoing beam from the optical fiber is increased by a factor of approximately 1000 by using a grin lens .
the return beams are received by another optical fiber and amplified by a si femtowatt photoreceiver with a gain of @xmath21 .
for a fsi system , drifts and vibrations occurring along the optical path during the scan will be magnified by a factor of @xmath22 , where @xmath12 is the average optical frequency of the laser beam and @xmath23 is the scanned frequency . for the full scan of our laser , @xmath24 .
small vibrations and drift errors that have negligible effects for many optical applications may have a significant impact on a fsi system . a single - frequency vibration may be expressed as @xmath25 , where @xmath26 , @xmath27 and @xmath28 are the amplitude , frequency and phase of the vibration respectively . if @xmath29 is the start time of the scan , eq .
( 2 ) can be re - written as @xmath30/c \end{array } \eqno{(3)}\ ] ] if we approximate @xmath31 , the measured optical path difference @xmath32 may be expressed as @xmath33 \times \\ \sin[\pi f_{vib}(t+t_0)+\phi_{vib } ] \end{array } \eqno{(4)}\ ] ] where @xmath34 is the true optical path difference without vibration effects .
if the path - averaged refractive index of ambient air @xmath35 is known , the measured distance is @xmath36 .
if the measurement window size @xmath37 is fixed and the window used to measure a set of @xmath38 is sequentially shifted , the effects of the vibration will be evident .
we use a set of distance measurements in one scan by successively shifting the fixed - length measurement window one f - p peak forward each time .
the arithmetic average of all measured @xmath38 values in one scan is taken to be the measured distance of the scan ( although more sophisticated fitting methods can be used to extract the central value ) . for a large number of distance measurements @xmath39
, the vibration effects can be greatly suppressed .
of course , statistical uncertainties from fringe and frequency determination , dominant in our current system , can also be reduced with multiple scans .
averaging multiple measurements in one scan , however , provides similar precision improvement as averaging distance measurements from independent scans , and is faster , more efficient , and less susceptible to systematic errors from drift . in this way
, we can improve the distance accuracy dramatically if there are no significant drift errors during one scan , caused , for example , by temperature variation .
this multiple - distance - measurement technique is called slip measurement window with fixed size , shown in fig.2 .
however , there is a trade off in that the thermal drift error is increased with the increase of @xmath39 because of the larger magnification factor @xmath40 for a smaller measurement window size . in order to extract the amplitude and frequency of the vibration , another multiple - distance - measurement technique called slip measurement window with fixed start point
is shown in fig.2 . in eq .
( 3 ) , if @xmath29 is fixed , the measurement window size is enlarged one f - p peak for each shift , an oscillation of a set of measured @xmath38 values reflects the amplitude and frequency of vibration .
this technique is not suitable for distance measurement because there always exists an initial bias term including @xmath29 which can not be determined accurately in our current system .
the typical measurement residual versus the distance measurement number in one scan using the above technique is shown in fig.3(a ) , where the scanning rate was 0.5 nm / s and the sampling rate was 125 ks / s .
measured distances minus their average value for 10 sequential scans are plotted versus number of measurements ( @xmath39 ) per scan in fig.3(b ) . the standard deviations ( rms ) of distance measurements for 10 sequential scans
are plotted versus number of measurements ( @xmath39 ) per scan in fig.3(c ) .
it can be seen that the distance errors decrease with an increase of @xmath39 .
the rms of measured distances for 10 sequential scans is 1.6 @xmath1 if there is only one distance measurement per scan ( @xmath41 ) .
if @xmath42 and the average value of 1200 distance measurements in each scan is considered as the final measured distance of the scan , the rms of the final measured distances for 10 scans is 41 nm for the distance of 449828.965 @xmath1 , the relative distance measurement precision is 91 ppb .
some typical measurement residuals are plotted versus the number of distance measurements in one scan(@xmath39 ) for open box and closed box data with scanning rates of 2 nm / s and 0.5 nm / s in fig.4(a , b , c , d ) , respectively .
the measured distance is approximately 10.4 cm .
it can be seen that the slow fluctuations of multiple distance measurements for open box data are larger than that for closed box data .
the standard deviation ( rms ) of measured distances for 10 sequential scans is approximately 1.5 @xmath1 if there is only one distance measurement per scan for closed box data . by using multiple - distance - measurement technique ,
the distance measurement precisions for various closed box data with distances ranging from 10 cm to 70 cm collected in the past year are improved significantly , precisions of approximately 50 nanometers are demonstrated under laboratory conditions , as shown in table 1 .
all measured precisions listed in the table 1 .
are the rms of measured distances for 10 sequential scans .
two fsi demonstration systems , air fsi and optical fiber fsi , are constructed for extensive tests of multiple - distance - measurement technique , air fsi means fsi with the laser beam transported entirely in the ambient atmosphere , optical fiber fsi represents fsi with the laser beam delivered to the interferometer and received back by single - mode optical fibers .
.distance measurement precisions for various setups using the multiple - distance - measurement technique .
[ cols="^,^,^,^,^ " , ] based on our studies , the slow fluctuations are reduced to a negligible level by using the plastic box and pvc pipes to suppress temperature fluctuations .
the dominant error comes from the uncertainties of the interference fringes number determination ; the fringes uncertainties are uncorrelated for multiple distance measurements . in this case
, averaging multiple distance measurements in one scan provides a similar precision improvement to averaging distance measurements from multiple independent scans , but is faster , more efficient and less susceptible to systematic errors from drift .
but , for open box data , the slow fluctuations are dominant , on the order of few microns in our laboratory .
the measurement precisions for single and multiple distance open - box measurements are comparable , which indicates that the slow fluctuations can not be adequately suppressed by using the multiple - distance - measurement technique .
a dual - laser fsi system@xcite intended to cancel the drift error is currently under study in our laboratory ( to be described in a subsequent article ) . from fig.4(d ) , we observe periodic oscillation of the distance measurement residuals in one scan , the fitted frequency is @xmath43 hz for the scan .
the frequency depends on the scanning rate , @xmath44 . from eq.(4 ) , it is clear that the amplitude of the vibration or oscillation pattern for multiple distance measurements depends on @xmath45 $ ] .
if @xmath26 , @xmath27 are constant values , it depends on the size of the distance measurement window .
subsequent investigation with a ccd camera trained on the laser output revealed that the apparent @xmath03 hz vibration during the 0.5 nm / s scan arose from the beam s centroid motion . because the centroid motion is highly reproducible ,
we believe that the effect comes from motion of the internal hinged mirror in the laser used to scan its frequency .
the measurable distance range is limited in our current optical fiber fsi demonstration system for several reasons . for a given scanning rate of 0.25 nm / s ,
the produced interference fringes , estimated by @xmath46 , are approximately 26400 in a 40-second scan for a measured distance ( @xmath47 ) of 60 cm , that is @xmath48 fringes / s , where @xmath23 is the scanned frequency and @xmath49 is the speed of light .
the currently used femtowatt photoreceiver has 3 db frequency bandwidth ranging from 30 - 750 hz , the transimpedance gain decreases quickly beyond 750 hz .
there are two ways to extend the measurable distance range .
one straightforward way is to extend the effective frequency bandwidth of the femtowatt photoreceiver ; the other way is to decrease the interference fringe rate by decreasing the laser scanning rate .
there are two major drawbacks for the second way ;
one is that larger slow fluctuations occur during longer scanning times ; the other is that the laser scanning is not stable enough to produce reliable interference fringes if the scanning rate is lower than 0.25 nm / s for our present tunable laser .
in addition , another limitation to distance range is that the intensity of the return beam from the retroreflector decreases inverse - quadratically with range .
in order to test the vibration measurement technique , a piezoelectric transducer ( pzt ) was employed to produce vibrations of the retroreflector . for instance , the frequency of the controlled vibration source was set to @xmath50 hz with amplitude of @xmath51 . for @xmath52 distance measurements in one scan ,
the magnification factor for each distance measurement depends on the scanned frequency of the measurement window , @xmath53 , where , @xmath12 is the average frequency of the laser beam in the measurement window , scanned frequency @xmath54 ghz , where i runs from 1 to @xmath39 , shown in fig.5(a ) .
the distance measurement residuals for 2000 distance measurements in the scan are shown in fig.5(b ) , the oscillation of the measurement residuals reflect the vibration of the retroreflector .
since the vibration is magnified by a factor of @xmath55 for each distance measurement , the corrected measurement residuals are measurement residuals divided by the corresponding magnification factors , shown in fig.5(c ) . the extracted vibration frequencies and amplitudes using this technique are @xmath56 hz , @xmath57 , respectively , in good agreement with expectations .
another demonstration was made for the same vibration frequency , but with an amplitude of only @xmath58 nanometers .
the magnification factors , distance measurement residuals and corrected measurement residuals for 2000 measurements in one scan are shown in fig.6(a ) , fig.6(b ) and fig.6(c ) , respectively .
the extracted vibration frequencies and amplitudes using this technique are @xmath59 hz , @xmath60 nanometers .
in addition , vibration frequencies at 0.1 , 0.5 , 1.0 , 5 , 10 , 20 , 50 , 100 hz with controlled vibration amplitudes ranging from 9.5 nanometers to 400 nanometers were studied extensively using our current fsi system .
the measured vibrations and expected vibrations all agree well within the 10 - 15% level for amplitudes , 1 - 2% for frequencies , where we are limited by uncertainties in the expectations .
vibration frequencies far below 0.1 hz can be regarded as slow fluctuations , which can not be suppressed by the above analysis techniques . for comparison ,
nanometer vibration measurement by a self - aligned optical feedback vibrometry technique has been reported@xcite .
the vibrometry technique is able to measure vibration frequencies ranging from 20 hz to 20 khz with minimal measurable vibration amplitude of 1 nm . our second multiple - distance - measurement technique demonstrated above has capability to measure vibration frequencies ranging from 0.1 hz to 100 hz with minimal amplitude on the level of several nanometers , without a _ priori _ knowledge of the vibration strengths or frequencies .
dispersive elements such as a beamsplitter , corner cube prism , etc . in the interferometer can create an apparent offset in measured distance for an fsi system , since the optical path length of the dispersive element changes during the scan . the small opd change caused by dispersion
is magnified by a factor of @xmath40 and has a significant effect on the absolute distance measurement for the fsi system . the measured optical path difference @xmath32
may be expressed as @xmath61 where @xmath62 and @xmath63 refer to the opd at times t and t0 , respectively , @xmath64 and @xmath65 are the wavelength of the laser beam at times t and t0 , c is the speed of light , d1 and d2 are true geometrical distances in the air and in the corner cube prism , @xmath66 and @xmath67 are the refractive index of ambient atmosphere and the refractive index of the corner cube prism for @xmath64 , respectively . the measured distance @xmath36 , where @xmath35 is the average refractive index around the optical path .
the sellmeier formula for dispersion in crown glass ( bk7)@xcite can be written as , @xmath68 where , the beam wavelength @xmath69 is in unit of microns , @xmath70 , @xmath71 , @xmath72 , @xmath73 , @xmath74 , @xmath75 .
if we use the first multiple - distance - measurement technique described above to make 2000 distance measurements for one typical scan , where the corner cube prism is used as retroreflector , we observe a highly reproducible drift in measured distance , as shown in fig.7 , where the fitted distance drift is @xmath76 microns for one typical scan using a straight line fit .
however , there is no apparent drift if we replace the corner cube prism by the hollow retroreflector .
numerical simulations have been carried out using eq.(5 ) and eq.(6 ) to understand the above phenomena .
for instance , consider the case d1 = 20.97 cm and d2 = 1.86 cm ( the uncertainty of d2 is 0.06 cm ) , where the first and the last measured distances among 2000 sequential distance measurements are denoted @xmath77 and @xmath78 , respectively . using the sellmeier equation ( eq.(6 ) ) for modeling the corner cube prism material ( bk7 ) dispersion , we expect @xmath79 = 373.876 microns and @xmath80 = 367.707 microns .
the difference between @xmath77 and @xmath78 is @xmath81 microns which agrees well with our observed @xmath76 microns drift over 2000 measured distances .
the measured distance shift and drift strongly depend on d2 , but are insensitive to d1 .
a change of 1 cm in d1 leads to a 3-nanometer distance shift , but the same change in d2 leads to a 200-micron distance shift . if a beamsplitter is oriented with its reflecting side facing the laser beam , then there is an additional dispersive distance shift .
we have verified this effect with 1-mm and a 5-mm beam splitters .
when we insert an additional beamsplitter with 1 mm thickness between the retroreflector and the original beamsplitter in the optical fiber fsi system , we observe a 500 microns shift on measured distance if @xmath35 is fixed consistent with the numerical simulation result . for the 5-mm beam splitter ( the measured thickness of the beam splitter
is @xmath82 mm ) , the first 20 scans were performed with the beamsplitter s anti - reflecting surface facing the optical fibers and the second 20 scans with the reflecting surface facing the optical fibers .
the expected drifts ( @xmath83 ) for the first and the second 20 scans from the dispersion effect are 0 and @xmath84 microns , respectively .
the measured drifts by averaging measurements from 20 sequential scans are @xmath85 microns and @xmath86 microns , respectively .
the measured values agree well with expectations .
in addition , the dispersion effect from air@xcite is also estimated by using numerical simulation .
the expected drift ( @xmath83 ) from air dispersion is approximately -0.07 microns for an optical path of 50 cm in air , this effect can not be detected for our current fsi system .
however , it could be verified by using a fsi with a vacuum tube surrounding the laser beam ; the measured distance with air in the tube would be approximately 4 microns larger than for an evacuated tube . in summary , dispersion effects can have a significant impact on absolute distance measurements , but can be minimized with care for elements placed in the interferometer or corrected for , once any necessary dispersive elements in the interferometer are understood .
some major error sources are estimated in the following ; \1 ) error from uncertainties of fringe and scanned frequency determination .
the measurement precision of the distance @xmath87 ( the error due to the air s refractive index uncertainty is considered separately below ) is given by @xmath88 . where @xmath87 , @xmath13 , @xmath23 , @xmath89 , @xmath90 , @xmath91 are measurement distance , fringe numbers , scanned frequency and their corresponding errors . for a typical scanning rate of 0.5 nm / s with a 10 nm scan range , the full scan time is 20 seconds .
the total number of samples for one scan is 2.5 ms at a sampling rate of 125 ks / s .
there is about a 4@xmath05 sample ambiguity in fringe peak and valley position due to a vanishing slope and the limitation of the 12-bit sampling precision .
however , there is a much smaller uncertainty for the f - p peaks because of their sharpness .
thus , the estimated uncertainty is @xmath92 for one full scan for a magnification factor @xmath93 .
if the number of distance measurements @xmath94 , the distance measurement window is smaller , the corresponding magnification factor is @xmath95 , where , @xmath12 is the average frequency of the laser beam , @xmath96 ghz .
one obtains @xmath97 , @xmath98 .
\2 ) error from vibrations . the detected amplitude and frequency for vibration ( without controlled vibration source )
are about 0.3 @xmath1 and 3.2 hz .
the corresponding time for @xmath94 sequential distance measurements is 5.3 seconds .
a rough estimation of the resulting error gives @xmath99 for a given measured distance @xmath100 meters .
\3 ) error from thermal drift .
the refractive index of air depends on air temperature , humidity and pressure ( fluctuations of humidity and pressure have negligible effects on distance measurements for the 20-second scan ) .
temperature fluctuations are well controlled down to about @xmath20(rms ) in our laboratory by the plastic box on the optical table and the pipe shielding the volume of air near the laser beam . for
a room temperature of 21 @xmath101 , an air temperature change of @xmath102 will result in a 0.9 ppm change of air refractive index .
for a temperature variation of @xmath20 in the pipe , @xmath94 distance measurements , the estimated error will be @xmath103 , where the magnification factor @xmath104 .
the total error from the above sources , when added in quadrature , is @xmath105 , with the major error sources arising from the uncertainty of fringe determination and the thermal drift .
the estimated relative error agrees well with measured relative spreads of @xmath106 in real data for measured distance of about 0.45 meters . besides the above error sources
, other sources can contribute to systematic bias in the absolute differential distance measurement .
the major systematic bias comes from the uncertainty in the fsr of the f - p used to determine the scanned frequency range .
the relative error would be @xmath107 if the fsr were calibrated by a wavemeter with a precision of @xmath108 .
a wavemeter of this precision was not available for the measurements described here .
the systematic bias from the multiple - distance - measurement technique was also estimated by changing the starting point of the measurement window , the window size and the number of measurements , the uncertainties typically range from 10 to 50 nanometers . systematic bias from uncertainties in temperature , air humidity and barometric pressure scales
are estimated to be negligible .
an optical fiber fsi system was constructed to make high - precision absolute distance and vibration measurements .
a design of the optical fiber with grin lens was presented which improves the geometrical efficiency significantly .
two new multiple - distance - measurement analysis techniques were presented to improve distance precision and to extract the amplitude and frequency of vibrations .
absolute distance measurement precisions of approximately 50 nm for distances ranging from 10 cm to 70 cm under laboratory conditions were achieved using the first analysis technique . the second analysis technique measures vibration frequencies ranging from 0.1 hz to 100 hz with minimal amplitude of a few nanometers .
we verified an expected dispersion effect and confirmed its importance when dispersive elements are placed in the interferometer .
major error sources were estimated , and the observed errors were found to be in good agreement with expectation .
fox - murphy , d.f .
howell , r.b .
nickerson , a.r .
weidberg , `` frequency scanned interferometry(fsi ) : the basis of a survey system for atlas using fast automated remote interferometry '' , nucl .
inst . meth .
a383 , 229 - 237(1996 ) american linear collider working group(161 authors ) , `` linear collider physics , resource book for snowmass 2001 '' , prepared for the department of energy under contract number de - ac03 - 76sf00515 by stanford linear collider center , stanford university , stanford , california .
hep - ex/0106058 , slac - r-570 299 - 423(2001 ) p. a. coe , `` an investigation of frequency scanning interferometry for the alignment of the atlas semiconductor tracker '' , doctoral thesis , st .
peter s college , university of oxford , keble road , oxford , united kingdom , 1 - 238(2001 ) | in this paper , we report high - precision absolute distance and vibration measurements performed with frequency scanned interferometry using a pair of single - mode optical fibers .
absolute distance was determined by counting the interference fringes produced while scanning the laser frequency .
a high - finesse fabry - perot interferometer(f - p ) was used to determine frequency changes during scanning .
two multiple - distance - measurement analysis techniques were developed to improve distance precision and to extract the amplitude and frequency of vibrations . under laboratory conditions , measurement precision of @xmath0 50 nm
was achieved for absolute distances ranging from 0.1 meters to 0.7 meters by using the first multiple - distance - measurement technique .
the second analysis technique has the capability to measure vibration frequencies ranging from 0.1 hz to 100 hz with amplitude as small as a few nanometers , without a _ priori _ knowledge . |
neutrino flavor oscillations @xcite with wavelength comparable to the earth - sun distance @xcite represent a solution @xcite to the deficit of solar @xmath4 s @xcite observed in the four pioneering underground experiments homestake @xcite , kamiokande @xcite , sage @xcite , and gallex @xcite , as compared to the standard solar model predictions @xcite .
the recent superkamiokande data @xcite confirm the deficit , and can be interpreted within the vacuum oscillation hypothesis as well @xcite . the planned borexino solar neutrino experiment @xcite ( in construction ) , designed to detect monochromatic @xmath5be neutrinos ( @xmath6 mev ) , is expected to test this hypothesis with unprecedented sensitivity @xcite .
an update of the current neutrino flux measurements @xcite is given in table i. figure 1 shows our vacuum oscillation fit to the data of table i , as obtained from a @xmath7 analysis ( including solar model uncertainties as in @xcite ) .
we have assumed , for simplicity , two neutrino families .
it can be seen that four regions ( a , b , c , and d ) are allowed at 95% c.l .
, the absolute minimum being located within the solution b ( @xmath8 and @xmath9 ) .
various tests can be envisaged to discriminate among the four solutions in fig . 1 .
in particular , since the neutrino oscillation length is proportional to the pathlength - to - energy ratio @xmath0 , deviations of event distributions from the expected shape in either @xmath2 or @xmath3 ( or related parameters ) represent direct tests of neutrino vacuum oscillations ( see , e.g. , @xcite ) .
it turns out that , in general , superkamiokande is more ( less ) sensitive than borexino to @xmath3-related ( @xmath2-related ) spectral shape deviations ; therefore , the two experiments provide complementary tools to study the vacuum oscillation hypothesis .
the purpose of this work is to investigate in detail the tests of the vacuum oscillation hypothesis that can be performed at superkamiokande and borexino , and their interplay .
the results will be shown in a form that makes easy to derive the experimental accuracy needed to perform a specific test .
the plan of this paper is as follows . in sec .
ii we discuss the tests of energy spectra deviations , in the light of the recent superkamiokande results . in sec .
iii we discuss in detail the fourier analysis of the signal , building upon our previous work @xcite . in sec .
iii we apply these tests to superkamiokande and borexino , both separately and jointly . in sec .
iv we draw our conclusions . some technical aspects of our analysis are described in the appendix .
both superkamiokande and borexino can measure the energy spectrum of recoil electrons from neutrino scattering .
the standard ( i.e. , no oscillation ) superkamiokande spectrum can be found in fig . 4 of ref .
@xcite . for completeness
, we show in fig .
2 the standard electron energy spectrum in borexino ( details about the inclusion of energy threshold and resolution effects are given in the appendix ) .
the main contribution to the spectrum in fig .
2 is given by the 0.86 mev @xmath5be line , which is responsible for the compton edge at @xmath10 mev .
the edge is smeared by the finite energy resolution .
the rise at low energies is due to @xmath11 neutrinos , while the tail at high energies is basically due to cno neutrinos .
the prospective analysis window is also indicated in fig . 2 .
the standard neutrino fluxes have been taken from @xcite .
distortions of the neutrino energy spectrum due to oscillations are generally reflected ( although somewhat degraded ) in the electron energy spectrum .
an effective parametrization of such distortions is given by the fractional variation of the average kinetic energy @xmath12 of the electron , an approach extensively developed in @xcite and applied to the superkamiokande data in @xcite .
in particular , the analysis @xcite of the most recent ( 374 day ) measurements of the electron spectrum in superkamiokande @xcite gives a fractional deviation @xmath13 such deviation is consistent with zero ( no oscillation ) at the level of @xmath14 , and disfavors scenarios predicting negative values for @xmath15 .
concerning borexino , the expected _ shape _ variations of the energy distribution are very small , since the main contribution to the electron spectrum comes from a monoenergetic source of neutrinos ( @xmath5be ) , rather than from a continuous source as in superkamiokande ( @xmath16b ) .
neutrino oscillations are expected to change significantly the global borexino rate but not its energy spectrum .
the variations of the solar neutrino pathlength @xmath2 due to the eccentricity of the earth orbit produce a geometrical @xmath17 modulation of the neutrino flux .
additional semiannual modulations are expected in the presence of neutrino oscillations @xcite .
the fourier analysis of the measured flux represents an effective tool to study both kinds of @xmath2-related modulations @xcite . in this section
we outline the fourier analysis of the signal observable in solar neutrino experiments . in the first three subsections
, we describe the general formalism and the results for the `` no oscillation '' and `` @xmath18 oscillation '' cases ( see also ref .
@xcite for further details ) . in the last two subsections we generalize the analysis to @xmath19 oscillations and
then discuss a useful check of both the symmetry properties and the estimated uncertainties of the signal .
the earth orbit radius , @xmath2 , varies periodically in time @xmath20 around its average value , @xmath21 km , according to @xmath22 where @xmath23 is the orbit eccentricity and @xmath24 yr ( @xmath25 at the perihelion ) .
terms of @xmath26 or higher are negligible for our purposes .
the neutrino signal @xmath27 is also , in general , a periodic function . assuming a constant background @xmath28 , the total observed neutrino rate @xmath29 is : @xmath30 for symmetry reasons , the analysis can be restricted to the time interval @xmath31 $ ] .
it is understood that events collected in subsequent half - years must be symmetrically folded in this interval .
the data sample consists then of @xmath32 events collected at different times @xmath33 , with @xmath34 $ ] and @xmath32 equal to the total sum of background and signal events , @xmath35 .
notice that , in general , one can determine @xmath36 and @xmath37 but can not distinguish background and signal on an event - by - event basis .
the expansion of the signal in terms of fourier components @xmath38 reads @xmath39 where @xmath27 is the time - averaged signal @xmath40 the @xmath41-th harmonic corresponds to a period of @xmath42 yr .
the explicit form of @xmath38 reads @xmath43 where eqs .
( [ fntheo ] ) and ( [ fnexpt ] ) represent the theoretical definition and the experimental determination of the @xmath38 s , respectively @xcite . assuming purely statistical fluctuations of the signal and of the background , the variance of the @xmath38 s reads : @xmath44 it turns out that the values of the @xmath45 s are @xmath46 in all cases of practical interest .
therefore , to a good approximation , the one - sigma statistical error @xmath47 affecting @xmath38 is given by @xmath48 for _ any _ @xmath41 .
the correlations between the statistical errors of different harmonics are also negligible @xcite . finally , the general expression of the signal @xmath27 expected in the presence of oscillations is @xmath49 , \label{sosc}\ ] ] where @xmath50 is the neutrino energy , @xmath51 is the neutrino energy spectrum , @xmath52 ( @xmath53 ) is the @xmath1 ( @xmath54 , @xmath55 ) interaction cross section , and @xmath56 is the @xmath1 survival probability , which varies in time through @xmath57 @xcite .
it is understood that the cross sections @xmath58 must be corrected for energy threshold and resolution effects , as described in the appendix . in the standard ( no oscillation ) case , characterized by @xmath59 ,
the signal @xmath27 varies as @xmath60 .
the standard fourier components are simply given by @xmath61 i.e. , only the first harmonic is nonzero and measures the earth s orbit eccentricity . in this case
, @xmath1 is a linear combination of two mass eigenstates @xmath63 characterized by a mixing angle @xmath64 @xmath65 and by a squared mass difference @xmath66 , @xmath67 the corresponding @xmath1 survival probability is given by @xmath68 the fourier components can be cast in the following , compact form @xcite : @xmath69 where the ( detector - dependent ) functions @xmath70 are given by @xmath71 and the universal ( i.e. , detector - independent ) functions @xmath72 are given by @xmath73\ , \ ] ] @xmath74 where @xmath75 and @xmath76 is the bessel function of order @xmath41 .
notice that , although our calculations are of @xmath77 ) , all orders in @xmath78 are kept , since @xmath79 may be large . in this case , @xmath1 is a linear combination of three mass eigenstates @xmath81 , usually parametrized in one of the following two forms : @xmath82 where @xmath83 , @xmath84 , and @xmath85 ( see , e.g. , @xcite ) . in addition , we assume a hierarchy of mass differences : @xmath86 the hypothesis ( [ m2 ] ) covers most of the situations of phenomenological interest @xcite . in this case
, the @xmath19 oscillation probability is given by @xmath87 we have worked out the corresponding fourier components , which read : @xmath88 -4\ , u^2_{e1 } u^2_{e2}\,d_n(\delta m^2 ) } { 1 - 4\ , u^2_{e3}(1-u^2_{e3})\,r- 4\,u^2_{e1 } u^2_{e2}\,d_0(\delta m^2)}\\ & = & \frac{\varepsilon\,\delta_{n1}(1-s^2_{2\phi}r ) -s^2_{2\omega}c^4_\phi\,d_n(\delta m^2 ) } { 1-s^2_{2\phi}r - s^2_{2\omega}c^4_\phi d_0(\delta m^2 ) } \label{f3nu}\end{aligned}\ ] ] where the functions @xmath70 are defined in eq .
( [ dn ] ) and @xmath89 notice that @xmath90 for @xmath91 , as it should . _ a priori _
, the signal @xmath92 must obey the symmetry @xmath93 in the one - year interval @xmath94 $ ] ( either with or without oscillations ) .
we have made use of this property in eqs .
( [ s][fnexpt ] ) .
the experimental test of such symmetry property is not without merit , since its failure might signal systematic , time - dependent effects , such as unexpected variations of the detection efficiency or of the background level . within our approach
, this is equivalent to check that the `` sine '' fourier components @xmath95 , defined as @xmath96 are identically zero , as they should ( both in the standard case and in the presence of possible oscillations ) : @xmath97 notice that , in eq .
( [ gnexpt ] ) , the event `` arrival times '' are folded in the interval @xmath94 $ ] and not in @xmath98 $ ] .
the experimentally inferred @xmath95 s will be distributed around zero with a variance @xmath99 . as far as statistical fluctuations are concerned
, the calculation of @xmath99 is analogous to @xmath100 @xcite and gives : @xmath101 however , as already noticed , the @xmath45 s are generally @xmath46 and thus can be neglected in the above equation .
therefore , the statistical uncertainty @xmath102 is approximately equal for all the @xmath95 s and has the same expression as @xmath103 [ eq .
( [ sigma ] ) ] . in conclusion , it is useful to check that the semiannual variations of the solar neutrino signal are indeed symmetric in time .
this implies that the `` sine '' fourier components @xmath95 defined in eq .
( [ gnexpt ] ) should form a distribution with mean value @xmath104 and standard deviation @xmath105 .
this test is universal , i.e. it is valid with or without oscillation effects .
any deviation of the mean and variance of the @xmath95 distribution from their standard values ( 0 and @xmath106 , respectively ) would indicate the presence ( and the magnitude ) of systematic effects beyond the purely statistical fluctuations of either the background or the signal .
these effects ( if any ) should then be accounted for by the experimentalists , before performing an unbiased analysis of the time variations of the signal .
in this section we analyze the tests of the vacuum oscillation hypothesis that can be performed at superkamiokande and borexino , both separately and jointly .
we consider two observables for each experiment .
in particular , we analyze @xmath107 and @xmath108 for superkamiokande ( sk ) , and @xmath108 and @xmath109 for borexino ( bx ) : @xmath110 all the above variables are zero in the standard ( no oscillation ) case . in the oscillation range of interest ,
fourier components with @xmath111 ( @xmath112 ) are not relevant for superkamiokande ( borexino ) @xcite . for simplicity , we will consider only @xmath18 oscillations , and the corresponding preferred regions a , b , c , and d of fig . 1 .
figure 3 shows the four solutions a , b , c , and d ( gray regions ) and the no oscillation point ( star ) in the plane charted by the parameters @xmath107 ( the fractional deviation of the mean electron kinetic energy ) and @xmath113 ( the deviation of the first fourier component from its standard value ) .
also shown is the horizontal band allowed at @xmath114 by the @xmath107 datum of eq .
( [ dt ] ) .
figure 3 evidences that solutions c and d ( which predict large , negative values for @xmath107 ) are highly disfavored by the superkamiokande measurement of eq .
( [ dt ] ) .
in particular , solution c is disfavored at @xmath115 and solution d at @xmath116 . on the other hand , the datum of eq .
( [ dt ] ) is unable to discriminate among the solutions a , b , and the no oscillation point at the @xmath117 level , although solution a seems to be preferred .
we can summarize these findings by saying that , under the hypothesis of @xmath18 oscillations , the _ combined _ data of table i and eq . ( [ dt ] ) select the solutions a and b in fig . 1 , corresponding to the following approximate ranges ( at 95% c.l . ) for the neutrino mass - mixing parameters : @xmath118 notice that the spread of the above parameters is only about @xmath119 . also notice that @xmath18 maximal mixing @xmath120 is allowed only in a restricted range of @xmath66 ( @xmath121@xmath122 ev@xmath123 ) .
concerning @xmath124 , we can not infer its value from the limited data which are publicly available @xcite . at any rate , the estimated uncertainty @xmath103 of @xmath124
[ see eq .
( [ sigma ] ) ] for @xmath125 yr of data taking is about @xmath126 @xcite too large to discriminate any of the solutions a , b , c , and d. significantly higher statistics are needed to reduce such error
. an interesting feature of fig .
3 is the tight correlation between the variables @xmath108 and @xmath127 , which parametrize @xmath2-related and @xmath3-related spectral distortions , respectively .
such correlation can be traced to the @xmath0 dependence of the oscillation probability , as emphasized in @xcite .
one can use such correlation to `` predict '' the value of @xmath108 in superkamiokande for a given value of @xmath107 in eq .
( [ dt ] ) .
more precisely , from fig . 3
one derives that the values of @xmath108 compatible with both the horizontal band and the solution a should lie in the range @xmath128@xmath129 . in other words ,
one expects an enhancement of the semiannual modulations of the neutrino flux ( relatively to the purely geometrical one ) in the range @xmath130=30$]@xmath131 at @xmath132 .
this is a clear prediction that needs , however , several years of data taking at superkamiokande to be tested . figure 4 shows the four solutions a , b , c , and d ( gray regions ) in the plane charted by the parameters @xmath133 and @xmath134 ( the deviation of the first two fourier components from their standard values ) .
the no oscillation point ( star ) corresponds to the origin .
the amplitude of the second harmonic appears to be generally smaller than the first ; nevertheless , both should be detectable in a sample of a few thousand events .
e.g. , for @xmath135 events , the expected statistical error of @xmath124 and @xmath109 is only about @xmath136 [ see eq .
( [ sigma ] ) ] , ensuring clear detection of semiannual modulations ( provided that systematics do not dominate the error budget ) .
however , the four solutions a , b , c , d are rather close to each other in the fourier parameter space , and it might be difficult to distinguish among them using only these two variables . by comparing figs . 3 and 4 , it can be noticed that the solutions a and b ( currently favored by superkamiokande data ) predict rather different values for the first fourier harmonic in superkamiokande and borexino , as a result of the different energy ranges probed by these two experiments . in particular , within solution a it is always @xmath137 , while @xmath133 can be either positive or negative .
therefore , at present the sign of the semiannual variations in borexino is unpredictable .
it is interesting to notice that , both in @xmath18 and @xmath19 oscillations , the fourier component ratio @xmath138 depends only on @xmath66 and not on the mixing angle(s ) [ see eqs .
( [ f2nu ] ) and ( [ f3nu ] ) ] .
therefore , such ratio can constrain the value of @xmath66 in a model - independent way .
figure 5 shows the four solutions a , b , c , and d ( gray regions ) in the four planes charted by the borexino observables @xmath133 and @xmath134 ( @xmath139-axes ) and by the superkamiokande observables @xmath113 and @xmath140 ( @xmath141-axes ) .
also shown is the horizontal band allowed at @xmath114 by the @xmath107 datum of eq .
( [ dt ] ) . within such band , the values of @xmath133 ( as well as those of @xmath134 ) can be either positive or negative , as observed in the previous subsection .
figure 5 shows that the present indeterminacy in the sign of @xmath133 can not be resolved by increasing the accuracy of the superkamiokande data .
the similarity between the upper and lower panels in fig .
5 is due to the tight correlation between the superkamiokande variables @xmath113 and @xmath140 .
this similarity should be reflected in the experimental data , if vacuum oscillations indeed occur .
when experimental data will be available for all the four observables charting the panels in fig . 5 , one of the four solutions should be easily spotted in at least one of the four panels .
however , in the unlucky situation of data points close to the no oscillation case , it might be difficult to distinguish between such case and solution b. the possibility to separate the no oscillation point from solution b would then depend decisively on the reduction of the @xmath140 uncertainty ( see upper panels of fig . 5 ) . in any case
, the reader can judge the discriminating power of the two experiments by drawing prospective data points and error bars in each panel of fig .
we have studied tests of vacuum oscillations of solar neutrinos in the superkamiokande and borexino experiments , both separately and jointly .
the tests are sensitive to either energy or time variations of the neutrino flux .
the interplay between such tests has been investigated , in the light of the most recent data ( 374 day ) from the superkamiokande experiment .
the results have been displayed in a graphical form ( figs .
35 ) that allows to determine easily the experimental accuracy needed to test the vacuum oscillation solution(s ) .
we have found that : ( i ) the total neutrino rates measured by solar @xmath4 experiments can be fitted in four distinct regions of the mass - mixing parameter space ; ( @xmath142 ) two of the four solutions are strongly disfavored by the superkamiokande energy spectrum data ; @xmath143 the energy spectrum data do not discriminate significantly ( at present ) the remaining two solutions between them and from the no oscillation case ; ( @xmath144 the amplitude of semiannual variations of the solar @xmath4 flux in superkamiokande is predicted to be about 3060% in excess of the purely geometric one ( at @xmath145 ) ; @xmath146 the sign of semiannual variations ( due to oscillations ) in borexino is not determined by present data ; @xmath147 the _ joint _ information coming from the energy spectrum data ( superkamiokande ) and from the fourier transform of the solar neutrino rates ( superkamiokande , borexino ) can provide powerful tests of the vacuum oscillation solutions . besides , we have generalized the fourier transform formalism to three - flavor oscillations , and we have discussed a useful check of the time symmetry of the signal .
we are grateful to j. g. learned for useful discussions and helpful suggestions .
el thanks the centro de fsica das interaces fundamentais at the instituto superior tcnico of the university of lisbon , portugal , for kind hospitality during the preparation of this work .
bf thanks the dipartimento di fisica of bari , italy , for hospitality .
in the calculation of the expected signal [ eq .
( [ sosc ] ) ] , it is understood that the @xmath148-@xmath149 cross sections @xmath150 @xmath151 have to be properly corrected to take into account the detector energy resolution and the analysis window for each experiment . here
we give some details about such corrections .
both in borexino and in superkamiokande , the finite energy resolution due to the photon statistics implies that the _ measured _ kinetic energy @xmath152 of the scattered electron is distributed around the _ true _ kinetic energy @xmath153 according to a resolution function @xmath154 of the form @xcite : @xmath155\ , \ ] ] where @xmath156 and @xmath157 kev and 0.47 mev for borexino @xcite and superkamiokande @xcite , respectively . on the other hand ,
the distribution of the true kinetic energy @xmath153 for an interacting neutrino of energy @xmath3 is dictated by the differential cross section @xmath158 , that we take from @xcite .
the kinematic limits are : @xmath159 concerning the measured kinetic energy , the present analysis window for superkamiokande is @xmath160=[6.5{\rm\ mev}-m_e,\,20{\rm\ mev}-m_e]$ ] . the prospective analysis window for borexino ( as used in this paper ) is @xmath161 $ ] mev . for assigned values of @xmath162 , @xmath163 , and @xmath164 , the corrected cross section @xmath165 is defined as : @xmath166 it is useful to reorder the integrands , obtaining @xmath167 where the function @xmath168 , which embeds the detector specifications @xmath162 , @xmath163 , and @xmath164 , is given by @xmath169\ , \ ] ] with @xmath170 as in eq .
( [ delta ] ) and @xmath171 borexino collaboration , c. arpesella _ et al .
_ , `` infn borexino proposal , '' vols .
i and ii , edited by g. bellini , r. raghavan , _ et al .
_ ( university of milan , 1992 ) ; j. benziger , f. p. calaprice _
et al . _ , `` proposal for participation in the borexino solar neutrino experiment , '' ( princeton university , 1996 ) .
borexino collaboration , g. bellini _ et al .
_ in _ taup 95 _ , proceedings of the 4th international workshop on theoretical and phenomenological aspects of underground physics , toledo , spain , edited by a. morales , j. morales , and j. a. villar [ nucl .
( proc . suppl . )
* b48 * , 547 ( 1996 ) ] , p. 363 ; g. bellini , in _ la thuile 97 _ , proceedings of the 11th rencontres de physique de la valle daoste , la thuile , italy , 1997 , ed . by m.
greco ( frascati physics series , vol .
9 ) , p. 27 .
k. lande ( homestake collaboration ) , in _
@xmath44 _ ,
proceedings of the 4th international solar neutrino conference , heidelberg , germany , 1997 , ed by w. hampel ( max planck institute , heidelberg , 1998 ) , p. 85 .
v. n. gavrin ( sage collaboration ) , in _ news about snu s _ , international workshop on solar neutrinos ( institute of theoretical physics , santa barbara , ca , 1997 ) .
proceedings available at the url http://www.itp.ucsb.edu/online/snu .
k. inoue , in _
taup 97 _ , proceedings of the 5th international workshop on topics in astroparticle and underground physics ( gran sasso , italy , 1997 ) , to appear in nucl .
b ( proc . suppl . ) ; also available at the same url as in ref .
@xcite .
y. suzuki , in _ neutrinos in astro , particle and nuclear physics _ , proceedings of the 19th international school of nuclear physics ( erice , italy , 1997 ) , to appear ; also available at the same url as ref .
@xcite .
fig . 1 .
vacuum oscillation solutions to the solar neutrino deficit in the usual mass - mixing plane , as derived from a fit to the data of table i. the four regions a , b , c , and d , are allowed at 95% c.l .
fig . 2 .
energy spectrum of recoil electrons in borexino .
the main components are shown separately .
the smearing effect of the energy resolution is also shown .
the arrows indicate the prospective energy window assumed in the analysis .
superkamiokande ( sk ) experiment .
map of the four vacuum oscillation solutions a , b , c , and d in the plane charted by @xmath107 ( the fractional deviation of the mean electron kinetic energy ) and by @xmath108 ( the deviation of the first fourier component from its standard value ) .
notice the strong correlation between these two variables , which is induced by the @xmath172 dependence of the oscillation probability .
the star at the origin corresponds to the standard ( no oscillation ) case .
the superkamiokande datum on @xmath107 is also shown ; it disfavors solution c at @xmath115 and solution d at @xmath173 . the part of the solution @xmath174 which is favored by the sk datum at @xmath145 corresponds to an expected fourier amplitude @xmath175@xmath176 .
borexino ( bx ) experiment .
map of the four vacuum oscillation solutions a , b , c , and d , in the plane charted by the first two fourier components .
the star at the origin corresponds to the no oscillation case .
there is some overlap among the four solutions .
superkamiokande ( sk ) _ vs _ borexino ( bx ) .
map of the four vacuum oscillation solutions a , b , c , and d in the planes charted by the sk observables @xmath107 and @xmath108 ( ordinates ) and by the bx observables @xmath108 and @xmath109 ( abscissae ) .
future measurements of the @xmath38 s in sk and bx , together with increasingly accurate @xmath107 data from sk , are expected to spot one of the solutions . | the vacuum oscillation solution to the solar neutrino problem predicts characteristics variations of the observable neutrinos rates , as a result of the @xmath0 dependence of the @xmath1 survival probability ( @xmath2 and @xmath3 being the neutrino pathlength and energy , respectively ) .
the @xmath3-dependence can be studied through distortions of the recoil electron spectrum in the superkamiokande experiment .
the @xmath2-dependence can be investigated through a fourier analysis of the signal in the superkamiokande and borexino experiments .
we discuss in detail the interplay among such observable variations of the signal , and show how they can help to test and constrain the vacuum oscillation solution(s ) .
the analysis includes the 374-day superkamiokande data . |
in the paradigm for the power source of an agn , the continuum luminosity and spectral temperature are related to the mass by the relation @xmath1 where @xmath2 is the average photon energy , @xmath3 cm is the radius , and @xmath4 is the bolometric luminosity .
the temperature in the part of the disk is @xmath5 where @xmath6 and @xmath7 is the in units of the ( @xmath8 being the efficiency ) , so that @xmath9 at the limit .
when the approaches the rate , the thin disk solution in the inner region is probably not valid , and it has to be replaced by a hot disk solution ( e.g. wandel & liang 1991 ) .
the spectrum of a multi - is given by integrating the local spectrum over the entire disk , @xmath10 dr $ ] , where @xmath11 is the planck function and @xmath12 is the transition radius from the region to the inner optically thin region . if the disk is radially extended ( @xmath13 ) , the spectrum is almost flat ( @xmath14 ) and cuts off beyond @xmath15 , approximately @xmath16 where @xmath17 and @xmath18 are the normalization and cutoff frequency . for a kerr black hole , malkan ( 1990 )
finds @xmath19 , where @xmath20 .
this can be written as @xmath21 where @xmath4 is the _ observed _ luminosity , and we have included a bolometric correction of 10 . if the region extends down to a radius @xmath12 , and the euv cutoff energy is @xmath22 , then @xmath23 when the regime extends close to the inner disk edge , the spectrum for the case peaks at the photon energy @xmath24 . for a kerr ( eq .
[ equ : eco ] ) @xmath25 . plotting the cutoff frequency versus the luminosity
, one can infer the mass .
alternatively , if the mass can be estimated independently , it is possible to estimate the cutoff frequency ( fig . 1 ;
wandel & petrosian 1988 ) .
it is possible to estimate the mass from the broad emission - line profile , assuming the velocity width is induced by a keplerian velocity dispersion .
when the broad line region size is estimated by reverberation mapping , this technique is particularly reliable .
wandel , peterson & malkan ( 1999 ) have used a sample of agn with reverberation data to calibrate the mass estimate obtained from photoionization models , finding the relation @xmath26 where @xmath27 is the ionization parameter , @xmath28 is the density in units of 10@xmath29 , and @xmath30 is the h@xmath31 fwhm in units of @xmath32 . using this calibrated relation , it is possible to estimate masses for large samples , even without reverberation data ( wandel , malkan & peterson , in preparation ) . combining equations [ equ : mph ] and [ equ : eco ] , we have @xmath33 which is independent of luminosity and only weakly dependent on the unknown parameters @xmath27 and @xmath34 . using eq .
[ equ : ecov ] with @xmath35=1 , we find the mass and cutoff energy distribution for a large sample of quasars , seyfert 1 galaxies and nls1s ( fig . 1 ) .
note that @xmath22 is anticorrelated with the mass , and in particular that different agn categories group in different regions of the @xmath36-@xmath22 plane : quasars have more massive and low cutoff energies , seyfert 1 galaxies have intermediate masses ( @xmath37@xmath38 ) , and nls1s have low masses ( @xmath39@xmath40 ) and high cutoff energies .
this is consistent with the large soft x - ray excesses observed for many of the nls1s .
the diagonal dotted lines in fig
. 1 give the @xmath41 ratio .
it appears that quasars tend to be in the 0.011 range , seyferts in the 0.0010.1 range , and nls1s are all near with @xmath42 0.11 .
compact dark masses , probably massive ( mbhs ) , have been detected in the cores of many normal galaxies using stellar dynamics ( kormendy & richstone 1995 ) . the mbh mass appears to correlate with the galactic bulge luminosity , with the mbh having about one percent of the mass of the spheroidal bulge ( magorrian 1998 , richstone 1998 ) .
wandel & mushotzky ( 1986 ) have found an excellent correlation between the virial mass included within the narrow line region ( of order tens to hundreds of pc from the center ) and the mass estimated from x - ray variability in a sample of seyfert 1 galaxies , while a group of seyfert 2 galaxies systematically deviated from this relation .
a -bulge relation similar to that of normal galaxies has been reported between the mbh of bright pg quasars and their host galaxies ( laor 1998 ) , but the mass of the elliptical host ( or bulge , for spiral hosts ) estimates have large uncertainties .
a significantly lower to bulge mass ratio has been found for the 17 seyfert 1 galaxies with reverberation data ( wandel 1999 ) .
while the average is lower than the magorrian relation by a factor of 20 , the record belongs to the nls1 galaxy ngc 4051 , for which wandel ( 1999 ) finds a ratio that is a factor of 200 lower than the magorrian relation .
apparently , this result suggests an intrinsic difference between the central of seyfert galaxies and those of normal galaxies .
actually the difference may be ( at least partly ) due to a selection effect .
in angular - resolution limited methods ( which are applied for detecting mbhs in normal galaxies ) , the mbh detection limit is correlated with bulge luminosity : more luminous bulges have a higher detection limit because the stellar velocity dispersion is higher ( the faber - jackson relation ) . in order to detect the dynamic effect of a mbh ,
it is necessary to observe closer to the center , while the most luminous galaxies tend to be at larger distances .
so , for a given angular resolution , the mbh detection limit is higher .
another effect could bias the pg quasars .
being the brightest nearby quasars , they may represent a subset of massive ( and well - fed ) .
smaller ones would not appear as bright and would not be included in the quasar host galaxy survey .
the blr method is not subject to this constraint , making seyfert 1 galaxies good candidates for detecting low - mass mbhs .
seyfert 1 galaxies also have more reliable bulge mass estimates ; as they are nearer and have a lower nuclear brightness , their host type and bulge magnitudes can be estimated more easily ( whittle 1992 ) . finally , we concentrate on the difference between seyfert 1 galaxies and nls1s .
ngc 4051 , the only nls1 in the sample of 17 seyfert 1 galaxies with reverberation data , shows a dramatically lower to bulge mass ratio and higher @xmath0 ratio ( wandel 1999 ) .
reverberation data on more nls1s are required before one can conclude that nls1s have intrinsically lower to bulge mass ratios and smaller black holes .
boller , th . ,
brandt , w.n . and
fink , h. 1996 , , 305 , 53 .
boroson , t.a . and green , r.f .
