article
stringlengths 0
745k
| abstract
stringlengths 0
92.7k
|
|---|---|
the processing of information based on the laws of quantum physics @xcite has become a very active field of research during the last decade . for the experimental demonstration of fundamental key results of quantum information theory ,
ion - trap based systems have played a major role .
the success of ion trap experiments can be attributed to the fact that encoding quantum information in either hyperfine ground or meta - stable excited atomic states provides well - defined quantum bits ( qubits ) with long coherence times .
the use of lasers for manipulating the qubit state allows for precisely switchable interactions with low decoherence rates .
the fundamental operations of qubit initialization , arbitrary single qubit manipulation and quantum state - detection have already been used in atomic clocks with single ions for many years .
contrary to other realizations @xcite of quantum information processing , the most demanding operation in ion traps consists in the realization of an entangling gate operation . because of the repulsive coulomb force , the inter - ion distance is orders of magnitude bigger than the characteristic length scale of any state - dependent interaction between ions in ground or low - lying excited states . in all current experiments creating entangled ions @xcite ,
gate operations rely on interactions that are mediated by the vibrational degrees of freedom of the ion string .
these gate operations fall into two categories : 1 .
quantum gates induced by a laser beam that interacts with a single ion at a time as originally proposed in the seminal paper by i. cirac and p. zoller @xcite and later realized by the innsbruck ion trapping group @xcite . in these gates , a single ion is entangled with a vibrational mode @xcite of the ion string and the entanglement is subsequently transferred from the vibrational mode to the internal state of a second ion .
2 . quantum gates induced by a bichromatic laser that collectively interacts with two or more ions . here , a vibrational mode becomes transiently entangled with the qubits before getting disentangled at the end of the gate operation , resulting in an effective interactions between the qubits capable of entangling them .
gates of this type were first proposed by g. milburn @xcite , a. srensen , k. mlmer @xcite and e. solano @xcite , and subsequently realized by ion trapping groups in boulder , ann arbor and oxford @xcite .
even though both classes of gates are applicable to hyperfine qubits as well as optical qubits ( i.e. qubits encoded in hyperfine states or in states linked by a dipole - forbidden transition with an optical wavelength ) , current experiments with optical qubits have relied on the former and experiments with hyperfine qubits on the latter type of interaction . in any case , the main goal consists in demonstrating fast operations creating entanglement with high fidelity .
the purpose of the present paper is to discuss bichromatic gate operations with a focus on implementations using an optical transition .
it turns out that for optical transitions , gate operations are achievable by illuminating the ions with an amplitude - modulated laser beam that is resonant with the qubit transition .
the paper is organized as follows : section [ sec : bichromaticgates ] reviews different methods of realizing bichromatic quantum gates and discusses properties that are specific to their application to optical qubits . in section [ sec : effectivehamiltonian ] , an effective hamiltonian for the laser - ion interaction will be derived by going into a reference frame rotating at non - uniform speed in order to eliminate non - resonant excitations of the qubit transition that do not couple to the vibrational mode . in this way
, it will be shown that for a single ion qubit the interaction is well described by a hamiltonian @xmath2 where the coupling strength @xmath3 is proportional to the laser intensity in the limit of low intensities but starts to saturate at higher intensities and where @xmath4 is a component of the pauli spin operator @xmath5 coupling to a vibrational mode of the ion described by creation and annihilation operators @xmath6 , @xmath7 .
furthermore , it will be shown that @xmath8 depends not only on the particular type of gate operation but also on the laser intensity and the relative phase between the two frequencies of the bichromatic field
. equations ( [ heffzz ] ) , ( [ heffms ] ) , ( [ mspropagator2 ] ) describing the action of the gates based on @xmath0 and on @xmath1 ( mlmer - srensen ) interactions are the key results of the paper . for the mlmer - srensen gate , the result will be compared to the analysis presented in @xcite .
in addition , the performance of @xmath0 and @xmath1 gates will be compared .
section [ sec : shapedpulses ] shows how to use pulse - shaping of the laser intensity as well as spin echo techniques to make the gates more robust against fluctuations of the control parameters .
the hamiltonian @xmath9 describes a harmonic oscillator oscillating at frequency @xmath10 and driven by a force with frequency @xmath11 and coupling strength @xmath12 . going into an interaction picture defined by @xmath13 yields the hamiltonian @xmath14 where @xmath15 . under the action of the driving force ,
an oscillator that is initially in a coherent state remains in a coherent state . for a force that is slightly detuned from resonance , the coherent state maps out a circle in phase space and returns to the initial state after a period @xmath16
this operation multiplies the oscillator state by a phase factor whose magnitude is given by the ratio of the strength of the force and the detuning as shown in @xcite . in order to allow for variations of the driving field s strength
, we generalize the hamiltonian to @xmath17 and calculate its propagator @xmath18 by using the baker - campbell - hausdorff relation @xmath19 for the displacement operator @xmath20 . for the propagator , we find @xmath21 where @xmath22 , @xmath23 and @xmath24 in case of a driving force with constant amplitude , @xmath25 , one obtains @xmath26 and @xmath27 . after a time @xmath28 , @xmath29 ,
the coherent state returns to its initial state in phase space with its phase changed by an amount @xmath30 . by making this phase change depend on the internal states of a pair of ions , an entangling gate operation can be achieved . for @xmath31 where @xmath32 is an operator acting on the qubit states , the propagator ( [ propagatordrivenosc ] ) is replaced by @xmath33 choosing the interaction time @xmath34 such that @xmath35 thus realizes a propagator that depends nonlinearly on @xmath32 and does not alter the vibrational state .
the interaction of a single ion qubit resonantly excited by a monochromatic laser field with frequency @xmath36 is usually described by performing a rotating - wave approximation with respect to the optical frequency to obtain the hamiltonian @xmath37 here , @xmath38 is the detuning from the qubit transition frequency @xmath39 , @xmath10 is the frequency of the ion s vibrational mode of interest , and @xmath40 with the pauli matrices @xmath41 . the strength of the laser - ion coupling is characterized by the rabi frequency @xmath12 , and the strength of processes involving changes in the vibrational state is determined by the value of the lamb - dicke parameter @xmath42 .
equation ( [ laserionhamiltonian ] ) represents a hamiltonian in an interaction picture that is defined with respect to the hamiltonian @xmath43 describing the ion qubit in the absence of any laser - ion interactions . if @xmath44 , the lamb - dicke approximation @xmath45 is used to simplify ( [ laserionhamiltonian ] ) .
the resulting three terms describe excitations on the carrier , the lower and the upper motional sideband , respectively .
the generalization of the hamiltonian to the case of two and more ions is straightforward . for the sake of simplicity
, calculations in section [ sec : effectivehamiltonian ] will be limited to the case of the laser coupling to the centre - of - mass mode along the axis of the ion string where all ions experience the same coupling strength .
a detailed account of laser - ion interactions is given in @xcite .
a hamiltonian as described by ( [ drivenoscillator ] ) was employed for the first time in an experiment @xcite creating a schrdinger cat state with a single ion using @xmath46 , i.e. a coupling to the motional mode that depended on the internal energy eigenstate of the ion .
later , it was realized that the same type of coupling could be used to entangle a pair of ions by performing a conditional phase gate @xcite . in the experimental realizations
@xcite , spin - dependent forces acting on a pair of hyperfine or zeeman ground states have been realized by near resonant driving of raman transitions between vibrational states ( see figure [ fig1:motionalcoupling_hyperfine ] ) .
for this purpose , two non - copropagating laser beams with frequencies @xmath47 form a moving standing wave with difference frequency @xmath48 close to the frequency @xmath10 of a vibrational mode .
the ac - stark shift of the qubit states @xmath49,@xmath50 results from a non - resonant coupling to another atomic state @xmath51 that is made qubit state - dependent by properly chosen polarizations .
since the laser field exhibits a strong spatiotemporal modulation , the resulting potential gradients induce a force acting on the qubits that is state - dependent and that couples to the vibrational mode by displacing the qubit along a circle in phase space .
when the ions couple to the centre - of - mass mode ( stretch mode ) , the coupling to the mode can be made to disappear when both qubits are in the same quantum state by choosing an ion spacing that is an odd ( even ) integer multiple of half the wavelength of the moving standing wave . disregarding unimportant global and single qubit phases ,
this coupling is then described by an operator @xmath52 where @xmath53 is a collective spin component of the qubits .
are coupled via the excited state @xmath51 . a similar coupling not shown in the figure exists for the qubit state @xmath49 .
the laser detunings @xmath54 from the mediating states are large compared with trap frequency @xmath10 and the qubit level spacing @xmath39 to avoid spontaneous emission from state @xmath51 .
constructive interference of paths 1 and 2 is achieved for counter - propagating laser beams .
( b ) the coupling is maximized for counter - propagating laser beams with frequencies @xmath55 , @xmath56 forming a moving standing wave along the axis of vibration of the motional mode of interest.,width=415 ] the situation is different for qubits encoded in atomic states connected by a narrow optical transition . for coupling motional states @xmath57 , here , the other qubit state @xmath50 serves to mediate the coupling .
similarly , a coupling between the states @xmath58 is mediated by the state @xmath49 . to achieve a strong coupling , a detuning @xmath59 from the intermediate states
can be chosen that is smaller than the transition frequency between vibrational states provided that the decay rate of the metastable state is small compared to @xmath10 .
for @xmath60 , the two interfering paths shown in figure [ fig1:motionalcoupling_optical]a connecting levels @xmath57 have equal strength . since the detunings from the mediating states now have opposite signs , destructive interference is achieved for counter - propagating beams whereas the coupling is maximized for co - propagating beams . in the limit of small excitation ( @xmath61 ) ,
the coupling strength @xmath62 on the raman transition between @xmath63 and @xmath64 is given by @xmath65 .
the states @xmath66 and @xmath67 are coupled with equal strength but opposite sign . for stronger excitation ,
the carrier transition is non - resonantly excited which leads to a saturation of @xmath62 .
as long as the intensities of the bichromatic beams are equal , there is no overall ac - stark shift due to excitation of the carrier transition and the first motional sidebands because ac - stark shifts caused by the two laser fields exactly cancel each other . in the case of two ions excited on the centre - of - mass mode
, a driven quantum mechanical oscillator is realized with collective atomic oscillator @xmath52 .
in addition to the coupling of vibrational states , there is another small mlmer - srensen coupling @xcite that does not exist for the case of hyperfine or zeeman qubits : collective spin flips between the states @xmath68 and @xmath69 occur by processes involving a blue and a red photon that are mediated by the states @xmath70 and @xmath71 ( see figure [ fig : msgate_scheme ] ) . a similar process involving either two blue or two red photons couples the states @xmath72 and @xmath73
. is mediated by the states @xmath74 and vice versa on the narrow qubit transition .
since spontaneous scattering from the mediating state is small , the detuning can be made small compared to the trap frequency . for @xmath75 , the coupling is maximized by choosing a copropagating beam geometry .
( b ) optimum coupling is achieved for co - propagating laser beams with frequencies @xmath55 , @xmath56 propagating along the axis of vibration of the motional mode of interest . as the bichromatic laser field
could also be described by a monochromatic laser field that is amplitude - modulated with a frequency @xmath48 close to the vibrational frequency , no spatially varying ac - stark shifts are involved in the coupling . , width=453 ] in contrast to @xmath0 gates that do not change the internal states of the ions , the @xmath1 gate operations first investigated by a. srensen , k. mlmer @xcite and others @xcite relies on collective spin flips @xmath76 , @xmath77 by processes coupling to the lower and upper motional sidebands as illustrated in figure [ fig : msgate_scheme ] .
satisfying @xmath78 is tuned close to the upper and lower motional sideband of the qubit transition .
the field couples the qubit states @xmath79 via the four interfering paths shown in the figure .
similar processes couple the states @xmath80 with the same strength provided that the rabi frequencies of the light fields @xmath47 are equal.,width=302 ] it can be shown that the hamiltonian governing the action of the gate is described by setting @xmath81 @xcite . for a properly chosen coupling strength , the gate operations maps the product state basis @xmath82 onto a basis of entangled states . for hyperfine qubits , a detailed discussion of advantageous beam geometries is presented in @xcite . in the case of optical qubits
, it is again possible to choose a pair of co - propagating beams for performing the gate operation .
the only difference to the @xmath0 gate consists in the choice of laser frequencies @xmath83 required for achieving a resonant coupling .
formally , the gate operation is equivalent to a @xmath0 interaction in a rotated basis . to stress this analogy ,
the mlmer - srensen gate operation is often also called a @xmath1 gate .
the possibility of choosing co - propagating laser beams for performing either @xmath0 or @xmath1 gates is attractive from an experimental point of view .
the light field could be generated by passing a laser beam through an acousto - optical modulator driven by two radio - frequency fields and subsequently coupling the first - order diffracted beams into a single - mode optical fibre , thus realizing a simple and stable setup .
gate operations could also be achieved by phase - modulating a laser with a modulation frequency close to @xmath10 by choosing a modulation index where the carrier strength vanishes . however , this approach has the strong disadvantage that even small changes of the modulation index from the desired value give rise to light resonant with the transition which is having desastrous effects on the gate performance .
similarly , modulating at frequency close to @xmath84 and using the carrier and one of the first sidebands is problematic because of light shifts induced by the other sidebands . ]
if the rabi frequencies @xmath85 , @xmath86 of the blue and the red detuned laser beam are equal , light shifts due to the non - resonant excitation of the carrier transition and the first - order sidebands are exactly cancelled .
light shifts arising from coupling to other zeeman transitions or far - detuned dipole transitions could be cancelled by suitably balancing the ratio @xmath87 .
we are interested in deriving an effective hamiltonian that accurately describes the dynamics on a pair of optical qubits induced by a co - propagating bichromatic laser field with frequencies @xmath88 , where the detuning is either close to half the vibrational frequency or close to the vibrational frequency , i. e. @xmath89 or @xmath90 with @xmath91 . as optical qubits interacting with lasers typically have smaller lamb - dicke parameters than hyperfine qubits coupled by raman transitions , transient non - resonant excitation of the carrier transition is expected to play an important role for @xmath92 . whereas usually non - resonant interactions are taken into account only qualitatively after having derived an effective hamiltonian , in the following calculation they will be eliminated right at the beginning by going into a reference frame rotating at non - uniform speed .
the hamiltonian for the bichromatic laser field we are interested in , is given by @xmath93 the laser is assumed to interact collectively with @xmath94 ions on the axial centre - of - mass mode .
here , @xmath95 , and @xmath7 , @xmath6 denote operators annihilating and creating phonons .
it is also possible to interpret this interaction as being due to a single resonant laser beam that is amplitude - modulated with modulation frequency @xmath96 .
the optical phase of the laser field is denoted @xmath97 , and the phase @xmath98 accounts for a time difference between the start of the gate operation and the maximum of the amplitude modulation on the laser beam . using the picture of an amplitude - modulated resonant beam ,
it is obvious that there are fast dynamical processes on the carrier transition with a periodicity given by @xmath16 that excite the ions to the other state in the first half of the period and transfer it back to the original state in the second half .
we are not really interested in exactly calculating the dynamical evolution of the quantum state on this fast time scale .
rather , we would like to know the time evolution at the instances @xmath99 .
it is useful to rewrite ( [ bichrham ] ) as @xmath100 where we used the displacement operator @xmath101 and the definitions @xmath102 , @xmath103 , @xmath104 and @xmath105 . for @xmath106 or @xmath107 as required by either @xmath0 or @xmath1 interactions , the hamiltonian is periodic in time , i. e. @xmath108 with period @xmath16 .
note that we assume the two - photon couplings to be strictly resonant for the moment . in the next step
, we will get rid of the fast non - resonant carrier oscillation by going into another interaction picture defined by @xmath109 .
writing @xmath110 we obtain the interaction hamiltonian @xmath111 where @xmath112 now , we can approximate the time evolution over the course of an oscillation period by a magnus expansion of the propagator @xcite in order to obtain @xmath113 + \ldots\right)\right\}.\label{bakercampbellhausdorffpropagator}\ ] ] from now on , the phase @xmath97 will be set to zero to simplify the notation .
the formulas in the remainder of section [ sec : effectivehamiltonian ] are easily generalized to the case of arbitrary @xmath97 by making the replacements @xmath114 and @xmath115 . in the following two subsections , effective hamiltonians for the @xmath0 gate and the @xmath1 gate will be derived starting from ( [ bakercampbellhausdorffpropagator ] ) .
we are interested in obtaining hamiltonians of the form given in ( [ drivenoscillator ] ) that are valid in the regime @xmath61 .
in addition , the calculation is going to yield correction terms for the case when @xmath61 no longer strictly holds and additional terms that do not commute with the atomic operator @xmath116 in ( [ drivenoscillator ] ) . towards this aim , terms proportional to higher orders of the expansion parameter @xmath117
will be dropped . in the calculation , bessel functions @xmath118
will be evaluated at @xmath119 .
these functions will be kept till the end of the calculation and expanded in @xmath120 only for the final analysis . for the @xmath0 gate , we set @xmath106 .
we start by calculating the first term @xmath121 appearing in the exponent of ( [ bakercampbellhausdorffpropagator ] ) . here , it is important to note that the integrand @xmath122 is a periodic function of time with period @xmath16 so that all its fourier components except the constant term @xmath123 will average to zero when they are integrated over one period .
the only non - zero fourier components of the function @xmath124 are @xmath125 and @xmath126 . since all non - zero fourier components of @xmath127 are even integer multiples of @xmath96 , the @xmath128-term of @xmath129 averages to zero .
therefore , we obtain @xmath130 in the lamb - dicke limit , @xmath131 and the components @xmath132 and @xmath133 are the only relevant ones .
finally , we have @xmath134\\ & = & ae^{i\frac{4\omega}{\delta}\sin(\delta t+\zeta)}+a^\dagger e^{-i\frac{4\omega}{\delta}\sin(\delta t+\zeta)}\\ & = & \sum_{n=-\infty}^\infty ( aj_n(\frac{4\omega}{\delta})+ a^\dagger j_n(-\frac{4\omega}{\delta}))e^{in\zeta}e^{in\delta t}\\ & = : & \sum_{n=-\infty}^\infty a_n e^{in\delta t}\,,\end{aligned}\ ] ] where @xmath135 is a bessel function , @xmath136 with @xmath137 and @xmath138 . in the following , the argument @xmath139 of the bessel functions @xmath135 will often be dropped to keep the notation simple .
it is convenient to express @xmath140 as @xmath141 note that the linear transformation ( [ rotatedops ] ) preserves the usual lie algebra commutation relations for the operators @xmath142 .
the four terms @xmath143,@xmath144 , and to a lesser degree @xmath145 , @xmath146 , contribute to @xmath147 . evaluating @xmath148 as well as the other terms , we arrive at the effective hamiltonian @xmath149 where @xmath150 .
this hamiltonian describes a spin - dependent force that starts to saturate when the rabi frequency goes up .
while the atomic operator @xmath151 coincides in the limit of weak excitation with the operator @xmath152 obtained from second - order perturbation theory , it depends on the phase @xmath98 between the blue- and the red - detuned laser beams in the limit of strong excitation . for the periodic hamiltonian @xmath153 the second order contribution to the effective hamiltonian @xmath147
is given by @xmath154\ , .
\label{heffii2}\ ] ] after evaluating the commutators @xmath155 $ ] , @xmath156 $ ] , the effective hamiltonian @xmath157 is obtained ( the contribution of the commutator @xmath158\propto(\eta\omega)^2(\omega/\delta)^6 $ ] is insignificant ) .
if the detuning @xmath89 slightly deviates from half the oscillation frequency @xmath10 , the hamiltonian is given by @xmath159 is obtained .
the second order term @xmath160 account for collective spin flip processes caused by a mlmer - srensen interaction .
if this interaction did not exist , the propagator could be calculated in the same way as for the driven harmonic oscillator described by ( [ drivenoscillator ] ) . in the limit @xmath61 , where @xmath161 , the time evolution from t=0 to @xmath162 would create a mapping of quantum states
@xmath163 described by the operator @xmath164 with @xmath165 for @xmath166 ions , the operator @xmath167 performs a conditional phase gate if @xmath168 . for @xmath169 and weak excitation ( @xmath61 )
, this requires setting the coupling strength @xmath170 with @xmath171 in the limit where @xmath61 no longer holds , saturation effects reduce the geometric phase @xmath172 picked up in the gate operation . for @xmath170 , we would now have @xmath173 where @xmath174 counts the number of trap cycles during the gate operation . for @xmath175 and a gate time of 100 trap cycles ,
@xmath172 is reduced by about @xmath176 .
the smaller the lamb - dicke factor gets , the more important saturation effects become for a given gate time .
the mlmer - srensen interaction contributes a term to the propagator @xmath167 which is now approximately described by @xmath177 with @xmath178 for the ratio @xmath179 , @xmath180 if @xmath44 , the contribution from the second term @xmath181 is comparatively small . up to now
, we have disregarded the fact that the effective hamiltonian is valid only for times @xmath182 where @xmath183 .
therefore , the gate time @xmath184 needs to fulfil @xmath185 as well as @xmath186 , with integer @xmath187 . combining both conditions , we find @xmath188 in writing equation ( [ approxpropagatorzz ] ) , terms arising from the non - vanishing commutator @xmath189 $ ] were neglected .
using the abbreviations @xmath190 and @xmath191 , it is convenient to rewrite @xmath192 with @xmath193 since @xmath194 and @xmath195 commute .
the time evolution induced by @xmath194 is given by the propagator @xmath196 with @xmath197 and @xmath198 , and for the interaction hamiltonian @xmath199 one finds @xmath200 here , the displacement operator @xmath201 was expressed by the real - valued operators @xmath202 and @xmath203 .
for the special case @xmath169 this is equivalent to @xmath204 the last expression shows that the interaction hamiltonian @xmath205 describes collective spin flips between the levels @xmath206 and @xmath207 that go along with displacements of the vibrational state . for a phase gate operation , @xmath208 .
minimum uncertainty states of motion are not conserved by the interaction .
the formalism developed so far can be employed to study the mlmer - srensen gate as hamiltonian ( [ bichrham ] ) also describes the bichromatic laser field of the mlmer - srensen gate .
since the laser frequencies are set close to the blue and red sideband resonance , the only difference is that @xmath90 instead of @xmath183 , thus changing the values of the fourier components @xmath209 used to express @xmath210 .
taking into account the leading terms in first and second order for the calculation of ( [ bakercampbellhausdorffpropagator ] ) , one finds the effective hamiltonian @xmath211 instead of ( [ heffzz ] ) .
integrating from @xmath212 to @xmath213 and neglecting commutators involving @xmath214 in the magnus expansion , yields the propagator @xmath215 with @xmath216 the contribution @xmath217 comes from the counter - rotating term , the red laser coupling to the blue sideband and vice versa . for weak excitation , we have @xmath218 for @xmath166 ions , an entangling gate operation is achieved by setting @xmath219 which amounts to setting the coupling strength @xmath12 to @xmath220 in the limit where @xmath61 is no longer valid but where @xmath221 still holds , we find for the correction terms in ( [ lambdat ] ) when keeping @xmath170 @xmath222 and for the ratio @xmath223 for @xmath175 and a gate operation that is performed within 100 trap cycles , the correction terms to @xmath224 have a relative strength of @xmath225 and @xmath226 , respectively , and the @xmath227 interaction is less that @xmath228 of the @xmath229 term . therefore , the interaction is quite well approximated by using the propagator ( [ mspropagator ] ) with @xmath230 set to zero .
then , one obtains for arbitrary @xmath231 @xmath232 where @xmath233 in the reference frame of the original hamiltonian ( [ bichrham ] ) , the laser - ion interaction is therefore well described by the propagator @xmath234 undergoing a mlmer - srensen interaction with @xmath235 , @xmath175 , @xmath236 and @xmath169 .
the calculations are based on ( [ mspropagator2 ] ) .
the values chosen reproduce the curves shown in figure 3(b ) of reference @xcite . counting from above at @xmath237 ,
the curves represent the populations @xmath238 , @xmath239 and the coherences @xmath240 and @xmath241 . at @xmath242 ( @xmath243 ) ,
the ions are in a maximally entangled state .
( b ) same as ( a ) but with @xmath244 .
if the gate operation starts in an intensity minimum of the amplitude - modulated laser beam , the non - resonant carrier oscillations are much stronger . at @xmath245
, the quantum state is no longer maximally entangled .
( c ) fidelity @xmath246 of creating the maximally entangled state @xmath247 near the optimum calculated from ( [ mspropagator2 ] ) .
the upper curve corresponds to @xmath169 , the lower one to @xmath244 .
the points on top of the upper curve represent the fidelity for @xmath169 and were obtained by a numerical integration of the hamiltonian ( [ bichrham ] ) after applying the lamb - dicke approximation .
( d ) infidelity @xmath248 of the gate at @xmath245 for @xmath169 and a state with @xmath249 .
the solid line is a numerical integration of ( [ bichrham ] ) in the lamb - dicke approximation , the dash - dotted line is based on the full hamiltonian .
the arrow labelled @xmath250 denotes the optimum rabi frequency predicted by ( [ omegams ] ) , @xmath251 the value of @xmath12 chosen in @xcite , @xmath3 the rabi frequency predicted by ( [ lambda]).,width=529 ] this propagator can be used to calculate the dynamics of expectation values of interest for the qubits .
it is possible to derive simple expressions by tracing over the motional states if the vibrational mode is in a thermal state . for this
, it is useful to note that @xmath252 where @xmath253 denotes the projector onto the subspace spanned by eigenvectors of @xmath254 with eigenvalue @xmath255 .
moreover , the diagonal elements of the displacement operator in the number - state representation are given by @xmath256 where @xmath257 denotes a laguerre polynomial @xcite . since
the generating function of @xmath258 is given by @xcite @xmath259 summation over a thermal state with number state population @xmath260 and mean phonon number @xmath261 simply yields @xmath262 in the case of two ions , @xmath169 , and an initial qubit state @xmath263 , the expectation value @xmath264 of the observable @xmath32 is given by @xmath265 where @xmath266 refers to the trace of the qubit state space and @xmath267 with @xmath268 and @xmath269 . as an example , the time evolution of @xmath270 is explicitely given by @xmath271 with @xmath272,@xmath273,@xmath274 containing the time dependent terms .
other quantities of interest could be calculated in the same way .
a propagator similar to ( [ mspropagator2 ] ) was calculated in ref .
@xcite for the case @xmath169 .
the authors argued that the non - resonant excitation of the carrier transition could be neglected in a first step and obtained in this way a hamiltonian of the type described by ( [ drivenoscillator ] ) that could be integrated exactly . in a second step , they considered the influence of the previously neglected non - resonant excitations . while this treatment yields correct results for @xmath169 , it fails to predict the dependence of the gate operation on @xmath98 via the angle @xmath275 as given by ( [ mspropagator2 ] ) .
figure [ fig3:ms2000gate_comparison ] shows the time evolution of matrix elements for the same parameters as used in @xcite for the cases @xmath169 and @xmath244 . in the latter case , the amplitude of the non - resonant carrier oscillations is considerable and the input state @xmath276 is never perfectly mapped to a maximally entangled state . for @xmath277 , the effect of a non - zero value of @xmath275
is fairly small for current gate realizations using hyperfine qubits where the lamb - dicke parameter @xmath42 is considerable .
however , it becomes crucial for the realization of fast gates on optical qubits with small @xmath42 since in this case the gate requires a larger value of @xmath12 to achieve the same gate speed .
gate as a function of the phase @xmath98 for a gate operation taking place in 25 trap cycles with @xmath278 , @xmath279 and @xmath280 .
the figure shows the distance @xmath281 between the exact propagator @xmath282 obtained by numerical integration of ( 5 ) and the gate operation @xmath283 ( solid line ) .
the points on top of the line denote @xmath284 where @xmath285 is the propagator given by ( 27 ) , thus demonstrating that this equation is a very good approximation to the exact solution .
the dashed - dotted line shows @xmath281 with @xmath286 and the dotted line @xmath287 with @xmath288 as predicted by simple second - order perturbation theory .
more details regarding the distance measure @xmath289 are given in the text .
[ fig5:comparemolmergate],width=302 ] figure [ fig5:comparemolmergate ] shows a comparison of the different propagators for the gate operation with @xmath278 taking place in 25 trap cycles ( @xmath290 ) . for the prediction of the required coupling strength @xmath12 for a gate operation realizing @xmath283 , equation ( [ lambda ] )
was iteratively solved to yield @xmath291 .
the propagator @xmath282 was obtained from a numerical integration of ( [ bichrham ] ) .
then , @xmath282 was compared to @xmath292 , to the propagator of ( [ mspropagator2 ] ) , to @xmath293 and to the prediction @xmath288 of second - order perturbation theory .
since the exact propagator does not perfectly return the motional state to the initial state at the end of the gate , the following procedure was applied for the calculation of the distance @xmath289 between the propagators : we assume that the ions are initially in the motional ground state and that a cooling mechanisms returns the motional state to the ground state at the end of the gate operation without affecting the qubit states .
this turns the unitary evolution into a quantum process acting only on the internal states of the ions . for the comparison of two quantum processes @xmath294 , @xmath295 ,
the processes are mapped using the jamiolkowski isomorphism onto density matrices @xmath296 , @xmath297 for which the distance @xmath298 is calculated @xcite .
the results show that the propagator given by ( [ mspropagator2 ] ) correctly predicts the coupling strength as well as the operator realized by the gate operation .
it also becomes obvious that @xmath299 considerably deviates from the operation generated by the hamiltonian ( [ bichrham ] ) unless @xmath169 . the main advantage of the @xmath0 interaction on optical qubits appears to be its insensitivity to changes in the optical path length . in the limit of weak excitation
, the gate operation tolerates changes that occur within the gate operation as in each elementary process a photon is absorbed and another one emitted into the same laser beam ( this property does not hold for hyperfine qubits since here raman beams in a counter - propagating configuration are used ) .
if higher rabi frequencies are used , the interaction rather becomes @xmath300 which make it susceptible to path - length fluctuations within the gate time .
still , if amplitude - shaped pulses are applied ( see section [ sec : shapedpulses ] ) , the gate operation tolerates changes of the path length that occur between consecutive applications of the gate .
this is not the case for the mlmer - srensen gate which becomes robust against changes between gate operations but remains susceptible to changes occurring within the gate when the gate is sandwiched between @xmath301 pulses applied to both qubits .
.comparison of @xmath0 and @xmath1 gates .
the first row gives the rabi frequency @xmath12 required to peform an entangling gate operation as a function of @xmath42 and the gate duration .
the latter is expressed as the number of trap oscillation periods @xmath302 .
the second row lists the reduction in coupling strength for this kind of gate due to saturation effects .
the third row compares the unwanted to the desired coupling strength.[gatecomptable ] [ cols="^,^,^",options="header " , ] the @xmath0 gate , however , seems to be much less favorable with respect to the following criteria : ( i ) the rabi frequency that is required for performing the gate operation in a given time , ( ii ) the strength of saturation effects reducing the coupling for the rabi frequency needed for the gate operation , and ( iii ) the ratio between the desired and the unwanted spin - spin couplings .
table [ gatecomptable ] shows a comparison of the gates with respect to these criteria , thus summarizing the results ( [ omegazz ] ) , ( [ thetatausat ] ) , ( [ kappaovertheta ] ) for the @xmath0 gate and ( [ omegams ] ) , ( [ lambdatsat ] ) , ( [ lambdaovermu ] ) for the @xmath1 gate . for all three criteria ,
the mlmer - srensen performs better .
having a low rabi frequency is also of interest when it comes to non - resonant excitation of other vibrational modes or light shifts induced by excitation of far - detuned dipole transitions .
in the limit of fast gate operations , the hamiltonians ( [ heffzz ] ) , ( [ heffms ] ) become sensitive to the phase @xmath98 which is related to the intensity of the bichromatic laser field at the start of the gate operation .
it is therefore interesting to shape the intensity of the bichromatic laser field during the gate operation so that the atomic operator @xmath303 , with @xmath304 , appearing in the hamiltonians becomes time - dependent but independent of @xmath98 at the beginning and at the end of the gate when the intensity is low . in this way
, the gate could be made insensitive to @xmath98 by an adiabatic process where @xmath305 evolves from @xmath306 at the start of the gate operations to a @xmath98-dependent operator @xmath307 and back to @xmath306 .
however , the state @xmath272 of the vibrational mode generally does not return to its original state at the end of the gate under the action of the propagator ( [ propagatorharmosc ] ) when the coupling @xmath273 is made time - dependent .
there is , however , a class of shaped pulses with the property @xmath35 that can be constructed in the following way : by applying an amplitude - shaped pulse twice with a sign change in the coupling between the two pulses , i. e. @xmath308 , one obtains the propagator @xmath309 because the first and the second pulse displace the motional state into opposite directions but by an equal amount ( see ( [ propagatorharmosc ] ) ) . a quantum state that is displaced along a circle in phase space by an off - resonant force of constant magnitude ,
@xmath310 , @xmath311 $ ] , can be viewed as a special case of this pulse form with @xmath312 . for the bichromatic gates based on the interactions ( [ heffzz ] ) , ( [ heffms ] )
, the sign change can be accomplished by either shifting the phase @xmath98 during the action of the second pulse by an amount @xmath301 ( @xmath313 ) , respectively , or by changing the overall phase of the laser by @xmath313 during the second pulse ( i.e. @xmath314 ) . ) without lamb - dicke approximation . for parameters
@xmath315 , @xmath278 , the gate takes place during 50 trap oscillation periods .
the pulses are switched on and off during 8 trap cycles using a @xmath316-profile . during the second pulse ,
the phase of the blue- and the red - detuned beam is shifted by @xmath313 with respect to the first pulse .
( a ) time evolution of the populations @xmath317 . when starting from state @xmath318 at time @xmath212 .
the dotted line shows the coupling strength @xmath319 , the dashed line is the average number of vibrational quanta .
( b ) infidelity of the final state as a function of the phase @xmath98 .
the upper curve shows the strong influence of the phase for a gate operation with constant coupling strength @xmath320 where a high - fidelity operation is achieved only for @xmath169 . for the amplitude - shaped gate ,
the fidelity is practically independent of @xmath98 .
similar results are also obtained for other input states . for a realistic calculation of @xmath321 , decoherence caused by spontaneous decay of the metastable state would have to be taken into account .
, width=453 ] figure [ fig4:shapedpulses ] illustrates the use of amplitude - shaping in order to make the mlmer - srensen gate operation robust against fluctuations in the phase @xmath98 between the blue- and the red - detuned laser beam . in this example , an entangling gate is accomplished within @xmath322 trap oscillation periods by a pair of laser pulses with @xmath323 , for @xmath324 with @xmath325 .
as shown in figure [ fig4:shapedpulses]a , the pulses are switched on and off within eight trap cycles .
after the first pulse , the vibrational state has not returned to its initial state .
it is only after the second pulse that the correlations between the vibrational state and the qubit states vanish again . using this technique ,
the initial state @xmath318 is mapped to the target state @xmath326 with an infidelity of below @xmath327 ( see figure [ fig4:shapedpulses]b ) .
this is in sharp contrast to the case of an excitation of the same duration with constant amplitude where the infidelity depends on the phase @xmath98 and varies between @xmath228 and @xmath328 .
similar results are also obtained for other input states .
spin echo pulses can be combined with amplitude - shaped laser pulses to make the @xmath0 gate more robust against imperfections .
it is possible to implement the conditional phase gate operation by having the motional state perform two circles in phase space so that the gate pulse can be split up into two separate pulses .
since the quantum states @xmath207 , @xmath206 as well as the states @xmath329 , @xmath330 pick up the same phases @xmath331 , @xmath332 , it is possible to exchange the states @xmath333 , @xmath334 by a collective @xmath313-pulse sandwiched between the two gate pulses and to exchange the populations at the end of the gate sequence again .
the first spin - echo @xmath313-pulse inverts the direction of the force on the motional state so that the motional state returns to the initial state after the second spin - dependent pulse .
in contrast to the case of shaped pulses without spin - echo , there is no need for changing the phase @xmath98 of the second pulse or the sign of the coupling strength @xmath12 .
the spin echo procedure is advantageous for the following reasons : 1 .
the gate becomes more robust against unequal light intensities on the ions .
single qubit phases arising from light shifts are transformed into an unimportant global phase . in the context of this gate , light shifts
will mainly be due to an imbalance in the power of the blue and the red - detuned laser beams and also due to very off - resonant excitation of dipole transitions .
in addition , a light shift @xmath335 occurs if the average frequency @xmath36 of the bichromatic light field does not exactly coincide with the atomic transition frequency @xmath39 .
however , this light shift will be fairly small as @xmath336 .
collective spin flips arising from the term @xmath337 in ( [ heffzz ] ) can be cancelled to first order by choosing rotation axes for the @xmath313-pulses on ion 1 and ion 2 that differ by @xmath338 ( x - rotation on ion 1 and y - rotation on ion 2 ) .
this effectively changes the sign of the rotation angle @xmath339 occurring in ( [ approxpropagatorzz ] ) for the second pulse and eliminates the spin flip contribution of the interaction . to perform different @xmath313-pulses on both ions requires , however , either a different trap frequency that changes the distance between the ions by @xmath340 or an additional laser beam . in the limit of short gate operations , spin echos
become somewhat less efficient in cancelling perturbations described by @xmath152 interactions as the gate interaction @xmath341 no longer commutes with @xmath152 for @xmath342 . for the mlmer - srensen gate operation , where @xmath343 $ ] is not a small quantity , spin echos seem to be of limited use .
if , however , the gate interaction is sandwiched between a pair of collective @xmath301 pulses to turn it into a @xmath0 interaction , spin echos are helpful for cancelling perturbations occurring between consecutive gates . also , it should be noted that a spin - echo like technique was already proposed in ref .
@xcite in order to cope with number - state dependent ac - stark shifts that arise if the gate is implemented by illuminating ion 1 with a red - detuned laser beam and ion 2 with a blue - detuned laser beam instead of using a bichromatic light field for both ions .
collective laser - ion interactions with bichromatic laser beams are capable of performing both @xmath0 gates as well as mlmer - srensen gate operations .
the analysis shows that it is important to include non - resonant excitation of the carrier transition for the precise calculation of the gate operation .
while the paper was focused on the case of qubit states linked by a weak optical transition , the discussion of the mlmer - srensen gate interaction applies also to hyperfine qubits where non - resonant carrier excitation also occurs in the limit of fast gate operations . for optical qubits
, the required laser beams can chosen to be co - propagating which allows for a robust and experimentally easily realizable setup where an acousto - optical modulator is used in single - pass configuration to create the bichromatic light field .
in a direct comparison of the gates , the mlmer - srensen interaction seems to be advantageous in terms of required laser power and gate accuracy while the @xmath0 interaction has the advantage of being robust even against certain path length fluctuations occuring during the gate operation in the limit of weak driving where @xmath344 . for gate durations coming close to @xmath345 , control of the phase @xmath98 between the red- and the blue - detuned laser beams
is of vital importance unless the gate is performed using amplitude - shaped laser pulses . in this case , the requirements are strongly relaxed and the gates appear to be very promising for experimental realization .
the possibility of using a single laser beam for global single qubit and entangling operations opens also interesting perspectives for creating multi - particle entangled states with more than two ions .
the operations using this beam could be combined with an off - resonant strongly focussed beam capable of inducing @xmath346 operations on individual qubits in order to create a larger variety of complex entangled states .
this work was supported by the austrian science fund ( fwf ) , dto and the eu network scala .
the author wishes to thank w. dr for discussions and j. benhelm for helpful comments and a critical reading of the manuscript .
0 monroe c , meekhoff d m , king b e , itano w m , and wineland d j 1995 , phys . rev . lett . * 75 * 4714 .
milburn g j 1999 , _ preprint _ , arxiv : quant - ph/9908037 .
milburn g j , schneider s , and james d f v 2000 fortschr
. phys . * 48 * 801 .
srensen a , and mlmer k 1999 phys .
lett . * 82 * 1971 .
srensen a , and mlmer k 2000 phys .
a * 62 * 022311 .
solano e , de matos filho r l , and zagury n 1999 phys .
a * 59 * r2539 .
leibfried d , blatt r , monroe c , and wineland d j 2003 rev .
phys . * 75 * 281 .
monroe c m , meekhof d m , king b e , and wineland d j 1996 science * 272 * 1131 .
wallentowitz s , and vogel w 1995 phys .
* 75 * 2932 .
solano e , de matos filho r l , and zagury n 2001 phys .
lett . * 87 * 060402 .
kim k _ et al .
_ 2007 , _ preprint _ , arxiv:0710.3950 .
lee p j , brickman k - a , deslauriers l , haljan p c , duan l - m , and monroe c 2005 j. opt .
b : quantum semiclass
* 7 * s371 .
magnus w 1954 commun .
pure appl . math .