1992 , apjs , 80 , 109 .
laor , a. 1998 , , 505 , l83 .
magorrian , j. 1998 , aj , 115 , 2285 .
malkan , m.a .
1990 , in iau colloquium no .
129 , `` structure and emission properties of accretion disks '' , eds .
c. bertout et.al . ,
editions frontiers : paris , p. 165 .
richstone , d. , et al .
1998 , nature , 395 , a14 .
wandel , a. 1997 , , 430 , 131 .
wandel , a. 1999 , , 509 , 39 .
wandel , a. and boller , th .
1998 , , 331 , 884 .
wandel , a. and liang , e.p .
1991 , , 380 , 84 .
wandel , a. and mushotzky , r.f .
1986 , , 306 , l61 .
wandel , a. , peterson , b.m . and malkan , m.a . 1999 , , 526 , 579
wandel , a. and petrosian , v. 1988 , , 329 , 11 .
whittle , m. 1992 , , 79 , 49 . | recently , reliable mass estimates for the central black holes in agn became feasible due to emission - line reverberation techniques . using this method as a calibrator ,
it is possible to determine black hole masses for a wide range of agn , in particular nls1s .
do nls1s have smaller black holes than ordinary seyfert 1 galaxies ?
are their black holes smaller compared to the sizes of their host galaxies ?
do they have larger @xmath0 ratios ?
do nls1s have hotter accretion disks ?
i confront these questions with accretion disk theory and with the data , showing that the above may well be the case .
# 1#2#3#4 = # 1 cm # 2 cm .3 cm galaxies : active ; quasars : general ; quasars : emission lines ; accretion disks ; black holes |
[ sec : intro ] a _ reflection framework _ is a planar structure made of _ fixed - length bars _ connected by _ universal joints _ with full rotational freedom .
additionally , the bars and joints are symmetric with respect to a reflection through a fixed axis .
the allowed motions preserve the _ length _ and _ connectivity _ of the bars and _ symmetry _ with respect to some reflection .
this model is very similar to that of _ cone frameworks _ that we introduced in @xcite ; the difference is that the symmetry group @xmath0 acts on the plane by reflection instead of rotation through angle @xmath1 .
when all the allowed motions are euclidean isometries , a reflection framework is _ rigid _ and otherwise it is _ flexible_. in this paper , we give a _ combinatorial _ characterization of minimally rigid , generic reflection frameworks .
formally a reflection framework is given by a triple @xmath2 , where @xmath3 is a finite graph , @xmath4 is a @xmath0-action on @xmath3 that is free on the vertices and edges , and @xmath5 is a vector of non - negative _ edge lengths _ assigned to the edges of @xmath3 .
a _ realization _
@xmath6 is an assignment of points @xmath7 and a representation of @xmath0 by a reflection @xmath8 such that : @xmath9 the set of all realizations is defined to be the _ realization space _ @xmath10 and its quotient by the euclidean isometries @xmath11 to be the configuration space .
a realization is _ rigid _ if it is isolated in the configuration space and otherwise _
flexible_. as the combinatorial model for reflection frameworks it will be more convenient to use colored graphs . a _ colored graph _
@xmath12 is a finite , directed , the orientation of the edges do not play a role , but we give the standard definition for consistency . ]
graph @xmath13 , with an assignment @xmath14 of an element of a group @xmath15 to each edge . in this paper @xmath15 is always @xmath0 .
there is a standard dictionary ( * ? ? ?
* section 9 ) associating @xmath16 with a colored graph @xmath12 : @xmath13 is the quotient of @xmath3 by @xmath15 , and the colors encode the covering map via a natural map @xmath17 . in this setting ,
the choice of base vertex does not matter , and indeed , we may define @xmath18 and obtain the same theory .
we can now state the main result of this paper .
[ theo : reflection - laman ] a generic reflection framework is minimally rigid if and only if its associated colored graph is reflection - laman . the _ reflection - laman graphs _ appearing in the statement are defined in section [ sec : matroid ] .
genericity has its standard meaning from algebraic geometry : the set of non - generic reflection frameworks is a measure - zero algebraic set , and a small _ geometric _ perturbation of a non - generic reflection framework yields a generic one .
as in all known proofs of `` maxwell - laman - type '' theorems such as theorem [ theo : reflection - laman ] , we give a combinatorial characterization of a linearization of the problem known as _
infinitesimal rigidity_. to do this , we use a _ direction network _ method ( cf .
@xcite ) . a _ reflection direction network _
@xmath19 is a symmetric graph , along with an assignment of a _ direction _
@xmath20 to each edge .
the _ realization space _ of a direction network is the set of solutions @xmath21 to the system of equations : @xmath22 where the @xmath0-action @xmath23 on the plane is by reflection through the @xmath24-axis .
a reflection direction network is determined by assigning a direction to each edge of the colored quotient graph @xmath12 of @xmath16 ( cf .
* lemma 17.2 ) ) .
since all the direction networks in this paper are reflection direction networks , we will refer to them simply as `` direction networks '' to keep the terminology manageable .
a realization of a direction network is _ faithful _ if none of the edges of its graph have coincident endpoints and _ collapsed _ if all the endpoints are coincident .
a basic fact in the theory of finite planar frameworks @xcite is that , if a direction network has faithful realizations , the dimension of the realization space is equal to that of the space of infinitesimal motions of a generic framework with the same underlying graph . in @xcite
, we adapted this idea to the symmetric case when all the symmetries act by rotations and translations .
as discussed in ( * ? ? ?
* section 1.8 ) , this so - called `` parallel redrawing trick '' described above does _ not _ apply verbatim to reflection frameworks .
thus , we rely on the somewhat technical ( cf .
* theorem b ) , ( * ? ? ?
* theorem 2 ) ) theorem [ theo : direction - network ] , which we state after giving an important definition .
let @xmath19 be a direction network and define @xmath25 to be the direction network with @xmath26 .
these two direction networks form a _
special pair _ if : * @xmath19 has a faithful realization .
* @xmath27 has only collapsed realizations .
[ theo : direction - network ] let @xmath12 be a colored graph with @xmath28 vertices , @xmath29 edges , and lift @xmath16 .
then there are directions @xmath30 such that the direction networks @xmath19 and @xmath27 are a special pair if and only if @xmath12 is reflection - laman . briefly
, we will use theorem [ theo : direction - network ] as follows : the faithful realization of @xmath19 gives a symmetric immersion of the graph @xmath3 that can be interpreted as a framework , and the fact that @xmath27 has only collapsed realizations will imply that the only symmetric infinitesimal motions of this framework correspond to translation parallel to the reflection axis . in this paper , all graphs
@xmath31 may be multi - graphs .
typically , the number of vertices , edges , and connected components are denoted by @xmath28 , @xmath32 , and @xmath33 , respectively .
the notation for a colored graph is @xmath12 , and a symmetric graph with a free @xmath0-action is denoted by @xmath16 . if @xmath16 is the lift of @xmath12 , we denote the fiber over a vertex @xmath34 by @xmath35 , with @xmath36 , and the fiber over a directed edge @xmath37 with color @xmath38 by @xmath39
. we also use _ @xmath40-sparse graphs _ @xcite and their generalizations . for a graph @xmath13 , a _
@xmath40-basis _ is a maximal @xmath40-sparse subgraph ; a _
@xmath40-circuit _ is an edge - wise minimal subgraph that is not @xmath40-sparse ; and a _
@xmath40-component _ is a maximal subgraph that has a spanning @xmath40-graph .
points in @xmath41 are denoted by @xmath42 , indexed sets of points by @xmath43 , and direction vectors by @xmath30 and @xmath44 .
realizations of a reflection direction network @xmath19 are written as @xmath21 , as are realizations of abstract reflection frameworks .
context will always make clear the type of realization under consideration .
lt is supported by the european research council under the european union s seventh framework programme ( fp7/2007 - 2013 ) / erc grant agreement no 247029-sdmodels .
jm is supported by nsf cdi - i grant dmr 0835586 .
[ sec : matroid ] in this short section we introduce the combinatorial families of sparse colored graphs we use .
let @xmath12 be a @xmath0-colored graph .
since all the colored graphs in this paper have @xmath0 colors , from now on we make this assumption and write simply `` colored graph '' .
we recall two key definitions from @xcite . the map @xmath18 is defined on cycles by adding up the colors on the edges .
( the directions of the edges do nt matter for @xmath0 colors .
similarly , neither does the traversal order . )
as the notation suggests , @xmath45 extends to a homomorphism from @xmath46 to @xmath0 , and it is well - defined even if @xmath13 is not connected .
let @xmath12 be a colored graph with @xmath28 vertices and @xmath32 edges .
we define @xmath12 to be a _ reflection - laman graph _ if : the number of edges @xmath47 , and for all subgraphs @xmath48 , spanning @xmath49 vertices , @xmath50 edges , @xmath51 connected components with non - trivial @xmath45-image and @xmath52 connected components with trivial @xmath45-image @xmath53 this definition is equivalent to that of _ cone - laman graphs _ in ( * ? ? ?
* section 15.4 ) .
the underlying graph @xmath13 of a reflection - laman graph is a @xmath54-graph .
another family we need is that of _ ross graphs _ ( see @xcite for an explanation of the terminology ) .
these are colored graphs with @xmath28 vertices , @xmath55 edges , satisfying the sparsity counts @xmath56 using the same notations as in ( [ eq : cone - laman ] ) .
in particular , ross graphs @xmath12 have as their underlying graph , a @xmath57-graph @xmath13 , and are thus connected @xcite .
a _ _ ross - circuit__. ] is a colored graph that becomes a ross graph after removing _ any _ edge .
the underlying graph @xmath13 of a ross - circuit @xmath12 is a @xmath57-circuit , and these are also known to be connected @xcite , so , in particular , a ross - circuit has @xmath58 , and thus satisfies ( [ eq : cone - laman ] ) on the whole graph . since ( [ eq : cone - laman ] ) is always at least ( [ eq : ross ] ) , we see that every ross - circuit is reflection - laman . because reflection - laman graphs are @xmath54-graphs and subgraphs that are
@xmath57-sparse are , in addition , ross - sparse , we get the following structural result . [ prop : ross - circuit - decomp ] let @xmath12 be a reflection - laman graph . then each @xmath57-component of @xmath13 contains at most one ross - circuit , and in particular , the ross - circuits in @xmath12 are vertex disjoint . [
sec : reflection-22 ] the next family of graphs we work with is new
. a colored graph @xmath12 is defined to be a _
graph , if it has @xmath28 vertices , @xmath47 edges , and satisfies the sparsity counts @xmath59 using the same notations as in ( [ eq : cone - laman ] ) .
the relationship between ross graphs and reflection-@xmath57 graphs we will need is : [ prop : ross - adding ] let @xmath12 be a ross - graph .
then for either * an edge @xmath37 with any color where @xmath60 * or a self - loop @xmath61 at any vertex @xmath62 colored by @xmath63 the graph @xmath64 or @xmath65 is reflection-@xmath57 . adding @xmath37 with any color to a ross @xmath12 creates either a ross - circuit , for which @xmath58 or a laman - circuit with trivial @xmath45-image . both of these types of graph meet this count , and so the whole of @xmath64 does as well .
it is easy to see that every reflection - laman graph is a reflection-@xmath57 graph .
the converse is not true .
[ prop : reflection - laman - vs - reflection-22 ] a colored graph @xmath12 is a reflection - laman graph if and only if it is a reflection-@xmath57 graph and no subgraph with trivial @xmath45-image is a @xmath57-block.@xmath66 let @xmath12 be a reflection - laman graph , and let @xmath67 be the ross - circuits in @xmath12 .
define the _ reduced graph _
@xmath68 of @xmath12 to be the colored graph obtained by contracting each @xmath69 , which is not already a single vertex with a self - loop ( this is necessarily colored @xmath63 ) , into a new vertex @xmath70 , removing any self - loops created in the process , and then adding a new self - loop with color @xmath63 to each of the @xmath70 . by proposition [ prop : ross - circuit - decomp ]
the reduced graph is well - defined .
[ prop : reduced - graph ] let @xmath12 be a reflection - laman graph . then its reduced graph is a reflection-@xmath57 graph .
let @xmath12 be a reflection - laman graph with @xmath71 ross - circuits with vertex sets @xmath72 . by proposition [ prop : ross - circuit - decomp ] ,
the @xmath73 are all disjoint .
now select a ross - basis @xmath74 of @xmath12 .
the graph @xmath48 is also a @xmath57-basis of @xmath13 , with @xmath75 edges , and each of the @xmath73 spans a @xmath57-block in @xmath48 .
the @xmath40-sparse graph structure theorem ( * ? ? ?
* theorem 5 ) implies that contracting each of the @xmath73 into a new vertex @xmath70 and discarding any self - loops created , yields a @xmath57-sparse graph @xmath76 on @xmath77 vertices and @xmath78 edges .
it is then easy to check that adding a self - loop colored @xmath63 at each of the @xmath70 produces a colored graph satisfying the reflection-@xmath57 counts ( [ eq : ref22a ] ) with exactly @xmath79 edges .
since this is the reduced graph , we are done .
a _ map - graph _ is a graph with exactly one cycle per connected component .
a _ reflection-@xmath80 _
graph is defined to be a colored graph @xmath12 where @xmath13 , taken as an undirected graph , is a map - graph and the @xmath45-image of each connected component is non - trivial .
[ lemma : reflection-22-decomp ] let @xmath12 be a colored graph . then @xmath12 is a reflection-@xmath57 graph if and only if it is the union of a spanning tree and a reflection-@xmath80 graph . by ( * ? ? ?
* lemma 15.1 ) , reflection-@xmath80 graphs are equivalent to graphs satisfying @xmath81 for every subgraph @xmath48 .
thus , ( [ eq : ref22a ] ) is @xmath82 the second term in ( [ eq : ref22redux ] ) is well - known to be the rank function of the graphic matroid , and the lemma follows from the edmonds - rota construction @xcite and the matroid union theorem . in the next section
, it will be convenient to use this slight refinement of lemma [ lemma : reflection-22-decomp ] .
[ prop : reflection-22-nice - decomp ] let @xmath12 be a reflection-@xmath57 graph .
then there is a coloring @xmath83 of the edges of @xmath13 such that : * the @xmath45-image of every subgraph in @xmath84 is the same as in @xmath12 .
* there is a decomposition of @xmath84 as in lemma [ lemma : reflection-22-decomp ] in which the spanning tree has all edges colored by the identity .
it is shown in ( * ? ? ?
* lemma 2.2 ) that @xmath45 is determined by its image on a homology basis of @xmath13 .
thus , we may start with an arbitrary decomposition of @xmath12 into a spanning tree @xmath85 and a reflection-@xmath80 graph @xmath86 , as provided by lemma [ lemma : reflection-22-decomp ] , and define @xmath83 by coloring the edges of @xmath85 with the identity and the edges of @xmath86 with the @xmath45-image of their fundamental cycle in @xmath85 in @xmath12 . proposition [ prop : reflection-22-nice - decomp ] has the following re - interpretation in terms of the symmetric lift @xmath16 : [ prop : reflection - laman - decomp - lift ] let @xmath12 be a reflection-@xmath57 graph . then for a decomposition , as provided by proposition [ prop : reflection-22-nice - decomp ] , into a spanning tree @xmath85 and a reflection-@xmath80 graph @xmath86 : * every edge @xmath87 lifts to the two edges @xmath88 and @xmath89 .
( in other words , the vertex representatives in the lift all lie in a single connected component of the lift of @xmath85 . ) * each connected component of @xmath86 lifts to a connected graph .
[ sec : direction - network ] we recall , from the introduction , that for reflection direction networks , @xmath0 acts on the plane by reflection through the @xmath24-axis , and in the rest of this section @xmath90 refers to this action . the system of equations ( [ eq : dn - realization1])([eq : dn - realization2 ] ) defining the realization space of a reflection direction network @xmath19 is linear , and as such has a well - defined dimension . let @xmath12 be the colored quotient graph of @xmath16 . to be realizable at all , the directions on the edges in the fiber over @xmath91 need to be reflections of each other .
thus , we see that the realization system is canonically identified with the solutions to the system : @xmath92 from now on , we will implicitly switch between the two formalisms when it is convenient .
let @xmath12 be a colored graph with @xmath32 edges .
a statement about direction networks @xmath19 is _ generic _ if it holds on the complement of a proper algebraic subset of the possible direction assignments , which is canonically identified with @xmath93 .
some facts about generic statements that we use frequently are : * almost all direction assignments are generic .
* if a set of directions is generic , then so are all sufficiently small perturbations of it . * if two properties are generic , then their intersection is as well .
* the maximum rank of ( [ eq : colored - system ] ) is a generic property .
we first characterize the colored graphs for which generic direction networks have strongly faithful realizations .
a realization is _ strongly faithful _ if no two vertices lie on top of each other .
this is a stronger condition than simply being faithful which only requires that edges not be collapsed .
[ prop : ross - realizations ] a generic direction network @xmath19 has a unique , up to translation and scaling , strongly faithful realization if and only if its associated colored graph is a ross graph .
to prove proposition [ prop : ross - realizations ] we expand upon the method from ( * ? ? ?
* section 20.2 ) , and use the following proposition .
[ prop : reflection-22-collapse ] let @xmath12 be a reflection-@xmath57 graph .
then a generic direction network on the symmetric lift @xmath16 of @xmath12 has only collapsed realizations .
since the proof of proposition [ prop : reflection-22-collapse ] requires a detailed construction , we first show how it implies proposition [ prop : ross - realizations ] .
let @xmath12 be a ross graph , and assign directions @xmath30 to the edges of @xmath13 such that , for any extension @xmath64 of @xmath12 to a reflection-@xmath57 graph as in proposition [ prop : ross - adding ] , @xmath30 can be extended to a set of directions that is generic in the sense of proposition [ prop : reflection-22-collapse ] .
this is possible because there are a finite number of such extensions . for this choice of @xmath30 ,
the realization space of the direction network @xmath19 is @xmath94-dimensional . since solutions to ( [ eq : colored - system ] )
may be scaled or translated in the vertical direction , all solutions to @xmath19 are related by scaling and translation .
it then follows that a pair of vertices in the fibers over @xmath62 and @xmath95 are either distinct from each other in all non - zero solutions to ( [ eq : colored - system ] ) or always coincide . in the latter case , adding the edge @xmath37 with any direction does not change the dimension of the solution space , no matter what direction we assign to it .
it then follows that the solution spaces of generic direction networks on @xmath19 and @xmath96 have the same dimension , which is a contradiction by proposition [ prop : reflection-22-collapse ] .
@xmath66 it is sufficient to construct a specific set of directions with this property .
the rest of the proof gives such a construction and verifies that all the solutions are collapsed .
let @xmath12 be a reflection-@xmath57 graph .
[ [ combinatorial - decomposition ] ] combinatorial decomposition + + + + + + + + + + + + + + + + + + + + + + + + + + + we apply proposition [ prop : reflection-22-nice - decomp ] to decompose @xmath12 into a spanning tree @xmath85 with all colors the identity and a reflection-@xmath80 graph @xmath86 . for now , we further assume that @xmath86 is connected . [ [ assigning - directions ] ] assigning directions + + + + + + + + + + + + + + + + + + + + let @xmath44 be a direction vector that is not horizontal or vertical .
for each edge @xmath87 , set @xmath97 .
assign all the edges of @xmath86 the vertical direction .
denote by @xmath30 this assignment of directions .
: the @xmath24-axis is shown as a dashed line .
the directions on the edges of the lift of the tree @xmath85 force all the vertices to be on one of the two lines meeting at the @xmath24-axis , and the directions on the reflection-@xmath80 graph @xmath86 force all the vertices to be on the @xmath24-axis . ]
[ [ all - realizations - are - collapsed ] ] all realizations are collapsed + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + we now show that the only realizations of @xmath19 have all vertices on top of each other . by proposition [ prop : reflection - laman - decomp - lift ] @xmath85 lifts to two copies of itself , in @xmath3 .
it then follows from the connectivity of @xmath85 and the construction of @xmath30 that , in any realization , there is a line @xmath98 with direction @xmath44 such that every vertex of @xmath3 must lie on @xmath98 or its reflection .
since the vertical direction is preserved by reflection , the connectivity of the lift of @xmath86 , again from proposition [ prop : reflection - laman - decomp - lift ] , implies that every vertex of @xmath3 lies on a single vertical line , which must be the @xmath24-axis by reflective symmetry .
thus , in any realization of @xmath19 all the vertices lie at the intersection of @xmath98 , the reflection of @xmath98 through the @xmath24-axis and the @xmath24-axis itself .
this is a single point , as desired .
figure [ fig : ref-22-collapse ] shows a schematic of this argument . [
[ x - does - not - need - to - be - connected ] ] @xmath86 does not need to be connected + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + finally , we can remove the assumption that @xmath86 was connected by repeating the argument for each connected component of @xmath86 separately .
@xmath66 the full theorem [ theo : direction - network ] will reduce to the case of a ross - circuit . [ prop : ross - circuit - pairs ] let @xmath12 be a ross circuit with lift @xmath16 . then there is an edge @xmath99 such that , for a generic direction network @xmath100 with colored graph @xmath101 : * the solution space of @xmath100 induces a well - defined direction @xmath20 between @xmath62 and @xmath95 , yielding an assignment of directions @xmath30 to the edges of @xmath13 . * the direction networks @xmath19 and @xmath102 are a special pair . before giving the proof , we describe the idea .
we are after sets of directions that lead to faithful realizations of ross - circuits . by proposition [ prop : reflection-22-collapse ]
, these directions must be non - generic .
a natural way to obtain such a set of directions is to discard an edge @xmath37 from the colored quotient graph , apply proposition [ prop : ross - realizations ] to obtain a generic set of directions @xmath103 with a strongly faithful realization @xmath104 , and then simply set the directions on the edges in the fiber over @xmath37 to be the difference vectors between the points .
proposition [ prop : ross - realizations ] tells us that this procedure induces a well - defined direction for the edge @xmath37 , allowing us to extend @xmath103 to @xmath30 in a controlled way .
however , it does _ not _ tell us that rank of @xmath19 will rise when the directions are turned by angle @xmath105 , and this seems hard to do directly .
instead , we construct a set of directions @xmath30 so that @xmath19 is rank deficient and has faithful realizations , and @xmath27 is generic .
then we make a perturbation argument to show the existence of a special pair .
the construction we use is , essentially , the one used in the proof of proposition [ prop : reflection-22-collapse ] but turned through angle @xmath105 .
the key geometric insight is that horizontal edge directions are preserved by the reflection , so the `` gadget '' of a line and its reflection crossing on the @xmath24-axis , as in figure [ fig : ref-22-collapse ] , degenerates to just a single line .
let @xmath12 be a ross - circuit ; recall that this implies that @xmath12 is a reflection - laman graph .
[ [ combinatorial - decomposition-1 ] ] combinatorial decomposition + + + + + + + + + + + + + + + + + + + + + + + + + + + we decompose @xmath12 into a spanning tree @xmath85 and a reflection-@xmath80 graph @xmath86 as in proposition [ prop : reflection - laman - decomp - lift ] . in particular
, we again have all edges in @xmath85 colored by the identity . for now , we _ assume that @xmath86 is connected _ , and we fix @xmath99 to be an edge that is on the cycle in @xmath86 with @xmath106 ; such an edge must exist by the hypothesis that @xmath86 is reflection-@xmath80 .
let @xmath107 .
furthermore , let @xmath108 and @xmath109 be the two connected components of the lift of @xmath85 .
for a vertex @xmath110 , we denote the lift in @xmath108 by @xmath111 and the lift in @xmath109 by @xmath112 .
we similarly denote the lifts of @xmath113 and @xmath114 by @xmath115 and @xmath116 .
[ [ assigning - directions-1 ] ] assigning directions + + + + + + + + + + + + + + + + + + + + the assignment of directions is as follows : to the edges of @xmath85 , we assign a direction @xmath44 that is neither vertical nor horizontal . to the edges of @xmath86 we assign the horizontal direction .
define the resulting direction network to be @xmath19 , and the direction network induced on the lift of @xmath48 to be @xmath117 .
[ [ the - realization - space - of - tildegvarphimathbfd ] ] the realization space of @xmath19 + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + figure [ fig : ross - circuit - special - pair ] contains a schematic picture of the arguments that follow .
[ lemma : rc - proof-1 ] the realization space of @xmath19 is @xmath94-dimensional and parameterized by exactly one representative in the fiber over the vertex @xmath62 selected above . in a manner similar to the proof of proposition [ prop : reflection-22-collapse ] , the directions on the edges of @xmath85 force every vertex to lie either on a line @xmath98 in the direction @xmath44 or its reflection .
since the lift of @xmath86 is connected , we further conclude that all the vertices lie on a single horizontal line .
thus , all the points @xmath118 are at the intersection of the same horizontal line and @xmath98 or its reflection .
these determine the locations of the @xmath119 , so the realization space is parameterized by the location of @xmath120 . inspecting the argument more closely
, we find that : in any realization @xmath21 of @xmath19 , all the @xmath118 are equal and all the @xmath119 are equal .
because the colors on the edges of @xmath85 are all zero , it lifts to two copies of itself , one of which spans the vertex set @xmath121 and one which spans @xmath122 .
it follows that in a realization , we have all the @xmath118 on @xmath98 and the @xmath119 on the reflection of @xmath98 . in particular , because the color @xmath123 on the edge @xmath99 is @xmath63 , we obtain the following .
[ lemma : rc - proof-5 ] the realization space of @xmath19 contains points where the fiber over the edge @xmath99 is not collapsed . : the @xmath24-axis is shown as a dashed line .
the directions on the edges of the lift of the tree @xmath85 force all the vertices to be on one of the two lines meeting at the @xmath24-axis .
the horizontal directions on the connected reflection-@xmath80 graph @xmath86 force the point @xmath118 to be at the intersection marked by the black dot and @xmath119 to be at the intersection marked by the gray one . ]
[ [ the - realization - space - of - tildegvarphimathbfd-1 ] ] the realization space of @xmath124 + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + the conclusion of lemma [ lemma : rc - proof-1 ] implies that the realization system for @xmath19 is rank deficient by one .
next we show that removing the edge @xmath99 results in a direction network that has full rank on the colored graph @xmath74 .
[ lemma : rc - proof-2 ] the realization space of @xmath19 is canonically identified with that of @xmath124 . in the proof of lemma [ lemma : rc - proof-1 ] , that @xmath86 lifts to a connected subgraph of @xmath3
was not essential .
because a horizontal line is preserved by the reflection , realizations will take on the same structure provided that @xmath86 lifts to a subgraph with two connected components .
removing @xmath99 from @xmath86 leaves a graph @xmath125 with this property since @xmath125 is a tree .
it follows that the equation corresponding to the edge @xmath99 in ( [ eq : colored - system ] ) was dependent .
[ [ the - realization - space - of - tildegvarphimathbfdperp ] ] the realization space of @xmath27 + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + next , we consider what happens when we turn all the directions by @xmath105 .
[ lemma : rc - proof-3 ] the realization space of @xmath27 has only collapsed solutions .
this is exactly the construction used to prove proposition [ prop : reflection-22-collapse ] .
[ [ perturbing - tildegvarphimathbfd ] ] perturbing @xmath19 + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + to summarize what we have shown so far : * @xmath19 has a @xmath94-dimensional realization space parameterized by @xmath120 and identified with that of a full - rank direction network on the ross graph @xmath74 .
* there are points @xmath21 in this realization space where @xmath126 .
* @xmath19 has a @xmath63-dimensional realization space containing only collapsed solutions .
what we have not shown is that the realization space of @xmath19 has _ faithful _
realizations , since the ones we constructed all have many coincident vertices .
proposition [ prop : ross - realizations ] will imply the rest of the theorem , provided that the above properties hold for any small perturbation of @xmath30 , since some small perturbation of _ any _ assignment of directions to the edges of @xmath74 has only faithful realizations .
[ lemma : rc - proof-4 ] let @xmath127 be a perturbation of the directions @xmath103 on the edges of @xmath48
. if @xmath127 is sufficiently close to @xmath103 , then there are realizations of the direction network @xmath128 such that @xmath126 .
the realization space is parameterized by @xmath120 , and so @xmath129 varies continuously with the directions on the edges and @xmath120 .
since there are realizations of @xmath130 with @xmath131 , the lemma follows .
lemma [ lemma : rc - proof-4 ] implies that any sufficiently small perturbation of the directions assigned to the edges of @xmath48 gives a direction network that induces a well - defined direction on the edge @xmath99 which is itself a small perturbation of @xmath132 .
since the ranks of @xmath100 and @xmath27 are stable under small perturbations , this implies that we can perturb @xmath103 to a @xmath133 that is generic in the sense of proposition [ prop : ross - realizations ] , while preserving faithful realizability of @xmath134 and full rank of the realization system for @xmath135 .
the proposition is proved for when @xmath86 is connected .
[ [ x - need - not - be - connected ] ] @xmath86 need not be connected + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + the proof is then complete once we remove the additional assumption that @xmath86 was connected .
let @xmath86 have connected components @xmath136 .
for each of the @xmath137 , we can identify an edge @xmath138 with the same properties as @xmath99 above .
assign directions to the tree @xmath85 as above . for @xmath139 , we assign directions exactly as above . for each of the @xmath140 with @xmath141 , we assign the edges of @xmath142 the horizontal direction and @xmath138 a direction that is a small perturbation of horizontal . with this assignment
@xmath30 we see that for any realization of @xmath19 , each of the @xmath140 , for @xmath141 is realized as completely collapsed to a single point at the intersection of the line @xmath98 and the @xmath24-axis .
moreover , in the direction network on @xmath143 , the directions on these @xmath137 are a small perturbation of the ones used on @xmath86 in the proof of proposition [ prop : reflection-22-collapse ] . from this
is follows that , in any realization @xmath27 , is completely collapsed and hence full rank .
we now see that this new set of directions has properties ( a ) , ( b ) , and ( c ) above required for the perturbation argument . since that argument makes no reference to the decomposition , it applies verbatim to the case where @xmath86 is disconnected .
@xmath66 the easier direction to check is necessity .
[ [ the - maxwell - direction ] ] the maxwell - direction + + + + + + + + + + + + + + + + + + + + + if @xmath12 is not reflection - laman , then it contains either a laman - circuit with trivial @xmath45-image , or a violation of @xmath54-sparsity . if there is a laman - circuit with trivial @xmath45-image , the parallel redrawing theorem ( * ? ? ?
* theorem 4.1.4 ) in the form ( * ? ? ?
* theorem 3 ) implies that this subgraph has no faithful realizations for @xmath144 only if it does in @xmath145 if rank - deficient .
a violation of @xmath54-sparsity implies that the realization system ( [ eq : colored - system ] ) of @xmath27 has a dependency , since the realization space is always at least @xmath63-dimensional .
[ [ the - laman - direction ] ] the laman direction + + + + + + + + + + + + + + + + + + + now let @xmath12 be a reflection - laman graph and let @xmath74 be a ross - basis of @xmath12 . for any edge @xmath146
, adding it to @xmath48 induces a ross - circuit which contains some edge @xmath99 having the property specified in proposition [ prop : ross - circuit - pairs ] .
note that @xmath147 is again a ross - basis .
we therefore can assume ( after edge - swapping in this manner ) for all @xmath146 that @xmath37 has the property from proposition [ prop : ross - circuit - pairs ] in the ross - circuit it induces .
we assign directions @xmath103 to the edges of @xmath48 such that : * the directions on each of the intersections of the ross - circuits with @xmath48 are generic in the sense of proposition [ prop : ross - circuit - pairs ] .
* the directions on the edges of @xmath48 that remain in the reduced graph @xmath68 are perpendicular to an assignment of directions on @xmath148 that is generic in the sense of proposition [ prop : reflection-22-collapse ] .
* the directions on the edges of @xmath48 are generic in the sense of proposition [ prop : ross - realizations ] .
this is possible because the set of disallowed directions is the union of a finite number of proper algebraic subsets in the space of direction assignments .
extend to directions @xmath30 on @xmath13 by assigning directions to the remaining edges as specified by proposition [ prop : ross - circuit - pairs ] . by construction , we know that : [ lemma : laman-1 ] the direction network @xmath19 has faithful realizations .
the realization space is identified with that of @xmath100 , and @xmath103 is chosen so that proposition [ prop : ross - realizations ] applies .
[ lemma : laman-2 ] in any realization of @xmath25 , the ross - circuits are realized with all their vertices coincident and on the @xmath24-axis .
this follows from how we chose @xmath30 and proposition [ prop : ross - circuit - pairs ] . as a consequence of lemma [ lemma : laman-2 ] , and the fact that we picked @xmath30
so that @xmath143 extends to a generic assignment of directions @xmath149 on the reduced graph @xmath68 we have : the realization space of @xmath27 is identified with that of @xmath150 which , furthermore , contains only collapsed solutions .
observe that a direction network for a single self - loop ( colored @xmath63 ) with a generic direction only has solutions where vertices are collapsed and on the @xmath24-axis .
consequently , replacing a ross - circuit with a single vertex and a self - loop yields isomorphic realization spaces .
since the reduced graph is reflection-@xmath57 by proposition [ prop : reduced - graph ] and the directions assigned to its edges were chosen generically for proposition [ prop : reflection-22-collapse ] , that @xmath27 has only collapsed solutions follows .
thus , we have exhibited a special pair , completing the proof .
@xmath66 [ [ remark ] ] remark +
+ + + + + it can be seen that the realization space of a direction network as supplied by theorem [ theo : direction - network ] has at least one degree of freedom for each edge that is not in a ross basis .
thus , the statement can not be improved to , e.g. , a unique realization up to translation and scale .
[ sec : reflection - laman - proof ] let @xmath151 be a reflection framework and let @xmath12 be the quotient graph . the configuration space , which is the set of solutions to the quadratic system ( [ eq : lengths-1])([eq : lengths-2 ] )
is canonically identified with the solutions to : @xmath152 where @xmath23 acts on the plane by reflection through the @xmath24-axis .
( that `` pinning down '' @xmath23 does not affect the theory is straightforward from the definition of the configuration space : it simply removes rotation and translation in the @xmath153-direction from the set of trivial motions . )
computing the formal differential of ( [ eq : colored - lengths ] ) , we obtain the system @xmath154 where the unknowns are the _ velocity vectors _ @xmath155 .
a standard kind of result ( cf .
@xcite ) is the following .
thus , we define a realization to be _ infinitesimally rigid _ if the system ( [ eq : colored - inf ] ) has maximal rank , and _ minimally infinitesimally rigid _ if it is infinitesimally rigid but ceases to be so after removing any edge from the colored quotient graph .
[ prop : rigidity - vs - directions ] let @xmath6 be a realization of a reflection framework with @xmath23 acting by reflection through the @xmath24-axis . define the direction @xmath20 to be @xmath157 .
then the rank of ( [ eq : colored - inf ] ) is equal to that of ( [ eq : colored - system ] ) for the direction network @xmath158 . the , more difficult , `` laman direction '' of the main theorem follows immediately from theorem [ theo : direction - network ] and proposition [ prop : rigidity - vs - directions ] : given a reflection - laman graph theorem [ theo : direction - network ] produces a realization with no coincident endpoints and a certificate that ( [ eq : colored - inf ] ) has corank one .
@xmath66 the statement of proposition [ prop : rigidity - vs - directions ] is _ exactly the same _ as the analogous statement for orientation - preserving cases of this theory .
what is different is that , for reflection frameworks , the rank of @xmath158 is _ not _ , the same as that of @xmath159 . by proposition [ prop : reflection-22-collapse ] ,
the set of directions arising as the difference vectors from point sets are _ always non - generic _ on reflection - laman graphs , so we are forced to introduce the notion of a special pair as in section [ sec : direction - network ] .
matthew berardi , brent heeringa , justin malestein , and louis theran .
rigid components in fixed - latice and cone frameworks . in _ proceedings of the @xmath160 annual canadian conference on computational geometry ( cccg ) _ , 2011 .
url http://arxiv.org/abs/1105.3234 .
mike develin , jeremy l. martin , and victor reiner .
rigidity theory for matroids .
_ comment .
_ , 820 ( 1):0 197233 , 2007 .
issn 0010 - 2571 .
doi : 10.4171/cmh/89 .
url http://dx.doi.org/10.4171/cmh/89 .
audrey lee and ileana streinu .
pebble game algorithms and sparse graphs .
_ discrete math .
_ , 3080 ( 8):0 14251437 , 2008 .
issn 0012 - 365x .
doi : 10.1016/j.disc.2007.07.104 .
url http://dx.doi.org/10.1016/j.disc.2007.07.104 .
ileana streinu and louis theran .
slider - pinning rigidity : a maxwell - laman - type theorem .
_ discrete & computational geometry _ , 440 ( 4):0 812837 , 2010 .
issn 0179 - 5376 .
doi : 10.1007/s00454 - 010 - 9283-y .
url http://dx.doi.org/10.1007/s00454-010-9283-y .
walter whiteley .
some matroids from discrete applied geometry . in j.
bonin , james g. oxley , and b. servatius , editors , _ matroid theory _ ,
volume 197 of _ contemporary mathematics _ ,
pages 171311 .
american mathematical society , 1996 . | we give a combinatorial characterization of generic minimally rigid reflection frameworks .
the main new idea is to study a pair of direction networks on the same graph such that one admits faithful realizations and the other has only collapsed realizations . in terms of infinitesimal rigidity ,
realizations of the former produce a framework and the latter certifies that this framework is infinitesimally rigid .
= 1 |
autonomous motion is an important issue in active soft matter , with possible applications ranging from energy harvesting @xcite to microfluidic transport @xcite and mixing @xcite .
recently realized microswimmers carry , as active element , heat - absorbing or catalytic parts which generate temperature or concentration gradients in the surrounding liquid . for particles dispersed in bulk phases ,
these thermodynamic forces give rise to an effective slip velocity , which in turn implies self - propulsion at a speed of the order of ten microns per second pax05,qia13 . much higher velocities can be achieved for active particles trapped at a liquid interface , where the self - generated temperature or concentration gradient induces a non - uniform interface tension and a marangoni flow . upon laser - heating one side of a centimeter - size object floating on water , oka09 observed self - propulsion at several cm / s .
the mutual marangoni advection of camphor releasing gel particles results in dynamical self assembly @xcite recent theoretical studies dealt with marangoni propulsion due to a non - uniform surfactant concentration .
the resulting marangoni stress comprises a dipolar term that is proportional to the inverse distance from its source .
thus @xcite calculated the marangoni flow for an asymmetric active disk , whereas @xcite considered the motion of ellipsoidal particles in terms of a reciprocal theorem .
accounting for advective non - linearities in the surfactant concentration , @xcite found spontaneous rotation of particles floating on a liquid droplet . in the present paper we consider marangoni propulsion driven by a non - uniform temperature or bulk concentration field , the dipolar term of which decays with the square of inverse distance .
assuming small reynolds , marangoni , pclet , and capillary numbers , we rely on stokes equation , linearize the field dependent tension , and neglect the interface deformation .
for a spherical particle with an off - center heat source , we calculate both the long - range marangoni flow @xmath2 driven by the non - uniform interface tension and the short - range velocity @xmath3 imposed by no - slip condition on the particle surface .
consider a colloidal particle trapped at a liquid interface . for the sake of simplicity
we discuss a sphere that is trapped at midplane , as shown in fig . 1 . retaining the first two terms of the multipole expansion of the temperature profile we find @xmath4where
@xmath5 is the total power absorbed by the particle and @xmath6 the sum of the thermal conductivities of the lower and upper fluid phases .
the strength and orientation of the dipole term are given by the position of the absorption area with respect to the particle center , @xmath7 , as shown in fig . 1 for a point - like heat source
moreover , @xmath8 depends on the conductivity contrast of the particle and fluid phases . in the case of a chemical marangoni effect , @xmath9 and @xmath10
are the solute concentration and diffusivity in the fluid phases and @xmath5 is the chemical activity at the particle surface .
the thermal gradient induces a marangoni flow along the liquid interface .
the temperature dependence of the interface tension , @xmath11 , results in the hydrodynamic boundary condition @xmath12where @xmath13 is the stress discontinuity across the interface , @xmath14 the ( downward oriented ) normal vector , and @xmath15 the tension gradient . the stress tensor @xmath16 comprises viscous and pressure terms , @xmath17 , with the velocity components @xmath18 and where the viscosity @xmath19 takes values @xmath20 and @xmath21 in the two phases .
the fluid velocity is continuous at the interface , and its normal component must vanish , @xmath22one more boundary condition is provided by the no - slip condition at the particle surface,@xmath23where @xmath24 is the velocity field of the fluid and @xmath25 the particle velocity , which is necessarily parallel to the interface .
since there are no external body forces acting on the particle we have @xmath26where the first term gives the net force exerted by the surface tension along the contact line with element @xmath27 , and the second one , the surface integral of the stress tensor with the oriented surface element @xmath28 .
for later convenience we rewrite the above conditions for the velocity and stress compenents in spherical coordinates . with the axes defined in fig .
1 we have @xmath29 and [ 12 ] @xmath30inserting the temperature gradient in ( [ 2 ] ) , we find @xmath31the zero normal velocity at the interface involves the polar component only , @xmath32 finally , imposing no - slip at the surface of the particle moving at velocity @xmath33 along the @xmath34-axis , we have @xmath35 the cosine @xmath36 turns out to be a more convenient coordinate than the polar angle @xmath37 .
thus we use the shorthand notation @xmath38note that @xmath39 is positive everywhere , whereas @xmath40 and @xmath41 on the upper and lower halfspaces , respectively .
finally we note that the particle is not allowed to rotate about the horizontal axis . in other words , the point @xmath8 in fig . 1
is confined to the interface plane .
this condition could be imposed through appropriate surface functionalization ; it is satisfied , for example , if heating occurs through a small metal patch which itself is trapped at the fluid phase boundary .
the velocity field is solution of stokes equation @xmath42 , which results in a set of coupled differiential equations for the three velocity components @xmath43 and the pressure @xmath44 . with the vector laplace operator one has [ 1 ] @xmath45where the derivative with respect to the polar angle is replaced according to @xmath46 .
the first term of each equation is given by the scalar laplace operator @xmath47similarly , we have for the incompressibility condition @xmath48for later use we give the components of the symmetrized stress in spherical coordinates , @xmath49 in view of ( [ 2 ] ) we need the off - diagonal components @xmath50 and @xmath51 . the above relations are valid in each of the fluid phases , albeit with different viscosities @xmath20 and @xmath52 .
we write the fluid velocity as the sum of two terms , @xmath53the first of which accounts for the source field ( [ 12 ] ) with the boundary condition ( [ 13 ] ) .
this field fully describes the marangoni flow induced by a pointlike heat source . for a particle of finite size , the second term @xmath1
is required in order to assure the no - slip condition at its surface .
this is achieved by chosing @xmath1 such that it satisfies both ( [ 13 ] ) and ( [ 14 ] ) , yet does not contribute to the marangoni stress .
since the marangoni stress ( [ 12 ] ) results in a jump of the velocity derivatives , the velocity is not analytic at the interface .
the isotropic component of @xmath55 varies with the distance as @xmath56 , and the dipolar contribution as @xmath57 .
this implies that @xmath2 consists of axially symmetric and dipolar terms proportional to @xmath58 and @xmath59 , respecively .
the marangoni flow of a heat source at the origin is obtained from the stream function @xmath60 with the velocity scale @xmath61 , @xmath62both radial and polar components are proportional to the inverse distance ; @xmath63 diverges on the positive @xmath64-axis for the minus sign , and on the negative one for the plus sign .
we look for solutions of stokes equation that vary with distance as @xmath65 and that are singular at @xmath66 . with the coordinates defined in fig .
1 , it is clear that that radial and polar components are proportional to @xmath67 , and @xmath68 .
thus we have @xmath69inserting this ansatz in ( [ 1 ] ) results in four coupled equations for the functions @xmath70 ; the incompressibility condition involves @xmath71 and @xmath72 only .
by successive elimination we obtain a single fourth - order differential equation for @xmath71 , @xmath73from its solution one readily constructs the remaining components and the related pressure ; discarding regular and logarithmic terms we have @xmath74with coefficients @xmath75 and @xmath76 .
note that the velocity field diverges along the vertical axis @xmath66 .
in view of the stress jump ( [ 12 ] ) it is clear that the velocity above and below the interface can not be given in closed form .
analytic solutions in the halfspaces @xmath77 are obtained by putting @xmath78then the singular factors in ( [ 34 ] ) disappear , and the relevant solution is given by @xmath79 with @xmath80 and where the plus and minus signs are valid in the upper and lower halfspaces , respectively .
the axisymmetric part corresponds to ( [ 22 ] ) , whereas the terms proportional to @xmath76 are obtained from ( 34 ) .
the contributions with @xmath81 and @xmath82 are well - known regular flow patterns @xcite that have been added in order to meet the boundary conditions at the interface .
the velocity field is symmetric with respect to the interface plane and continuous at @xmath83 , yet its derivatives are not and lead to a stress discontinuity along the interface.finally , the corresponding pressure reads as @xmath84the pressure vanishes at @xmath83 and thus is continuous at the interface .
since the viscosity @xmath19 in general takes different values in the upper and lower phases , the normal component of the pressure gradient is discontinuous .
now we determine the parameters @xmath61 and @xmath85 from the boundary conditions . in view of (
[ 13 ] ) for the normal velocity component we put @xmath86 and find @xmath87 from ( [ 3 ] ) we calculate the off - diagonal stress components @xmath88 and @xmath51 at both sides of the interface . their difference @xmath89 is proportional to the sum of the viscosities @xmath90 , @xmath91inserting the temperature gradient in the boundary conditions ( [ 12 ] ) , one readily obtains the velocity scale @xmath61 and the remaining coefficients @xmath92 , @xmath93(for a liquid - air interface one has @xmath94 and @xmath95 . )
then the velocity field reads [ 5 ] @xmath96where the upper and lower signs occur in the upper and lower halfspaces , respectively .
the corresponding pressure @xmath97 is proportional to @xmath98 and thus vanishes at the interface .