* 7 * 649 .
cahill k e and glauber r j 1969 phys . rev . *
177 * 1857 .
abramowitz m and stegun i a 1972 _ handbook of mathematical functions _
( new york : dover publications ) p 784 | in ion traps , entangling gate operations can be realized by a bichromatic pair of laser beams that collectively interact with the ions . in this paper , a new method of modelling the laser - ion interaction
is introduced that turns out to be superior to standard techniques for the description of gate operations on optical qubits .
the treatment allows for a comparison of the performance of gates based on @xmath0 and @xmath1 interactions on optical transitions where the bichromatic laser field can be realized by an amplitude - modulated laser resonant with the qubit transition . shaping the amplitude of the bichromatic laser pulse
is shown to make the gates more robust against experimental imperfections . |
sf and nk acknowledge the support of the epsrc grant ep / j019526/1 . 10 _ anomalous transport : foundations and applications _ , edited by r. klages , g. radons , and i.m .
sokolov ( wiley - vch , weinheim , 2007 ) .
g. drazer and d.h .
zanette , phys .
e * 60 * , 5858 ( 1999 ) .
weeks and d.a .
weitz , chem .
phys . * 284 * , 361 ( 2002 ) .
g. seisenberger _
et al . _ ,
science * 294 * , 1929 ( 2001 ) .
saxton and k. jacobson , ann .
biophys .
* 26 * , 373 ( 1997 ) .
k. ritchie _ et al .
_ , biophys .
j. * 88 * , 2266 ( 2005 ) .
m. weiss , h. hashimoto , and t. nilsson , biophys .
j. * 84 * , 4043 ( 2003 ) .
i. golding and e.c .
cox , phys .
lett . * 96 * , 098102 ( 2006 ) .
tolic - norrelykke _ et al .
_ , phys .
lett . * 93 * , 078102 ( 2004 ) .
f. hfling and t. franosch , rep .
76 * , 046602 ( 2013 ) .
e. barkai , y. garini , and r. metzler , phys . today * 65 * 29 ( 2012 )
. a. serg _ et al .
_ , nat .
methods * 5 * , 687 ( 2008 ) .
k. jaqaman _ et al .
. methods * 5 * , 695 ( 2008 ) .
k. norregaard _ et al .
_ , phys .
* 16 * , 12614 ( 2014 ) .
y. he _ et al .
_ , phys .
lett.*101 * , 058101 ( 2008 ) .
a. lubelski , i.m .
sokolov , and j. klafter , phys .
lett.*100 * , 250602 ( 2008 ) .
y. meroz , i.m .
sokolov and j. klafter , phys .
* 110 * , 090601 ( 2013 ) .
i. bronstein _ et al .
_ , phys .
lett . * 103 * , 018102 ( 2009 ) .
t. neusius _ et al .
_ , phys .
100 , 188103 ( 2008 ) .
m. saxton , biophys .
j. * 81 * , 2226 ( 2001 ) .
weigel _ et al .
_ , proc .
usa * 108 * , 6438 ( 2011 ) .
sokolov , soft matter * 8 * , 9043 ( 2012 ) .
bressloff and j.m .
newby , rev .
phys . * 85 * , 135 ( 2013 ) .
r. metzler , e. barkai , and j. klafter , phys .
lett.*82 * , 3563 ( 1999 ) .
r. metzler and j. klafter , phys .
reports * 339 * , 1 ( 2000 ) .
e. heinsalu _ et al .
_ , phys .
lett . * 99 * , 120602 ( 2007 ) .
m. magdziarz , a. weron , and j. klafter , phys .
lett.*101 * , 210601 ( 2008 ) .
henry , t.a.m .
langlands , and p. straka , phys .
lett . * 105 * , 170602 ( 2010 ) . s. eule and r. friedrich , euro .
lett . * 86 * , 30008 ( 2009 ) .
i.m . sokolov and j. klafter ,
* 97 * , 140602 ( 2006 ) .
shushin , phys .
e * 78 * , 051121 ( 2008 ) .
shkilev , journal of experimental and theoretical physics , * 114 * , 830 ( 2012 ) .
s. fedotov , a.o .
ivanov and a.y .
zubarev , math . model .
nat . phenom .
* 8 * , 28 ( 2013 ) .
e. abad , s.b .
yuste , and k. lindenberg , phys .
e * 81 * , 031115 ( 2010 ) .
s. fedotov and s. falconer , phys .
e * 87 * , 052139 ( 2013 ) .
t. neusius , i.m . sokolov and j.c .
smith , phys .
e * 80 * , 011109 ( 2009 ) .
wong _ et al .
_ , phys .
* 92 * , 178101 ( 2004 ) . d.r .
cox and h.d .
miller , _ the theory of stochastic processes _ ( methuen , london , 1965 ) .
chechkin , r. gorenflo , and i.m .
sokolov , j. phys .
a : math . gen . * 38 * , l679 ( 2005 ) .
s. fedotov , phys .
e * 88 * , 032104 ( 2013 ) .
to derive eq .
( [ main_eq ] ) we use the structured probability density function @xmath105 with the residence time @xmath52 as auxiliary variable .
this density gives the probability that the particle position @xmath106 at time @xmath7 is at the point @xmath6 and its random residence time @xmath43 at point @xmath6 is in the interval @xmath107 the density @xmath108 obeys the balance equation @xmath109we consider only the case when the residence time of random walker at @xmath110 is equal to zero , so the initial condition is @xmath111where @xmath112 is the initial density .
the boundary condition in terms of residence time variable ( @xmath113 can be written as @xcite @xmath114we solve ( [ basic ] ) by the method of characteristics for @xmath115 @xmath116where @xmath117the solution eq .
( [ je ] ) is written in terms of the integral arrival rate @xmath118 and in terms of the survival function eq .
( [ ffff ] ) @xmath119our purpose now is to derive the master equation for the probability density@xmath120let us introduce the integral escape rate to the right @xmath121 and the integral escape rate to the left @xmath122 as@xmath123note that the integration with respect to the residence time @xmath52 in ( deng ) and ( [ i1 ] ) involves the upper limit @xmath124 where we have a singularity due to the initial condition ( [ initial ] )
. then the boundary conditions ( [ arr1 ] ) can be written in a simple form : @xmath125it follows from ( [ initial ] ) , ( [ je ] ) and ( [ i1 ] ) that @xmath126where @xmath127 and @xmath128 substitution of ( initial ) and ( [ je ] ) to ( [ deng ] ) , gives @xmath129the balance equation for probability density @xmath5 can be written as@xmath130let us find a closed equation for @xmath5 by expressing integral rates @xmath131 and @xmath132 in terms of the density @xmath133 we apply the laplace transform @xmath134 to ( [ i55 ] ) , and ( [ p11 ] ) , and obtain @xmath135which after the inversion of the laplace transform and using the shift theorem gives @xmath136the memory kernels @xmath137 and @xmath138 are defined by laplace transforms @xmath139now we consider the sub - diffusive case where @xmath140 is inversely proportional to the residence time @xmath141 @xmath142for simplicity we consider @xmath143 it is straightforward to generalize to non - homogeneous systems by considering space dependent @xmath144 and space dependent anomalous exponent @xmath145 , this case we consider elsewhere @xcite . from eqs .
( [ ffff ] ) and ( s33 ) it follows that the survival function has a power - law dependence @xmath146 the waiting time density functions @xmath147 are @xmath148using the tauberian theorem their laplace transforms are @xmath149 as @xmath150 , where @xmath151 . from ( [ new5 ] )
we obtain the laplace transforms @xmath152therefore , the integral escape rates to the right @xmath153 and to the left @xmath154 in the sub - diffusive case are @xmath155 \slash ( 2 g ) .
\label{gen_i}\]]by introducing the total integral escape rate @xmath156and expanding the right - hand side of eq .
( [ balan ] ) to second order in jump size @xmath9 we obtain the following fractional equation @xmath157 + \frac{a^{2}}{2}\frac{\partial ^{2}i}{\partial x^{2 } } , \label{sup_main}\]]which using eq .
( [ gen_i ] ) leads to the main equation of the paper eq .
( [ main_eq ] ) .
now we derive the equation for the stationary solution eq .
( [ stat_eq ] ) . writing the escape rate @xmath10 in laplace form @xmath158and taking the limit @xmath159
corresponding to @xmath160 , we obtain the stationary escape rate @xmath161where the stationary density is defined in a standard way @xmath162 . taking the time derivative in eq .
( [ sup_main ] ) to zero and substituting eq .
( [ i_st ] ) we obtain the stationary advection - diffusion equation @xmath163 + \frac{d^2}{dx^2}\left [ d_{f}(x)p_{st}(x)\right ] = 0 .
\label{stat_eq_2}\]]integrating eq .
( [ stat_eq_2 ] ) and taking into account that the flux of the particles is zero we obtain eq .
( [ stat_eq ] ) . | we propose a model of sub - diffusion in which an external force is acting on a particle at all times not only at the moment of jump .
the implication of this assumption is the dependence of the random trapping time on the force with the dramatic change of particles behavior compared to the standard continuous time random walk model .
constant force leads to the transition from non - ergodic sub - diffusion to seemingly ergodic diffusive behavior .
however , we show it remains anomalous in a sense that the diffusion coefficient depends on the force and the anomalous exponent . for the quadratic potential
we find that the anomalous exponent defines not only the speed of convergence but also the stationary distribution which is different from standard boltzmann equilibrium .
recently it has become clear that anomalous diffusion measured by a non - linear growth of the ensemble averaged mean squared displacement @xmath0 with the anomalous exponent @xmath1 is as widespread and important as normal diffusion with @xmath2 @xcite .
sub - diffusion with @xmath3 was observed in many physical and biological systems such as porous media @xcite , glass - forming systems @xcite , motion of single viruses in the cell @xcite , cell membranes @xcite , and inside living cells wei03,gol06,tolic04 .
many examples of sub - diffusive processes in biological systems can be found in recent reviews @xcite .
nowadays new tools are available including super - resolution light optical microscopy techniques to deal with biological in vivo data which allows to monitor a large number of trajectories at the single - molecule level and at nanometer resolution @xcite . using these techniques it is possible to discriminate between anomalous ergodic processes where the ensemble and time averages coincide and non - ergodic processes where ensemble and time averages have different behavior @xcite .
two important observations have been made about anomalous transport in living cells : ( 1 ) anomalous transport is usually a transient phenomenon before transition to normal diffusion or saturation due to confined space bro09,neu08,sax01 ( 2 ) ergodic and non - ergodic processes may coexist as it was observed in plasma membrane @xcite .
several models are proposed to describe ergodic and non - ergodic anomalous processes such as non - ergodic continuous time random walk ( ctrw ) with power - law tail waiting times , ergodic anomalous process generated by fractal structures , fractional brownian - langevin motion characterized by long correlations and time dependent diffusion coefficient @xcite .
the standard ctrw model for sub - diffusion of a particle in an external field @xmath4 randomly moving along discrete one - dimensional lattice can be described by the generalized master equation for the probability density @xmath5 to find the particle at position @xmath6 at time @xmath7 @xmath8where @xmath9 is the lattice spacing and @xmath10 is the total escape rate from @xmath11 @xmath12here @xmath13 is a constant timescale and @xmath14 is the riemann - liouville fractional derivative defined by @xmath15the probabilities of jumping to the right @xmath16 and to the left @xmath17 are @xmath18series expansion of eq .
( [ mm ] ) together with eq . ( [ ieq ] ) and eq .
( [ ww ] ) leads to the fractional fokker - planck equation ( ffpe ) @xcite @xmath19 \mathcal{d}_{t}^{1-\mu } p , \label{eq : ffpe}\]]where the generalized diffusion @xmath20 the stationary solution of eq .
( [ eq : ffpe ] ) is the boltzmann distribution .
there exist a huge literature on this equation met99,met00 and its generalization for time dependent forces hei07,mag08,henry10,eule09,sok06,shu08,shkilev12 .
one of the main assumptions in this literature , which is not always clearly stated is that , as long as a random walker is trapped at a particular point @xmath11 , the external force @xmath4 does not influence the particle .
it is clear from eq .
( [ ieq ] ) that the escape rate @xmath10 does not depend on the external force @xmath21 the force only acts at the moment of escape inducing a bias .
the question is how to take into account the dependence of the escape rate on @xmath22 to the author s knowledge this is still an open question .
one of the main aims of this letter is to propose a model which deals with this problem .
we find that the dependence of escape rate on force drastically changes the form of the master equation ( [ mm ] ) and ffpe ( [ eq : ffpe ] ) .
we observe transient anomalous diffusion and transition from non - ergodic to normal ergodic behavior .
however , we show that this seemingly normal process could be still anomalous masked by normal behavior .
our findings suggest that a closer inspection of experimental results could be necessary in order to discriminate between normal and anomalous processes . _
model. _ we consider a random particle moving on a one dimensional lattice under assumption that an external force acts on a particle at all times not only at the moment of jump as in eq .
( [ mm ] ) .
the implication of this assumption is the dependence of the random trapping time on the external force ( not just jumping probabilities as in ( [ ww ] ) ) .
some discussion of situation when the external force influence the rates and jumps can be found in @xcite .
the main physical idea behind our model is that there exists two independent mechanism of escaping from the point @xmath6 with two different random residence times .
the first mechanism is due to external force with the escape rate proportional to @xmath21 the second one is the sub - diffusive mechanism involving the rate inversely proportional to the residence time .
the latter generates the power law waiting time distribution with the infinite first moment .
regarding the first mechanism , we define the jump process from the point @xmath6 as follows .
we assume that the rate of jump to the right @xmath23 from @xmath6 to @xmath24 is @xmath25 when @xmath26 and the rate of jump to the left @xmath27 from @xmath6 to @xmath28 is @xmath29 when @xmath30 for this jump model the random waiting time @xmath31 at the point @xmath6 is defined by the exponential survival probability @xmath32 involving the external force @xmath4 @xmath33where @xmath34 is the intensity of jumps due to force field . for example , one can think of the escape rate @xmath23 that is defined in terms of the potential field @xmath35 that is @xmath36 > 0,$ ] there @xmath37 for @xmath38 the second mechanism involves the sub - diffusive random walk with the escape rate @xmath39 from the point @xmath6 , which is inversely proportional to the residence time @xmath40 in this case the random waiting time @xmath41 at the point @xmath6 is defined by the survival probability @xmath42the question now is how to implement the jumping process due to external force into the sub - diffusive random walk scheme ? when the random walker makes a jump to the point @xmath6 , it spends some random time ( residence time ) before making another jump to @xmath24 or @xmath28 .
let us denote this residence time @xmath43 .
the key point of our model is that we define this residence time as the minimum of two : @xmath41 and @xmath31 @xmath44for the anomalous sub - diffusive case this model could lead to the drastic change in the form of the fractional master equation .
the main reason for this is that the external force @xmath4 plays the role of tempering factor preventing the random walker to be anomalously trapped at point @xmath6 . because of the independence of two mechanisms , in our model the rate of jump @xmath23 to the right from @xmath6 to @xmath24 and the rate of jump @xmath45 to the left from @xmath6 to @xmath28 can be written as the sum@xmath46and@xmath47although it is straightforward to consider general @xmath48 , @xmath49 , for simplicity in what follows we consider @xmath50 . in our model
the asymmetry of random walk occur only from the force dependent rate .
let us explain the main idea of eqs .
( [ t+ ] ) and ( [ t- ] ) .
the external force @xmath26 increases the sub - diffusive rate of jumps to the right @xmath51 and does not change the sub - diffusive rate of jumps to the left .
the essential property of eqs . ( [ t+ ] ) and ( [ t- ] ) is that the rate @xmath39 depends on the residence time variable @xmath52 .
this dependence makes any model involving the probability density @xmath5 non - markovian . for the markov case with @xmath53 @xmath54
has a meaning of the mean residence time at the point @xmath6 .
when the parameter @xmath55 and the rates are @xmath56 @xmath57 , we obtain the standard fractional fokker - planck equation ( eq : ffpe ) . notice that eq .
( [ min ] ) is consistent with the expression for the effective escape rate @xmath58 as a sum of two rates @xmath59 .
similar situation has been considered in @xcite .
after incorporation of the force dependent escape rates we can obtain generalized master equation ( see supplementary materials for the derivation ) . by expanding the rhs of the master equation to the second order in jump size @xmath9 , we get a fractional diffusion equation @xmath60 \right ] - \label{main_eq}\]]@xmath61 .\ ] ] this equation is fundamentally different from the classical ffpe ( eq : ffpe ) because it involves the external force in both terms on the right hand side .
one can see that the force @xmath4 not only determines the advection term as in eq .
( [ eq : ffpe ] ) , but also plays the role of tempering parameter through the factor @xmath62 . similar factor occurs in sub - diffusive equation with the death or evanescent process @xcite .
however , here we consider the system with constant total number of particle . the stationary solution @xmath63 of eq .
( [ main_eq ] ) obeys the standard equation@xmath64 = 0 .
\label{stat_eq}\]](see a supplement material for details ) .
interesting property of this equation is that the effective diffusion constant @xmath65 depends on the external force and anomalous exponent @xmath66this fact implies that the boltzmann distribution is no longer stationary solution of ( [ stat_eq ] ) . for the quadratic potential @xmath67 with @xmath68
we find that for large @xmath6 the stationary density @xmath69 has the form @xmath70where @xmath71 is a constant .
one can see that the form of stationary density is determined by the anomalous exponent @xmath72 . in this case
the particles spread further compared to the boltzmann case .
the reason is the dependence of the effective diffusion constant @xmath65 on force @xmath21 note that for the sub - diffusive fractional fokker - planck equation ( [ eq : ffpe ] ) the anomalous exponent only determines the slow power law relaxation rate , while the stationary density converges to boltzmann equilibrium which does not depend on @xmath73 _ numerical simulations. _ we consider two particular cases : ( 1 ) constant force @xmath74 corresponding to the linear potential and ( 2 ) the quadratic potential @xmath75 both in the infinite domain .
we concentrate on the behavior of the density function @xmath5 , the mean @xmath76 and the variance @xmath77 calculated using an ensemble of trajectories from the initial distribution @xmath78 .
we also calculate the time averaged variance of a single trajectory of length @xmath79 , @xmath80 , where @xmath81 , @xmath82 .
this quantity become a standard tool to assess ergodic properties of a system been equivalent to its ensemble averaged counterpart only for ergodic case .
when the external force @xmath74 is constant , we observe the transition from sub - diffusion at short times to seemingly normal diffusion at long times .
the density function changes from the distinct sub - diffusive shape for short times to the gaussian shape propagator at longer times ( see the inset of figure [ fig1 ] ) .
the average position of the ensemble behaves as @xmath83 .
the ensemble averaged variance @xmath84 grows as a power law for short times , @xmath85 , and transition to a normal diffusive linear growth @xmath86 for longer times .
however , in this case the diffusion coefficient @xmath87 depends on the force @xmath74 and anomalous exponent @xmath72 .
we conclude that although the variance @xmath88 is linearly proportional to time , this dependence reveals the anomalous nature of the process even in the limit @xmath89 .
numerical calculations confirms the analytical result for the diffusion coefficient eq .
( [ d_f ] ) ( see figure [ fig1 ] ) .
second observation is that the power law behavior at short times involves the exponent @xmath90 which depends on force @xmath74 .
for @xmath91 they are estimated to be @xmath92 for @xmath93 , @xmath94 for @xmath95 and @xmath96 for @xmath97 .
this can be interpreted as an enhancement of sub - diffusion coursed by the constant force .
such enhancement should be taken into account in the analysis of biological experiments where sub - diffusion usually appears as transient before the transition to the normal diffusion @xcite .
for the large value of @xmath74 the exponent @xmath98 tends to one while in the small force limit @xmath99 .
the time averaged variance calculated for constant force grows linearly @xmath100 and shows minor scatter between single trajectories ( figure [ fig2 ] ) . after averaging over different trajectories
, it grows with the coefficient @xmath101 which is equal to the ensemble average value .
this shows that the non - ergodic sub - diffusive system becomes an ergodic one .
now we consider the quadratic potential @xmath102 the system becomes again non - ergodic despite the tempering affect of the force . to confirm this
we calculate the time averaged variance ( inset of figure fig2 ) .
as expected it shows large fluctuations among different trajectories typical for non - ergodic systems .
note that even with this typical behavior , it can be easily distinguished in experiments since in our case the mean of the time averaged variance converges to a constant , while for standard ctrw in a bounded region it grows as a power of the anomalous exponent , @xmath103 .
regarding the shape of the stationary density , numerical simulations are in good agreement with analytical results eq .
( [ stat ] ) ( see figure fig3 ) . _
summary. _ in this letter we have presented a model of anomalous sub - diffusive transport in which the force acts on the particle at all times not only at the moment of jump .
this leads to the dependence of jumping rate on the force with the dramatic change of particles behavior compared to the standard ctrw model .
we have derived a new type of fractional diffusion equation which is fundamentally different from the classical fractional fokker - planck equation . in our model
the force @xmath4 not only appears in the drift term as in eq .
( [ eq : ffpe ] ) , but also determines the structure of the diffusion term controlling the spread of particles .
the constant external force leads to the natural tempering of the broad waiting time distribution and , as a result , to the transition to a seemingly normal diffusion ( linear growth of the mean squared displacement ) and equivalence of the time and ensemble averages .
however , this may lead to a wrong conclusion in analyzes of experimental results on transient sub - diffusion @xcite that the process is normal for large times .
we have found that contrary to normal diffusion process in the external force field , the diffusion coefficient depends on the force and anomalous exponent .
this fact implies that the boltzmann distribution is no longer stationary solution .
external perturbations and noise fluctuations are not separable which reflects the non - markovian nature of the process even for large times .
our results would be possible to test in experiments , for example , by considering a bead which is moving sub - diffusively in an actin network .
the motion of such a beat can be described by a random walk type of dynamics @xcite .
force - measurements could be realized by using optical trap and tweezers which are the nano - tools capable of performing such measurements on individual molecules and organelles within the living cell @xcite .
when the force is constant the dependence of the measures diffusion coefficient on the strength of the force would reveal the predicted power law behavior @xmath104 . for quadratic potential
it could be possible to retrieve the form of the stationary profile ( [ stat ] ) with the slow decay compared to boltzmann distribution for large @xmath6 . |
according to the currently favoured galaxy formation scenario , the cold dark matter ( cdm ) cosmogony , galaxies are built up from the assembly of small structures . in this paradigm
the assembly mechanism plays a key role in understanding the formation history of galaxies .
recent cold dark matter cosmological numerical simulations predict the existence of a large population of _
subhalos_. comparisons with the observed population of dwarf galaxies and detailed predictions of the present - day subhalo population , dark or luminous , have become important tests of the cdm galaxy formation paradigm @xcite .
most studies to date use large cosmological simulations and classify their properties statistically .
however , to properly investigate these processes , one needs to perform high resolution idealised simulations of subhalo evolution within the cdm paradigm @xcite .
alternatively , in this study , we investigate one important consequence of subhalo disruption : the formation and evolution of tidal tails . by adopting initial conditions motivated by the cdm simulations
, we can focus our computational resources on understanding the dynamical mechanism .
satellite galaxy tidal tails are an important observable fossil signature to help understand the formation history of the milky way and to test cdm theory as a consequence .
tails and streams provide information about the galactic halo mass model as well as the evolutionary history of the observed satellite galaxy . in the cdm model
, galaxies are embedded in massive dark matter halos .
estimating dark matter halo structure is essential to understand galaxy formation and tidal tail morphology probes halo structure @xcite .
several space missions , for example the esa astrometric satellite gaia @xcite , are planned to measure the position and motion of stars in the milky way with very high accuracy , in the near future .
together with ground - based radial velocity experiments , e.g. rave@xcite , these surveys will provide full phase space information .
accurate six dimensional phase space information of milky way stars will provide observational information of the tidal tail and hence the formation history of the milky way .
the time is ripe to carry out a detailed theoretical study of satellite galaxy disruption and the induced tidal tail morphology . in this study
we perform numerical simulations of satellite galaxy disruption and its induced tidal tail morphology within the cdm cosmogony .
the objective of this study is to understand the physical processes responsible for satellite galaxy disruption rather than reproducing the evolutionary history of any individual milky way satellite galaxy .
in particular , satellite disruption in n - body simulations is produced by escaping satellite particles .
in addition , the gravitational shock , which is caused by the slowly varying host halo potential as the satellite goes through its orbit , changes the satellite s internal structure . an initially stable satellite galaxy and accurate numerical integration of a satellite particle s orbit
are necessary to represent these physical processes correctly .
we investigate satellite galaxy evolution by performing high resolution and low noise n - body simulations with such stable initial conditions .
in addition , we can estimate any trends of satellite tidal tail morphology with satellite properties from our simulations , even though we do not reproduce the evolutionary history of any specific milky way satellite .
our simulation results show that the gravity of the satellite alters the location of the tidal tails relative to the satellite orbit .
the satellite decelerates ( accelerates ) the leading ( trailing ) tail beyond the tidal radius , which is proportional to the satellite mass . for more massive satellites , this results in the leading tail being located well inside and the trailing tail being located well outside of the satellites orbit , rather than tracking the original satellite orbit @xcite .
since the satellite torques the tidal tail , the distribution of the tidal tail in the observational plane is rather different from predictions that exclude such satellite torquing .
in addition , the simulations provide six - dimensional phase - space information of the tidal tail to compare with upcoming astrometric measurements . in section , [
sec : simulation ] we describe how we make stable satellite initial conditions and provide a brief overview of the simulation algorithm . in section ,
[ sec : satellite ] we investigate satellite disruption and the formation of the tidal tail , including the effects of the satellite potential on the tidal tail , and in section [ sec : observation ] , we investigate the observational consequences . in section [ sec : summary ]
, we summarise our results and discuss their importance .
.initial properties of the three satellite models [ cols="^,^,^,^",options="header " , ] the initial conditions of our simulations are motivated by the cdm cosmology .
cdm cosmological simulations suggest that dark matter halos have a universal density profile ( * ? ? ?
* hereafter nfw ) , @xmath3 , where @xmath4 is a scale length characterised by the concentration parameter @xmath5 and @xmath6 is the virial radius of the halo . although there are some disagreements regarding the accuracy of this simple formula in describing halos in numerical simulations and in comparisons with observed galaxies , it remains the accepted cdm halo density profile and we adopt it for our study .
we represent the host halo by a static nfw halo potential .
hence our simulations ignore the effects of dynamical friction and the subsequent reaction of the host halo .
obviously this is not realistic but since the satellite masses of interest are often much smaller than the host halo mass , the consequences are minor . moreover ,
the prime motivation of this study is to understand the physical processes responsible for satellite disruption and tidal tail formation and not the evolutionary history of the satellite .
dynamical friction is not vital to understand these processes . , @xmath7 , and @xmath8 .
] before tidal truncation , the satellites have an nfw density profile .
however , the nfw profile extends to infinity and real astronomical systems have a finite size .
the conventional solution limits the size of an isolated dark matter halo to its virial radius @xcite .
the host halo s tidal field then determines the size of satellite halo .
the host halo s tidal field affects a satellite halo even before the satellite halo passes within the host halo s virial radius . as a result , it is computationally expensive to simulate the entire evolution of a satellite
. remember , the objective of our study is to understand the physical processes responsible for tidal tail formation not the reproduction of a particular tail feature .
therefore , we place the satellite in the host halo on the desired orbit to start and include the host halo s tidal field when we generate a satellite s initial phase - space distribution .
it is natural to characterise the tidal length scale in the satellite by the radius of the x - point that this satellite would have at some fiducial galactocentric radius in its orbit .
we call this fiducial radius the _ tidal distance_. in other words , the satellite on a circular orbit at the tidal distance would have the x - point @xmath9 . at this point
, the gravitational force from the satellite exactly balances the gravitational force from the host halo and non - inertial centrifugal force .
the details of the iterative procedure that we use to generate the satellite s initial condition is as follows .
first , we truncate the virial radius limited nfw satellite halo at the x - point radius , @xmath9 , using the error function , @xmath10\}$ ] .
we then compute the distribution function using eddington inversion and calculate a new satellite density profile by integrating the distribution function over velocity .
the parameter @xmath11 in the error function truncation formula is increased from zero until a smooth phase - distribution function results .
we iterate these steps until the density distribution - function pair is converged .
figure [ fig : procedure ] shows an example of how this procedure modifies an initial satellite halo . at the conclusion of the procedure , the effective tidal radius is approximately 75% of the initial x - point radius radius .
we characterise a satellite halo by its _ initial _ maximum circular velocity ; figure [ fig : procedure ] demonstrates that this velocity is only weakly affected by the truncation procedure .
we denote the outer radius of the satellite after the truncation procedure as the _ effective tidal radius_. it is smaller than @xmath9 owing to the truncation with the error function and the eddington inversion process
. finally , we use an acceptance - rejection algorithm to generate each particle halo realization .
the satellite is made up @xmath12 particles .
since the initial satellites are already truncated , satellite particles are ejected only through interactions with the host halo during the simulation .
we simulate a set of satellite realizations with the same tidal distance but different initial maximum rotation velocities .
we investigate the effects of a satellite s size and orbit on its tidal tail morphology by varying them separately but keeping the other parameters fixed . in detail , we choose a @xmath13 nfw model for both the host halo potential and for the satellites .
we generate three different size satellites , which we refer to as the massive satellite , the low - mass satellite , and the tiny - mass satellite .
we use the maximum rotation velocity , @xmath14 , as a measure of satellite size since the continuous mass loss makes mass an inexact measure .
we use a @xmath14 of 0.45 , 0.16 , and 0.08 times @xmath15 for the massive , low - mass , and tiny satellite , respectively . the tidal distance for all three satellites
is @xmath16 .
we also set the galactocentric orbital radius to @xmath16 for our circular orbit simulations .
after our truncation procedure is complete , the initial mass of the massive satellite is 0.018 @xmath17 , the low - mass satellite 0.001 @xmath17 , and the tiny - mass satellite 0.0001 @xmath17 . converting our simulation units to a milky way size galaxy system and evolving for a few satellite orbits
, the low - mass satellite roughly corresponds in mass to the sagittarius dwarf galaxy halo @xcite .
the massive and tiny - mass satellites are an order of magnitude more and less massive , respectively .
the properties of these satellite halos is summarised in table [ tbl : satellites ] in units of the virial quantities of the host halo .
all satellite initial conditions in this study have @xmath18 particles .
we evolve each of the three satellites on three different orbits with the same energy but with different eccentricities .
we define the eccentricity of the orbits as @xmath19 where @xmath20 and @xmath21 are the apocentre and the pericentre of a satellite .
the first orbit is circular ( @xmath22 ) at @xmath16 .
the second orbit has an @xmath23 with a pericentre of @xmath24 and an apocentre of @xmath25 , and the third orbit has @xmath26 with a pericentre of @xmath27 and an apocentre of @xmath28 .
the third orbit is particularly relevant cosmologically since its _ circularity _
( @xmath29),where @xmath30 is the angular momentum and @xmath31 is the angular momentum of a circular orbit with the same energy . ]
is 0.5 , which is the median @xmath29 of subhalos in a sample taken from recent cosmological simulations @xcite .
we quote results using the following _ system _ units unless otherwise specified : @xmath32 , @xmath7 , and @xmath8 .
the timestep for our n - body simulations is @xmath33 system time units .
therefore , one circular orbit is made of about @xmath34 timesteps because one circular orbit period is @xmath35 and a tidal distance corresponding to @xmath16 with an eccentricity @xmath36 .
the virial radius orbit begins with @xmath37 and a tidal distance corresponding to @xmath38 with @xmath23 . owing to gravitational heating
, the evolution of the two satellites is different but the profiles approximately agree when the total mass lose is the same . ]
our satellite subhalo is initially truncated without considering its evolutionary history and without including any gravitational heating .
this is crudely consistent with our initial condition generation procedure that assumes an equilibrium configuration at some radius inside the host halo to start .
fortunately , this idealised setup does not produce an unrealistic satellite mass loss history . @xcite and @xcite performed a quantitative study of nfw subhalo evolution in a host halo using idealised n - body simulations and claim that satellites on two different orbits have similar mass and velocity profiles after losing the same amount of mass . to check this , we compared the evolved density profiles of two low - mass satellites on orbits with @xmath26 but at two different tidal distances : 1 ) the radius of a circular orbit with the same energy ; and 2 ) the host halo virial radius , @xmath6 . the first test describes the satellite evolution scenario adopted for this study .
the second test describes the cosmologically - motivated scenario of a satellite entering the host halo for the first time .
certainly , these two satellites have quite different evolutionary histories .
however , when the bound mass of the two satellites is scaled to the same value , their evolved density profiles are similar , as shown in figure [ fig : profile_comp ] .
this test , together with the results of @xcite and @xcite , suggests that our tidally truncated satellite models are a fair representation of cdm subhalos ( see figure 10 in * ? ? ?
in addition , although tidal heating , which is sensitive to a satellite s structure , plays an important role in the satellite disruption process , we will show that the tidal tail morphology does not depend on a satellite s inner structure but only on a satellite s mass and orbit .
for the gravitational potential solver , we use a three - dimensional self - consistent field algorithm ( scf , also known as an _ expansion _ algorithm , e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
this algorithm produces a bi - orthogonal basis set of density - potential pairs from which it computes the gravitational potential of a n - body system , given the mass and positions of the particles . for an arbitrary basis ,
e.g. spherical bessel functions , the expansion generally requires a large number of terms to achieve convergence , which introduces small - scale noise as well as requiring greater computational expense .
the situation was dramatically improved by @xcite using a numerical solution of the sturm - liouville equation to match the lowest - order pair to the equilibrium profile , and therefore , the expansion series converges rapidly . here
, we use the current density profile as the zero - order basis function . for our purposes ,
this expansion algorithm is attractive for two reasons .
first , the expansions can be chosen to follow structure over an interesting range of scales and simultaneously suppresses small - scale noise .
in contrast , noise from two - body scattering can arise at all scales in direct - summation , tree algorithm , and mesh based codes .
small - scale scattering can give rise to a diffusion in conserved quantities , which can lead to unphysical outcomes particularly for studies of long - term galaxy evolution ( see * ? ? ?
* ; * ? ? ?
second , the expansion algorithm is computationally efficient ; the computational time only increases linearly with particle number .
hence , the expansion algorithm permits the use of a much larger number of particles than most other algorithms for the same computational cost . an accurate potential solver for a cuspy halo demands a precise determination of the expansion centre , @xmath39 .
this is the major disadvantage of the expansion algorithm relative to a lagrangian potential solver such as a tree code .
we developed and tested the following algorithm for evolving cuspy dark matter halos with an expansion code : 1 . at time
step @xmath40 , we compute @xmath41 from the centre of mass of the @xmath42 most bound particles ; 2 . to evaluate the expansion centre at time step @xmath43 , a predicted centre @xmath44 is estimated from a linear least squares solution using the previous @xmath45 centres : @xmath46 ; 3 . for @xmath47 , we set @xmath48 .
4 . to reduce truncation error , we separately track the motion relative to the satellite s centre and the motion of the centre itself .
the linear least squares estimator for the expansion centre @xmath49 reduces the poisson noise from the @xmath42 particles used to determine each of the @xmath41 . for our simulations
we have adopted @xmath50 and @xmath51 and have verified that this centring scheme maintains the cusp while the satellite orbits in a host halo for situations where the tidal field is insignificant .
time - dependent forcing by the host halo s tidal field adds energy to the satellite , driving mass loss and , ultimately , disruption .
these forces are a combination of the differential force from the host halo and the non - inertial forces from the satellite orbit .
the work done against the satellite s gravitational potential results in mass loss . in addition , these forces _ deform _ the outer density contours of the satellite . to understand the evolution of the ejecta
, one must also consider the gravitational field of the satellite .
the gravitational force from the satellite decelerates ( accelerates ) the leading ( trailing ) tail , modifying the energy and angular momentum of the ejecta well past the point of escape . the conserved quantities of the ejecta , then , may be dramatically different than that of the satellite centre of mass .
the strength of the satellite gravity increases with satellite mass , of course .
these effects combine to make the morphology of tidal tails more complicated than previously suggested @xcite , especially for a massive satellite .
we investigate the causes of these effects and their consequences in detail below . at @xmath52 ( top - left ) , @xmath53 ( top - right ) , @xmath54 ( bottom - left ) , and @xmath55 ( bottom - right ) .
recall that the orbital period for the circular orbit in this simulation is @xmath56 .
the colour scale is logarithmic in the dark matter mass density , increasing from blue to red , and is fixed for all times @xmath57 .
each panel has a linear size of 2 host - halo virial radii . for this simulation
, the tail particles do not feel the gravitational force of the satellite after escape .
the circles show the satellite orbit .
the multiple streams in the tail owe to phase crowding near apocentre for initially prograde and retrograde orbits.,title="fig : " ] at @xmath52 ( top - left ) , @xmath53 ( top - right ) , @xmath54 ( bottom - left ) , and @xmath55 ( bottom - right ) .
recall that the orbital period for the circular orbit in this simulation is @xmath56 .
the colour scale is logarithmic in the dark matter mass density , increasing from blue to red , and is fixed for all times @xmath57 .
each panel has a linear size of 2 host - halo virial radii . for this simulation
, the tail particles do not feel the gravitational force of the satellite after escape .
the circles show the satellite orbit .
the multiple streams in the tail owe to phase crowding near apocentre for initially prograde and retrograde orbits.,title="fig : " ] at @xmath52 ( top - left ) , @xmath53 ( top - right ) , @xmath54 ( bottom - left ) , and @xmath55 ( bottom - right ) .
recall that the orbital period for the circular orbit in this simulation is @xmath56 .
the colour scale is logarithmic in the dark matter mass density , increasing from blue to red , and is fixed for all times @xmath57 .
each panel has a linear size of 2 host - halo virial radii .
for this simulation , the tail particles do not feel the gravitational force of the satellite after escape .
the circles show the satellite orbit .
the multiple streams in the tail owe to phase crowding near apocentre for initially prograde and retrograde orbits.,title="fig : " ] at @xmath52 ( top - left ) , @xmath53 ( top - right ) , @xmath54 ( bottom - left ) , and @xmath55 ( bottom - right ) . recall that the orbital period for the circular orbit in this simulation is @xmath56 .
the colour scale is logarithmic in the dark matter mass density , increasing from blue to red , and is fixed for all times @xmath57 .
each panel has a linear size of 2 host - halo virial radii . for this simulation
, the tail particles do not feel the gravitational force of the satellite after escape .
the circles show the satellite orbit .
the multiple streams in the tail owe to phase crowding near apocentre for initially prograde and retrograde orbits.,title="fig : " ] we begin by describing the dynamics and morphology of the tidal tails in a simulation that _ ignores _ the gravitational field of the satellite past the tidal radius .
figure [ fig : framesltc3 ] shows a sequence of snapshots of the low - mass satellite ( @xmath58 ) on a circular orbit at @xmath16 .
we use units where @xmath59 , @xmath60 , and newton s gravitational constant @xmath32 , together which defines a natural time unit . scaled to the milky way , 0.5 natural time units is approximately 1 gyr .
appealing to the standard zero - velocity roche potential , which balances the effective gravitational potential in the rotating frame with the halo potential , we expect the mass to become unbound in the vicinity of the lagrange or x - points .
indeed , we observe the double cometary appearance of tails leading and trailing the satellite , enforced by the conservation of angular momentum . for this halo model ,
the leading ejecta orbits faster than the satellite and has a position angle of @xmath61 measured from the positive vertical axis , the direction of satellite s instantaneous motion .
the trailing tail moves slower than the satellite and has a position angle of @xmath62 .
since the simulation in figure [ fig : framesltc3 ] ignores the satellite s gravity beyond the tidal radius , the orbit of the tidal tail merely represents the kinematic condition of the tail material just when it escapes from the satellite .
an example of a randomly chosen orbit in the leading tail is shown in figure [ fig : orbtc3 ] .
as expected , the orbit describes a rosette with its apocentre at the radius of the satellite orbit . the energy and angular momentum lost during the escape changes the conserved quantities of the ejecta orbits from that of the satellite orbit ; the leading ( trailing ) ejecta lose ( gain ) energy during deformation .
moreover , the distribution of the tails fills a wide region about the satellite orbit .
this reflects the broad distribution of phases for orbits at escape .
hence , the width of the tail is nearly the same as the distance between the apocentre and the pericentre of a typical rosette orbit .
each tail has several distinct _
streamers _ filling a common envelope .
the two primary streams in each tail demarcate the escape of the most extreme prograde and retrograde orbits .
the originally prograde orbits have lower specific angular momentum and , therefore , smaller pericentres and larger epicyclic amplitudes .
in contrast , originally retrograde orbits have larger pericentres and smaller epicyclic amplitudes .
distinct streamers result from the phase caustics near apocentre , similar to shells in elliptical galaxies caused by merger ejecta with a velocity dispersion much smaller than its new orbital velocity .
this mechanism , illustrated in figure [ fig : framesltc3 ] , is the _ massless _ description of tail formation .
this massless description assumes that the orbital energy and angular momentum of a tail is the same as those of a satellite ; this yields a simple easy - to - compute prescription for the tails location . but including the gravitational attraction of the satellite on the ejecta at all times.,title="fig : " ] but including the gravitational attraction of the satellite on the ejecta at all times.,title="fig : " ] but including the gravitational attraction of the satellite on the ejecta at all times.,title="fig : " ] but including the gravitational attraction of the satellite on the ejecta at all times.,title="fig : " ] ( compare with figure [ fig : framesltc3 ] left panel , and figure [ fig : framesltc2 ] middle panel , respectively ) .
the two steamers are clear in left panel but very weak in the right panel .
although very weak , the second streamers can be identified near the tidal radius .