( [ 5 ] ) and ( [ 5a ] ) constitute the solution for a pointlike interface particle carrying temperature monopole and dipole moments . the latter results in a radial flow velocity at the interface proportional to @xmath99 . from elementary symmetry considerations
it is clear that both contributions to ( [ 6 ] ) have finite components along the @xmath34-axis only .
thus the contact line element @xmath100 reduces to the component @xmath101 , and the non - uniform interface tension leads to a net force @xmath102 .
the drag force exerted on the surface element @xmath103 is given by the stress component @xmath104 ; from ( [ 3 ] ) and ( [ 5 ] ) one finds @xmath105integrating the linearized tension @xmath106 along the contact line and the stress over the particle surface , we find that the tension and drag forces cancel each other , @xmath107
now we turn to the additional velocity @xmath1 which is required in order to satisfy the stick boundary condition ( [ 14 ] ) at the particle surface , @xmath108 in addition , @xmath1 has continuous derivatives and its normal component vanishes at the phase boundary . in other words
, @xmath109 does not contribute to the marangoni stress ( [ 12 ] ) and satisfies the condition ( [ 13 ] ) .
we start from the general analytic solution of stokes equation ( [ 1 ] ) which is given by the well - known series @xcite [ 20 ] @xmath110 \\ v_{\varphi } ^{p } & = u\sum_{n , m}\left [ -\frac{a^{n}}{r^{n}}\left ( \frac{n-2}{% n(n+1)}p_{nm}+\frac{q_{nm}}{n+1}\frac{a^{2}}{r^{2}}\right ) \frac{\partial _ { \varphi } x_{nm}}{s}+\frac{a^{n+1}}{r^{n+1}}\,s_{nm}s\partial _ { c}x_{nm}% \right]\end{aligned}\]]where @xmath111 and @xmath112 .
the angle - dependent basis functions @xmath113 involve associated legendre polynomials,@xmath114for example @xmath115 , @xmath116 , and @xmath117 .
the inhomogeneous solutions with coefficients @xmath118 are related to the pressure @xmath119since the source comprises monopole and dipole moments only , the above series has finite terms with @xmath111 and @xmath120 .
the only components that contribute to the drag force are the stokeslet terms @xmath121 , with @xmath120 . here
we are interested in the force along the @xmath34-axis and thus retain the coefficient @xmath122 only .
the relevant component of the stress tensor reads @xmath123 in view of ( [ 6 ] ) and ( [ 18 ] ) , it is clear that the surface integral of @xmath124 vanishes , @xmath125 in physical terms this means that there is no additional drag force arising from @xmath124 and thus no stokeslet contribution to the flow @xmath126 . in order not to interfere with the above solution @xmath2 and the conditions ( [ 12 ] ) and ( [ 5 ] )
, the additional term @xmath109 satisfies @xmath127putting @xmath83 in both velocity and stress components , one finds that all coefficients vanish except for @xmath128with @xmath111 these coefficients are determined from the boundary condition at the particle surface ( [ 5 ] ) .
we consider the velocity contributions that do not depend on the azimuthal angle @xmath129 .
inserting the velocity field @xmath1 in ( 19 ) , we find that the even coefficients @xmath130 and @xmath131 are determined by the axisymmetric part of @xmath132 and @xmath133 , according to@xmath134expanding these equations in terms of legendre polynomials one obtains @xmath135the coefficient @xmath136 provides the leading term in inverse powers of distance , @xmath137
. thus @xmath1 decays faster than the marangoni flow @xmath138 .
now we turn to the dipolar terms @xmath139 in @xmath1 . performing the derivatives with respect to @xmath129 , separating factors of @xmath140 and @xmath141 , and dividing by the velocity scale @xmath61
, we obtain from eq .
( [ 19 ] ) the relations [ 23 ] @xmath142where the upper and lower signs are valid in the halfspaces @xmath40 and @xmath41 , respectively .
the first equation is solved by expanding in terms of the polynomials @xmath143 , @xmath144 we have used @xmath115 and the coefficients @xmath145/s}{% \int \mathrm{d}cp_{2n-1}^{1}(c)^{2}}=\frac{(-1)^{n}(2n)!(8n^{2}-4n-3)}{% 2^{2n}n!(n+1)!(2n-1)(2n-3)},\]]with @xmath111 the first terms read @xmath146 @xmath147 . the second and third equations in ( [ 23 ] ) can be conveniently separated in two parts , one that relates @xmath148 to the coefficients @xmath149 and @xmath150 , and a second one relating the terms proportional to @xmath151 to the coefficients @xmath152,@xmath153for @xmath154 equations ( [ 24 ] ) and ( [ 25 ] ) are readily solved , @xmath155the coefficients @xmath156 are determined by multiplying the second equation in ( [ 25 ] ) with @xmath39 and expanding in terms of @xmath157 ; thus we obtain @xmath158 now we turn to the case @xmath159 . noting @xmath160 and @xmath161 one readily determines the coefficient @xmath162and the particle velocity @xmath163it is worth noting that the coefficient @xmath164 corresponds to the amplitude of the source - doublet of a diagrammatic expansion @xcite .
the factor @xmath165 could be obtained equally well from the reciprocal theorem developed by @xcite .
the total velocity field consists of two contributions , @xmath2 * * * * and * * * * @xmath1 , which describe the long - range behavior and the flow in the vicinity of the trapped particle , respectively . the first one is entirely determined by the temperature field ( [ temp ] ) , whereas the second one depends on the particle size and shape .
the dipolar part of the marangoni flow ( [ 5 ] ) behaves as @xmath166 , whereas to leading order * * * * @xmath1 decays as @xmath167 and is given by the coefficients @xmath168 , which comprise the source - doublet of a diagrammatic expansion @xcite . in an alternative approach , the first contributions to * * * * @xmath1 could be constructed in terms of an image system by successive reflections at the particle surface and at the interface @xcite . the dipolar component of the marangoni flow varies with the square of the inverse distance , @xmath169 .
this power law is related to the fact heat or a molecular solute diffuse in a bulk phase , and that temperature or concentration obey the 3d diffusion equation . on the other hand ,
a surfactant diffuses along the 2d interface only , and the resulting marangoni flow behaves as @xmath170 lau12 . throughout this paper
we have assumed that the interface plane coincides with the particle s midplane , requiring the contact angle @xmath171 . as a consequence ,
the velocity field @xmath172 is perfectly symmetric with respect to the interface . in the general case @xmath173 , the particle center is above or below the interface
. then the marongoni flow @xmath2 * * * * is still symmetric , whereas the contribution arising from the boundary condition at the particle surface , * * * * @xmath1 , is no longer the same above and below the interface .
this would double the number of independent coefficients in the boundary conditions at the particle surface and modify the numerical factor in ( [ 29 ] ) .
we do not expect a change of the qualitative results ; in particular the far - field @xmath2 is not affected by the vertical particle position . according to ( [ 29 ] )
a non - uniformly heated colloidal sphere moves in the direction opposite to its self - generated temperature gradient . with @xmath61 from eq .
( [ 15 ] ) , the self - propulsion velocity is given by the tension derivative @xmath174 , the particle s heat absorption rate @xmath5 and dipole moment @xmath8 , and the fluid viscosity @xmath175 and heat conductivity @xmath176 . in terms of the excess temperature at the particle surface , @xmath177
, the velocity scale reads as @xmath178with the parameters of a water - air interface at room temperature , @xmath179 pa.s and @xmath180 nk@xmath181m@xmath181 , we find that hot particles with an excess temperature @xmath182 k induce a marangoni flow with velocity scale @xmath61 of the order of @xmath183 m / s .
the self - propulsion velocity @xmath184 then depends on the reduced temperature dipole moment @xmath151 . for a particle that is heated through laser - absorbing molecules dispersed in the particle s bulk material ,
the asymmetry @xmath151 is significantly smaller than unity . on the other hand ,
if the heating occurs through a metal patch fixed on its surface , one has @xmath185 and @xmath186 .
this estimate agrees qualitatively with the velocity of several cm / s observed by @xcite for a macroscopic floater with a hot spot at one side .
note that , at constant excess temperature , @xmath61 is independent of the particle size .
the stationary temperature profile ( [ temp ] ) accounts for diffusive heat flow from an immobile source but neglects convective transport and the particle s motion .
this approximation is valid for small marangoni number @xmath187 , where @xmath188 is the thermal diffusivity . for micron - size particles and @xmath189 m@xmath190/s for water ,
one find that the linear approximation ceases to be valid at @xmath191 cm / s . in the case of a chemical marangoni effect ,
the relevant parameter is provided by the pclet number @xmath192 of the molecular solute with diffusivity @xmath193 . since molecules diffuse more slowly than heat ( @xmath193 is about hundred times smaller than @xmath188 )
, the linear approximation is restricted to velocities @xmath194mm / s .
finally , convective acceleration is negligible for small reynolds number @xmath195 , with the kinematic viscosity @xmath196 ; this condition holds true for @xmath197 m / s .
we briefly discuss two non - linear effects , heat advection by the fluid velocity and distortion of the temperature field due to the particle motion .
the temperature profile of an immobile heat source is determined by @xmath198 . for small marangoni number
the advective term is negligible , and the excess temperature varies with the inverse distance . for @xmath199
advection enhances the heat flow and modifies the temperature profile in the vicinity of the particle ( @xmath200 ) .
an additional effect occurs at finite particle velocity @xmath184 , where heat diffusion is efficient at short distances only , thus resulting in strong retardation effects in the far - field . for a hot particle moving in negative @xmath201-direction
, the @xmath58 term of ( [ temp ] ) is distorted according to @xmath202with @xmath203 . at large distances ( @xmath204 )
, the diffusive heat transport does not catch up the particle s motion , resulting in an asymmetric temperature profile that decays more rapidly in front of the moving particle ( @xmath205 ) and thus significantly modifies the marangoni stress ( [ 2 ] ) .
we conclude with a brief discussion of collective effects arising from the interplay of self - propulsion and hydrodynamic interactions ; the latter are well approximated by the far - field contribution @xmath2 .
in addition to the self - generated velocity @xmath206 , each particle is advected in the marangoni flow of its neighbors @xmath207 , resulting in the drift - diffusion equation for the probability density @xmath208 , @xmath209with the 2d gradient @xmath210 , the rotation operator about the vertical axis @xmath211 , and the diffusion coefficients @xmath193 and @xmath212 .
the self - generated marangoni flow of spheres does not exert a viscous torque , in contrast to sufficiently asymmetric particles @xcite and in contrast to the effective slip velocity in bulk phases @xcite ; the torque exerted by the vorticity of the advection flow is small . evaluating ( [ 5 ] ) at the interface , one readily obtains the advection velocity as a gradient field @xmath213 , where the effective single - particle potential @xmath214depends on the distance vector @xmath215 and the temperature dipole @xmath216 of the neighbor @xmath217 . for @xmath218
both the self - propulsion velocity @xmath219 and the dipole term in @xmath220 are small . for chemically active particles , eq . ( [ 30 ] ) leads to the keller - segel model and its rich dynamical behavior @xcite . for an external source field , such as the excess temperature due to laser heating of absorbing particles , an outward marangoni flow ( @xmath221 ) results in a radially symmetric repulsive pair interaction , and the steady state @xmath222 is given by the effective overall potential @xmath223 .
this potential favors an ordered phase of hexatic symmetry , which was indeed observed for camphor boats floating on water @xcite .
a more complex behavior is expected for a non - zero temperature dipole @xmath8 . during its rotational diffusion time
the particle travels over a distance @xmath224 .
if this length is smaller than the nearest - neighbor distance @xmath225 of the hexatic lattice , self - propulsion merely enhances the effective diffusion coefficient @xcite . with a rotational diffusion time of @xmath226
s , micron - size particles may attain a velocity @xmath184 of several mm / s , resulting in a length @xmath227 of several centimeters .
then the self - generated motion destroys the hexatic lattice and may lead to disordered phases where the particle orientations @xmath228 play the role of quenched random fields .
the author thanks mireille bousquet - mlou , thomas bickel , and guillermo zecua for stimulating discussions and helpful remarks .
this work was supported by agence nationale de la recherche through contract anr-13-is04 - 0003 , and by cnrs / idex bordeaux through peps propulsion de micro - nageurs par effet marangoni .
qian , b. , montiel , d. , bregulla , a. , cichos , f. , yang , h. 2013 harnessing thermal fluctuations for purposeful activities : the manipulation of single micro - swimmers by adaptive photon nudging . _ chem .
* 4 * , 1420 . | we study auto - propulsion of a interface particle , which is driven by the marangoni stress arising from a self - generated asymmetric temperature or concentration field .
we calculate separately the long - range marangoni flow @xmath0 due to the stress discontinuity at the interface and the short - range velocity field * * * * @xmath1 imposed by the no - slip condition on the particle surface ; both contributions are evaluated for a spherical floater with temperature monopole and dipole moments .
we find that the self - propulsion velocity is given by the amplitude of the source doublet which belongs to short - range contribution * * * * @xmath1 .
hydrodynamic interactions , on the other hand , are determined by the long - range marangoni flow @xmath2 ; its dipolar part results in an asymmetric advection pattern of neighbor particles , which in turn may perturb the known hexatic lattice or even favor disordered states . |
the combinations of different observations astrophysical data continuously testing the theoretical models and the bounds of the parameters .
different observations of the sneia @xcite , large scale redshift surveys @xcite , the measurements of the cosmic microwave background ( cmb ) @xcite and wmap @xcite indicate that our universe is presently expanding with acceleration .
standard big bang cosmology with perfect fluid fails to accommodate the observational fact . in einstein s gravity ,
the cosmological constant @xmath11 ( which has the equation of state @xmath12 ) is a suitable candidate which derive the acceleration , but till now there is no proof of the origin of @xmath11 .
now assume that there is some unknown matter which is responsible for this accelerating scenario which has the property that the positive energy density and sufficient negative pressure , know as dark energy @xcite . the scalar field or quintessence
@xcite is one of the most favored candidate of dark energy which produce sufficient negative pressure to drive acceleration in which the potential dominates over the kinetic term . in the present cosmic concordance @xmath11cdm model the universe is formed of @xmath13
26% matter ( baryonic + dark matter ) and @xmath13 74% of a smooth vacuum energy component .
the thermal cmb component contributes only about 0.01% , however , its angular power spectrum of temperature anisotropies encode important information about the structure formation process and other cosmic observables .
+ if we assume a flat universe and further assume that the only energy densities present are those corresponding to the non - relativistic dust - like matter and dark energy , then we need to know @xmath14 of the dust - like matter and @xmath15 to a very high accuracy in order to get a handle on @xmath16 or @xmath17 of the dark energy @xcite
. this can be a fairly strong degeneracy for determining @xmath18 from observations .
tonry data set with the 230 data points @xcite alongwith the 23 points from barris et al @xcite are valid for @xmath19 .
another data set consists of all the 156 points in the `` gold '' sample of riess et al @xcite , which includes the latest points observed by hst and this covers the redshift range @xmath20 . in einstein s gravity and in the flat model of the frw universe ,
one finds @xmath21 , which are currently favoured strongly by cmbr data ( for recent wmap results , see @xcite ) . in a simple analysis for the most recent riess data set gives a best - fit value of @xmath14 to be @xmath22 .
this matches with the value @xmath23 obtained by riess et al @xcite . in comparison ,
the best - fit @xmath14 for flat models was found to be @xmath24 @xcite .
the flat concordance @xmath11cdm model remains an excellent fit to the union2 data with the best - fit constant equation of state parameter @xmath25(stat)@xmath26(stat+sys together ) for a flat universe , or @xmath27(stat)@xmath28(stat+sys together ) with curvature @xcite .
chaplygin gas is the more effective candidate of dark energy with equation of state @xmath29 @xcite with @xmath30 .
it has been generalized to the form @xmath31 @xcite and thereafter modified to the form @xmath32 @xcite .
the mcg best fits with the 3 year wmap and the sdss data with the choice of parameters @xmath33 and @xmath34 @xcite which are improved constraints than the previous ones @xmath35 @xcite .
+ in recent years , loop quantum gravity ( lqg ) is a outstanding effort to describe the quantum effect of our universe .
nowadays several dark energy models are studied in the frame work of loop quantum cosmology ( lqc ) .
quintessence and phantom dark energy models @xcite have been studied in the cosmological evolution in lqc . when the modified chaplying gas coupled to dark matter in the universe
is described in the frame work lqc by jamil et al @xcite who resolved the famous cosmic coincidence problem in modern cosmology . in another study @xcite
the authors studied the model with an interacting phantom scalar field with an exponential potential and deduced that the future singularity appearing in the standard frw cosmology can be avoided by loop quantum effects . here
we assume the frw universe in lqc model filled with the dark matter and the mcg type dark energy .
we present the hubble parameter in terms of the observable parameters @xmath14 , @xmath36 and @xmath2 with the redshift @xmath3 . from stern data set
( 12 points ) , we obtain the bounds of the arbitrary parameters by minimizing the @xmath8 test .
the best - fit values of the parameters are obtained by 66% , 90% and 99% confidence levels .
next due to joint analysis with bao and cmb observations , we also obtain the bounds and the best fit values of the parameters ( @xmath9 ) by fixing some other parameters @xmath4 and @xmath7 . from the best fit of distance modulus @xmath10 for our theoretical mcg model in lqc
, we concluded that our model is in agreement with the union2 sample data .
we consider the flat homogeneous and isotropic universe described by frw metric , so the modified einstein s field equations in lqc are given by @xmath37 and @xmath38 where @xmath39 is the hubble parameter defined as @xmath40 with @xmath41 is the scale factor .
where @xmath42 is called the critical loop quantum density , @xmath43 is the dimensionless barbero - immirzi parameter .
when the energy density of the universe becomes of the same order of the critical density @xmath44 , this modification becomes dominant and the universe begins to bounce and then oscillate forever .
thus the big bang , big rip and other future singularities at semi classical regime can be avoided in lqc .
let us note here it has been suggested that @xmath45 by the black hole thermodynamics in lqc . in lqg , this parameter is fixed by the requirement of the validity of bekenstein - hawking entropy for the schwarzschild black hole .
the physical solutions are allowed only for @xmath46 . for @xmath47
, it is called bounce .
the maximum value of the hubble factor @xmath39 is reached for @xmath48 and the maximum value of hubble factor is @xmath49 . + here
@xmath50 and @xmath51 , where @xmath52 is the density of matter ( with vanishing pressure ) and @xmath53 , @xmath54 are respectively the energy density and pressure contribution of some dark energy .
now we consider the universe is filled with modified chaplygin gas ( mcg ) model whose equation of state(eos ) is given by @xmath55 we also consider the dark matter and and the dark energy are separately conserved and the conservation equations of dark matter and dark energy ( mcg ) are given by @xmath56 and @xmath57 from first conservation equation ( 4 ) we have the solution of @xmath52 as @xmath58 from the conservation equation ( 5 ) we have the solution of the energy density as @xmath59^{\frac{1}{\alpha+1}}\ ] ] where @xmath6 is the integrating constant , @xmath60 is the cosmological redshift ( choosing @xmath61 ) and the first constant term can be interpreted as the contribution of dark energy .
so the above equation can be written as @xmath62 @xmath63^{\frac{1}{\alpha+1}}\ ] ] where @xmath64 is the present value of the dark energy density .
+ in the next section , we shall investigate some bounds of the parameters in loop quantum cosmology by observational data fitting .
the parameters are determined by @xmath15-@xmath3 ( stern ) , bao and cmb data analysis @xcite .
we shall use the @xmath8 minimization technique ( statistical data analysis ) from hubble - redshift data set to get the constraints of the parameters of mcg model in lqc .
from the solution ( 8) of mcg and defining the dimensionless density parameters @xmath65 and @xmath66 we have the expression for hubble parameter @xmath39 in terms of redshift parameter @xmath3 as follows ( @xmath67 ) @xmath68 @xmath69^{1/2}\end{aligned}\ ] ] @xmath70 @xmath71^{1/2}\ ] ] from equation ( 9 ) , we see that the value of @xmath39 depends on @xmath72 so the above equation can be written as @xmath73 where @xmath74 @xmath69^{1/2}\end{aligned}\ ] ] @xmath70 @xmath71^{1/2}\ ] ] now @xmath75 contains four unknown parameters @xmath76 and @xmath7 .
now we will fixing two parameters and by observational data set the relation between the other two parameters will obtain and find the bounds of the parameters .
@xmath77 * table 1 : * the hubble parameter @xmath15 and the standard error @xmath78 for different values of redshift @xmath3 . using observed value of hubble parameter at different redshifts ( twelve data points ) listed in observed hubble data by @xcite
we analyze the model .
the hubble parameter @xmath15 and the standard error @xmath78 for different values of redshift @xmath3 are given in table 1 .
for this purpose we first form the @xmath8 statistics as a sum of standard normal distribution as follows : @xmath79 where @xmath15 and @xmath80 are theoretical and observational values of hubble parameter at different redshifts respectively and @xmath78 is the corresponding error for the particular observation given in table 1 . here , @xmath81 is a nuisance parameter and can be safely marginalized .
we consider the present value of hubble parameter @xmath2 = 72 @xmath82 8 kms@xmath83 mpc@xmath83 and a fixed prior distribution . here
we shall determine the parameters @xmath76 and @xmath7 from minimizing the above distribution @xmath84 .
fixing the two parameters @xmath85 , the relation between the other parameters @xmath86 can be determined by the observational data .
the probability distribution function in terms of the parameters @xmath76 and @xmath7 can be written as @xmath87 where @xmath88 is the prior distribution function for @xmath2 .
we now plot the graph for different confidence levels . in early stage
the chaplygin gas follow the equation of state @xmath89 where @xmath90 .
so , as per our theoretical model the two parameters should satisfy the two inequalities @xmath90 and @xmath30 .
now our best fit analysis with stern observational data support the theoretical range of the parameters .
the 66% ( solid , blue ) , 90% ( dashed , red ) and 99% ( dashed , black ) contours are plotted in figures 1 , 2 and 4 for @xmath91 and @xmath92 .
the best fit values of @xmath5 and @xmath6 are tabulated in table 2 .
@xmath93 * table 2 : * @xmath15-@xmath3 ( stern ) : the best fit values of @xmath5 , @xmath6 and the minimum values of @xmath8 for different values of @xmath4 .
+ fig.1 fig.2 fig.3 + figs . 1 - 3
show that the variation of @xmath6 with @xmath5 for @xmath94 , @xmath95 with @xmath92 respectively for different confidence levels .
the 66% ( solid , blue ) , 90% ( dashed , red ) and 99% ( dashed , black ) contours are plotted in these figures for the @xmath15-@xmath3 ( stern ) analysis .
the method of joint analysis , the baryon acoustic oscillation ( bao ) peak parameter value has been proposed by @xcite and we shall use their approach .
sloan digital sky survey ( sdss ) survey is one of the first redshift survey by which the bao signal has been directly detected at a scale @xmath13 100 mpc .
the said analysis is actually the combination of angular diameter distance and hubble parameter at that redshift .
this analysis is independent of the measurement of @xmath2 and not containing any particular dark energy . here
we examine the parameters @xmath5 and @xmath6 for chaplygin gas model from the measurements of the bao peak for low redshift ( with range @xmath97 ) using standard @xmath8 analysis .
the error is corresponding to the standard deviation , where we consider gaussian distribution .
low - redshift distance measurements is a lightly dependent on different cosmological parameters , the equation of state of dark energy and have the ability to measure the hubble constant @xmath2 directly .
the bao peak parameter may be defined by @xmath98 here @xmath99 is the normalized hubble parameter , the redshift @xmath100 is the typical redshift of the sdss sample and the integration term is the dimensionless comoving distance to the to the redshift @xmath101 the value of the parameter @xmath102 for the flat model of the universe is given by @xmath103 using sdss data @xcite from luminous red galaxies survey .
now the @xmath8 function for the bao measurement can be written as @xmath104 now the total joint data analysis ( stern+bao ) for the @xmath8 function may be defined by @xmath105 according to our analysis the joint scheme gives the best fit values of @xmath5 and @xmath6 in table 3 .
finally we draw the contours @xmath4 vs @xmath5 for the 66% ( solid , blue ) , 90% ( dashed , red ) and 99% ( dashed , black ) confidence limits depicted in figures @xmath106 for @xmath91 and @xmath92 .
+ @xmath107 * table 3 : * @xmath15-@xmath3 ( stern ) + bao : the best fit values of @xmath5 , @xmath6 and the minimum values of @xmath8 for different values of @xmath4 .
+ fig.4 fig.5 fig.6 + the contours are drawn for 66% ( solid , blue ) , 90% ( dashed , red ) and 99% ( dashed , black ) confidence levels in figs .
4 - 6 which show the variations of @xmath6 against @xmath5 for @xmath108 with @xmath92 respectively for the @xmath15-@xmath3+bao joint analysis .
one interesting geometrical probe of dark energy can be determined by the angular scale of the first acoustic peak through angular scale of the sound horizon at the surface of last scattering which is encoded in the cmb power spectrum cosmic microwave background ( cmb ) shift parameter is defined by @xcite .
it is not sensitive with respect to perturbations but are suitable to constrain model parameter .
the cmb power spectrum first peak is the shift parameter which is given by @xmath109 + fig.7 fig.8 fig.9 + the contours are drawn for 66% ( solid , blue ) , 90% ( dashed , red ) and 99% ( dashed , black ) confidence levels in figs .
7 - 9 which show the variations of @xmath6 against @xmath5 for @xmath110 with @xmath92 respectively for the @xmath15-@xmath3+bao+cmb analysis . where @xmath111 is the value of redshift at the last scattering surface . from wmap7 data of the work of komatsu et al @xcite
the value of the parameter has obtained as @xmath112 at the redshift @xmath113 .
now the @xmath8 function for the cmb measurement can be written as @xmath114 now when we consider three cosmological tests together , the total joint data analysis ( stern+bao+cmb ) for the @xmath8 function may be defined by @xmath115 now the best fit values of @xmath5 and @xmath6 for joint analysis of bao and cmb with stern observational data support the theoretical range of the parameters given in table 4 .
the 66% ( solid , blue ) , 90% ( dashed , red ) and 99% ( dashed , black ) contours are plotted in figures 7 - 9 for @xmath91 and @xmath92 .
@xmath116 * table 4 : * @xmath15-@xmath3 ( stern ) + bao + cmb : the best fit values of @xmath5 , @xmath6 and the minimum values of @xmath8 for different values of @xmath4 . the supernova type ia experiments provided the main evidence for the existence of dark energy .
since 1995 , two teams of high-@xmath3 supernova search and the supernova cosmology project have discovered several type ia supernovas at the high redshifts @xcite .
the observations directly measure the distance modulus of a supernovae and its redshift @xmath3 @xcite . now
, take recent observational data , including sne ia which consists of 557 data points and belongs to the union2 sample @xcite .
+ from the observations , the luminosity distance @xmath117 determines the dark energy density and is defined by @xmath118 and the distance modulus ( distance between absolute and apparent luminosity of a distance object ) for supernovas is given by @xmath119 + 25\ ] ] the best fit of distance modulus as a function @xmath10 of redshift @xmath3 for our theoretical model and the supernova type ia union2 sample are drawn in figure 10 for our best fit values of @xmath7 , @xmath4 , @xmath5 and @xmath6 . from the curves , we see that the theoretical mcg model in lqc is in agreement with the union2 sample data .
+ fig.10 + in fig.10 , @xmath120 vs @xmath3 is plotted for our model ( solid line ) and the union2 sample ( dotted points ) .
modified chaplygin gas ( mcg ) is one of the candidate of unified dark matter - dark energy model .
we have considered the frw universe in loop quantum cosmology ( lqc ) model filled with the dark matter ( perfect fluid with negligible pressure ) and the modified chaplygin gas ( mcg ) type dark energy .
we present the hubble parameter in terms of the observable parameters @xmath0 , @xmath1 and @xmath2 with the redshift @xmath3 and the other parameters like @xmath4 , @xmath5 , @xmath6 and @xmath7 .
we have chosen the observed values of @xmath121 , @xmath122 and @xmath2 = 72 kms@xmath83 mpc@xmath83 . from stern data set
( 12 points ) , we have obtained the bounds of the arbitrary parameters by minimizing the @xmath8 test .
next due to joint analysis of bao and cmb observations , we have also obtained the best fit values and the bounds of the parameters ( @xmath9 ) by fixing some other parameters @xmath92 and @xmath91 . the best - fit values and bounds of the parameters
are obtained by 66% , 90% and 99% confidence levels are shown in figures 1 - 9 for stern , stern+bao and stern+bao+cmb analysis .
the distance modulus @xmath10 against redshift @xmath3 has been drawn in figure 10 for our theoretical model of the mcg in lqc for the best fit values of the parameters and the observed sne ia union2 data sample .
so our predicted theoretical mcg model in lqc permitted the observational data sets .
the observations do in fact severely constrain the nature of allowed composition of matter - energy by constraining the range of the values of the parameters for a physically viable mcg in lqc model .
we have checked that when @xmath44 is large , the best fit values of the parameters and other results of lqc model in mcg coincide with the results of the ref . [
29 ] in einstein s gravity .
when @xmath44 is small , the best fit values of the parameters and the bounds of parameters spaces in different confidence levels in lqc distinguished from einstein s gravity for mcg dark energy model .
also , in particular , if we consider the generalized chaplygin gas ( @xmath123 ) , the best fit value of critical barbero - immirzi parameter @xmath43 is 0.2486 , where we have assumed the values of other parameters @xmath91 and @xmath124 for our convenience . in summary
, the conclusion of this discussion suggests that even though the effect that quantum aspect of gravity have on the cmb are small , cosmological observation can put upper bounds on the magnitude of the correction coming from quantum gravity that may be closer to the theoretical expectation than what one would expect .
the authors are thankful to iucaa , pune , india for warm hospitality where part of the work was carried out . | we have considered the frw universe in loop quantum cosmology ( lqc ) model filled with the dark matter ( perfect fluid with negligible pressure ) and the modified chaplygin gas ( mcg ) type dark energy .
we present the hubble parameter in terms of the observable parameters @xmath0 , @xmath1 and @xmath2 with the redshift @xmath3 and the other parameters like @xmath4 , @xmath5 , @xmath6 and @xmath7 . from stern data set
( 12 points ) , we have obtained the bounds of the arbitrary parameters by minimizing the @xmath8 test .
the best - fit values of the parameters are obtained by 66% , 90% and 99% confidence levels .
next due to joint analysis with bao and cmb observations , we have also obtained the bounds of the parameters ( @xmath9 ) by fixing some other parameters @xmath7 and @xmath4 . from the best fit of distance modulus @xmath10 for our theoretical mcg model in lqc
, we concluded that our model is in agreement with the union2 sample data . |
the fundamental principle behind the creation of dissipative solitons is that their stability relies upon the simultaneous balance of conservative and dissipative ingredients in the underlying system @xcite .
these are the diffraction and self - focusing nonlinearity in the conservative part of the system , and linear and nonlinear loss and gain terms in the dissipative part .
well - known physical realizations of such systems are offered by lasing @xcite and plasmonic @xcite cavities , the respective models being based on the complex ginzburg - landau ( cgl ) equations with the cubic - quintic ( cq ) set of gain and loss terms , combined with the background linear loss @xcite .
this combination is well known to maintain stable localized modes @xcite .
the cgl equations constitute a generic class of dissipative pattern - formation models @xcite , which find many other applications , including bosonic condensates of quasi - particles in solid - state media @xcite , reaction - diffusion systems @xcite , and superconductivity @xcite .
originally , the cgl equation of the cq type was introduced @xcite as a model for the creation of stable two - dimensional ( 2d ) localized modes . following this work ,
similar models were derived or proposed as phenomenological ones in various settings .
many 1d and 2d localized states , i.e. , dissipative solitons , have been found as solutions of such equations @xcite-@xcite .
an essential ingredient of advanced laser cavities is a transverse periodic grating , which can be fabricated by means of available technologies jena .
in addition to the permanent gratings , virtual photonic lattices may be induced in photorefractive crystals as interference patterns by pairs of pump beams with the ordinary polarization , which illuminate the crystal along axes @xmath0 and @xmath1 , while the probe beam with the extraordinary polarization is launched along @xmath2 @xcite . a 2d cavity model with the grating
was introduced in ref .
it is based on the cq - cgl equation including the cellular ( lattice ) potential , which represents the grating .
in fact , the laser cavity equipped with the grating may be considered as a photonic crystal built in the active medium .
periodic potentials also occur in models of passive optical systems , which are driven by external beams and operate in the temporal domain , unlike the active systems which act in the spatial domain @xcite .
localized vortices , alias vortex solitons , are an important species of self - trapped modes in 2d settings . in uniform media ,
dissipative vortex solitons can not be stable without the presence of a diffusion term , in the framework of the cgl equation ( see , e.g. , ref .
however , this term is absent in models of waveguiding systems ( it may sometimes be present in temporal - domain optical models @xcite ) .
nevertheless , compound vortices , built as complexes of four peaks pinned to the lattice potential , may be stable in models including the grating in the absence of the diffusion @xcite . using this possibility ,
stable 2d @xcite and 3d @xcite vortical solitons have been found in the framework of cgl equations including trapping potentials . in a majority of previous works ,
the studies of various 2d localized patterns have been focused on their stabilization by means of the lattice potentials .
another relevant issue is mobility of 2d dissipative solitons in the presence of the underlying lattice ( dissipative solitons may move without friction only if the diffusion term is absent , therefore the mobility is a relevant issue for the diffusion - free models of laser cavities ) .
localized modes can be set in motion by the application of a kick to them , which , in the context of the laser - cavity models , implies launching a tilted beam into the system .
recently , the mobility of kicked 2d fundamental solitons in the cq - cgl equation with the cellular potential was studied in ref .
it has been demonstrated that the kicked soliton , hopping through the periodic structure , leaves in its wake various patterns in the form of single- or multi - peak states trapped by the periodic potential . in the case of periodic boundary conditions ( b.c . ) , which correspond to an annular system , the free soliton completes the round trip and hits the pattern that it has originally created .
depending on parameters , the free soliton may be absorbed by the pinned mode ( immediately , or after several up to five cycles of quasi - elastic collisions ) , or the result may be a regime of periodic elastic collisions , which features periodic cycles of passage of the moving soliton through the quiescent one . a natural extension of the analysis performed in ref .
@xcite is the study of the mobility of kicked soliton complexes , such as dipoles , quadrupoles , and compound vortices , and various scenarios of the dynamical pattern formation initiated by such moving complex modes , in the framework of the 2d cq - cgl equation with the lattice potential .
this is the subject of the present work .
in fact , such configurations are truly two - dimensional ones , while the dynamical regimes for kicked fundamental solitons , studied in ref .
@xcite , actually represent quasi-1d settings .
the model is formulated in section ii , which is followed by the presentation of systematic numerical results for dipoles , quadrupoles , and vortices of two types , onsite and offsite - centered ones ( alias rhombuses " and squares " ) in sections iii , iv , and v , respectively .
the paper is concluded by section vi .
an essential finding is that the interaction of a freely moving dipole with pinned patterns , originally created by the same kicked dipole , gives rise to new outcomes under the periodic b.c . in particular ,
the quiescent dipole can be absorbed ( cleared " ) by the moving one , which may have obvious applications to the design of all - optical data - processing schemes , where one may need to install or remove a blocking soliton .
also noteworthy is the transformation of an unstable vortex by a strong kick into a stable moving four - soliton cluster .
the cq - cgl equation with a periodic potential is written as @xmath3 u. \label{cgl}\ ] ] it describes the evolution of the amplitude of electromagnetic field @xmath4 along propagation direction @xmath5 , with transverse laplacian @xmath6 .
parameter @xmath7 is the linear - loss coefficient , @xmath8 is the cubic gain , @xmath9 the quintic loss , and @xmath10 the quintic self - defocusing coefficient ( it accounts for the saturation of the kerr effect if @xmath11 ) .
the 2d periodic potential with amplitude @xmath12 is taken in the usual form , @xmath13 $ ] , where the normalization of the field and coordinates is chosen so as to make the normalized period equal to @xmath14 , which is always possible .
the total power of the field is also defined as usual,@xmath15 we solved cgl equation ( [ cgl ] ) by means of the fourth - order runge - kutta algorithm in the @xmath5-direction , and five - point finite - difference scheme for the computation of the transverse laplacian @xmath16 .
periodic boundary conditions ( b.c . ) were used for the study of kicked dipoles and quadrupoles , and absorbing b.c . for kicked vortices . in the latter case , the absorbing b.c .
was implemented by adding a surrounding linear - absorption strip to the computation box .
the absorption coefficient varies quadratically with @xmath17 and @xmath18 from zero at the internal border of the strip to a value large enough to induce complete absorption of any outgoing pulse , at its external border .
this smooth variation , if the width of the strip is not too small , allows one to suppress any reflection from the absorption strip .
values of coefficients chosen for numerical simulations are @xmath19 , @xmath20 , @xmath21 , @xmath22 , and @xmath23 .
this choice corresponds to a set of parameters for which the initial static configurations for the dipoles , quadrupoles , and vortices are stable ( in - phase bound states of two dissipative solitons are possible too , but , unlike the dipoles , with the phase shift of @xmath24 between the bound solitons , they are unstable ) .
the kick is applied to then in the usual way , by adding the linear phase profile to the initial field:@xmath25where @xmath26 .
the key parameters are length @xmath27 of kick vector @xmath28 , and angle @xmath29 which it makes with the @xmath17-axis , i.e. , @xmath30 in the laser setup the kick corresponds to a small deviation of the propagation direction of the beam from the @xmath5 axis . if @xmath31 is the full wave number and @xmath32 is the deviation angle , the length of the transverse wave vector in physical units is @xmath33 , which corresponds to @xmath27 in the normalized form .
below , we investigate the influence of kick parameters @xmath27 and @xmath29 , defined as per eq .
( theta ) , on a variety of multi - soliton complexes , which are created by moving dipoles , quadrupoles , or vortices ( of both onsite- and offsite - centered types ) in the 2d cgl medium with the cellular potential .
in this section we consider the simplest soliton complex in the form of a stable vertical dipole , which consists of a pair of solitons aligned along the @xmath18-axis and mutually locked with phase difference @xmath14 , which is shown in fig .
[ dipoleinit ] . the same color code as in fig .
[ absdipoleinit ] is used in all figures showing amplitude distributions throughout the paper .
first , the dipole is set in motion by the application of the kick in the horizontal ( @xmath17 ) direction ( i.e. , transversely to the dipole s axis ) , as per eqs .
( [ kick ] ) and ( [ theta ] ) with @xmath34 . as shown in fig .
[ dipolek016908theta0 ] , the moving dipole multiplies into a set of secondary ones , similar to the outcome of the evolution of the kicked fundamental soliton @xcite .
each newly created dipole features the fixed phase shift @xmath14 between two constituent solitons , and the entire pattern , established as the result of the evolution , is robust . the particular configuration displayed in fig .
[ dipolek016908theta0 ] is a chain of five trapped dipoles , and a free one , which has wrapped up the motion and reappears from the left edge , moving to the right , due to the periodic b.c .
then , the free dipole will hit the pinned chain , and will be absorbed by it , yielding a pattern built of five quiescent dipoles . immediately after the collision , the pattern features intrinsic oscillations , which are gradually damped . the snapshot shown in fig .
[ dipolek016908theta0 ] corroborates an inference made from the analysis of numerical results : the largest number of the dipoles generated by the initially kicked one is six , including one moving dipole and five identical quiescent ones .
it is worthy to note that , as seen in fig .
[ energdipolek016908theta0 ] , in this case the total power ( [ p ] ) of the finally established set of six dipoles is close to the net power corresponding to _ seven _ quiescent ones , which is explained by the observation that the power of the stably moving dipole is , approximately , twice that of its quiescent counterpart . to study the outcome of this dynamical pattern - formation scenario in a systematic form , we monitored the number of output solitons as a function of the kick s strength , @xmath27 .
these results are summarized in fig .
nombdipole , which provides an adequate overall characterization of the interactions , including a potential possibility to use these interactions for the design of data - processing setups . , applied to the vertical dipole in the horizontal direction.,width=377 ] below the threshold value of the kick s strength , whose numerically found value is @xmath35the kicked dipole exhibits damped oscillations , remaining trapped near a local minimum of the cellular potential .
then , as seen in fig .
nombdipole , the number of dipoles initially increases steeply with @xmath27 , reaching ( as mentioned above ) a maximum of six at @xmath36 .
it is worthy to mention that this value is different from those , ranging in interval @xmath37 $ ] , in which the maximum number of secondary solitons is reached in the case when the kick is applied to a fundamental soliton @xcite .
this observation suggests that building the structures by the kicked dipole does not merely reduce to the earlier studied regime of the pattern formation by the individual solitons forming the dipole . with the further increase of @xmath27 , the number of solitons in the output decreases by increasingly broad steps . for the sake of the completeness of the description of the 2d system , we have also simulated essentially quasi-1d dynamical regimes initiated by the motion of the dipole kicked at angle of @xmath38 , i.e. , in the longitudinal direction , see eq .
( [ theta ] ) .
this setting implies the possibility to generate not only new dipoles but fundamental solitons as well .
it was found that the minimum value of the kick which is necessary to set the dipole in motion is smaller in this case than the one given by eq .
( [ thr0 ] ) : @xmath39the results obtained for this configuration are summarized in table table_dipole_theta0.5pi . above the threshold value ( [ thr_pi/2 ] ) , additional moving solitons are created : one at @xmath40 $ ] and two in a narrow interval @xmath41 $ ] .
then , for @xmath42 $ ] , a new moving dipole appears , which , as well as the original one , is oriented along the direction of the motion , and accompanied by two moving solitons . for @xmath43 $ ]
, we have one moving soliton less , and at @xmath44 $ ] the original dipole disappears in the course of the propagation , thus leaving one moving dipole and two moving solitons in the system . at @xmath45 $ ] , we observe the same pattern as for @xmath46 $ ] ( two dipoles and one moving soliton ) .
then , for @xmath47 $ ] , the dipole splits into two traveling solitons , with the upper one leaving a pinned soliton at the site which it originally occupied . at higher values of the kick s strength
, the same pattern appears , except that the solitons do not leave anything behind them , just traveling through the lattice .
.the number of dipoles and fundamental solitons in the established pattern versus the kick s strength @xmath27 directed along the dipole s axis ( @xmath48 ) . in the right column ,
a newly emerging dipole ( if any ) is counted as two solitons .
[ cols="^,^,^ " , ] + the above consideration was performed for a long system , before the collision of the freely moving dipole with the static pattern left in its wake , which should take place in the case of periodic b.c . in the application to laser - cavity settings , the periodic b.c . in the direction of @xmath49
are relevant , corresponding to the cavity with the annular shape of its cross section .
the study of dynamical pattern - formation scenarios with the periodic b.c .
is also interesting in terms of the general analysis of models based on the cgl equations @xcite .
thus , under the periodic b.c . , the freely moving dipole observed in fig .
dipolek016908theta0 will complete the round trip and will hit the trapped chain of quiescent dipoles .
results of extensive simulations of this setting are summarized in the list of three different outcomes of the collisions , which feature persistent or transient dynamics ( all the regimes were observed for @xmath34 , i.e. , the transversely kicked dipole ) : * the regime of the periodic elastic collisions , corresponding to the periodically recurring passage of the moving dipole through the quiescent one , see fig .
[ absdipolek01865theta0 ] .
this outcome takes place for @xmath50 $ ] .
note that , according to fig .
[ nombdipole ] , in this region the pattern left in the wake of the kicked dipole indeed amounts to another single quiescent dipole .
* the transient regime , which features several quasi - elastic collisions , before the moving dipole is eventually absorbed by the pinned pattern , which is a bound complex of two dipoles , see fig .
this transient regime occurs around @xmath51 , in which case fig .
[ absdipolek01865theta0 ] confirms that the moving dipole leaves a set of two additional dipoles in its wake .
* the regime of clearing the obstacle " , opposite to the previous one : it features several elastic collisions , before the pinned dipole is absorbed by the moving one , see fig .
this happens for @xmath52 $ ] and around @xmath53 ( in this region , fig .
[ nombdipole ] confirms that the moving dipole creates , originally , a single quiescent one ) . in other cases ,
the freely moving dipole is absorbed by the quiescent pattern as a result of the first collision .
it is relevant to stress that , while the first two above - mentioned regimes have been reported in ref .
@xcite for the motion of kicked fundamental solitons , the third regime ( clearing the obstacle `` ) is a new one , which was not found for the fundamental solitons .
another characteristic feature of the latter regime is that it eventually leads to the splitting of the surviving single dipole into unbound fundamental solitons , as shown in fig .
[ absdipolek01884theta0sep ] . to analyze the splitting ,
we have identified position @xmath54 of the field maximum in each soliton ( its center ) , and values of phases at these points ( mod @xmath55 ) , as functions of evolution variable @xmath5 . as a result
, it has been found that the splitting of the dipole and the loss of the phase correlation between the splinters starts in a latent form '' at @xmath56 , and becomes explicit at @xmath57 , see figs .
[ absdipolek01884theta0sepy ] and diffphasedipolek01884theta0 .
the two splinter solitons get completely separated at @xmath58 .
the splitting also leads to the appearance of the velocity difference between the solitons ( the velocity is defined as @xmath59 ) , as seen in fig .
[ dipolek01884theta0abs](b ) .
a quadrupole is composed of four soliton - like power peaks , which are mutually locked with phase difference @xmath14 between adjacent ones , see an example of the offsite - centered ( alias square - shaped " ) quadrupole in fig . quadrinit .
although this mode carries no vorticity , simulations demonstrate that it is a very robust one .
we here aim to investigate dynamical regimes initiated by the application of the horizontal kick ( kick ) to the quadrupole .
the quadrupole is set in motion by the kick whose strength exceeds the respective threshold , @xmath60cf .
( [ thr0 ] ) and ( [ thr_pi/2 ] ) .
the horizontal motion of the kicked quadrupole splits it into two vertical dipoles , and generates a set of additional vertically arranged quiescent _ soliton pairs _ , with a phase shift of @xmath61 between them .
the dependence of the total number of solitons in the eventually established pattern on the kick s strength , @xmath62 , is shown in fig .
[ evoquad ] . because these simulations were subject to the periodic b.c .