, title="fig : " ] ( compare with figure [ fig : framesltc3 ] left panel , and figure [ fig : framesltc2 ] middle panel , respectively ) .
the two steamers are clear in left panel but very weak in the right panel .
although very weak , the second streamers can be identified near the tidal radius .
, title="fig : " ] in contrast , figure [ fig : framesltc2 ] repeats the simulation including the gravity of both the halo and the satellite at all times . at early times ( upper - right panel )
, the evolution is similar .
however , at later times ( lower panels ) , the effects of the satellite gravity are marked .
the continued acceleration of the tail by the satellite after escape decreases the internal velocity dispersion and narrows or focuses the tail as a consequence .
the streamers in figure [ fig : framesltc3 ] become less distinct when accelerated by the gravity of the satellite and the host halo together for the same reason ( see figure [ fig : frameslsate ] ) .
similarly , the acceleration of the ejecta by the satellite also decreases the angular separation between the streamers .
although the multi - streamer feature is diminished as the satellite gravitational field accelerates the ejecta , the feature can still be seen very close to the tidal radius . the importance of a satellite s gravity increases with mass and , therefore , we begin with a study of the tail produced by a massive satellite .
figure [ fig : framesmtc1 ] shows snapshots of a massive satellite ( @xmath63 ) on an circular orbit at @xmath16 , where once again the tail particles always feel the gravitational force from the satellite .
the overall evolution of the satellite and its disruption time is similar to the less massive satellite shown in figure [ fig : framesltc2 ] .
however , the long - term acceleration of the ejected material by the remaining satellite significantly alters these orbits .
as the tail continues to lose mass , the leading and trailing tails evolve to positions that are well inside and well outside the satellite s orbit and hence does not trace the satellite orbit at all @xcite .
the leading tail significantly tilts toward the centre of the halo and almost points directly there at late times .
the trailing tail is distributed throughout a wide annulus in the outer halo .
this difference results from the torque applied by the satellite _ well _ after escape .
orbits in the leading tail that lose energy and angular momentum fall toward the centre of halo , while orbits in the trailing tail gain energy and angular momentum and spread over a wide range of radii in the outer halo .
we show the tail morphology for our satellites with three different masses ( see table [ tbl : satellites ] ) on circular orbits at @xmath64 in figure [ fig : framescirc ] .
the tidal tails in the low - mass and tiny - mass satellites ( @xmath65 and @xmath66 , respectively ) very roughly follow the satellite orbit , with the leading and trailing tail located inside and outside of the satellite orbit .
compared to figure [ fig : framesltc3 ] , it is clear that the differences decrease with the satellite mass . as we described in section
[ sec : simulation ] , the low - mass satellite corresponds to the sagittarius dwarf spheroidal galaxy halo and the tiny - mass satellite corresponds to the draco dwarf spheroidal galaxy halo .
figure [ fig : compcirc ] shows the evolution of the distance from the host halo centre and the satellite s gravitational potential for an ensemble average of 10 randomly sampled particles near the tip of the leading tail in the three satellites .
the tail from a massive satellite receives a larger torque and a larger shift to smaller energies and angular momentum than the tail from a lower - mass satellite .
the bottom panel in figure [ fig : compcirc ] shows that the decay results from interactions with the satellite potential .
figure [ fig : compcirc ] also shows that the satellite potential remains important in the low - mass and tiny - mass satellites when the tail is close to the satellite but it is unimportant when the tail is far from satellite .
the satellite potential always remains significant for the massive satellite tail .
the long - term influence of the satellite on the tail morphology makes any inference of the satellite orbit from the tidal tail impractical , especially for satellites on non - circular orbits ( see section [ sec : ecctail ] ) .
the leading tail from the low - mass and the tiny - mass satellites in figure [ fig : framescirc ] exhibits kinks .
the kinks are a consequence of the epicyclic motion of the tail orbits and of acceleration by the satellite at subsequent apocentres .
figure [ fig : orbtc2 ] shows the ensemble averaged distance and positions for a sample of leading tail particles orbits taken from the low - mass satellite simulation shown in figure [ fig : framesltc2 ] .
the kink occurs at the first apocentre of the ejecta , after it is decelerated by the satellite during and subsequent to its escape .
the deceleration during escape tends to correlate the phases of the ejected orbits and results in a narrowing of the tidal tail s width .
in contrast , the satellite potential accelerates the trailing tail particles , which increases the peri- and apocentres of the trailing tail .
the analogous kink in the trailing tail is not so obvious because of its lower orbital frequencies .
however , a plot analogous to figure [ fig : orbtc2 ] does show a similar oscillation with lower angular frequency .
the large changes in the orbits of escaping particles orbits are easily understood using a restricted three - body approach .
consider a satellite of mass @xmath67 in circular orbit at galactocentric radius @xmath68 in a halo of mass @xmath69 . in the frame of reference moving with a satellite of vanishingly small mass ,
the effective potential is symmetric about the satellite centre .
although orbital energy and angular momentum are not conserved , this system admits a conserved quantity , the jacobi constant : @xmath70 where @xmath71 and @xmath72 are the orbital energy and angular momentum and @xmath73 is the satellite s angular frequency about the host halo .
this expression is easily derived by identifying a perfect time derivative in the inner product of the velocity vector and newton s equations of motion in the rotating frame of reference ( * ? ? ?
* section 3.3.2 ) .
an isocontour of the jacobi constant passes through the x - points , @xmath9 , and demarcates the bound and unbound trajectories as shown in figure [ fig : jacobi]a . as the satellite mass increases
, the inversion symmetry about the satellite centre is broken and the unstable points separate as shown in figure [ fig : jacobi]b . for small - mass satellites ,
therefore , the tidal force is symmetric about the satellite centre leading to symmetric tidal tails as seen in globular clusters .
however , for large - mass satellites , the asymmetry in the tidal force leads to asymmetric mass loss .
now consider the mass lost through the inner ( outer ) critical point , @xmath9 .
such orbits will have an inward ( outward ) velocity and unbound values of the jacobi constant .
the force from the satellite continues to affect the orbit beyond the tidal radius in this restricted problem as in the n - body simulations .
moreover , the smaller the mass of the satellite , the closer the radius is to that of the satellite , and the ejected orbit
_ lingers _ near the original satellite orbit , partly offsetting the smaller gravitational force .
for this reason , the orbit does not take on the orbital actions of the satellite but continues to be torqued by the satellite .
one may estimate the scaling of this energy change by computing the work done in the satellite frame on the escaping tail particle ; this naturally takes into account the lingering .
begin with the standard restricted three - body problem with generalised forces .
assuming that the satellite orbits in the @xmath74-@xmath75 plane and using hamilton s equations , one may compute the @xmath76-component torque on an escaping particle and the change in angular momentum of the escaping particle after an interval @xmath57 becomes @xmath77 where @xmath78 is the hamiltonian , @xmath79 is the azimuthal coordinate conjugate to @xmath80 and @xmath81 is the gravitational potential of the satellite .
the second equality in equation ( [ eq : ejtorque ] ) owes to the @xmath79 independence of all the other terms in @xmath78 .
we may consider an escaping orbit in the limit that the mass of the satellite @xmath67 is much smaller than the mass of halo @xmath69 and use perturbation theory to evaluate equation ( [ eq : ejtorque ] ) . to do this ,
let the unperturbed orbit be the circular orbit that passes through the x - point , @xmath9 at @xmath82 .
expanding to lowest contributing order in @xmath83 , after some straightforward algebra and taking the limit @xmath84 , one may show that @xmath85 where @xmath86 is the azimuthal orbital frequency .
finally , it follows that @xmath87 from the conservation of the jacobi constant ( equation [ eq : jacobi ] ) which yields : @xmath88 since @xmath89 , @xmath68 , and @xmath90 are constant , equation ( [ equ : scale ] ) implies that the work done is proportional to @xmath0 . in other words , the change in the orbital energy of the escaping particle decreases as the satellite mass decreases but only weakly !
-axis is the ratio of satellite mass @xmath67 to total halo mass @xmath69 and the @xmath75-axis is the magnitude of the energy change , @xmath91 .
the straight line is the relation @xmath0 . ] , and the jacobi constant , @xmath92 , for two ensembles of particles ejected from the low - mass satellite selected from the leading and trailing tail , respectively .
the value @xmath92 is nearly conserved while the energy changes owing to work done during escape . changes in @xmath92 are caused by mass loss and the resulting evolution of the satellite s gravitational potential . ] although the derivation of the scaling assumes @xmath93 , we demonstrate numerically that it applies over all values of interest by integrating the equations of motion in the rotating potential .
we adopt @xmath94 and choose values of the jacobi constant that are 1% larger than the critical value passing through @xmath9 with zero velocity .
the initial motion , in the rotating frame , is along ( or against ) the direction of rotation for inner ( outer ) escapees .
figures [ fig : m10em4][fig : m10em2 ] show the resulting trajectories and conserved quantities for @xmath95 . for inner ( outer ) escapees , the energy and angular momentum decrease ( increase ) after the initial transient for @xmath96 .
figure [ fig : scaletest ] shows that the energy change for ensembles of orbits in the leading tail chosen as follows . the initial position is chosen to be 2% of @xmath9 outside of the x - point and the velocities
are chosen to have a normal distribution in the satellite frame with a dispersion that is 2% of the satellite s circular velocity at @xmath9 .
the orbits in figures [ fig : m10em4][fig : m10em2 ] are representative members of these ensembles .
the magnitude of the energy change @xmath91 is defined as the ensemble average of @xmath97 where @xmath98 is the initial energy and @xmath99 is the minimum energy along the orbit .
these numerical values are consistent with the predicted scaling @xmath0 .
circumstance may eventually bring the ejected particle close to the satellite once again , as seen in figure [ fig : m10em2 ] .
the radial extent of the annulus covered by the orbit increases only gradually with increasing satellite mass , reflecting the same weak dependence on @xmath83 . in the simulations ,
the mass loss and subsequent reequilibration of the satellite potential causes small deviations in the jacobi constant even though the satellite orbit remains circular as shown in figure [ fig : jacobie ] .
nonetheless , the restricted three - body dynamics explains most of the features seen in the orbit evolution .
although we still expect some of the insight gained from the restricted three - body problem to carry over to the evolution of a satellite on an _ eccentric _ orbit , this more complex situation requires direct simulation .
figure [ fig : framesecc ] shows the tidal tails of a massive and low - mass satellite on an eccentric , @xmath23 , orbit .
the tidal tail morphology for eccentric satellite orbits is significantly more complex than for circular satellite orbits and varies more strongly with satellite mass . in the left panel of figure [ fig : framesecc ] ,
the massive satellite has dramatically decelerated the leading tail , which now reaches the host halo centre and forms an inner `` reservoir '' of ejecta .
the deceleration by the satellite causes the leading tail to appear close to radial . in the right panel of figure
[ fig : framesecc ] , the multiply segmented tail from the low - mass satellite is caused by two mechanisms .
first , during each satellite orbit , the leading tail forms during the approach to pericentre .
after pericentre , the tidal strain and the mass - loss rate diminishes resulting in a gap in the tail .
second , deceleration by the satellite changes the orbits of the newly disconnected leading tail , producing a distinct segment .
but for the massive satellite on an @xmath26 orbit at @xmath52 , @xmath53 , @xmath54 , and @xmath55 in the top - left , top - right , bottom - left panels , and bottom - right panels , respectively.,title="fig : " ] but for the massive satellite on an @xmath26 orbit at @xmath52 , @xmath53 , @xmath54 , and @xmath55 in the top - left , top - right , bottom - left panels , and bottom - right panels , respectively.,title="fig : " ] but for the massive satellite on an @xmath26 orbit at @xmath52 , @xmath53 , @xmath54 , and @xmath55 in the top - left , top - right , bottom - left panels , and bottom - right panels , respectively.,title="fig : " ] but for the massive satellite on an @xmath26 orbit at @xmath52 , @xmath53 , @xmath54 , and @xmath55 in the top - left , top - right , bottom - left panels , and bottom - right panels , respectively.,title="fig : " ] but for times @xmath100 , @xmath101 , and t=@xmath54 from left to right , respectively , as the the satellite moves from pericentre to apocentre .
owing to the deceleration by the satellite potential , the leading tail falls toward the centre of the host halo.,title="fig : " ] but for times @xmath100 , @xmath101 , and t=@xmath54 from left to right , respectively , as the the satellite moves from pericentre to apocentre .
owing to the deceleration by the satellite potential , the leading tail falls toward the centre of the host halo.,title="fig : " ] but for times @xmath100 , @xmath101 , and t=@xmath54 from left to right , respectively , as the the satellite moves from pericentre to apocentre .
owing to the deceleration by the satellite potential , the leading tail falls toward the centre of the host halo.,title="fig : " ] figure [ fig : framesmtk ] shows the evolution of a massive satellite on an e=0.74 orbit .
initially , the leading tail points directly toward the halo centre but the strong deceleration by the satellite eventually fills the inner halo with ejecta .
figure [ fig : framesmtk2 ] provides a finer time sampling of the evolution between pericentre and apocentre for the same simulation .
instantaneously , the morphology can be very complex and the position angle of the leading tail can vary significantly from its nearly radial average . there is little correlation between the tail location and the satellite orbit .
the location of the inner ejecta , e.g. its outer turning points , is determined by the host halo potential , the time - varying satellite potential , and the satellite orbit in combination . therefore , unlike streams from very low - mass satellites , the tail orientation is not directly informative .
however , through dynamical modelling , the location of the inner ejecta may provide constraints on combinations of satellite properties and its history , and the galaxy potential .
we have demonstrated that tail morphology depends sensitively on the satellite mass and orbit . for modest to high - mass satellites , the ejected tails have orbits that differ significantly from that of the progenitor satellite . in this section ,
we illustrate the observational implications of these results .
the observational implications for milky way streams can be summarised by projecting the tail star counts and radial velocity signatures against the sky with the observer at the centre of host halo .
figure [ fig : aitoff ] shows aitoff projections of number density and mean radial velocity for the massive satellite ( top panels ) and the low - mass satellite ( bottom panels ) with an @xmath23 orbit ( the same simulations described in figure [ fig : framesecc ] at the same time , @xmath102 ) .
the aitoff projection covers the entire sky , @xmath103 and @xmath104 , and the pixel size is @xmath105 .
the number density of the particles ( left panels ) and the mean radial velocity ( right panels ) are coded by colour .
the contours in all the panels represent the particle number density .
velocity outliers at low number density are trimmed by setting to @xmath106 all the pixels with fewer than 10 particles .
the satellites are located at @xmath107 and @xmath108 and move in the positive @xmath109 direction .
the radial velocity signatures of the massive and low - mass satellites are distinctly different .
these qualitative differences are a direct consequence of the large energy and angular momentum changes of the ejecta orbits leading to the phase wrapping of the leading tail and the dramatic broadening of the trailing tail ( see section [ sec : ecctail ] ) .
this causes the lower overall mean velocity values with a more rapid angular variation around the sky .
in contrast , the mean velocity of the leading and trailing tails for the low - mass satellite are smooth and slowly vary around the sky . quite clearly
, the debris from the massive satellite will not show the distinct kinematic and spatial signatures that have been exploited in recent observational campaigns .
near @xmath110 and @xmath111 , one observes a region of receding orbits surrounded in longitude by regions of approaching orbits .
figure [ fig : framesecc ] ( right snapshot ) shows that line - of - sight projections will encounter strong leading and trailing tails from same satellite at different radii .
the closer the tail to the observer , the larger the angular extent perpendicular to the motion of the stream .
their velocity signature in the aitoff projection occurs as a single line of sight cuts through these tails at different radii .
the aitoff projections contain most of the information that one might obtain from combined kinematic
photometric surveys such as rave @xcite .
in particular , these results show that tail morphology depends on satellite mass .
therefore , a wide range of kinematic `` template '' models may be required to best exploit the information implicit in observed halo stars . and are centred at a longitude of @xmath112 .
the width of the distribution in @xmath113 increases with satellite mass . for comparison ,
the satellite trajectories for halos with flattenings of @xmath114 are also shown .
although all three simulations are performed in a spherical host halo potential , the tail locus is better matched for a satellite trajectory with @xmath115 . ]
radial velocity
orbital longitude diagrams are frequently used to characterise large - scale kinetic features in the milky way .
figure [ fig : vr_l ] shows radial velocity
orbital longitude diagrams for the ejecta of satellites with orbits having @xmath23 for each of our three masses .
we convert simulation units to milky way units by assuming a virial radius of 250 kpc and a total mass of @xmath116 @xcite . in figure
[ fig : vr_l ] , the sun has @xmath117 kpc and the galactic plane and the satellite s orbital plane are coincident .
here we adopt the sagittarius longitudinal coordinate system described in @xcite for the orbital longitude .
all satellites have @xmath118 kpc and move in the @xmath75 direction . therefore
, the satellite has @xmath119 and longitude increases along the trailing tail ( in the @xmath120 direction ) .
the radial velocity is measured from the halo centre .
the spread in @xmath121 is proportional to the satellite mass , as expected from the previous discussion and hence the mean velocity will be an unbiased diagnostic of the satellite orbit only for very low mass satellites .
although we have only modelled the dark matter , it is likely that the @xmath122 space distribution for stellar and dark matter ejecta will be similar in most cases since the internal satellite velocity dispersion plays only a minor role in shaping the ejecta distribution .
@xcite use m giants from the two micron all - sky survey ( 2mass , * ? ? ?
* ) to map the position and velocity distributions of tidal debris from the sagittarius dwarf spheroidal galaxy .
assuming that tidal tails approximately align with the satellite trajectory , the authors note that the radial velocity distribution of tidal debris suggests an prolate milky way halo with an axis ratio of @xmath123 .
however , our results demonstrate that the tails do _ not _ follow the satellite orbit . in figure
[ fig : vr_l ] , we also plot satellite trajectories for three different halo flattenings to compare with the particle distributions of our simulations evolved in a spherical host halo .
following @xcite , we flatten our host halo parallel to the satellite s motion and compute point - mass satellite trajectories to compare with our simulated @xmath124 diagrams .
surprisingly , the distributions of the low - mass satellite and the tiny - mass satellite tidal tails most closely matches a @xmath125 halo
. the gravitational acceleration by the satellite shifts the tail location in the radial velocity distribution and this trend is degenerate with the effects of halo flattening .
for instance , the location of the leading tails decelerated by a massive satellite is degenerate with the trajectories of tails in an oblate halo with no satellite deceleration .
we have not attempted to model the milky way in sufficient detail to estimate the halo flattening including satellite deceleration .
however , the degeneracy between halo flattening and the shift caused by the satellite gravitational acceleration suggests that the @xcite conclusions may be biased and a more careful analysis including the full dynamics of the halo - satellite interaction is necessary .
several groups have proposed phase - space - based detection diagnostics for moving groups associated with disrupted dwarf galaxy and star cluster streams . @xcite
proposed using the intrinsic correlation of moving groups radial energy and galactocentric radius to identify disrupted systems .
the procedure is as follows . assuming a spherical gravitational potential for the outer galaxy , one estimates the radial energy @xmath126 and the galactocentric radius @xmath127 of the putative ejecta stars from observations .
then , assuming that all of the debris from a single satellite has the same orbital energy , @xmath71 , and angular momentum , @xmath128 , conservation of energy implies a simple linear relationship in @xmath129 : @xmath130 .
hence , linear features in the observed @xmath131@xmath129 diagram indicate the detection of a tidal stream .
recently @xcite used this method to support the detection of stellar streams in the sloan digital sky survey . , plotted against inverse galactocentric radius , @xmath129 , for the low - mass satellite with an @xmath23 orbit .
particles with the same energy and angular momentum as the satellite will lie on a particular straight line .
the zero velocity curve describes the lowest possible energy at a given radius for this host halo .
the material bound to the satellite follows the straight locus at all times ( top left ) .
the escaped particles are shown at @xmath132 ( top right ) , @xmath133 ( bottom left ) and @xmath134 ( bottom right ) , respectively .
only one particle out of 250 are plotted for visibility . because the leading ( trailing ) tail loses ( gains ) energy , the ejecta deviates from the predicted straight line with significant scatter .
times in gyrs are scaled to the milky way . ]
but for the tiny - mass satellite .
] however , as we have now seen , a massive satellite will modify the conserved quantities of the ejecta orbits and change their location in @xmath1 space .
figures [ fig : er_dsph2 ] and [ fig : er_dsph3 ] show the @xmath1 diagrams for the low - mass and tiny - mass satellite simulations on an @xmath23 orbit . for clarity ,
we have reduced the point density by randomly sampling the simulation phase space and plot the bound particles at five different times in the upper - left panels .
we calculate the expected linear relation from the satellite s initial position and velocity .
the bound material in low - mass and tiny - mass satellites lies along the predicted linear relation at all times .
we plot the tail particles at three different times in the other three panels .
as one can see in figure [ fig : er_dsph2 ] , the deviation of the tail particles from the predicted locus and the scatter in @xmath71 at fixed @xmath129 for the low - mass satellite is large . especially at late times , e.g. the bottom right panel in figure [ fig : er_dsph2 ] , the tail nearly fills the region between the zero velocity curve and the predicted locus .
however , one can see from figure [ fig : er_dsph3 ] that tail particles from the tiny - mass satellite do follow the predicted linear relation .
we conclude that the @xcite diagnostic can only detect streams from very low - mass satellites such as globular clusters . orbit and its ejected tail particles plotted in @xmath71@xmath80 space .
_ top left : _ the phase - space distribution at @xmath52 and @xmath134 for the bound satellite and its tail ( subsampled as in figure [ fig : er_dsph2 ] ) .
density plots of the phase - space distribution are shown at @xmath52 ( top right ) , at @xmath134 ( bottom left ) and at @xmath134 for the ejected tail particles alone ( bottom right ) . ]
but for the tiny - mass satellite ] motivated by the prospect of six - dimensional phase - space data from future astrometric missions , @xcite proposed to identify phase - mixed satellite debris by a cluster analysis in @xmath135 space .
we explore the consequences of tail evolution on this approach using the same two simulations in figure [ fig : psp_dsph2 ] for the low - mass satellite and in figure [ fig : psp_dsph3 ] for the tiny - mass satellite . for simplicity
, we assume that we know the orbital orientation and consider only the @xmath71@xmath80 projection .
the top - left panels show the distribution at 0 gyr ( @xmath136 ) and at 8 gyr ( @xmath64 ) when scaled to the milky way halo in @xmath71@xmath80 space . once again , we randomly sampled the material to improve clarity .
the top - right and bottom - left panels show density estimates in @xmath71@xmath80 space for the satellite at 0 gyr and at 8 gyr , respectively . in the bottom - right panels we show the density of only the tail particles at 8 gyr .
the overall position of the satellite and its tail changes little from 0 gyr to 8 gyr , although the shape of the distribution shifts . in both figures , there are two or more peaks at high and low energy with respect to the satellite owing to decelerations and accelerations of tail particles by the satellite potential .
moreover , the tidal field is nonaxisymmetric and this leads to spatial correlations in the energy and angular momentum of the least bound satellite particles , which in turn leads to the production of several apparently disassociated phase - space clumps before disruption .
the observational detection of `` s''- or `` z''-shaped tidal tails in globular clusters ( e.g. * ? ? ? * ; * ? ? ? * ; * ? ? ?
* ) promises sensitive statistical tests of the galaxy s gravitational potential and has renewed the quest for streams from larger satellites . for globular clusters , i.e. very low mass satellites ,
the tidal tail morphology is easily interpreted .
@xcite however , for massive satellites , the bisymmetry that leads to this simple morphology is broken by the interaction of the host halo s gravitational field and the self - gravity of the satellite itself .
we present new dynamical aspects and morphologies of tidal tails produced in satellites of significant mass , @xmath137 .
there are two dynamical principles that affect the tail production for massive satellites .
first , the leading and trailing x - points , points where the attractive force of the host halo and satellite are balanced at zero velocity , do not occur at equal distances from the centre nor do they have the same equipotential value for large - mass satellites ( see figure [ fig : jacobi ] ) .
second , the escaped ejecta in the leading ( trailing ) tail continues to be decelerated ( accelerated ) by the satellite s gravity leading to large offsets of the ejecta orbits from the satellite s original orbit ( see figure [ fig : framesecc ] ) .
we show that this is consistent with hill - jacobi theory ( generalised to dark - matter halos ) for satellites on circular orbits . in particular
, the effect of the satellite s self gravity on the tail decreases only weakly with decreasing satellite mass , proportional to @xmath0 ( see section [ sec : circtail ] ) and , therefore , the acceleration by the satellite after escape is important for dwarfs and dark halos of modest mass .
these findings have several important and useful theoretical and observational consequences .
first , for a finite mass satellite , the morphology of the leading and trailing tails will be different owing to the gradient in the underlying halo potential across the satellite .
in addition , the tail ejection occurs over a range of azimuth relative to the x - point owing to the dynamical response of the originally prograde and retrograde orbits to the tidal and non - inertial acceleration .
these effects should be observable in high resolution imaging for both dwarf spheroidal and globular clusters ( see figs .
[ fig : framesecc][fig : framesmtk2 ] ) .
second , the radial velocity of tail particles will be displaced from that of the satellite orbit .
the magnitude of the displacement is proportional to the satellite mass .
these trends distort the ejecta from the gravitationally bound satellite trajectory in the @xmath122 plane in much the same sense as a satellite trajectory in a flattened halo ( see figure [ fig : vr_l ] ) .
in other words , in fitting the @xmath124 diagram for tidal tails to satellite orbits of different flattenings , the satellite mass is covariant with halo flattening , i.e. the _ shape _ parameter @xmath138 .
therefore , a constraint on the milky way halo shape using tidal streams requires mass - dependent modelling . finally , the acceleration of ejecta by a massive satellite during escape spreads the velocity distribution and obscures the signature of a well - defined `` moving group '' in phase space ( see figs . [
fig : er_dsph2][fig : psp_dsph3 ] ) .
although we believe that the physical effects described in this study are robust , our intentionally idealised simulations ignore several possibly relevant processes .
first , the dynamical friction and the self - gravitation of the tail are ignored , although in all but the most extreme mass satellites their effects on the tail morphology will be negligible , since the mass in the tail is very small .
second , we assume a smooth and static spherical host halo potential . in reality , over time , as the host halo mass grows its shape may change , and the ejecta will be perturbed by substructure .
these time dependent effects will not affect the applicability of the dynamics described here but will complicate the prediction of observational signatures .
finally , we have not included the physics of a dissipational baryonic component that may have slightly different kinematics than that of a dark collisionless component . in spite of these shortcomings
, our study elaborates the details of satellite tidal tail production and the dynamics that bear on the interpretation of observed streams . as an example ,
@xcite and @xcite find that satellite tails follow the satellite orbit for dwarf galaxies whose mass is negligible compared to the galaxy mass .
the mass of these satellites is usually similar to or less than our tiny - mass satellite . using these simulation results
, @xcite developed an efficient numerical method to investigate the detectability and interpretation of tidal debris tails .
however , we have demonstrated here that the gravity of the satellite for @xmath139 will change the actions of the tidal ejecta to mimic halo flattening ( see section [ sec : vr - l ] , figure [ fig : vr_l ] ) .
in addition to the spatial distribution , the velocity distribution of the tail is affected by the satellite potential ( see section [ sec : psp ] ) .
we would like to note that @xcite have also noticed a systematic distance offset for leading ( trailing ) tails inside ( outside ) the orbit of a satellite owing to the satellite potential .
however , they did not focus on the tail morphology . in summary
, we have shown that the interplay between the satellite and the host halo results in a complex tail morphology whose amplitude scales weakly with mass .
although these findings complicate the interpretation of stellar streams and moving groups , the intrinsic mass dependence provides additional leverage on both the halo and on the progenitor satellite properties .
a statistical study of these trends will further constrain the dark halo potential and the mass accretion history of the milky way .
we would like to thank an anonymous referee for many useful comments .
this work was supported in part by nsf ast-0205969 , and nasa atp nag5 - 12038 , nags-13308 , & nng04gk68 g . | we investigate the dynamical mechanisms responsible for producing tidal tails from dwarf satellites using n - body simulations . we describe the essential dynamical mechanisms and morphological consequences of tail production in satellites with masses greater than 0.0001 of the host halo virial mass .
we identify two important dynamical co - conspirators : 1 ) the points where the attractive force of the host halo and satellite are balanced ( x - points ) do not occur at equal distances from the satellite centre or at the same equipotential value for massive satellites , breaking the morphological symmetry of the leading and trailing tails ; and 2 ) the escaped ejecta in the leading ( trailing ) tail continues to be decelerated ( accelerated ) by the satellite s gravity leading to large offsets of the ejecta orbits from the satellite orbit .
the effect of the satellite s self gravity decreases only weakly with a decreasing ratio of satellite mass to host halo mass , proportional to @xmath0 , demonstrating the importance of these effects over a wide range of subhalo masses .
not only will the morphology of the leading and trailing tails for massive satellites be different , but the observed radial velocities of the tails will be displaced from that of the satellite orbit ; both the displacement and the maximum radial velocity is proportional to satellite mass .
if the tails are _ assumed _ to follow the progenitor satellite orbits , the tails from satellites with masses greater than 0.0001 of the host halo virial mass in a spherical halo will _ appear _ to indicate a flattened halo .
therefore , a constraint on the milky way halo shape using tidal streams requires mass - dependent modelling .
similarly , we compute the the distribution of tail orbits both in @xmath1 space ( lynden - bell & lynden - bell 1995 ) and in @xmath2 space ( helmi & de zeeuw 2000 ) , advocated for identifying satellite stream relics .
the acceleration of ejecta by a massive satellite during escape spreads the velocity distribution and obscures the signature of a well - defined `` moving group '' in phase space . although these findings complicate the interpretation of stellar streams and moving groups , the intrinsic mass dependence provides additional leverage on both halo and progenitor satellite properties .
galaxies : evolution galaxies : interaction galaxies : haloes galaxies : kinematics and dynamics method : n - body simulation method : numerical |
10 r. brak and j. w. essam , 4183 ( 2004 ) r. brak , j. de gier and v. rittenberg , 4303 ( 2004 ) f. h. jafarpour and s. zeraati , 011119 ( 2010 ) b. derrida , m.r .
evans , v. hakim and v. pasquier 1493 ( 1993 ) g. m. schtz , 4265 ( 1993 ) h. hinrichsen , 3659 ( 1996 ) r. a. blythe and m. r. evans , r333 ( 2007 ) f. h. jafarpour and s. r. masharian , 051124 ( 2009 ) | we have shown that the partition function of the asymmetric simple exclusion process with open boundaries in a sublattice - parallel updating scheme is equal to that of a two - dimensional one - transit walk model defined on a diagonally rotated square lattice .
it has been also shown that the physical quantities defined in these systems are related through a similarity transformation .
recently , much attention has been devoted to the study of two - dimensional walk models and their relations with the one - dimensional driven - diffusive systems @xcite .
the reason is that the partition function of some of these walk models can be related to those of some of the one - dimensional out - of - equilibrium driven - diffusive systems with open boundaries .
one of the most simple and important system of this type is the asymmetric simple exclusion process ( asep ) with open boundaries @xcite .
+ the asep is a widely studied driven - diffusive system which reveals a wealth of interesting critical phenomena .
the asep is defined on a one - dimensional discrete lattice .
the classical particles are allowed to enter the lattice only from the left boundary and leave the lattice only from the right boundary .
the particles also hop unidirectionally toward the right boundary .
it is known that a single product shock measure has a simple random walk dynamics in this system , without any constraints on the parameters of the system , provided that we consider a discrete - time updating scheme . in this case
the particles are injected into the leftmost lattice site with the probability @xmath0 if the target site is empty .
they are also extracted from the rightmost lattice site with the probability @xmath1 if it is already occupied .
we assume that in the bulk of the lattice a particle at a given site @xmath2 deterministically move to the site @xmath3 provided that it is not occupied by other particles , which means that double occupation is prohibited .
+ in this paper we consider the asep with open boundaries in the sublattice - parallel update @xcite .
we are looking for a two - dimensional walk model which shares similar critical properties with the asep .
it turns out that there exists such a two - dimensional walk model as we will describe it later .
these systems have equal partition functions and the physical quantities in these systems are related through a simple transformation .
+ the asep is defined on a lattice of length @xmath4 .
the time evolution of the probability distribution vector @xmath5 is given by the following master equation : @xmath6 the sublattice - parallel updating scheme is defined as follows : we divide the bulk dynamics consists into two half time steps . in the first half time
step the pairs of neighboring sites ( @xmath7 ) for @xmath8 and also the first and the last lattice sites are updated . in the second half time
step the pairs of neighboring sites ( @xmath9 ) for @xmath10 are updated .
considering this definition , the transfer matrix @xmath11 in ( [ time evol ] ) is given by the multiplication of two factors @xmath12 defined as : @xmath13 where @xmath14 , @xmath15 and @xmath16 are given by : @xmath17 the matrix @xmath18 is written in the basis @xmath19 when @xmath20 stands for an empty lattice site and @xmath21 stands for an occupied lattice site .
the matrices @xmath22 and @xmath23 are also written in the basis @xmath24 . + following @xcite we define two different types of product shock measures @xmath25 ( @xmath10 ) and @xmath26 ( @xmath27 ) as follows : @xmath28 note that the position of the shock in @xmath25 lies between two consecutive sites @xmath29 and @xmath30 . at the same time
the shock position lies between two consecutive sites @xmath30 and @xmath31 in @xmath32 .
the authors in @xcite have shown that the time evolution of the shock position in ( [ shocks ] ) governed by ( [ time evol ] ) is similar to that of a simple random walker .
the shock position hops to the left and to the right with the probabilities @xmath33 and @xmath34 respectively provided that : @xmath35 the steady - state of the system @xmath36 which obeys @xmath37 can be written as a superposition of the shocks defined in ( [ shocks ] ) as follows : @xmath38 provided that : @xmath39 in which @xmath40 .
the partition function of the system @xmath41 ( which is also the normalization factor ) is also given by : @xmath42 in the steady - state and depending on whether @xmath33 is larger or smaller than @xmath34 , the system undergoes a phase transition .
the coexistence line occurs where @xmath43 . in both phases
the density of particles on the lattice have exponential behaviors while on the coexistence line it is linear because of considering a superposition of shocks with unbiased random walk dynamics .
+ in @xcite the author has rigorously shown that the steady - state of the asep in a sublattice - parallel update can be obtained using a matrix product method recently reviewed in @xcite . according to this method
we assign an operator to each state of a lattice site in the steady - state .
the two operators @xmath44 and @xmath45 are associated with the presence of a particle and a vacancy at an even lattice site respectively .
the two operators @xmath46 and @xmath47 are also associated with the presence of a single particle and a vacancy at an odd lattice site respectively . for a system of length @xmath4
we associate a product of @xmath4 operators to each configuration of the system consisting of particles and vacancies at different lattice sites .
the unnormalized weight associated with any configuration can be obtained by considering a matrix element of this product .
+ using the matrix product method and in the same basis mentioned above one can rewrite the steady - state of the asep as follows : @xmath48^{\otimes
n } \vert v \rangle \rangle\ ] ] in which the normalization factor is given by : @xmath49 it can be verified that the following two - dimensional matrix representation generates the same steady - state introduced in ( [ ss1 ] ) : @xmath50 given that : @xmath51 note that the two - dimensional matrix representation ( [ matrixrep ] ) also generates the same partition function ( [ z1 ] ) using ( [ z2 ] ) .
in comparison to the matrix representation introduced in @xcite we have adopted an upper triangular matrix representation .
+ in what follows we will define a one - transit walk model consisting of @xmath52 steps on a diagonally rotated square lattice and explain how the partition function of this system can be related to that of the asep studied above .
we will then discuss how the physical quantities in both systems are related through a similarity transformation .
+ consider a random walker moving on a two - dimensional diagonally rotated square lattice .
the random walker starts from the origin @xmath53 and takes @xmath52 consecutive steps until it gets to the point @xmath54 .
the random walker moves in the north - east ( ne ) or the south - east ( se ) direction .
the random walker never takes two consecutive steps in the ne direction along the path ; however , it can take two consecutive steps in the se direction only once during the journey . in this case the random walker moves along a path which crosses the horizontal axis only once .
a typical one - transit path is given in figure ( [ fig1 ] ) .
sketch of a simple one - transit path.,width=288 ] let us assume that the random walker moves along a weighted path . we assign a fugacity to the steps that the random walker takes as follows : *
the fugacity @xmath55 to each step in se direction and the fugacity @xmath56 to each step in ne direction if the ending point is @xmath57 for an odd @xmath2 ( @xmath58 ) .
* the fugacity @xmath59 to each step in se direction and the fugacity @xmath60 to each step in ne direction if the ending point is @xmath57 for an even @xmath2 ( @xmath61 ) . *
the fugacity @xmath21 to all other steps . in order to find the weight of a given path one can simply multiply the fugacities of different steps of the path . the partition function of this system can be easily calculated and one finds : @xmath62 in which : @xmath63 the first term in ( [ z3 ] ) does not play any important role in the critical behavior of the system .
the second term in ( [ z3 ] ) , which is given explicitly in ( [ z4 ] ) , can be reinterpreted as the partition function of a two - dimensional walk model on a diagonally rotated square lattice ; however , one should assign the fugacities to the contact points ( instead of the steps ) in a different way .
we assign a fugacity @xmath64 to each contact point @xmath57 for an even @xmath2 and a fugacity @xmath65 to each contact point @xmath57 for an odd @xmath2 if the contact point is above the horizontal axis .
we also assign a fugacity @xmath66 to each contact point @xmath57 for an even @xmath2 and a fugacity @xmath67 to each contact point @xmath57 for an odd @xmath2 if the contact point is below the horizontal axis . in all cases
we do not assign any fugacity to the contact points with the horizontal axis if they come from the first upward step or the first downward step .
the partition function of this walk model given in ( [ z4 ] ) can now be written using a transfer matrix method as follows : @xmath68 in which the transfer matrix @xmath11 is a two - step transfer matrix defined as @xmath69 .
note that @xmath70 and @xmath71 are each related to two consecutive steps i.e. @xmath70 ( @xmath71 ) is the transfer matrix of two consecutive steps ending to the point @xmath57 for an odd ( even ) @xmath2 ; therefore , one can rewrite @xmath70 and @xmath71 as : @xmath72 by assigning four base vectors to the vertices and contact points of a path one can find a matrix representation for the transfer matrices and the vectors introduced in ( [ tmz ] ) and ( [ tms ] ) . the vectors @xmath73 and @xmath74 for @xmath75 are assigned to the vertices and the contact points of the path as shown in figure ( [ fig2 ] ) .
the base vectors associated with the vertices and contacts.,width=192 ] in an appropriate basis one can write : @xmath76 with the following completeness property : @xmath77 in which @xmath78 is a @xmath79 identity matrix . considering the definition of the weighted walk model
the only nonzero matrix elements are : @xmath80 all other matrix elements are zero . now using ( [ trans ] )
one can easily construct the transfer matrices in the above mentioned basis : @xmath81 and the vectors @xmath82 and @xmath83 are : @xmath84 before going any farther let us investigate the asymptotic behavior of the partition function of the one - transit walk model given in ( [ z4 ] ) in the large-@xmath85 limit .
it can be easily seen that in the large-@xmath85 limit the partition function shows two different behaviors depending on whether @xmath86 or @xmath87 which results in two different phases .
this means that the phase diagram of the walk model depends only on @xmath88 and @xmath89 and not the four independent fugacities .
it turns out that for the case @xmath86 most of the contacts with the horizontal axis will be from below while in the case @xmath87 they will be from above . on the coexistence line where @xmath90 the number of contacts both from above and below the horizontal axis will vary linearly along the horizontal axis .