, the free dipole completes the round trip to collide with the quiescent pattern .
the number of solitons was counted just before this collision . in the case
where there is no motion in the system ( no free dipole emerges ) , the count of the number of solitons is straightforward .
applied to the stable offsite - centered quadrupole .
each dipole counts as two solitons.,width=377 ] the result is quite different from that reported in the previous section for the pattern formation by the kicked dipole , cf .
[ nombdipole ] . above the threshold value ( [ quadr ] ) ,
the number of fundamental solitons in the emergent pattern increases and remains constant in a wide interval of values of @xmath27 , _
viz_. , six solitons for @xmath63 $ ] .
then , the number of the solitons increases to its maximum , which is 14 at @xmath64 $ ] , @xmath65 and @xmath66 $ ] .
note that the increase is not monotonous .
for example , 12 solitons are generated at @xmath67 $ ] and @xmath68 $ ] .
subsequently , in the interval of @xmath69 $ ] , the soliton number varies between 8 and 16 .
the largest number of solitons , 18 , is reached at @xmath70 .
then , the soliton number drops to 6 , and this value remains constant over a relatively broad interval , @xmath71 $ ] . at still larger values of @xmath27 , no additional solitons are generated by the initially moving quadrupole , which in this case again splits into two dipoles . at @xmath70 ,
the simulations generate a set of 18 solitons ( the largest number , as said above ) . at first ,
two moving dipoles are actually produced by the splitting of the original quadrupole , see fig .
[ absquadrk02339theta0crea ] .
the faster dipole [ whose trajectory is characterized by a larger slope ( velocity ) , @xmath72 moves without creating new solitons , while the slower one creates several of them , namely , the third moving dipole and six quiescent ones , which brings the total number of solitons to 18 , as said above . the total energy increases up to about 24 times the energy of a quiescent soliton , which corresponds to the 12 such solitons , plus the 3 moving dipoles , with the energy of a moving soliton being about twice that of a quiescent one ( see fig .
[ energquadrk02339theta0 ] ) . due to the periodic b.c . , the three moving dipoles hit the previously generated quiescent chain , one after the other ( see fig . [ absquadrk02339theta0crea ] ) .
as a result , two first dipoles are captured by the chain increasing the number of the bound solitons in it , while the third moving dipole is absorbed without adding new solitons to the chain .
this complex interaction results in a chain of 8 quiescent dipoles ( equivalent to 16 solitons ) .
the so generated dipole train originally features intrinsic oscillations , which are eventually damped , see fig .
[ absquadrk02339theta0onde ] .
note that fig .
[ absquadrk02339theta0crea ] shows only the constituent fundamental solitons on line @xmath73 , in terms of fig .
absquadrk02339theta0max , their counterparts on the line of @xmath74 showing the same picture . as mentioned above and shown in fig .
[ vitquadrk03theta0 ] , at @xmath75 the initial quadrupole splits into two dipoles , which move at different velocities , without the formation of additional soliton pairs .
each dipole keeps the phase difference of @xmath14 between the constituent solitons .
it is well known that the lattice potential supports localized vortical modes of two types , the onsite- and offsite - centered ones @xcite .
first , we consider the pattern - formation dynamics for horizontally kicked rhombic vortices built of four fundamental solitons with an empty site in the center , which carry the total phase circulation of @xmath55 , corresponding to the topological charge @xmath76 , see fig .
[ absrhombusinit ] . a weak horizontal kick , with @xmath77 , excites oscillations of the constituent fundamental solitons which built the vortex ,
while vorticity @xmath78 is kept ( i.e. , phase differences between the adjacent solitons remain very close to @xmath61 ) , see fig .
[ diffphaserhombusk001 ] . a stronger kick ( for instance , @xmath79 ) destroys the vortical phase structure , and transforms the vortex into a quadrupole , as shown in fig . diffphaserhombusk002 . at @xmath80 and @xmath81 ,
[ rho_k0_10_theta_amp ] and rho_k0_15_theta_amp , respectively , the kick completely destroys the vortices , which are replaced by apparently random clusters of quiescent fundamental solitons .
note that , although the results shown in figs .
rho_k0_10_theta_amp-[rho_k0_15_theta_amp ] have been obtained with absorbing b.c .
, rather than periodic ones , this circumstance does not affect the results .
the same type of b.c . is used below . at @xmath82 , generated by the kicked rhombic vortex for @xmath80 .
the color code is the same as in fig .
1a.,width=377 ] , but for @xmath81 .
the color code is the same as in fig .
1a.,width=377 ] unlike their onsite - centered counterparts , quiescent offsite - centered vortices , such as the one shown in fig .
[ squareinit ] , are unstable in the entire parameter space of eq .
( [ cgl ] ) which we have explored , in agreement with the general trend of the offsite - centered vortices to be more fragile than their onsite - centered counterparts @xcite . as a result of the instability development
, they are transformed into stable quadrupoles .
nevertheless , results displayed below confirm that it is relevant to consider dynamical pattern formation by unstable kicked vortices as this type .
first , we consider the application of the horizontal kick ( [ kick ] ) corresponding to @xmath34 and varying strength @xmath27 .
the fundamental solitons building the vortex oscillate without setting in progressive motion below the threshold , @xmath83 , cf .
( [ thr0 ] ) , ( [ thr_pi/2 ] ) , and ( [ quadr ] ) . actually , it may happen , in this case , that a new soliton is created and starts moving in the horizontal direction , but the energy is not sufficient to stabilize it , and the new soliton decays eventually , while the initial solitons which compose the offsite - centered vortex are recovered at the original positions .
the inner phase structure of the unstable offsite - centered vortices is destroyed in the course of the oscillations , and it transforms into a quadrupole , in accordance with the above - mentioned fact that this is the outcome of its instability in the absence of the kick .
the increase of @xmath27 leads to formation of new 2d patterns . at @xmath81 , the right vertical pair ( column ) of the fundamental solitons , which are a part of the original vortical square ,
start to duplicate themselves , while moving to the right ( in the direction of the kick ) , see fig .
[ sqc_k0_15 ] .
a noteworthy effect is breaking of the symmetry between the top and bottom solitons in the column by the kick , only the bottom soliton succeeding to create a horizontal array of additional solitons ( three ones , in total ) . * * * * in this case , fig .
[ sqc_k0_15_energ ] shows that the eventual value of the total power ( [ p ] ) oscillates between values corresponding to the cumulative power of @xmath84 or @xmath85 quiescent fundamental solitons .
the resulting pattern develops a disordered form , which keeps to oscillate randomly , as fig .
[ sqc_k0_15_energ ] clearly demonstrate . .
the red horizontal lines show power levels corresponding to @xmath86 quiescent solitons.,width=377 ] at somewhat higher values of @xmath27 ( for example , @xmath87 ) , the original four - soliton set is transformed into a quiescent three - soliton complex , while an extra dipole and separate free solitons are created and travel through the lattice , see fig .
[ sqc_k0_2 ] .
finally , a still stronger kick applied to the square - shaped vortex transforms it into a square - shaped cluster of four solitons moving as a whole , see figs .
[ sqc_k0_25_xyz ] and [ sqc_k0_30_xyz ] , which display the result in the 3d form . in the former case , at @xmath88 ,
the cluster leaves behind a copy of one of the original solitons , while at @xmath89 the moving cluster is the single emerging mode .
although the clusters are dynamically stable , they do not carry the vortical phase structure .
, which is transformed into a stably moving four - soliton cluster .
the chromatic progression indicates the propagation direction .
the vertical rod represents the additional quiescent fundamental soliton , left in the wake of the moving four - soliton cluster.,width=377 ] but for @xmath90 . in this case , the unstable vortex is entirely transformed into the stable moving cluster.,width=377 ] we have also studied the application of the kick to the offsite - centered vortex in other directions , i.e. , varying angle @xmath29 in eq .
( [ theta ] ) .
first , as seen in fig
. [ sqc_k0_15_theta](a ) , in the case of @xmath91 and @xmath81 , the kick again breaks the symmetry between the top and bottom rows of the solitons , generating an array of additional solitons in the bottom horizontal row .
further , to check that the numerical code is compatible with the global symmetry of the setting , we also considered equivalent angles , @xmath92 , @xmath93 and @xmath94 .
the results , shown in fig .
[ sqc_k0_15_theta ] , evidence the possibility of controlling the direction of the emission of the soliton array by the direction of the initial kick .
further , running the computations for varying @xmath29 and moderate values of @xmath27 , we have concluded that there is a threshold angle @xmath95 , so that the emission towards any of the four equivalent directions , corresponding to directions @xmath96 , @xmath61 , @xmath14 or @xmath97 , occurs provided that the orientation of the kick belongs to a certain range around this direction , _
viz_. , @xmath98 , with @xmath99 .
if the kick s orientation falls into interstices between these ranges , namely,@xmath100 $ ] , solitons arrays are not generated . in the latter case
, the square vortex transforms into a quadrupole .
these results can be explained by noting that the intrinsic phase circulation in the vortex is directed counterclockwise ( from @xmath17 to @xmath18 ) .
then , as schematically shown ( for example ) for @xmath101 in fig .
sens_vort , the superposition of the externally applied kick ( the phase gradient ) and the intrinsic phase flow gives rise to the largest local phase gradient at the position of the bottom right soliton , in the positive horizontal direction , therefore the array is emitted accordingly .
it is instructive too to perform the simulations for the vortex with the opposite vorticity ( @xmath102 instead of @xmath103 ) . in this way
, the expected symmetry reversal has been verified ( not shown here in detail ) : the same results as above are obtained , with angle @xmath29 replaced by its counterpart , which is symmetric with respect to the closest coordinate axis .
we note too that identical results were obtained using both periodic and absorbing boundary conditions .
the objective of this work is to extend the analysis of the dynamical pattern - formation scenarios in the cq - cgl ( cubic - quintic complex ginzburg - landau ) equation with the 2d cellular potential .
the equation is the model for laser cavities with built - in gratings , represented by the spatially periodic potential .
recently , the quasi-1d pattern - formation scenarios , initiated by the moving fundamental solitons , were studied in this model . here
, we have systematically analyzed the fully 2d scenarios , produced by kicking compound modes , _
viz_. , dipoles , offsite - centered quadrupoles , and vortices of two different types , onsite- and offsite - centered ones . the motion of the kicked compound through the cellular potential leads to the generation of diverse multi - peak patterns pinned to the lattice , which the moving object leaves in its wake .
in the annular system with periodic boundary conditions , the persistently traveling dipole hits the pinned pattern from the opposite direction . in this way ,
several dynamical regimes are initiated , including the periodically recurring elastic passage of the free dipole through the quiescent one , and transient regimes , which lead , after a few quasi - elastic collisions , to absorption of one dipole by the other . in the case of vortices ,
the dependence of the outcome on the magnitude and direction of the kick was investigated too .
in particular , a noteworthy result is that a strong kick transforms the original offsite - centered vortex ( which is unstable by itself ) into a clean stably moving four - soliton cluster .
the analysis can be extended by considering two - component systems ( which would take the polarization of light into account ) , collisions between independently created moving modes , and the motion of kicked solitons in inhomogeneous lattices .
eventually , the analysis may be generalized for the three - dimensional setting , which is not relevant to optics , but may be realized , in principle , in terms of bose - einstein condensates of polariton - exciton quasiparticles @xcite .
the work of dm was supported in part by a senior chair grant from the rgion pays de loire , france .
support from the romanian ministry of education and research ( project pn - ii - id - pce-2011 - 3 - 0083 ) is also acknowledged by this author .
99 n. akhmediev and a. ankiewicz ( eds . ) , _ dissipative solitons _ , lect .
notes phys .
* 661 * , springer , berlin , 2005 ; n. akhmediev and a. ankiewicz ( eds . ) , _ dissipative solitons : from optics to biology and medicine _ ,
notes phys .
* 751 * , springer , berlin , 2008 .
s. barland , j. r. tredicce , m. brambilla , l. a. lugiato , s. balle , m. giudici , t. maggipinto , l. spinelli , g. tissoni , t. kndl , m. miller , and r. jger , nature ( london ) * 419 * , 699 ( 2002 ) ; z. bakonyi , d. michaelis , u. peschel , g. onishchukov , and f. lederer , j. opt .
b * 19 * , 487 ( 2002 ) ; e. a. ultanir , g. i. stegeman , d. michaelis , c. h. lange , and f. lederer , phys .
lett . * 90 * , 253903 ( 2003 ) ; p. mandel and m. tlidi , j. opt .
b : quantum semiclass . opt .
* 6 * , r60 ( 2004 ) ; n. n. rosanov , s. v. fedorov , and a. n. shatsev , appl .
b * 81 * , 937 ( 2005 ) ; c. o. weiss and ye .
larionova , rep .
phys . * 70 * , 255 ( 2007 ) ; n. veretenov and m. tlidi , phys .
a * 80 * , 023822 ( 2009 ) ; m. tlidi , a. g. vladimirov , d. pieroux , and d. turaev , phys . rev .
lett . * 103 * , 103904 ( 2009 ) ; p. genevet , s. barland , m. giudici , and j. r. tredicce , phys .
lett . * 104 * , 223902 ( 2010 ) ; p. grelu and n. akhmediev , nature photonics , * 6 * , 84 ( 2012 ) ; j. jimnez , y. noblet , p. v. paulau , d. gomila , and t. ackemann , j. opt .
* 15 * , 044011 ( 2013 ) ; c. fernandez - oto , m. g. clerc , d. escaff , and m. tlidi , phys . rev . lett . *
110 * , 174101 ( 2013 ) .
n. lazarides and g. p. tsironis , phys .
e * 71 * , 036614 ( 2005 ) ; y. m. liu , g. bartal , d. a. genov , and x. zhang , phys .
lett . * 99 * , 153901 ( 2007 ) ; e. feigenbaum and m. orenstein , opt . lett . *
32 * , 674 ( 2007 ) ; i. r. gabitov , a. o. korotkevich , a. i. maimistov , and j. b. mcmahon , appl .
phys . a * 89 * , 277 ( 2007 ) ; a. r. davoyan , i. v. shadrivov , and y. s. kivshar , opt
17 * , 21732 ( 2009 ) ; k. y. bliokh , y. p. bliokh , and a. ferrando , phys . rev .
a * 79 * , 041803 ( 2009 ) ; e. v. kazantseva and a. i. maimistov , _ ibid_. * 79 * , 033812 ( 2009 ) ; y .- y .
lin , r .- k .
lee , and y. s. kivshar , opt
. lett . * 34 * , 2982 ( 2009 ) ; a. marini and d. v. skryabin , _ ibid_. * 81 * , 033850 ( 2010 ) ; a. marini , d. v. skryabin , and b. a. malomed , opt .
* 19 * , 6616 ( 2011 ) .
j. anglin , phys .
* 79 * , 6 ( 1997 ) ; f. t. arecchi , j. bragard , and l. m. castellano , opt . commun . * 179 * , 149 ( 2000 ) ; j. keeling and n. g. berloff , phys . rev . lett . * 100 * , 250401 ( 2008 ) ; b. a. malomed , o. dzyapko , v. e. demidov , and s. o. demokritov , phys .
b * 81 * , 024418 ( 2010 ) .
o. thual and s. fauve , j. phys .
( paris ) * 49 * , 1829 ( 1988 ) ; s. fauve and o. thual , phys .
lett . * 64 * , 282 ( 1990 ) ; w. van saarloos and p. c. hohenberg , phys
* 64 * , 749 ( 1990 ) ; v. hakim , p. jakobsen , and y. pomeau , europhys .
lett . * 11 * , 19 ( 1990 ) ; b. a. malomed and a. a. nepomnyashchy , phys .
a * 42 * , 6009 ( 1990 ) ; p. marcq , h. chat , r. conte , physica d * 73 * , 305 ( 1994 ) ; n. akhmediev and v. v. afanasjev , phys . rev . lett . * 75 * , 2320 ( 1995 ) ; h. r. brand and r. j. deissler , phys . rev
. lett . * 63 * , 2801 ( 1989 ) ; v. v. afanasjev , n. akhmediev , and j. m. soto - crespo , phys .
e * 53 * , 1931 ( 1996 ) ; j. m. soto - crespo , n. akhmediev , and a. ankiewicz , phys .
lett . * 85 * , 2937 ( 2000 ) ; h. leblond , a. komarov , m. salhi , a. haboucha , and f. sanchez , j. opt . a :
pure appl .
opt . * 8 * , 319 ( 2006 ) ; w. h. renninger , a. chong , and f. w. wise , phys .
a * 77 * , 023814 ( 2008 ) ; j. m. soto - crespo , n. akhmediev , c. mejia - cortes , and n. devine , opt .
express * 17 * , 4236 ( 2009 ) ; d. mihalache , proc .
romanian acad .
a * 11 * , 142 ( 2010 ) ; y. j. he , b. a. malomed , d. mihalache , f. w. ye , and b. b. hu , j. opt .
b * 27 * , 1266 ( 2010 ) ; d. mihalache , rom .
phys . * 63 * , 325 ( 2011 ) ; c. mejia - cortes , j. m. soto - crespo , r. a. vicencio , and m. i. molina , phys . rev . a * 83
* , 043837 ( 2011 ) ; d. mihalache , rom . j. phys . * 57 * , 352 ( 2012 ) ; o. v. borovkova , v. e. lobanov , y. v. kartashov , and l. torner , phys .
a * 85 * , 023814 ( 2012 ) .
crasovan , b. a. malomed , and d. mihalache , phys . rev .
e * 63 * , 016605 ( 2001 ) ; phys .
a * 289 * , 59 ( 2001 ) ; d. mihalache , d. mazilu , f. lederer , y. v. kartashov , l .- c .
crasovan , l. torner , and b. a. malomed , phys .
rev . lett . * 97 * , 073904 ( 2006 ) ; d. mihalache , d. mazilu , f. lederer , h. leblond , and b. a. malomed , phys . rev .
a * 76 * , 045803 ( 2007 ) ; _ ibid_. * 75 * , 033811 ( 2007 ) ; d. mihalache and d. mazilu , rom . rep . phys . *
60 * , 749 ( 2008 ) .
v. skarka and n. b. aleksi , phys .
96 * , 013903 ( 2006 ) ; n. b. aleksi , v. skarka , d. v. timotijevi , and d. gauthier , phys .
a * 75 * , 061802 ( 2007 ) ; v. skarka , d. v. timotijevi , and n. b. aleksi , j. opt .
a : pure appl
. opt . * 10 * , 075102 ( 2008 ) ; v. skarka , n. b. aleksi , h. leblond , b. a. malomed , and d. mihalache , phys .
105 * , 213901 ( 2010 ) .
h. deng , h. haug , and y. yamamoto , rev .
phys . * 82 * , 1489 ( 2010 ) ; b. deveaud - pldran , j. opt .
b * 29 * , a138 ( 2012 ) ; n. g. berloff and j. keeling , ( ed .
a. bramati and m. modugno ) , _ universality in modelling non - equilibrium pattern formation in polariton condensates _ ,
physics of quantum fluids , springer , pp .
19 - 38 ( 2013 ) | we analyze pattern - formation scenarios in the two - dimensional ( 2d ) complex ginzburg - landau ( cgl ) equation with the cubic - quintic ( cq ) nonlinearity and a cellular potential .
the equation models laser cavities with built - in gratings , which stabilize 2d patterns .
the pattern - building process is initiated by kicking a compound mode , in the form of a dipole , quadrupole , or vortex which is composed of four local peaks .
the hopping motion of the kicked mode through the cellular structure leads to the generation of various extended patterns pinned by the structure . in the ring - shaped system ,
the persisting freely moving dipole hits the stationary pattern from the opposite side , giving rise to several dynamical regimes , including periodic elastic collisions , i.e. , persistent cycles of elastic collisions between the moving and quiescent dissipative solitons , and transient regimes featuring several collisions which end up by absorption of one soliton by the other .
another noteworthy result is transformation of a strongly kicked unstable vortex into a stably moving four - peaked cluster .
pattern formation , dissipative soliton , spatial soliton , vortex , complex ginzburg - landau equation , nonlinear dynamics |
the determination of geometric parameters of extrasolar planets has an essential role in inferring their densities and hence their compositions , masses and ages .
this information leads to refinements of the models of planetary systems and yields important constraints on planet formation @xcite . during the last several years
there is a sharp rise in the detections of transiting extra - solar planets ( tep ) mainly by the wide - field photometric variability surveys : ( i ) ground - based observations as superwasp @xcite , hatnet @xcite , ogle - iii @xcite , tres @xcite , etc . ;
( ii ) space missions as _ corot _ @xcite and _ kepler _ @xcite .
the _ kepler _ mission produced a real bump of the number of exoplanet candidates , dozens of them yet confirmed ( * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* etc . ) .
the space - based missions , especially _ kepler _ , have the photometric ability to detect even transiting terrestrial - size planets @xcite . the recent increasing number of the planet - candidate discoveries and the increasing precision of the observations allow to investigate fine effects such as : ( a ) : : rotational and orbital synchronization and alignment @xcite ; ( b ) : : zonal flows and violent atmospheric dynamics due to large temperature contrast between day - sides and night - sides of the planets @xcite ; ( c ) : : departures from sphericity of the planets and spin precession @xcite ; ( d ) : : rings and satellites @xcite ; ( e ) : : stellar spots @xcite ; ( f ) : : dependence of the derived planetary radius from the limb - darkening coefficients @xcite ; ( g ) : : irradiation by the parent star @xcite , planet transmission spectra @xcite , atmospheric lensing due to atmospheric refraction @xcite , rayleigh scattering , cloud scattering , refraction and molecular absorption of starlight in the planet atmosphere @xcite , etc .
the study of such fine effects requires very precise methods for determination of parameters of the planetary systems from the observational data .
recently , we established that the synthetic light curves generated by the widely - used codes for planet transits deviated from the expected smooth shape .
this motivated us to search for a new approach for the direct problem solution of the planet transits .
we managed to realize this idea successfully for the case of orbital inclination @xmath0 and linear limb - darkening law @xcite .
this paper presents continuation of the new approach for arbitrary orbital inclinations and arbitrary limb - darkening laws .
the global parameters of the configurations of teps are different from those of eclipsing binary systems .
the main geometric difference is that the radii of the two components of teps are very different , while those of ebs are comparable . as a result ,
almost all models based on numerical integration over the stellar surfaces of the components give numerical errors , especially around the transit center .
the overcoming of this problem required specific approaches and models for the study of teps .
the first solution of the direct problem of the planet transits is that of @xcite .
they derived analytical formulae describing the light decreasing due to covering of stellar disk by a dark ( opaque ) planet in cases of quadratic and nonlinear limb - darkening laws .
the formulae of ( * ? ? ?
* further m&a solution ) contain several types of special functions ( beta function , appel s hypergeometric function and gauss hypergeometric function , complete elliptic integral of the third kind ) . to generate synthetic transits
@xcite created idl and fortran codes occultsmall ( for small planet ) , occultquad ( for quadratic limb - darkening law ) and occultnl ( for nonlinear limb - darkening laws ) that are based on numerical calculations of the special function values .
these codes are widely used by many investigators for analysis of observed transits and improved later by different authors . in the meantime @xcite
obtained analytical solution for the particular case of total transit and uniform stellar disk ( i.e. neglecting the limb - darkening effect ) , @xcite made calculation of the secondary planet eclipses while @xcite studied the problem of eccentric orbits .
the second direct problem solution for the planet transits was made by @xcite .
he derived analytical formulae for the computation the light curves of planet transits for arbitrary limb - darkening laws .
this approach is similar to that of the kopal s @xmath1 functions and the derived formulae contain different special functions ( elliptic integrals of the first , second and third order ) .
most of the codes for inverse problem solutions of planet transits are based on the solution .
for instance , the recent packages tap and autokep @xcite .
@xcite estimated the fit quality of the inverse problem corresponding to the solution .
it should be noted that the inverse problem solution of the planet transits is not a trivial task .
the known codes for stellar eclipses are not applicable for the analysis of most planet transits due to the non - effective convergence of the differential corrections in cases of observational precisions poorer than 1/10 the depth of planet transit .
ebop @xcite is the only model for ebs , which heavily modified version jktebop @xcite can be applied successfully to teps .
recently , solutions of the whole inverse problem based on simultaneous modeling of photometric and spectral data of exoplanets performed using the markov - chain monte carlo ( mcmc ) code were proposed ( * ? ? ?
* ; * ? ? ?
* etc . ) .
but their subroutines for fitting of the transits also use the solution for quadratic limb - darkening law .
for instance , the subroutine exofast_occultquad of the package exofast @xcite is a new improved version of occultquad .
an opportunity to generate synthetic transit light curve as well as to search for fit to own observational data is provided from the website exoplanet transit database ( etd ) .
it is based on the occultsmall routine of the solution that uses the simplification of the planet trajectory as a straight line over the stellar disk @xcite .
let s consider configuration from a spherical planet with radius @xmath2 orbiting a spherical star with radius @xmath3 on circle orbit with radius @xmath4 , period @xmath5 and initial epoch @xmath6 .
let s the line - of - sight is inclined at an angle @xmath7 to the orbital plane of the planet .
usually , it is assumed that the limb - darkening of the main - sequence stars may be represented by the linear function @xmath8\ ] ] where @xmath9 is the light intensity at the center of the stellar disk depending on the stellar temperature , @xmath10 and @xmath11 is the angle between the normal to the current point of the stellar surface and the line of sight .
@xcite found that the more accurate limb - darkening functions are the quadratic law @xcite @xmath12\ ] ] and `` nonlinear '' law @xmath13\ ] ] which is a taylor series in @xmath14 to fourth order in 1/2 .
square - root law is proposed by @xcite @xmath15\ ] ] while @xcite proposed logarithmic law for early stars @xmath16.\ ] ] our solution can be applied for arbitrary limb - darkening law @xmath17 where @xmath18 is an arbitrary function of @xmath14 .
the possibility to use an arbitrary limb - darkening law is one of the main advantages of our approach .
the luminosity of the planetary system at phase @xmath19 out - of - transit is @xmath20 where @xmath21 is the stellar luminosity and @xmath22 is the planet luminosity .
the phase @xmath19 is calculated by the period @xmath5 and the initial epoch @xmath23 .
the planet luminosity @xmath22 is variable out of the eclipse ( because more of the day - side of the planet is visible after the occultation while more of the night - side of the planet is visible before the transit ) . due to the relatively small size and low temperature of the planet
, it can be assumed that its disk is uniform and its luminosity does not change during the transit , i.e. @xmath24 where @xmath25 depends on the planet temperature .
the luminosity during the transit is @xmath26 where @xmath27 is the light decrease due to the covering of the star by the planet .
it might be expressed in the form @xmath28 where the integration is on the stellar area @xmath29 covered by the planet .
hence , the solution of the direct problem for the planet transit is reduced to a calculation of a surface integral .
it is not a trivial task due to the nonuniform - illuminated stellar disk .
if we assume the out - of - transit flux to be @xmath30=1 , then its decreasing during the transit is described by the expression @xmath31 for the next considerations we use coordinate system whose origin coincides with the stellar center .
the axis @xmath32 is along the line - of - sight and the @xmath33 plane coincides with the visible plane .
we choose the axis @xmath34 to be along the projection of the normal to the orbit on the visible plane .
taking into account that the stellar isolines of equal light intensity are concentric circles with radius @xmath35 we may calculate the light decrease as a sum ( integral ) of the contributions of differential uniformly - illuminated arcs with central angles 2@xmath36 and area @xmath37 ( fig . [ fig01 ] ) . in this way
we transform the surface integral ( [ eq11 ] ) to a linear one @xmath38 where @xmath39 the integrand of our main equation ( [ eq13 ] ) is different from that of the main equation ( 2 ) of the solution . as a result , the methods of numerical calculation of these integrals are different .
the integration limits @xmath40 and @xmath41 are the extremal radii of the stellar isolines that are covered by the planet at orbital phase @xmath19 .
these limits depend on the configuration parameters .
it is appropriate to assume the separation @xmath4 as a size unit and to work with dimensionless quantities : relative radius of the planet @xmath42 ; relative radius of the star @xmath43 ; relative radius of the stellar isoline @xmath44 .
then the equation for the light decrease during the transit can be rewritten in the following form : @xmath45}\ ] ] where @xmath46 @xmath47 ^ 4\ ] ] assuming the center of the transit to be at phase 0.0 we consider only the phase interval ( 0 , 0.5 ) because in the case of spherical star and planet the transit light curve in the range ( -0.5 , 0 ) is symmetric to that in the range ( 0 , 0.5 ) . for a circular orbit the coordinates ( in units @xmath4 ) of the planet center at phase @xmath19
are @xmath48 the coordinates @xmath49 ( in units @xmath4 ) of the intersection points of the stellar brightness isoline with radius @xmath50 and the planet limb are respectively ( fig .
[ fig01 ] ) : @xmath51 where we have introduced the designations @xmath52 further we will derive the expressions for the angle @xmath53 and the integration limits in equation ( [ eq16 ] ) for different combinations of geometric parameters and at different phases . the transit is partial if the orbital inclination is into the range @xmath54 where @xmath55 the phases of outer contacts star - planet are @xmath56 ( beginning of the transit ) and @xmath57 ( end of the transit ) where @xmath58 the partial transit occurs into the phase range [ 0 , @xmath59 .
for the sake of brevity we will not write further the dependence of @xmath60 on @xmath19 . ( a.1 ) : : if @xmath61 ( fig .
[ fig01 ] ) the light decreasing is calculated by ( [ eq16 ] ) where limits and expression for @xmath53 are given in table [ tab01 ] ( case a.1 ) .
( a.2 ) : : if @xmath62 then the integral @xmath63 in ( [ eq16 ] ) can be presented as a sum of two integrals @xmath64 ( fig .
[ fig02 ] ) whose limits and expressions for 2@xmath65 are given in table [ tab01 ] ( case a.2 ) .
( a.3 ) : : if @xmath66 the integral @xmath63 ( fig .
[ fig03 ] ) is a sum of three integrals @xmath67 whose limits and expressions for @xmath53 are given in table [ tab01 ] ( case a.3 ) .
[ cols="^,^,^,^,^",options="header " , ] if the orbital inclination is into the range @xmath68 where @xmath69 then the transit develops from partial to total . in this case
the planet does not cover the star center at any phase ( fig .
[ fig04 ] ) .
the phases of the inner contact star - planet are @xmath70 ( during planet entering ) and @xmath71 ( during planet exit ) where @xmath72 into the phase range [ @xmath73 the transit is partial and geometry is similar to the case a. into the phase range [ 0 , @xmath74 the transit is total .
( b.1 ) : : if @xmath75 then the integral @xmath63 is calculated by ( [ eq16 ] ) where limits and expression for @xmath53 are given in table [ tab01 ] ( case b.1 ) .
( b.2 ) : : if @xmath76 the integral @xmath63 is a sum of three integrals @xmath67 whose attributes are given in table [ tab01 ] ( case b.2 ) .
, case b.2 ] if the orbital inclination is into the range @xmath77 the transit develops from partial to total and finally the planet covers the stellar center . the phases at which the planet limb touches the stellar center are @xmath78 ( before the transit center ) and @xmath79 ( after the transit center ) where @xmath80 into the phase range [ @xmath73 the transit is partial and the geometry is similar to the case a. if @xmath81 then the integral @xmath63 is calculated by ( [ eq16 ] ) where limits and expression for @xmath53 are given in table [ tab01 ] ( case a.1 ) .
if @xmath76 and @xmath82 then the integral @xmath63 is presented as a sum of two integrals @xmath64 whose attributes are the same as those of the case a.2 ( table [ tab01 ] ) .
if @xmath76 and @xmath83 then the integral @xmath63 is presented as a sum of three integrals @xmath67 whose attributes are the same as those of the case a.3 ( table [ tab01 ] ) . into the phase range
[ @xmath84 the transit is total , outside the stellar center .
the geometry is similar to case b. if @xmath75 then the integral @xmath63 is calculated by ( [ eq16 ] ) where limits and expression for @xmath53 are given in table [ tab01 ] ( case b.1 ) .
if @xmath76 then the integral @xmath63 is presented as a sum of three integrals @xmath67 whose attributes are the same as those of the case b.2 ( table [ tab01 ] ) . into the phase range [ 0 ,
@xmath85 the planet covers the stellar center and there are three subcases depending on the phase @xmath86 that corresponds to the moment when @xmath87 .
( c.1 ) : : if @xmath88 then into the phase range [ @xmath89 ( for which @xmath90 ) the integral @xmath63 is presented as a sum of six integrals @xmath91 ( fig . [ fig06 ] ) whose attributes are given in table [ tab01 ] ( case c.1 ) .
( c.2 ) : : if @xmath88 then into the phase range [ 0 , @xmath92 ( for which @xmath93 ) the integral @xmath63 is presented as a sum of six integrals @xmath91 ( fig . [ fig07 ] ) whose attributes are given in table [ tab01 ] ( case c.2 ) . ( c.3 ) : : if @xmath94 then into the whole phase range [ 0 , @xmath85 is fulfilled @xmath90 and the integral @xmath63 is presented as a sum of six integrals @xmath91 ( fig .
[ fig06 ] ) whose attributes are the same as those of case c.1 .
note : the condition @xmath94 is satisfied for @xmath95 where @xmath96 ] ] finally , it should be noted that for the arbitrary limb - darkening law ( [ eq06 ] ) the stellar luminosity l@xmath97 is given by the integral @xmath98 it has analytical solution for all known limb - darkening functions .
particularly , for the wide - used limb - darkening laws ( equations from ( [ eq01 ] ) to ( [ eq05 ] ) ) the stellar luminosity is calculated by the formulae @xmath99
the integral in equation ( [ eq16 ] ) can not be solved analytically .
that is why we had to carry out a numerical solution of this integral .
for this reason we wrote the code tac - maker ( * * t**ransit * * a**nalytical * * c**urve ) whose input parameters are : * radius of the orbit @xmath4 ; * period @xmath5 and initial epoch @xmath100 ; * radius of the star @xmath3 ; * radius of the planet @xmath2 ; * orbital inclination @xmath7 ; * temperature of the star @xmath101 ; * temperature of the planet @xmath102 ; * coefficients of the limb - darkening @xmath103 ; * step in phase @xmath104 ; * parameter of precision @xmath105 of the numerical calculations of the integrals .
the code tac - maker provides the possibility to choose the limb - darkening law from a list of the known wide - spread functions ( 1 5 ) , or to write arbitrary function @xmath106 ) .
this is a significant advantage of the proposed approach as even now the accuracy of the most used quadratic limb - darkening law is worse than the achieved from _ kepler _ for large planets with @xmath107 @xcite .
moreover , the code tac - maker allows to obtain the stellar limb - darkening coefficients from the transit solution and to compare them with the theoretical values of @xcite calculated for different temperatures , surface gravities , metal abundances and micro - turbulence velocities .
the code is written in python 2.7 language with a graphical user interface . at the very
beginning the code checks which condition ( case a , case b , or case c ) is satisfied for the given combination of configuration parameters ( stellar and planet radii and orbital inclination ) . after that the code calculates the characteristic phases @xmath108 for the respective case .
further the code makes numerical calculation of the integral ( [ eq16 ] ) for the current phase and chosen function @xmath109 using the scipy package .
finally , the code repeats the procedure for each phase of the corresponding phase ranges .
the output results flow as data file ( phase @xmath19 , flux @xmath110 ) .
the code allows to search for solutions of observed transits by the method of trials and errors varying the input parameters .
the data file might be in format magnitude or flux and @xmath111 or phase .
an estimate of the fit quality is the calculated value of @xmath112 .
moreover , the two plots of the current solution showing the observational data with the synthetic transit as well as the corresponding phase distribution of the residuals allow fast finding of a good fit . to validate our approach we used comparison with the wide - spread solution , particularly we compared the synthetic light curves generated by the code tac - maker and those produced by the code occultnl for the same configuration parameters and the same limb - darkening law .
for this purpose we applied the freely available version of the last code without any changes , particularly with its default precision , while our code worked with the default precision of the scipy package for the numerical calculations of the integrals .
we established that for linear limb - darkening law the two synthetic transits coincide ( fig .
[ fig08 ] , top ) . however , the detailed review reveals the meandering course of the solution around the smooth course of the tac - maker transit curve ( fig .
[ fig08 ] , second panel ) .
the phase derivatives of the fluxes ( fig .
[ fig08 ] , third panel ) exhibit more clearly which one of the two solutions is precise : the derivative of the tac - maker solution has almost linear course while that of the solution reveals plantigrade shape .
this result allows us to assume that our solution of the direct problem for the planet transit in the case of linear limb - darkening law is more accurate than that of . for more detailed comparison of the two approaches we analyzed the differences ( residuals ) between the flux values of the tac - maker solution and the solution for the same configuration parameters and phases .
we established that they depend on different parameters ( limb - darkening coefficients , inclination , phase , planet radius , planet temperature ) .
, @xmath113 , @xmath114 , @xmath115 , @xmath116 k , @xmath117 .
_ top panel _
: the coincidence of the two solutions for the linear limb - darkening law and moderate precision of the method ; _ second panel _ : the difference of the two solutions for small part of the transit in big scale ; _ third panel _ : the first derivatives of the two solutions for small part of the transit in big scale ; _ bottom panel _ : the residuals for the chosen part of the transit in big scale . ] for linear limb - darkening law the residuals vary with the phase by oscillating way around level 0 ( fig . [ fig09 ] )
which analysis led us to the following conclusions .
( a ) : : the frequencies and amplitudes of the oscillating residuals depend on the limb - darkening coefficients ( fig .
[ fig09 ] , top ) .
the residuals are zero for @xmath118 that means that the solution is precise for uniform star .
+ and different linear limb - darkening coefficients ; _ bottom _ for different orbital inclinations and @xmath119 ( all residuals oscillate around level 0 but are shifted vertically for a good visibility ) . ] ( b ) : : the frequencies of the residual oscillations decrease to the central part of the transit for all orbital inclinations , especially for small inclinations ( fig .
[ fig09 ] , bottom ) .
( c ) : : the amplitudes of the residuals are bigger for partial transits than for total ones ( fig .
[ fig09 ] , bottom ) .
( d ) : : the residual values increase with the planet radius ( as it should be expected ) .
we present comparison with those nonlinear limb - darkening laws which are available in the code occultnl .
( 1 ) : : the comparison of the synthetic light curves generated for the same configuration parameters and quadratic limb - darkening law revealed that those produced by the code occultnl exhibited ( fig .
[ fig10 ] ) : ( i ) small - amplitude oscillations similar to those for the linear limb - darkening law ; ( ii ) the fluxes into the transit are systematically smaller than ours ( the underestimation increases with the coefficient @xmath120 of the quadratic term ) .
( 2 ) : : the comparison of the synthetic light curves generated for the same configuration parameters and squared root limb - darkening law revealed that those produced by the code occultnl suffered from the same disadvantages as those for quadratic limb - darkening law ( fig .
[ fig10 ] ) . ; _ bottom _ panel : residuals for the code tac - maker corresponding to different values of @xmath105 . ] the detailed comparison ( validation ) of our approach with the wide - spread method demonstrated that the difference between them did not exceed @xmath121 for the considered three types limb - darkening laws .
this leads to the conclusion that for variety of model parameters the new approach gives reasonable and expected synthetic transits which practically coincide with those of the solution .
there is a possibility to increase the precision of the synthetic transits generated by the code occultnl ( e. agol , private communication ) .
its convergence criterion is the maximum change to be less than the product of two multipliers , the factor @xmath122 ( @xmath123 is an integer ) and the transit depth .
we established that the decreasing of @xmath124 from its default value @xmath125 leads to the following effects ( fig .
[ fig10 ] , top and middle panels ) : ( i ) decreasing of the amplitudes of the oscillating residuals ; ( ii ) increasing of their frequencies ; ( iii ) decreasing of the underestimation of the fluxes for the whole transit . on the other hand the default value of @xmath105 is @xmath126 ( as it is defined for the package scipy ) .
its increasing leads to some small imperfections of the numerical calculations made by our code ( fig . [ fig10 ] , bottom ) . the two considered approaches of the direct problem solution lead to different definitions of the precision parameters of the corresponding codes .
therefore , it is not reasonable to compare the precision of the generated synthetic transits for equal values of the parameters @xmath105 and @xmath124 .
though , we established that the two methods allow one to reach the desired ( practically arbitrary ) precision of calculations by appropriate choice of the corresponding precision parameter . moreover , for the best precisions , the differences between the transit fluxes generated with the two codes are below @xmath127 .
the computational speed is another important characteristic of the codes for transit synthesis , especially when they are used for the solution of the corresponding inverse problem . in principle
, this speed depends on two parameters : time precision ( step in phase ) and space precision ( size of the differential element of the integrals ) .
the thorough comparison of the computational speeds of different codes for the solution of the same problem requires these codes to be written in the same languages , to be run on the same computers and for the equal model parameters .
that is why it is not reasonable to compare the computational speed of our code with the others .
nevertheless , we made some tests which revealed the following results .
( a ) : : the reducing of @xmath124 ( i.e. the increasing of the precision of the synthetic transit ) by one order leads to 3 - 5 times increase of the computational time while the reducing of @xmath124 from @xmath125 to @xmath128 ( with five orders ) requires around 270 times longer computational time for the code occultnl . (
b ) : : the decreasing of @xmath105 from @xmath129 to @xmath128 ( with seven orders ) requires around 15 times longer computational time .
( c ) : : the formal comparison of the absolute computational speeds of the python code tac - maker and idl code occultnl revealed that our code is slightly faster for high - precision calculations while for the low - precision calculations the code occultnl is faster than the code tac - maker . from these results as well as from the consideration that each idl code is faster than its python version we conclude that the code tac - maker possesses good computational speed for the high - precision calculations .
( d ) : : the new subroutine exofast_occultquad @xcite is around 2 orders faster than its progenitor occultquad .
moreover , we established that exofast_occultquad ( with idl , fortran and python versions ) is considerably improved version of occultquad not only concerning the computational speed but also concerning the precision ( we had found some bugs of occultquad ) . although the computational speed of exofast_occultquad is higher than those of tac - maker and occultnl , it should be remembered that exofast_occultquad can produce transits only for quadratic limb - darkening law .
the previous transit solutions consider the planet as a black lens and the corresponding codes do not fit the planet temperature @xmath102 . as a result
only a formal parameter , the equilibrium temperature of the planet depending on the stellar heating ( i.e. , on stellar temperature and planet distance ) , can be calculated ( out of the procedure of transit modelling ) .
the increasing precision of the observations raises the problem of the planet temperature , i.e. to study the effect of the planet temperature on the transit and to search for a possibility to determine this physical parameter from the observations .
the new approach allowed such tests to be directly carried out because @xmath102 is an input parameter of its own ( unlike the previous methods ) . as a result , we established that increasing of the planet temperature from 0 k to 1000 k causes the decreasing of the transit depth up to @xmath121 while the increasing of the planet temperature from 0 k to 2000 k leads to shallower transit up to @xmath130 .
the contribution of the planet temperature on the transit depth increases both with the ratio @xmath131 and limb - darkening coefficients .
moreover , it rapidly increases with wavelength .
the obtained estimations revealed that the effect of the planet temperature is lower than the precision of the present optical photometric observations , even than that of the _ kepler _ mission .
hence , the determination of the real planet temperature from the observed transit in optics is postponed for the near future . however , the sensibility of the observations at longer wavelengths to the planet emission is higher .
recently , there have been reports that the observed ir fluxes of some hot jupiter systems during occultation are higher than those corresponding to their equilibrium temperatures @xcite .
these results imply that the determination of the planet temperature from the observed transits is forthcoming . until such observational precision
is reached the code tac - maker could be used for theoretical investigations of the effects of different thermal processes ( as ohmic heating , tidal heating , internal energy sources , etc . ) on the planet transits .
although the analytical solutions of the problems are very useful , the modern astrophysical objects and configurations are quite complex to allow analytical descriptions . instead of analytical solutions we are now able to use fast numerical computations .
this paper presents a new solution of the direct problem of the transiting planets .
it is based on the transformation of the double integrals describing the light decrease during the transit to linear ones .
we created the code tac - maker for generation of synthetic transits by numerical calculations of the linear integrals .
the validation of our approach was made by comparison with the results of the wide - spread method for modelling of planet transits .
it was demonstrated that our method gave reasonable and expected synthetic transits for linear , quadratic and squared - root limb - darkening laws and arbitrary combinations of input parameters .
the main advantages of our approach for the planet transits are : ( 1 ) : : it gives a possibility , for the first time , to use an arbitrary limb - darkening law @xmath106 ) of the host star ; ( 2 ) : : it allows acquisition of the stellar limb - darkening coefficients from the transit solution and comparison with the theoretical values of @xcite ; ( 3 ) : : it gives a possibility , in principle , to determine the planet temperature from the observed transits .
our estimations reveal that the effect of the non - zero planet temperature to the transit depth is lower than the precision of the present optical photometric observations .
however , the higher sensibility of the observations at longer wavelengths ( ir ) to the planet emission implies that the determination of the planet temperature from the observed transits is forthcoming .
these properties of our approach and the practically arbitrary precision of the calculations of the code tac - maker reveal that our solution of the planet transit problem is able to meet the challenges of the continuously increasing photometric precision of the ground - based and space observations .
we plan to build an inverse problem solution for the planet transit ( for determination of the configuration parameters ) on the basis of our direct problem solution and by using the derived simple analytical expressions in this paper to obtain initial values of the fitted parameters .
the code tac - maker is available for free download from the astrophysics source code library or its own site .
the research was supported partly by funds of projects do 02 - 362 , do 02 - 85 , and ddvu 02/40 - 2010 of the bulgarian scientific foundation .
we are very grateful to eric agol , the referee of the manuscript , for the valuable recommendations and useful notes .
alonso r. , brown t. m. , torres g. , latham d. w. , sozzetti a. , mandushev g. , belmonte j. a. charbonneau d. et al .
, 2004 , apj , 613 , l153 arnold l. , schneider j. , 2006 , dies.conf , 105 baglin a. , auvergne m. , boisnard l. , lam - trong t. , barge p. , catala c. , deleuil m. , michel e. et al . , 2006 , in 36th cospar scientific assembly .
held 16 - 23 july 2006 , in beijing , china .
bakos g. , noyes r. w. , kovcs g. , stanek k. z. , sasselov d. d. , domsa i. , 2004 , pasp , 116 , 266 batygin , k. , stevenson , d. j. , bodenheimer , p. h. , 2011 , apj 738 , 1 bodenheimer , p. , lin , d. , mardling , r. , 2001 , apj 548 , 466 bodenheimer , p. ; laughlin , g. ; lin , d. , 2003 , apj 592 , 555 borucki w. j. , koch d. , basri g. , batalha n. , brown t. , caldwell d. , caldwell j. , christensen - dalsgaard j. et al . , 2010 , sci , 327 , 977 borucki w. j. , koch d. g. , brown t. m. , basri g. , batalha n. m. , caldwell d. a. , cochran w. d. , dunham e. w. et al . , 2010 ,
apj , 713l , 126 burrows a. , rauscher e. , spiegel d. s. , menou k. , 2010 , apj , 719 , 341 carter j. a. , winn j. n. , 2010 , apj , 716 , 850 claret a. , 2000 , a&a , 363 , 1081 claret a. , 2004 , a&a , 428 , 1001 collier cameron a. , wilson d. m. , west r. g. , hebb l. , wang x .- b . , aigrain s. , bouchy f. , christian d. j. , et al .
2007 , mnras , 380,1230 croll b. , jayawardhana r. , fortney j. j. , lafrenire d. , albert l. , 2010 , apj , 718 , 920 cody a. m. , sasselov d. d. , 2002 , apj , 569 , 451 diaz - cordoves j. , gimenez a. , 1992 , a&a , 259 , 227 dittmann j. a. , close l. m. , green e. m. , fenwick m. , 2009 , apj , 701 , 756 dunham e. w. , borucki w. j. , koch d. g. , batalha n. m. , buchhave l. a. , brown t. m. , caldwell d. a. , cochran w. d. et al . , 2010 , apj , 713 , l136 eastman j. , gaudi s. , agol e. , 2013 , pasp , 125 , 83 etzel p. b. , 1975 , mst , 1 etzel p. b. , 1981 , psbs.conf , 111 gazak j. z. , johnson j. a. , tonry j. , dragomir d. , eastman j. , mann a. w. , agol e. , 2012 , adast , 2012 , 30 gibson n. p. , aigrain s. , pollacco d. l. , barros s. c. c. , hebb l. , hrudkov m. , simpson e. k. , skillen i. et al
. , 2010 , mnras , 404 , l114 gillon m. , demory b .- o . , triaud a. h. m. j. , barman t. , hebb l. , montalbn j. , maxted p. f. l. , queloz d. et al . , 2009 , a&a , 506 , 359 gimnez a. , 2006 , a&a , 450 , 1231 guillot , t. , showman , a. p. , 2002
, a&a 385 , 156 hbrard g. , dsert j .- m . , daz r. f. , boisse i. , bouchy f. , lecavelier des etangs a. , moutou c. , ehrenreich d. et al . , 2010 , a&a , 516 , a95 hubbard w. b. , fortney j. j. , lunine j. i. , burrows a. , sudarsky d. , pinto p. , 2001
, apj , 560 , 413 hui l. , seager s. , 2002 , apj , 572 , 540 jackson b. , barnes r. , greenberg r. , 2008 , mnras , 391 , 237 jackson b. , miller n. , barnes r. , raymond s. , fortney j. , greenberg r. , 2010 , mnras , 407 , 910 kipping d. m. , 2008 , mnras , 389 , 1383 kipping d. m. , 2010 , mnras , 407 , 301 kipping d. , bakos g. , 2011 , apj , 733 , 36 kjurkchieva d. p. , dimitrov d. p. , 2012 ,
iaus , 282 , 474 klinglesmith d. a. , sobieski s. , 1970 , aj , 75 , 175 knutson h. a. , charbonneau d. , allen l. e. , fortney j. j. , agol e. , cowan n. b. , showman a. p. , cooper c. s. et al . , 2007 , natur , 447 , 183 koch d. g. , borucki w. j. , rowe j. f. , batalha n. m. , brown t. m. , caldwell d. a. , caldwell j. , cochran w. d. et al . , 2010 , apj , 713 , l131 kopal z. , 1950 , harci , 454 , 1 latham d. w. , borucki w. j. , koch d. g. , brown t. m. , buchhave l. a. , basri g. , batalha n. m. , caldwell d. a. et al . , 2010 , apj , 713 , l140 mandel k. , agol e. , 2002 , apj , 580 , l171 miller n. , fortney j. j. , jackson b. , 2009 , apj , 702 , 1413 pl , a. , bakos g. . , torres g. , noyes r. w. , fischer d. a. , johnson j. a. , henry g. w. , butler r. p. et al . , 2010 , mnras , 401 , 2665 penev k. , jackson b. , spada f. , thom n. , 2012 , apj 751 , 96 penev k. , sasselov d. , 2011 , apj 731 , 67 poddan s. , brt l. , pejcha o. , 2010 , newa , 15 , 297 pollacco d. l. , skillen i. , collier cameron a. , christian d. j. , hellier c. , irwin j. , lister t. a. , street r. a. et al . , 2006 , pasp , 118 , 1407 pollacco d. l. , skillen i. , collier cameron a. , loeillet b. , stempels h. c. , bouchy f. , gibson n. p. hebb , l. et al .
2008 , mnras , 385 , 1576 popper d. m. , etzel p. b. , 1981 , aj , 86 , 102 rabus m. , alonso r. , belmonte j. a. , deeg h. j. , gilliland r. l. , almenara j. m. , brown t. m. , charbonneau d. et al . , 2009 , a&a , 494 , 391 seager s. , malln - ornelas g. , 2003 , apj , 585 , 1038 seager s. , sasselov d. d. , 2000 , apj , 537 , 916 seager s. , whitney b. , 2000 , aspc , 212 , 232 showman a. , guillot t. , 2002 , a&a 385 , 166 southworth j. , 2008 , mnras , 386 , 1644 southworth j. , maxted p. f. l. , smalley b. , 2004a , mnras , 351 , 1277 southworth j. , maxted p. f. l. , smalley b. , 2004b , mnras , 349 , 547 steffen j. h. , fabrycky d. c. , ford e. b. , carter j. a. , dsert j .- m . , fressin f. , holman m. j. , lissauer j. j. et al . , 2012 , mnras , 421 , 2342 udalski a. , paczynski b. , zebrun k. , szymanski m. , kubiak m. , soszynski i. , szewczyk o. , wyrzykowski l. et al . , 2002 , aca , 52 , 1 winn j. n. , albrecht s. , johnson j. a. , torres g. , cochran w. d. , marcy g. w. , howard a. w. , isaacson h. et al .
, 2011 , apj , 741 , l1 wu y. , lithwick y. , 2012 , astro - ph : arxiv:1210.7810 | we present a new solution of the direct problem of planet transits based on transformation of double integrals to single ones . on the basis of our direct problem solution
we created the code tac - maker for rapid and interactive calculation of synthetic planet transits by numerical computations of the integrals .
the validation of our approach was made by comparison with the results of the wide - spread mandel & agol ( 2002 ) method for the cases of linear , quadratic and squared root limb - darkening laws and various combinations of model parameters .
for the first time our approach allows the use of arbitrary limb - darkening law of the host star .
this advantage together with the practically arbitrary precision of the calculations make the code a valuable tool that faces the challenges of the continuously increasing photometric precision of the ground - based and space observations .
[ firstpage ] methods : analytical methods : numerical planetary system binaries : eclipsing |
the most prominent feature in the magnetic excitation spectrum of yba@xmath1cu@xmath2o@xmath3 ( ybco ) and bi@xmath1sr@xmath1cacu@xmath1o@xmath11 ( bscco ) cuprates is the so - called `` 41mev resonance''@xcite at the antiferromagnetic ( af ) wave vector @xmath12 .
its energy @xmath13 is @xmath14mev in optimally doped samples and decreases with underdoping@xcite down to @xmath15mev .
recently the magnetic response has also been studied by averaging the neutron - scattering data over the in - plane 2d brillouin zone@xcite .
the resulting local magnetic excitation spectrum @xmath16 shows the above - mentioned resonance and a second , hump - like feature at an energy @xmath17 above @xmath13 .
in contrast to @xmath13 , @xmath17 depends only weakly on the doping level . within the calculation to be presented in the following the ` hump '
is naturally explained by particle
hole ( ph ) excitations across the maximum d - wave gap @xmath9 .
the energy @xmath18 comes out almost independent of doping .
the resonance , on the other hand , emerges from a ph - bound state in the magnetic ( spin - flip ) channel and shows a strong doping dependence .
our starting point is the doped mott insulator .
we study the @xmath5@xmath7-model on a simple square lattice of cu-3d orbitals for each of the two coupled cuo@xmath1 layers ( planes ) in ybco or bscco : @xmath19 in the subspace with no doubly occupied orbitals , the electron operator on a cu - lattice site @xmath20 is denoted @xmath21 with spin index @xmath22 ; @xmath23 is the spin - density operator .
a cu - site is specified through @xmath24 $ ] , where @xmath25 indicates the cu - position within one cuo@xmath1-plane and @xmath26 selects the layer .
@xmath27 denotes the effective intra- and inter - layer cu cu - hopping matrix elements , and @xmath28 the antiferromagnetic super exchange . to deal with the constraint of no double occupancy , the standard auxiliary - particle formulation @xmath29 is used .
the fermion @xmath30 creates a singly occupied site ( with spin @xmath31 ) , the `` slave '' boson @xmath32 an empty one out of the ( unphysical ) vacuum @xmath33 .
the constraint now takes the form @xmath34 . in mean - field theory
the constraint is relaxed to its thermal average @xmath35 .
together with the number @xmath36 of doped holes per cu - site , it fixes the fermion and boson densities to @xmath37 the derivation of mean - field equations is presented in ref .. the dynamical spin susceptibility is given in units of @xmath38 as @xmath39 where @xmath40 is the in - plane wave vector , @xmath41 denotes the distance of cuo@xmath1 planes within a double - layer sandwich . for the even @xmath42 and odd @xmath43 mode susceptibilities a rpa - type expression
is obtained , @xmath44 the irreducible part @xmath45 consists of a particle hole ( ph ) bubble of fermions as is known from bcs theory , @xmath46 \frac{f(s'e ' ) - f(s e)}{\omega + se - s'e ' + i0_+}\ ] ] with @xmath47 and @xmath48 for the modes @xmath49 .
boson excitations do not enter @xmath50 on mean - field level .
fermions obey an effective dispersion @xmath51 , @xmath52 , @xmath53 - 4 \widetilde { t}'\cos(k_x ) \cos(k_y ) - \widetilde { t}^\perp({\bf k } ) e^{i \tilde{p}_z}\ ] ] and d - wave gap function @xmath54 , @xmath55 , @xmath56 + \delta^{\perp 0 } e^{i \tilde{p}_z}\ ] ] these enter the usual quasi - particle energy @xmath57 , @xmath58 .
formally , vertex corrections to the simple bubble eq.([eqn - bubb ] ) have to be taken into account . however , these have almost no effect in the energy - range below @xmath59 and are therefore ignored in the following@xcite . from feynman
s variation principle the effective hopping parameters are determined as@xcite @xmath60 , @xmath61 , @xmath62 . for
the bare nearest and next - nearest neighbor hopping we assume @xmath63 , @xmath64 , and for the inter - plane hopping@xcite @xmath65 ^ 2 + t^\perp_0 $ ] with @xmath66 and @xmath67 .
we assume an in - plane superconducting order parameter @xmath9 with equal amplitude and phase in both layers .
the self - consistent solution of the mean - field equations then leads to a vanishing inter - plane gap @xmath68 .
magnetic excitations in the superconducting phase are described by quasi particles ( the fermions ) in a bcs - type d - wave pairing state .
these propagate with effective hopping parameters @xmath69 , @xmath70 strongly reduced from the bare parameters @xmath71 by the small gutzwiller factor @xmath36 .
the bubble eq.([eqn - bubb ] ) describes spin - flip ph - excitations of these particles , which are subject to the mode - dependent final - state interaction in eq .
( [ eqn - birpa ] ) , @xmath72\ ] ] the inter - plane exchange is chosen as @xmath73 . the destruction of the antiferromagnetic ( af ) state of the 1/2-filled system by hole doping@xcite is missing in mean - field theory
. the necessary correlations of fermions and bosons are not contained , and the af order vanishes at an unphysically high doping level@xcite @xmath74 .
therefore we assume@xcite a renormalization @xmath75 of the in - plane nearest neighbor exchange .
using @xmath76 reduces @xmath77 down to @xmath78 , which is consistent with experiment and makes the study of underdoped systems possible .
note that the above - mentioned renormalization @xmath79 of the quasi - particles comes out of the self - consistent calculation , whereas @xmath80 is a phenomenological model .
our assumption of @xmath81 being independent of doping leads to an af correlation length@xcite @xmath82 at @xmath83 , which agrees with known experimental@xcite and theoretical@xcite results .
from the susceptibility eq .
( [ eqn - birpa ] ) the local magnetic excitation spectrum is determined from @xmath84 fig .
[ fig - chi2d ] shows @xmath85 for the odd and even mode at @xmath86 in the superconducting state .
a resonance is clearly visible in the odd @xmath43 mode , at an energy @xmath87mev for @xmath88 near optimal doping . when @xmath36 is reduced ( underdoping ) the resonance moves to lower energies and gains spectral weight .
the resonance appears at the same energy@xcite as in @xmath89 for fixed wave vector @xmath90 .
in addition , both modes @xmath91 show a broad peak ( ` hump ' ) at energies @xmath92 above @xmath13 .
in contrast to @xmath13 the hump - maxima @xmath93 are almost independent of doping , located somewhat below @xmath59 ( @xmath94 for @xmath88 ) .
integrated odd- and even - mode susceptibilities @xmath85 for hole filling @xmath95 .
parameters are @xmath63 , @xmath64 , @xmath96 , @xmath97 .
curves are calculated with a damping fwhm@xmath98mev . *
inset : * @xmath40-integrated bubble spectrum @xmath99 for @xmath100 . the maximum is located close to @xmath101 . ]
the resonance emerges from a pole in eq .
( [ eqn - birpa ] ) at wave - vector @xmath102 and energy @xmath13 , driven by the effective interaction eq .
( [ eqn - effj ] ) in the odd @xmath43 mode .
since @xmath13 is slightly below the threshold @xmath103 to the particle hole ( ph ) continuum , the resonance appears undamped , i.e. , as a @xmath104-function . due to the inter - layer coupling @xmath105 the interaction eq .
( [ eqn - effj ] ) is weaker in the even @xmath42 mode , and the resonance in @xmath106 is shifted up into the ph - continuum , becoming almost suppressed .
consequently , in wave - vector space a sharp peak around @xmath102 is visible only in the odd @xmath43 mode .
this is demonstrated in the top panel of fig.[fig-2dplots ] .
the bottom panel of that figure shows the magnetic response at a higher energy close to the hump - maxima @xmath93 .
the intensity is much reduced compared to the resonance . however , the magnetic excitations at @xmath107 occupy almost the whole 2d brillouin zone , and despite their small amplitude they contribute to the wave vector integrated susceptibility eq .
( [ eqn - chi2d ] ) .
the ` hump ' can be traced back to ph - excitations across the maximum gap @xmath9 : at @xmath90 the irreducible particle
hole bubble @xmath108 shows a log - type van hove singularity ( vhs ) at @xmath109 , remnant of the density of states of the d - wave superconductor . moving off @xmath102 this vhs splits into three vhs that disperse very weakly throughout the brillouin zone , leading to a soft maximum ( ` hump ' ) at @xmath59 in the @xmath110-integrated ( local ) bubble spectrum @xmath111
this is shown in the inset of fig.[fig - chi2d ] .
when the final - state interaction eq.([eqn - effj ] ) is taken into account , the resonance is obtained in the odd mode , and the hump is pulled down to @xmath112 .
also is the doping dependence of @xmath9 compensated : @xmath93 are independent of doping , while @xmath9 increases with underdoping .
two experimental groups studied the wave - vector integrated magnetic response @xmath113 in underdoped ybco . refs . reported a line shape for ybco@xmath114 , which agrees quite well with our theoretical result for @xmath115 .
a ` hump ' in @xmath116 ( even ) appears at @xmath117mev , @xmath118 ( odd ) shows a similar structure at a somewhat lower energy @xmath119mev .
the well - known resonance appears only in @xmath118 at @xmath120mev . in refs .
two underdoped samples ybco@xmath121 and ybco@xmath122 have been studied . in the even ( `` optical '' ) mode of ybco@xmath121
a hump appears around @xmath123mev , whereas the odd ( `` acoustic '' ) mode shows a weak hump - like structure at @xmath124mev , separated from the resonance at @xmath125mev .
these features tend to move up in energy in the more underdoped sample ybco@xmath122 , while the resonance in @xmath118 shifts down to @xmath126mev .
although the detailed experimental line shapes are not unique , the qualitative features of our calculation are found in the neutron - scattering spectra . in particular we reproduce the different dependencies on doping level of the resonance at @xmath13 in the odd mode and the hump - like feature at @xmath93 in both modes . also is @xmath127 of the odd mode lower than the @xmath128 of the even mode . theory and experiments
can also be compared quantitatively .
the measured neutron - scattering intensities@xcite are of the same order as the theoretical ones in fig .
[ fig - chi2d ] ( using @xmath129mev , i.e. , @xmath130ev ) . the maximum of the hump in the even , odd mode in fig .
[ fig - chi2d ] occurs at @xmath131mev , @xmath132mev , in good agreement with the measurements ref . on ybco@xmath121 at low temperature . , @xmath59 , the ph - threshold @xmath103 , and the resonance energy @xmath13 as function of doping at @xmath83 .
the local magnetic excitation spectrum @xmath133 is characterized by two energy scales that behave differently with hole filling @xmath36 ( doping ) .
the resonance energy @xmath13 follows the particle
hole threshold @xmath103 , which in underdoped systems is determined by the chemical potential @xmath134 of the fermions as @xmath135 . the maxima of the humps , on the other hand ,
are determined by the gap @xmath9 through @xmath136 .
when @xmath36 is reduced from optimal doping , @xmath137 and therefore @xmath13 decrease quickly , while @xmath9 increases .
this is displayed in fig .
[ fig - dope ] .
it has been noted above that the mean - field theory describes magnetic excitations in terms of quasi particles ( qp ) ( the fermions ) with a reduced fermi velocity @xmath138 .
hence in underdoped systems the qp s chemical potential comes out ( much ) smaller than the gap , @xmath139 , and thus determines the scale for @xmath13 .
this leads to the observed decoupling of the resonance energy from the gap @xmath9 , while @xmath9 is still visible through the ` humps ' in the local spectrum @xmath113 .
this work has been supported by the deutsche forschungsgemeinschaft through sfb195 and the nsf under mrsec program no.dmr98-08941 . | the wave - vector integrated dynamical spin susceptibility @xmath0 of yba@xmath1cu@xmath2o@xmath3 cuprates is considered .
@xmath4 is calculated in the superconducting state from a renormalized mean - field theory of the @xmath5@xmath6@xmath7-model , based on the slave - boson formulation . besides the well - known `` 41mev resonance '' a second , much broader peak ( ` hump ' ) appears in @xmath8 .
it is caused by particle
hole excitations across the maximum gap @xmath9 .
in contrast to the resonance , which moves to lower energies when the hole filling is reduced from optimal doping , the position of this ` hump ' at @xmath10 stays almost unchanged .
the results are in reasonable agreement with inelastic neutron - scattering experiments .
pacs numbers : 71.10.fd , 74.25.ha , 74.72.bk , 75.20.hr |
dwarf spheroidal galaxies ( dsph ) galaxies are believed to be the most dark matter ( dm ) dominated stellar systems known .
they have low stellar content and are poor in , or entirely devoid of , gas .
they are widely thought of as the smallest cosmological structures containing dm in the universe @xcite and are regarded as key objects in the formation of larger galaxies . according to the @xmath0-cdm cosmological model , complex structures are formed hierarchically @xcite in the potential wells of dm haloes , having density distributions following a navarro , frenk & white ( nfw ) profile @xcite . the dsph galaxies would be formed first in small haloes , and then become involved in the mechanisms of forming larger , complex objects .
the dsph galaxies are characterized by absolute magnitudes in the range @xmath1
m@xmath2 @xmath3 @xcite .
their total estimated masses , considering both the stars and the dm halo , is of the order of @xmath4 m@xmath5 to @xmath6 m@xmath5 , within their half - light radii . with the advent of the sloan digital sky survey ( sdss
* ) , many new faint and ultra - faint dsph galaxies were detected around the milky way ( mw ) ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* and many more ) .
many of these dwarfs are less luminous than a globular cluster ( or even an open cluster ) yet they exhibit high velocity dispersions ( given their luminous mass ) ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) and , similarly to the so - called classical dsphs , are not rotationally supported @xcite .
should these objects be in virial equilibrium , they are the most dark matter ( dm ) dominated objects known in the universe .
they would exhibit mass - to - light ( m / l ) ratios of more than a thousand ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
recent works have demonstrated that dsph galaxies are dm dominated objects @xcite .
they have observed that the stellar velocity dispersion of the classical dsphs is of the order of @xmath7 kms@xmath8 and that it remains approximately constant with distance from the centre of the galaxy .
there are several models that attempt to explain the origin of dsph galaxies by considering different mechanisms .
some of them are based on tidal and ram - pressure stripping @xcite . in these models ,
the dsph galaxies are formed due to the interaction between a rotationally supported dwarf irregular galaxy and a mw - sized host galaxy .
these models show that dsph galaxies tends to appear near a spiral galaxy , but they do not explain the presence of distant isolated dsph galaxies , such as tucana and cetus .
the model proposed by @xcite considers a mechanism known as resonant stripping and can be used to explain the formation of isolated dsph galaxies . basically , these objects are formed after encounters between dwarf disc galaxies and larger systems , in a process driven by gravitational resonances .
there is another model that explains the formation of dwarf galaxies based on energy and momentum conservation when gas - rich galaxies interact @xcite . in this model
the dsph galaxies are regarded as second - generation objects , devoid of dm , forming in the tidal tails of gas - rich interacting galaxies .
all the models cited above consider the interaction between two or more galaxies to explain the formation of dsph galaxies .
there are models which consider dwarf galaxies in isolation @xcite but they usually take only a smooth gas distribution into account and/or focus on higher masses for the dwarfs .
@xcite , for example , perform cosmological re - simulations of dwarf haloes in the mass - range @xmath9 to @xmath10 m@xmath5 using an sph code with a star formation recipe . where the density of the gas is sufficiently high ( amongst other , more sophisticated criteria to mimic a correct global star formation ) , they convert gas particles into star particles , having masses of about @xmath11 to @xmath12 m@xmath5 .
they do not take into account , that those masses in stars would form in an association or in a small dissolving cluster , because it is below their resolution , nor that more massive star clusters could form in a single star forming event . here , we present a different approach .
we also perform numerical simulations of isolated galaxies ( i.e. no other galaxy is involved ) but we distribute the newly formed stars into dissolving star clusters within a dark matter halo , which allow the formation of objects that resemble classical dsph galaxies . our model is based on the assumption that stars never form in isolation but in hierarchical structures ( i.e. star clusters ) ( e.g. * ? ? ?
* ; * ? ? ?
the star formation events range from slowly forming stars in small clusters and associations to intense star - bursts , in gas - rich environments , typically producing a few to a few hundred young star clusters , within a region of just a few hundred pc ( e.g. * ? ? ?
those star clusters form embedded inside a molecular gas cloud and eventually will expel their remaining gas via various feedback processes such as stellar winds , uv - radiation and finally the onset of supernovae ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* and many more and references therein ) .
if the star formation efficiency ( sfe ) is low then the star clusters are not able to form bound entities and instead they will disperse their stars and dissolve . in this paper ,
we propose that the dynamical evolution of these star clusters , i.e. their dissolution due to gas expulsion , may explain the formation of classical dsph galaxies . in our scenario
we simulate scs within a dm halo , as a natural extension of the work of @xcite .
the star clusters form with low sfe and , thus , are designed to dissolve inside the dm halo to form the luminous component of the dsph galaxy .
we follow the evolution of the star clusters within the dm halo for 10 gyr , and then we measure the properties of the object . in our models , we observe fossil remnants of the formation process in the velocity space of the final objects of our simulations , providing predictions for kinematic observations in the future that could test our scenario . in the next section
we explain our idea in more detail followed by sect .
[ sec : setup ] in which we describe the setup of our simulations .
we present the results of our fiducial model in sect .
[ sec : res ] and discuss them in the final section .
our fiducial model is supported by two widely accepted theories .
one of them is the structure formation in the @xmath0-cold dark matter ( @xmath0-cdm ) paradigm . in this scenario
galaxies form hierarchically in the potential well of dm haloes and small haloes form first .
@xmath0-cdm cosmology with its large n - body models of the universe ( e.g. millennium ii simulation of * ? ? ? * ) and the local universe around the mw ( via lactea incite simulation of * ? ? ?
* ) predict that a galaxy like our mw should be surrounded by many of small dm haloes , which should / could host a dwarf galaxy as the luminous component .
the second theory describes the star formation inside of galaxies .
it is now widely accepted that stars form in a clustered mode ( e.g. * ? ? ?
* and follow - up publications ) . in extreme star bursts as in collisions of gas - rich galaxies massive star clusters or even cluster complexes , i.e. clusters of massive star clusters ( e.g. the knots in the antennae * ? ? ?
* ) are formed .
low star formation rates form open clusters or associations , which have low sfe and are more likely to dissolve . in our scenario
we simulate the dm halo of a dwarf galaxy . in its central region ,
i.e. where the luminous component is formed out of the assembled baryons , we insert star clusters which have a low sfe and therefore are designed to dissolve .
these star clusters orbit the central region of the halo , while they are dissolving , thereby forming the faint luminous component of the dwarf .
we have to use dissolving star clusters to form the low - density luminous component , as the merging of bound star clusters will only form bound compact objects ( e.g. * ? ? ?
but as the dsph galaxies are expected to only form stars with a very low star formation rate , this assumption might be justified ( see e.g. * ? ? ?
dsph galaxies which show globular clusters might have formed some star clusters with a high sfe , which allowed some of them to survive and still orbit their dwarf galaxy .
we follow the evolution of the dwarf galaxy for ten gyr and measure its properties at the end .
[ cols=">,^,^,^,^,^,^,^ " , ] = 5.8 cm = 5.8 cm = 5.8 cm = 5.8 cm = 5.8 cm = 5.8 cm in left panel of fig .
[ fig : sigma ] we show the line - of - sight velocity dispersion profiles ( @xmath13 ) for our four simulations and the crosses again show the average of three of our realizations and the three cartesian sight - lines . even though the profiles seem to be quite below @xmath7
kms@xmath8 ( we see a mean central velocity dispersion of about @xmath14 kms@xmath8 ) the profiles are almost flat , out to a radius of @xmath15 pc .
we see similar velocity dispersion profiles in leo ii and sextans @xcite .
the lines show the dispersion profiles of each sight - line and each realisation . here
we see again mostly flat profiles but some profiles show wiggles , dips or bumps .
while observers usually would have large error - bars and would use a smooth curve to fit through the data , we argue that those deviations from a smooth curve are real and due to our formation process .
we also included error - bars for the data - points of the single simulations , single sight - line data - points . but these error - bars are smaller than the actual size of our small symbols as each data - point is based on 1,000 to 60,000 single velocities .
in fact these wiggles and bumps might be the only possible way , with the observations we have today , to verify our scenario .
only realisation 2 shows a very high dispersion , which we discuss in a separate section below .
again we exclude this realisation from our mean values .
the results of our four realisations are also given in tab .
[ tab : vel ] .
= 4.5 cm finally , in the right panel of fig .
[ fig : sigma ] we show the radial velocity of our particles of simulation 1 as function of their radius . in the top panel
we show one percent of our particles to emphasise how well we are able to resolve the velocity space .
the lower panel shows one particle in 1,000 , which compares with the best available observational data , present to date .
again , in this low - resolution plot , such as could be obtained from current observed data , we do not see a strong sign of coherent motion , even though an inspection by eye of the full simulation data ( upper panel ) of fig [ fig : sigma ] would suggest some bi - modality in the very centre .
if we use all stars within a radius of @xmath15 pc to compute an overall velocity dispersion we get @xmath16 kms@xmath8 which resembles the values observed in classical dsphs as reported by @xcite .
this mean value of the velocity dispersion velocity is calculated considering the lines - of - sight along the three coordinates , because the sight - line of the dsph galaxy towards us with respect to our formation scenario is unknown . in fig .
[ fig : overview ] , we show the dynamics of the first realisation within @xmath15 pc as two - dimensional pixel maps with a resolution of @xmath17 pc . in the left panel
we show the line - of - sight velocity dispersion calculated for each pixel separately .
we see regions of high velocity dispersion of more than @xmath7 kms@xmath8 in the central area but also at the locations where the off - centre density peaks are . in high resolution
the distribution of dispersions is far from being smooth .
in fact we see even more sub - structure than in the density plots . while the dm cusp was able to erase most of the structure in positional space , the structures in velocity space survived .
we call these structures in velocity space which stem from the formation process of the dwarf fossil remnants. still we see high values of velocity dispersion throughout the visible part ( surface brightness above @xmath18 magarcsec@xmath19 ) of the dsph galaxy . in the middle panel
we show the mean velocity of all particles within a pixel .
the figure shows clearly regions which seem to exhibit coherent streaming motions with velocities differences of up to @xmath20 kms@xmath8 .
this means , even after @xmath7 gyr of evolution we still see the stars of dissolved star clusters , which were on rather circular orbits in coherent motion .
the panel shows at least two structures of opposing radial velocities .
a similar scenario was proposed by @xcite to explain the velocity structure of umi ( though in the plane of the sky and therefore with no detectable offset in mean velocity ) .
we define a @xmath21-parameter : @xmath22 where @xmath23 is the mean velocity measured in the radial bin or later in a pixel and @xmath24 is the velocity dispersion measured within a projected @xmath15 pc radius .
this parameter allows us to quantify how strong coherent motions in the object are .
the function @xmath21 corresponds to a re - normalization of the mean velocities , which allows to see easily regions where the mean velocities are distinct from zero .
higher values of @xmath21 correspond to regions where streaming motions and/or rotation are present .
while the value of the @xmath21-parameter depends on the strength of the velocity deviations in our model , the significance of those values , i.e. if they could be due to noise in the data - sample , depend on the number of velocities in the sample .
if we would apply a strict criterion to regard only the deviations which have @xmath21-values higher than possible @xmath25-deviations ( due to noise ) as real , then we have to have values larger than @xmath26 , @xmath27 and @xmath28 for sample sizes of @xmath7 , @xmath29 and @xmath30 stars respectively . for more detailed values
see table 1 in @xcite .
the right panel shows the @xmath21-parameter .
if no streaming motion is present we expect the @xmath21 values to be randomly distributed without correlation . as our high resolution plots have more than 300 and up to 3000 particles in each pixel , the @xmath21 values we see of up to @xmath31 are definitely significant and not due to noise produced by low particle numbers .
we speculate above that our fossil remnants could be related with star clusters which dissolved on rather circular orbits . in fig .
[ fig : s28 ] we show the particles of one of our dissolved clusters .
its initial orbit is shown as a green line on top of the particles , which got spread out to much larger distances than the original orbit .
still they form a ring - like structure , where at the upper end of the ring are more particles with negative velocities , while at the lower end we see particles with positive velocities .
but the ring is too large for its ends to be the explanation of the two places of high delta values we see in the right panel of fig .
[ fig : overview ] ( top right and lower left area ) .
if we zoom in ( right panel ) , we see that even though the ring is much larger there are areas where the particles of this dissolved cluster have predominantly positive or negative velocities ( circles ) .
these areas coincide with the pixels of high delta values . in fig .
[ fig : vdist ] we focus now on one pixel with a relatively high @xmath21-value of @xmath32 .
it is located in the upper right area of fig .
[ fig : overview ] .
we plot the histogram of radial velocities of all 808 particles in this pixel ( taken from all dissolved scs , including the one we show in fig .
[ fig : s28 ] ) .
the top histogram , which has a good resolution of @xmath33 kms@xmath8 , shows that we have a complicated structure with several peaks and that the majority of particles have negative velocities .
if we fit a gaussian to this data we obtain a mean value of @xmath34 kms@xmath8 and a @xmath35 kms@xmath8 ( skewness : @xmath36 ; kurtosis : @xmath37 ) .
if we would have a normal non - rotating velocity distribution , the mean value has to be zero and the velocity dispersion reflects the strength of the potential according to the jeans equation .
in reality , our histogram reflects many different streams of particles from many different dissolved star clusters rather than a smooth distribution of a uniform object in virial equilibrium .
therefore , the peak is shifted and the measured velocity dispersion is much higher than expected .
this is clearly visible if we look at the cumulative histograms of the different dissolved star clusters , which are shown as the thin , solid lines ( black ) in fig .
[ fig : vdist ] .
larger areas without lines show where some star clusters deposit a lot of stars ,
i.e. we have a stream of stars from that particular cluster inside this pixel .
we also see that we do not have @xmath18 large areas inside this pixel , i.e. we do not find stars from every cluster in this pixel .
this reflects the complicated velocity structure we have throughout the dwarf galaxy .
the particles of the different scs are not distributed evenly and smoothly throughout the whole available phase - space . some of them still occupy distinct regions , reflecting the former orbit of the sc they came from .
the better the resolution is , with which we view the object ( smaller areas , larger particle samples ) the more visible are these structures . in the lower panel
we plot the same velocity sample with half the resolution and the complicated multi - peak structure disappears , only the shifted mean and the large @xmath38-value remain .
if we sample the same amount of velocities randomly from all over the dwarf ( red , dashed histograms in fig .
[ fig : vdist ] ) the distribution is indistinguishable from a normal gaussian ( skewness : 0.055 ; kurtosis : 3.305 ) .
the mean value is centred on zero ( @xmath39 ) and @xmath40 kms@xmath8 ( remember , we sample from the whole object , i.e. also from the very outer parts showing low dispersions ; the velocity dispersion of all particles within @xmath15 pc is still @xmath41 kms@xmath8 ) .
the result of these complicated velocity structures or fossil remnants as we dub them is that we could see regions with high velocity dispersion and regions with deviations in mean velocity , if we have high enough resolution , i.e. having both : small pixel size and still enough stellar velocities per pixel . if we do not have the necessary resolution we still could see bumps and wiggles in the velocity dispersion profiles and we might observe a velocity dispersion that is enhanced relative to the equilibrium value for a smooth model . if we use the jeans equations to estimate the mass of the system we could then be biased to larger values of the mass .
again we have to discuss our singled out model separately .
[ fig : sigma ] shows that this model exhibits a cold core with a rising velocity dispersion outside the nucleus , reaching up to @xmath42 kms@xmath8 and dropping again to values similar to the other random realisations . as said in the previous section
, this simulation exhibits a nucleus , which we do not find in the known dsph galaxies of the mw .
but as expected for a nucleus inside a prominent dm halo we see a dip in the velocity dispersion .
the maximum in the dispersion is due to the fact that in this particular model quite a lot of the luminous mass got trapped in the nucleated centre instead of being more distributed like the other models .
= 4.1 cm = 4.1 cm = 4.1 cm = 4.1 cm = 4.1 cm = 4.1 cm = 4.1 cm = 4.1 cm we try to compare our model with the available fornax data in fig .
[ fig : lores ] .
the top left panel shows the velocity differences of the central part of our model with a resolution of @xmath29 pc per pixel .
overlaid is a contour of the surface brightness . within this contour
the velocity structure could be mistaken as a signal of rotation around the minor axis or a velocity gradient of about @xmath20 kms@xmath8 . underneath
we see that the pixels showing the strongest deviation have a @xmath21-parameter of about @xmath43 .
now with our particle resolution we have more than 10,000 velocities in each pixel and we can be sure that this velocity structure is real and not due to noise .
still , with this low resolution most signals of our complicated velocity structure are erased .
the biggest structure is just visible in the top panel due to a dark red and a dark blue pixel and in the bottom panel through one dark pixel on either side of the object .
we conclude therefore , that having only low resolution one can not see coherent motions ( several pixels together showing the same behaviour ) but rather has to search for symmetric pixels with high deviations .
on the right hand side we try to produce a similar figure of the inner part of fornax .
fornax has almost 2,000 radial velocities publicly available . in the region
printed we have about 800 velocities .
still , we have just around 20 velocities per pixel ( minimum value : 8) even with pixels this large .
there are pixels which show high deviations in the mean velocity but first of all the strength of these deviations is still within the possible noise ( about a one - sigma deviation from the system velocity , a @xmath25-deviation due to noise would result in delta values of @xmath26 for ten stars or @xmath44 for @xmath17 stars ) and second we can not detect any symmetry between the pixels with high values .
so we have to conclude that the presently available data of fornax does not show any visible fossil remnant , as we do not have sufficient data to disentangle possible sub - structure in velocity space from the possible level of noise . once again
: even though our model and fornax show deviations from the systemic velocity of the same strength , in our model these deviations are well above any noise and show symmetries , while fornax lacks sufficient data to enhance any structure , if present , above the possible level of noise .
if we would degrade our data to the same low resolution in pixel size _ and _ particle sampling , as available for fornax , not a single fossil remnant would remain visible above the noise level .
in this section we discuss in more detail the observational detectability in real dsphs of the features we have seen in our simulation . if we analyse fig .
[ fig : smooth ] in terms of detectability we find that the visible sub - structure in the surface brightness ( bottom left panel ) amounts to tiny differences in actual mass - densities . within a @xmath45 pc pixel
the difference between the dark red and red pixels amounts to a difference of about @xmath17 m@xmath5 . in our simulations we can easily resolve such mass differences . in reality , regarding the fact that most of this mass - difference will be in faint low - mass stars , which are hard to detect at the distance of the dsph galaxies even with modern telescopes , these sub - structures will be below the detection limit .
maybe the next generation telescopes like e - elt , tmt or gmt will be able to make those structures visible .
but effects like slightly off - centre nuclei ( sextans ) , secondary density peaks ( ursa minor ) or dents in the contours ( draco ) ( as seen in @xcite ) are now potentially explainable with our scenario as low - resolution counterparts of a more complicated structure due to the formation process . looking at the velocity dispersion profiles of our fiducial models , we find similar radial profiles to some of the classic dwarfs of the mw . if we recall the profiles of the different realisations of our standard model in fig .
[ fig : sigma ] we see that some of them show not only an almost flat profile but also bumps or dents . in our simulations
these deviations from a smooth curve are not due to noise ( as we have 1,000 to 60,000 velocities in one bin ) but are intrinsic deviations stemming from our formation scenario .
they are hints to the fact that a more complicated velocity structure exists in our model . in reality
@xcite we see similar bumps in carina , leo i and sculptor , even though the observational error - bars are still much larger than the deviations . even though some of the classical dwarfs have now more than 2,000 radial velocities , it is still not feasible to produce 2d velocity maps of the dispersion or the mean velocity with a sufficiently high resolution to detect fossil remnants , as we see in fig .
[ fig : lores ] where we compare our simulations with fornax data set .
if with next generation telescopes several thousands of radial velocities more for each dsph galaxy would become available , significant sub - structure in velocity space might emerge above the expected level of noise . in our fiducial model
we found a maximum @xmath21-value of @xmath46 , if we degrade our resolution to a pixel - size of @xmath29 pc . in order for this feature to be detectable above the noise at a statistically significant level
, we would need at least @xmath29 velocities per pixel ( to be above a @xmath25 noise level ) .
another interesting property of our model is that if you would draw an imaginary line along the maximum extension of our object in surface brightness ( red / orange area in the left panels of fig .
[ fig : smooth ] or grey area in the middle panels of the same figure , i.e. the projected major axis ) , the position angle of this line would be different to a similar line drawn through the maximum extension of high velocity dispersion ( red / orange area in the left panel of fig . [
fig : overview ] ) .
furthermore , a line through the two regions of high mean velocity deviations ( blue and red area in the middle panel of fig .
[ fig : overview ] or two main dark areas in the right panels of the same figure ) , again would show a different position angle .
in other words , if one would fit ellipses of constant surface brightness and constant velocity dispersion to our data shown in fig . [
fig : smooth ] and fig .
[ fig : overview ] ( left panel ) , their position angles would differ .
furthermore , searching for the position angle along which we see the highest velocity gradient in our data ( middle panel of fig .
[ fig : overview ] ) , this angle would differ from the other two .
this is something which might only occur in our formation scenario and therefore would be a measurable quantity .
unluckily , a perfect alignment would not exclude our scenario .
just a perfect alignment in all known dwarfs would render our scenario very unlikely .
one important ingredient of our models is the low sfe and therefore the subsequent dissolution of the scs . only with the dissolution of the scs we are able to lower the phase space density of the stars sufficiently to obtain low luminosity objects like dsph galaxies .
merging scs without expansion always lead to compact objects , even without the presence of a dm halo @xcite .
we see the difference in our model 2 , which had a chance merger early on in the centre , thereby preventing the merged scs from dissolving . as such objects as model 2
are not observed , we could conclude that the scs which formed the known dsph had even lower sfes and slower sfr than adopted in our model .
last but not least , this paper only talks about our fiducial model , i.e. one particular set of parameters that without much tweaking results in an object which resembles ( as shown in the section above ) a classical dwarf spheroidal galaxy in terms of appearance , scale - length and velocity dispersion ( without matching one particular dwarf of the mw in every aspect ) . a thorough investigation of the vast possible parameter space ( @xmath47 , @xmath48 , @xmath49 , shape of the dm halo , distribution and virial ratio of the scs , ... ) will be dealt with in the follow - up papers we wish to publish .
finally , one should bear in mind that after all , our model is still a toy model making a lot of simplifications ( all scs form at the same time , have the same mass and sfe , gas is lost entirely from the system , ... ) , some justified , some because we have no better knowledge . also here
we have to investigate a much larger parameter space , which we will do in the future .
a somewhat delayed formation of some of the scs might be a plausible explanation of the spread in metallicity we see in dsph galaxies .
if not , a low sfr with low sfe also implies that we will lack high - mass stars and strong super - nova explosions @xcite which might be able to blow the remaining gas out of the dm halo . retaining the gas could then lead to another epoch of star formation later on .
in this paper we tested a possible scenario for the formation of dsph galaxies with numerical simulations using the particle mesh code superbox . in our simulations we considered the evolution of @xmath50 star clusters within a dark matter halo with an nfw profile .
we observe that these star clusters dissolve due to the low sfe @xmath51 assumed and form an extended luminous object which resembles a dsph galaxy and after @xmath7 gyr of evolution has a projected diameter of about @xmath52 kpc . one major result to take away from our models
is that even @xmath7 gyr of evolution is not enough to erase sub - structures stemming from our formation process inside of the dsph .
these substructures are mainly visible in velocity space . because of these sub - structures , i.e.coherent motions , the measurements of the velocity dispersion of the object may be affected .
for this reason our fiducial model can mimic the observations of a classical dsph galaxy even within a halo that only has @xmath53 m@xmath5 , a value somewhat lower than the canonical values @xmath54 @xcite obtained by using the measured velocity dispersion and assuming complete virial equilibrium without coherent motions .
the external structural remnants of the formation process in our model are very faint , too faint to be observed in current data sets .
the central structures which are within @xmath15 pc , have observable luminosities comparable to the luminosity of the dsph galaxies of the mw .
but the sub - structures stemming from the formation process visible in the two - dimensional brightness distribution of our model are only visible through our ultra - high particle resolution and are currently unobservable , as long as we can not see all the stars of the dwarf . if such structures are present , only future telescopes like the e - elt might be able to observe them . in velocity space
the object has the overall properties of a typical classical dsph galaxy .
we show that the velocity dispersion of this object is around @xmath7 kms@xmath8 and that it has a flat velocity dispersion profile in projection .
again this resembles closely the observations of the known classical dwarfs . besides we show the existence of streaming motions in this object even after @xmath7 gyr of evolution .
these streams are fossil remnants of the dissolving star clusters considered in our simulations , mainly of those with more circular orbits . as mentioned above these kinematical sub - structures only become visible in high - resolution pixel - maps .
the current state of the art measurements in real dsph galaxies using up to and more than 2,000 velocities in total is astonishing but do still not give enough resolution to find those fossil remnants , predicted by our model , in reality .
our model showed one strong fossil remnant , which might be detectable in real observations possible today .
this structure is still visible if we reduce our resolution to a resolution possible with present day observational data .
it might be an explanation for the ring - like structure suspected in umi or it could be easily mistaken as a velocity gradient throughout the luminous component , as for example observed with uma ii and her .
it is important to note that these structures have 1d counterparts .
we see in the dispersion profiles dents and bumps . in our simulations
these are not errors of the measurement but real features .
the same behaviour is visible in some of the mw dsph galaxies . in observational papers
these dents or bumps are smoothed over because the data - points have much larger error - bars but according to our scenario some of them could be real .
our simulations show that our new scenario for the formation of dsph galaxies , joining two basic astronomical principles , indeed works .
we are not only able to reproduce the observational data we have today , but our models also provide observers with predictions for future high - precision and high - resolution observations .
+ * acknowledgments : * pa wishes to thank w. dehnen for his help with the nfw profiles .
m. metz for his support with superbox .
pa gratefully acknowledges support from a conicyt phd studentship , mecesup fms0605 , fondap no . 15010003 , basal cata grant pfb-06/2007 , fondecyt no .
1100540 and fondecyt no .
mf acknowledges financial support through fondecyt grant no .
1095092 and basal .
miw acknowledges the royal society for support .
rs is funded through a comite mixto grant and fondecyt no .
3120135 .
belokurov v. , zucker d.b .
, evans n.w .
, wilkinson m.i .
, irwin m.j . , hodgkin s. , bramich d.m . , irwin j.m . , gilmore g. , willman b. , vidrih s. , newberg h.j . , wyse r.f.g .
, fellhauer m. , hewett p.c . ,
cole n. , bell e.f . , beers t.c .
, rockosi c.m . , yanny b. , grebel e.k . , schneider d.p .
, lupton r. , barentine j.c . , brewington h. , brinkmann j. , harvanek m. , kleinman s.j .
, krzesinski j. , long d. , nitta a. , smith j.a . ,
snedden , s.a .
2006 , apj , 647 , l111 belokurov v. , zucker d.b .
, evans n.w .
, kleyna j.t . , koposov s. , hodgkin s.t . , irwin m.j .
, gilmore g. , wilkinson m.i . , fellhauer m. , bramich d.m .
, hewett p.c . , vidrih s. , de jong j.t.a . , smith j.a . ,
rix h .- w .
, bell e.f . ,
wyse r.f.g . ,
newberg h.j .
, mayeur p.a . , yanny b. , rockosi c.m . , gnedin o.y . ,
schneider d.p . , beers t.c .
, barentine j.c . , brewington h. , brinkmann j. , harvanek m. , kleinman s.j . , krzesinski j. , long d. , nitta a. , snedden , s.a . 2007 ,
apj , 654 , 897 bressert e. , bastian n. , gutermuth r. , megeath s.t .
, allen l. , evans n.j.ii , rebull l.m . , hatchell j. , johnstone d. , bourke t.l .
, cieza l.a . , harvey p.m. , merin b. , ray t.p . , tothill n.f.h .
2010 , mnras , 409 , l54 koposov s.e . , gilmore g. , walker m.g . , belokurov v. , evans n.w .
, fellhauer m. , gieren w. , geisler d. , monaco l. , norris j.e .
, okamoto s. , penarrubia j. , wilkinson m. , wyse r.f.g . , zucker d.b .
2011 , apj , 736 , 146 willman b. , dalcanton j.j . , martinez - delgado d. , west a.a .
, blanton m.r .
, hogg d.w .