+ at this point we show that how the partition function of the walk model ( [ z4 ] ) is related to that of the asep given in ( [ z1 ] ) .
it will also leads us to understand how the physical quantities in these systems are related .
let us assume that : @xmath91 now the following similarity transformation connects the two above mentioned systems : @xmath92 in which : @xmath93 it is also necessary to have the following relations : @xmath94 provided that @xmath95 and @xmath96 satisfy ( [ conditions1 ] ) which is achievable since @xmath97 , @xmath98 and @xmath99 are free to be chosen .
it is now easy to check that the partition functions of the two systems in different descriptions given by ( [ z1 ] ) , ( [ z2 ] ) and ( [ z4 ] ) are equal .
+ let us now investigate how the physical quantities in the walk model are related to those in the asep .
we start with the probability of finding a contact at the lattice site @xmath57 for an even @xmath2 from above @xmath100 or below @xmath101 the horizontal axis which is given by the following expressions : @xmath102 in which @xmath103 and @xmath104 are the contact operators at the lattice site @xmath57 from above and below the horizontal axis , respectively .
using the matrix representations ( [ rep1 ] ) and ( [ rep2 ] ) , it is easy to calculate the expressions ( [ contacts1 ] ) and ( [ contacts2 ] ) analytically . + one can also calculate the probability of finding a contact at the lattice site @xmath57 for an odd @xmath2 from above @xmath105 or below @xmath106 the horizontal axis which is given by the following expressions : @xmath107 the contact operators @xmath108 and @xmath109 have the same definitions mentioned above .
+ now we show that how the density of the particles in the asep is related to the probabilities of contacts given in ( [ contacts1])-([contacts4 ] ) . according to the matrix product formalism the density of the particles in an even site @xmath110 ( @xmath2 even ) in the asep in the sublattice - parallel update
is given by : @xmath111 using ( [ st ] ) , ( [ contacts1 ] ) and ( [ contacts2 ] ) and after some straightforward calculations it turns out that ( [ evenden ] ) can be written as : @xmath112 on the other hand , in the same system and using the matrix product formalism the density of the particles in an odd site @xmath110 ( @xmath2 odd ) is given by : @xmath113 using ( [ st ] ) , ( [ contacts3 ] ) and ( [ contacts4 ] ) it is not difficult to show that ( [ oddden ] ) can be written as : @xmath114 in @xcite the authors have already studied both the phase diagram and also the behaviors of ( [ deneven ] ) and ( [ denodd ] ) in the large-@xmath85 limit ; therefore , we do not repeat their results here and the reader can directly refer to those references . on the other hand ,
a description of the phase diagram of the same system based on the traveling shocks has been done in @xcite .
+ in this paper we aimed to show that the steady - state of the asep with open boundaries in a sublattice - parallel update can be described using a two - dimensional walk model .
we showed that the partition functions of these models are equal by properly defining the fugacities assigned to the steps ( or the contact points ) of the random walker . on the other hand
, the probability of finding a particles at a given lattice site in the asep is connected to the probability that the dyck path touches the horizontal axis at that point .
this is valid for both even and odd lattice sites as obtained in ( [ deneven ] ) and ( [ denodd ] ) . |
one of recent important developments in study of the ads / cft correspondence is the duality between gluon scattering amplitudes in @xmath5 super yang - mills theory and the area of the minimal surface in ads spacetime surrounded by the closed light - like wilson loops . from the duality one can compute the gluon scattering amplitudes at strong coupling . in the case of the 4-point amplitude , alday and
maldacena @xcite showed that the dimensionally regularized area agrees with the formula conjectured by bern , dixon and smirnov ( bds ) @xcite based on the perturbative analysis .
see @xcite for further developments .
the duality between gluon scattering amplitudes and wilson loops is shown to hold at weak coupling @xcite , which implies that the amplitude is invariant under dual conformal symmetry in momentum space @xcite . in superstring theory ,
this symmetry is interpreted as the symmetry of ads spacetime and invariance of the action under the combination of bosonic and fermionic t - duality@xcite .
the anomalous conformal ward identity constrains the structure of the 4 and 5-point amplitudes , which agrees with the bds formula .
but for higher @xmath1-point amplitudes there arises some ambiguities in the finite remainder part of the amplitude which can be written in terms of the conformal invariant cross - ratio of the external gluon momenta @xcite .
in fact , the explicit calculations of the two - loop 6-point gluon scattering amplitude and the hexagon wilson loop @xcite shows that they agree with each other but differ from the bds formula by finite term , which depends on three independent cross - ratios of the mandelstam variables .
this discrepancy from the bds ansatz was also observed at strong coupling by studying zigzag rectangular @xcite and a wavy circular wilson lines @xcite .
but the precise evaluation of the finite deviation from the bds formula is difficult to obtain since the exact solution of the minimal surface for higher - point amplitudes is not yet known . in a previous paper @xcite
, we constructed the minimal surfaces corresponding to the 4 , 6 and 8 point amplitudes numerically and evaluate the area in the radial cut - off regularization .
the light - like segments of the boundary is the same as the cut and glue type surface @xcite .
we showed that the numerical solutions differ from the cut and glue type surface and the area is consistent with the ir behavior of the amplitude . in this paper
we will study the area of the discretized surfaces for the 6 and 8-point amplitudes by applying conformal transformation and compare the area to the conjectured bds formula numerically .
this analysis gives a test of the duality between gluon scattering amplitudes and the wilson loops at strong coupling .
this paper is organized as follows : in section 2 , we review the radial cut - off regularization and a numerical approach to the construction of the minimal surface . in section 3 , we apply the conformal transformation to the 4-point amplitudes and compare it with the exact formula of the area in the radial cut - off regularization .
we propose a method to compare the numerical data with the bds formula without using the exact formula of the area in the radial cut - off regularization . in section 4
, we apply this method to the minimal surfaces corresponding to the 6 and 8-point amplitudes and compare the numerical solutions with the bds formula .
section 5 is devoted to discussion .
in this section we review the radial cut - off regularization of the minimal surface in ads spacetime and a numerical approach to get discretized version of minimal surface .
we consider the surface which is surrounded by the curve @xmath6 made of light - like segments @xmath7 .
this corresponds to the @xmath8-point gluon amplitude with on - shell momenta @xmath9 ( @xmath10 , @xmath11 ) .
the coordinates @xmath12 ( @xmath13 ) and the radial coordinate @xmath14 are the poincar coordinates in ads@xmath15 spacetime with the metric @xmath16 and @xmath17 is the radius of ads@xmath15 .
the nambu - goto action in the static gauge @xmath18 is given by @xmath19 here @xmath20 is the derivative with respect to @xmath21 ( @xmath22 ) .
the euler - lagrange equations become @xmath23 where @xmath24 is the lagrangian of the action . by solving these non - linear partial differential equations ,
one obtains the minimal surface @xmath25 and @xmath26 .
we consider the @xmath27-point amplitude for two incoming particles with momenta @xmath28 and @xmath29 and outgoing particles with momenta @xmath30 and @xmath31 .
for the momentum configuration in the @xmath32-space @xmath33 the wilson loop is represented by the square with corners at @xmath34 .
the boundary condition for the euler - lagrange equations is given by @xmath35 alday and maldacena @xcite found the exact solution of the nonlinear differential equations ( [ eq : euler1 ] ) , which is given by @xmath36 the above solution corresponds to the @xmath37 solution , where @xmath38 and @xmath39 are the mandelstam variables defined by @xmath40 and @xmath41 .
the general @xmath42 solution is obtained by scale and boost transformation of the @xmath37 solution : @xmath43 where @xmath44 is a parameter for the scale transformation and @xmath45 is a boost parameter .
after the conformal transformation , the momenta become @xmath46 the mandelstam variables @xmath38 and @xmath39 are given by @xmath47 using the dimensional regularization for the @xmath48-brane ( @xmath49 ) , the area is shown to agree with the bds formula at strong coupling @xcite . in this paper
we will use the radial cut - off regularization instead . in the radial cut - off regularization scheme
we introduce a cut - off @xmath50 in the radial direction@xcite . for general @xmath42 solution ,
the regularized area is surrounded by the cut - off curve @xmath51 in the @xmath52-plane : @xmath53 the action is evaluated by substituting the solution ( [ eq:4pt ] ) into ( [ eq : ngaction ] ) .
the result is @xmath54=\int_s dy_1 dy_2 { 1\over ( 1-y_1 ^ 2)(1-y_2 ^ 2)}. \label{eq : int1}\ ] ] where @xmath55 is the region surrounded by the curve @xmath51 .
we can put @xmath56 by rescaling @xmath57 . for fixed @xmath58
, @xmath59 takes the value in the range @xmath60 , where @xmath61 in ( [ eq : int1 ] ) , the integral over @xmath59 yields @xmath54=\int_{-\sqrt{{1-r_c^2\over 1-b^2 r^2_c}}}^{\sqrt{{1-r_c^2\over 1-b^2 r^2_c } } } dy_1 f(y_1,r_c ) , \label{eq : int4rcb}\ ] ] where @xmath62 expanding @xmath63 in @xmath50 we get @xmath64 after the integral over @xmath58 in ( [ eq : int4rcb ] ) , we obtain the action @xmath65 $ ] in the radial cut - off regularization .
we note that o(@xmath66 ) terms in ( [ eq : int4rcb ] ) also contribute a constant term .
we then obtain @xmath67&= & { 1\over4}\log^2\left({r_c^2\over -8\pi^2 s}\right ) + { 1\over4}\log^2\left({r_c^2\over -8\pi^2 t}\right ) -{1\over4 } \log^2({s\over t } ) + a_0 + o(r_c^2\log r_c^2 ) .
\nonumber\\ \label{eq : rad4pt1}\end{aligned}\ ] ] evaluating the constant @xmath68 numerically up to @xmath69 ( @xmath70 ) terms in ( [ eq : frc1 ] ) , we get @xmath71 this shows that the finite term numerically agrees with the bds formula @xcite @xmath72 since @xmath73 .
motivated from the analysis of the 4-point amplitude , the @xmath8-point amplitude is expected to have the structure @xcite @xmath74={1\over8}\sum_{i=1}^{n}\left(\log { r_c^2\over -8\pi^2 s_{i , i+1}}\right)^2 + f_n(p_1,\cdots , p_n)+o(r_c^2\log r_c^2 ) , \label{eq : nptactrc1}\ ] ] where @xmath75 and @xmath76 .
we have factored out the cusp anomalous dimension in the above formula .
the first term in ( [ eq : nptactrc1 ] ) characterizes infra - red divergences of the amplitude .
the function @xmath77 is a finite remainder part of the amplitude and takes the form @xmath78 the term @xmath79 is given by the bds formula which is written in terms of the mandelstam variables @xmath80}=(p_i+\cdots+p_j)^2.\ ] ] the explicit formula for @xmath81 @xcite is @xmath82 where @xmath83 - 1 } \log\left({-t^{[r]}_{i}\over -t^{[r+1]}_{i}}\right ) \log\left({-t^{[r]}_{i+1}\over -t^{[r+1]}_{i}}\right ) + d_{n , i}+l_{n , i}+{3\over2}\zeta_{2}. \label{eq : bds1}\ ] ] here @xmath84 and @xmath85 are defined by @xmath86}_{i } t^{[r+2]}_{i-1}\over t^{[r+1]}_i t^{[r+1]}_{i-1}}\right ) , \nonumber\\ l_{2m+1,i}&= & -{1\over2 } \ln\left({-t^{[m]}_{i}\over -t^{[m]}_{i+m+1}}\right ) \ln\left({-t^{[m]}_{i+1}\over -t^{[m]}_{i+m}}\right),\end{aligned}\ ] ] for @xmath87 and @xmath88}_{i } t^{[r+2]}_{i-1}\over t^{[r+1]}_i t^{[r+1]}_{i-1}}\right ) -{1\over2}{\rm li}_2\left(1-{t^{[m-1]}_{i } t^{[m+1]}_{i-1}\over t^{[m]}_i
t^{[m]}_{i-1}}\right ) , \nonumber\\ l_{2m , i}&= & -{1\over4 } \log\left({-t^{[m]}_{i}\over -t^{[m]}_{i+m+1}}\right ) \log\left({-t^{[m]}_{i+1}\over -t^{[m]}_{i+m}}\right),\end{aligned}\ ] ] for @xmath89 . @xmath90
denotes the dilogarithm function .
the term @xmath91 , called the remainder function , is a function of cross - ratios in momentum space : @xmath92 which represents a deviation from the bds formula .
the @xmath93 satisfies the anomalous dual conformal identities but the function @xmath91 is itself conformally invariant and is not determined by conformal symmetry . for the 6-point amplitude @xcite , the remainder part @xmath94 is a function of the cross - ratios @xmath95}t_{4}^{[2 ] } \over t_{1}^{[3 ] } t_{3}^{[3 ] } } , \nonumber\\ u_2&=&{x_{24}^2 x_{15}^2\over x_{25}^2 x_{14}^2 } = { t_2^{[2 ] } t_1^{[4]}\over t_2^{[3]}t_1^{[3 ] } } , \nonumber\\ u_3&=&{x_{35}^2 x_{26}^2\over x_{36}^2 x_{25}^2 } = { t_3^{[2 ] } t_2^{[4]}\over t_3^{[3]}t_2^{[3]}}.\end{aligned}\ ] ] the bds formula of the 6-point amplitude for specific momentum configurations will be discussed in sect .
4 . although exact formula for the minimal surface for @xmath96-point amplitudes is not yet known so far , we can study minimal surface by solving numerically the euler - lagrange equations on the square lattice with spacing @xmath97 where @xmath98 is a positive integer .
at each site @xmath99 ( @xmath100 ) , we assign the variables @xmath101=y_0(-1+h i,-1+h j),\quad r[i ,
j]=r(-1+h i,-1+h j).\ ] ] for the 4-point amplitude , we discretize the differential equations by the central difference method with the boundary conditions @xmath102= y_0(-1+hi,-1 ) , \quad y_0[i , m]=y_0(-1+hi,1 ) , \nonumber\\ & & y_0[0,j]= y_0(-1,-1+hj ) , \quad y_0[m , j]=y_0(1,-1+hj ) , \nonumber\\ & & r[i,0]=r[i , m]=r[0,j]=r[m , j]=0.\end{aligned}\ ] ] then we obtain @xmath103 nonlinear simultaneous equations for @xmath104 $ ] and @xmath105 $ ] and use newton s method to find a numerical solution . in this paper
we use the @xmath106 lattice data @xcite , where the newton method is repeatedly applied until the discrete equation is satisfied up to @xmath107 and the area of the obtained surface does not change up to @xmath108 .
the area is approximately evaluated as @xmath109h^2 $ ] , where @xmath110 $ ] and @xmath111 are the discretized lagrangian at a lattice point @xmath112 and the lattice spacing , respectively . the area @xmath55 becomes large as @xmath98 increases , which is due to the ir divergent behavior near cusps . in @xcite
we have defined the area of the surface in the radial cut - off regularization @xmath113=\sum_{(i , j)\in a[r_c]}l[i , j ] h^2,\ ] ] where @xmath114 $ ] denotes the set of lattice points @xmath112 satisfying @xmath115<r_c$ ] . in this paper , we calculate the area of the conformally boosted minimal surface by evaluating @xmath116=\sum_{(i , j)\in a[r_c , b]}l[i , j ] h^2 , \label{eq : discs4b}\ ] ] where @xmath117 $ ] is made of the points @xmath112 satisfying @xmath118={r[i , j]\over 1+b y_0[i , j]}<r_c.\ ] ] since it is difficult to estimate the finite @xmath50 correction from the integral formula ( [ eq : int4rcb ] ) in the case of the 4-point amplitude , we will compare the area ( [ eq : discs4b ] ) with numerically evaluated integral ( [ eq : int4rcb ] ) .
we can also construct the minimal surface for the 6 and 8-point amplitudes whose boundary conditions are the same as the cut and glue type surface obtained from the 4-point amplitude @xcite .
in this section , in order to confirm the validity of our numerical approach , we compare the integral formula @xmath119 $ ] for the 4-point amplitude with the area @xmath120 $ ] of the discretized minimal surface .
we evaluate the area by using @xmath106 lattice data @xcite .
the exact radial cut - off area @xmath119 $ ] is evaluated numerically by using mathematica .
we plot @xmath65 $ ] and @xmath121 $ ] at some values of @xmath45 ( fig . [
fig : b00 ] ) , where we can see that our numerical approach agrees with the exact formula in the region @xmath122 . in tables
[ table:4ptarea1 ] and [ table:4ptarea2 ] , we compare the values of @xmath121 $ ] , @xmath65 $ ] and @xmath123 $ ] at @xmath124 , @xmath125 by evaluating the ratios @xmath126-s_{4}[r_c])/ s^{dis}_{4}[r_c]$ ] and @xmath127-s_{4}^{dis}[r_c])/ s_{4}^{dis}[r_c]$ ] . here @xmath128 $ ] is given by dropping the @xmath129 corrections in ( [ eq : rad4pt1 ] ) , which is equal to the bds formula up to the constant term @xmath130&=&{1\over4}\log^2\left ( { r_c^2 ( 1-b)^2\over 16}\right ) + { 1\over4}\log^2\left ( { r_c^2 ( 1+b)^2\over 16}\right ) \nonumber\\ & & -{1\over4}\left\{\log\left({1+b\over 1-b}\right)^2\right\}^2 -{\pi^2\over3}. \label{eq : bds4pt2}\end{aligned}\ ] ] we can see that the discretized minimal surface area agrees with the exact formula for @xmath131 within @xmath132 .
it also differs from the bds formula ( [ eq : bds4pt2 ] ) about @xmath133 .
the @xmath106 data numerically reproduces the analytical result of the 4-point amplitude for @xmath131 . from the ratio @xmath134
, it is found that that finite @xmath50 corrections become small when @xmath50 decreases . .
the area of the discretized surface , the integral formula , the bds amplitudes and their differences ( divided by @xmath135 ) at @xmath136 [ cols="^,^,^,^,^,^",options="header " , ]
in this paper we studied the area of the minimal surfaces in ads spacetime surrounded by the light - like boundary which corresponds to the 4 , 6 and 8-points gluon scattering amplitudes with specific momentum configurations @xcite .
for all the solutions , it is found that the remainder function @xmath91 is independent of @xmath45 and the @xmath91-independent difference of the areas with different boost parameters obtained from the discretized minimal surface is consistent with the bds formula up to finite @xmath50 corrections .
it would be interesting to study the 6-point solutions with various momentum configuration ( hexagon for example ) .
we can determine numerically the remainder function @xmath137 as a function of @xmath138 , @xmath139 and @xmath140 , where at some values we could compare this with the result obtained in @xcite .
the present numerical approach will be helpful to determine the exact functional form of the remainder function @xmath91 via the ads / cft correspondence
. it would be also an interesting problem to estimate the finite @xmath50 correction analytically .
the integral formula ( [ eq : int4rcb ] ) of the 4-point solution can be expanded in @xmath50 and be evaluated by using hypergeometric function as @xmath141&= & -{2\sqrt{\pi}\over \sqrt{1-b^2 r_c^2 } } \sum_{n=0}^{\infty}{1\over 2n+1 } ( 1-r_c^2)^{n+1 } { \gamma(n+{3\over2})\over \gamma(2+n ) } { } _ 2f_1({1\over2 } , { 3\over2}+n ; 2+n ; { 1-r_c^2\over 1-b^2 r_c^2 } ) .
\nonumber\\ \label{eq : ef2}\end{aligned}\ ] ] because of complexity of this formula , it is difficult to estimate the finite @xmath50 corrections at this moment . the authors would like to thank koh iwasaki for collaboration in early stage of this work .
the work of k. i. is supported in part by the grant - in - aid for scientific research from ministry of education , science , culture and sports of japan .
z. bern , l. j. dixon and v. a. smirnov , `` iteration of planar amplitudes in maximally supersymmetric yang - mills theory at three loops and beyond , '' phys .
d * 72 * ( 2005 ) 085001 arxiv : hep - th/0505205 .
s. abel , s. forste and v.v .
khoze , `` scattering amplitudes in strongly coupled n=4 sym from semiclassical strings in ads '' , jhep * 0802 * ( 2008 ) 042 , arxiv:[0705.2113 ] + e. i. buchbinder , `` infrared limit of gluon amplitudes at strong coupling , '' phys .
b654 * ( 2007 ) 46 , arxiv : 0706.2015[hep - th ] .
+ s. ryang , `` conformal so(2,4 ) transformations of the one - cusp wilson loop surface . ''
b659 * ( 2008 ) 894 , arxiv:0710.1673[hep - th ] + a. popolitov , `` on coincidence of alday - maldacena - regularized sigma - model and nambu - goto areas of minimal surfaces '' , arxiv:0710.2073[hep - th ] + g. yang , `` comment on the alday - maldacena solution in calculating scattering amplitude via ads / cft '' , jhep * 0803 * ( 2008 ) 010 , arxiv : 0711.2828[hep - th ] + a. jevicki , k. jin , c. kalousios and a. volovich , `` generating ads string solutions '' , jhep * 0803 * ( 2008 ) 032 , arxiv:0712.1193[hep - th ] + z. komargodski , `` on collinear factorization of wilson loops and mhv amplitudes in n=4 sym '' , jhep * 0805 * ( 2008 ) 019 , arxiv:0801.3274[hep - th ] + r. brower , h. nastase and h. j.scnitzer and c .- i .
tan , `` implications of multi - regge limits for the bern - dixon - smirnov conjecture '' , arxiv:0801.3891 [ hep - th ] + a. mironov , a. morozov , and t. n. tomaras , `` on n - point amplitudes in n=4 sym , '' jhep * 0711 * ( 2007 ) 021 , arxiv:0708.1625 [ hep - th ] .
+ a. mironov and a. morozov and t. tomaras , `` some properties of the alday - maldacena minimum '' , phys . lett . *
b659 * ( 2008 ) 723 , arxiv:0711.0192 + h. itoyama , a. mironov and a. morozov , `` boundary ring : a way to construct approximate ng solutions with polygon boundary conditions : i. @xmath142-symmetric configurations , '' nucl .
b * 808 * , 365 ( 2009 ) arxiv:0712.0159 [ hep - th ] .
+ h. itoyama and a. morozov , `` boundary ring or a way to construct approximate ng solutions with polygon boundary conditions .
polygons which admit an inscribed circle , '' prog .
* 120 * , 231 ( 2008 ) arxiv:0712.2316 [ hep - th ]
. + z. komargodski and s.s razamat , `` planar quark scattering at strong coupling and universality '' , jhep * 0801 * ( 2008 ) 044 , arxiv : 0707.4367 [ hep - th ] + j. mcgreevy and a. sever , `` quark scattering amplitudes at strong coupling , '' jhep * 0802 * ( 2007 ) 015 , arxiv : 0710.0393[hep - th ] + k. ito , h. nastase and k. iwasaki , `` gluon scattering in @xmath5 super yang - mills at finite temperature , '' prog . theor .
phys . * 120 * , 99 ( 2008 ) arxiv:0711.3532 [ hep - th ]
. + y. oz , s. theisen and s. yankielowicz , `` gluon scattering in deformed n=4 sym'',phys . lett .
* b662 * ( 2008 ) 297 , arxiv:0712.3491 [ hep - th ] + m. kruczenski , `` a note on twist two operators in n=4 sym and wilson loops in minkowski signature '' , jhep * 0212 * ( 2002 ) 024 , hep - th/0210115 l. f. alday and j. maldacena , `` comments on gluon scattering amplitudes via ads / cft , '' jhep * 0711 * , 068 ( 2007 ) arxiv:0710.1060 [ hep - th ] .
h. itoyama , a. mironov and a. morozov , `` anomaly in n = infinity alday - maldacena duality for wavy circle , '' jhep * 0807 * , 024 ( 2008 ) arxiv:0803.1547 [ hep - th ] .
+ d. galakhov , h. itoyama , a. mironov and a. morozov , `` deviation from alday - maldacena duality for wavy circle , '' arxiv:0812.4702 [ hep - th ] .
drummond , g.p .
korchemsky and e. sokatchev , `` conformal properties of four - gluon planar amplitudes and wilson loops '' , arxiv:0707.0243[hep - th ] + a. brundhuber , p. heslop and g. travaglini , `` mhv amplitudes in n=4 super yang - mills and wilson loops '' , nucl .
* b794 * ( 2008 ) 231 , arxiv:0707:1153[hep - th ] , + j.m .
drummond , j. henn , g.p .
korchemsky and e. sokatchev , `` on planar gluon amplitudes / wilson loops duality '' , arxiv.0709.2368[hep - th ] j.m .
drummond , j. henn , g.p .
korchemsky and e. sokatchev , `` conformal ward identities for wilson loops and a test of the duality with gluon amplitudes '' , arxiv:0712.1223[hep - th ] + j. m. drummond , j. henn , g. p. korchemsky and e. sokatchev , `` dual superconformal symmetry of scattering amplitudes in n=4 super - yang - mills theory , '' arxiv:0807.1095 [ hep - th ] .
r. ricci , a. a. tseytlin and m. wolf , `` on t - duality and integrability for strings on ads backgrounds , '' jhep * 0712 * ( 2007 ) 082 arxiv:0711.0707 [ hep - th ] .
+ n. berkovits and j. maldacena , `` fermionic t - duality , dual superconformal symmetry , and the amplitude / wilson loop connection , '' jhep * 0809 * , 062 ( 2008 ) arxiv:0807.3196 [ hep - th ] .
+ n. beisert , r. ricci , a. a. tseytlin and m. wolf , `` dual superconformal symmetry from ads5 x s5 superstring integrability , '' phys .
d * 78 * , 126004 ( 2008 ) arxiv:0807.3228 [ hep - th ] .
drummond , j. henn , g.p .
korchemsky and e. sokatchev , `` hexagon wilson loop = six - gluon mhv amplitude '' , arxiv:0803.1466[hep - th ] + z. bern , l.j .
dixon , d.a .
kosower , r. roiban , m. spradlin , c. vergu and a. volovich , `` the two - loop six - gluon mhv amplitude in maximally supersymmetric yang - mills theory '' , arxiv:0803.1465[hep - th ] | we construct numerically the minimal surface in ads spacetime surrounded by the light - like segments , which are dual to the 4 , 6 and 8-point gluon scattering amplitudes in @xmath0 super yang - mills theory .
we evaluate the area of the minimal surface in the radial cut - off regularization and compare these areas with the formula conjectured by bern , dixon and smirnov ( bds ) , which is modified by the remainder function of cross - ratios of external momenta for @xmath1-point amplitudes . in our momentum configuration cross - ratios
are constant .
we calculate the difference of areas with different conformal boost parameters , which is independent of the remainder function , and find that its dependence on the boost parameter is numerically consistent with the bds formula .
tit / hep-592 + january , 2009 + revised version * discretized minimal surface and the bds conjecture in @xmath2 super yang - mills theory at strong coupling * .75em 1.0 cm suguru dobashi@xmath3 and katsushi ito@xmath4 2.5em @xmath3 _ department of accelerator and medical physics , + national institute of radiological sciences , + 4 - 9 - 1 anagawa , inage - ku , chiba 263 - 8555 , japan
_ 1.5em @xmath4 _ department of physics + tokyo institute of technology + 2 - 12 - 1 , ookayama , meguro - ku , tokyo , 152 - 8551 , japan _ 4.5em = 0.7 cm |
both the current noise and the admittance provide information about the dynamics of mesoscopic conductors . however , whereas the current fluctuations have been widely studied in interacting systems , it is not the case of the quantum admittance for which only few theoretical works are available @xcite .
moreover , the inter - relations between these quantities have not been deeply explored despite the fact that they verify the simple relation : @xmath0=s(-\omega)-s(\omega)$ ] , which means that a non - vanishing real part of the admittance @xmath1 is responsible of the asymmetry of the noise @xmath2 obtained in interacting systems @xcite .
the admittance is a quantity which deserves to be known , especially as its experimental determination expends rapidly . indeed , following the pioneer works by gabelli and co - workers @xcite ,
measurements of the quantum admittance in various mesoscopic systems have been achieved recently . in particular ,
high - frequency admittance has been measured for a quantum point contact in the quantum hall regime @xcite , determination of the admittance has been performed for a carbon nanotube double quantum dot @xcite , both high - frequency admittance and noise have been measured in superconductor / insulator / superconductor junction @xcite , and dynamic admittance has been studied in a quantum dot coupled to a two dimensional electron gas @xcite .
it is thus needed to develop in parallel theoretical studies of quantum admittance . here
, we consider a two dimensional electron gas in the fractional quantum hall regime and we carefully look at the inter - relations between admittance , which was calculated in ref .
@xcite , and current - current correlations between edge states along which the fractional charges are propagated .
the paper is organized as follows : in sec .
ii , we present the model used to describe the chiral edge states in the fractional quantum hall regime and the method of calculation of the current - current correlations generated by a constriction ( see figure [ diagrams ] ) . in sec .
iii , we give the expressions of the auto- and cross - correlators in terms of current , admittance and integral which involve all the transmission amplitude through the constriction . in sec .
iv , we discuss the results in various limits .
we conclude in sec .
a two - dimensional electron gas in the fractional quantum hall regime is modeled in the framework of the tomonaga - luttinger theory @xcite by the hamiltonian : @xmath3\nonumber\\ & & + \frac{\gamma_\mathrm{b}}{4\pi}e^{i[\phi_l(x)+\phi_r(x)]/\sqrt{2}-i\overline{e}\chi(\tau)/(\hbar c)}+hc~,\end{aligned}\ ] ] where @xmath4 and @xmath5 are the bosonic fields associated with the left ( @xmath6 ) and right ( @xmath7 ) moving electrons , @xmath8 is the fermi velocity , and @xmath9 is included in order to treat the applied voltage .
the energy @xmath10 corresponds to the backscattering amplitude at the constriction ( see fig .
[ diagrams ] ) and @xmath11 is the fractional charge associated to the filling factor @xmath12 . to calculate transport properties
, we use a refermionization method @xcite which applied for @xmath13 .
note that eq .
( [ hamiltonian ] ) describes incompressible chiral edge modes which do not correspond to the half filled states since those ones are expected to be compressible @xcite .
however , the results obtained by the refermionization method provide interesting features on the relation between admittance and current correlations in such a system that are worth to study .
the admittance is obtained by calculating the photo - assisted current , i.e. , the response to an ac voltage superimposed to a dc voltage : @xmath14 , associated to eq .
( [ hamiltonian ] ) and taking the derivative of its first order harmonic with respect to @xmath15 @xcite : @xmath16 \nonumber\\ & & \times\big[f(\hbar\omega+\overline{e}v_\mathrm{dc})-f(-\hbar\omega+\overline{e}v_\mathrm{dc})\big]d\omega~,\end{aligned}\ ] ] where @xmath17 is the fermi - dirac distribution function , and @xmath18 is the transmission amplitude through the constriction .
next , we calculate the fourier transform of the non - symmetrized correlations between currents in branches @xmath19 and @xmath20 associated to eq .
( [ hamiltonian ] ) with @xmath21 : @xmath22 with @xmath23 . for chiral edge states
, we have @xmath24 , where @xmath25 for right ( @xmath26 ) moving branch and @xmath27 for left ( @xmath28 ) moving branch .
@xmath29 is the density operator which is related to the density operators @xmath30 associated to the new fermionic operators @xmath31 and @xmath32 introduced in the refermionization procedure through the relation @xcite : @xmath33/2 $ ] . note that @xmath34 and @xmath35 are free fields that are affected neither by the applied voltage nor by the backscattering .
thus , we have : @xmath36~,\end{aligned}\ ] ] where @xmath37 .
all the possible correlators found by varying @xmath19 , @xmath20 , @xmath38 and @xmath39 are shown on figure [ diagrams ] .
since we have four branches , the total number of correlators equals @xmath40 . and @xmath41 in the black boxes indicate the associated time arguments of the current operators in eq .
( [ definition]).,width=321 ] they correspond to @xmath42 and @xmath43 and take real values .
there are four auto - correlators equal in pairs ( the indexes used below refer to the diagram numbers in figure [ diagrams ] ) .
the calculation gives : @xmath44 where @xmath45 is the quantum of conductance , and @xmath46^{-1}$ ] @xcite .
in addition , we have : @xmath47\mathrm{re}\{y(\omega)\}~,\end{aligned}\ ] ] where : @xmath48\nonumber\\ & & -\frac{e}{4\pi}\big[n(\tilde\sigma\hbar\omega+\sigma ev_\mathrm{dc})-n(\tilde\sigma\hbar\omega)\big]\nonumber\\ & & \times\int_{-\infty}^{\infty}d\omega f(\hbar\omega-\sigma \overline{e}v_\mathrm{dc})\big|t(\omega+\tilde\sigma\omega)+t^*(\omega)\big|^2\bigg\}~,\nonumber\\\end{aligned}\ ] ] is an even function in terms of frequency and voltage .
the dc current reads as : @xmath49d\omega~.\end{aligned}\ ] ] the auto - correlators @xmath50 and @xmath51 are those of charges free to propagate along the edge states ( the constriction is not yet reached ) , whereas the auto - correlators @xmath52 and @xmath53 are those of charges that have been reflected or transmitted through the constriction .
it explains why the transmission amplitude @xmath54 appears in these two last auto - correlators only .
they corresponds to @xmath55 or / and @xmath56 ( in the latter case , we take the limits @xmath38 and @xmath39 close to zero in order to focus on what happens near the constriction ) .
there are twelve cross - correlators , some of them take real values whereas the others take complex values .
the two last correlators of the first line in figure 1 are equal to zero : @xmath57 , since as long as the constriction is not reached , the carriers in two independent chiral branches are not correlated .
the two last correlators of the second line in figure [ diagrams ] are equal to each other : @xmath58 , with : @xmath59\mathrm{re}\{y(\omega)\}~.\end{aligned}\ ] ] the two first correlators of the third and fourth lines in figure [ diagrams ] are related through the relation : @xmath60 , with : @xmath61~.\end{aligned}\ ] ] the two last correlators of the third and fourth lines in figure [ diagrams ] are related through @xmath62 with : @xmath63~.\end{aligned}\ ] ] note that the knowledge of @xmath64 and @xmath1 fully determine the auto- and cross - correlators .
in this section , we consider the excess correlators @xcite , i.e. the difference between their values at finite dc voltage @xmath65 and their values at zero voltage : @xmath66 with @xmath67 $ ] .
the calculations show that the excess correlators are equal four to four .
since the carriers that are ahead the constriction do not feel the voltage , the excess correlators for the diagrams 1 , 2 , 3 and 4 of the first line in figure [ diagrams ] vanish : @xmath68}(\omega)=0~.\end{aligned}\ ] ] the excess correlators associated to diagrams 5 , 6 , 7 et 8 coincide with each other and are related to @xmath64 and to the real part of the excess admittance : @xmath69}(\omega)=\delta s_{0}(\omega)-\hbar\omega\big[2n(\hbar\omega)+1\big]\mathrm{re}\{\delta y(\omega)\}~.\nonumber\\\end{aligned}\ ] ] the excess correlators associated to diagrams 9 , 10 , 11 and 12 are related to the excess admittance : @xmath70}(\omega)=\hbar\omega n(\hbar\omega)\delta y(\omega)~,\end{aligned}\ ] ] as well as the diagrams 13 , 14 , 15 and 16 : @xmath71}(\omega)=\hbar\omega n(\hbar\omega)\delta y^*(\omega)~.\end{aligned}\ ] ] the expressions of the total excess noise @xmath72 and backscattering excess noise @xmath73 can also be derived . indeed from figure
[ diagrams ] , we deduce from the definitions of the correlators that @xmath74 and @xmath75 , which lead to : @xmath76 since @xmath77 is even in frequency and the real part of @xmath78 is odd in frequency , eq .
( [ noise_t ] ) verifies the relation @xmath79 as it should . in the following
we study the frequency profile of the excess correlators with emphasize on various limits : first the low temperature limit in order to understand how the backscattering amplitude strength affects this profile , next the weak backscattering regime in order to characterize the effect of temperature , and finally the high temperature limit . in figure [ figure2](a )
is plotted the correlator @xmath80 at temperature much smaller than the applied voltage .
we observe that this excess correlator is symmetric in frequency .
it is due to the fact that the diagrams 5 , 6 , 7 and 8 of figure [ diagrams ] corresponding to this correlator are symmetrical under time inversion , thus @xmath81 which immediately leads to @xmath82 .
the real and imaginary parts of @xmath83 are plotted in figures [ figure2](b ) and [ figure2](c ) .
this excess correlator has the very particular property to be non zero only at negative frequency .
we can interpret this result as a time inversion symmetry breaking which is understandable by looking at diagrams 9 , 10 , 11 and 12 of figure [ diagrams ] .
comparing now figures [ figure2](a ) and [ figure2](b ) , we remark that at negative frequency we have @xmath84 , which means that the contribution of the term @xmath77 in eq .
( [ deltas5 ] ) is negligible in that regime . since at zero temperature , we have @xmath85 , where @xmath86 is the heaviside function , we can approximate the correlators as : @xmath87\mathrm{re}\{\delta y(\omega)\}~.\end{aligned}\ ] ] thus , from the fact that @xmath88 in the low temperature limit , we find that the excess correlators are entirely determined by the excess admittance in that regime . as a consequence ,
the singularities observed at @xmath89 in figures [ figure2](a ) , [ figure2](b ) and [ figure2](c ) for weak backscattering amplitudes are those of the admittance ( see figure 2 of ref .
these singularities disappear when the backscattering amplitude increases .
note that since the correlators @xmath90 and @xmath91 appearing in the definitions of the total and backscattering excess noises are symmetric in frequency , the asymmetry of @xmath72 and @xmath92 comes from the diagrams 9 , 11 , 13 and 15 of figure [ diagrams ] . ,
( b ) real part and ( c ) imaginary part of @xmath93 in units of @xmath94 , as a function of frequency in units of @xmath95 , for various values of the backscattering amplitude @xmath10 at temperature @xmath96 .
the backscattering amplitudes are @xmath97 ( solid red line ) , @xmath98 ( dashed green line ) , @xmath99 ( purple dotted line ) and @xmath100 ( dash - dotted black line).,title="fig:",width=234 ] , ( b ) real part and ( c ) imaginary part of @xmath93 in units of @xmath94 , as a function of frequency in units of @xmath95 , for various values of the backscattering amplitude @xmath10 at temperature @xmath96 .
the backscattering amplitudes are @xmath97 ( solid red line ) , @xmath98 ( dashed green line ) , @xmath99 ( purple dotted line ) and @xmath100 ( dash - dotted black line).,title="fig:",width=234 ] , ( b ) real part and ( c ) imaginary part of @xmath93 in units of @xmath94 , as a function of frequency in units of @xmath95 , for various values of the backscattering amplitude @xmath10 at temperature @xmath96 . the backscattering amplitudes are @xmath97 ( solid red line ) , @xmath98 ( dashed green line ) , @xmath99 ( purple dotted line ) and @xmath100 ( dash - dotted black line).,title="fig:",width=234 ] we turn now our attention to the behavior of the correlators when the backscattering amplitude is much smaller than the voltage . in figures
[ figure3](a ) , [ figure3](b ) and [ figure3](c ) are plotted @xmath80 and the real and imaginary parts of @xmath93 at various temperature for @xmath97 .
the correlator @xmath80 is equal to zero at @xmath101 under the condition that both temperature and backscattering are small in comparison to the voltage ( see solid red lines on figures [ figure2](a ) and [ figure3](a ) ) . concerning the correlator @xmath83 , both the backscattering and the thermal effects are able to give a non - vanishing contribution at @xmath102 , whereas only thermal effects can give a non - vanishing contribution at @xmath103 , which stays however small for what concerns the imaginary part of @xmath93 ( see figure [ figure3](c ) ) .
in units of @xmath94 , as a function of frequency in units of @xmath95 , for various temperature at weak backscattering amplitude @xmath97 . the temperatures are @xmath96 ( solid red line ) , @xmath104 ( dashed green line ) , @xmath105 ( purple dotted line ) and @xmath106 ( dash - dotted black line ) . the same curves for ( b ) the real part and ( c ) the imaginary part of @xmath93.,title="fig:",width=234 ] in units of @xmath94 , as a function of frequency in units of @xmath95 , for various temperature at weak backscattering amplitude @xmath97 .
the temperatures are @xmath96 ( solid red line ) , @xmath104 ( dashed green line ) , @xmath105 ( purple dotted line ) and @xmath106 ( dash - dotted black line ) . the same curves for ( b ) the real part and ( c ) the imaginary part of @xmath93.,title="fig:",width=234 ] in units of @xmath94 , as a function of frequency in units of @xmath95 , for various temperature at weak backscattering amplitude @xmath97 .
the temperatures are @xmath96 ( solid red line ) , @xmath104 ( dashed green line ) , @xmath105 ( purple dotted line ) and @xmath106 ( dash - dotted black line ) . the same curves for ( b ) the real part and ( c ) the imaginary part of @xmath93.,title="fig:",width=234 ] for temperature much higher than the voltage , i.e. , at equilibrium , we find from eq .