, barentine j.c . ,
brewington h.j . , harvanek m. , kleinman s.j . , krzesinski j. , long d. , neilsen e.h.jr . ,
nitta a. , snedden s.a .
2005 , apj , 626 , l85 zucker d.b . , belokurov v. , evans n.w . , wilkinson m.i .
, irwin m.j . , sivarani t. , hodgkin s. , bramich d.m .
, irwin j.m . , gilmore g. , willman b. , vidrih s. , fellhauer m. , hewett p.c . , beers t.c . , bell e.f .
, grebel e.k . ,
schneider d.p . ,
newberg h.j . ,
wyse r.f.g . ,
rockosi c.m . , yanny b. , lupton r. , smith j.a . , barentine j.c . , brewington h. , brinkmann j. , harvanek m. , kleinman s.j .
, krzesinski j. , long d. , nitta a. , snedden s.a .
2006 , apj , 643 , l103 | we use numerical simulations to study a formation scenario for dwarf spheroidal galaxies in which their stellar populations are the products of the dissolution of open star clusters and stellar associations within cosmological dark matter haloes .
this paper shows that this process gives rise to objects which resemble the observed dwarf spheroidal satellites of the milky way without invoking external influences .
the presence of long - lived kinematic substructures within the stellar components of these objects affects their projected velocity dispersions .
we find that this in turn affects mass estimates based on the projected velocity dispersion profiles which may over - estimate the actual dark matter halo mass depending on the amount of substructure which is present .
our models make predictions about the detailed kinematic and photometric properties of the dsphs which can be tested using future observations .
[ firstpage ] galaxies : dwarfs galaxies : star clusters methods : n - body simulations |
in the standard picture ( e.g. ) subdwarf b ( sdb ) stars are core helium burning stars on the extended horizontal branch ( ehb ) .
they consist of a he core with a mass of @xmath00.48@xmath1 , the canonical mass for a he core flash on the first red giant branch ( rgb ) , and a thin , inert hydrogen shell .
evolutionary models of ehb stars calculated by and adopted an enhanced mass loss on the rgb , however , without specifying a particular mechanism .
recent radial velocity ( rv ) surveys revealed that a large fraction of all sdb stars resides in close binaries (; ; ) .
@xcite ( @xcite , @xcite ) performed a theoretical investigation of possible close binary channels for the formation of sdbs : stable roche lobe overflow , common envelope ejection , and merging of two he core white dwarf ( wds ) .
they concluded that it is possible to explain _ all _ sdb stars as the result of close binary evolution .
the helium - rich subdwarf o stars ( he - sdo ) are hotter than sdbs , but have similar surface gravities .
their formation is still a mystery . in one scenario
they are explained as an advanced stage of post - ehb evolution ( , ) .
another scenario proposes the formation of he - sdos via a phase as extreme he stars after the merging of two he core wds ( cf .
discussion in ) .
spy , the eso * s*upernovae type ia * p*rogenitor surve*y * ( @xcite @xcite , @xcite ) , is a programme dedicated to search for short period binary wds ( double degenerates dds ) .
the aim of spy is the detection of dd progenitors of supernovae type ia by means of a survey for rv variations .
snia progenitor candidates should be close enough to merge within one hubble time due to gravitational wave radiation and the combined mass should exceed the chandrasekhar limit for wds . the spy input catalogue collected wd candidates brighter than @xmath2 from a variety of source catalogues ( cf . ) .
the sdbs analyzed in this paper were selected from the hamburg / eso survey @xcite , the wd catalogue , and the hamburg quasar survey @xcite . as a by - product ,
spy produced high accuracy rv measurements of more than 46 sdb stars and 23 he - sdo stars , which were included in our sample because of misclassifications in the input catalogues .
thus spy produces a large sample of known close binary sdb stars , offering an independent data set and allowing for a substantial improvement of the statistics .
the he - sdo sample included in spy is the first he - sdo sample systematically checked for rv variations .
this allows us to investigate the role of binarity for their formation and their possible link with the sdb stars .
spy is carried out with the high - resolution spectrograph uves at the ut2 telescope of eso vlt . with our instrument setup
we achieve nearly complete spectral coverage from 3200 to 6650with only two @xmath080 wide gaps .
since spy was implemented as a bad weather program , we used a wide ( @xmath3 ) slit to minimize slit losses .
the resulting spectral resolving power is @xmath4 ( 0.36 at h@xmath5 ) or better , if seeing disks were smaller than the slit width . due to the nature of the project ,
two spectra at different `` random '' epochs separated by at least one day are observed ( cf . , for details ) .
the rv variations were measured differentially between both observed spectra with a cross correlation routine .
this allowed to include all suitable lines ( h , he , and metal lines ) for the rv determination in a very flexible way .
we routinely achieved a rv accuracy of 2km / s or better .
rv variations could be checked for 46 single - lined sdbs from the complete spy sample ( , @xcite ) .
their distribution in the temperature - gravity plane is displayed in fig .
[ f : hr ] .
our rv measurements yielded 18 rv variable binaries , i.e. 39% of this sample .
this indicates a high fraction of close binaries in the spy sample , conforming the importance of binary channels for the formation of sdbs .
however , somewhat surprisingly our result points to a lower fraction of detections than in the sample , who found 58% rv variable sdbs . for a quantitative comparison with other samples and an evaluation of the true fraction of binaries one needs to know the detection efficiency , i.e. the chance that a binary escaped detection , because of unfavorable inclination angles or phasing of the observations .
the detection efficiency obviously depends on the orbital period and the mass of the unseen companion .
we performed a monte carlo simulation of our observed sample for two different choices of the mass of the invisible companion : @xmath6 , the mass of a typical wd , and @xmath7 for a typical low mass main sequence star .
for every star we calculated detection probabilities taking into account the timing of exposures and the rv accuracy .
results for the sample were coadded yielding the effective sample size ( i.e. the size of an equivalent sample with 100% binary detection rate ) plotted in fig .
[ f : detection ] as function of orbital period .
for an alternative interpretation we indicate the relative percentage of the detection rate .
the detection probabilities are quite high for short period systems , exceeding 90% for systems with a wd companion and periods shorter than one day .
.corrections for detection efficiency for the sdb and the he - sdo samples .
calculations were done for the two distributions discussed in the text , the canonical sdb mass of @xmath8 , and different companion masses . [ cols= " < ,
< , < , < , < , < " , ] the very low number of close binaries in our he - sdo sample provides evidence that he - sdos , as a class , are not successors of ehb stars .
formation of he - sdos by merging of two he core wds would be in agreement with our finding . how do our sdb results compare with the binary population synthesis of ? @xcite produced several simulated samples of sdbs resulting from their binary evolution channels with different parameter choices and presented a best fit sample ( their simulation set 2 ) .
they could reproduce the properties of observed sdb binaries ( from ) and the distribution in the temperature - gravity plane ( cf . , @xcite ) .
observational selection effects against sdbs with visible cool companions ( gk selection ) and against stars above the ehb ( strip selection ) were simulated .
they predict an observable binary frequency for their best fit model of 45% ( the remainder resulting from merging ) , which compares well to our 40% , if we apply the same criteria to the spy sample
. however , two problems remain .
simulated a population with thin disk characteristics , which should be representative for the sample , but this sample yielded a binary frequency of 69% . on the other hand predict a low binary frequency for stars above the ehb , because this region should contain many sdbs resulting from merging .
although a systematic observational investigation of stars in this region has still to be done , it appears that the binary frequency in this region is higher than expected ( see fig .
[ f : hr ] ) from the theoretical simulation .
a further complication might arise if some or all of the mergers produce he - sdos , instead . in summary
the new theoretical simulations represent the properties of the observed sdb sample well , but some unsolved problems remain .
future investigations will help to decide , if these can be solved by minor adjustments or imply larger revisions .
ahmad , a. & jeffery , c.s .
: 2003 , _ a&a _ 402 , 335 dcruz , n.l . ,
dorman , b. , rood , r.t .
, & oconnell , r.w .
: 1996 , _ apj _ 466 ,
359 dorman , b. , rood , r.t . , &
oconnell , r.w .
: 1993 , _ apj _ 419 ,
596 dreizler , s. , heber , u. , werner , k. , moehler , s. , de boer , k.s . : 1990 , _
a&a _ 235 , 234 edelmann , h. , heber , u. , lisker , t. , green , e.m .
: 2003 , _ ap&ss _ , these proceedings hagen , h .- j . ,
groote , d. , engels , d. , & reimers , d. : 1995 ,
_ a&as _ 111 , 195 han , z. , podsiadlowski , p. , maxted , p.f.l . ,
marsh , t.r .
, & ivanova , n. : 2002 , _ mnras _ 336 , 449 han , z. , podsiadlowski , p. , maxted , p.f.l .
, & marsh , t.r . : 2003 , _ mnras _ 341 , 669 heber , u. : 1986 , _ a&a _ 155 , 33 lisker , t. , heber , h. , napiwotzki , r. , et al . : 2003a , _ ap&ss _ , these proceedings lisker , t. , heber , h. , napiwotzki , r. , et al . : 2003b , _ a&a _ , submitted maxted , p.f.l . ,
heber , u. , marsh , t.r . , & north , r.c . : 2001 , _ mnras _ 326 ,
1391 mccook , g.p .
& sion , e.m . : 1999 , _ apjs _ 121 , 1 morales - rueda , l. , maxted , p.f.l . , marsh , t.r . ,
north , r.c . , & heber , u. : 2003 , _ mnras _ 338 752 napiwotzki , r. , christlieb , n. , drechsel , h. , et al .
: 2001a , _ an _ 322 , 411 napiwotzki , r. , edelmann , h. , heber , u. , et al .
: 2001b , _ a&a _ 378 , l17 napiwotzki , r. , christlieb , n. , drechsel , h. , et al .
: 2003 , _ eso messenger _ 112 , 25 saffer , r.a . , livio , m. , & yungelson , l.r .
: 1998 , _ apj _ 502 , 394 wesemael , f. , winget , d.e . , cabot , w. , van horn , h.m . , & fontaine , g. : 1982 : _ apj _ 254 , 221 wisotzki , l. , koehler , t. , groote , d. , & reimers , d. : 1996 , _ a&as _ 115 , 227 | we present the results of a radial velocity ( rv ) survey of 46 subdwarf b ( sdb ) and 23 helium - rich subdwarf o ( he - sdo ) stars .
we detected 18 ( 39% ) new sdb binary systems , but only one ( 4% ) he - sdo binary .
orbital parameters of nine sdb and sdo binaries , derived from follow - up spectroscopy , are presented .
our results are compared with evolutionary scenarios and previous observational investigations . |
in the union @xmath0 , fig . [ f1 ] , b and a , of the unit straight strip@xmath10 and a rectangle of length @xmath11 and a small width @xmath12@xmath13 \right\ } , \label{0}\ ] ] we consider the spectral neumann problem@xmath14 [ ptb ] newcaduna.epsc:/users/desktop/documents/google drive / articoli / articoli - in - preparazione / cardonedurantenazarov1/graphics / newcaduna__1.pdf where @xmath15 is the laplace operator , @xmath16 @xmath17 is the spectral parameter and @xmath18 is the directional derivative , @xmath19 stands for the unit outward normal defined everywhere at the boundary @xmath20 except for corner points , i.e. vertices of the rectangle ( [ 0 ] ) .
since a solution of the problem ( [ 1 ] ) , ( [ 2 ] ) may get singularities at these points , the problem ought to be reformulated as the integral identity @xcite@xmath21 where @xmath22 is the natural scalar product in the lebesgue space @xmath23 and @xmath24 stands for the sobolev space .
the symmetric bilinear form on the left - hand side of ( [ 3 ] ) is closed and positive in @xmath24 so that problem ( [ 1 ] ) , ( [ 2 ] ) is associated ( * ? ? ?
* ch 10 ) with a positive self - adjoint operator @xmath25 in @xmath26 whose spectrum @xmath27 is continuous and covers the closed positive semi - axis @xmath28 the domain @xmath29 of @xmath30 of course , belongs to @xmath31 but is bigger than @xmath32 due to singularities of solutions at the corner points , see , e.g. , ( * ? ? ?
the point spectrum @xmath33 of @xmath25 can be non - empty and the main goal of our paper is to single out a particular value of the length parameter @xmath34 such that the operator @xmath25 wins an eigenvalue @xmath35 embedded into the continuous spectrum .
the corresponding eigenfunction @xmath36 decays exponentially at infinity and is called a trapped mode , cf .
@xcite and @xcite .
our central result formulated below in theorem [ theoremex ] , roughly speaking , demonstrates that an eigenvalue @xmath37 exists in the interval @xmath38 for @xmath39 with @xmath40 only .
asymptotics of @xmath17 and @xmath41 are constructed , too .
problem ( [ 1 ] ) , ( [ 2 ] ) is a model of an acoustic waveguide with hard walls , cf .
@xcite , but is also related in a natural way to the linear theory of surface water - waves , cf .
@xcite . indeed , the velocity potential @xmath42 satisfies the laplace equation in the channel @xmath43 of depth @xmath44 with the neumann condition ( no normal flow ) at its vertical walls and horizontal bottom as well as the spectral steklov condition ( the kinetic one ) on the free horizontal surface@xmath45 after factoring out the dependence on the vertical variable @xmath46 , @xmath47 see , e.g. , @xcite and @xcite , the water - wave problem reduces to the two - dimensional neumann problem ( [ 1 ] ) , ( [ 2 ] ) for the function @xmath48 in ( [ 4 ] ) and the parameter @xmath17 determined from the equation@xmath49 we will not discuss separately this interpretation of our problem but in the next section present some asymptotic formulas for eigenvalues of the laplace operator with either dirichlet , or mixed boundary conditions . imposing the dirichlet condition@xmath50 instead of the neumann condition ( [ 2 ] ) , creates the positive cut - off value @xmath51 of the continuous spectrum @xmath52 of the dirichlet problem ( [ 1 ] ) , ( [ 5 ] ) which provides an adequate model of a quantum waveguide , cf .
the interval @xmath53 stays now below the continuous spectrum and therefore may contain eigenvalues composing the discrete spectrum @xmath54 of the problem . as follows from a result in @xcite ,
the multiplicity @xmath55 is equal to @xmath56 for a small @xmath2 .
although the paper @xcite deals with a regular ( smooth ) perturbation of the wall , it is possible to select two smooth shallow pockets boxes as in fig . [ f2 ] , a and b , and to extend the existence and uniqueness result in @xcite for the box - shaped perturbations by means of the max - min principle , see , e.g. , ( * ? ? ?
* thm 10.2.2 ) . however , the attendant asymptotic formula @xmath57 can not be supported by these results because an application in @xcite of a change of variables which transforms @xmath0 into @xmath58 requires certain smoothness properties of the boundary @xmath59 which are naturally absent , fig .
[ f1 ] , b. in section [ sect8.3 ] we will explain how our approach helps to justify formula ( [ 6 ] ) .
local perturbations of quantum waveguides in @xmath60 , @xmath61 , are intensively investigated during last two decades and many important results on the existence and asymptotic behavior of their discrete spectrum have been published .
we mention a few of them , namely @xcite for the slightly curved and twisted cylindrical waveguides , @xcite for cranked waveguides and @xcite for the laplacian perturbed by a small second - order differential operator with compactly supported coefficients .
we also refer to @xcite for non - local perturbations , fig .
[ f3 ] , a and to @xcite for alternation of the dirichlet and neumann boundary conditions .
[ ptb ] caduna2.epsc:/users/desktop/documents/google drive / articoli / articoli - in - preparazione / cardonedurantenazarov1/graphics / caduna2__2.pdf in the literature one finds much less results on eigenvalues embedded into the continuous spectrum , cf .
the review papers @xcite .
first of all we describe an elegant method @xcite which is based on imposing an artificial dirichlet condition and had become of rather wide use in proving the existence of embedded eigenvalues but only in symmetric waveguides .
[ ptb ] caduna3.epsc:/users/desktop/documents/google drive / articoli / articoli - in - preparazione / cardonedurantenazarov1/graphics / caduna3__3.pdf let us consider an auxiliary mixed boundary value problem and supply the helmholtz equation ( [ 1 ] ) with the neumann condition on the lower lagged wall and the dirichlet condition on the upper straight wall , see fig .
[ f1 ] , b , @xmath62 problem ( [ 1 ] ) , ( [ 7 ] ) , has the continuous spectrum @xmath63 and in section [ sect8.1 ] we will show the existence of only one eigenvalue@xmath64 in the discrete spectrum @xmath65 .
following @xcite we extend the corresponding eigenfunction @xmath66 as an odd function in @xmath67 from @xmath0 onto the bigger waveguide @xmath68 drawn in fig .
[ f3 ] , b and obtained as the union of the strip @xmath69 and the box @xmath70 .
owing to the dirichlet condition in ( [ 7 ] ) at the midline of @xmath68 , this extension @xmath71 is a smooth function everywhere in @xmath72 except at corner points and inherits from @xmath66 an exponential decay as @xmath73 .
clearly@xmath74 and , thus , @xmath75 is an eigenfunction of the neumann problem ( [ 888 ] ) while the corresponding eigenvalue ( [ 8 ] ) belongs to the continuous spectrum @xmath76 of this problem .
we emphasize that the method @xcite requires the mirror symmetry of the waveguide and can not be applied to the asymmetric waveguide @xmath77 in fig .
[ f1 ] , b. the detected embedded eigenvalue @xmath37 of the neumann problem ( [ 888 ] ) is stable with respect to small symmetric perturbations of the waveguide walls but any violation of the symmetry may lead it out from the spectrum and turn it into a point of complex resonance , cf .
@xcite and , e.g. , @xcite .
the intrinsic instability of embedded eigenvalues requests for special techniques to detect them as well as to construct their asymptotics . in the present paper
we use a criterion for the existence of trapped modes ( see @xcite and theorem [ theorema ] below ) and a concept of enforced stability of eigenvalues in the continuous spectrum , cf .
@xcite . in view of the mirror symmetry about the @xmath78-axis ,
notice the difference with the above mentioned assumption in @xcite , we truncate the waveguide @xmath0 and consider the neumann problem@xmath79 is its right half ( overshaded in fig .
[ f1 ] , b)@xmath80 clearly , the even in @xmath81 extension of an eigenfunction @xmath82 of problem ( [ 9 ] ) , ( [ 10 ] ) becomes an eigenfunction of the original problem ( [ 1 ] ) , ( [ 2 ] ) .
searching for an eigenvalue@xmath83 we will show in section [ sect7 ] that , first , problem ( [ 9 ] ) , ( [ 10 ] ) can not get more than one eigenvalue in @xmath53 and , second , the mixed boundary value problem in ( [ 11 ] ) with the dirichlet condition at the truncation segment @xmath84 instead of the neumann condition as in ( [ 10 ] ) , does not have eigenvalues ( [ 12 ] ) .
these mean that an eigenfunction of problem ( [ 1 ] ) , ( [ 2 ] ) associated with the eigenvalue ( [ 12 ] ) is always even in the variable @xmath81 .
in this way , we will be able to describe the part @xmath85 of the point spectrum in the entire waveguide @xmath0 . in what follows
we skip the subscript @xmath34 .
hence , we regard ( [ 3 ] ) as an integral identity serving for problem ( [ 9 ] ) , ( [ 10 ] ) in @xmath86 and denote @xmath87 the corresponding self - adjoint operator in @xmath88 , cf . section [ sect1.1 ] . in the absence of the mirror symmetry about a midline of a waveguide
the modern literature gives much less results on the existence of embedded eigenvalues .
a distinguishing feature of an eigenvalue in the continuous spectrum is its intrinsic instability with respect to a variation of the waveguide shape while all eigenvalues in the continuous spectrum stay stable . in this way to detect an eigenvalue in the neumann waveguide @xmath77 ,
see ( [ 1 ] ) , ( [ 2 ] ) , a fine tuning of the parameter - dependent shape is needed , namely the length @xmath89 of the perturbation box ( [ 0 ] ) must be chosen specifically in dependence of its height @xmath90 . a method to construct particular waveguide shapes
which support embedded eigenvalues , was developed in @xcite on the basis of a sufficient condition @xcite for the existence of exponentially decaying solutions trapped modes in waveguides to elliptic problems in domains with cylindrical outlets to infinity . as a result , several examples of eigenvalues in the continuous spectrum were constructed without requiring a geometrical symmetry of waveguides which are obtained from the straight unit strip by either singular @xcite , or regular @xcite perturbations of the boundary , see fig .
[ f5 ] , a and b , respectively ; for a non - local smooth perturbation , see @xcite . to this end , a notion of the augmented scattering matrix @xcite was used together with certain traditional asymptotic procedures in domains with small holes and cavities , cf .
* ch.4 , 5 and 2 ) , or in domains with smoothly varied boundaries , cf .
xii , 6.5 ) .
[ ptb ] caduna5.epsc:/users/desktop/documents/google drive / articoli / articoli - in - preparazione / cardonedurantenazarov1/graphics / caduna5__4.pdf it is very important in the above - mentioned approach to detect embedded eigenvalues that the asymptotic procedures in use are completed in such a way they provide both , a formal derivation of asymptotic expansions and an operator reformulation of the diffraction problem in the framework of the perturbation theory of linear operators , see , e.g. , @xcite , which helps to conclude a smooth ( actually analytic ) dependence of scattering matrices on geometrical parameters describing the waveguide shape .
moreover , this permits to reformulate the sufficient condition @xcite for the existence of trapped modes as a nonlinear abstract equation and to fulfil the condition by means of the contraction principle , cf .
sections [ sect2.3 ] and [ sect4.1 ] , [ sect4.2 ] below . the box - shaped perturbation ( [ 0 ] ) of the strip ( [ 00 ] ) can be regarded as a combination of regular and singular perturbations , respectively outside and inside neighborhoods of the corner points but unfortunately the authors do not know a tool to reduce the problem ( [ 1 ] ) , ( [ 2 ] ) ( or ( [ 3 ] ) in the variational form ) to an abstract equation in a fixed ( independent of @xmath90 ) banach space and to confirm necessary properties of the scattering matrices in the waveguide @xmath91 .
thus , in order to support each step of our procedure to detect an embedded eigenvalue and to establish its uniqueness , we have to obtain a certain new result for the problems ( [ 1 ] ) , ( [ 2 ] ) and/or ( [ 9 ] ) , ( [ 10 ] ) which can not be deduced from the general perturbation theory .
although several approaches and tricks proposed in our paper work also for other shapes like in fig .
[ f6 ] , we focus the analysis on the particular shape in fig . [ f1 ] , b. we proceed in section [ sect2 ] with introducing different waves in @xmath92 : oscillatory and exponential for @xmath93 and linear in @xmath81 at the threshold @xmath94 . then on the basis of the mandelstam energy principle , cf .
@xcite we perform the classification @xmath95 for the introduced waves and impose two types , physical and artificial , of radiation conditions at infinity .
the corresponding diffraction problems give rise to two scattering matrices @xmath96 and @xmath97 .
due to the restriction of the boundary value problem ( [ 1 ] ) , ( [ 2 ] ) onto the semi - infinite waveguide ( [ 11 ] ) the matrix @xmath96 reduces to classical scalar reflexion coefficient but the augmented scattering matrix @xmath97 is of size @xmath98 because the artificial radiation conditions involve the exponential waves in addition to the oscillatory waves .
the above - mentioned criterion for the existence of embedded eigenvalues is formulated in terms of the matrix @xmath97 , see theorem [ theorema ] and note that its proof is completed by theorem [ theoremma ] about solutions of the problem ( [ 9 ] ) , ( [ 10 ] ) with a fast exponential decay . in section [ sect3 ] we construct formal asymptotic expansions of the augmented scattering matrix which are justified in section [ sect6.4 ] . in order to detect an embedded eigenvalue in section [ sect4 ] we need the main asymptotic and first correction terms .
the two - fold nature of the box - shaped perturbation manifests itself in different anstze for the diagonal entries @xmath99 and @xmath100 of the matrix @xmath97 . in the first case the asymptotic procedure looks like as for a regular perturbation of the boundary , that is the boundary layer phenomenon does not influence the main asymptotic term @xmath99 in the expansion@xmath101 in the second case the correction term @xmath102 in the expansion @xmath103 results from the boundary layer phenomenon while the regular expansion affects higher - order terms only .
it should be emphasized that the augmented scattering matrix is unitary and symmetric and the main asymptotic term in the expansion@xmath104 of the anti - diagonal entries can be computed by means of both the approaches . in section [ sect4 ]
we first reduce the criterion @xmath105 from theorem [ theorema ] to an abstract equation and second solve it with the help of the contraction principle .
finally we formulate theorems [ theoremex ] and [ theoremun ] on the existence and uniqueness of the embedded eigenvalue .
these assertions are proved in the next three sections . in section [ sect5 ] formulations of the problem ( [ 1 ] ) , ( [ 2 ] ) in the kondratiev spaces ( therem [ theoremka ] ) and weighted spaces with detached asymptotics ( theorem [ theoremkmm ] ) are presented as well as the operator formulation of the radiation condition at infinity . at the same time , key results for the particular box - shaped perturbation ( [ 0 ] ) , are displayed in sections [ sect5.3 ] and [ sect5.5 ] where we verify the absence of trapped modes with a fast decay and clarify the dependence of majorants in a priori estimates for solutions on the small and spectral parameters @xmath106 and @xmath107 . in section [ sect6 ]
we evaluate remainders in the asymptotic formulas ( [ s11])-([s01 ] ) for the augmented scattering matrix theorem [ theoremasy ] while the boundary layer phenomenon brings additional powers of @xmath108 into bounds of estimates .
other necessary properties of the matrix are clarified in section [ sect7 ] where the uniqueness of the embedded eigenvalue is verified , too .
again kinks of the perturbation profile seriously complicate the analysis .
conclusive remarks are collected in section [ sect8 ] where , in particular , we discuss essential simplifications of the analysis within the discrete spectrum and a certain hardship for detecting eigenvalues near higher thresholds of the continuous spectrum .
for the spectral problem ( [ 12 ] ) , the limit ( @xmath109 ) problem ( [ 1 ] ) , ( [ 2 ] ) in the straight strip @xmath110 has two oscillating waves@xmath111 by the sommerfeld principle , see , e.g. , @xcite , the waves @xmath112 and @xmath113 are outgoing and incoming , respective , in the one - sided waveguide ( [ 11 ] ) according to their wavenumbers @xmath114 and @xmath115 in this way , the inhomogeneous problem ( [ 9 ] ) , ( [ 10])@xmath116 with , for example , compactly supported right - hand side @xmath117 is supplied with the radiation condition@xmath118 in section [ sect5 ] we will give an operator formulation of problem ( [ 14 ] ) , ( [ 15 ] ) . at the threshold@xmath119 in addition to the oscillating
waves@xmath120 there appear the staying and growing waves @xmath121 which can not be classified by the sommerfield principle because of zero wavenumber .
however , as was observed in @xcite , that , first , waves ( [ 13 ] ) verify the relations@xmath122 and , second , the linear combinations@xmath123 together with waves ( [ 17 ] ) fulfil formulas ( [ 19 ] ) at @xmath109 as well .
here , @xmath124 is the kronecker symbol , @xmath125 in the first case and @xmath126 in the second case , and @xmath127 is a symplectic , that is sesquilinear and anti - hermitian form@xmath128 this form emerges from the green formula in the truncated waveguide @xmath129 @xmath130 and therefore does not depend on the length parameter @xmath131 for any of introduced waves and their linear combinations .
hence , we skip the subscript @xmath132 in ( [ 19 ] ) and ( [ 20 ] ) . for waves ( [ 19 ] )
, the sign of @xmath133 coincides with the sign of the wavenumber and therefore indicates the propagation direction .
analogously , we call the wave @xmath134 outgoing and the wave @xmath135 incoming in the waveguide @xmath136 so that the problem ( [ 14 ] ) with @xmath94 ought to be supplied with the following threshold radiation condition@xmath137 in section [ sect5 ] we will prove that this formulation of the problem at the threshold ( [ 16 ] ) provides an isomorphism in its operator setting . as was demonstrated in @xcite , the form @xmath138
is closely related to the umov - poyting vector @xcite so that both radiation conditions ( [ 15 ] ) and ( [ 23 ] ) arise from the mandelstam ( energy ) principle , see @xcite . in the case (
[ 12 ] ) the incoming wave in ( [ 13 ] ) generates the following solution of problem ( [ 9 ] ) , ( [ 10]):@xmath139 here , the remainder @xmath140 decays as @xmath141 and @xmath142 is the reflection coefficient which satisfies @xmath143 due to conservation of energy . in the same way , in the case (
[ 16 ] ) we can determine the solutions @xmath144 where @xmath145 , @xmath146 and the coefficients @xmath147 form the ( threshold ) scattering matrix @xmath148 of size @xmath149 according to the normalization and orthogonality conditions ( [ 19 ] ) for waves ( [ 17 ] ) , ( [ 20 ] ) and the relation @xmath150 , the matrix @xmath148 is unitary and symmetric ( cf .
@xcite ) that is @xmath151 where @xmath152 stands for transposition and @xmath153 is the adjoint matrix .
the reflection coefficient @xmath142 ought to be regarded as a scattering matrix of size @xmath154 in view of the only one couple of waves ( [ 13 ] ) which are able to drive energy along the waveguide ( [ 11 ] ) .
for example , dealing with the next couple of waves@xmath155 which are decaying @xmath156 and growing @xmath157 one readily finds that @xmath158 but@xmath159 as was observed in @xcite and mentioned above , formula ( [ 28 ] ) annihilates the projection on the @xmath160-axis of the umov - poyting vector and therefore waves ( [ 27 ] ) can not drive energy . in the papers @xcite ( see @xcite for general elliptic systems ) the linear combinations of exponential waves@xmath161 were introduced .
it is remarkable that , thanks to ( [ 28 ] ) and ( [ 29 ] ) , waves ( [ 13 ] ) and ( [ 30 ] ) enjoy conditions ( [ 19 ] ) with @xmath162 the latter allows us to determine the solutions@xmath163 to compose the coefficient matrix @xmath164 and to assure its unitary and symmetry property , cf .
( [ 26 ] ) .
moreover , since @xmath165 , this matrix is symmetric , see @xcite again .
in section [ sect5 ] we will give an operator formulation of the problem ( [ 14 ] ) , at @xmath93 with the radiation condition@xmath166 we recognize this condition as artificial because the right - hand side of ( [ 32 ] ) involves the exponentially growing wave @xmath167 , see ( [ 30 ] ) and ( [ 27 ] ) . a reason to consider problem ( [ 14 ] ) , ( [ 32 ] ) and the augmented scattering matrix @xmath97 is explained by the following assertion .
[ theorema ] problem ( [ 9]),([10 ] ) with the spectral parameter ( [ 12 ] ) has a trapped mode @xmath168 if and only if @xmath169 in other words , equation ( [ 33 ] ) provides a criterion for the existence of a trapped mode in the one - sided waveguide ( [ 11 ] ) .
a proof of theorem [ theorema ] can be found , e.g. , in @xcite and ( * ? ? ?
* thm 2 ) but , since the criterion ( [ 33 ] ) plays the central rule in our analysis , we here give the condensed proof .
the unitary property of @xmath97 demonstrates that @xmath170 thus the solution ( [ 31 ] ) with @xmath145 becomes a trapped mode because formulas ( [ 30 ] ) and ( [ 27 ] ) assure that@xmath171 hence , ( [ 33 ] ) is a sufficient condition . to verify the necessity ,
we first assume that the decomposition@xmath172 of a trapped mode @xmath173 has a coefficient @xmath174 then @xmath175 becomes a linear combination of solutions ( [ 31 ] ) , namely , according to ( [ 30 ] ) , we have@xmath176 owing to the exponential decay of @xmath175 , coefficients of the oscillating waves @xmath112 must vanish so that @xmath177 @xmath178 moreover , coefficients of the exponential waves @xmath179 and @xmath180 , respectively , are @xmath181 and @xmath182 recalling our assumption @xmath183 we see that @xmath184 and , therefore , ( [ 33 ] ) holds true . if @xmath185 in ( [ 35 ] ) , the trapped mode @xmath186 gains very fast decay rate @xmath187 in section [ sect5.3 ] we will show with a new argument that such trapped modes do not exist for a small @xmath188 [ remarkso ] .
the relationship between the augmented scattering matrix and the reflection coefficient in ( [ 24 ] ) looks as follows:@xmath189 see , e.g. , ( * ? ? ?
* thm 3 ) .
note that , in view of ( [ 34 ] ) , the last term in ( [ 36 ] ) becomes null in the case @xmath105 when @xmath190
in this section we derive asymptotic expansions by means of a formal asymptotic analysis and postpone their justification to section [ sect6 ] .
perturbation of the straight waveguide drawn in fig .
[ f1],b and in fig .
[ f3 ] , b , ought to be regarded as a combination of regular and singular perturbations , see , e.g. , ( * ? ? ?
* ch.xii , 6.5 ) and ( * ? ? ?
* ch . 2 and 4 ) , respectively . for a regular perturbation of the boundary ,
an appropriate change of variables , which differs from the identity in magnitude @xmath191 only , is usually applied in order to convert the perturbed domain into the reference one . in this way ,
differential operators in the problem gain small perturbations but asymptotics can be constructed with the help of standard iterative procedures like decompositions of a perturbed operator in the neumann series .
singular perturbations of boundaries need much more delicate analysis because they require for a description of asymptotics in the stretched coordinates which , for the domain @xmath192 , see ( [ 11 ] ) , take the form @xmath193 notice that the change @xmath194 and setting @xmath109 transform @xmath136 into the upper half - plane @xmath195 with a semi - infinite step , fig .
[ f4 ] , a , @xmath196 [ ptb ] caduna4.epsc:/users/desktop/documents/google drive / articoli / articoli - in - preparazione / cardonedurantenazarov1/graphics / caduna4__5.pdf as a result , the singular perturbation of the waveguide wall gives rise to the boundary layer phenomenon described by solutions to the following problem:@xmath197 the laplace equation is caused by the relation @xmath198 which singles out the laplacian as the main asymptotic part of the helmholtz operator .
the neumann problem ( [ 40 ] ) with a compactly supported datum @xmath199 admits a solution @xmath200 as @xmath201 provided @xmath202 otherwise , a solution grows at infinity like @xmath203 and loses the intrinsic decay property of a boundary layer so that traditional asymptotic procedures become much more sophisticated , see ( * ? ? ?
* ch . 2 , 4 ) and @xcite .
however , we will see that in our particular problem the boundary perturbation does not affect the main asymptotic term and the first correction term does not include a boundary layer .
we search for an eigenvalue of the problem ( [ 9 ] ) , ( [ 10 ] ) in the form@xmath204 where the correction coefficient @xmath205 is to be found in section [ sect4 ] .
recalling the normalization factors in ( [ 27 ] ) and ( [ 13])@xmath206 we guess at the following asymptotic anstze for entries of the augmented scattering matrix@xmath207 but aim to calculate the terms @xmath208 and @xmath209 only .
we further estimate the remainders in section [ sect6 ] . using definitions of waves ( [ 13 ] ) and ( [ 27 ] ) , ( [ 30 ] ) , we take relations ( [ 43 ] ) and ( [ 44 ] ) into account and rewrite the decomposition ( [ 31 ] ) of the special solution @xmath210 as follows:@xmath211 here and everywhere in this section , ellipses stand for lower - order terms inessential in our formal asymptotic analysis . in ( [ 45 ] ) , the taylor formula@xmath212 was applied so that expansion ( [ 46 ] ) becomes meaningful under the restriction @xmath213 with a fixed @xmath132 , i.e. for @xmath214 in view of the above observation we employ the method of matched asymptotic expansions , cf .
@xcite , in the interpretation @xcite .
namely , we regard ( [ 45 ] ) as an outer expansion and introduce the inner expansion@xmath215 at the same time , coefficients of @xmath216 and @xmath217 on the right - hand side of ( [ 45 ] ) exhibit a behavior at infinity of the terms @xmath218 and @xmath219 in ( [ 47 ] ) because the upper bound @xmath132 for @xmath81 can be chosen arbitrary large .
thus , they must satisfy@xmath220 the formal passage to @xmath109 transforms the waveguide ( [ 11 ] ) into the semi - infinite strip @xmath221 while due to ( [ 42 ] ) the neumann problem ( [ 9 ] ) , ( [ 10 ] ) converts into@xmath222 this limit problem has two bounded solutions@xmath223 comparing with ( [ z0 ] ) , we set@xmath224 since @xmath225 the function @xmath219 also satisfies the homogeneous equation ( [ 481 ] ) but the neumann condition becomes inhomogeneous because of the boundary perturbation . for @xmath145
, we have @xmath226 to determine the datum @xmath227 we observe that the function ( [ 50 ] ) satisfies the neumann condition ( [ 10 ] ) everywhere on @xmath228 , except at the lower side @xmath229 of the box - shaped perturbation @xmath230 $ ] in ( [ 11 ] ) .
furthermore , we obtain @xmath231 and , hence,@xmath232{l}0,\text { \ \ } x\in\partial\pi_{+}^{0}\setminus\overline{\upsilon^{0}},\\ g_{p}^{\prime},\text { \ \ } x\in\upsilon^{0}. \end{array } \right . \label{52}\ ] ] although the neumann datum ( [ 52 ] ) is not smooth and has a jump at the point @xmath233 the problem ( [ 51 ] ) with @xmath145 has a solution in @xmath234 such that @xmath235 this fact is a direct consequence of the elliptic theory in domains with cylindrical outlets to infinity ( see the key works @xcite and , e.g. , the monographs @xcite , but also may be derived by the fourier method after splitting @xmath236 into the rectangle @xmath237 and the semi - strip @xmath238 a simple explanation how to apply the above - mentioned theory can be found in the introductory chapter 2 of @xcite , the review @xcite and section [ sect5 ] of this paper .
the solution ( [ 5222 ] ) is defined up to the term @xmath239 and , therefore , the coefficient @xmath240 can be taken arbitrary .
other coefficients in ( [ 5222 ] ) are computed by application of the green formula in the long ( @xmath132 is big ) rectangle @xmath241 indeed , we send @xmath132 to @xmath242 and obtain@xmath243 in the same way we deal with the functions ( [ 491 ] ) and ( [ 5222 ] ) that results in the equality@xmath244 comparing ( [ z1 ] ) and ( [ 5222 ] ) , we arrive at the relations @xmath245 which together with our calculations ( [ 55 ] ) and ( [ 56 ] ) give us the following formulas:@xmath246 we emphasize that @xmath247 the necessary computations are completed .
it should be mentioned that , to determine the correction terms @xmath248 and @xmath249 in the anstze ( [ 44 ] ) , one has to make another step in our asymptotic procedure , see the next section , but they are of no further use .
let us construct asymptotics of the entries @xmath250 in the augmented scattering matrix .
these asymptotics are not of very importance in our analysis of eigenvalues but are much more representational than the analogous formulas ( [ 44 ] ) because the ansatz for @xmath251 indeed involves the boundary layer concentrated near the ledge of the box - shape perturbation in ( [ 11 ] ) . using ( [ 43 ] ) , ( [ 60 ] ) and ( [ 46 ] ) , we rewrite the decomposition ( [ 31 ] ) of @xmath252 as follows:@xmath253 then we accept in a finite part of @xmath136 the ansatz@xmath254 referring to ( [ 61 ] ) , we fix the behavior of its terms at infinity@xmath255 comparing ( [ z00 ] ) with ( [ 491 ] ) , ( [ 492 ] ) , we conclude that @xmath256 and , hence,@xmath257 let us describe the correction terms in ( [ 60 ] ) . the function ( [ 63 ] ) satisfies the equation ( [ 9 ] ) with @xmath94 and leaves the discrepancies @xmath258 in the boundary condition ( [ 10 ] ) on the big @xmath259 and small @xmath260 sides of the rectangle @xmath261 , respectively .
the discrepancy ( [ z22 ] ) is similar to ( [ 5111 ] ) and appears as the datum ( [ 52 ] ) in the problem ( [ 51 ] ) with @xmath262 to compensate for ( [ z02 ] ) , we need the boundary layer @xmath263 where @xmath264 are stretched coordinates ( [ 38 ] ) and @xmath265 is a solution of the neumann problem ( [ 40 ] ) in the unbounded domain ( [ 39 ] ) with the right - hand side@xmath266{l}0,\text { \ \ } \xi_{1}\neq0,\\ 1,\text { \ \ } \xi_{1}=0,\text { \ \ } \xi_{2}\in\left ( -1,0\right ) , \end{array } \right .
\ \text{for } \xi\in\partial\xi.\ ] ] this solution , of course , can be constructed by an appropriate conformal mapping but we only need its presentation at infinity@xmath267 the constant @xmath268 is arbitrary but the coefficient @xmath269 can be computed by the green formula in the truncated domain @xmath270 with @xmath271@xmath272 here , @xmath273 is the polar coordinates system .
we fix @xmath274 in ( [ z05 ] ) and observe that @xmath275 in the polar coordinate system @xmath276 centered at the point @xmath277 see fig .
[ f4 ] , b. applying the method of matched asymptotic expansions in the traditional manner , cf .
@xcite , ( * ? ? ?
2 ) , as well as ( * ? ? ?
5 ) , @xcite for ledge - shaped perturbation of domains , we consider ( [ 62 ] ) as an outer expansion in a finite part of the waveguide while @xmath278 becomes the main term of the inner expansion in the vicinity of the ledge of the box - shaped perturbation in ( [ 11 ] ) . in view of ( [ z03 ] ) and ( [ z06 ] ) ,
the standard matching procedure proposes @xmath279 as a singular solution of the homogeneous problem ( [ 481 ] ) , ( [ 482 ] ) with the following asymptotic condition at the point @xmath280@xmath281 arguing in the same way as for the function @xmath282 we conclude that the problem ( [ 51 ] ) , ( [ z07 ] ) has a solution @xmath279 which admits the representation ( [ 5222 ] ) with @xmath283 under the restriction @xmath284 needed due to the logarithmic singularity in ( [ z07 ] ) .
the coefficient @xmath285 is arbitrary but @xmath286 and @xmath287 can be computed again by means of the green formula in the domain @xmath288 and the limit passage @xmath289 @xmath290 cf .
( [ 54 ] ) , ( [ 55 ] ) and ( [ z08 ] ) .
dealing with @xmath291 and @xmath292 we take into account the equality @xmath293 and obtain that@xmath294 inserting @xmath295 and @xmath279 into the green formula in @xmath296 , we take into account the inhomogeneous boundary condition on @xmath297 and derive that @xmath298{c}0={\displaystyle\int_{0}^{1 } } \left .
\cos\left ( \pi x_{2}\right ) \partial_{1}z_{0}^{\prime}\left ( x\right ) \right\vert _ { x_{1}=r}dx_{2}-{\displaystyle\int_{0}^{l-\delta } } \left .
\cos\left ( \pi x_{2}\right ) \partial_{2}z_{0}^{\prime}\left ( x\right ) \right\vert _ { x_{2}=0}dx_{1}-\\ -\delta{\displaystyle\int_{0}^{\pi } } \left .
\left ( \cos\left ( \pi x_{2}\right ) \partial_{r}z_{0}^{\prime } \left ( x\right ) -z_{0}^{\prime}\left ( x\right ) \partial_{r}\cos\left ( \pi x_{2}\right ) \right ) \right\vert _ { r=\delta}d\varphi=\\ = \dfrac{1}{2}c_{0}^{1}\left ( 4\mu\right ) ^{-1/4}\pi^{2}ls_{10}^{0}\left ( 1-i\right ) + \left ( 2\pi\right ) ^{1/2}\sin\left ( \pi l\right ) + o\left ( 1\right ) . \end{array } \label{zc1}\ ] ] we now compare coefficients in the expansions ( [ 5222 ] ) , @xmath283 , and ( [ z01 ] ) . according to the calculations ( [ zc ] ) and ( [ zc1 ] ) we derive the formulas@xmath299 calculation of coefficients in the anstze ( [ 44 ] ) and ( [ 60 ] ) is completed .
it is worth to underline that the expression ( [ anti ] ) for the main asymptotic term of @xmath300 can be derived from the cumbersome relations ( [ 57 ] ) and ( [ 58 ] ) as well .
we opt for the form@xmath301 of the spectral and length parameters which support a trapped mode . here , @xmath302 is fixed but small @xmath303 @xmath304 are to be determined . if @xmath305 and @xmath306 , the equalities @xmath307 and @xmath308 hold due to ( [ 57 ] ) and ( [ anti ] )
. we purpose to choose the small increments @xmath309 and @xmath304 in ( [ t1 ] ) such that the criterion ( [ 33 ] ) for the existence of a trapped mode is satisfied .
since @xmath99 is complex , the criterion furnishes two equations for two real parameters @xmath309 and @xmath310 it is convenient to consider the other equations @xmath311 which , for small @xmath90 and @xmath303 @xmath312 are equivalent to @xmath313 indeed , from formulas ( [ 58 ] ) , ( [ anti ] ) and ( [ 64 ] ) together with estimates ( [ est ] ) it follows that@xmath314 since @xmath97 is unitary and symmetric , see section [ sect2.2 ] , the second assumption in ( [ t2 ] ) means that @xmath315 with some @xmath316 and , furthermore,@xmath317 hence , @xmath318 when @xmath309 , @xmath304 and @xmath2 are small so that @xmath319 due to ( [ t3]),([t2 ] ) .
we have proved that ( [ t2])@xmath320([33 ] ) but the inverse implication ( [ 33])@xmath320([t2 ] ) is obvious . by virtue of (
[ 58 ] ) , ( [ s11 ] ) , and ( [ t1 ] ) , the first equation ( [ t2 ] ) turns into@xmath321 the formulas ( [ anti ] ) , ( [ s01 ] ) , and a simple algebraic calculation convert the second equation ( [ t2 ] ) into@xmath322 and thus@xmath323 we search for a solution @xmath324 of the transcendental equations ( [ t4 ] ) , ( [ t5 ] ) in the closed disk@xmath325 and rewrite them in the condensed form@xmath326 where @xmath327 is a nonlinear operator involving asymptotic remainders from formulas ( [ s11 ] ) and ( [ s01 ] ) for the augmented scattering matrix @xmath328 . the estimates ( [ est ] ) and proposition [ propositionanal ] below demonstrate that the operator is smooth in @xmath329 with @xmath330 , @xmath331 and , furthermore,@xmath332 hence , for any @xmath333 , there exists @xmath334 such that @xmath327 with @xmath335 is a contraction operator in the disk ( [ t6 ] ) . by the banach contraction principle , the abstract equation ( [ t7 ] ) has a unique solution @xmath336 and the estimate @xmath337 is valid .
this solution depends on @xmath90 and determines the spectral and length parameters ( [ t1 ] ) supporting a trapped mode in the perturbed waveguide ( [ 11 ] ) according to the criterion ( [ 33 ] ) from theorem [ theorema ] reformulated as ( [ t2 ] ) .
based on the performed formal calculations , we will prove in the next three sections the following existence and uniqueness theorems .