( [ s0_exp ] ) that : @xmath107\mathrm{re}\{y(\omega)\}~,\end{aligned}\ ] ] which leads to the cancellation of @xmath90 in agreement with the evolution of the curves shown in figure [ figure3](a ) , and to : @xmath108 in agreement with the fluctuation - dissipation theorem since the real part of the admittance correspond to the ac conductance .
the cancellation of @xmath90 is related to the fact that the dominant contribution in that limit is the thermal noise which is voltage independent .
since the excess correlator is defined as the difference between its values at finite and zero voltages , there is an exact compensation . from these results ,
we understand that in that regime too , the excess correlators are entirely determined by the excess admittance .
our study shows that the current , the quantum admittance and the current - current correlators between the edge states of a fractional quantum hall gas are quantities that are closely related .
indeed , since all these quantities can be expressed in terms of the transmission amplitude through the constriction which coupled the edge states , it is possible to obtain direct relations between them .
accordingly , the singularities that are observed in the correlators in the weak backscattering and low temperature regimes at frequencies equal to @xmath109 , are directly related to the singularities of the quantum admittance . moreover , in both low and high temperature limits , we have shown that all the current correlators are fully determined by the admittance and we have established explicit relations between them .
these relations can be use to find the value of the finite - frequency noise from the measurement of the admittance which is easier to perform .
the authors would like to thank t. martin and i. safi .
99 m. bttiker , a. prtre , and h. thomas , phys . rev
. lett . * 70 * , 4114 ( 1993 ) .
y. hamamoto , t. jonckheere , t. kato , and t. martin , phys .
b * 81 * , 153305 ( 2010 ) .
a. cottet , c. mora , and t. kontos , phys .
b * 83 * , 121311(r ) ( 2011 ) . a. crpieux , phys .
b * 87 * , 155432 ( 2013 ) .
m. filippone , k. le hur , c. mora , cond - mat - arxiv:1304.4454 ( 2013 ) .
i. safi , c. bena , and a. crpieux , phys .
b * 78 * , 205422 ( 2008 ) .
i. safi and p. joyez , phys .
b * 84 * , 205129 ( 2011 ) .
r. zamoum , a. crpieux , and i. safi , phys .
b * 85 * , 125421 ( 2012 ) . j. gabelli , g. fve , j .-
berroir , b. plaais , a. cavanna , b. etienne , y. jin , and d.c .
glattli , science * 313 * , 49 ( 2006 ) .
j. gabelli , g. fve , t. kontos , j .-
berroir , b. plaais , d.c .
glattli , b. etienne , y. jin , and m. bttiker , phys .
lett . * 98 * , 166806 ( 2007 ) .
j. gabelli , g. fve , j .-
berroir , and b. plaais , rep .
. phys . * 75 * , 126504 ( 2012 ) .
m. hashisaka , k. washio , h. kamata , k. muraki , and t. fujisawa , phys .
b * 85 * , 155424 ( 2012 ) . s.j .
chorley , j. wabnig , z.v .
penfold - fitch , k.d .
petersson , j. frake , c.g .
smith , and m.r .
buitelaar , phys .
lett . * 108 * , 036802 ( 2012 ) .
j. basset , h. bouchiat , and r. deblock , phys .
b * 85 * , 085435 ( 2012 ) .
t. frey , p.j .
leek , m. beck , j. faist , a. wallraff , k. ensslin , t. ihn , and m. bttiker , phys .
b * 86 * , 115303 ( 2012 ) .
s. tomonaga , prog .
phys . * 5 * , 544 ( 1950 ) .
luttinger , j. math . phys . * 4 * , 1154 ( 1963 ) .
chamon , d.e . freed , and x.g .
wen , phys .
b * 53 * , 4033 ( 1996 ) .
halperin , p.a .
lee , and n. read , phys .
b * 47 * , 7312 ( 1993 ) .
g. murthy and r. shankar , rev .
phys . * 75 * , 1101 ( 2003 ) .
a. crpieux , p. devillard , and t. martin , phys .
b * 69 * , 205302 ( 2004 ) .
note that at positive frequency @xmath110 corresponds to the bose - einstein distribution function . in experiments ,
measurement of excess correlators rather than correlators allows one to remove the unwanted sources of noise . | in this work , we focus on the finite frequency current - current correlations between edge states in a fractional quantum hall two dimensional gas and on their relations to the quantum admittance . using a refermionization method
, we calculate these quantities within the same framework .
our results apply whatever the values of backscattering amplitude , frequency , voltage and temperature , allowing us to reach different regimes .
auto - correlations and cross - correlations exhibit distinct frequency dependencies that we discuss in detail . |
understanding thermodynamic properties of electrolyte systems is a long - standing problem@xcite which , in recent years , has attracted increased attention due to controversial results on the nature of criticality in coulomb systems.@xcite experiments@xcite suggest that the critical region of solutions of some organic salts can be well described by classical behavior , while for other electrolyte systems the ising - like description with non - classical behavior is more appropriate.@xcite this dichotomy has greatly stimulated theoretical attempts to understand the thermodynamics of ionic systems .
criticality in simple nonionic fluids can be successfully described and analyzed by the renormalization group ( rg ) method , and it is reasonable to suggest that this method could also be used to investigate also coulomb systems . however , in order to proceed with rg calculations , a physically meaningful and well based mean - field theory should be developed.@xcite most theoretical studies of ionic systems concentrate on the simplest model of electrolytes , the so called restricted primitive model ( rpm ) , in which ions are viewed as equal size particles of positive and negative charges of equal magnitude .
currently , there are two theoretical directions in the development of the mean - field description of ionic fluids .
the first approach is based on integral equations for correlation functions@xcite , while the second approach extends the original debye - hckel ( dh ) theory.@xcite comprehensive theoretical analysis,@xcite which utilizes for example thermodynamic energy bounds , and comparison with current monte carlo simulations,@xcite indicate that theories based on dh theory may provide a better description of the thermodynamics of electrolytes in critical regions .
so far most of the theoretical efforts in the investigation of charged systems have been devoted to continuum models . however , lattice models are also important for understanding the criticality in coulomb systems , since the ising model , which is a lattice gas model , has been crucial for the description of critical phenomena in nonionic systems.@xcite there are few numerical@xcite and analytical@xcite results for lattice ionic systems which show that the phase diagram differs significantly from continuum models . the structures of simple cubic ( sc ) and body - centered cubic ( bcc ) lattices allow for charge distribution with the appearance of a long - range order phase at low temperatures similar to that of an ionic crystal
. however , this ordering decreases entropy and for high temperatures the disordered phase is thermodynamically more stable . as a result , there is an order - disorder phase transition line which ends up at the tricritical point,@xcite while for continuum rpm systems only a gas - liquid coexistence can be found .
recently , a systematic investigation of lattice models of electrolytes has been presented.@xcite in this investigation the lattice restricted primitive model ( lrpm ) , with charged particles occupying sites on a general @xmath0-dimensional lattice , has been considered using debye - hckel theory . by solving exactly the lattice version of the debye - hckel equation ,
closed expressions for thermodynamic properties of general @xmath0-dimensional ionic systems have been obtained . for three - dimensional lattice coulombic systems specific calculations , which included pairing and dipole - ion solvation , yielded a gas - liquid phase separation at low densities .
however , by taking into account the lattice symmetry , it has been shown that for sc and bcc lattices this gas - liquid phase separation is thermodynamically unfavorable , and the order - disorder phase transitions with the tricritical point will dominate , in agreement with monte carlo simulation results.@xcite the thermodynamics of lattice anisotropic ionic systems have not been studied yet , although they may provide important information on thermodynamics of real electrolytes . in addition , the lattice stretching , which leads to anisotropy , can be viewed as analogous of lowering the spatial dimensionality of the system . however
, dh - based calculations for the continuum@xcite and lattice@xcite models of electrolytes predict an _ increase _ of gas - liquid critical temperatures for lower dimensions .
thus , for lattice anisotropic models , gas - liquid phase separation may reappear along with distinct order - disorder phase transitions .
this possibility raises the question of the precise determination of phase diagrams for ionic systems on anisotropic lattices . in this article , we present a theoretical investigation of anisotropic lattice models of electrolytes using the debye - hckel method,@xcite which has the advantage that it accounts for both electrostatic screening and sublattice ordering in a unified framework .
we investigate the ionic systems on three - dimensional lattices obtained by anisotropic stretching of simple cubic and body - centered cubic lattices .
the paper is organized as follows . in sec .
ii we provide a combined dh and mean - field ordering description of ions on simple tetragonal lattices .
the similar analysis for stretched body - centered lattices is given in sec .
a discussion and conclusions are presented in secs .
our derivation for anisotropic lattice electrolytes follows closely the debye - hckel approach for isotropic lattice ionic systems.@xcite we consider a system of equal numbers of positive and negative ions with the total density @xmath1 on a tetragonal lattice , which is obtained by stretching the simple cubic lattice , with unit cell dimensions @xmath2 with an anisotropy factor defined as the ratio of lattice parameters , @xmath3 .
the case @xmath4 corresponds to the isotropic simple cubic lattice electrolytes which have been studied in detail earlier.@xcite the debye - hckel approach implies that we can construct the total free energy of the system by summing consecutively the terms which describe interactions between different species .
charged particles interact through the lattice coulomb potential , otherwise they behave as ideal particles with additional hard - core on - site exclusions . thus the total free energy density is given by @xmath5 .
the ideal lattice gas contribution can be written as @xmath6 where @xmath7 is the reduced dimensionless density and @xmath8 is the unit lattice cell volume .
the other term in @xmath9 comes from the coulombic interactions of the free ions and includes the effect of screening . in order to find an expression for the debye - hckel contribution to the free energy density ,
we need the electric potential felt by an ion due to all other ions . this potential can be found by fixing an arbitrary ion at the origin and solving the linearized poisson - boltzmann equation @xmath10 where @xmath11 is the inverse squared debye screening length , with @xmath12 and @xmath13 for three - dimensional lattices .
the lattice laplacian used in eq.([linpb ] ) can be presented in the form which incorporates the geometry of the lattice , @xmath14 with @xmath15,\ ] ] where @xmath16 ; @xmath17 and @xmath18 are the unit vectors along the corresponding lattice directions .
then eq.([linpb ] ) can be easily solved by fourier transformation to yield for @xmath19 @xmath20 where we introduced the anisotropic lattice function @xmath21 @xmath22 is the number of nearest neighbors , and @xmath23 and @xmath24 .
note that the vectors @xmath25 describe the reciprocal lattice .
to compute the electric potential at the origin due to the surrounding ions , it is recalled that no other ion can be placed at that site , and hence the appropriate equation is the laplace s equation , @xmath26 this enables us to write @xmath27 which yields @xmath28 then from eqs .
( [ fiofr ] ) and ( [ fiof0.1 ] ) it follows that @xmath29 following the dh approach for isotropic lattices,@xcite we introduce the integrated anisotropic lattice green s function @xmath30 which has been evaluated exactly in terms of a product of two complete elliptic integrals of the first kind by joyce.@xcite then the total potential at the origin due to the surrounding ions , @xmath31 , takes the form @xmath32.\ ] ] by using the debye charging procedure , the reduced electrostatic free energy density can be calculated explicitly , yielding @xmath33.\ ] ] the chemical potential , @xmath34 , is then given by @xmath35,\ ] ] where , following the continuum dh theory@xcite and the dh theory for isotropic cubic lattices,@xcite the reduced temperature is defined as @xmath36 and for the reduced density we obtain @xmath37 knowing the free energy density and the chemical potential allows us to calculate the pressure , @xmath38 $ ] , yielding @xmath39.\ ] ] eqs .
( [ f.id ] ) , ( [ f.dh ] ) , ( [ chem.potential ] ) and ( [ pressure ] ) provide a full thermodynamic description of the simple tetragonal lattice model of electrolytes .
the thermodynamics at the critical region can be investigated by analyzing the spinodal , which is determined by the condition @xmath40 .
( [ chem.potential ] ) we obtain @xmath41 with @xmath42 . the phase transitions and the gas - liquid coexistence can be studied by analyzing the pressure and the chemical potential in different phases .
the predicted gas - liquid coexistence curves for simple tetragonal lattices are shown in fig.1 .
the critical temperature increases monotonically as the anisotropy parameter decreases and reaches the value of @xmath43 at @xmath44 : see fig.2a . at the same time , the critical density shows a non - monotonic behavior with two maxima and a minimum , and finally approaches the value @xmath45 at @xmath44 , as shown in fig.2b . lowering the anisotropy parameter
can be visualized as the stretching the lattice along one direction , and the limit of @xmath46 corresponds to an infinite distance between the layers .
thus the anisotropic lattice model at @xmath44 is equivalent to the 2d coulomb system on square lattice , for which the pure dh theory@xcite predicts the critical parameters to be @xmath47 and @xmath45 .
the apparent discrepancy between our results for the critical temperature and the results for two - dimensional lattice electrolytes@xcite can be easily explained by analyzing the linearized poisson - boltzmann eq.([linpb ] ) . in our calculations
we used the three - dimensional coefficient @xmath48 , while in two dimensions this coefficient is equal to @xmath49 , which explains the factor 2 in the difference in the corresponding values of the critical temperatures . at the limit of large @xmath50 the lattice
is stretched along 2 directions , and thus the anisotropic lattice with @xmath51 corresponds to the 1d lattice coulomb system .
our predictions for critical parameters in this case are @xmath52 and @xmath45 , while for one - dimensional lattice electrolytes the dh - based calculations give @xmath53 and @xmath45 . in this case
, the difference in critical temperatures can be attributed to our definition of the reduced temperature in eq.([temp ] ) .
note , however , that the dh method is incorrect in describing one - dimensional ionic systems.@xcite the overall agreement between our estimates of critical parameters of strongly anisotropic lattice models of electrolytes and the results for 1d and 2d ionic lattice systems supports our arguments that anisotropic stretching is analogous to lowering of spatial dimensionality for simple cubic lattices .
a surprising feature of the predicted coexistence curves is the critical density dependence on the lattice anisotropy as exhibited in fig.2b .
it shows two maxima ( at @xmath54 and @xmath55 ) and one minimum ( at @xmath4 ) ; this picture is probably the result of geometric and packing effects .
the debye - hckel approach presented above describes only the low - density behavior of the system . to obtain the full thermodynamic description of simple tetragonal lattices we have to take into account the lattice symmetry .
a simple tetragonal lattice , similarly to a simple cubic lattice , can be viewed as consisting of two intercalated sublattices .
ions of opposite signs can be distributed unequally between these sublattices , thus reducing the electrostatic contribution to the free energy . at the same time
, unequal distribution of charged particles lowers the entropy which leads to an increase in the total free energy .
the competition between these factors determine the thermodynamics and phase behavior of the system .
we consider again a simple tetragonal cubic lattice with @xmath56 charged particles .
the overall system is neutral , and there are @xmath57 ( @xmath58 ) positive ( negative ) particles in sublattice @xmath59 , and @xmath60 ( @xmath61 ) positive ( negative ) particles in sublattice @xmath62 . assuming that sublattice @xmath59 has an excess of positive ions , the corresponding order parameter can be defined as @xmath63 this order parameter has a positive value in the ordered phase , while it equals to zero in the disordered phase . the nonzero charge density on each sublattice produces an additional `` background '' potential @xmath64 which
, however , does not change the correlation functions.@xcite this potential can be found using the linearized poisson - boltzmann equation @xmath65 because of the symmetry between sublattices , we have @xmath66 . by using the definition of the lattice laplacian ( [ laplac1])-([laplac2 ] ) , and following the approach outlined for isotropic lattice electrolytes,@xcite we obtain for the potential @xmath67 due to all ions except the one fixed at the origin the following expression @xmath68 where @xmath69 is given in eq.([psi.1 ] ) .
the electrostatic part of the total free energy then follows again from the debye charging process , while the entropic contribution can be calculated as for isotropic lattices,@xcite yielding @xmath70 with @xmath71,\ ] ] where @xmath72 and @xmath73 are given by eqs.([f.id ] ) and ( [ f.dh ] ) .
the knowledge of the total free energy allows us to investigate the possibility of sublattice ordering .
it can be done by looking for minima of @xmath74 for nonzero values of the order parameter @xmath75 .
this procedure leads us to the equation describing the @xmath76-line , along which second - order phase transitions occur , @xmath77 the anticipated tricritical point can be found by calculating the intersection of the @xmath76-line with the spinodal @xmath78 , and this analysis yields the equation for the tricritical point,@xcite @xmath79}{\partial ( x^2)}\right]\right|_{x^2=4(2+\alpha^2)}+\frac{1}{1-\rho_{tri}^ * } - \frac{3}{2}=0.\ ] ] the resulting tricritical densities and temperatures for different values of the anisotropic parameter @xmath50 are presented in figs.2a and 2c .
the tricritical temperature is a decreasing function of the anisotropy parameter , and it reaches its maximal value of @xmath80 at @xmath44 , which is exactly twice the value of the tricritical temperature for the ionic system on the two - dimensional square lattice ( see eq.(70 ) of ref .
this deviation is again the result of using different dimension - dependent coefficients @xmath81 , as was argued above .
both critical and tricritical temperatures vanish at large anisotropies , however , @xmath82 becomes smaller than @xmath83 for @xmath84 .
the behavior of the tricritical density is different .
it has a minimal value for the isotropic lattice ( @xmath4 ) , and it reaches the maximal values of @xmath85 and @xmath86 for @xmath44 and @xmath87 respectively .
the phase diagrams for simple tetragonal lattices are presented in fig.3 . for weakly anisotropic lattices
there are only order - disorder phase transitions , while for strongly anisotropic lattices the gas - liquid phase separation reappears at low densities .
consider a system of equal numbers of positive and negative ions on the body - centered tetragonal lattice with @xmath88 unit lattice cell . using the symmetry of the lattice and applying eqs.([laplac1 ] ) and ( [ laplac2 ] ) the lattice laplacian for tetragonal body - centered lattice is given by @xmath89,\ ] ] where @xmath3 and the summation runs over all @xmath90 neighbors in a cell .
when @xmath91 this reduces to the well known lattice laplacian for the body centered cubic lattice.@xcite because of the special symmetry of body - centered lattice , the lattice function @xmath92 in contrast to the simple tetragonal lattice , is independent of the anisotropy parameter @xmath50 . thus the isotropic bcc lattice green s function @xmath93 can be used for calculation of the thermodynamic properties .
then the potential at the origin due to the surrounding ions takes the form @xmath94,\ ] ] and the electrostatic free energy is given by @xmath95.\ ] ] furthermore , the ordering free energy does not depend on the type of lattice @xcite and is the same both for the simple and body centered lattices .
therefore we can use the corresponding expression from ( [ f.ord ] ) and estimate the tricritical point from eq.([tri ] ) by using @xmath96 instead of the simple tetragonal green s function @xmath97 $ ] .
phase diagrams for electrolytes on the tetragonal body - centered lattices are shown in fig.4 .
analysis of phase diagrams for the body - centered lattice models of ionic systems indicate that both critical and tricritical densities are independent of the degree of lattice stretching , while @xmath98 and @xmath99 are decreasing functions of @xmath50 .
these relations can be understood by analyzing the corresponding equation for the spinodal , @xmath100 where @xmath101 .
since the lattice green s function @xmath102 is independent of the anisotropy parameter @xmath50 , the critical point can be described by the parameter @xmath103 with @xmath104 being _ independent _ of @xmath50 , while @xmath105 .
then , utilizing eq.([density ] ) the critical density @xmath106 is also independent of @xmath50 .
similar arguments can now be applied for the analysis of the tricritical point .
the thermodynamic behavior of the tetragonal body - centered lattice electrolytes is then different from that of the simple tetragonal lattice ionic systems . since critical and tricritical temperature decay at the same rate as functions of the anisotropy parameter @xmath50 , while @xmath107 and @xmath108 are constant , the possible low - density gas - liquid coexistence is suppressed by order - disorder phase transitions with a tricritical point at any degree of anisotropy , as shown in fig.4 .
our analysis of simple tetragonal lattice models of electrolytes based on the dh approach indicates that , similarly to isotropic sc ionic lattice systems , at weak anisotropies ( @xmath109 ) the possible low - density gas - liquid phase separation is metastable , and the sublattice ordering is always thermodynamically more favorable .
however , for strongly anisotropic simple lattices ( @xmath110 or @xmath111 ) the gas - liquid coexistence reappears and the phase diagram becomes more complex , with critical , tricritical and triple points , as shown in fig.3 .
the explanation of this phenomenon is the following .
for weakly anisotropic lattices the ordering of ions of opposite signs on different sublattices decreases significantly the total free energy , and order - disorder phase transitions with a tricritical point determines the phase diagram of the system . for strongly anisotropic lattices
this ordering is less significant at low densities and the gas - liquid phase separation is restored .
another way of looking at this phenomenon , as was discussed above , is the analogy between lattice stretching and lowering of the space dimensionality . as was shown before,@xcite at low dimensions the critical temperature is increasing and thus gas - liquid coexistence occurs again at low densities .
however , the thermodynamic behavior of coulomb systems on body - centered tetragonal lattices is very different .
at all degrees of anisotropy , the sublattice ordering is thermodynamically more stable , and gas - liquid phase separation is always suppressed .
this is the result of the special symmetry of body - centered lattices .
our pure debye - hckel treatment of anisotropic lattice electrolytes assumed that there are only free ions and empty sites in the system .
however , at low temperatures the formation of strongly bound neutral dimers , or bjerrum pairs,@xcite is a highly favorable process .
such pairing can be viewed as a reversible chemical reaction,@xcite and this process can be treated in a systematic way .
another important contribution to the free energy of electrolytes is the ion - dipole solvation energy.@xcite the formation of ion pairs and their interactions with single ions have a strong effect on the thermodynamic properties and phase diagrams of electrolytes.@xcite our theoretical method can be extended to include these effects , although , due to anisotropy of the lattice system , there will be more than one type of bound neutral dimers .
based on the comparison with the continuum and isotropic lattice electrolytes,@xcite we predict that , if we take these effects into account , both critical and tricritical temperatures will decrease , while critical and tricritical densities will increase .
however , this means that our qualitative conclusions on thermodynamics and phase diagrams of anisotropic cubic lattices will not change .
it is interesting to note that similar phase diagrams of lattice electrolytes have been obtained by ciach and stell.@xcite they considered a mean - field theory of electrolytes with single - ion lattice potential on the isotropic sc lattice and additional short - range interactions .
note that this treatment neglects the cooperative screening , which is thermodynamically important and which is included in our dh - based theory .
however , the origins of similar complex phase diagrams in both models are different . in the model of ciach and
stell the gas - liquid phase coexistence is driven by short - range interactions , while in our model the lattice stretching makes the sublattice ordering less thermodynamically favorable at low densities and , as a result , gas - liquid phase separation is restored .
we have extended the debye - hckel method to treat three - dimensional anisotropic lattice models of electrolytes .
phase diagrams for different degrees of anisotropy have been obtained . for weakly anisotropic simple tetragonal lattices , the order - disorder phase transitions with a tricritical point
suppress the possibility of low - density gas - liquid phase transitions .
however , for strongly anisotropic lattices gas - liquid phase coexistence is restored .
thus the lattice anisotropy for simple tetragonal lattices mimics the lowering of the spatial dimensionality .
however , the thermodynamics of the body - centered tetragonal lattice ionic systems is very different .
there is no gas - liquid separation and the phase diagram has only order - disorder phase transitions for all degrees of anisotropy .
this is the consequence of the special symmetry of body - centered lattices .
the relevance of our results for understanding the thermodynamics of real ionic fluids remains unclear .
however , our method may be more useful for description of the thermodynamics of real ionic crystals with defects.@xcite furthermore , numerical simulations of the anisotropic lattice ionic models with lattice coulombic potentials are clearly needed in order to check the validity of our theoretical predictions .
acknowledgement is made to the donors of the american chemical society petroleum research fund ( grant # 37867-g6 ) for support of this research .
we also acknowledge the support of the camille and henry dreyfus new faculty awards program ( under grant no .
nf-00 - 056 ) .
we would like to thank prof .
fisher , prof .
panagiotopoulos and prof .
m. robert for critical comments and useful suggestions . * figure captions * + + fig.1 .
gas - liquid coexistence curves for simple tetragonal cubic lattices predicted by pure dh theory for different values of the lattice anisotropy parameter @xmath50 .
critical parameters as a function of degree of anisotropy : ( a ) critical and tricritical temperatures ; ( b ) critical density ; ( c ) tricritical density . + fig.3 .
phase diagrams of electrolytes on simple tetragonal lattices with sublattice ordering for different degrees of anisotropy : ( a ) for @xmath4 and @xmath112 ;
( b ) for @xmath113 .
dashed lines show the metastable gas - liquid coexistence curves predicted by pure dh theory .
phase diagrams for ionic systems on body - centered tetragonal lattices with sublattice ordering .
the gas - liquid coexistence curves predicted by pure dh theory are shown by dashed lines . | systems of charged particles on anisotropic three - dimensional lattices are investigated theoretically using debye - hckel theory .
it is found that the thermodynamics of these systems strongly depends on the degree of anisotropy . for weakly anisotropic simple cubic lattices ,
the results indicate the existence of order - disorder phase transitions and a tricritical point , while the possibility of low - density gas - liquid coexistence is suppressed . for strongly anisotropic lattices
this picture changes dramatically : the low - density gas - liquid phase separation reappears and the phase diagram exhibits critical , tricritical and triple points . for body - centered lattices ,
the low - density gas - liquid phase coexistence is suppressed for all degrees of anisotropy .
these results show that the effect of anisotropy in lattice models of electrolytes amounts to reduction of spatial dimensionality . |
ising susceptibility , high - temperature series , low - temperature series , scaling function , irrelevant variables , differentiably finite functions , scaling fields .
since onsager s @xcite celebrated solution of the ising model free energy in 1944 , followed by yang s @xcite proof of onsager s result for the spontaneous magnetization in 1952 , almost half a century has passed during which time many , if not most of the world s most able mathematical physicists have devoted themselves to the problem of elucidating the susceptibility . attempting to list all these contributions would produce a bibliography of prohibitive length , and one that would inevitably commit many sins of omission .
therefore rather than attempt this , we will only make mention of those papers that have directly motivated our work here , and crave the forgiveness of those who we have inadvertently offended . while much of the notation for describing the square lattice ising model is standard , we begin by defining our notation here both for the benefit of the more casual reader and to emphasize those cases where we deviate from convention . the interactions in the two perpendicular directions are taken to be @xmath6 but we also often set @xmath7 to discuss the isotropic lattice . for high temperatures , @xmath8 and
@xmath9 are appropriate variables for series expansions @xcite , while for low temperatures , we use @xmath10 and @xmath11 instead .
thus , @xmath12 in many cases , high - temperature and low - temperature formulas can be obtained from each other by kramers - wannier duality with a simple interchange of primes and stars .
the critical temperature is defined by the condition @xmath13 .
a conventional high - temperature variable is @xmath14 , while an often - used low - temperature variable is @xmath15 .
the translations between these and our variables are @xmath16 and @xmath17 and similarly for the primed variables . in studying the critical behavior ,
we will use both the variable @xmath18 and more frequently @xmath19 to parameterize deviations from the critical temperature . to leading order
, @xmath20 an elliptic parameterization will be useful , and to that end we define the elliptic modulus , @xmath21 let @xmath22 be the spin at lattice site @xmath23 and define the two - point function @xmath24 in terms of the correlation functions the susceptibility is @xmath25 where @xmath26 is the magnetization . in 1956
syozi and naya @xcite presented an approximation to the anisotropic high - tempera - ture susceptibility which gave the correct critical point , correct critical exponent , an amplitude estimate that was wrong by less than 1% , and reproduced the first 8 terms of the series expansion .
it was also exact along the disorder line @xmath27 the condition for the disorder line is that @xmath28 along this line the correlations decay exponentially , and the partition function factorizes @xcite .
the disorder condition is of no particular interest in the case of the nearest neighbor square lattice ising model . ] in 1976 a celebrated paper by wu , mccoy , tracy and barouch @xcite showed how the high- and low - temperature expansions of the susceptibility could be understood in terms of a multi - particle expansion , with an odd number of particles being appropriate at high temperatures and an even number at low temperatures . with this interpretation
it became clear that the result of @xcite was just the lowest order , or one - particle approximation to the susceptibility . in terms of the elliptic modulus @xmath29 ( [ mod ] )
the high- and low - temperature susceptibilities can be written @xmath30 and @xmath31 respectively , where @xmath32 is the sum over all lattice separations of the @xmath33-particle contribution to the two - point function , and was first given @xcite as a @xmath34-fold multiple integral .
it was subsequently shown that this integral can be reduced to a @xmath33-fold integral of the form @xmath35 where @xmath36 is a fermionic determinant and @xmath37 is an algebraic function .
this reduction has been achieved by various routes @xcite .
the factor , @xmath36 , appearing in the integrand , which can be expressed in terms of pfaffians @xcite , has been found to have a product form , first by palmer and tracy in the low temperature regime @xcite , and then independently by yamada in both the high- and low - temperature regimes @xcite . from the product form
it readily follows that the first non - zero term in ( [ e1 ] ) is @xmath38 from the original expression one could only conclude that each integral entered at order @xmath39 so clearly massive cancellations occur . while this comes as a surprise if handling the integral @xcite directly , it follows straightforwardly @xcite from the product form . in terms of the elliptic modulus , the first terms in the high- and low - temperature expansions for the isotropic ( @xmath40 ) susceptibility are @xmath41 and @xmath42 where e and k are the complete elliptic integrals . ] of the second and first kind respectively .
anisotropic versions of these formulae are given in section [ sect : integrals ] .
simple forms for higher terms in the expansion are not known .
in @xcite one of us gave compelling evidence that unlike the free - energy and spontaneous magnetization , the _ anisotropic _ susceptibility @xmath43 is not _ differentiably finite .
_ a series in @xmath44 variables , @xmath45 , is said to be _ differentiably finite _ or _ d - finite _ if and only if it satisfies a system of @xmath44 partial differential equations of the form @xmath46 where the @xmath47 are polynomials and for each @xmath48 , @xmath49 is not the null polynomial , see _ e.g. _ proposition 2.2 in @xcite .
thus the expression for the susceptibility was shown to be in a different and less tractable class of function than other known properties of the ising model .
the evidence for this was based on the observation ( not proved ) that the anisotropic susceptibility @xmath50 as a function of @xmath51 with @xmath52 fixed has a natural boundary on the unit circle @xmath53 for the _ isotropic _ susceptibility , another of us @xcite provided strong confirmation ( though again , not a proof ) of this observation by showing that the circle @xmath54 in the complex @xmath55 plane is a natural boundary .
these two observations are discussed further in section [ sect : circle ] , following a discussion of the general anisotropic case in section [ sect : si ] . in section [
sect : dfinite ] we also prove the important result that while @xmath56 is not d - finite , @xmath57 _ is _ d - finite for all @xmath58 two other directions in which we have achieved substantial progress are in the generation of series coefficients for the individual series @xmath59 continuing work initiated in @xcite , and even greater progress in obtaining the coefficients of the total series @xmath60 using nonlinear partial difference equations for the correlation functions @xcite . in order to generate the series for the total susceptibility @xmath61 or @xmath62 without computing separately the @xmath33-particle contributions , a more efficient method of series generation
is obtained by first returning to the expression ( [ 1.8 ] ) of the susceptibility as the sum over all lattice separations of the two - point correlation functions . in the scaling limit
, the two - point functions were found to satisfy a nonlinear differential equation of painlev type @xcite which was then solved to give the leading scaling terms @xmath63 and @xmath64 exactly @xcite .
in 1980 , a discrete analogue of this equation was discovered by mccoy and wu @xcite which holds for the two - point functions of the lattice ising model at arbitrary temperature , and which reduces to the painlev equation in the scaling limit . in the same year , a simple set of partial difference equations was derived by perk @xcite , which reproduced the equation of mccoy and wu and provided an additional equation .
also in the same year , an unrelated set of difference equations , which can be used to compute the correlation functions on the diagonal , @xmath65 , was obtained by jimbo and miwa @xcite .
many of these developments are described in some detail in @xcite .
the difference equations are valid for arbitrary temperature and were used by kong , _ et al . _
@xcite to obtain exactly the leading short distance " constant terms in the susceptibility both at the ferromagnetic and anti - ferromagnetic points at @xmath66 this work was later extended to give the amplitudes of the term @xmath67 @xcite .
here we dramatically extend that work by obtaining all terms in the short - distance " part @xmath68 ( see ( [ 6e ] ) and ( [ 9b ] ) ) of the susceptibility to o@xmath69 details necessary for the generation of @xmath70 appear in section [ sect : difference ] while in section [ sect : sdc ] we describe the numerical analysis of the @xmath70 that leads us to conclude that the short - distance " terms have the form @xmath71 although clearly nonanalytic at @xmath72 we have denoted these short - distance " terms in ( [ 6f ] ) by @xmath73 as a further reminder that they also include the analytical background . "
the actual coefficients in ( [ 6f ] ) can be found in the appendix .
the quadratic difference equations of perk @xcite can also be used to generate high- and low - temperature series for @xmath60 and as shown in section [ sect : difference ] the series coefficients can be obtained in polynomial time ! while some people have expressed the view that a polynomial time algorithm for the computation of the series coefficients constitutes a solution , it is clearly preferable to have a closed form expression .
nevertheless , a polynomial time algorithm is equivalent to a complete solution if one seeks only the series coefficients , as to expand any closed form expression also takes polynomial time .
as the history of the development of these key nonlinear recurrences described above shows , the ingredients for such an algorithm have existed unexploited in the literature for many years .
analysis of the resulting series of hitherto unimaginable length combined with the short - distance " knowledge contained in ( [ 6f ] ) leads to a solidly based conjecture specifying completely the remaining scaling " part of the susceptibility @xmath60 of the isotropic ising model . near the anti - ferromagnetic point the short - distance " terms are complete and we simply have for @xmath74 @xmath75 near the ferromagnetic point we conjecture ( for @xmath74 or @xmath76 ) @xmath77 where the scaling - amplitude functions @xmath78 have ( possibly asymptotic ) expansions in integer powers of @xmath79 without any of the powers of @xmath80 present in @xmath81 the leading terms are @xmath82 where @xmath83 in fact the breakdown in equality is dramatic , and we estimate that @xmath84 and @xmath85 where more accurate values of these and further terms in ( [ 9d ] ) through order @xmath86 are given in the appendix . in section [
sect : anal ] we describe the short - distance " subtraction and analysis on which the assumed form of the expansion of the scaling - amplitude function @xmath78 ( [ 9d ] ) is based , while the analysis leading to the numerical values of the coefficients in ( [ 9d ] ) is sketched in section [ sect : ca ] .
aharony and fisher @xcite have predicted a scaling - amplitude function @xmath87 that is equal above and below @xmath88 on the assumption that the ising model critical region can be described entirely by two nonlinear scaling fields .
our exact result ( [ 9d ] ) is clearly different and furthermore the explicit expansion ( cf .
( 22 - 24 ) in @xcite ) @xmath89 differs from ( [ 9d ] ) at order @xmath90 this is unequivocal evidence for the presence of at least one , and almost surely two , irrelevant operators changes sign in ( [ 9d ] ) and ( [ 9c ] ) as we change from @xmath91 to @xmath92 .
there is further possible evidence for irrelevant operators in the short - distance " terms ( [ 6f ] ) which contain powers of @xmath93 beyond the first starting at @xmath94 , and thus are not of the energy " form given by the nonlinear field analysis .
an important numerical study investigating corrections to scaling was that of gartenhaus and mccullough @xcite who confirmed the @xmath95 form in ( [ 9c ] ) through o@xmath96 and provided a good estimate of the term linear in @xmath79 in @xmath97 in ( [ 9b ] ) .
estimates of terms to o@xmath98 in @xmath99 in ( [ 6e ] ) were obtained in @xcite .
an attempt @xcite to go beyond this using longer series than available in @xcite was inconclusive other than to indicate the necessity of terms beyond that predicted by aharony and fisher @xcite .
our numerical work began in part as a modest attempt to improve on @xcite but expanded to where it now clearly quantifies the effect of nonlinear scaling fields @xcite as well as irrelevant operators .
as anisotropy is a marginal operator , extending the present calculation to the anisotropic square lattice would , we expect , be extremely helpful in better understanding the effect of the irrelevant operator(s ) we have identified .
both this and a study of the susceptibility on the triangular and hexagonal lattices are projects we hope to tackle in the near future .
the layout of the paper is as follows . in section [
sect : defs ] we define the model and give some useful parameterizations . in section [
sect : integrals ] we show how the key integrals referred to above may be simplified , and provide a short proof of the assertion that @xmath100 , so defined , is d - finite , even though , as we have seen , @xmath60 is presumably not . in section [
sect : difference ] the computation of the susceptibility series from the correlation functions by means of nonlinear partial difference equations is shown to be achievable in polynomial time . in section [
sect : qseries ] we first discuss the isotropic series in @xmath29 , and then the @xmath101 series form of the susceptibility . in this subsection some regularity features of the coefficients are discovered and the consequences partially developed . in section [
sect : numerical ] we summarize our numerical work , state our conjecture giving the complete analytic structure of the isotropic susceptibility , and quantify the effect of irrelevant variables in the scaling fields . finally , in section [ sect : scaling ] we review scaling theory as it applies to the two - dimensional ising model , and comment on the relevance of our results to this theory and to the renormalization group .
the high- and low - temperature series for @xmath102 are given on the www at site www.ms.unimelb.edu.au/@xmath103tonyg .
in this section of the paper it will be convenient to formulate the general anisotropic model . our later numerical work is confined to the isotropic model .
the complementary modulus of the elliptic modulus @xmath29 ( [ mod ] ) is given by @xmath104 . here the prime is not related to the anisotropy .
the moduli , @xmath29 and @xmath105 , are related to the elliptic nome , @xmath101 , by @xmath106 the connection between the two moduli yields a well - known theta function identity .
( see section 13.20 of bateman @xcite for these and other formulas . ) in terms of these variables the magnetization takes a particularly simple form @xcite @xmath107 for @xmath108 and @xmath109 otherwise .
the theta functions introduced above have useful infinite sum and product forms , which we will subsequently use . for zero argument
these are @xmath110 thus , @xmath111 in the elliptic parameterization , one variable is taken to be either @xmath29 or @xmath101 .
we take the other variable to be the anisotropy parameter .
onsager @xcite used the variable @xmath112 defined by @xmath113 we will also need the related variable @xmath114 in some contexts , it will be useful , following ref .
@xcite , to use instead the variables @xmath115 and @xmath116 defined by @xmath117 obviously @xmath118 .
the isotropic values of these variables are @xmath119 and @xmath120 .
we observe that @xmath121 where @xmath122 is the variable of ref .
@xcite , defined by @xmath123 in which the spontaneous magnetization has a simple product form .
the elliptic modulus of ref .
@xcite is related to our @xmath29 by a landen transformation @xcite . in reference
@xcite it was also noted that the first terms of the isotropic ( @xmath40 ) high - temperature susceptibility have a simple product expression , which breaks down at order @xmath124 .
this is explained by the expansion of the high - temperature susceptibility in multi - particle states described in the introduction .
the contribution of one - particle states has a product form , and the first three - particle state contributes at order @xmath124 .
this one - particle product form is @xmath125 ( strictly speaking the product form above is not the product form of ref .
@xcite which actually breaks down only at order @xmath126 .
they differ by a factor which first contributes at order @xmath124 and the absence of this factor in ref .
@xcite exactly compensates for the addition of the first term of @xmath127 . )
attempts to do the same for the low - temperature series have failed as we do not know of a product form for @xmath128 .
the function k has one , namely @xmath129 while the function e is given in terms of @xmath101 by the formula @xcite
in this section we present integral expressions for the @xmath100 and show that they define d - finite functions .
we will start with the form given by yamada @xcite for the @xmath33-particle contribution @xmath100 in terms of elliptic variables , and derive from it an expression in terms of trigonometric / hyperbolic variables .
the integral expression is @xmath131 where the modulus of the complete elliptic integral @xmath132 is @xmath29 , and this same modulus is assumed in all jacobi elliptic functions which appear below .
considered as a function of one of the @xmath133 , the integrand has a single simple pole on the real axis which derives from the last factor in the integrand .
the contour of integration is deformed in the vicinity of the pole so that only half the residue is taken .
the function @xmath36 can be written as the product @xmath134 and @xmath135 , @xmath136 and @xmath137 are given by @xmath138 note that the trigonometric / hyperbolic variables , @xmath139 and @xmath140 are related to the elliptic variables , @xmath133 and @xmath141 by the functional equation @xmath142 the choice of @xmath141 rather than @xmath112 as anisotropy parameter is arbitrary because of the symmetry under interchange of horizontal and vertical lattice axes .
choosing @xmath141 at this point results in expressions for @xmath139 and @xmath140 which are equivalent to those of onsager .
the mapping between the elliptic parameterization and the trigonometric / hyperbolic one was described in ref .