[ theoremex]let @xmath302 .
there exist @xmath338 , @xmath339 and @xmath340 , @xmath341 such that , for any @xmath342 , the estimate@xmath343 is valid and the problem ( [ 1 ] ) , ( [ 2 ] ) in the waveguide @xmath344 with the box - shaped perturbation ( [ 0 ] ) of length @xmath345 has an eigenvalue@xmath346 the eigenvalue ( [ t10 ] ) is unique in the interval @xmath347 .
[ theoremun]let @xmath302 and @xmath348 there exist @xmath349 such that , for any @xmath350 , the waveguide @xmath0 with the length parameter @xmath351 \label{t11}\ ] ] does not support a trapped mode in the case @xmath352 where @xmath353 is taken from theorem [ theoremex ] .
let @xmath354 be the kondratiev ( weighted sobolev ) space composed from functions @xmath48 in @xmath355 with the finite norm@xmath356 where @xmath357 is the exponential weight index .
if @xmath358 , functions in @xmath354 decay at infinity in the semi - infinite waveguide ( [ 11 ] ) but in the case @xmath359 a certain exponential growth is permitted while the rate of decay / growth is ruled by @xmath360 clearly , @xmath361 .
the space @xmath362 of smooth compactly supported functions is dense in @xmath363 for any @xmath360 by a solution of the problem ( [ 14 ] ) in @xmath364 , @xmath316 , we understand a function @xmath365 satisfying the integral identity@xmath366 where @xmath367 is an ( anti)linear continuous functional on @xmath368 and @xmath369 is an extension of the lebesgue scalar product up to a duality between an appropriate couple of weighted spaces . in view of ( [ k1 ] )
all terms in ( [ k2 ] ) are defined correctly so that the problem ( [ k2 ] ) is associated with the continuous mapping@xmath370 if @xmath371 that is @xmath372 , then the functional@xmath373 belongs to @xmath374 . clearly , @xmath375 is the adjoint operator for @xmath376 .
let us formulate some well - know results of the theory of elliptic problems in domains with cylindrical outlets to infinity ( see the key papers @xcite and , e.g. , the monographs @xcite ) .
this theory mainly deals with the classical ( differential ) formulation of boundary value problems , however as was observed in @xcite , passing to the weak formulation involving integral identities of type ( [ k2 ] ) does not meet any visible obstacle .
the only disputable point , namely the dependence of constants on the small parameter @xmath90 , we will be discussed in section [ sect5.5 ] . [ theoremka](see @xcite )
let @xmath377 $ ] .
\1 ) the operator @xmath378 is fredholm if and only if@xmath379 in the case @xmath380 with @xmath381 the range @xmath382 is not closed subspace in @xmath383 \2 ) let @xmath384 and let @xmath385 be a solution of problem ( [ k2 ] ) with the weight index @xmath386 and the right - hand side @xmath387 . then the asymptotic decomposition@xmath388 and the estimate@xmath389 are valid , where @xmath390 is the asymptotic remainder , @xmath391 and @xmath392 are coefficients depending on @xmath393 and @xmath48 , the waves @xmath394 are given by ( [ 13 ] ) and @xmath395 by ( [ 20 ] ) for @xmath94 but by ( [ 30 ] ) for @xmath396 .
the factor @xmath397 in ( [ k6 ] ) is independent of @xmath393 and @xmath48 but may depend on @xmath398 .$ ] as was mentioned @xmath399 and , hence , kernels and co - kernels of these operators are in the relationship@xmath400 in the next assertion we compare the indexes ind@xmath401@xmath402 notice that ind@xmath403ind@xmath404 according to ( [ k7 ] ) .
[ theoremkb ] ( see ( * ? ? ?
3.3.3 , 5.1.4 ( 4 ) ) ) if @xmath405 see ( [ k4 ] ) , then@xmath406 we emphasize that the last @xmath407 is nothing but the number of waves detached in ( [ k5 ] ) . from ( [ k7])-([k8 ] )
, it follows that @xmath408 in this section we prove that , for @xmath409 $ ] and @xmath410 , there holds the formula@xmath411 which , in particular , completes the proof of theorem [ theorema ] , cf .
our assumption @xmath412 for trapped mode ( [ 35 ] ) while @xmath413 leads to @xmath414 .
clearly , @xmath415 for any @xmath358 that is the limit problem ( [ 9 ] ) , ( [ 10 ] ) in the straight semi - strip @xmath416 can not have a trapped mode .
however , as was mentioned in section [ sect1.4 ] , formula ( [ m1 ] ) does not follow by a standard perturbation argument and , moreover , @xmath417 for some @xmath418 and @xmath419 .
[ theoremma]let @xmath384 be fixed .
there exists @xmath420 such that , for @xmath421 and @xmath409 $ ] , formula ( [ m1 ] ) is valid .
* let us assume that , for some @xmath422 $ ] and an infinitesimal positive sequence @xmath423 , the homogeneous problem ( [ 9 ] ) , ( [ 10 ] ) has a solution @xmath424 .
we denote by @xmath425 the restriction of @xmath426 onto the semi - strip @xmath221 . under the normalization condition@xmath427
we are going to perform the limit passage @xmath428 in the integral identity@xmath429 where @xmath430 is obtained from a test function @xmath431 by the even extension over the @xmath81-axis .
if we prove that @xmath432 @xmath425 converges to @xmath433 weakly in @xmath434 and , therefore , strongly in @xmath435 @xmath436 @xmath437 @xmath438 @xmath439 with any smooth function @xmath265 in the rectangle @xmath440 \times\left [ -1,1\right ] , $ ] then the limit passage in ( [ m2 ] ) and ( [ m0 ] ) gives @xmath441 by a density argument , the integral identity ( [ m3 ] ) is valid with any @xmath442 and , therefore , @xmath443 because the limit problem in @xmath236 can not get a non - trivial trapped mode .
let us confirm facts @xmath444 .
we write @xmath90 instead of @xmath445 first , we apply a local estimate , see , e.g. , @xcite , to the solution @xmath48 of the problem ( [ 9 ] ) , ( [ 10 ] ) with @xmath446 as a given right - hand side:@xmath447 here , @xmath448 and @xmath449 are rectangles such that @xmath450 and , therefore , the right - hand side of ( [ m5 ] ) is less than @xmath451 according to ( [ m0 ] ) .
second , we split @xmath48 as follows:@xmath452 where @xmath453 is a cut - off function , @xmath454 for @xmath455 and @xmath456 for @xmath457 the components in ( [ m55 ] ) satisfy the integral identities@xmath458 third , inserting @xmath459 into ( [ m6 ] ) and taking ( [ m0 ] ) , ( [ m5 ] ) into account yield@xmath460 the problem ( [ m7 ] ) needs a bit more advanced argument .
it is posed in the semi - strip @xmath461 independent of @xmath90 and , thus , the following a priory estimate in the kondratiev space , see @xcite and , e.g. , ( * ? ? ?
* thm 5.1.4 ( 1)),@xmath462 involves some constants @xmath463 independent of @xmath90 . in this way , formulas ( [ m8 ] ) and ( [ m9 ] ) , ( [ m5 ] ) , ( [ m1 ] ) assure that@xmath464 thus , the convergence in @xmath465 occurs along a subsequence which is still denoted by @xmath466 . the last step of our consideration uses integration in @xmath467 and @xmath468 of the newton - leibnitz formula@xmath469 where @xmath470 is a cut - off function , @xmath471 for @xmath472 and @xmath473 for @xmath474 . as a result , we obtain the estimate@xmath475 while referring to ( [ m10 ] ) again .
this provides ( ii ) as well as ( iii ) because @xmath476 theorem [ theoremma ] is proved .
let @xmath93 and @xmath478 , cf .
theorem [ theoremka ] .
the pre - image @xmath479 of the subspace @xmath480 in @xmath481 for the operator @xmath482 consists of functions in the form ( [ k5 ] ) . introducing the norm @xmath483 as the left - hand side of ( [ k6 ] )
makes @xmath484 a hilbert space but this hilbert structure is of no use in our paper .
the restriction @xmath485 of @xmath486 onto @xmath487 inherits all properties of @xmath488 in particular , @xmath489 is a fredholm operator with ind@xmath490ind@xmath491 see ( [ k9 ] ) .
thus , the restriction @xmath492 of @xmath493 onto the subspace@xmath494 of codimension @xmath495 becomes of index zero .
[ theoremkmm]let @xmath496 , @xmath497 and @xmath90 be the same as in theorem [ theoremma ] .
then the operator @xmath498 actualizes the isomorphism@xmath499 since the decomposition ( [ k5 ] ) of a function @xmath500 loses the incoming waves @xmath113 and @xmath501 due to the restriction in ( [ k11 ] ) , @xmath502 has to be interpreted as an operator of the problem ( [ k11 ] ) with the radiation condition ( [ 23 ] ) at @xmath503 and ( [ 32 ] ) at @xmath504 theorem [ theoremkmm ] says that such problem is uniquely solvable , while its solution in the form@xmath505 obeys the estimate@xmath506 if @xmath507 \label{kk3}\ ] ] with a fixed @xmath508 the coefficient in the estimate ( [ k6 ] ) can be chosen independent of @xmath509 $ ] with some @xmath510 .
this fact originates in the smooth dependence of the waves ( [ 13 ] ) and ( [ 27 ] ) , ( [ 30 ] ) on the parameter ( [ kk3 ] ) and the following consideration . by multiplying @xmath48 with
the same cut - off function @xmath511 as in ( [ m55 ] ) , we reduce the problem ( [ k2 ] ) onto the semi - strip @xmath461 , namely , inserting @xmath512 with any @xmath513 as a test function , we obtain for @xmath514 the integral identity@xmath515 moreover,@xmath516 cf . the right - hand side of ( [ k6 ] ) . by restriction ( [ kk3 ] )
, @xmath17 stays at a distance from the thresholds @xmath517 and @xmath518 so that we may choose the same weight index @xmath497 for all legalized @xmath17 .
hence , a general result in @xcite , see also @xcite , on the basis of a perturbation argument provides a common factor @xmath519 in the estimate ( [ k6 ] ) for ingredients of the asymptotic representation ( [ k5 ] ) of the solution @xmath520 to the problem ( [ kk4 ] ) in the @xmath90-independent domain @xmath521 since the weight @xmath522 is uniformly bounded in @xmath523 , the evident relation@xmath524 allows us to extend the above mentioned estimate over the whole waveguide @xmath136 .
similarly , in the case ( [ kk3 ] ) the factor @xmath525 in ( [ kk2 ] ) can be fixed independent of @xmath90 , too . the desired eigenvalue ( [ 42 ] ) is located in the vicinity of the threshold @xmath518 and the above consideration becomes unacceptable .
moreover , the normalization factor @xmath526 in ( [ 27 ] ) is big so that the independence property of @xmath527 and @xmath525 is surely lost .
thus , our immediate objective is to modify the estimates in order to make them homotype for all small @xmath2 .
we emphasize that a modification of the normalization factor does not suffice because the waves @xmath528 in ( [ 27 ] ) become equal at @xmath109 .
we follow a scheme in @xcite and define for @xmath529 the linear combinations of the exponential waves ( [ 27])@xmath530 cf .
( [ 30 ] ) .
a direct calculation demonstrates that the new waves ( [ w1 ] ) together with the old waves ( [ 13]),@xmath531 still satisfy the normalization and orthogonality conditions ( [ 19 ] ) but additionally are in the relationship@xmath532 in other words , the waves ( [ w1 ] ) and ( [ w2 ] ) smoothly become the waves ( [ 20 ] ) and ( [ 13 ] ) introduced in section [ sect2.1 ] at the threshold @xmath94 .
the first property of @xmath533 allows us to repeat considerations in sections [ sect5.4 ] , [ sect2.2 ] and compose the space @xmath534 of functions satisfying the new , so - called artificial , radiation condition @xmath535 cf .
( [ kk1 ] ) , to determine the solutions @xmath536 of the homogeneous problem ( [ k2 ] ) , @xmath537@xmath538 cf .
( [ 31 ] ) and to detect a unitary and symmetric artificial scattering matrix @xmath539 . at the same time
, the second property of @xmath540 assures that , for a fixed @xmath90 , the operator@xmath541 of the problem ( [ k2 ] ) , @xmath386 , with the radiation condition ( [ u5 ] ) depends continuously on the spectral parameter @xmath542 $ ] , @xmath543 , when the domain of @xmath544 is equipped with the norm @xmath545 of a weighted space with detached asymptotics , cf . the left - hand side of ( [ kk3 ] ) . recalling our reasoning in section [ sect5.3 ] and the beginning of this section ,
we conclude that the operator ( [ u7 ] ) is an isomorphism while its norm and the norm of the inverse are uniformly bounded in @xmath546 , \text { \ \ } \varepsilon \in\left [ 0,\varepsilon_{0}\right ] .
\label{u8}\ ] ] furthermore , by the fourier method , entries of the matrix @xmath547 can be expressed as weighted integrals of solutions ( [ u6 ] ) , this matrix is continuous in both arguments ( [ u8 ] ) and the limit matrix@xmath548 is nothing but augmented scattering matrix at the thresholds and its diagonal form is due to the explicit solutions ( [ 491 ] ) and ( [ 492 ] ) in the semi - strip @xmath549,@xmath550 we resume that above consideration and find out a unique solution @xmath551 of the problem ( [ k2 ] ) with @xmath386 , @xmath552 and the artificial radiation condition ( [ u5 ] ) .
moreover , the estimate@xmath553 is valid , where @xmath268 is independent of both parameters ( [ u8 ] ) .
we now search for a solution @xmath554 of the same integral identity but the radiation condition from section [ sect5.4 ] in the form@xmath555 the unknown coefficients @xmath556 should be fixed such that the decomposition ( [ kk1 ] ) is satisfied . to this end , we insert into the right - hand side of ( [ u13 ] ) formulas ( [ u5 ] ) , ( [ u6 ] ) and ( [ w1 ] ) , ( [ w2 ] ) , we compare the resultant coefficients of the waves ( [ 13 ] ) , ( [ 27 ] ) in ( [ u13 ] ) with those in ( [ kk1 ] ) and arrive at the following systems of linear algebraic equations for the unknowns @xmath557 , @xmath558 and @xmath559 , @xmath560@xmath561 solving the system ( [ u15 ] ) with the help of the cramer s rule , a simple calculation gives the determinant@xmath562 and the estimates@xmath563 because @xmath564 due to ( [ u9 ] ) and ( [ u8 ] ) .
in view of the first relation in ( [ u14 ] ) we obtain that@xmath565 collecting formulas ( [ u15 ] ) , ( [ u16 ] ) and ( [ u12 ] ) , ( [ u10 ] ) adjusts the inequality ( [ kk2 ] ) as well as theorem [ theoremkmm ] .
[ theoremkum ] let @xmath566 $ ] , @xmath567 $ ] and @xmath384 .
the solution ( [ kk1 ] ) of the problem ( [ k2 ] ) with @xmath386 and @xmath568 admits the estimate@xmath569 where @xmath570 does not depend on @xmath17 , @xmath90 and @xmath393 .
the reformulation ( [ t2 ] ) of the criterion ( [ 32 ] ) implicates the coefficients @xmath99 and @xmath571 in the decomposition ( [ 31 ] ) of the special solution @xmath210 of the problem ( [ 1 ] ) , ( [ 2 ] ) and this section is devoted to the justification of the formal asymptotic expansions ( [ 44 ] ) .
we emphasize that the similar expansions ( [ 60 ] ) of other entries in the augmented scattering matrix @xmath97 can be verified in the same way but actually we had used in section [ sect4.1 ] much simpler relation ( [ t3 ] ) only .
in section [ sect3 ] we applied the method of matched asymptotic expansions and our immediate objective becomes to compose a global approximation solution from the inner and outer expansions ( [ 47 ] ) and ( [ z1 ] ) . to this end
, we employ several smooth cut - off functions:@xmath572 where @xmath573 we set@xmath574 this construction needs explanations .
first , the expressions ( [ j3 ] ) and ( [ j4 ] ) of the outer and inner types are multiplied with the cut - off functions @xmath575 and @xmath576 whose support overlap so that the sum ( [ j5 ] ) of terms matched in section [ sect3.2 ] attend the global approximation ( [ j2 ] ) twice , i.e. in @xmath577 and @xmath578 , but we compensate for this duplication by subtracting @xmath579 .
moreover , the formula for commutators @xmath580 = \left [ \delta,\chi_{\infty}\right ] + \left [ \delta , x_{\varepsilon}\right ] $ ] demonstrates that @xmath581 \left ( \mathfrak{z}^{out}-\mathfrak{z}^{mat}\right ) + \left [ \delta , x_{\varepsilon}\right ] \left ( \mathfrak{z}^{in}-\mathfrak{z}^{mat}\right ) : = \nonumber\\ : = \mathcal{f}^{\varepsilon}=\chi_{\infty}\mathcal{f}^{out}+x_{\varepsilon } \mathcal{f}^{in}-\chi_{\infty}x_{\varepsilon}\mathcal{f}^{mat}+\mathcal{f}^{oma}+\mathcal{f}^{ima}.\nonumber\end{gathered}\ ] ] second , the function @xmath218 is properly defined by the formula ( [ 50 ] ) in the whole waveguide but @xmath219 needs an extension from @xmath416 onto @xmath136 denoted by @xmath582 in ( [ j4 ] ) . since the neumann datum ( [ 52 ] ) in the problem ( [ 51 ] ) , @xmath145 , has a jump at the point @xmath583 , the solution @xmath584 gets a singular behavior near this point . a simple calculation based on the kondratiev theory @xcite
( see also ( * ? ? ?
2 ) ) demonstrates that @xmath585 where @xmath586 are polar coordinates in fig .
[ f4 ] , b and @xmath587 is a smooth function in the closed rectangle @xmath588 of any fixed length @xmath132 .
we emphasize that the solution @xmath219 has no singularities at the corner points @xmath589 and @xmath590 cf .
2 ) , but third derivatives of @xmath587 are not bounded when @xmath591 .
the extension @xmath582 in ( [ j7 ] ) where @xmath587 is smoothly continued through the segment @xmath592 , \text{\ } x_{2}=0\right\ } $ ] .
finally , we mention that the correction term @xmath219 in ( [ 47 ] ) was determined in section [ sect3.2 ] up to the addendum @xmath593 but putting @xmath594 in the expansion ( [ 5222 ] ) defines uniquely the function @xmath219 as well as its value @xmath595 according to ( [ j7 ] ) .
notice that we also must take @xmath596 by virtue of ( [ 56 ] ) .
the extension @xmath582 of @xmath219 is smooth everywhere in a neighborhood of @xmath597 except at the point @xmath598 where it inherits a singularity from ( [ j7 ] ) .
using the partition of unity @xmath599 makes the last term in ( [ j4 ] ) smooth in @xmath600 but produces additional discrepancy in the helmholtz equation ( [ 1 ] ) .
first of all , we observe that @xmath601 in @xmath236 according to definition of waves in ( [ 13 ] ) and ( [ 30 ] ) . in view of the factor @xmath602 from ( [ j1 ] ) the first term on the right of ( [ j6 ] )
moreover , the taylor formulas ( [ 43 ] ) and ( [ 46 ] ) assure that@xmath603 on the rectangle @xmath604 \times\left [ 0,1\right ] $ ] where supports of coefficients in the commutator @xmath605 $ ] are located in .
hence@xmath606 let us consider the sum @xmath607 outside the finite domain @xmath608 it is equal to@xmath609 where formulas ( [ 42 ] ) and ( [ t1 ] ) are taken into account .
we now use the representations ( [ 50 ] ) and ( [ 5222 ] ) to conclude that@xmath610 inside @xmath608 we have@xmath611 ( \widehat{z}_{1}^{\prime}-z_{1}^{\prime}\left ( l,0\right ) ) .\ ] ] the inequality@xmath612 is evident .
because of the singularity @xmath613 in ( [ j7 ] ) , estimates of other two terms in ( [ j11 ] ) involve the weight function@xmath614 cf . the hardy inequality ( [ hardy ] ) . since @xmath615 in @xmath236 satisfies the helmholtz equation from ( [ 51 ] ) , we have@xmath616 observing that , according to the third line in ( [ j1 ] ) , coefficients in the commutator @xmath617 = 2\nabla\chi _ { \varepsilon}\cdot\nabla+\delta\chi_{\varepsilon}$ ] gets orders @xmath618 and @xmath619 respectively but vanish outside the set @xmath620 , we conclude that @xmath621 \left ( \widehat { z}_{1}^{\prime}\left ( x\right ) -z_{1}^{\prime}\left ( l,0\right ) \right ) & = 0,\text { \ \ } x\in\pi_{+}^{\varepsilon}\left ( 3l/2\right ) \setminus \cap^{\varepsilon},\label{j16}\\ \varepsilon^{1/2}\left\vert \left [ \delta,\chi_{\varepsilon}\left ( \tau\right ) \right ] \left ( \widehat{z}_{1}^{\prime}\left ( x\right ) -z_{1}^{\prime}\left ( l,0\right ) \right ) \right\vert & \leq c\varepsilon^{1/2}\left ( \varepsilon^{-1}\left\vert \ln r\right\vert + \varepsilon^{-2}r\left\vert \ln r\right\vert \right ) \leq\nonumber\\ & \leq c\varepsilon^{1/2}\left\vert x_{1}\right\vert \left ( \varepsilon + r\right ) ^{-2}\rho\left ( x\right ) , \text { } x\in\cap^{\varepsilon } .\nonumber\end{aligned}\ ] ] finally , we mention that the support of the term @xmath622 in ( [ j6 ] ) belongs to the rectangle @xmath623 \times\left [ 0,1\right ] , $ ] see the first line of ( [ j1 ] ) , where the remainder @xmath624 in ( [ 5222 ] ) gets the exponential small order @xmath625 , and hence@xmath626 \left ( \mathfrak{z}^{in}\left ( x\right ) -\mathfrak{z}^{mat}\left ( x\right ) \right ) \right\vert = \varepsilon^{1/2}\left\vert \left [ \delta , x_{\varepsilon } \right ] \widetilde{z}_{1}^{\prime}\left ( x\right ) \right\vert \leq c\varepsilon^{1/2}e^{-3\sqrt{\pi}/\varepsilon}. \label{j17}\ ] ] it remains to consider discrepancies in the neumann condition ( [ 2 ] ) . since the cut - off functions @xmath576 and @xmath575 can be taken dependent on the longitudinal coordinate @xmath81 only , the asymptotic solution ( [ j2 ] ) satisfies the homogeneous neumann condition everywhere on @xmath627 , except on the sides @xmath259 and @xmath628 of the rectangle ( [ 0 ] ) , cf .
sections [ sect3.2 ] and [ sect3.3 ] .
furthermore , @xmath218 does not depend on @xmath81 and @xmath219 is multiplied in ( [ j4 ] ) with the cut - off function @xmath629 in the radial variable @xmath630 thus , @xmath631 on the short side @xmath632 regarding the trace @xmath633 of @xmath634 on the long side
@xmath259 we take the formulas ( [ 51])-([52 ] ) into account and , similarly to ( [ j14 ] ) and ( [ j16 ] ) , obtain@xmath635 we emphasize that differentiation in @xmath78 eliminates @xmath636 in the first term of ( [ j7 ] ) .
first of all , we observe that @xmath637 as @xmath638 by virtue of the definition ( [ j1 ] ) of @xmath576 and @xmath602 .
thus , in view of ( [ 31 ] ) and ( [ j2 ] ) , ( [ j3 ] ) the difference @xmath639 loses the incoming waves @xmath640 and falls into the space @xmath641 .
moreover , the decomposition ( [ kk1 ] ) of @xmath642 contains the coefficients @xmath643 and @xmath644 defined in ( [ s11 ] ) and ( [ s01 ] ) . the integral identity ( [ k2 ] ) with @xmath386 serving for @xmath645 , involves the functional@xmath646 where @xmath647 is given in ( [ j6 ] ) and @xmath648 .
if we prove the inclusion @xmath649 , then the estimate ( [ u18 ] ) adjusted by the weighting factor @xmath650 demonstrates that @xmath651 we fix some test function @xmath652 .
the classical one - dimensional hardy inequality@xmath653 in a standard way , cf .
* ch.1 , 4 ) , leads to the relation@xmath654 where @xmath655 is the weight factor ( [ j14 ] ) .
moreover , introducing the new weight factor @xmath656 , we derive the weighted trace inequality@xmath657 here , we took into account that @xmath658 and used a cut - off function @xmath659 , @xmath660 for @xmath661 and @xmath471 for @xmath662 .
the inclusion @xmath663 is obvious because @xmath647 has a compact support . to estimate the norm @xmath664 we apply inequalities obtained in the previous section .
since @xmath665 , the estimates ( [ j17 ] ) gives@xmath666 by the formula ( [ j9 ] ) , we have@xmath667 recalling ( [ j10 ] ) and ( [ j12 ] ) , ( [ j14 ] ) , ( [ j16 ] ) yields@xmath668 finally , we derive from ( [ j18 ] ) and ( [ j22 ] ) the following estimate of the last scalar product in ( [ j20]):@xmath669 collecting the obtained inequalities we conclude that the functional in ( [ j20 ] ) meets the estimate@xmath670 we are in position to formulate the main technical result in the paper .
since the norm @xmath671 in the space with detached asymptotics contains the coefficients @xmath672 and @xmath673 in the representation ( [ kk1 ] ) of @xmath645 estimates of the asymptotic remainders in ( [ s01 ] ) and ( [ s11 ] ) follow directly from ( [ j23 ] ) and ( [ u18 ] ) , ( [ 42 ] ) . a similar outcome for @xmath100
can be obtained by repeating word by word calculations in the previous section based on the formal asymptotic ( [ 60 ] ) , ( [ 64 ] ) .
we only mention that the discrepancy ( [ z02 ] ) on the small side @xmath628 of the box @xmath674 can be considered as follows:@xmath675 where a weighted trace inequality of type ( [ j22 ] ) is applied .
[ theoremasy]remainders in the asymptotic forms ( [ s11])-([s01 ] ) enjoy the estimate @xmath676 we finally note that formulas ( [ t3 ] ) , ( [ t999 ] ) is the only issue in the paper which deals with @xmath100 but requires much less accurate information .
let @xmath90 be fixed small and positive .
we take @xmath679 and make the change of coordinates@xmath680 where @xmath681 is a smooth cut - off function , @xmath682 for @xmath683 and @xmath684 for @xmath685 . if @xmath677 is small
, this change is nonsingular .
moreover , it transforms @xmath0 into @xmath686 and turns the helmholtz operator @xmath687 into the second - order differential operator @xmath688 whose coefficients depend smoothly on @xmath689 and @xmath690 clearly , @xmath691 .
owing the fourier method , we can rewrite the element @xmath692 of the augmented scattering matrix as the integral@xmath693 over the rectangle @xmath694 where @xmath695 according to ( [ j51 ] ) . due to the general result in the perturbation theory of linear operators , see , e.g. , @xcite , the special solution @xmath696 @xmath697 rewritten in the coordinates @xmath698 depend smoothly on @xmath699the coefficient @xmath700 in ( [ j53 ] ) is also a smooth function whose exact form is of no need .
thus , the element ( [ j53 ] ) inherits this smooth dependence while the remainder @xmath701 in the representation ( [ s11 ] ) gets the same property according to the formula ( [ 58 ] ) for @xmath702 written in the variables ( [ t1 ] ) .
similar operations apply to @xmath703 and @xmath704 finally , recalling our examination in section [ sect5.5 ] and theorem [ theoremasy ] , we formulate the result .
[ propositionanal ] the remainders in the asymptotic formulas ( [ s11 ] ) , ( [ 58 ] ) and ( [ s01 ] ) , ( [ anti ] ) satisfy the inequality@xmath705
let us adapt a trick from @xcite for the box - shaped perturbation ( [ 0 ] ) and conclude with the uniqueness mentioned in theorem [ theoremex ] .
assume that there exists an infinitesimal sequence @xmath423 such that the problem ( [ 9 ] ) , ( [ 10 ] ) in the semi - infinite waveguide @xmath707 has two eigenvalues @xmath708 and @xmath709 while @xmath710,\text { \ \ } \widehat{\lambda}_{j}^{\varepsilon_{k}}\rightarrow0,\text { \ \ } j=1,2 .
\label{u0}\ ] ] in what follows we write @xmath90 instead of @xmath445 the corresponding eigenfunctions @xmath711 and @xmath712 are subject to the normalization and orthogonality conditions@xmath713 cf .
( [ m0 ] ) . repeating with evident changes our arguments in section [ sect5.3 ]
we observe that the restrictions @xmath714 of @xmath715 onto @xmath236 converge to @xmath716 weakly in @xmath717 and strongly in @xmath718 .
furthermore , the limits satisfy the formula ( [ m4 ] ) and the following integral identity , see ( [ m5]):@xmath719 any solution in @xmath717 with @xmath410 of the homogeneous neumann problem ( [ u1 ] ) in the semi - strip @xmath720 is a linear combination of two bounded solutions ( [ 491 ] ) and ( [ 492]):@xmath721 let us prove that @xmath722 in ( [ u2 ] ) . since the trapped mode @xmath715 has an exponential decay at infinity , the green formula in @xmath723 with it and the bounded function @xmath724 assures that @xmath725 the local estimate ( [ m5 ] ) in @xmath726 @xmath727 and formulas in ( [ u0 ] ) , ( [ u00 ] ) allow us to compute the limit of the left - hand side of ( [ u3 ] ) and obtain that@xmath728 inserting ( [ u2 ] ) into ( [ u4 ] ) , we see that @xmath729 indeed .
[ remarkuu1 ] .
if @xmath730 and @xmath731 in ( [ u0 ] ) , one may use the green formula in @xmath723 with four bounded functions @xmath724 and @xmath732 . in this way one derives the equalities @xmath733 ( see @xcite for details ) and concludes that a small neighborhood of the threshold @xmath734 can contain only eigenvalues indicated in ( [ u0]).@xmath735the same reasoning show that the problem ( [ 9 ] ) , ( [ 10 ] ) can not get an eigenvalue @xmath736 as @xmath737 .
@xmath477 since @xmath729 the limit normalization ( [ m4 ] ) shows that @xmath738 @xmath739 moreover , theorem [ theoremka ] ( 2 ) applied to the trapped mode @xmath740 gives the formula@xmath741 where @xmath742 @xmath268 is independent of @xmath90 according to the content of section [ sect5.5 ] and the waves @xmath394 in ( [ 13 ] ) and @xmath743 in ( [ 27 ] ) do not appear in the expansion of @xmath715 due to the absense of decay at infinity .
since the right - hand side of ( [ u6bis ] ) is uniformly bounded in @xmath744 , @xmath302 ( see section [ sect5.5 ] again ) , we have@xmath745 along a subsequence of @xmath746 moreover , the last equality in ( [ u00 ] ) turns into@xmath747 we multiply this relation with @xmath748 as @xmath749 and derive from ( [ u7 ] ) the absurd formula @xmath750 thus , there can exist at most one eigenvalue indicated in ( [ u0 ] ) .
in section [ sect1.3 ] we have changed the original problem ( [ 1 ] ) , ( [ 2 ] ) in @xmath752 for the neumann problem ( [ 9 ] ) , ( [ 10 ] ) in the half @xmath136 of the waveguide while assuming that an eigenfunction is even in @xmath81 . replacing ( [ 10 ] ) by the mixed boundary conditions@xmath753 we deal with the alternative , namely an eigenfunction is odd in @xmath81 and therefore vanishes at @xmath754 the variational formulation of the problem ( [ 9 ] ) , ( [ u8bis]),@xmath755 involves a subspace of functions in @xmath756 which are null on @xmath757 evident modifications of considerations in sections [ sect5 ] and [ sect6 ] adapt all our results to the mixed boundary value problem ( [ 9 ] ) , ( [ u8bis ] ) .
the only but important difference is that the solutions ( [ 491 ] ) , ( [ 492 ] ) of the limit neumann problem in the half - strip @xmath758 now turn into the following ones:@xmath759 thus , supposing that , for an infinitesimal sequence @xmath760 the problem ( [ 9 ] ) , ( [ u8bis ] ) in @xmath761 has an eigenvalue @xmath708 with the properties ( [ u0 ] ) at @xmath762 we obtain that a non - trivial limit @xmath763 , cf .
( [ u2 ] ) , of the corresponding eigenfunction @xmath764 becomes @xmath765 now , in contrast to section [ sect7.1 ] , we may insert @xmath766 into the green formula in @xmath723 together with one of three bounded functions @xmath767 and @xmath768 .
similarly to ( [ u3 ] ) and ( [ u4 ] ) , these possibilities allow us to conclude that @xmath769 , @xmath770 and , hence , @xmath771 .
the observed contradiction and remark [ remarkuu1 ] which remains true for the problem ( [ 9 ] ) , ( [ u8bis ] ) , confirm the absence of eigenvalues in a small neighborhood of the threshold @xmath772 for any @xmath773 , the limit neumann problem in @xmath774 has the solutions@xmath775 where @xmath776 , @xmath777 and , thus , the augmented scattering matrix takes the form@xmath778{cc}1 & 0\\ 0 & i \end{array } \right ) .
\label{u12bis}\ ] ] to fulfill the criterion ( [ 33 ] ) , the perturbation @xmath674 in the waveguide @xmath136 has to turn the right - hand bottom element of the matrix ( [ u12bis ] ) into @xmath779 that can not be made , e.g. , for any @xmath780 , $ ] @xmath781 and a small @xmath90 .
the latter fact may be easily verified by either constructing asymptotics in sections [ sect3 ] , [ sect6 ] , or applying a perturbation argument as in section [ sect5.5 ] .
notice that we have succeeded in sections [ sect3.2 ] , [ sect4.2 ] to construct asymptotics ( [ s11 ] ) with the main term @xmath307 because the spectral parameter ( [ 42 ] ) stays too close to the threshold and the augmented scattering matrix in @xmath236 is not continuous at @xmath503 , cf .
section [ sect5.5 ] . in the case of the mixed boundary value problem ( [ 9 ] ) , ( [ u8bis ] )
evident changes in solutions ( [ u11bis2 ] ) give the matrix @xmath782 instead of ( [ u12bis ] ) but our final conclusion remains the same .
the material of the previous three sections proves the last assertion in theorem [ theoremex ] .
the interval @xmath53 where the eigenvalue ( [ t10 ] ) is unique in the waveguide @xmath0 with @xmath783 and a fixed @xmath784 can be enlarged up to @xmath785 @xmath786 due to remark [ remarkuu1 ] . moreover , enhancing our consideration in section [ sect7.3 ] by dealing with the exponential waves @xmath787 and the augmented scattering matrix of size @xmath788 , cf .
@xcite , confirms that any @xmath789 can not be an eigenvalue as well .
we will discuss the higher thresholds @xmath790 with @xmath791 in section [ sect8.2 ] . to confirm theorem [ theoremun ] , we use a similar reasoning . in this
way,,we recall the asymptotic formulas ( [ s01 ] ) , ( [ anti ] ) and ( [ est ] ) and observe that @xmath792 can not vanish for a small @xmath90 when the length parameter ( [ t11 ] ) stays outside the segment@xmath793 \label{u13bis}\ ] ] if @xmath17 belongs to ( [ u13bis ] ) , the uniqueness of the solution @xmath794 of the abstract equation ( [ t7 ] ) which is equivalent to the criterion in theorem [ theorema ] follows from the contraction principle .
let us consider the mixed boundary value problem ( [ 1 ] ) , ( [ 7 ] ) . as in section [ sect1.3 ] we reduce it to the half ( [ 11 ] ) of the perturbed waveguide @xmath795 , cf .
( [ 9 ] ) , ( [ 10]):@xmath796 if @xmath797 stays below the continuous spectrum @xmath798 of the problem ( [ x1 ] ) , there is no oscillating wave but deal with the exponential waves@xmath799 and , similarly to ( [ 27 ] ) , ( [ 30 ] ) , compose the linear combinations@xmath800 the conditions ( [ 28 ] ) , ( [ 29 ] ) and ( [ 19 ] ) with @xmath801 are satisfied and we may determine the augmented scattering matrix @xmath802 which now is a scalar .
theorem [ theorema ] remains valid and , therefore , the equality@xmath803 states a criterion for the existence of a trapped mode .
constructing asymptotics of @xmath97 and solving the equation ( [ x2 ] ) yield the relation ( [ 8 ] ) for an eigenvalue in the discrete spectrum of the problems ( [ x1 ] ) and ( [ 1 ] ) , ( [ 7 ] ) . repeating arguments from sections [ sect5 ] - [ sect7 ] proves estimates of the asymptotic remainders as well as the uniqueness of the eigenvalue @xmath804 however , for any @xmath805 the latter conclusion requires to explain a distinction between analysis of isolated and embedded eigenvalues .
the main difference is caused by the application of the criterion ( [ 33 ] ) which in the case of the scalar @xmath97 changes just into one equation@xmath806 which is equivalent to ( [ x2 ] ) because @xmath807 as a result we may satisfy ( [ x3 ] ) by choosing @xmath689 and do not need the additional parameter @xmath808 in ( [ t1 ] ) which was used in section [ sect4 ] to solve the system ( [ t2 ] ) of two transcendental equations .
in other words , the absence of oscillating waves crucially restricts a possible position of @xmath809 to the unit circle on the complex plane while the entry @xmath99 in the previous unitary matrix @xmath97 of size @xmath810 see section [ sect2.2 ] , can step aside from @xmath811 and a fine tuning by means of @xmath677 is necessary to assure the equality ( [ 33 ] ) .
a straightforward modification of our approach may be used for an attempt to construct embedded eigenvalues near the thresholds @xmath812 @xmath791 of the continuous spectrum@xmath813 of the problem ( [ 9 ] ) , ( [ 10 ] ) in @xmath814 at the same time , the number of oscillating outgoing waves at the threshold @xmath790 equals @xmath815 and , therefore , size of the augmented scattering matrix becomes @xmath816 in this case the fine tuning needs at least @xmath815 free parameters , cf .
@xcite , instead of only one @xmath677 in section [ sect4 ] .
additional parameters can be easily introduced when the perturbed wall is a broken line like in fig .
[ f6 ] , a , with @xmath817 @xmath818 and @xmath815 .
the amplification of the augmented scattering matrix does not affect the criterion ( [ 33 ] ) in theorem [ theorema ] , in sections [ sect3 ] and [ sect4 ] . in the mirror symmetry with respect to the line @xmath819
is denied , see fig .
[ f6 ] , b , then we have to analyze the problem ( [ 1 ] ) , ( [ 2 ] ) in the intact waveguide @xmath752 where the augmented scattering matrix gets rise of size even in the case @xmath820 in this sense the box - shaped perturbation is optimal because it demonstrates all technicalities but reduces the computational details to the necessary minimum .
a preliminary assessment predicts that embedded eigenvalues of the problem ( [ 9 ] ) , ( [ 10 ] ) in @xmath136 do not appear near any threshold @xmath790 with @xmath821 but we are not able to verify this fact rigorously .
all procedures described above can be applied to detect eigenvalues of the helmholtz equation ( [ 1 ] ) in the quantum waveguide @xmath752 , cf .
@xcite , with the dirichlet condition ( [ 5 ] ) .
however , the asymptotic structures must be modified a bit due to the following observation .
the correction term @xmath822 in the inner asymptotic expansion@xmath823 cf .
( [ 62 ] ) , must be found out from the mixed boundary value problem in the semi - strip@xmath824 according to the kondratiev theory @xcite , see also ( * ? ? ?
2 ) , a solution of this problem admits the representation@xmath825 where @xmath826 has the decay @xmath827 is smooth everywhere in @xmath828 except at the point @xmath829 and behaves as@xmath830 while @xmath586 is the polar coordinates system centered at @xmath831 .
the singularity in ( [ x11 ] ) leads the function @xmath832 out from the sobolev space @xmath833 .
nevertheless , the solution @xmath822 still lives in appropriate kondratiev space with a weighted norm so that the coefficient @xmath834 in ( [ x10 ] ) can be computed by inserting @xmath835 and @xmath836 into the green formula in @xmath837 . to compensate for the singularity
, one may construct a boundary layer as a solution of the dirichlet problem in the unbounded domain ( [ 39 ] ) in fig .
[ f4 ] , a. the above commentary exhibits all changes in the asymptotic analysis in section [ sect3.2 ] . as for the justification scheme in section [ sect6 ] , it should be noted that , due to the dirichlet condition ( [ 5 ] ) , the inequality ( [ hardy ] ) of hardy s type takes the form@xmath838 and sheds the factor @xmath839 in the weight function ( [ j14 ] ) . as a result , the factor @xmath840 occurring in ( [ 8 ] ) and ( [ est ] ) for the neumann case , disappears from the asymptotic remainder in ( [ 6 ] ) for the dirichlet condition .
agmon s. , douglis a. , nirenberg l. , estimates near the boundary for solutions of elliptic partial differential equations satisfying general boundary conditions .
pure appl .
( 1959 ) 623 - 727 .
maslov , an asymptotic expression for the eigenfunctions of the equation @xmath841 with boundary conditions on equidistant curves and the propagation of electromagnetic waves in a waveguide , soviet physics dokl . 3 ( 1959 ) 11321135 .
mazya v.g .
, nazarov s.a . ,
plamenevskij b.a .
, asymptotic theory of elliptic boundary value problems in singularly perturbed domains , operator theory : advances and applications , 112 , birkhuser verlag , basel ( 2000 ) .
mazya v.g .
, plamenevskii b.a .
, estimates in @xmath842 and hlder classes and the miranda - agmon maximum principle for solutions of elliptic boundary value problems in domains with singular points on the boundary , amer
soc . transl . 123
( 1984 ) 1 - 56
. nazarov s.a .
, properties of spectra of boundary value problems in cylindrical and quasicylindrical domains , sobolev spaces in mathematics .
international mathematical series , 9 .
new york : springer , 2008 , 261309 . | we consider the spectral neumann problem for the laplace operator in an acoustic waveguide @xmath0 obtained from a straight unit strip by a low box - shaped perturbation of size @xmath1 where @xmath2 is a small parameter .
we prove the existence of the length parameter @xmath3 with any @xmath4 such that the waveguide @xmath5 supports a trapped mode with an eigenvalue @xmath6 embedded into the continuous spectrum .
this eigenvalue is unique in the segment @xmath7 $ ] and is absent in the case @xmath8 the detection of this embedded eigenvalue is based on a criterion for trapped modes involving an artificial object , the augmented scattering matrix .
the main technical difficulty is caused by corner points of the perturbed wall @xmath9 and we discuss available generalizations for other piecewise smooth boundaries .
keywords : acoustic waveguide , neumann problem , embedded eigenvalues , continuous spectrum , box - shaped perturbation , asymptotics msc : 35p05 , 47a75 , 49r50 , 78a50 . |
in the treatment of the motion of _ extended _ bodies in classical field theory , the derivation of radiation reaction forces is based upon certain expansions of the retarded field potentials in powers of the retardation @xcite .
the resulting local equations of motion involve derivatives of the acceleration and generally suffer from the existence of unphysical runaway solutions . under certain model circumstances ,
we trace the origin of these problems to the expansion of the functions of the retarded arguments resulting in the replacement of the original nonlocal delay equations of motion by local higher - derivative equations that exhibit runaway solutions . in this general context
, the properties of the delay equations that appear in classical field theory were first studied by l. bel @xcite .
although our approach is rather general , for the sake of concreteness we discuss physical situations involving only the gravitational interaction .
consider , for instance , inspiraling compact binaries that are expected to be promising sources of gravitational radiation .
for a binary that is comprised of two compact objects neutron stars or black holes with , say , approximately equal masses @xmath0 and @xmath1 in nearly circular orbits about each other , the relative orbital radius decays because orbital energy is emitted in the form of gravitational radiation .
the dynamics of a usual binary system can be adequately described using the post - newtonian approximation scheme that is valid in case the gravitational field is everywhere ` weak ' and the motion is slow , that is @xmath2 , where @xmath3 is the characteristic orbital speed and @xmath4 is the speed of light .
although einstein s equations have a hyperbolic character associated with the retarded gravitational interaction , the standard post - newtonian approximation scheme of general relativity deals with functions of instantaneous coordinate time @xmath5 rather than the retarded time @xmath6 , where ( for the binary system ) @xmath7 is the effective distance between the bodies ( approximately the relative orbital radius ) .
the gravitational potentials , which are originally functions of the retarded time , are expanded in taylor series about @xmath5 using the effective small parameter in this expansion that can be written as @xmath8 , where @xmath9 is the relative orbital frequency and @xmath10 is the binary period . because the gravitational waves emitted by the binary have an effective frequency of @xmath11 and wavelength @xmath12 , the small parameter in the expansion
can be reduced to the ratio @xmath13 . due to the observational fact
that in typical astronomical systems @xmath14 , the first few terms of such an expansion can be used to derive the post - newtonian equations of motion that describe , for instance , the orbital evolution of the binary pulsars discovered by hulse and taylor @xcite .
the post - newtonian equations of motion of binary stars are similar to the abraham - lorentz form in electrodynamics @xcite but , because of the tensorial character of the gravitational field , these equations involve not only the third , but the fourth and fifth derivatives of the stars positions with respect to time as well .
schematically , the equation of the relative orbital motion reads @xmath15 where @xmath16 is the radius vector connecting the stars , the overdot denotes differentiation with respect to time , @xmath17 and per reduced mass @xmath18 is the newtonian force , @xmath19 is the post - newtonian force , @xmath20 is the post - post - newtonian force and @xmath21 is the gravitational radiation reaction force responsible for the decay of the orbital period ( @xmath22 ) associated with the emission of gravitational waves by the binary system . in the quadrupole approximation under consideration here ,
the gravitational waves carry away energy and angular momentum , but not linear momentum @xcite ; therefore , the total momentum of the binary system is conserved and this fact is responsible for the absence of a force term proportional to @xmath23 in .
moreover , all tidal , spin - orbit , and spin - spin interactions are neglected in ; the only parameters in equation are the masses and the separation between the centers of mass of the members of the binary system .
we mention that relativistic hydrodynamical ( euler ) equations similar to system have been derived to describe the motion of the fluid elements of the stars @xcite .
the higher time - derivative equations of the form can not be used directly to predict the dynamical evolution of a physical system because of the existence of so - called runaway modes that have been much discussed in the literature on electrodynamics but not in connection with astrophysical problems involving gravitational radiation reaction and the calculation of templates of the gravitational waves emitted by coalescing binaries @xcite .
in analogy with electrodynamics , the existence of these runaway modes suggests that the truncated equations of the form may not correctly predict the qualitative behavior of the solutions of the original true dynamical delay - type equations @xcite that involve the retarded time @xmath6 .
moreover , the existence of runaway modes can cause serious difficulties for numerical integration in addition to the problems associated with the inaccuracies inherent in the approximation of higher - order derivatives by finite differences @xcite . in case
the post - newtonian expansion parameter @xmath24 is sufficiently small , we will provide in the following section a theoretical basis for eliminating the runaway solutions by replacing system with a new model that is a system of second - order ordinary differential equations .
within this theory , high - order vector differential equations like equation are not the desired approximate equations of motion , and they should not be used for numerical integration . rather , system must be viewed merely as an intermediate step in the derivation of the physically correct , second - order model equation with no runaway solutions that faithfully approximates the dynamics of the underlying delay - type equation . for illustration purposes , we apply this approach in section [ sec:3 ] to a discussion of one - dimensional gravitational dynamics of a two - body system .
as expected , the reduced model predicts the correct motion of the binary except possibly when @xmath24 is not small and residual terms that have been neglected in equation start to play a significant role .
indeed , for an inspiraling binary system , the effective delay @xmath25 increases to some noticeable finite value as the system approaches coalescence .
motivated by this physical scenario , we introduce a simple model involving variable delay in section [ sec:4 ] that can be expressed as a duffing - type differential - delay equation .
this model is then analyzed to show some specific behavior of this delay equation that is not predicted by expansion in powers of the delay .
finally , section [ sec:5 ] contains a discussion of our results .
the delay - type equations of motion with retarded arguments are usually too complicated for mathematical analysis ; therefore , we limit our discussion in this section to equations with constant delays . although this is an unrealistic restriction in general , we note that for an astrophysical binary system consisting of compact point - like neutron stars or black holes moving around each other along circular orbits , the delay is almost constant .
in fact , a close approximation to this delay is the ratio @xmath26 , where @xmath7 , the radius of the relative orbit , is changing very slowly due to the emission of gravitational energy in the form of gravitational waves .
taking into account the last remark , let us consider a family of delay differential equations of the form @xmath27 where @xmath28 is viewed as a real dimensionless parameter and @xmath29 is a variable in @xmath30 ; intuitively , the constant delay @xmath28 corresponds in effect to @xmath31 .
the members of this family are examples of a more general and widely studied class called retarded functional differential equations ( see @xcite ) . using the delay equation as an abstraction for the retarded - time model that is supposed to be approximated by a system of the form
, we will discuss an approach for extracting the ` correct ' dynamical equations of motion from system that eliminates the runaway solutions .
our approach assumes the existence of an attractor for the underlying delay - type equation .
we will rely on the work of bel @xcite for ( numerical ) evidence in favor of the existence of attractors in the retarded equations of motion with space - dependent delays that appear in electrodynamics ; but , we know of no mathematical proof for the existence of attractors for these equations or for the similar delay - type equations of astrophysics .
indeed , the proof of the existence of attractors for delay - type equations with space - dependent delays remains a challenging mathematical problem of physical significance . for the delay equation , however , if @xmath32 is sufficiently small , then the corresponding member of the family
has a global @xmath33-dimensional attractor such that the restriction of the delay equation to this attractor is equivalent to a first - order system @xmath34 of ordinary differential equations .
we will eventually outline a proof of this result .
but , let us first discuss our approach to eliminating the runaway solutions . the solutions of the delay equation approach the attractor exponentially fast ; therefore , the system @xmath34 on this attractor determines ( asymptotically ) the true dynamical behavior of the system , hence we consider it to be the ` correct ' physical model .
on the other hand , it is easy to see that if equation is expanded in the small parameter @xmath28 and truncated at some order @xmath35 , then an @xmath35th - order ordinary differential system @xmath36 akin to system is obtained such that the coefficient of the @xmath35th - order time derivative of @xmath29 contains the factor @xmath37 and is therefore singular in the limit as @xmath38 . for @xmath39 and sufficiently small
, the high - order differential equation @xmath36 has an equivalent first - order system @xmath40 that has an @xmath33-dimensional ( invariant ) slow manifold .
moreover , this slow manifold has corresponding stable and unstable manifolds ; in effect , the first - order system has ( physical ) solutions that are asymptotically attracted to the slow manifold and ( unphysical or runaway ) solutions that are asymptotically repelled from the slow manifold .
the restriction of the first - order singularly perturbed system of differential equations to its slow manifold is of course an @xmath33-dimensional first - order system of ordinary differential equations @xmath41 on this @xmath33-dimensional manifold .
our main result states that _ in appropriate local coordinates , the system @xmath41 on the slow manifold agrees to order @xmath35 in @xmath28 with the first - order system @xmath34 on the global attractor of the underlying delay differential equation ; therefore , the system @xmath41 , which can be obtained directly from the high - order differential equation @xmath36 , is a faithful approximation of the ` correct ' physical model . _
generalizing to the abraham - lorentz type equation ( analogous to @xmath36 ) , the correct physical model is obtained as the system of ordinary differential equations ( analogous to @xmath41 ) that determines the motion on the slow manifold of a corresponding first - order system that is viewed as being singularly perturbed relative to the small parameter @xmath24 . while there is evidence that the mathematical assertions in the scenario just proposed are valid , some of these assertions have not yet been rigorously justified in full generality , even for the case of fixed delays . in the remainder of this section
we will provide some evidence , in the case of fixed delays , for the existence of a global attractor and for the claim that the dynamical system on this attractor is well approximated by the dynamical systems on the slow manifolds of singularly perturbed first - order systems obtained by truncations of the expansion of the delay equation in powers of the delay .
our approach for the elimination of runaway solutions is equivalent to the procedure of iterative reduction ( also called order reduction ) that is often used to eliminate runaway solutions by means of the evaluation of the higher time - derivative terms in equations like ( [ 5der ] ) by the repeated substitution of the equations of motion and the subsequent reduction of the resulting equation to one of the second order ( cf .
thus , our approach provides a theoretical framework for the rigorous justification of iterative reduction ( cf .