@xcite for the isotropic case .
the formulas below are a generalization of this mapping .
it is simpler to use the trigonometric parameterization than it is to use the elliptic parameterization for numerical series generation .
the variables @xmath139 and @xmath140 defined above satisfy the identities given in appendix 2 of onsager s paper @xcite , @xmath143 with @xmath144 notice that it is @xmath112 which appears in these formulas rather than @xmath141 .
the mapping from the variables @xmath133 and @xmath112 to the variables @xmath139 and @xmath140 is conformal , as is seen from the formulas , also given by onsager @xmath145 finally , onsager gives the functional equation @xmath146 to make the change of variables , we first note that @xmath147 corresponds to @xmath148 , and @xmath149 corresponds to @xmath150 . using eq .
( [ conformal ] ) , we see that @xmath151 which implies @xmath152[3.13 ] the condition @xmath153 mod @xmath154 , which results from having performed one of the integrations , is assumed . in terms of @xmath139 , @xmath29 and @xmath115 ,
the quantities @xmath136 and @xmath155 can be expressed as @xmath156 , \\
\sinh\gamma_n&= \cot^2\alpha\sqrt{(\xi-\cos\omega_n)^2-(\cot\alpha)^{-4}}\end{aligned}\ ] ] with @xmath157 in the isotropic case , these reduce to @xmath158 here @xmath159 for high - temperature and @xmath160 for low temperature , although the distinction is irrelevant as the dependence of the integrand on @xmath161 and @xmath10 is symmetric .
finally @xmath162 when @xmath163 or @xmath164 , the integrals can be rewritten in terms of known functions as was noted in the introduction for the isotropic case .
when @xmath165 glasser @xcite showed that the integrals can be written as an integral involving the square root of a polynomial of degree higher than 4 .
such integrals are often called _ hyperelliptic _ integrals , and are a special case of _ abelian _ integrals .
the corresponding anisotropic expressions are @xmath166,\\ { { \hat\chi}^{({2})}}=&\frac{k^{1/2}}{1+k}\sqrt{(k^{1/2}-k^{-1/2})^2 + 4\csc^2 2\alpha } \,{{\hat\chi}^{({2})}_{\text{iso}}}.\end{aligned}\ ] ] as we show in section [ sect : qseries ] , we can say more about the general form of the expansion , based on inspection of the long series published by nickel @xcite , as well as more recent extensions reported here , giving hope that there is still more regularity to be found . as mentioned in the introduction , there were observations , first in @xcite and then in @xcite , that strongly suggested the susceptibility of the ising model is a function with a natural boundary unlike the free - energy or magnetization .
we expand and clarify those observations here in the light of our new knowledge of the general anisotropic case discussed in section [ sect : si ] and the additional numerical work on the isotropic limit described in section [ sect : numerical ] .
first , we expect that as described in @xcite , the @xmath167 given by the integrals ( [ 3.13 ] ) multiplied by the factors outside the summation in eq .
( [ 1a ] ) or ( [ 1b ] ) will be singular at the symmetry points of the integrand and where the denominator factor @xmath168 vanishes .
the symmetry point condition requires all @xmath139 to be equal and given by @xmath169 @xmath170 the vanishing of the denominator factor requires the @xmath136 , now all equal , to be given by @xmath171 @xmath172 equivalently , from the explicit formula ( [ 3.14 ] ) , @xmath173 with @xmath174 from ( [ alpha ] ) and @xmath175 given by ( [ 3.16 ] ) , we find ( [ last ] ) can be reduced to @xmath176 with @xmath177 as discussed above .
it will be noted that the left - hand side of ( [ 3.24 ] ) is the denominator in the onsager integral for the free - energy and thus we find the ( to us ) surprising result that the singularity of @xmath167 , a property of the ising model in a magnetic field , is intimately connected with a property in zero field .
the full @xmath60 , being a sum of @xmath167 , will naively be expected to be singular at a dense set of points and thus have the onsager line ( [ 3.24 ] ) as a natural boundary .
the presence of natural boundaries has implications for expansions about the physical singularity points @xmath178 that are necessary to understand corrections to scaling .
we briefly explore some of these implications but restrict ourselves for simplicity to the ferromagnetic point @xmath179 in the isotropic model .
we also make the plausible assumption that the singularity in each @xmath167 closest to @xmath72 is the most important for determining , in expansions of @xmath60 , the @xmath180 large @xmath181 asymptotics and this considerably simplifies the discussion map to points on the imaginary axis in the complex @xmath79 plane . ] .
let @xmath182 be that singularity in @xmath167 that is closest to the ferromagnetic @xmath183 ; @xmath184 is fixed by @xmath185 with @xmath186 choose the branch - cut arising from this singularity to lie along the imaginary @xmath79 axis and directed away from @xmath187 take @xmath188 to be a point on the branch - cut .
provided the ( positive ) deviation @xmath189 is not too large , the discontinuity across the cut can be estimated from the linearized singularity equations ( 14 ) in @xcite and ( 12 ) in @xcite . for @xmath190 a more general result that includes
the first order correction is @xmath191^{(j^2 - 3)/2 } \nonumber \\ & \times&\left[1 + \delta\theta_j\sin{\theta_j } \left\{\frac{\cos{\theta_j}}{4\sin^2\theta_j } + \frac{j^2 - 4}{8\cos{\theta_j}(1+\cos{\theta_j } ) } - \frac{1}{2}\right\ } + { \rm o}(\delta\theta_j^2)\right]\end{aligned}\ ] ] where @xmath192 the last equality is valid only in the large @xmath33 limit . to obtain the discontinuity in say @xmath61
, we must first sum the discontinuities in @xmath167 , @xmath33 odd , and this we can crudely estimate by integrating over @xmath33 , keeping only the leading exponential factors in ( [ new26 ] ) and ( [ new27 ] ) and making a small angle approximation as well .
essentially the same formula is obtained for the discontinuity in @xmath62 on summing over @xmath33 even ; in either case we find @xmath193^{j^2/2}\ ] ] where the sum is over @xmath33 odd or @xmath33 even , and the lower limit is simply the restriction to those @xmath33 values that contribute , while the upper limit roughly defines the limit of validity of the linearized approximation .
the precise value of this limit is not important since the integrand has a maximum well below the limit . for large @xmath33
only the maximum of the integrand matters and ( [ new29 ] ) reduces to @xmath194 in which the numerical coefficient of @xmath195 is @xmath196 with @xmath122 the smallest real solution of @xmath197 keeping terms such as the correction term in ( [ new27 ] ) and the @xmath198 in ( [ new28 ] ) in the steepest descent analysis lead to o@xmath199 corrections to the exponent in ( [ new30 ] ) so we conjecture that the right hand side of ( [ new30 ] ) is exactly the leading exponential .
it is interesting that the discontinuity ( [ new30 ] ) is similar to that found in weak coupling field theory expansions , but the mechanism producing the cut here does not seem to be related in any way to instantons . the additional singularities that @xmath60 has at @xmath200 namely terms such as the divergent scaling " @xmath63 or the
short- distance " powers of @xmath201 are a complication we do not know how to handle in any rigorous fashion in ( [ 9b ] ) formed a convergent sequence , then a subtraction process similar to that described in section [ sect : anal ] could be carried out here . the numerical evidence from section [ sect : anal ] is that the point @xmath72 in the scaled and pole subtracted @xmath202 is no longer a branch point singularity and hence the final result ( [ eqcut ] ) is justifiable and would apply directly to the scaling - amplitude function @xmath203 unfortunately , we do not have any independent information to suggest that the short - distance " sum ( [ 6f ] ) is convergent and thus the argument leading to ( [ eqcut ] ) is at best suggestive that one or the other ( or both ) of the short - distance " or scaling " sequences are asymptotic . ] . as a consequence we will simply ignore them and make the simplest , yet reasonable , assumption that they make an additive contribution not relevant for understanding the effect of the natural boundary .
then the cut discontinuity ( [ new30 ] ) would imply a divergent behavior in the @xmath79 expansion of @xmath60 . that is to say ,
the coefficient of @xmath180 in the limit @xmath204 will diverge as @xmath205 with @xmath206 .
this follows from a contour integral around the origin distorted to run on either side of the cut imaginary axis .
the contribution of the cut discontinuity to the coefficient , @xmath207 , of @xmath180 in the expansion is then @xmath208 we know little about the cut discontinuity on the circle @xmath54 other than what we have deduced near @xmath209 as given by ( [ new30 ] ) .
however , the fact that the amplitudes of the singularities of @xmath167 on @xmath54 vary dramatically with order @xmath33 almost certainly implies there will be no cancellation of singularities in the sum of @xmath167 that defines @xmath210 furthermore , the variation with order means that there is no length scale at which , as one approaches the circle of singularities @xmath211 @xmath102 can be smooth .
it is these two points taken together that we consider overwhelming evidence that , in the isotropic case at least , @xmath60 has a natural boundary that is the entire @xmath54 circle .
we do not imply by this that all points on the circle are equally singular " . as argued above , the existence of an asymptotic expansion about @xmath209 seems likely .
a very different situation arises at a point such as @xmath212 while the singularity in @xmath167 on the circle @xmath213 nearest @xmath209 lies at a distance @xmath214 for large @xmath215 the corresponding nearest distance from @xmath216 is @xmath217 furthermore this latter singularity is larger in its leading amplitude than the former by a factor roughly @xmath218 the reduction in distance and dramatic increase in amplitude suggests that an asymptotic expansion about @xmath216 is not possible , but this has not been proved .
for the anisotropic case we have not analyzed ( [ 3.13 ] ) in detail so we do not have the necessary amplitude information to make the same claim directly .
however , we can take the extreme anisotropic limit of @xmath219 infinitesimal ( but not @xmath109 ) and find that the onsager line ( [ 3.24 ] ) has now come very close to the circle @xmath220 at this point we can connect to the work in @xcite .
there it was observed that if @xmath43 was written as @xmath221 the @xmath222 would be singular at a dense set of points on @xmath223 as @xmath224 furthermore , if as in the discussion above , @xmath52 is infinitesimal ( but not @xmath109 ) the amplitudes of the singularities vary dramatically with order @xmath44 and the same conclusion as in the isotropic case is reached . given that the onsager line ( [ 3.24 ] ) is a natural boundary in two extremes , it seems highly probable that it is also a natural boundary at all intermediate anisotropy values .
we conclude this section by contrasting the above behavior of the susceptibility with that of the free - energy and magnetization which are only singular at an isolated set of points , not a dense set .
this is precisely what one expects for a d - finite function , as in that case we have the following let @xmath225 be a d - finite series in @xmath226 with rational coefficients . for @xmath227 let @xmath228 be the set of poles of @xmath229 ;
let @xmath230 then @xmath231 has only a finite number of accumulation points .
this is observed in practice .
the anisotropic magnetization and free - energy each have exactly one accumulation point @xcite , while the ( non - d - finite ) susceptibility appears to have an infinite number @xcite .
[ sect : dfinite ] this remarkable result , that @xmath167 ( or equivalently @xmath100 ) is d - finite while @xmath60 is not , follows from the results of lipshitz @xcite ( see also zeilberger @xcite ) , who gives several basic definitions and theorems concerning d - finiteness of series in several variables .
the integrand of the trigonometric / hyperbolic form of @xmath100 is an algebraic function of the variables @xmath232 and @xmath233 , and is thus a d - finite function in these same variables .
then by theorem 2.7 of ref .
@xcite , integrating over one or more variables preserves d - finiteness which implies the result .
the term d - finite is synonymous with _
holonomic _ in much of the literature .
kashiwara and kawai @xcite have shown that any feynman diagram is holonomic , whereas an infinite sum of such diagrams may not be .
this is just the phenomenon we observe here . at first glance
it appears that kashiwara s definition of holonomic differs from that used here , but this is not so .
the point is that the definition of d - finite functions of more than one variable requires that the underlying system of partial differential equations be such that only a finite number of initial conditions are needed to specify the function .
such systems are called maximally over - determined " or holonomic " in the analysis literature . in the single variable case ,
the question of a finite number of initial conditions is clearly automatically satisfied . motivated by the above observation
, we have attempted to find linear differential equations with polynomial coefficients in @xmath233 , or equivalently , linear recurrences for the series coefficients , of @xmath127 and @xmath234 . with
the available series of order @xmath235 and @xmath236 respectively , calculated using the methods of numerical integration described in @xcite , we have ruled out any such recurrences of depth 14 with coefficients of degree 15 for @xmath127 , and of depth 6 with coefficients of degree 7 for @xmath234 .
thus while these functions are provably d - finite , it is clear that the generating differential equation will be a fairly cumbersome object .
we have also attempted to fit these series as polynomials in the complete elliptic integrals k and e with polynomial coefficients in @xmath29 and obtained similar negative results .
in this section we give details of our more efficient method of series generation based on summing the correlation functions , obtained by means of nonlinear recurrences , as outlined in the introduction .
the equations we used for generation of the very long series are the ones given by perk @xcite . here
we present a slight generalization due to mccoy and wu @xcite which keeps track of the separate multi - particle components .
the expansions of the two - point correlation functions in multi - particle components are analogous to the corresponding expansions ( [ 1a ] ) and ( [ 1b ] ) for the susceptibility @xmath237 with @xmath238 and for @xmath239 @xmath240 the fugacity @xmath241 is associated with the number of particles , and @xmath242 . with these definitions the quadratic partial difference equations are @xmath243\nonumber\\ & \quad+[c^*(m , n;\lambda)^2-c^*(m-1,n;\lambda)c^*(m+1,n;\lambda)]=0\\ \label{pd2 } & { s'}^2 [ c(m , n;\lambda)^2-c(m-1,n;\lambda)c(m+1,n;\lambda)]\nonumber\\ & \quad+[c^*(m , n;\lambda)^2-c^*(m , n-1;\lambda)c^*(m , n+1;\lambda)]=0\\ \label{pd3 } & ss'[c(m , n;\lambda)c(m+1,n+1;\lambda)-c(m , n+1;\lambda)c(m+1,n;\lambda ) ] \nonumber \\ & \quad = c^*(m , n;\lambda)c^*(m+1,n+1;\lambda)-c^*(m , n+1;\lambda ) c^*(m+1,n;\lambda).\end{aligned}\ ] ] the object @xmath244 is the correlation function on the dual lattice and is obtained from @xmath245 by replacing @xmath219 with @xmath246 and @xmath161 with @xmath247 .
equation ( [ pd3 ] ) holds for all @xmath248 and @xmath1 .
when @xmath249 equations ( [ pd1 ] ) and ( [ pd2 ] ) must be replaced by @xmath250 we do not know of @xmath251 versions of these equations .
these equations are nearly enough to determine all two - point functions completely . for the isotropic expansion ,
all that is lacking is either the high or the low temperature set of diagonal correlations ( @xmath65 ) . from either one of these the other
can be obtained using equation ( [ pd3 ] ) with @xmath65 .
when @xmath251 we have used the integral formula ( [ ci ] ) to compute the diagonal correlation functions . for the rest of this section we focus on the case @xmath252 where two superior methods for obtaining the diagonal correlations are available . from the purely computational point of view the difference equations of jimbo and miwa @xcite are almost certainly the most efficient and to be preferred . however , from the point of view of understanding the analytical structure of the correlations the original toeplitz determinants @xcite are better .
we also find that for the numerical computations we have carried out so far the evaluation of the determinants is only a small fraction of the total project time so efficiency is not yet an issue .
we will restrict ourselves in the following to the isotropic lattice .
in that case and for @xmath253 the diagonal correlation @xmath254 is the determinant of an @xmath255 toeplitz matrix with elements @xmath256 that are the integrals @xcite @xmath257 these integrals apply both above and below @xmath88 and furthermore since @xmath258 corresponds to @xmath259 one can establish from the integral in ( [ 7a ] ) the relations @xmath260 explicit formulae for @xmath261 for small @xmath44 can be given in terms of elliptic integrals e and k but these expressions are not particularly enlightening and are not needed here .
rather we need the series expansions of ( [ 7a ] ) , either in @xmath262 @xmath10 or @xmath79 depending on the application .
the series expansions in @xmath161 and @xmath10 both for @xmath261 and @xmath254 are completely straightforward with computer packages such as maple that automatically handle the multiple precision arithmetic required .
furthermore , these same packages can be set to treat the @xmath70 in the recursion formulae ( [ pd1])([pd3 ] ) as series and thus very little programming is necessary to generate the susceptibility .
admittedly , some steps need to be taken to conserve time and/or memory resources but this is very hardware dependent and will not be described here .
what is worth noting , however , is that for a series to order @xmath1 the recursion formulae require o@xmath263 multiplications of series of length @xmath1 and thus in a naive implementation , o@xmath264 multiplications .
since the word length grows linearly with @xmath1 the algorithm has complexity of at most o@xmath0 .
there are more efficient ways to multiply long series and numbers with a large number of digits @xcite but we have not found it necessary to explore these options . timing proportional to @xmath265 is observed in practice , in our implementation of the recursion in maple .
we have generated high - temperature series of order 323 and low - temperature series of order 646 for the isotropic susceptibility .
the entire calculation took 123 hours on a 500mhz dec alpha with 21164 processor running maple v version 5.1 .
we have also obtained shorter anisotropic series in this way ( either the nearest off - diagonal correlation functions , or additional assumptions are needed ) . as
lattice anisotropy is a marginal operator , we hope that an extension of this calculation will be very revealing .
the series in @xmath79 is most easily obtained by expressing the toeplitz element integral ( [ 7a ] ) in terms of hypergeometric functions . to show this connection we start by writing @xmath266 in terms of the real integrals @xmath267 where @xmath268 the required symmetry @xmath269 is explicit in eqs .
( [ 7aa ] ) and ( [ 7ab ] ) and by standard integration by parts manipulation one can show that the cosine integral satisfies the differential equation @xmath270 while the sine integral can be expressed as the derivative @xmath271 furthermore , a direct evaluation of the integral in ( [ 7ab ] ) for small @xmath79 yields @xmath272 with @xmath273 and this initial condition together with ( [ 7ac ] ) completely determines @xmath274 since ( [ 7ac ] ) can be recognized as the hypergeometric differential equation in the variable @xmath275 we can immediately write the solution as @xcite @xmath276 on taylor expansion we now obtain @xmath277\tau^2/2 + \nu^2 ( \ell_{\nu } - 1)\tau^3/4 \\
& \mbox { } + [ \nu^3(\ell_{\nu } - 5/4)-\nu - 2\nu^{-1}]\tau^4/16 + [ \nu^4(\ell_{\nu } - 3/2)-4\nu^2(\ell_{\nu } - 1/2)]\tau^5/64 + { \rm
o}(\tau^6 ) \nonumber\end{aligned}\ ] ] which can be extended as required . from the expression ( [ 7b ] ) one can conclude that @xmath278 as an @xmath255 determinant , will contain logarithmic terms @xmath279 with @xmath280 and @xmath281barring cancellation .
however , there is cancellation and the key qualitative features of the cancellation can be deduced simply by using an alternative representation for the determinant . in particular , by systematically subtracting rows and columns one can show that an equivalent determinant has matrix elements @xmath282 which are of the same integral form as the @xmath283 in eq .
( [ 7a ] ) except for the replacement of the exponential @xmath284 by the product @xmath285 the new matrix is no longer of toeplitz form but for our purposes here it is the better representation because of the powers of @xmath286 which have the effect of shifting the @xmath80 singularity of the integrals to higher order in @xmath287 indeed one can show from the new integrals that the leading singular behavior of each matrix element @xmath288 is proportional to @xmath289 and this is sufficient to show that in the logarithmic terms @xmath290 in @xmath254 one must have @xmath291 we have not attempted to pursue this argument to deduce @xmath254 analytically but rather have resorted to a numerical small @xmath1 evaluation of @xmath254 using ( [ 7b ] ) and then fitting to obtain formulae valid for general @xmath292 our results for the leading logarithm term are summarized by the expression @xmath293 in which we have used @xmath294 to denote the discrete logarithm , i.e. @xmath295 the product factor in the first braces of ( [ 7d ] ) is to be understood as unity for @xmath296 and in the final brace pair the coefficients of @xmath297 are polynomials in @xmath1 of degree @xmath298 furthermore , with the @xmath299 factor extracted explicitly as in ( [ 7d ] ) these coefficients of @xmath297 vanish if @xmath300 with @xmath301 the critical correlation factor in ( [ 7d ] ) is @xmath302 and approaches @xmath303 as @xmath304 where @xcite @xmath305 .
we have used ( [ 7d ] ) , extended or truncated to some order in @xmath79 , as input to the quadratic recursion formulae to generate all correlation products @xmath306 as series in @xmath79 within an octant @xmath307 @xmath308 while most of our results are numerical , they are consistent with the assumption that the structure observed on the diagonal remains true on the octant .
that is , if we define @xmath309 @xmath310 and set @xmath311 ( even shell ) or @xmath312 ( odd shell ) then on these even / odd shells for large @xmath313 the correlations are of the form @xmath314 where the @xmath315 are still taylor series in @xmath79 as in ( [ 7d ] ) but with coefficients that are now ( possibly asymptotic ) laurent series in @xmath316 rather than polynomial in @xmath292 the highest power of @xmath1 multiplying @xmath297 remains @xmath317 considerable care must be exercised in the numerical work since the recursion formulae are unstable .
a toy recursion relation of structure similar to the ones we use in the ising study illustrates this nicely .
let @xmath318 which is to be applied to all possible @xmath44 and iterated forward in @xmath319 the recursion ( [ 7h ] ) has as a solution a constant , say @xmath320 but is susceptible to a steady state growth of errors so that @xmath321 is also a possible solution .
on substituting ( [ 7i ] ) into ( [ 7h ] ) one finds the growth constant @xmath115 must satisfy @xmath322 so that @xmath323 this is very close to what we find in our numerical work and implies that if we want some number of digits @xmath324 that are accurate at the outer edge of the octant on a shell specified by @xmath325 we need to start with digits @xmath326 in practice we have worked as high as @xmath327 with @xmath328 using the automatic multiple precision facility of maple . in section [ sect : sdc ] we use these results to calculate the `` short - distance '' ( including analytic background ) terms in the susceptibility .
in this section we state some conjectures for the short distance behavior of the @xmath251 model introduced in section [ sect : difference ] .
we arrived at these conjectures by inspection of series obtained from a combination of the integrals ( [ ci ] ) and the difference equations ( [ pd1])([pd3 ] ) .
our interest in this model is motivated by several considerations .
firstly , it enables us to see how the analytic structure of @xmath60 evolves as successive contributions @xmath329 or @xmath330 are added .
secondly , the correlation function @xmath245 for small values of @xmath248 and @xmath1 may be required as initial conditions for certain series generation algorithms .
thirdly , we hope that the presence of an additional parameter which can be varied will provide some insight into the ising model itself .
finally , the deformations of the elliptic functions that appear in our conjectures may be of intrinsic mathematical interest .
following the lead of ref .
@xcite we make a change of variable from the modulus , @xmath29 , to the nome @xmath101 .
examination of the @xmath33-particle contributions to the isotropic susceptibility and two - point functions reveals that there is much regular structure .
we arrive at exact conjectures for @xmath331 , @xmath332 , @xmath333 , @xmath334 and @xmath335 as functions of @xmath33 and @xmath101 .
these provide the first terms in the short distance expansion of the susceptibility .
although more complete results have been obtained for the correlation functions , we will demonstrate the method by which we derived our conjectures using the susceptibility as an example . to make our observations more concrete , we reproduce tables of series coefficients for @xmath336 . as an example of how to interpret the table
, we read @xmath337 .coefficients of the series @xmath338 .
the numbers in the tables are the coefficients of @xmath339 starting at @xmath340 in the upper left corner and reading down and to the right . [ cols= " < , < , < ,
< , < , < , < " , ] inspecting table [ table : summary ] , we make the conjecture that the series can be sensibly decomposed as @xmath341 \label{decomp}\end{gathered}\ ] ] in fact , we observe that as @xmath33 tends to larger and larger values , the first row of coefficients tends towards the expansion in @xmath29 of @xmath342 . ) for @xmath343 and using the identity @xmath344 to obtain the expansion of @xmath345 in @xmath29 . ]
the first correction comes in at order @xmath346 .
hence we make the change of variable from @xmath29 to @xmath101 in @xmath347 and divide the result by @xmath348 . for all @xmath33
we obtain a series of the form @xmath349 the sequence of terms starting at order @xmath350 appears again to be fitted by a recognizable function of @xmath101 , at least until contributions appear at order @xmath351 or @xmath352 .
subtracting this assumed product form yields a series whose first correction enters at order @xmath351 .
the coefficients of the terms of orders lying between @xmath351 and @xmath352 are not independent of @xmath33 as was the case previously , but vary linearly with @xmath33 .
the constant part has a product form , and the @xmath33-dependent part may as well be we do not have sufficiently many terms to make a firm conjecture . likewise
the terms between @xmath352 and @xmath353 depend quadratically on @xmath33 , and the correction at @xmath353 appears to vary as the fourth power of @xmath33 . from the long series we have produced ,
we have been able to conjecture that @xmath354 \right ] \end{split}\ ] ] where @xmath355 these are consistent with the expressions for the correlation functions in the following section .
that is to say , summing the correlation functions given in the following subsection gives terms that agree , as far as they should , with the above expression .
similarly , summing ( [ split ] ) over @xmath33 gives a series that agrees to the appropriate ( low ) order with the known expansion for @xmath356 in this subsection we write down some conjectured results for the short - distance correlation functions which we have derived empirically .
for these cases , unlike the susceptibility , the general correction term is apparent from the series and we are able to formulate exact conjectures .
let us define operators , @xmath357 and @xmath358 , which convert power series in @xmath359 to power series in @xmath101 according to the rules @xmath360 then we conjecture that @xmath361 -4\left[\frac{1}{\theta_3(0,q)}\right.\nonumber\\ & \qquad\left.+\frac{\theta_2(0,q)}{2\theta_3 ^ 2(0,q)\theta_2(0,q^4 ) } -\frac{8}{\theta_3(0,q)\theta_2 ^ 4(0,q^{1/2})\theta_2(0,q^4 ) } q^{1/4}\frac{d}{dq^{1/4}}\left[\phi_0\frac{z^2}{(1+z^2)^2(1-z^2)}\right ] \right]\nonumber\\ & \qquad\times\phi_0\frac{z^{j+2}}{(1+z^2)^{j+2}(1-z^2)},\\ \text{and}\nonumber\\ & 2^{-j}(1-k^2)^{1/4}{{\hat c}^{({j})}}(2,1)=\frac{1}{\theta_3(0,q ) } \left(8q^{5/16}\frac{\theta_2(0,q^{1/4})}{\theta_2 ^ 5(0,q^{1/2 } ) } \frac{d}{dq^{1/4}}\left[\phi_1\frac{z^{j+1}(1+z)}{(1+z^2)^{j+1}(1-z^2)}\right ] \right.\nonumber\\ & \qquad+\left(\phi_1\frac{z(1+z)}{1-z^4}\right)^{-1 } \phi_1\frac{z^{j+1}(1+z)}{(1+z^2)^{j+1}(1-z^2)}\nonumber\\ & \qquad\left.\times\left[\theta_4(0,q)- 8q^{5/16}\frac{\theta_2(0,q^{1/4})}{\theta_2 ^ 5(0,q^{1/2 } ) } \frac{d}{dq^{1/4}}\left[\phi_1\frac{z(1+z)}{1-z^4}\right]\right ] \right)\end{aligned}\ ] ] as noted above , these are necessary , but not sufficient , for the generation of the @xmath33-particle contributions to the correlation functions , being some of the initial conditions for the recurrences .
the main result in this section is a conjecture completely specifying the analytic structure of the susceptibility @xmath60 of the isotropic ising model in the vicinity of the critical point both in the neighborhood of the ferromagnetic point @xmath209 and the anti - ferromagnetic point @xmath362 . the conjecture , contained in eqs . ( [ 6f][9d ] )
is based on what we believe is overwhelming numerical evidence that is obtained by disentangling the short- distance " and scaling " parts of @xmath60 in a manner described below . in section [ sect : sdc ]
we give the assumptions and numerical procedures we use to derive the short - distance " part of @xmath60 in ( [ 6f ] ) with the coefficients listed in the appendix .
then , in section [ sect : anal ] we describe the short - distance " subtraction and analysis on which the behavior of the scaling - amplitude function @xmath78 shown in ( [ 9d ] ) is based . finally , in section [ sect : ca ] we outline our fitting programs to determine the coefficients in the functions @xmath78 .
since there are no confluent singularities in @xmath78 whose amplitudes need to be found , the fitting procedure is very well - conditioned and the coefficients in the appendix are as determined to an accuracy of up to 20 digits . that the `` short - distance '' contribution to @xmath60 can be obtained from numerical values of @xmath363 for small @xmath364 and @xmath365 relies on certain assumptions about the behavior of the expansion coefficients in eq .
( [ 7 g ] ) .
in particular , we assume that ( [ 7 g ] ) remains valid up to @xmath1 of the order @xmath366 where it can , in principle , be matched term by term to a large distance expansion that properly describes the roughly exponential @xmath367 decay of correlations as @xmath368 explicit matching formed the basis of the previous calculations of terms in the short - distance " @xmath60 ( cf .
@xcite ) but this becomes extremely cumbersome at higher order .
our ability to go to high order here rests on the fact that we dispense with such explicit matching and rely instead on power counting to uniquely identify those terms that contribute to the scaling " and the short - distance " parts of @xmath60 separately .
we believe this is tantamount to the scaling argument that in the critical region there is a single length scale proportional to @xmath369 with @xmath370 and thus , in a way to be made more precise below , we can deduce the power law of the large distance contribution of any set of terms varying as @xmath371 by simply replacing @xmath371 with @xmath372 terms whose variation is as a fractional power of @xmath79 ( with possibly logarithmic multipliers ) are discarded as assumed contributions to the scaling " part of @xmath210 terms whose variation is predicted to be an integer power of @xmath79 ( with possibly logarithmic multipliers ) are assumed to be part of the short - distance " @xmath60 and are treated more carefully . to make the argument and assumptions more explicit we begin with some definitions that will be useful also for the subsequent analysis .
let the two dimensional sum ( [ 1.8 ] ) defining @xmath60 be reduced to a one dimensional sum by combining the contributions from all sites on the even and odd squares @xmath373 , @xmath374 and then further combining these into sum and difference combinations which are necessary for separating the ferromagnetic and anti - ferromagnetic contributions .
that is we write @xmath375 where the first sum simply defines the coefficients @xmath376 while the second is the actual lattice sum over the octant correlations defined by ( [ 7 g ] ) and extended by symmetry over the full square .
the coefficients @xmath377 in the expansion @xmath378 have the large @xmath1 asymptotic expansions @xmath379 as assumed in ( [ 7 g ] ) based on numerical evidence .
the ferromagnetic and anti - ferromagnetic cases have been treated separately in ( [ 7 m ] ) to emphasize that shell subtraction in ( [ 7k ] ) reduces the leading power of @xmath1 by unity .
the upper limit @xmath380 on the logarithmic powers in ( [ 7l ] ) depends on @xmath29 as discussed in connection with eq .
( [ 7d ] ) but its precise value is not needed in the following discussion .
we also introduce the partial sums @xmath381 where the @xmath382 in the second equality in ( [ 7n ] ) are numerical coefficients that are directly generated by the quadratic recursion relations and the subsequent lattice summations . in the final equality in ( [ 7n ] ) these expansion coefficients have been formally split into two terms .
this formal separation is to be understood in an asymptotic sense with @xmath383 a constant that is the @xmath1 independent part of @xmath384 as @xmath304 and which we can understand as an integration constant " .
the remainder @xmath385 , we will show in the case that @xmath386 , has an asymptotic expansion like the @xmath387 in ( [ 7l ] ) except for an extra factor of @xmath1 from the summation .
for @xmath388 the @xmath385 can be expressed as sums of such expansions with multiplying logarithmic factors @xmath389 we now come to our key assumption .
the partial sums in ( [ 7n ] ) in most cases diverge as @xmath304 because of the presence of large fractional powers of @xmath1 in the @xmath385 .
however , we assume that if the formal matching of the short and large distance expansions for @xmath363 had been carried out , these partial sums would in fact converge and furthermore the contribution of each power of @xmath1 term could be estimated by the replacement @xmath390 since all the powers of @xmath1 in @xmath385 are fractional , the result is that these terms contribute only to the scaling " part of @xmath210 thus the short - distance " part of @xmath60 comes entirely from the `` integration constant '' term @xmath383 in ( [ 7n ] ) , and on explicitly reinserting the factors of @xmath299 and @xmath391 we get @xmath392 at the ferromagnetic and anti - ferromagnetic points .
the last equality in ( [ 7o ] ) is eq .
( [ 6f ] ) and follows simply by expanding @xmath299 in a series in @xmath287 the rest of this section comprises a discussion of the numerical scheme we use to isolate the `` integration constants '' efficiently .
a technical problem arises in that one must somehow isolate the constant @xmath383 from an @xmath1-dependent sequence @xmath384 and this can be a very unstable procedure if the @xmath384 are combinations involving logarithms .
fortunately there are no logarithms in @xmath393 and one can iteratively remove the logarithmic factors in the remaining @xmath384 by working in sequence from @xmath394 to @xmath395 . to understand
precisely how to do the subtractions we first relate the unknown @xmath396dependent structure of the @xmath384 to the known , or rather assumed , simple structure of the @xmath397 from the definition of @xmath398 as a partial sum it follows that @xmath399 and by comparing coefficients in eqs .
( [ 7n ] ) and ( [ 7l ] ) we get @xmath400 in the case that @xmath386 there are no logarithmic factors on the right hand side of ( [ 7p ] ) and the equation is just a discrete first order differential equation in @xmath1 whose solution is an integration constant additive to an asymptotic series of the same form as in ( [ 7 m ] ) except for an extra power of @xmath292 the easiest way to determine the `` integration constant '' @xmath401 is to fit it together with unknown coefficients defining the asymptotic series part to a sequence of @xmath393 .
this is a stable numerical procedure , although if @xmath29 is large , @xmath401 is sub - dominant to many much larger terms and it is essential to calculate in high precision .
for example , we might start the recursive calculation of @xmath393 as described in the previous section with @xmath328 digits , end with @xmath402 digits at @xmath403 and continue with this number of digits in a @xmath404 matrix inversion to obtain the @xmath401 of interest while discarding the remaining 70 coefficients of the asymptotic series ! the results are slightly more accurate than the final answers that are given in the appendix . for @xmath388
the presence of logarithmic factors on the right - hand side of ( [ 7p ] ) makes fitting to @xmath384 impractical .
instead we define subtracted functions @xmath405 which we use as replacements for @xmath384 in all fitting procedures . that the fitting functions @xmath406 will give the same @xmath383 is obvious since the added terms in ( [ 7q ] ) all contain logarithmic factors and are thus not @xmath1-independent .
that the @xmath406 are logarithm free follows from @xmath407 and an argument by induction starting from @xmath386 .
( [ 7r ] ) can be verified using the definition ( [ 7q ] ) and the subtraction eq .
( [ 7p ] ) .
to summarize , we obtain @xmath383 numerically by using ( [ 7q ] ) iteratively , starting with @xmath408 to generate @xmath409 using procedures similar to that described above for the initial @xmath410 which are just the @xmath411 the results are given in the appendix .
our arguments above are at best plausible and to provide a rigorous proof justifying our numerical procedure we believe one would have to do three things .
first , one would have to show the @xmath363 expansion has the form ( [ 7 g ] ) we deduced on numerical grounds .
second , one would have to show there exists a corresponding asymptotic expansion valid at large @xmath292 third , one must show both expansions have a sufficiently large domain of validity that the matching we assumed could in fact be carried out to arbitrarily high order . on the other hand , we have numerical evidence for the validity of our procedure as described in the next section .
the short - distance " terms we have calculated , when used in a series subtraction process , leave a residual in high order series coefficients that is consistent with the complete elimination of all `` short - distance '' terms , both singular and analytic , to the o@xmath412 we have worked . while the main intent of section [ sect : anal ] is actually quite different , it is highly unlikely that the cancellations necessary to yield the expansion of the scaling - amplitude function @xmath413 of eq .
( [ 9d ] ) in pure integer powers of @xmath79 would have occurred had there been an error , numerical or otherwise , in the results of this section .
we conclude this section with a toy model example to illustrate the possible convergence properties of the short - distance expansion .
the main impediment to extending the calculation described above is that we do not have simple analytical expressions for the @xmath79 expansion of the @xmath70 in general .
an exception is on the diagonal where we know , cf .
( [ 7d ] ) , that the coefficient of @xmath414 is @xmath415 and a simple expression for the asymptotic expansion of @xmath416 at large @xmath1 can be found in au - yang and perk @xcite .
we can define the diagonal partial sums @xmath417 and , as in the analysis described above , ask for the short - distance coefficient @xmath418 which is the @xmath1 independent part of @xmath419 in the limit @xmath420 this term @xmath418 could be viewed as a partial contribution to the short - distance terms of interest but since we do not know what cancellations will occur when we include all @xmath421 we prefer to consider it only as a toy result that is suggestive for the convergence of the short - distance terms with order @xmath422 the term @xmath418 is also equal to the @xmath423 independent part of @xmath424 in the limit @xmath425 this latter expression is more convenient since we can add to it any term such as @xmath426 for integer @xmath29 without contributing to any @xmath423 independent term . if we expand such terms in series in @xmath427 we obtain as an equivalent sum @xmath428 e^{-\epsilon n}.\ ] ] now choose the @xmath429 in ( 6.10 ) such that the divergent terms in the asymptotic @xmath44 expansion of @xmath430 cancel the divergent terms in the gamma functions .
then the @xmath44 sum becomes convergent even with @xmath431 with the cancellation extended to include also some slowly decaying terms in @xmath432 thus requiring @xmath433 one obtains the explicit formula @xmath434\ ] ] which is very convenient for numerical work .
we find the @xmath418 calculated from ( [ t11 ] ) for @xmath435 are @xmath436 the sequence in ( [ t12 ] ) clearly shows the asymptotic nature of the toy expansion .
furthermore the magnitude of the terms is in semi - quantitative agreement with @xmath437 from eq .
( [ eqcut ] ) .
thus while we can not conclude that this will be how the short - distance susceptibility terms @xmath438 in ( [ 6f ] ) will behave , it is at least encouraging to note that there is no evidence for any behavior more singular than that predicted by our natural boundary analysis .
any numerical analysis program to deduce functions such as the scaling - amplitude functions @xmath78 from their series expansions is in essence a fitting routine and always presupposes a knowledge of the analytic structure of the result .
although we have very little exact knowledge of @xmath78 we do know that the multiplier of @xmath63 in @xmath439 ( cf .
( [ n2 ] ) ) contains @xmath93 terms and there is numerical evidence @xcite that this is true of the @xmath440 for @xmath441 also .
on the other hand , gartenhaus and mccullough @xcite found that the series for @xmath60 were consistent with the absence of logarithmic corrections in @xmath442 through order @xmath443 and this was subsequently confirmed for @xmath444 as well @xcite . scaling arguments on the question of logarithmic corrections are necessarily inconclusive because of the lack of information on amplitudes , some of which may vanish .
we know of no scaling argument that either definitely requires the presence of logarithmic terms or can definitely exclude them .
the natural boundary arguments given in section [ sect : circle ] preclude the possibility that @xmath60 has a convergent rather than asymptotic expansion about @xmath183 , but we know of no analytical argument that shows whether this applies to both the `` scaling '' and `` short - distance '' terms or just to one or the other .
in the variable @xmath445 ( see appendix ) is indicative of a convergent series with radius of convergence @xmath446 .
this implies a conjugate pair of singularities at @xmath447 .
if however the series is asymptotic , this observation will fail to hold .
] neither could we distinguish the convergent from the asymptotic expansions from the data we have . although there are singularities in @xmath60 dense on the line @xmath448 , the singularities close to @xmath72 are extremely weak and we have not detected any singularity closer to the ferromagnetic @xmath72 point than that arising out of @xmath449 at @xmath450 thus for all practical ( numerical ) purposes we can ignore the possible asymptotic nature of the `` scaling '' terms .
the possibility that the @xmath78 might be expanded in a series in * integer * powers of @xmath79 can be confirmed numerically without evaluating any of the coefficients in the expansion .
the trick is to generate the series for the susceptibilities divided by the factor @xmath451 .
any term in @xmath442 or @xmath444 that is * not * a pure integer power law @xmath180 , @xmath452 , will necessarily contribute to the high order coefficients of the series .
of course , the leading two terms @xmath453 in @xmath78 , which occur as poles proportional to @xmath454 and @xmath366 in the scaled @xmath202 , will also contribute , but the amplitude of these terms is known and so their contribution can be subtracted . similarly , given the accurate `` short - distance '' amplitudes in the appendix , the contribution of all `` short - distance '' terms can be similarly eliminated through order @xmath455 we find that the remaining high order series coefficients are plausibly consistent with the `` short - distance '' @xmath456 that has not been subtracted and thus that there is * no numerical evidence * for any powers of @xmath79 other than pure integers in the scaling - amplitude functions @xmath78 .
the rest of this section gives details of the analysis that is the basis for this conclusion while the following section reports on our procedures for estimating the coefficients in the @xmath79 expansions of @xmath78 .
as outlined above , our search for possible terms other than those with pure integer powers of @xmath79 in the @xmath78 involves the observation of the high order series terms in the scaled and pole subtracted susceptibility functions @xmath457 @xmath458 applicable for @xmath74 and @xmath76 respectively . since the procedures for @xmath459 and @xmath76 are not different in principle , we will restrict the discussion below to the @xmath76 case only . to implement the `` short - distance '' subtraction we can restrict ourselves to determining @xmath460 , the contribution to the coefficient of @xmath461 arising from the @xmath462 singularity in @xmath463 factor here and in the following
is included for the convenience of allowing us to use the `` short - distance '' amplitudes exactly as tabulated in the appendix .
it also simplifies the final formula ( [ 715 ] ) . ]
the contribution from @xmath464 follows trivially from @xmath460 by @xmath101 repetitions of the derivative operation @xmath465 because of the simple form @xmath466 note also that the contribution from the @xmath467 singularity is identical except for an overall sign @xmath468 and thus simply requires that in the end we set all odd power amplitudes to zero and double the even ones . to determine @xmath460 is a standard exercise in complex variable contour integration
; we deform the contour in an @xmath469 integral with integrand @xmath470 to surround the branch - cut @xmath471 and in a final step set @xmath472 the result is @xmath473 and although the integral ( [ 715 ] ) can not in general be done in closed form , expansions valid asymptotically for large @xmath44 are easy to generate with computer algebra packages such as maple . because our series are particularly long these asymptotic expansions are essentially exact and are also very convenient for the subsequent @xmath474 differentiations .
it is also necessary to subtract and/or smooth out the contributions from complex singularities on the circle @xmath475 the only significant ones for the present calculation are those at @xmath476 and @xmath477 the subtractions of the leading contributions from the latter singularities arising from @xmath478 can be obtained directly from eq .
( 28 ) in @xcite .
the result is @xmath479\end{aligned}\ ] ] where @xmath480 is a pochhammer symbol .
we find that the further smoothing to give the remainder @xmath481 is more than adequate . here as in @xcite @xmath482 suppresses the contributions from @xmath476 while @xmath483 reduces what has not been subtracted by ( [ 716 ] ) .
note that the smoothing in eq .
( [ 717 ] ) scales up the amplitude contributions from the singularities at @xmath484 by a factor @xmath485 we find that as we include higher and higher orders of the `` short - distance '' terms in the subtraction process the remainder amplitudes ( [ 717 ] ) decrease in a smooth fashion .
the amplitude @xmath486 we obtain after having subtracted all terms through @xmath487 listed in the appendix is , at @xmath488 , @xmath489 compared to @xmath490 one gets without pole subtraction in ( [ 8a ] ) .
the residual is very reasonably the amplitude we would expect from the @xmath456 terms and this has been confirmed by extending the `` short - distance '' amplitude sequence in the appendix by two terms using a crude pad analysis .
the result of this additional subtraction is to reduce @xmath486 at @xmath488 to @xmath491 analysis of the high temperature series leads to a similar conclusion and also that the anti - ferromagnetic `` short - distance '' terms in the appendix represent the susceptibility at this point completely with nothing left out .
although our analysis does not `` prove '' the absence of powers of @xmath79 other than pure integers , we can put very stringent bounds on the amplitudes of any possible singular terms . to make this comparison concrete , we suppose either of @xmath78 contains the singular term @xmath492 and for simplicity assume @xmath181 integer
. the contribution of this term to the amplitude of the coefficient of @xmath493 for @xmath74 or @xmath494 for @xmath76 in the high / low temperature series expansion of ( [ 8a ] ) will be about @xmath495 relative to the pole contribution and this is to be compared to the observed amplitude we obtain after subtracting or smoothing away the known singularities as best we can .
our current results are the amplitude bounds @xmath496 for all @xmath181 less than about 15 .
the task of determining @xmath78 numerically is enormously simplified by the _ a priori _ knowledge strictly speaking a conjecture based on the numerical work of the last section that @xmath78 has an expansion in integral powers of @xmath79 near @xmath497 the absence of any confluent terms means that many different analyses can be used efficiently and we report here on two independent calculations that give essentially identical results , thus again confirming the above conjecture .
we have not seriously attempted to optimize our analysis to give the most accurate numerical values possible and thus if it ever becomes necessary one could almost certainly improve on our coefficients as given in the appendix .
one particular feature of the @xmath78 expansion is worth noting here .
when the functions are scaled by @xmath299 the resulting series are numerically consistent with series in even powers of @xmath79 only .
we have also noted a similar simplifying role played by @xmath299 in the `` short - distance '' terms which ultimately trace back to the toeplitz determinant giving the diagonal correlations @xmath254 and to the quadratic recursion relations for general @xmath70 .
was used to simplify the scaling limit . ]
this result is not entirely unexpected .
the singular part of the free energy in zero field is an even function of @xmath79 and the magnetization @xmath498 is @xmath499 times an even function of @xmath287 nonlinear scaling field analysis then predicts , in the absence of corrections , that the `` scaling '' part of the susceptibility is @xmath500 times an even function of @xmath79 .
thus although our results for @xmath78 are not consistent with the complete absence of correction terms as discussed in section 7 the prediction that @xmath501 is even in @xmath79 does appear to be preserved to all orders . to determine the coefficients in @xmath78 we return to the unscaled @xmath202 of eqs .
( [ 1a],[1b ] ) so that the terms in @xmath78 are now singularities of the function and contribute to the high order coefficients in the series .
the cases @xmath76 and @xmath74 are again similar ; for @xmath76 the unwanted contributions from the `` short - distance '' part of @xmath60 are subtracted as in the @xmath161-plane analysis described in section [ sect : anal ] .
we use essentially the same smoothing except for a @xmath502 shift in power necessitated by the difference in the singularity structure generated by the @xmath60 rescaling of section [ sect : anal ] .
that is , we replace the remainder eq .
( [ 717 ] ) by @xmath503 and reduce the remainders @xmath486 in ( [ 719 ] ) by a least squares fitting to the unknown @xmath504 coefficients in a procedure similar to that described in @xcite .
fitting intervals @xmath505 are typically used , and an fft of the residuals is very useful as a diagnostic to interpret the observed oscillations in the residuals in terms of @xmath60 singularities on the circle @xmath506 . because the highest order terms in @xmath504 are not fixed ( i.e. known ) unlike the `` short - distance '' terms , they tend to float and become effective amplitudes that incorporate all the higher order effects including the `` short - distance '' contributions that have not been subtracted .
one technical result of this is that the residuals we observe in ( [ 719 ] ) are some six orders of magnitude smaller than the residuals we obtained in ( [ 717 ] ) . in part
this means that whereas ( [ 717 ] ) was more than adequate as a smoothing operation , ( [ 719 ] ) is marginally so .
in addition , we observe here for the first time one of the singularities from @xmath507 and using the cut information given in eqs .
( [ new27 ] , [ new28 ] ) can subtract it as @xmath508 where @xmath509 .
a similar analysis has been carried out for @xmath74 and our estimates for the coefficients of @xmath78 are given in the appendix .
the coefficients through o@xmath510 are unambiguously rational and have been fixed in the final fittings .
we have also set to zero all coefficients of odd powers of @xmath79 in the product @xmath501 to o@xmath511 but have allowed variable coefficients of @xmath512 @xmath513 and @xmath514 these latter coefficients are , as expected , sensitive to whether we stop the short - distance " subtraction at o@xmath412 or add in the additional terms we have estimated by pad methods .
the terms quoted in the appendix on the other hand are completely stable and thus we believe reliable except possibly for the last digit . when we relax the constraint of zero amplitude on individual odd @xmath391 terms in @xmath501 for integer @xmath515 we find no significant improvement in our fits and the resulting amplitudes are consistent with zero . for example , when @xmath76 , we find best fit coefficients of @xmath391 , @xmath29 odd , that in absolute magnitude are all less than @xmath516 .
for @xmath74 the corresponding bounds can be as much as @xmath517 times larger .
but in both cases these bounds are of a magnitude similar to what we estimate is the uncertainty in the even order coefficients given in the appendix .
thus we believe @xmath501 even in @xmath79 to be an exact symmetry of the scaling - amplitude function .
an alternative analysis using the traditional variable @xmath518 was also carried out .
the natural boundary singularities at @xmath54 are mapped to two circles @xmath519 . in this expansion variable ,
the ferromagnetic and anti - ferromagnetic critical points are at @xmath520 respectively , and all other points on the two circles are farther away from the origin .
hence the amplitudes of any other singularities are exponentially damped and may be neglected in the analysis .
the two analyses are in complete agreement , but the more detailed @xmath161-plane analysis provides greater precision . for contemplated future analyses on other lattices , for which less is known about the natural boundary singularities , it may be necessary to use the @xmath51-plane analysis .
the basic scaling ansatz for the singular part of the free - energy of the two - dimensional ising model is @xmath521 here @xmath522 are nonlinear scaling fields associated with the thermal field @xmath523 the magnetic field @xmath524 and the irrelevant fields @xmath525 the exponents @xmath526 are the thermal and magnetic exponents , and @xmath527 are the irrelevant exponents . and @xmath528 .
this scaling ansatz assumes only a single resonance , between the identity and the energy . that the dimensions are integers implies higher powers of @xmath529 .
the surprise is the power of @xmath530 at which higher powers of @xmath529 enter .
we thank andrea pelissetto for this clarification . ] .
the nonlinear scaling fields have expansions @xmath531 where @xmath532 are smooth functions of @xmath530 and @xmath533 if irrelevant fields are neglected , then the known zero field free energy forces the equalities @xmath534 and @xmath535 .
furthermore , the absence of logarithmic terms in the known magnetization and the divergent part of the susceptibility requires the derivatives @xmath536 and @xmath537 to vanish .
aharony and fisher have conjectured @xcite , almost certainly correctly , that there are no logarithms multiplying the leading power law divergence of all higher order field derivatives , in which case the @xmath538 are constants and analyticity on the critical isotherm for @xmath539 demands @xmath540 with all these constraints built in , one can show the scaling ansatz ( [ s1 ] ) together with the field expansions ( [ s2 ] ) lead to @xmath541 where @xmath542 @xmath543 @xmath544 and @xmath545 are constants and @xmath546 @xmath547 the free energy and magnetization eqs .
( [ s3 ] ) determine the scaling field coefficients @xmath548 and @xmath549 in ( [ s2 ] ) .
the presence of irrelevant scaling fields will be expected to manifest themselves as deviations from the predicted form of the susceptibility in ( 7.4 ) and/or as deviations from the unique prediction for the coefficient of @xmath550 working as usual in the temperature variable @xmath551 , we write the predicted isotropic susceptibility from ( 7.4 ) as @xmath552 where @xmath87 has already been given in eq .
( [ 9c ] ) .
the short - distance " contribution to @xmath60 is here predicted to be given by the sum of the term containing @xmath553 which arises as the mixing of the first two terms in the expansion of @xmath554 in ( [ s2 ] ) , and the analytic @xmath555 the clear implication of eq .
( [ 9d ] ) , which shows the exact @xmath78 is not equal to @xmath87 is that irrelevant variables do play a role , and the multipliers @xmath413 represent the contribution of a number of scaling fields . while there are suggestions in the literature for what these scaling fields might be is absent for the pure @xmath556 ising model susceptibility considered here . a mechanism in terms of generators of the energy family of the virasoro algebra is adduced by caselle et al
@xcite which gives rise to corrections at order @xmath557 as observed .
] it is unlikely that a unique identification could be made here since we are dealing with a single isolated model with no free parameters to vary .
a corresponding analysis of the anisotropic square , and the triangular and hexagonal lattices is likely to be enlightening in this regard . in @xcite
a study of difference equations is given , which implicitly outlines what is needed to obtain the difference equations for the hexagonal and triangular lattices .
additional material in this respect can also be found in @xcite .
note however that two different effects manifest themselves . at fourth
order in @xmath530 ( or , equivalently , @xmath79 ) scaling under the assumption of only two nonlinear scaling fields breaks down , as evidenced by the difference between the coefficients of @xmath557 in eqs .
( [ 9d ] ) and ( [ 9c ] ) .
however the corresponding high- and low - temperature amplitudes still satisfy @xmath558 for all @xmath559 for @xmath560 not only does simple scaling fail to hold , but this equality also breaks down . in the vicinity of the anti - ferromagnetic point in the high - temperature phase , @xmath60 is given exclusively by a `` short - distance '' term ( [ 6e ] ) . both @xmath99 and @xmath97 from eqs .
( [ 6e ] ) and ( [ 9b ] ) have expansions of the same form ( [ 6f ] ) , where the sum over @xmath181 is restricted to @xmath561 the coefficients in this expansion can be determined from the short - distance correlations , and accurate values for the expansions of @xmath438 are given in the appendix through o@xmath69 again we note that there are terms in these `` short - distance '' functions that are not of the aharony and fisher @xcite predicted form ( [ 7.5 ] ) based on the absence of irrelevant variables . we conclude with some speculative remarks .
we can not account physically for terms of order @xmath562 with @xmath563 and @xmath564 in @xmath565 , though we can see their origin mathematically , as discussed below eq .
( [ 7b ] ) and in footnote 12 . while terms without logarithms and terms of order @xmath566 are expected , it is surprising ( to us ) that higher powers of @xmath567 enter at the orders they do . from conformal field theory
, we have predictions for the irrelevant exponents @xmath568 the fact that @xmath569 for @xmath570 though @xmath571 is not equal to the corresponding term in ( [ 9c ] ) for @xmath572 suggests the presence of only a * single * irrelevant operator contributing at order @xmath573 while the breakdown of high - low temperature symmetry in @xmath78 at o@xmath574 suggests that more than one scaling operator couples to the lattice magnetization at this order . a corresponding study to that reported here on the triangular and honeycomb lattices , as well as on
the anisotropic square lattice is likely to be enlightening , and we hope to report on this in future . * acknowledgments * we would like to thank larry glasser , mireille bousquet - mlou and john cardy for many helpful discussions and calculations , andrea pelissetto for adding to our understanding of scaling theory and the provision of a preprint , barry mccoy , alan sokal and michael fisher for invaluable comments on earlier versions of the manuscript , and douglas abraham for useful comments on the history of the ising model .
ajg and wo would like to thank the australian research council for financial support , and jhhp thanks the nsf for support in part by nsf grant no .
phy 97 - 22159 .
here we list the expansion coefficients @xmath575 of the short - distance " functions @xmath438 defined in ( [ 6f ] ) to o@xmath412 .
the prefactor @xmath576 is understood to be expanded in a series in @xmath287 the leading constant and coefficient of @xmath577 have been reported previously @xcite .
the @xmath78 series as deduced in section [ sect : ca ] are also given .
@xmath578 \\ & - 0.002067088393167114141650194740281136875636\tau^{7 } \\ & + 0.05424693070421409615112542698595864778919\tau^{8 } \\ & + 0.0010601025315498815900774416057541651837\tau^{9 } \\ & - 0.049300253157082567741316861339709144063\tau^{10
} \\ & - 0.00026830064161204203467706137637400358\tau^{11 } \\ & + 0.0450270525719569186212816103126356308\tau^{12 } \\
& - 0.00034332683257234543036792535081332\tau^{13}\\ & - 0.041428586463052869356803144137620\tau^{14})\\ & + ( \log|\tau|)^2 \\ & ( 0.0093915698711458721317953318727075770649513654\tau^4 \\ & - 0.00869592546287923802156416645191752987912922\tau^6 \\ & + 0.007669481493104540876445085447422616885330\tau^{8 } \\ & + 0.00015428438297902275440225213783285077606\tau^{9 } \\ & - 0.0068054076881441249098452112921129773269\tau^{10 } \\ & - 0.000310520937481414524040686012525223279\tau^{11 } \\ & + 0.00611386643219454473116391019937140965\tau^{12 } \\ & + 0.000444606198235804033861443998682830\tau^{13}\\ & - 0.0055571002151161308034896964314679\tau^{14})\\ & + ( \log|\tau|)^3 \\ & ( - 0.000015771569138451840480001012621461738178\tau^{9 } \\ & + 0.0000344282066208887553647799856857753380\tau^{11}\\ & - 0.0000524427177487226174161583779149393\tau^{13})],\end{aligned}\ ] ] @xmath579 \\ & - 0.011741188869656263932121387296300743594029390\tau^4 \\ & - 0.01406604087566590060620992322775625815515533\tau^5 \\ & + 0.0131064546156258402249424759665220798848681\tau^6 \\ & + 0.012239696625538370626786005459530159711716\tau^{7 } \\ & - 0.01184019404541084813958321002995560636877\tau^{8 } \\ & - 0.0105854093023116661362507232392645147231\tau^{9 } \\ & + 0.010151560037724359473553197335905170854\tau^{10 } \\ & + 0.00908000411233119371610549453140718281\tau^{11 } \\ & - 0.0085420122287896456879459087815054030\tau^{12 } \\ & - 0.00771702694013238358077176900242074\tau^{13 } \\ & + 0.007123677682511208149032476379667\tau^{14}\\ & + \log|\tau| \\ & ( - 0.1553171901580110585934133538932734529992121600305\tau \\ & + 0.03206714814586975221843437287457551882247161782\tau^3 \\ & - 0.0077168875724615093064542922101962689299768599\tau^4 \\ & - 0.015675211573817078943430169665269657287132910\tau^5 \\ & - 0.00028554245153720354627897919087710530890677\tau^6 \\ & + 0.0096072545027321808179041903535130201897217\tau^{7 } \\ & + 0.004835406420625092236673413378307375358908\tau^{8 } \\ & - 0.00606499034448050751379194815071149626812\tau^{9 } \\ & - 0.0073400150414474023562454611060746360875\tau^{10 } \\ & + 0.003910356521403913091050321141297009252\tau^{11 } \\ & + 0.00870842744568158003036434762719697635\tau^{12 } \\ & - 0.002697783010884752101384006121375890\tau^{13 } \\ & - 0.0094056230380765607719474925088649\tau^{14})\\ & + ( \log|\tau|)^2 \\ & ( 0.01153371437882328027949011442761203640684043805\tau^4 \\ & - 0.011311734920691560067535056532207842716405684\tau^6 \\ \displaybreak[4 ] \\ & + 0.0100457687111988577404299867962466051265974\tau^{8 } \\ & - 0.000475698571097159420906450182271928179428\tau^{9 } \\ & - 0.00878397202228689639470985683437717938463\tau^{10 } \\ & + 0.0011571801729636538264100914359800686355\tau^{11 } \\ & + 0.007680651109512704070606639646988801296\tau^{12 } \\ & - 0.0018650912616201532939412831153215046\tau^{13 } \\ & - 0.00674470189451526288478200059343432\tau^{14 } ) \\ & + ( \log|\tau|)^3 \\ & ( 0.0000578997194764877297760067221144062249541\tau^{9 } \\ & - 0.00016991508824012890240796446744935908812\tau^{11}\\ & + 0.00032664884687465587957270016883093909\tau^{13})]\end{aligned}\ ] ] | we have made substantial advances in elucidating the properties of the susceptibility of the square lattice ising model .
we discuss its analyticity properties , certain closed form expressions for subsets of the coefficients , and give an algorithm of complexity o@xmath0 to determine its first @xmath1 coefficients . as a result ,
we have generated and analyzed series with more than 300 terms in both the high- and low - temperature regime .
we quantify the effect of irrelevant variables to the scaling - amplitude functions .
in particular , we find and quantify the breakdown of simple scaling , in the absence of irrelevant scaling fields , arising first at order @xmath2 , though high - low temperature symmetry is still preserved . at terms of order @xmath3 and beyond
, this symmetry is no longer present .
the short - distance terms are shown to have the form @xmath4 with @xmath5 .
conjectured exact expressions for some correlation functions and series coefficients in terms of elliptic theta functions also foreshadow future developments . |
depending on the mean front velocity , between @xmath158 and @xmath159 photos are recorded for each choice of parameters . to enhance the statistics in the transient propagation regime , for @xmath160 , we performed between @xmath161 and @xmath162 different realizations for each value of @xmath10 .
a total of 30 different values of @xmath10 have been investigated , spanning from @xmath163 to @xmath164 . , upward front propagation with supportive flow .
center : @xmath165 : relaxation pattern in the absence of flow .
right : @xmath166 , backward propagating front in adverse flow.,width=604 ] example of the extracted fronts are given on fig .
[ fronts ] . in the absence of the flow ( @xmath165 )
, the reaction fronts are uniformly moving through the glass beads with a constant velocity @xmath79 m/s and @xmath167 , such that initially rough fronts gradually relax toward a smooth front shape [ see movie 1 ] .
for a high flow strength ( @xmath168 ) , the fronts become rough and present self - affine structures [ see movie 2 and 3 ] .
the heterogeneous permeability field @xmath169 is generated according to a log - normal distribution , which is gaussian correlated over a length @xmath170 , as commonly assumed to model groundwater flow @xcite .
the flow is computed by solving the darcy - brinkman equation @xmath171 where @xmath172 is the fluid viscosity and @xmath173 is the body force driving the flow ( _ e.g. _ the buoyancy ) .
the boundary conditions are periodic in the lateral direction .
a convection - diffusion - reaction equation is then solved for the concentration @xmath174 of the auto - catalytic reactant ( iodide ) , normalized by the initial concentration of iodate , @xmath175 here @xmath1 and @xmath2 are the molecular diffusion and the reaction rate .
both equations are solved using lattice boltzmann schemes ( see @xcite for details ) . in the numerical study
, the parameters @xmath2 and @xmath1 have been chosen to keep constant the ratio @xmath176 , and the standard deviation of the log - normal distribution has been kept constant @xmath177 .
as in the experiments , the control parameter @xmath10 is determined from the relative velocity between the reaction velocity @xmath178 and the mean - flow velocity @xmath73 . the grid size is @xmath115 and the computational cost per realisation is around @xmath179 hours using @xmath180 cpus on a standard workstation
. for each value of @xmath10 , around four different realizations were performed , and over 800 fronts are computed for each simulation .
a total of hundred different values of @xmath10 have been numerically generated , with thirty points close to the transition point at @xmath95 .
rough interface fluctuations are generally characterized by the standard deviation of their height over a length scale @xmath148 , defining a roughness exponent @xmath23 . in the presence of `` normal scaling '' ( see @xcite and section [ s : sawtooth ] below for exceptions ) one has the choice between different methods . while we used several ones , a common choice , and
the one employed in the data presented in the present work is to measure @xmath181 ^
2\right>_{\delta x}}\right>_x\ , \ ] ] where @xmath182 correspond to an average over the interval @xmath148 and @xmath183 is an average over the whole interface . in the context of kinetic roughening , an initially flat front develops power - law distributed structures as a function of time @xmath70 , which in the absence of a length scale in the system implies the family - vicsek scaling @xcite @xmath184 the exponents @xmath23 and @xmath24 are the roughness and growth exponents .
the exponent @xmath185 is the dynamic exponent and characterizes the lateral correlation length growth @xmath186 for short time , before the front fluctuations reach the steady regime . in this scenario
, one can measure the growth exponent @xmath24 by determining the overall front width @xmath187 ^ 2}\sim t^\beta$ ] .
this scaling requires that the initial front starts as a straight line .
although this has been done in some experiments @xcite , it is generally hard to achieve experimentally : as the reaction is propagating through the glass beads , the initial fronts usually acquire a non - trivial shape as illustrated on fig [ fronts ] .
instead , one can use the front fluctuations in the steady state to determine the scaling exponents @xcite .
this is achieved by determining the local width for both spatial and temporal fluctuations in the stationary regime @xmath188 ^ 2\right>_{\delta x}}\right>_{x}\sim \delta x
^\alpha\;\text{,}\\ w(\delta t , x ) & = & \left<\sqrt{\left < [ h(x , t)-\left < h(t)\right>_{x}]^2\right>_{\delta
t}}\right>_{t}\sim \delta t ^\beta\;\text{.}\end{aligned}\ ] ] in order to improve the statistics , an additional average can be made in the stationary regime over a time window and space window respectively to determine @xmath84 and @xmath30 , @xmath189 alternatively , the fronts can be characterised by the 2-point correlation functions in the steady state @xmath190 ^ 2 \right>_{x , t } \sim \delta x^{2\alpha}\\ & & \left < \left[h(x , t+\delta t)-h(x , t ) -v \delta t \right>]^2 \right>_x \sim \delta t^{2\beta } \\ & & v \delta t : = \left
< h(x , t+\delta t)-h(x , t ) \right > \label{set2'}\ .\end{aligned}\ ] ] we found no significant differences in the estimates of the critical exponents when using either eqs .
( [ set1])-([set1 ] ) or ( [ set2])-([set2 ] ) .
the front displays static sawtooth shapes at large scales when @xmath10 becomes close to @xmath112 , in the negative qkpz regime , leading to distinct scaling properties of the fronts before and after formation of these large scale sawtooth shapes .
once the final static state is reached , the front exhibits a larger roughness than the one determined during the transient propagation , or excluding these inclined static regions . in the pinned phase with negative @xmath10 ,
the front exhibits a higher roughness exponent once the final static state is reached .
[ wl ] shows the corresponding @xmath84 functions : left , determined from the sawtooth shape fronts ; center , during the transient propagation , excluding the inclined static regions .
the temporal height fluctuations determined during the transient regime for the same values of @xmath10 are displayed on fig .
[ wt ] left end center respectively .
when @xmath10 becomes less negative , i.e. @xmath191 , both the sawtooth size and inclination become smaller , and no more distinction is necessary between the scalings of the static state and the transiently moving state . for the positive qkpz regime
, one can note on figs .
[ wl ] ( right ) and [ wt ] ( right ) that the thermal kpz exponents are recovered for @xmath192 . for higher value of @xmath10 , when @xmath193 , the mean flow velocity vanishes and the noise is too weak to generate height fluctuations .
this area corresponds to the hatched area in fig . 4 in the manuscript .
as a consequence , the temporal fluctuations become uncorrelated : @xmath194 , and the roughness of the front stays constant at @xmath195 .
+ as in the experiments , measuring a roughness exponent of the static front is more complex , and the same distinction between static fronts and moving fronts is necessary . as indicated on fig .
[ wlsimu2 ] , when @xmath196 , the roughness of the static front @xmath197 becomes close to one indicating the characteristic sawtooth shape .
[ wlsimu2 ] also shows the roughness of the front close to @xmath125 , with the value @xmath198 in agreement with the positive qkpz equation . in this regime , it is interesting to note that there is a discrepancy between the roughness measured in the moving regime ( @xmath199 ) and the one measured in the static regime , which should be compared to directed percolation @xcite . far from the two transitions when @xmath200 or @xmath201 , the two exponents are close to the `` thermal '' kpz ones .
indeed at @xmath202 , @xmath203 and @xmath204 and at @xmath205 , one has @xmath206 and @xmath204 .
as @xmath207 is decreased further on fig . 5 in the manuscript , the roughness of the numerical fronts drops to zero at @xmath208 and then increases again as @xmath10 approaches the second transition at @xmath121 .
we can note here that , deep inside the pinned phase , we expect the final front shape to depend on the initial condition .
both in the experiments and the numerics , when @xmath209 , the front moves only by a small amount before getting pinned .
in fact , it is worth noting that the poisson point disorder models ( first layer png as in @xcite or in a more realistic form in @xcite ) predict that the final pinned configuration has the same roughness as the initial one .
hence in the numerics it is expected that the roughness drops to zero as the initial condition is flat , whereas in the experiments the initial fronts are never flat as they have already traveled through the glass beads before the flow is switched on . when we get closer to @xmath121 , the front moves a longer distance before getting pinned , and thus one recovers a more universal behavior as @xmath196 .
as a control parameter , we define @xmath210 in the experiments and @xmath211 in the simulations .
one can note that the value @xmath78 is quite singular , since then the noise vanishes but not the driving force . in the porous medium
, it is quite unclear if one should use @xmath135 or @xmath212 m/s ( _ i.e. _ including or not the effect of tortuosity / porosity ) to determine @xmath10 . here
, we are using @xmath178 to define @xmath10 as suggested by the eikonal approximation .
indeed , the parameter @xmath178 was shown to lead to the right estimation of the sawtooths angle when the fronts are static @xcite . in the simulations , porosity and tortuosity are both equal to one , and @xmath213 . by analogy with the qkpz behavior , the constant @xmath214
is determined by fitting the right velocity branch far from its pinning threshold ( see figs .
[ vf_zoom ] ) .
as described below , we attribute this constant value to the average roughness of the front . a closer inspection of the fig . [ vf_zoom ]
right , shows that it can be fitted with a power - law @xmath123 , with @xmath215 as shown on fig .
[ vf_ffc ] .
this exponent is slightly larger than the theoretically predicted value @xmath124 @xcite .
one can note from fig .
[ vf_zoom ] left , that this exponent is hard to determine experimentally as the number of available data points close to @xmath95 is too small .
the reason for this limitation lies in experimental uncertainties , constraining the number of distinct @xmath10 to approximately 4 in this region . in the numerics , below the first transition , @xmath216 , and beyond the second one , @xmath217 ,
the velocity @xmath120 is almost linear with @xmath218 and @xmath219 , respectively .
both branches have different slopes and differ from the prediction of the kpz equation ( @xmath220 at large @xmath22 ) .
similarly , in the experiments , we note that both for negative @xmath221 and for positive @xmath222 , the linear branches have a different slope than one : @xmath223 and @xmath224 respectively .
in addition , for the supportive flow regime in the experiments , this slope becomes @xmath225 .
we attribute this behavior to a nonlinear interaction between the chemical reaction and the flow field , which enhances mixing and front velocity as studied in @xcite . indeed , since flow heterogeneities necessarily enhance the mixing and thus the reaction rate , one can expect at first order in @xmath226 to have an effective chemical velocity @xmath227 leading to an average front velocity @xmath228 for adverse flow ( @xmath229 ) , and @xmath230 for supportive flow ( @xmath231 ) .
for a thin front , the local front velocity follows the eikonal equation : @xmath232 where @xmath131 is the normal of the front , @xmath132 the curvature and @xmath133 the local flow velocity .
indeed , this equation is similar to the `` flux - line model '' of kardar @xcite where the chemical velocity plays the role of the lorentz force and the disordered flow of the random force . after projection
the eikonal approximation yields @xmath233 \;\text { , } \\ & & { \rm with } \;\ ; s=\nabla h \;\ ; { \rm and } \;\
; \vec u(\vec r ) = \bar u \vec e_y + \delta \vec u(\vec r)\;\text{.}\end{aligned}\ ] ] considering that the flow is highly anisotropic , @xmath234 , this gives @xmath235 finally , assuming small gradients and neglecting higher - order terms , this leads to the qkpz equation , @xmath236 after normalizing by @xmath135 , we can write the qkpz equation with the following parameters : @xmath237 near the transition at @xmath112 , the small - gradient limit may not be quantitatively accurate in this region as the sawtooth - pattern slope becomes large . in this regime , since pinning occurs where the flow field is slower than the chemical reaction , the transition results from the competition of two effects : ( i ) the probability to find a region of low flow velocity decreases as @xmath238 , ( ii ) when the sawtooths " get sharper , the probability to find new pinning points increases . based on extreme - value statistics and the png model @xcite ,
a more precise scenario has been proposed in @xcite .
in the experiment @xmath91 and @xmath93 is proportional to @xmath10 for large mean flow velocities , i.e. @xmath239 .
the expected kpz noise @xmath240 becomes then almost independent of @xmath10 .
@xmath241 can be deduced experimentally from the local flow - velocity fluctuations . the width @xmath242 of the corresponding flow - velocity pdf , at a given value of @xmath10 ,
can be estimated with tracer transport experiments @xcite .
[ sigma ] a ) shows the value of @xmath243 determined from tracer front dispersion experiments ; it shows that @xmath244 is proportional to @xmath10 for both signs of @xmath10 and @xmath245 .
the dependence of @xmath246 with @xmath10 is shown on fig .
[ sigma ] b ) .
one can note that @xmath247 for @xmath248 , with a different value depending on the the sign of @xmath10 .
upward propagating reaction fronts at large flow rate , for positive mean flow orientation at @xmath250 . in this configuration both the chemical wave velocity and the mean flow velocity are oriented towards the top . +
* http://www.fast.u-psud.fr/~atis/movie2_q025_up.avi * + downward propagating reaction fronts at large flow rate , for negative mean flow orientation at @xmath251 . in this configuration
the chemical wave velocity is oriented toward the top and the mean flow velocity toward the bottom .
+ * http://www.fast.u-psud.fr/~atis/movie3_q08_down.avi * + upward propagating reaction front at low flow rate , for negative mean flow orientation at @xmath252 . in this configuration
the chemical wave velocity is oriented towards the top and the mean flow velocity towards the bottom .
+ * http://www.fast.u-psud.fr/~atis/movie4_q011_down.avi
* + zero velocity reaction fronts for negative mean - flow orientation at @xmath253 .
the chemical wave velocity is oriented towards the top and the mean - flow velocity towards the bottom .
the front propagates at the chemical reaction velocity and stops when the flow is turned on .
+ * http://www.fast.u-psud.fr/~atis/movie5_q015_down.avi * + downward propagating reaction fronts at low flow rate , for negative mean flow orientation at @xmath254 . in this configuration
the chemical - wave velocity is oriented towards the top and the mean - flow velocity towards the bottom . *
http://www.fast.u-psud.fr/~atis/movie5_20_down_ssb.avi * + | self - sustained reaction fronts in a disordered medium subject to an external flow display self - affine roughening , pinning and depinning transitions .
we measure spatial and temporal fluctuations of the front in @xmath0 dimensions , controlled by a single parameter , the mean flow velocity .
three distinct universality classes are observed , consistent with the kardar - parisi - zhang ( kpz ) class for fast advancing or receding fronts , the quenched kpz class ( positive - qkpz ) when the mean flow approximately cancels the reaction rate , and the negative - qkpz class for slowly receding fronts .
both quenched kpz classes exhibit distinct depinning transitions , in agreement with the theory .
growing interfaces are ubiquitous in nature , appearing in situations as different as bacterial colonies @xcite , solidification @xcite , atomic layer deposition @xcite , liquid interfaces in porous media @xcite or crack propagation in heterogeneous materials @xcite .
the formation of scale - free structures in these systems raises the important question of universality in out - of - equilibrium phenomena . in this letter
, we consider the propagation of a reaction front inside a porous medium .
resulting from the balance between the molecular diffusion @xmath1 and the reaction rate @xmath2 , autocatalytic reactions can develop a traveling front . in the absence of an externally imposed flow
, the reaction front develops into a flat horizontal front , propagating with a constant velocity @xmath3 , and a stationary concentration profile of width @xmath4 .
when coupled with the heterogeneous flow field of the porous medium , the fronts become rough , and modify their behavior accordingly with the flow strength and mean orientation relative to the chemical reaction direction .
they propagate either downstream or upstream , or can remain frozen over a range of counter - flow rates , delimited by two distinct depinning transitions @xcite . until now ,
however , their universality classes have not been identified . using both experimental and numerical approaches
, we investigate their spatial and temporal scaling over the whole range of the externally imposed flow . in the vicinity of both depinning points ,
these reaction fronts display transient static configurations with distinct morphologies depending on the front propagation direction , displayed on fig .
[ field ] .
we show that this is a well - controlled system which encompasses several universality classes .
two important classes predicted by the theory , and discussed below , are : + ( i ) non - linear stochastic growth governed by the ( thermal ) kardar - parisi - zhang ( kpz ) equation ( [ kpzstoch ] ) , + ( ii ) growth where both the non - linearity and quenched disorder are present , described by the quenched kpz ( qkpz ) equation ( [ qkpz ] ) .
it divides into two subclasses , positive qkpz and negative qkpz , depending on the sign of the non - linearity @xmath5 . .
top : upward propagating front near @xmath6 ( @xmath7 ) .
bottom : backward propagating front near @xmath8 ( @xmath9).,width=325 ] it has been difficult to find unambiguous experimental realizations , due to long - range effects , quenched disorder , and a mixing of ( i ) and ( ii ) @xcite .
recently , experiments on turbulent liquid crystals @xcite made a precise connection with the theory of the kpz class experiencing a revolution of its own @xcite . the ( positive ) qkpz class @xcite and both kpz and qkpz classes in evaporating colloidal suspensions @xcite were observed .
remarkably , in the present system , by tuning a _ single _
parameter @xmath10 , one can observe _ all three classes_. versus the applied force @xmath10 ( resp .
@xmath11 ) , in adverse flow a ) experiments ( black dots with error bars ) , b ) numerics .
dashed lines are linear extrapolation of the advancing branch .
to put all data on one plot , axes are rescaled according to @xmath12 , @xmath13 .
insert : front velocity versus @xmath14 .
the continuous line corresponds to @xmath15.,width=321 ] the kpz equation @xcite was proposed as a generic model for an interface growing along its local normal,@xmath16 ^ 2 + \eta(x , t ) + f\ ; .\ ] ] its height @xmath17 is along the vertical axis , @xmath18 an effective stiffness due to diffusion , @xmath5 the non - linearity , and @xmath19 a gaussian white noise with @xmath20 and @xmath21 .
@xmath22 is an applied force , and up to a shift , proportional to the experimental applied force @xmath10 as shown below .
the surface can be characterized by two scaling exponents , the roughness @xmath23 , and the growth exponent @xmath24 , defined via @xmath25 ^ 2}\sim |x - x'|^{2\alpha}$ ] and @xmath26 ^ 2}\sim|t - t'|^{2\beta}$ ] . in @xmath27 dimensions , @xmath28 and @xmath29 @xcite . and right column : temporal fluctuations @xmath30 .
( a ) and ( b ) for @xmath31 ; ( c ) and ( d ) for @xmath32.,width=325 ] in a heterogeneous medium , the noise " acquires a static _ quenched _ component , described by the qkpz equation @xcite : @xmath33 ^ 2 + \bar \eta\big(x , h(x , t)\big ) + f\;.\ ] ] the case @xmath34 models a number of systems , and is a distinct universality class ( quenched edwards - wilkinson ) @xcite but does not seem to be relevant here ( it predicts @xmath35 and @xmath36 . ) in the kpz eq .
( [ kpzstoch ] ) one can eliminate the driving by setting @xmath37 . changing
then @xmath38 reverses the sign of the non - linear term @xmath5 , which is thus unimportant .
by contrast , the qkpz eq .
( [ qkpz ] ) does not allow for this change , since the term @xmath39 is not invariant .
the driving force @xmath22 is thus a new parameter of the problem , and its sign ( relative to the sign of @xmath5 ) matters . if the disorder is statistically invariant by parity , i.e. if @xmath40 has the same properties as @xmath41 , eq . ( [ qkpz ] ) is invariant under @xmath42 , @xmath43 , and @xmath44 , leading to two cases : positive qkpz when @xmath5 and @xmath22 have the same sign , and negative qkpz when they have opposite signs . in the moving phase , qkpz of
either sign crosses over to kpz at large scales .
this is seen e.g. in the limit of large mean interface velocity @xmath45 : consider eq .
( [ qkpz ] ) with white noise @xmath46 and perform the change @xmath47 .
the disorder then becomes @xmath48 i.e. the same noise as in the kpz eq .
( [ kpzstoch ] ) , identifying @xmath49 . as @xmath50 is decreased , the crossover from qkpz at short scales to kpz occurs at larger and larger scales .
the positive qkpz equation exhibits a depinning transition @xcite , well characterized in @xmath27 .
the mean interface velocity vanishes for @xmath51 ; for @xmath52 it moves with velocity @xmath53 . at @xmath54
the pinned interface outlines a transversal path on a directed percolation ( dp ) cluster @xcite with roughness @xmath55 , and a growth exponent @xmath56 . as discussed above the quenched nature of the noise
is relevant only when @xmath22 is close to @xmath54 .
the predictions for the negative qkpz class are different @xcite . in the pinned phase , @xmath57 and negative @xmath58 , the interface forms sawtooth configurations ( see fig .
[ field ] , bottom ) where alternating non - zero local average slopes @xmath59 help the system to remain pinned .
as @xmath22 decreases , the sawtooth slopes increase until there are discontinuous jumps at @xmath60 of both the average slope ( back to zero ) and the velocity @xmath50 , well evident in our experiment , see fig .