@xcite ) , a procedure that has been justified so far by physical intuition . for _ conservative _ higher time - derivative systems ,
the order reduction procedure has been investigated within the frameworks of lagrangian and hamiltonian dynamics by a number of authors ( see @xcite and the references cited therein ) .
in particular , it can be shown that under suitable conditions relevant , for instance , to euler - lagrange equations analogous to equation truncated at the fourth order higher - derivative lagrangians can be iteratively reduced by redefinitions of position variables @xcite .
returning to the delay equation , we note that it has an infinite - dimensional state space of initial conditions .
for example , if the delay @xmath28 is a fixed positive number , then the natural state space of initial conditions is the infinite - dimensional vector space of continuous @xmath30-valued functions on the interval @xmath42 $ ] .
this space endowed with the supremum norm is a banach space that we will denote by @xmath43 .
note that for an arbitrary continuous @xmath30-valued function @xmath44 defined on the interval @xmath45 $ ] , the function @xmath46 given by @xmath47 is in @xmath43 . under the assumption that @xmath48 is a smooth function and @xmath49
, there is a unique continuous solution @xmath50 of the corresponding delay equation in the family such that @xmath50 is uniquely defined for @xmath51 and @xmath52 ( see , for example , @xcite ) .
the state of the system at time @xmath53 is defined to be the function @xmath54 in @xmath43 . to see
that there is an attractor for the family in case @xmath28 is sufficiently small , it is convenient to introduce the fast time @xmath55 , valid for @xmath56 , so that with @xmath57 the family takes the form @xmath58 and each member of this family , parametrized by @xmath28 , has the same state space the continuous functions on the interval @xmath59 $ ] . for each @xmath56
the delay equation is equivalent to the corresponding member of the family .
for @xmath60 the corresponding differential equations are not equivalent , but this is of no consequence because we are only interested in the solutions of the family for @xmath56 . by viewing the unperturbed system , namely @xmath61 , as a delay equation with unit delay ,
it is clear that the solution with initial state @xmath62 is given by @xmath63 on the interval @xmath59 $ ] and by the constant @xmath64 for @xmath65 .
the initial state in @xmath43 thus evolves at time @xmath66 to its final constant state , the function defined on the interval @xmath59 $ ] with the constant value @xmath67 .
thus , we conclude that the @xmath33-dimensional space of constant functions on @xmath59 $ ] is a global attractor for the delay equation @xmath61
. moreover , the convergence to this attractor is faster than any exponential ( the solution reaches the attractor in finite time ) , and the dynamical system on this attractor is given by the ordinary differential equation @xmath61 .
if @xmath48 is appropriately bounded and @xmath28 is sufficiently small , then , because the contraction rate to the attractor is exponentially fast , the attractor persists in the family in analogy with the persistence of attractors in finite - dimensional dynamical systems ; in fact , each corresponding member of the family has an @xmath33-dimensional attractor in the state space @xmath43 and the restriction of the dynamical system to this attractor is an ordinary differential equation .
in particular , the family has a corresponding family of invariant manifolds ( that is , manifolds consisting of a union of solutions ) that depend smoothly on the parameter @xmath28 . to identify the dynamical system on an attractor of a delay equation ,
let us suppose that the delay equation has a family of @xmath33-dimensional invariant manifolds parametrized by @xmath28 .
moreover , let @xmath68 denote the local coordinate on these invariant manifolds , and let @xmath69 denote the solution with the initial condition @xmath70 on the invariant manifold corresponding to the parameter value @xmath28 .
because these solutions satisfy the delay equation , we have that @xmath71 therefore , the generator of the dynamical system on the attractor is the vector field @xmath72 under our assumption that this vector field is analytic in @xmath28 , it can be expanded as a taylor series about @xmath60 by differentiating the function @xmath73 with respect to @xmath28 . to this end
, we note that the partial derivatives of @xmath74 with respect to its first argument can be evaluated using equation , and partial derivatives with respect to its third argument vanish at @xmath60 since @xmath70 ; moreover , its mixed partial derivatives can be evaluated by differentiation of equation with respect to @xmath28 . as a concrete and instructive example of the construction of the dynamical system on an attractor ,
let us consider a simple case of equation by replacing it with @xmath75 where the function @xmath76 is the scalar linear function given by @xmath77 so that the associated family of delay equations is @xmath78 in this case , there is a corresponding family of solutions given by @xmath79 where @xmath80 is the unique _ real _ root of the equation @xmath81 , a fact that is easily checked by direct substitution of equation into the delay equation .
the dynamical system on the invariant manifold is generated by the family of vector fields @xmath82 . by an application of the lagrange inversion formula @xcite , the taylor series expansion of @xmath83
about @xmath60 is @xmath84 and its radius of convergence is @xmath85 .
the qualitative behavior of solutions of the system for small @xmath28 is clear : all solutions are attracted to a one - dimensional invariant manifold on which the dynamical system is the linear ordinary differential equation @xmath86 .
for example , if @xmath87 and @xmath28 is sufficiently small , then all solutions are attracted to the trivial solution @xmath88 .
let us now turn to the standard approach in physics where an underlying delay equation is expanded in powers of the delay to obtain an ordinary differential equation of motion . to illustrate this ,
let us consider a special scalar case of the delay equation given by @xmath89 and let us suppose , in analogy with the true dynamical delay - type equations that might arise in theories of electromagnetism and gravitation , that the true equation of motion for some process is the delay equation .
the result of expanding equation to order @xmath90 is the second - order differential equation ( an analogue of equation ) @xmath91,\ ] ] where a prime denotes differentiation with respect to @xmath29 .
although we only write the second - order expansion , we note that the coefficient of the @xmath35th - order time derivative of @xmath29 in the @xmath35th - order expansion is @xmath92 .
hence , the corresponding @xmath35th - order ordinary differential equation is singular in the limit as @xmath38 .
also , if we assume that system has a ( smooth ) family of attractors parametrized by @xmath28 , then the corresponding family of vector fields generating the dynamical systems on these attractors is given to second order in @xmath28 by @xmath93 ^ 2\\ & & \mbox{}+f'(x)[3f'(x)+g'(x)][f(x)+g(x)]\big\}+o(\tau^3).\end{aligned}\ ] ] the ` correct ' model ( that is , the dynamical system on the attractor in the original delay equation ) can be obtained by treating the expanded and truncated system akin to system as a singular perturbation problem , which can be analyzed using fenichel s geometric theory of singular perturbations @xcite .
a basic result of this theory states that if an @xmath35th - order singular perturbation problem with small parameter @xmath28 is recast as a first - order ( ` fast ' ) system and the corresponding unperturbed system has an invariant manifold that satisfies certain conditions ( normal hyperbolicity ) , then for sufficiently small @xmath28 each member of the family of perturbed first - order systems has an invariant slow manifold .
the dynamical system on this slow manifold for the perturbed first - order family obtained from the @xmath35th - order truncation of the delay equation is the desired faithful approximation to the correct model .
for example , let us recast the second - order ordinary differential equation as the first - order singular perturbation problem @xmath94 .
\end{aligned}\ ] ] using fenichel s theory , it is easy to show that each member of this family , corresponding to a sufficiently small value of @xmath32 , has a slow manifold . also , it is possible to prove that the family of vector fields on these manifolds agrees to order @xmath90 with the family of vector fields on the attractor in the state space of the underlying family of delay equations .
we note that these results are valid for the vector case of delay equation as well . for the delay equation , and also for more general families of delay equations where the delay is viewed as a small parameter , we conjecture that the slow vector field , for an appropriately defined first - order system that is equivalent to the @xmath35th - order truncation of the expansion of the family in powers of the small delay , agrees to order @xmath35 with the vector field on the attractor in the state space of the original delay equation .
we have just mentioned that this conjecture is true for the delay equation in case @xmath95 .
it can be shown that the conjecture is true in general for the linear delay equation @xmath96 , where @xmath29 is a variable in @xmath30 and @xmath97 is a nonsingular @xmath98 matrix @xcite .
as we have already discussed , singular equations of motion like system generally have unphysical runaway solutions . to eliminate these solutions and leave only the physical solutions , the singular system must be replaced by the dynamical system on the corresponding slow manifold . in effect ,
the truncated equations obtained from the underlying delay equation after expansion in the small delay must be replaced by the system obtained using iterative reduction ; this system is equivalent to the dynamical system on the slow manifold . without this replacement ,
the appearance of spurious runaway modes in inevitable , and their existence will cause overflows in numerical simulations .
to illustrate the singular perturbation procedure described in section [ sec:2 ] as a method for the elimination of runaway solutions , we examine an application of this approach to a one - dimensional abraham - lorentz equation of the form .
let us consider an ideal linear quadrupole oscillator ( that is , two masses @xmath0 and @xmath1 connected by a spring of negligible mass ) , where the only source of damping is the gravitational radiation reaction force associated with the emission of gravitational radiation due to the variable quadrupole moment of the system . a model for the ( dimensionless ) relative position @xmath99 of these particles , with gravitational radiation damping included , is the fifth - order ordinary differential equation @xmath100 where the small parameter is given by @xmath101 , @xmath102 is the reduced mass ( @xmath103 ) , @xmath104 is the spatial scale parameter and @xmath105 is the frequency of the ideal linear oscillator such that @xmath106 is the temporal scale parameter . in equation , we have neglected the newtonian gravitational interaction between @xmath0 and @xmath1 as well as all relativistic effects except for radiation damping .
let us note that this oscillator has an equilibrium solution @xmath107 . according to our general approach described in section [ sec:2
] , this system can be treated as a singular perturbation problem , where the physically correct dynamical system would be the system defined on the slow manifold of a corresponding , appropriately chosen , first - order system .
we emphasize that the fifth - order differential equation is not the correct physical model ; for example , to specify a solution , the initial position and its first four time derivatives must be given .
even with the obvious choice for these initial conditions that is , the initial conditions for a sinusoidal oscillation a numerical integration shows that such solutions do not oscillate ; rather , they are divergent . to obtain a system with the expected dynamical behavior of an under - damped oscillator , the differential equation must be replaced by its restriction to an appropriate slow manifold .
we will not carry out the complete reduction procedure here @xcite .
we note , however , that the system matrix for the linearization of system at the steady state solution @xmath108 has five distinct eigenvalues that are given to lowest order in the small parameter by @xmath109 for small @xmath110 , the first three eigenvalues are ` fast ' and the last two are ` slow ' .
this suggests that the nonlinear system has a two - dimensional slow manifold .
in fact , in accordance with our general scheme , the restriction of the dynamical system to this invariant manifold is a second - order system that gives the correct post - newtonian dynamics . in this case , the dynamical system on the slow manifold to first order in the small parameter is given by the second - order differential equation @xmath111 a unique solution of this equation is obtained by specifying only the initial relative position and velocity of the oscillating masses . for @xmath99 near the equilibrium state @xmath108 , the expected dynamics for the radiating system is revealed : the relative motion is an under - damped oscillator .
numerical integration of this equation using standard algorithms is stable and produces the expected result .
the iterative reduction procedure can also be used to obtain equation from equation .
even our simple example illustrates the necessity of reducing the higher - order equations of motion involving radiation reaction before numerical integration . for the more realistic hydrodynamic equations that include conservative post - newtonian terms as well as radiation reaction
, the corresponding euler equation must involve these forces in the _ reduced _ form , that is , they should contain at most the position and the velocity of the fluid element @xcite .
it is important to point out that the reduction procedure described in sections [ sec:2 ] and [ sec:3 ] can not in general be expected to produce a good approximation to the true dynamics for ` large ' delays . as a simple but revealing example , let us reconsider the scalar linear delay equation @xmath112 with @xmath113 . for small @xmath32
, we have already shown that all orbits in the state space are attracted to a one - dimensional attractor on which the dynamical system is given by the vector field with @xmath114 . for @xmath115
less than the radius of convergence of this series @xmath116 , the correct dynamical behavior of the delay equation is predicted by this vector field .
because , in this case , the zeroth - order approximation @xmath117 already has a hyperbolic structure ( that is , all solutions are attracted to the rest point at the origin exponentially fast ) , even the zeroth order approximation determines the qualitative dynamics for these values of @xmath28 . by inspection of this delay equation
, it might seem natural to conclude that the fixed delay @xmath28 does not influence the behavior for sufficiently large @xmath5 and the approximation @xmath117 remains valid for all fixed delays .
this is not true .
for instance , if @xmath118 , then the delay equation has the two - parameter family of exact solutions @xmath119 therefore , the qualitative behavior of the delay equation @xmath120 is certainly not predicted by the ordinary differential equation @xmath117 , or by the corrections to this equation within the radius of convergence of the slow vector field .
the transition of the dynamical behavior of this delay equation from a stable rest point to a periodic regime as @xmath28 increases is easily seen to be the result of a degenerate hopf bifurcation @xcite . indeed , we recall that @xmath121 is a solution of the delay equation under consideration if @xmath122 is a solution of the characteristic equation @xmath123 . for @xmath124 ,
the solutions of this equation have negative real parts and all such solutions are therefore attracted to the zero solution . if @xmath118 , then the characteristic equation has a pair of pure imaginary roots that give rise to the two - parameter family of periodic solutions . for @xmath125 ,
the characteristic equation has roots with positive real parts ; therefore , there are solutions that grow without bound .
nevertheless , for these values of @xmath28 , the delay equation has an attractor .
in fact , for @xmath126 , there is a two - dimensional attractor and the dynamical system on the attractor has the form @xmath127 corresponding to the roots @xmath128 of the characteristic equation @xmath129 with positive real parts . as @xmath28 increases further , the dimension of the attractor increases discontinuously by two at each @xmath130 , where @xmath131 .
the hopf bifurcation for delay equations with constant delays has been studied in detail .
for instance , a more sophisticated analysis ( see , for example , @xcite ) shows that @xmath132 is a supercritical hopf bifurcation value for the nonlinear scalar delay equation @xmath133x(t-\tau).\ ] ] moreover , this system has a nontrivial periodic orbit for each @xmath134 .
the delay - type equations of astrophysics generally do not have constant delays .
but as we have mentioned , for two coalescing neutron stars with nearly equal masses and on nearly circular orbits , the delays involved are almost constant ; in fact , this ` fast ' periodic motion evolves as a result of radiation damping on a timescale that is much longer than @xmath10 . during this ` slow ' evolution ,
the delay increases as the radius of the binary decreases due to the emission of gravitational waves .
motivated by this astrophysical scenario , we have studied an oscillator model with a time - dependent delay .
this example is not intended to be a realistic model , rather it is meant to illustrate some of the bifurcation phenomena that occur in delay equations with time - dependent delays .
our example is the second - order differential - delay equation @xmath135 where @xmath136 , @xmath137 and @xmath138 are constant system parameters and @xmath29 is viewed as the state variable of a ( duffing ) oscillator with variable delay @xmath139 such that @xmath140 .
here @xmath141 and @xmath142 are constants ; hence , @xmath143 is an exponentially decreasing or increasing function of time depending on whether @xmath141 is positive or negative , respectively . in any case
, we have a dynamic delay that is asymptotic to the constant value @xmath142 .
note that if @xmath144 , the delay is constant ; in this case , the corresponding second - order differential equation on the slow manifold ( to first order in @xmath145 ) is given by @xmath146 a form of van der pol s equation . in case @xmath147 , this differential equation typically has a stable limit cycle for @xmath148 .
but for @xmath149 ( that is when all forces are retarded ) , it is easy to prove that no periodic orbits exist and most solutions are unbounded . a typical plot of @xmath29 versus @xmath5 for system for @xmath150 is given in figure [ fig:1 ] , where the delay increases from its initial value @xmath151 to @xmath152 .
the initial response of the system ( where the delay is small ) is characterized by an oscillation as expected from equation , which follows from the expansion of equation to first order in @xmath153 .
but as @xmath145 increases , the qualitative behavior of the system is affected by three additional bifurcations not accounted for by equation . at the third bifurcation , the stable oscillation disappears .
additional bifurcations of the same type occur if @xmath142 is set to a larger value .
numerical experiments suggest that these bifurcations are not hopf bifurcations ; instead , they are ` center bifurcations ' , where at some parameter value there is a rest point of center type and one of the periodic orbits surrounding this rest point continues to exist as the parameter is changed .
the family with the parameter values as in figure [ fig:1 ] has a bifurcation of this type as @xmath145 increases through @xmath154 .
the behavior depicted in figure [ fig:1 ] is suggested by an analysis of the roots of the characteristic equation , @xmath155 for the linearization of the delay equation .
the bifurcation points ( corresponding to the existence of centers ) are given by @xmath156 where @xmath157 is a non - negative integer .
these are the values of @xmath145 such that the characteristic equation has pure imaginary roots .
a computation shows that if @xmath157 is even , then as @xmath145 increases a pair of pure imaginary roots crosses the imaginary axis into the right half - plane , and if @xmath157 is odd , then the roots cross into the left half - plane . under the assumption that the bifurcations are supercritical
, a stable limit cycle appears after the bifurcation in the first case ; in the second case , a stable limit cycle disappears . for the parameter values used to obtain figure [ fig:1 ] , the bifurcation values computed from equation are ( approximately ) @xmath158 , @xmath159 , @xmath160 , @xmath161 for @xmath162 such that @xmath163 . at @xmath164
a limit cycle appears , at @xmath165 the limit cycle disappears , and so on .
thus , these bifurcations account for the appearance and disappearance of oscillations in figure [ fig:1 ] .
we note that a similar sequence of bifurcations occurs whenever @xmath150 . on the other hand , if @xmath166 ( for example if @xmath149 ) , then all bifurcation points correspond to roots of the characteristic equation crossing into the right half - plane . in this case
, the bifurcations can be subcritical .
indeed , for @xmath149 , numerical simulations indicate that no limit cycle appears . as a result ,
solutions starting near the unstable rest point become unbounded .
the slow dynamical system , obtained by reduction from a truncation of an expansion of a delay equation in powers of the delay , approximates the dynamics on the global attractor of the delay equation as long as the delay is sufficiently small ; but , as our examples show , the ordinary differential equations obtained by expansion , truncation , and reduction _ can not _ be used in general to predict the correct dynamical behavior for sufficiently large delays .
we have mentioned , for example , that the dimension of the attractor of a family of delay equations , parametrized by the delay , can increase in dimension so that that the corresponding slow vector field is no longer defined on the attractor . but this is not the only possible scenario for the appearance of new attractor dynamics ; for example , the attractor could cease to exist or be a manifold for some values of the delay .
the abraham - lorentz type equation can be used , after reduction to a slow manifold , to predict the relative orbital motion of a relativistic binary system in the regime where the delay is sufficiently small .
the size of the maximum allowed delay would have to be computed on a case - by - case basis using the explicit form of the delay equation that models the dynamics of a coalescing pair of neutron stars .
the results of this section show that for sufficiently large delay the attractor does not in general correspond to the slow manifold .
the question remains whether such a divergence of behaviors could ever occur in the case of retarded equations of classical field theory .
this is an interesting open problem .
it is expected that interferometric gravitational wave detectors that are presently under construction will be able to detect signals from massive coalescing binary systems . for the analysis of such forthcoming data , it is important to have theoretically predicted wave forms ( ` templates ' ) for the relevant astrophysical processes . to this end ,
extensive computations are necessary that need to take gravitational radiation reaction into account @xcite .
the standard approach leads to higher time - derivative equations that involve runaway modes and inevitably produce incorrect results .
we have determined the source of the difficulty by investigating delay equations , which are essentially nonlocal , and the higher time - derivative equations that are obtained by truncations of the ( post - newtonian ) expansions in powers of the delay . for sufficiently small delays ,
a proper justification is provided for the usual method of replacing terms with higher - derivatives by terms with at most first derivatives using repeated substitution of the equations of motion ( ` iterative reduction ' ) .
we have shown that in the investigation of the solutions of higher - derivative equations that represent phenomena involving radiation reaction , it is essential to reduce such equations to the corresponding slow manifolds before numerical analysis .
our work suggests , however , that unexpected nonlocal phenomena could occur for sufficiently large delays that can not be predicted using the local equations of motion even after iterative reduction . | starting from delay equations that model field retardation effects , we study the origin of runaway modes that appear in the solutions of the classical equations of motion involving the radiation reaction force . when retardation effects are small , we argue that the physically significant solutions belong to the so - called _ slow manifold _ of the system and we identify this invariant manifold with the attractor in the state space of the delay equation .
we demonstrate via an example that when retardation effects are no longer small , the motion could exhibit _ bifurcation _ phenomena that are not contained in the local equations of motion . |
spin dependent transport phenomena in low - dimensional semiconductor structures have been of a lot of interest to the scientific community in recent years due to the potential applications in the highly emerging field of spintronics.@xcite intense research in this field was initiated after the proposal of spin field effect transistor by datta and das @xcite .
the principal aim of this field is to produce pure spin current and its manipulation on semiconductor nanostructure devices .
one important tool for generating pure spin current is the well known spin hall effect ( she ) .
@xcite in she spin - orbit interaction ( soi ) leads to a generation of spin current from an external electric voltage . for the happening of she , the charge carriers are either electrons in the conduction band or holes in the valence band in a iii - v semiconductor such as gaas
many @xmath1-doped semiconductors such as gaas , insb , si etc show four - fold degeneracy in their valence band around the @xmath2 point .
this kind of systems is described by @xmath3 luttinger@xcite hamiltonian .
the luttinger hamiltonian leads to two two - fold degenerate energy branches having different effective masses .
these two dispersion branches are known as the heavy - hole ( hh ) and light - hole ( lh ) bands which are described by @xmath4 and @xmath5 , respectively . for spin - orbit
coupled two - dimensional hole system ( 2dhs ) in a narrow quantum well , the @xmath3 luttinger hamiltonian can be projected onto @xmath6 hh states giving rise to an effective @xmath7 rashba hamiltonian@xcite provided that the hole density is sufficiently low . in 2dhs
the rashba soi is tri - linear in momentum . in recent years
an oscillatory quantum motion , called @xmath0 ( zb ) , has become a central point of research interest in various low - dimensional semiconducting systems .
the length and time scales corresponding to this trembling motion in vacuum are of the order of the compton wavelength @xmath8 nm and @xmath9 s , respectively . due to this ultra small length and time scale ,
it has not been possible to make an experimental verification of zb phenomenon in vacuum so far .
however an intense interest was initiated around 2005 by zawadzki@xcite who has drawn an analogy between the dirac equation of a free dirac electron and the @xmath10 theory of the band structure of a narrow gap semiconductor .
he was able to find out a new length scale @xmath11 nm which is much larger than @xmath12 .
later , the problem of zb of electrons and holes in iii - v semiconductor quantum wells in the presence of spin - orbit interaction was studied by schliemann@xcite et .
these pioneering works initiated tremendous motivation for theoretical studies in search of zb in various condensed matter systems such as crystalline solid,@xcite graphene,@xcite carbon nanotubes,@xcite luttinger liquid,@xcite superconductor,@xcite ultra - cold atom,@xcite topological insulators @xcite etc .
it was also reported that the origin of the minimal conductivity @xcite in graphene can be explained in the light of the peculiar phenomenon zb .
recently , a general theory for zb has been developed by david and cserti.@xcite winkler@xcite et .
al . considered a number of effective hamiltonians representing different systems and studied various consequences of zb oscillations .
the effect of an in - plane magnetic field on the zb oscillations in a rashba - dresselhaus system has been studied.@xcite very recently , a complex quantum motion known as `` super zitterbewegung '' in graphene has been studied@xcite theoretically using time - dependent two - band hamiltonian in the framework of rotating wave approximation . although the zb phenomenon in vacuum is not yet observed experimentally , however an optical and acoustic analog of relativistic zb
have been observed experimentally in optical super - lattice@xcite and in a two - dimensional sonic crystal@xcite , respectively .
also simulation@xcite of dirac particles has been performed recently using trapped ions and laser excitations .
most recently , the zb in @xmath13rb bose - einstein condensate is observed experimentally @xcite using direct imaging technique . in this work
we consider the long standing problem of zb of heavy holes with cubic rashba interaction in a iii - v semiconductor quantum well subjected to a perpendicular magnetic field .
we study time - evolution of a heavy hole represented by the gaussian wave packet .
we visualize and discuss how the hole wave packet evolves with time around the cyclotron orbit .
it is shown that as time goes on , the initial wave packet starts to diffuse asymmetrically along the cyclotron orbit while executing cyclotron motion . at a later time , depending on the spin - orbit strength , the wave packet splits into two peaks with unequal amplitudes .
after many more cycles , the wave packet diffuses entirely along the cyclotron orbit .
then interference occurs between tail and head of the diffused wave packet , which randomizes the wave packet along the entire cyclotron orbit .
the zb of heavy holes in various physical quantities , such as position , charge current and spin angular momentum , are studied analytically as well as numerically .
it is revealed that the zb in these observables are very much dependent on the initial wave vector of the gaussian wave packet and the rashba spin - orbit interaction .
this paper is organized in the following way . in section
ii we present all relevant theoretical details such as the hamiltonian , landau levels , its corresponding eigenfunction and time - evolution of initial hole wave packet using green s function approach as well as average values of various physical observables . numerical results and discussion
are given in section iii .
we summarize our findings in section iv .
we present detailed calculations in appendix a.
the dynamics of holes in the @xmath14 valence band of iii - v semiconductors with zinc - blende structure like gaas are well described within the framework of the luttinger model ( lm ) . in the spherical approximation@xcite of lm ,
the topmost valence band is four fold degenerate at the @xmath2 point with hh and lh bands corresponding to total spin @xmath15 .
strong quantum confinement in the semiconductor heterostructure along the growth direction removes the @xmath2 point degeneracy between the hh and lh bands . at very low temperature and low density
only the hh subbands are assumed to be occupied .
now , it is possible to obtain a @xmath16-cubic rashba hamiltonian from the luttinger hamiltonian by projecting the latter onto the hh states .
it is well known@xcite that the rashba spin - splitting for hh and lh bands are proportional to @xmath17 and @xmath16 , respectively .
the linear spin - splitting for lhs is similar to that for electrons.@xcite our present study mainly deals with only hhs and hence linear spin - splitting for electrons and lhs is completely ignored . in presence of a perpendicular magnetic field
the single particle hamiltonian @xcite of a 2dhs with rashba soi ( rsoi ) can be written as @xmath18 where @xmath19 with @xmath20 is the canonical momentum operator , @xmath21 is the vector potential corresponding to the external magnetic field @xmath22 , @xmath23 , @xmath24 is the effective mass of the heavy hole , @xmath25 is the rashba coupling coefficient and @xmath26 s are the pauli matrices .
also , @xmath27 , @xmath28 and @xmath29 is the effective lande g - factor .
one important point to be mentioned here that the pauli matrices represent an effective pseudo - spin with spin projection @xmath30 along the growth direction of the quantum well . for convenience , we assume @xmath31 and the corresponding vector potential in the landau gauge as @xmath32 . the hamiltonian @xmath33 in eq .
( [ ham1 ] ) commutes with @xmath34 .
hence the wave vector @xmath35 is a good quantum number . in matrix form
the hamiltonian takes the following form : @xmath36 where @xmath37 is the oscillator hamiltonian and @xmath38 is the creation ( annihilation ) operator . here ,
@xmath39 with @xmath40 and @xmath41 is the magnetic length . also , @xmath42 is the cyclotron frequency , the dimensionless rashba parameter @xmath43 and @xmath44 .
the operations of @xmath45 and @xmath46 on the oscillator wave functions @xmath47 are @xmath48 , @xmath49 and @xmath50 . for @xmath51 the energy eigenvalues@xcite are given by @xmath52,\end{aligned}\ ] ] where @xmath53 represents two spin - split energy branches , @xmath54 and @xmath55 .
the eigenfunctions corresponding to the eigenvalues given in eq .
( [ eigen ] ) are @xmath56 here , @xmath57 is a set of two quantum numbers , @xmath58 with @xmath59
. for @xmath60 , the energy levels are @xmath61 and the corresponding eigenfunctions are @xmath62 in order to describe the time - evolution of an initial gaussian wave packet it is necessary to construct an appropriate propagator . for this purpose
we follow the well known green s function technique as described in ref .
[ ] . in matrix form , the propagator or the green s function can be written as @xmath63 the matrix elements of eq .
( [ greenm ] ) are defined as @xmath64 here , the indices @xmath65 represent the upper / lower components of the wave functions .
the hole wave function at a later time @xmath66 is given by @xmath67 , where @xmath68 s are given by eqs .
( [ wavefnn ] ) and ( [ wavefn0 ] ) . in presence of the perpendicular magnetic field
, we represent an initial state of the hole by a gaussian wave packet with the initial spin polarization along the @xmath69 axis as given by @xmath70 where @xmath71 is the initial momentum of the wave packet along @xmath72 direction .
note that the initial state coincides with the coherent state of a charge particle in a magnetic field .
we have taken such wave function because the dynamics of coherent states in a magnetic field resembles the dynamics of a classical particle . the choice of the two component spinor considered in eq .
( [ initial ] ) is completely arbitrary .
it can have both nonzero components as discussed in refs .
[ ] . in general the time - evolution of the initial wave packet can be obtained with the help of the green s function as given by @xmath73 using eq .
( [ greenm ] ) , ( [ initial ] ) and ( [ time - evo ] ) one can obtain @xmath74 it is straightforward to calculate the required matrix elements of the green s function and these are given by @xmath75 and @xmath76 where @xmath77 $ ] and @xmath78 with @xmath79 , @xmath80 and @xmath81 . note that @xmath82 and @xmath83 .
it is easy to verify an interesting result that @xmath84 .
the components of the wave packet at a later time @xmath66 are given by @xmath85 and @xmath86 where @xmath87 and @xmath88 with @xmath89 .
in this section we find the average values of the position and velocity operators .
the time - dependent average value of the position operator is given by @xmath90 it is straightforward to calculate the average values of @xmath72 and @xmath91 using eqs .
( [ wavefn1 ] ) and ( [ wavefn2 ] ) .
detailed calculations are given in appendix a. the average values of @xmath72 and @xmath91 are given by @xmath92 and @xmath93 where @xmath94 .
the velocity operator is obtained from the commutation relation @xmath95/(i \hbar)$ ] .
the components of the velocity operator are given by @xmath96 and @xmath97 the average value of the velocity operator is given by @xmath98 where the index @xmath16 represents the @xmath72 and @xmath91 components of the velocity .
using eqs .
( [ wavefn1 ] ) , ( [ wavefn2 ] ) and ( [ vel_exp ] ) , after a lengthy but straightforward calculation , we finally obtain average values of the components of the velocity operator as @xmath99 and @xmath100 now we turn to concentrate on the zb in the spin angular momentum of a heavy hole in the presence of a perpendicular magnetic field . in this sub - section
we shall calculate the time - dependent average values of the components of the effective spin operator @xcite @xmath101 .
the spatio - temporal profile of the hole spin density is defined as @xmath102 the time - dependent average value of the spin operator is given by @xmath103 the components of the average values of the spin operators are @xmath104 @xmath105 and @xmath106
in this section we shall visualize and discuss how the spatial distribution of probability density for the heavy holes changes with time .
we shall also discuss the time dependencies of the expectation values of the position , current and spin operators and their various consequences .
for numerical calculations , we adopt the material parameters appropriate for gaas quantum wells .
we took @xmath107 with @xmath108 as the free electron mass , @xmath109 m@xmath110 and @xmath111 t. it was reported that the value of the effective lande g - factor for heavy hole is highly anisotropic@xcite , we take its value @xmath112 for gaas system .
the magnitude of rashba strength ( @xmath25 ) depends explicitly on the external parameters@xcite like electric field , detail of confinement and hence can be tuned experimentally .
so in this study , we take various values of @xmath25 in such a way that the corresponding length scale @xmath113 is of the order of few angstroms . in fig .
( 1 ) we show the time - evolution of the probability density @xmath114 + @xmath115 of the heavy hole . to do this we numerically evaluate the components of the wave packet @xmath116 and @xmath117 at a later time @xmath66 as described by eqs .
( [ wavefn1 ] ) and ( [ wavefn2 ] ) . the infinite series in eqs . ( [ wavefn1 ] ) and ( [ wavefn2 ] )
converge approximately when @xmath118 for @xmath119 .
however the convergence of these infinite series also depends on the value of @xmath120 . for larger @xmath120 larger
@xmath121 is required .
we have checked that @xmath122 is appropriate for the convergence when @xmath123 .
1(a)-1(i ) are plotted for @xmath124 and @xmath125 , respectively .
initially the wave packet is situated at the origin .
we know that in a perpendicular magnetic field a charge particle ( without soi ) moves in a cyclotron orbit with time period @xmath126 .
the radius of this circular orbit is @xmath127 .
the presence of soi modifies @xmath128 and @xmath129 marginally . in presence of the rsoi
the hole wave packet starts to diffuse asymmetrically around the cyclotron orbit while making circular motion as time goes on .
the diffused wave packet is making circular motion with time period @xmath130 and radius @xmath131 . at @xmath132 [ fig 1(c ) ]
the hole wave packet splits into two unequal peaks which are rotating with different velocities along the cyclotron orbit .
the difference in the peak amplitudes nearly vanishes around @xmath133 [ fig .
1(f ) ] . at the same time
, the wave packet diffuses almost along the entire cyclotron orbit .
now the interference effect begins to occur between the tail and head of the diffused wave packet and as a result a completely randomized pattern of hole wave packet is observed after some more cycles as shown in fig .
1(i ) . recalling the case of two - dimensional electron system ( 2des ) with linear rsoi in a perpendicular magnetic field where this kind of splitting of wave packet occurs around @xmath134@xcite for @xmath135 t. but in the present case of 2dhs
this splitting occurs at much lesser time than the 2des case .
moreover , the wave - packet does not diffuse along the circular orbit in the case of 2des with linear rashba term .
these features can be attributed to the cubic rashba term in the 2dhs . ,
( b ) @xmath136 , ( c ) @xmath132 , ( d ) @xmath137 , ( e ) @xmath138 , ( f ) @xmath133 , ( g ) @xmath139 , ( h ) @xmath140 , and ( i ) @xmath141 . in this case
@xmath25 is fixed in such a way that @xmath142 nm . in each plot
the color bar represents the dimensionless parameter @xmath143.,width=415 ] for different values of @xmath144 namely ( a ) @xmath142 nm , ( b ) @xmath145 nm , ( c ) @xmath146 nm , and ( d ) @xmath147 nm . in this case magnetic field and the initial wave vector are fixed to @xmath111 t and @xmath119 . in each plot
the color bar represents the dimensionless parameter @xmath143.,width=453 ] in fig . 2 spatial distribution of probability density at time @xmath132 is shown .
different panels are plotted for different values of @xmath25 such that @xmath142 nm , @xmath148 nm , @xmath149 nm , and @xmath150 nm . it is clear that as @xmath25 increases the wave packet diffuses very fast and covers the entire cyclotron orbit .
the diffusion rate of the wave packet increases with increase of the spin - orbit strength .
also strong @xmath25 helps to expedite the splitting and the randomization of the hole wave packet . ,
( b ) @xmath151 , ( c ) @xmath152 , and ( d ) @xmath153 . here in each panel solid and
dashed lines correspond to @xmath154 and @xmath155 , respectively . here
@xmath156 is defined as @xmath157.,width=434 ] , ( b ) @xmath151 , ( c ) @xmath152 , and ( d ) @xmath153 . here in each panel
dotted , solid and dashed lines correspond to @xmath142 , @xmath158 , and @xmath159 nm , respectively . here
we define @xmath157.,width=415 ] in fig .
3 we plot the average values of the position and current operators in @xmath72 and @xmath91 directions as a function of time @xmath66 for a fixed value of magnetic field @xmath111 t and fixed @xmath25 such that @xmath142 nm . here , we define the current operators as @xmath160 with @xmath161 , @xmath91 . in this case
we consider two different values of initial wave vector @xmath154 ( solid line ) and @xmath119 ( dashed line ) . when @xmath154 , oscillations appearing in @xmath162 and @xmath163 are persistent in time .
but in the case of higher @xmath120 , the amplitude of zb decreases at later time which shows some kind of localization of zb oscillations .
this is because higher landau levels are involving for higher values of @xmath120 . comparing figs .
3(a ) and 3(b ) we mention that @xmath164 is oscillating about zero for both values of @xmath120 whereas @xmath151 is oscillatory but always positive because eq .
( [ posy ] ) contains a constant term .
there is also a definite phase difference between @xmath165 and @xmath166 and @xmath167 and @xmath168 as clearly shown in fig .
3 . investigating all the graphs in fig .
( 3 ) one can conclude that although initially there is no phase difference but an increment in @xmath120 introduces a phase difference at later times .
figure 4 describes the time dependence of position and current operators for a fixed magnetic field @xmath111 t and @xmath119 .
different values of @xmath25 such that @xmath142 nm ( dotted ) , @xmath158 nm ( solid ) , and @xmath159 nm ( dashed ) have been considered .
it can be seen from fig .
4 that the amplitude of zb oscillation decreases as @xmath25 increases .
one interesting point is to be mentioned here that at stronger @xmath25 ( @xmath169 nm ) the zb oscillations start to reappear at large time . and @xmath91 components of the spin operator : ( a ) @xmath154 , ( b ) @xmath119 , and ( c ) @xmath123 . here in each panel solid and dashed lines represent @xmath170 and @xmath171 , respectively which are plotted in units of @xmath172.,width=415 ] component of spin operator : ( a ) @xmath173 , ( b ) @xmath154 , ( c ) @xmath119 , and ( d ) @xmath123 .
here in each panel @xmath174 is plotted in units of @xmath172.,width=415 ] the time - dependent average values of the spin components are shown in fig . 5 and 6 .
we consider various cases corresponding to the different values of the initial wave vector , namely @xmath173 , @xmath175 , @xmath155 and @xmath176 as mentioned in both figures .
figure 5 describes the variations of @xmath170 and @xmath171 with time @xmath66 for different values of @xmath120 .
it can be seen that @xmath170 and @xmath171 maintains same oscillatory pattern apart from a definite phase difference .
an increase in @xmath120 results a change in the oscillatory pattern significantly .
the amplitude of the zb oscillations , appearing in @xmath170 and @xmath171 increases as @xmath120 increases .
we have shown the time dependence of @xmath174 in fig .
it should be noted that @xmath174 does not vanish when there is no initial momentum i.e. @xmath173 .
this fact can be confirmed from eq .
( [ spinz ] ) which contains a non - zero term @xmath177 .
as @xmath120 increases the oscillatory pattern of @xmath174 changes abruptly .
when @xmath154 a beating - like pattern appears in the oscillations of @xmath174 . from fig
6(b ) , ( c ) , ( d ) it can be seen that the number of oscillations before the first beating node decreases drastically as @xmath120 increases . this complicated oscillatory pattern can be attributed to the fact that the infinite series in eqs .
( [ spinx]-[spinz ] ) converges for higher values of @xmath121 as @xmath120 increases . as a result
more and more frequencies appear in the oscillations because of the involvement of higher landau levels . comparing fig .
5 and 6 at @xmath178 @xmath179 where as @xmath180 and this is consistent with the fact that the initial wave packet is polarized along the @xmath69-direction .
the jittery motion in @xmath181 induces jittery motion in @xmath182 and @xmath183 .
in summary , we have studied wave packet propagation and @xmath0 of a heavy hole in iii - v semiconductor quantum wells .
we have visualized and discussed various consequences of the time - evolution of the hole wave packet along the cyclotron orbit .
the hole wave packet diffuses asymmetrically along the circular orbit while making cyclotron motion .
it is shown that the hole wave packet splits into two peaks with unequal amplitudes at certain time which depends on the spin - orbit interaction strength .
the two peaks rotate with different frequencies .
the amplitude of the two peaks become nearly equal as time goes on . after many cycles , tail and head of the diffused wave packet interfere with each other and produces a complete randomized pattern .
the diffusion rate of the wave packet increases with increase of the spin - orbit interaction strength .
also strong spin - orbit coupling expedite the splitting and the randomization of the wave packet .
our results for the hole is compared with an electron in a linear rashba system in presence of the magnetic field .
we have also studied zb phenomenon in position , current and spin angular momentum of the heavy hole .
the @xmath0 oscillations are very much sensitive to the initial momentum of the wave packet and the rashba spin - orbit coupling constant .
in this appendix we shall present detailed derivation of the average values of various physical observables like position and velocity operators .
the calculation of the average value of the @xmath91 component of the position operator is easier than that of the @xmath72 component .
let us first calculate average value of the @xmath72 component .
equation ( [ position ] ) can be written as @xmath184 , where @xmath185 .
the explicit form of @xmath186 is @xmath187 where @xmath188 , @xmath189 with @xmath190 . the above equation [ eq . ( [ posx1 ] ) ]
can be re - written as @xmath191 | in this work we study wave packet dynamics and @xmath0 , an oscillatory quantum motion , of heavy holes in iii - v semiconductor quantum wells in presence of a quantizing magnetic field .
it is revealed that a gaussian wave - packet describing a heavy hole diffuses asymmetrically along the circular orbit while performing cyclotron motion .
the wave packet splits into two peaks with unequal amplitudes after a certain time depending on spin - orbit coupling constant .
this unequal splitting of the wave packet is attributed to the cubic rashba interaction for heavy holes .
the difference in the peak amplitudes disappears with time . at a certain time
the two peaks diffuse almost along the entire cyclotron orbit . then tail and head of the diffused wave packet interfere and as a result a completely randomized pattern of the wave packet
is observed .
the diffusion rate of the wave packet increases with increase of the spin - orbit interaction strength .
also strong spin - orbit coupling expedite the splitting and the randomization of the wave packet .
we also study the @xmath0 in various physical observables such as position , charge current and spin angular momentum of the heavy hole .
the @xmath0 oscillations are very much sensitive to the initial wave vector of the gaussian wave packet and the strength of the rashba spin - orbit coupling . |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.