[ vf ] a ) . inside the pinned phase ,
the transitory dynamics is similar to positive qkpz @xcite , and the depinning was analyzed via a mapping to the first layer png model @xcite , a close cousin of kpz .
our experiment is made with the iodate arsenous acid ( iaa ) reaction , autocatalytic in iodide , with the concentration values @xmath61 = 7.5 $ ] mm and @xmath62_0 = 25 $ ] mm , such that the arsenous acid is in excess @xcite .
the resulting front has a velocity @xmath63 , and a width @xmath64 .
the front position is visualized with polyvinyl alcohol colored by transient iodine production @xcite .
the disordered flow is generated with a @xmath65 mixture of @xmath66 and @xmath67 mm diameters packed glass beads inside a transparent ( @xmath68 ) rectangular cell .
a range of injectors at the top of the cell can either suck out or inject unreacted fluid parallel to the vertical .
the bottom of the cell is dipped into a container with reacted solution to start the reaction , leading to a flat horizontal initial front .
the front propagates upwards in the absence of the flow , which is then switched on once the desired vertical position is reached .
we extract the location @xmath17 of the front at a given position @xmath69 and time @xmath70 from @xmath71 pixel digitized images with a 12-bit roper coolsnap hq video camera . to enhance the statistics
, we performed 3 to 6 different realizations for each of the 30 values of the applied flow velocity .
we define the control parameter as @xmath72 , whith @xmath73 the mean flow velocity and @xmath74 an _ ad - hoc _ constant @xcite , such that the front advances when @xmath75 or recedes when @xmath76 .
fig .
[ vf ] a ) displays the normalized front velocity , @xmath77 as function of @xmath10 . in the absence of flow ,
when @xmath78 , the reaction front propagates through the beads with a smooth shape at a constant velocity @xmath79 m/s [ supp .
movie 1 in @xcite ] .
when the flow is turned on , the front exhibits distinct self - affine scalings depending on @xmath10 , and quantified by the front width @xmath80 , and the standard deviation of the temporal height fluctuations @xmath81 @xcite . at large flow rates , when @xmath82 on fig .
[ vf ] , @xmath83 is a linear function of @xmath10 and the front propagates downstream for both orientations of the mean flow [ supp .
movie 2 and 3 in @xcite ] . fig .
[ w ] a ) shows the width @xmath84 and @xmath30 of saturated fronts , i.e. such that @xmath85 @xcite , determined for experiments realized at opposite mean - flow orientations .
they display both a similar roughness @xmath86 and @xmath87 and growth exponent @xmath88 and @xmath89 while @xmath90 .
as can be seen on fig .
[ h_beta ] a ) , at large front velocity of either orientation , the front exhibits scale - invariant fluctuations with statistical properties in agreement with the kpz class and the theoretical discussion around eq .
( [ eq3 ] ) .
in addition , since in the experiment @xmath91 , the expected kpz noise @xmath92 is almost independent of @xmath93 @xcite , and the amplitude of @xmath84 does not vary significantly with @xmath10 for a given flow orientation , as can be seen in fig .
[ w ] a ) .
when @xmath94 , some regions of the front pin to the flow heterogeneities . in this configuration ,
the front propagates mainly upstream , from the bottom to the top of the cell , while locally the front exhibits transiently static regions , as shown in fig .
[ field ] ( top ) .
note that the moving parts have a larger slope than the arrested or slowly propagating ones , leading to a lateral growth of the fronts in this regime ( supp .
movie 4 in @xcite ) .
when the opposite flow is amplified , the pinned portions become larger . the value @xmath95 for which the front eventually stops and remains static is @xmath96 . on fig .
[ w ] c ) and d ) for @xmath97 , the values @xmath98 and @xmath99 , are consistent with the theoretically predicted exponents of positive qkpz , @xmath100 ( see @xcite for additional measurements ) , suggesting that the front undergoes a depinning transition when @xmath101 . finally , when @xmath10 decreases below @xmath95 , the transient front propagation becomes very short .
for @xmath102 , the front is static almost instantaneously after the flow is turned on ( supp .
movie 5 in @xcite ) . when @xmath10 becomes negative , the front propagates in the direction opposite to the chemical reaction . for sufficiently small @xmath10 ,
@xmath103 , it quickly becomes static after a transient propagation and displays a particular sawtooth pattern @xcite .
one notes in fig .
[ field ] b ) that the front is slowed down or arrested when it reaches a certain slope , resulting in facet formation ( supp .
movie 6 in @xcite ) .
another depinning transition occurs at @xmath104 , below which triangular states become unstable and the front goes back to a phase moving from the top to the bottom of the cell . as indicated in fig .
[ h_beta ] , these triangular shapes lead to a different roughness exponent of the transiently moving parts compared to the final static front as @xmath105 @xcite .
the roughness and growth exponents of propagating regions have @xmath106 and @xmath107 , consistent with the observations in @xcite . however , the final static front roughness is larger : @xmath108 , increasing as the sawtooth slope rises when @xmath109 .
interestingly , a crossover from @xmath110 to @xmath111 for larger scales is visible in fig .
[ w ] d ) , underlining the second depinning transition at @xmath112 .
close to @xmath112 , the front pins to point - like regions , while close to @xmath95 the pinning regions extend horizontally .
this shows that receding fronts are consistent with negative qkpz , known for similar pinning processes and interface morphologies @xcite .
this model also predicts a first - order depinning transition , observed here as a jump in the @xmath113 curve . to better understand the behavior close to the transitions , lattice boltzmann simulations were performed in a @xmath114 disordered porous medium ( @xmath115 grid size ) , solving the convection - diffusion reaction and the darcy - brinkman equations @xcite .
fig .
[ vf ] b ) displays the numerical front velocity versus @xmath116 , where hatted quantities denote parameters in the simulations with @xmath117 . in good agreement with the experiments ,
two transitions occur at @xmath118 and @xmath119 and the velocity @xmath120 is almost linear in @xmath11 below the first transition and beyond the second one .
while the transition at @xmath121 is very abrupt , the second one at @xmath122 is more continuous .
moreover , the critical behavior can be fitted ( fig .
[ vf].b , insert ) with a power - law @xmath123 , with an exponent slightly larger than the theoretical value @xmath124 @xcite .
this is consistent with a first - order transition at @xmath121 and a second - order one at @xmath125 .
shown on fig .
[ h_betasim ] , the scaling exponents of the numerics are also in good agreement with the theory of the thermal kpz class for large @xmath126 . near the depinning transitions at @xmath125 and @xmath121 , the roughness exponents @xmath127 and @xmath128 are consistent with experiments and the positive and negative q - kpz predictions .
the remaining differences in the q - kpz pinned phase are due to different initial conditions : in the experiment , the fronts propagate without flow ( with their own roughness exponent at @xmath129 ) and then the flow is switched on , whereas in the simulation the initial front is flat ( see @xcite for details ) . the good agreement between experiments , numerical simulations , and theory for the different kpz universality classes
can be understood through the _ eikonal approximation _
@xcite .
for a thin front , the local front velocity follows the eikonal equation : @xmath130 , where @xmath131 is the normal of the front , @xmath132 the curvature and @xmath133 the local flow velocity . indeed , this equation is similar to the `` flux - line model '' of kardar @xcite where the chemical velocity plays the role of the lorentz force , and the disordered flow that of the random force . after projection and neglecting higher - order terms @xcite ,
the eikonal approximation yields @xmath134 assuming small gradients , and normalizing by @xmath135 , leads to eq . where @xmath136 and with the parameters @xmath137 , @xmath138 , and @xmath139 , whose small renormalization due to the neglected terms can be estimated @xcite .
the difference @xmath140 is related to the space average of the kpz term @xmath141 .
note that @xmath142 is independent of the front propagation direction , and fixed by the initial condition of the experiment .
negative qkpz describes then the backward moving fronts , i.e. @xmath143 , since performing @xmath144 is equivalent to measuring the front position along the @xmath145 axis .
finally , near the transition at @xmath112 , the slope of the sawtooths may be large .
although it correctly predicts the first - order transition , small - gradient qkpz may not be quantitatively accurate .
a more precise scenario was proposed in @xcite , based on the png model @xcite and extreme - value statistics . in conclusion ,
chemical - wave propagation coupled with the disordered flow in a porous medium , develop self - affine structures , with scaling exponents consistent with either kpz or qkpz classes .
remarkably , by tuning a single parameter , this system passes through three universality classes , providing a rich experimental setting to study growth phenomena .
slowly backward propagating fronts constitute beautiful experimental evidence of a chemical interface described by negative qkpz .
part of this phenomenology was recently observed in magnetic domain walls @xcite : it would be interesting to reach the thermal kpz class there by increasing the driving .
this work opens the door for further investigations on frozen pattern formation in out - of - equilibrium systems @xcite .
this work was supported by project procathet rtra triangle de la physique and psl grant anr-10-idex-0001 - 02-psl .
we are grateful to k. takeuchi , t. gueudre and a. rosso for useful discussions . 34 natexlab#1#1bibnamefont # 1#1bibfnamefont # 1#1citenamefont # 1#1url # 1`#1`urlprefix[2]#2 [ 2][]#2 , , , , , * * , ( ) . , * * , ( ) . , * * , ( ) . , , , , * * , ( ) . ,
* * , ( ) . , , , * * , ( ) . , , , , * * , ( ) .
, * * , ( ) . , * * , ( ) . , , , , , * * , ( ) . , , , , * * , ( ) . edwards , b. f. , phys . rev . lett . * 89 * , 104501 ( 2002 ) .
edwards , b. f. , chaos * 16 * , 043106 ( 2006 ) .
, , , , , , , * * , ( ) . , , , , , , , * * , ( ) . , * * , ( ) . , , , , , , * * , ( ) .
m. nicoli , r. cuerno , and m. castro , phys .
rev .
lett . * 111 * , 209601 ( 2013 ) and p. j. yunker , m. a. lohr , t. still , a. borodin , d. j. durian and a. g. yodh , phys .
rev .
lett . * 111 * , 209602 ( 2013 ) . ,
( ) .
i. corwin , arxiv:1106.1596 and references therein .
, , , , , , * * , ( ) .
, , , * * , ( ) . , , , * * , ( ) .
h. leschhorn et al . ,
ann .
phys . 6 , 1 ( 1997 ) . for reviews
see k. wiese and p. le doussal , markov processes relat .
fields * 13 * 777 ( 2007 ) and p. le doussal , int .
journal of modern physics b , 24 ( 2010 ) 3855 . , * * , ( ) . , * * , ( ) .
h. jeong , b. kahng and d. kim , phys .
rev .
lett . *
77 * 5094 ( 1996 ) and phys .
rev .
e * 59 * 1570 ( 1999 ) .
y. m. choi , h. j. kim , and i.m .
kim , phys .
rev .
e. * 66 * 047102 ( 2002 ) . , * * , ( ) . , , , * * , ( ) .
m. leconte , j. martin , n. rakotomalala , d. salin , phys .
rev .
lett . * 90 * , 128302 ( 2003 ) . , , , , , ,
* * , ( ) . supplementary material to this article .
we define : @xmath146 ^ 2\right>_{\delta x}}\right>_{x , t}$ ] ; here @xmath147 represents an average over a length @xmath148 , and @xmath149 a translational average over @xmath69 , as well as a temporal average over time @xmath70 , in the steady state .
temporal fluctuations are quantified by @xmath150 ^ 2\right>_{\delta t}}\right>_{x , t}$ ] , where @xmath151 is a time window .
both quantities exhibit power laws , @xmath152 for @xmath153 , and @xmath154 for @xmath155 , with @xmath156 and @xmath157 the spatial and temporal correlation lengths .
talon , l. , martin , j. , rakotomalala , n. , salin , d. and yortsos , y. water resour . res . *
39 * , 1135 - 1142 ( 2003 ) .
m. kardar , phys . rep . * 301 * , 85 ( 1998 )
. p. le doussal and k.j .
wiese , phys .
rev .
e * 67 * , 016121 ( 2003 ) .
t. gueudr , a. k. dubey , l. talon , and a. rosso , phys .
rev .
e * 89 * 041004 ( 2014 )
. k. w. moon , d. h. kim , s. c. yoo , c. g. cho , s. hwang , b. kahng , b. c. min , k. h. shin , and s. b. choe , phys .
rev .
let .
* 110 * , 107203 ( 2013 ) .
v. schaller et al .
pnas * 108 * , 19183 ( 2011 ) .
p. le doussal and t. giamarchi , phys .
rev .
b. 57 11356 ( 1998 ) .
* supplemental material for _ experimental evidence for three universality classes for reaction fronts in disordered flows _ * |
in this paper we study spectral properties of infinite matrices of the form @xmath12 where @xmath13 is an infinite sequence of @xmath1 s , and the box marks the entry at @xmath14 .
let @xmath15 denote the linear space of those complex - valued sequences @xmath16 for which @xmath17 , a hilbert space equipped with the norm @xmath18 .
then to each matrix @xmath19 with @xmath20 corresponds a bounded linear mapping @xmath21 , which we denote again by @xmath19 , given by the rule @xmath22 for @xmath23 .
following @xcite we will term a _
feinberg - zee hopping matrix_. further , in the case where each @xmath24 is an independent realisation of a random variable with probability measure whose support is @xmath25 , we will term @xmath19 a feinberg - zee _ random _ hopping matrix , this particular non - selfadjoint random matrix studied previously in @xcite .
but it is no loss of generality to restrict attention to matrices of the form as the case where the superdiagonal is also random can be reduced to by a simple gauge transformation ; see @xcite or ( * ? ? ?
* lemma 3.2 , theorem 5.1 ) . ]
the spectrum of a realisation @xmath19 of this random hopping matrix is given , almost surely , by ( e.g. , @xcite ) @xmath26 here @xmath27 denotes the spectrum of @xmath28 as an operator on @xmath15 .
note that implies that @xmath0 is closed .
equation holds whenever @xmath20 is _ pseudo - ergodic _ , which means simply that every finite sequence of @xmath1 s appears as a consecutive sequence somewhere in the infinite vector @xmath29 ; it is easy to see that @xmath29 is pseudo - ergodic almost surely if @xmath29 is random .
the concept of pseudo - ergodicity dates back to @xcite , as do the arguments that holds , or see @xcite for derived as a special case of more general limit operator results .
many of the above cited papers are concerned primarily with computing upper and lower bounds on @xmath0 . a standard upper bound for the spectrum
is provided by the numerical range .
it is shown in @xcite that , if @xmath20 is pseudo - ergodic , its numerical range @xmath30 , defined by @xmath31 , where @xmath32 is the inner product on @xmath15 , is given by @xmath33 this gives the upper bound that @xmath34 , the closure of @xmath35 .
other , sharper upper bounds on @xmath0 are discussed in section [ sec : pre ] below .
this current paper is related to the problem of computing lower bounds for @xmath0 via .
if @xmath36 is constant then @xmath28 is a laurent matrix and @xmath37 $ ] if @xmath38 , while @xmath39 $ ] if @xmath40 ; thus , by , @xmath41\cup { { \rm i}}[-2,2]\subset \sigma$ ] .
generalising this , if @xmath36 is periodic with period @xmath9 then @xmath27 is the union of a finite number of analytic arcs which can be computed by calculating eigenvalues of @xmath42 matrices ( see lemma [ lem : spper ] below ) . and , by , @xmath43 , where @xmath44 is the union of @xmath27 over all @xmath45 with period @xmath9 .
this implies , since @xmath0 is closed , that @xmath46 where @xmath47 .
we will call @xmath2 the _ periodic part _ of @xmath0 , noting that @xcite conjectures that equality holds in , i.e. that @xmath48 is dense in @xmath0 and @xmath49 . whether or not this holds is an open problem , but it has been shown in @xcite that @xmath48 is dense in the open unit disk @xmath50 , so that @xmath51 for a polynomial @xmath5 and @xmath52 , we define , as usual , @xmath53 and @xmath54 .
( we will use throughout that if @xmath55 is open then @xmath56 is open ( @xmath5 is continuous ) and , if @xmath5 is non - constant , then @xmath57 is also open , e.g. , ( * ? ? ? * theorem 10.32 ) . )
the proof of in @xcite depends on the result , in the case @xmath58 , that @xmath59 this implies that @xmath60 , for @xmath61 , where @xmath62 $ ] and @xmath63 , for @xmath64 .
thus @xmath65 , which is dense in @xmath66 , is also in @xmath48 , giving .
hagger @xcite makes a large generalisation of the results of @xcite , showing the existence of an infinite family , @xmath4 , containing monic polynomials of arbitrarily high degree , for which holds . for
each of these polynomials @xmath5 let @xmath67 ( here @xmath68 , @xmath69 , etc . )
hagger @xcite observes that , as a consequence of and , @xmath70 .
he also notes that standard results of complex dynamics ( e.g. , ( * ? ? ?
* corollary 14.8 ) ) imply that @xmath71 , so that @xmath72 ; here @xmath73 denotes the julia set of the polynomial @xmath5 .
( where @xmath74 , @xmath75 , etc . , we recall @xcite that the _ filled julia set
_ @xmath76 of a polynomial @xmath5 of degree @xmath77 is the compact set of those @xmath78 for which the sequence @xmath79 , the _ orbit _ of @xmath80 , is bounded .
further , the boundary of @xmath76 , @xmath81 , is the _ julia set _ of @xmath5 . ) the definition of the set @xmath4 in @xcite , while constructive , is rather indirect .
the first contribution of this paper ( section [ sec : poly ] ) is to make explicit the membership of @xmath4 . as a consequence
we show , in particular , that @xmath82 , for @xmath83 , where @xmath84 , and @xmath85 is the chebychev polynomial of the second kind of degree @xmath9 @xcite .
the second contribution of this paper ( section [ sec : interior ] ) is to say more about the interior points of @xmath2 .
previous calculations of large subsets of @xmath48 , precisely calculations of @xmath44 for @xmath9 as large as 30 @xcite , suggest that @xmath2 fills most of the square @xmath35 , but @xmath86 , the interior of @xmath2 , is known only to contain @xmath87 . using that the whole family @xmath88 , we prove that @xmath89 . this result
is then used to show that @xmath2 is the closure of its interior . using that @xmath90 , for @xmath10
, we also , in section [ subsec : b ] , construct new explicit subsets of @xmath2 and its interior ; in particular , extending , we show that @xmath91 for @xmath92 . in the final section [ sec : julia ] of the paper
we address a conjecture of hagger @xcite that , not only for every @xmath10 is @xmath71 ( which implies @xmath72 ) , but also the filled julia set @xmath93 .
this is a stronger result as , while the compact set @xmath73 has empty interior ( * ? ? ?
* summary 14.12 ) , @xmath76 contains , in addition to @xmath73 , all the bounded components of the open fatou set @xmath94 .
we show , by a counter - example , that this conjecture is false .
but , positively , we conjecture that @xmath95 for all @xmath10 , and we prove that this is true for a large subset of @xmath4 , in particular that @xmath96 for @xmath7 .
the results in this paper provide new information on the almost sure spectrum @xmath97 of the bi - infinite feinberg - zee random hopping matrix .
they are also relevant to the study of the spectra of the corresponding finite matrices
. for @xmath98 let @xmath99 denote the set of @xmath42 matrices of the form , so that @xmath100 and , for @xmath101 , @xmath102 , where @xmath103 ( this notation will be convenient , but note that @xmath104 is independent of the last component of @xmath105 . ) then @xmath106 is the set of eigenvalues of the matrix @xmath104 .
let @xmath107 so that @xmath108 is the union of all eigenvalues of finite matrices of the form . then
, connecting spectra of finite and infinite matrices , it has been shown in @xcite that @xmath109 , for @xmath64 , so that @xmath110 .
further , @xcite shows that @xmath108 is dense in @xmath48 , so that @xmath111 . in section [ subsec : a ] we build on and extend these results , making a surprising connection between the eigenvalues of the finite matrices and the spectra of the periodic operators associated to the polynomials in @xmath4 .
the result we prove ( theorem [ thm : dense ] ) , is key to the later arguments in section [ sec : interior ] .
[ [ notions - of - set - convergence . ] ] notions of set convergence .
+ + + + + + + + + + + + + + + + + + + + + + + + + + + we will say something below about set sequences , sequences approximating @xmath0 and @xmath2 from above and below , respectively . we will measure convergence in the standard haussdorf metric @xmath112 ( * ? ? ?
* f ) ( or see @xcite ) on the space @xmath113 of compact subsets of @xmath114 .
we will write , for a sequence @xmath115 and @xmath116 , that @xmath117 if @xmath118 as @xmath119 and @xmath120 for each @xmath9 ; equivalently , @xmath117 as @xmath121 if @xmath120 for each @xmath9 and , for every @xmath122 , @xmath123 , for all sufficiently large @xmath9 . similarly , we will write that @xmath124 if @xmath118 as @xmath119 and @xmath125 for each @xmath9 ; equivalently , @xmath124 if @xmath125 for each @xmath9 and , for every @xmath122 , @xmath126 , for all sufficiently large @xmath9 . the following observation , which follows immediately from ( * ? ? ?
* proposition 3.6 ) ( or see ( * ? ? ?
* lemma 4.15 ) ) , will be useful : [ eq : setconv ] if @xmath127 are closed and @xmath128 is bounded , then @xmath129 , as @xmath121 . [ [ spectra - of - periodic - operators . ] ] spectra of periodic operators .
+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + we will need explicit formulae for the spectra of operators @xmath19 with @xmath20 in the case when @xmath29 is periodic . for @xmath130 ,
let @xmath131 denote @xmath19 in the case that @xmath132 , for @xmath133 , and @xmath134 , for @xmath135
. for @xmath98 let @xmath136 denote the order @xmath9 identity matrix , @xmath137 the @xmath42 matrix which is zero except for the entry 1 in row @xmath9 , column @xmath138 , and let @xmath139 denote the transpose of @xmath137 . for @xmath98 , @xmath140 , and @xmath141 , let @xmath142 .
the following characterisation of the spectra of periodic operators is well - known ( see lemma 1 and the discussion in @xcite ) . [
lem : spper ] for @xmath98 and @xmath140 , @xmath143 key to our arguments will be an explicit expansion for the determinant in the above lemma , expressed in terms of the following notation .
for @xmath98 , @xmath130 , and @xmath78 , let @xmath144 for @xmath145 and @xmath78 , let @xmath146 , for @xmath147 , and define @xmath148 then , for @xmath98 and @xmath140 , expanding the determinant by laplace s rule by the first row and by the last row , we see that @xmath149 the following lemma follows easily by induction on @xmath9 , using .
the bounds on @xmath150 stated are used later in corollary [ cor : pkbound ] .
[ lem : qkbound ] if @xmath151 , for some @xmath98 , then @xmath150 is a monic polynomial of degree @xmath152 , and @xmath150 is even and @xmath153 if @xmath9 is odd , @xmath150 is odd and @xmath154 if @xmath9 is even .
further , @xmath155 , @xmath156 , and @xmath157 , for @xmath158 . for @xmath98 let @xmath159 denote the @xmath42 flip matrix , that is the @xmath42 matrix with entry @xmath160 in row @xmath161 , column @xmath162 , where @xmath163 is the kronecker delta . then @xmath164 so that @xmath165 .
for @xmath130 , let @xmath166 .
the first part of the following lemma is essentially a particular instance of a general property of determinants .
[ lem : qksym ] if @xmath140 , for some @xmath98 , and @xmath167 , then @xmath168 ; if @xmath169 , then @xmath170 suppose first that @xmath167
. then @xmath171 given by is the determinant of a matrix @xmath172 , and @xmath173 is the determinant of @xmath174 , so that @xmath175 . that holds if @xmath169 can be shown by an easy induction on @xmath9 , using , or directly by a gauge transformation .
here is the announced explicit expression for the determinant in lemma [ lem : spper ] .
[ lem : det ] ( * ? ? ?
* lemma 3 ) for @xmath98 , @xmath140 , @xmath78 , and @xmath176 , @xmath177 where @xmath178 is a monic polynomial of degree @xmath9 given by @xmath179 further , @xmath178 is odd ( even ) if @xmath9 is odd ( even ) .
since , from the above lemmas , @xmath178 is odd ( even ) and @xmath150 even ( odd ) if @xmath9 is odd ( even ) , implies that @xmath180 this last equation obtained using .
the following lemma , proved using these representations , makes clear that many different vectors @xmath105 correspond to the same polynomial @xmath178 .
[ lem : pksym ] if @xmath151 , for some @xmath98 , and @xmath167 or @xmath181 is a cyclic permutation of @xmath105 , then @xmath182 . if @xmath169 then @xmath183 .
using and we see that @xmath184 if @xmath181 is a cyclic shift of @xmath105 , i.e. , @xmath185 , @xmath186 , and @xmath187 , then the right - hand side is identically zero .
thus @xmath188 if @xmath181 is a cyclic permutation of @xmath105 . if @xmath167 then that @xmath189 follows from and lemma [ lem : qksym ] . if @xmath169 then that @xmath183 follows from and lemma [ lem : qksym ] .
call @xmath140 _ even _ if @xmath190 , and _ odd _ if @xmath191 .
then ( * ? ? ?
* corollary 5 ) , it is immediate from lemmas [ lem : spper ] and [ lem : det ] that @xmath192 ) , \mbox { if $ k$ is even } , \ ; { { \rm spec}\,}a_k^{{\rm per } } = p_k^{-1}({{\rm i}}[-2,2 ] ) , \mbox { if $ k$ is odd.}\ ] ] [ [ complex - dynamics . ] ] complex dynamics .
+ + + + + + + + + + + + + + + + + in section [ sec : julia ] below we show that filled julia sets , @xmath76 , of particular polynomials @xmath5 , are contained in the periodic part @xmath2 of the almost sure spectrum of the feinberg - zee random hopping matrix . to articulate and prove these results we will need terminology and results from complex dynamics .
throughout this section @xmath5 denotes a polynomial of degree @xmath77 .
we have defined above the compact set that is the filled julia set @xmath76 , the julia set @xmath193 , the fatou set @xmath194 ( the open set that is the complement of @xmath73 ) , and the orbit of @xmath195 .
it is easy to see that , if @xmath196 , i.e. , the orbit of @xmath197 is not bounded , then @xmath198 as @xmath121 , i.e. , @xmath199 , the _ basin of attraction of infinity_. we call @xmath52 _ invariant _ if @xmath200 , and _ completely invariant _ if both @xmath55 and its complement are invariant , which holds iff @xmath201 .
clearly @xmath202 is completely invariant .
we call @xmath197 a _ fixed point _ of @xmath5 if @xmath203 and a _ periodic point _ if @xmath204 , for some @xmath64 , in which case the finite sequence @xmath205 , where @xmath206 , @xmath207 , ... , is the _ cycle _ of the periodic point @xmath197 .
we say that @xmath197 is an _ attracting fixed point _ if @xmath208 , a _ repelling fixed point _
if @xmath209 , and a _ neutral fixed point _ if @xmath210 .
generalising , we say that a periodic point @xmath197 is attracting / repelling / neutral if @xmath211/@xmath212/@xmath213 , where @xmath214 . by the chain rule , @xmath215 , where @xmath206 and @xmath216 , @xmath217 .
the value @xmath218 is the _ multiplier _ of the neutral periodic point @xmath197 ( clearly @xmath219 ) .
if @xmath197 is a neutral periodic point with multiplier @xmath220 we say that it is _ rationally neutral _ if @xmath221 for some @xmath222 , otherwise we say that it is _ irrationally neutral_. we call @xmath197 a _ critical point _ of @xmath5 if @xmath223 . if @xmath224 is an attracting periodic point we denote by @xmath225 the _ basin of attraction _ of the cycle @xmath226 of @xmath197 , by which we mean @xmath227 . here , for @xmath228 and @xmath229 , @xmath230 .
it is easy to see that @xmath225 contains some neighbourhood of @xmath231 , and hence that @xmath225 is open .
we will make use of standard properties of the julia set captured in the following theorem .
we recall that a family @xmath232 of analytic functions is _ normal _ at a point @xmath195 if , in some fixed neighbourhood @xmath233 of @xmath197 , each @xmath234 is analytic and every sequence drawn from @xmath232 has a subsequence that is convergent uniformly either to some analytic function or to @xmath235 .
* summary 14.12 ) @xmath73 is compact with no isolated points , is uncountable , and has empty interior .
@xmath73 is completely invariant , @xmath236 , for every @xmath64 , @xmath237 @xmath73 is the closure of the repelling periodic points of @xmath5 , and , for all except at most one @xmath195 , @xmath238 the fatou set @xmath194 has one unbounded component @xmath239 .
it follows from that @xmath240 ; indeed , @xmath241 as a consequence of the maximum principle @xcite .
it may happen that this is the only component of @xmath194 so that @xmath242 .
this is the case if @xmath243 and @xmath244 , when @xmath245 $ ] @xcite . if @xmath194 has more than one component it either has two components ( for example , if @xmath246 and @xmath247 , when @xmath248 and @xmath249 ) , or infinitely many components ( * ? ? ?
* theorem iv.1.2 ) .
it follows from that @xmath250 and @xmath251 if @xmath224 is an attracting fixed point or periodic point ( * ? ? ?
* theorem iii.2.1 ) . arguing similarly ( * ? ? ?
* theorem 1.7 ) , @xmath252 , where @xmath253 , so that @xmath254 .
because @xmath73 is completely invariant and @xmath5 is an open map , the image @xmath255 of any component @xmath239 of @xmath194 is also a component of @xmath194 .
now consider the orbit of @xmath239 , i.e. @xmath256 . the following statement of possible behaviours is essentially sullivan s theorem @xcite .
[ thm : sullivan ] let @xmath239 be a component of @xmath194 .
then one of the following cases holds : \i ) @xmath257 , for some @xmath98 , in which case we call @xmath239 a _ periodic component _ of @xmath194 , and call the smallest @xmath9 for which @xmath257 the _ period _ of @xmath239 .
( if @xmath258 , when @xmath239 is invariant , we also term @xmath239 a _ fixed component _ of @xmath194 . )
\ii ) @xmath259 is a periodic component of @xmath194 for some @xmath260 , in which case we say that @xmath239 is a _ preperiodic component _ of @xmath194 .
the above theorem makes clear that the orbit of every component @xmath239 of @xmath194 enters a periodic cycle after a finite number of steps . to understand the eventual fate under iterations of @xmath5 of the components of the fatou set it is helpful to understand the possible behaviours of a periodic component
this is achieved in the _ classification theorem _ ( e.g. @xcite ) . to state
this theorem we introduce further terminology .
let us call a fixed component @xmath239 of @xmath194 a _ parabolic _ component if there exists a neutral fixed point @xmath261 with multiplier 1 such that the orbit of every @xmath262 converges to @xmath224 . call a fixed component @xmath239 of @xmath194 a _ siegel disk _ if @xmath5 is conjugate to an irrational rotation on @xmath239 , which means that there exists a conformal mapping @xmath263 with @xmath264 and an irrational @xmath265 such that @xmath266 where @xmath267 , @xmath268 , and @xmath269 .
it is easy to see that , for @xmath270 , @xmath271 iff @xmath272 , and @xmath273 .
thus every siegel disk contains a unique irrationally neutral fixed point ( the siegel disk fixed point ) .
[ thm : class ] _ classification theorem ( * ? ? ?
* theorem iv.2.1 ) .
_ if @xmath239 is a periodic component of @xmath194 with period @xmath98 , in which case @xmath239 is also a component of @xmath274 , then exactly one of the following holds : in @xcite gives a 4th option , that @xmath239 is a _ herman ring component _ of @xmath275 .
this option is excluded if @xmath5 is a polynomial @xcite . ]
\a ) @xmath239 contains an attracting periodic point @xmath224 which is an attracting fixed point of @xmath276 , and @xmath277 ; \b ) @xmath239 is a parabolic component of @xmath275 ; \c ) @xmath239 is a siegel disk component of @xmath275 .
the following proposition relates the above cases to critical points of @xmath5 ( see ( * ? ? ?
* theorems iii 2.2 and 2.3 , pp .
83 - 84 ) ) : [ prop : crit ] if @xmath239 is a periodic component of @xmath194 with period @xmath9 then either : ( i ) @xmath239 is a parabolic component of @xmath275 or contains an attracting periodic point , in which case @xmath278 contains a critical point of @xmath5 ; or ( ii ) @xmath239 is a siegel disk component of @xmath275 and there is a critical point @xmath224 of @xmath5 such that the orbit of @xmath224 is dense in @xmath279 .
the following proposition will do the work for us in section [ sec : julia ] .
[ prop : cd2a ] suppose that @xmath52 is bounded , open and simply - connected , that @xmath280 is closed , and that the orbit of every critical point in @xmath76 eventually lies in @xmath281 .
then @xmath282 that @xmath283 if @xmath284 follows from .
if @xmath285 then , by theorems [ thm : sullivan ] and [ thm : class ] , after a finite number of iterations the orbit of @xmath197 is in a periodic component of @xmath194 that is parabolic , part of the domain of attraction of an attracting periodic point , or is a siegel disk . in the first two cases it follows that @xmath286 as @xmath121 for some cycle @xmath231 , but also @xmath287 for some critical point @xmath224 by proposition [ prop : crit ] .
this last implies that @xmath288 is non - empty , and so @xmath289 for some @xmath9 . in the case
that the orbit of @xmath197 is eventually in a siegel disk then also @xmath289 for some @xmath9 for , if the orbit of every critical point @xmath290 is eventually in @xmath281 , it follows that the boundary of every siegel disk is in @xmath281 , and ( as @xmath55 is simply connected ) that every siegel disk is in @xmath55 .
[ [ previous - upper - bounds - on - sigma . ] ] previous upper bounds on @xmath0 .
+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + we have noted above that , if @xmath20 is pseudo - ergodic , then @xmath291 , given by .
similarly , the spectrum of @xmath292 is contained in the closure of _ its _ numerical range , so that is the polynomial @xmath58 , @xmath293 is @xmath294 in the notation of @xcite , so that @xmath293 is a superset of the second order numerical range . ] @xmath295 hagger @xcite introduces a new , general method for computing numerical ranges of infinite tridiagonal matrices via the schur test , which he applies to computing the numerical range of @xmath292 ( by expressing it as the direct sum of tridiagonal matrices ) .
these calculations show that @xmath296 ; indeed , the calculations in @xcite imply that @xmath297 , where @xmath298 is even and periodic with period @xmath299 , given explicitly on @xmath300 $ ] by @xmath301 by comparison , in polar form , @xmath302 figure [ figure2 ] includes a visualisation of @xmath303 and @xmath293 .
the bound , expressed concretely through , is the sharpest explicit upper bound on @xmath0 obtained to date .
it implies that @xmath0 is not convex since we also know ( see below ) that @xmath304 , @xmath305 , and @xmath306 are all in @xmath0 .
a different family of upper bounds was established in @xcite ( and see @xcite ) , expressed in terms of pseudospectra .
for a square matrix @xmath172 of order @xmath9 and @xmath307 let @xmath308 denote the @xmath309-pseudospectrum of @xmath172 ( with respect to the 2-norm ) , i.e. , @xmath310 , with the understanding that @xmath311 if @xmath80 is an eigenvalue , so that @xmath312 . (
here @xmath18 is the operator norm of a linear mapping on @xmath313 equipped with the @xmath314-norm . )
analogously to , for @xmath122 and @xmath64 , let @xmath315 which is the union of the pseudospectra of @xmath316 distinct matrices .
then it is shown in @xcite that @xmath317 where @xmath318 and @xmath319 is the unique solution in @xmath320 $ ] of the equation @xmath321 .
clearly , @xmath322 is a convergent family of upper bounds for @xmath0 that is in principle computable ; deciding whether @xmath323 requires only computation of smallest singular values of @xmath42 matrices ( see ( * ? ? ? * ( 39 ) ) ) .
explicitly @xmath324 , and @xmath325 is plotted for @xmath326 in @xcite .
but for these values @xmath327 , and computing @xmath325 for larger @xmath9 is challenging , requiring computation of the smallest singular value of @xmath316 matrices of order @xmath9 to decide whether a particular @xmath328 .
substantial numerical calculations in @xcite established that @xmath329 , providing the first proof that @xmath0 is a strict subset of @xmath303 , this confirmed now by the simple explicit bound and .
complementing the upper bounds on @xmath0 that we have just discussed , lower bounds on @xmath0 have been obtained by two methods of argument .
the first is that tells us that @xmath330 for every @xmath36 . in particular
this holds in the case when @xmath45 is periodic , when the spectrum of @xmath28 is given explicitly by lemmas [ lem : spper ] and [ lem : det ] , so that , as observed in the introduction , @xmath331 explicitly ( * ? ? ? * lemma 2.6 ) , in particular , @xmath332\cup { { \rm i}}[-2,2]\ \mbox { and } \ \pi_2
= \pi_1 \cup \{x\pm { { \rm i}}x : -1\leq x\leq 1\}.\ ] ] in the introduction we have defined @xmath333 and have termed @xmath334 , also a subset of @xmath0 since @xmath0 is closed , the periodic part of @xmath0 .
we have also recalled the conjecture of @xcite that @xmath49 .
let @xmath335 then it follows from lemma [ eq : setconv ] that @xmath336 if , as conjectured , @xmath49 , then complements ; together they sandwich @xmath0 by convergent sequences of upper ( @xmath325 ) and lower ( @xmath337 ) bounds that can both be computed by calculating eigenvalues of @xmath42 matrices .
figures [ figure2 ] and [ figure3 ] include visualisations of @xmath338 , indistinguishable by eye from @xmath339 , but note that the solid appearance of @xmath338 , which is the union of a large but finite number of analytic arcs , is illusory .
see @xcite for visualisations of @xmath340 for a range of @xmath9 , suggestive that the convergence is approximately achieved by @xmath341 . the same method of argument to obtain lower bounds was used in @xcite , where a special sequence @xmath36 was constructed with the property that @xmath342 , so that , by , @xmath343 .
the stronger result , that this new lower bound on @xmath0 is in fact also a subset of @xmath2 , was shown in @xcite , via a second method of argument for constructing lower bounds , based on surprising symmetries of @xmath0 and @xmath2 .
we will spell out in a moment these symmetries ( one of these described first in @xcite , the whole infinite family in @xcite ) , which will be both a main subject of study and a main tool for argument in this paper .
but first we note more straightforward but important symmetries . in this lemma and throughout @xmath344
denotes the complex conjugate of @xmath345 .
[ lem : symeasy ] ( * ? ? ?
* lemma 3.4 ) ( and see @xcite , ( * ? ? ?
* lemma 4 ) ) .
all of @xmath44 , @xmath346 , @xmath2 , and @xmath0 are invariant with respect to the maps @xmath347 and @xmath348 , and so are invariant under the dihedral symmetry group @xmath349 generated by these two maps .
to expand on the brief discussion in the introduction , @xcite proves the existence of an infinite set @xmath4 of monic polynomials of degree @xmath77 , this set defined constructively in the following theorem , such that the elements @xmath10 are symmetries of @xmath48 and @xmath0 in the sense that below holds .
[ thm : cs ] @xcite
let @xmath4 denote the set of those polynomials @xmath178 , defined by , with @xmath130 for some @xmath101 , for which it holds that : ( i ) @xmath350 and @xmath351 ; ( ii ) @xmath352 , where @xmath353 is the vector identical to @xmath105 but with the last two entries interchanged , so that @xmath354 and @xmath355
. then @xmath356 for all @xmath10 .
we will call @xmath4 _ hagger s set of polynomial symmetries for @xmath0 .
_ we remark that if @xmath10 then it follows from , by taking closures and recalling that @xmath5 is continuous , that also @xmath357 we note also that @xmath358 implies that @xmath359 , but not vice versa , and that @xmath360 iff @xmath361 further , we note that it was shown earlier in @xcite that holds for the particular case @xmath362 ( this the only element of @xmath4 of degree 2 , see table [ table1 ] ) ; in @xcite it was also shown , as an immediate consequence of and lemma [ lem : symeasy ] , that @xmath363 for @xmath362 .
whether this last inclusion holds in fact for all @xmath10 is an open problem . our first result is a much more explicit characterisation of @xmath4 .
[ prop : eqchar ] the set @xmath4 is given by @xmath364 , where @xmath365 consists of those vectors @xmath130 with @xmath101 , for which : ( i ) @xmath366 and @xmath367 ; and ( ii ) @xmath368 , or @xmath369 and @xmath370 , for @xmath371 , so that @xmath372 is a palindrome .
moreover , if @xmath373 , then @xmath374 it is clear from theorem [ thm : cs ] that what we have to prove is that , if @xmath140 with @xmath101 and @xmath350 , @xmath351 , then @xmath352 if @xmath368 or 3 ; further , if @xmath375 , then @xmath352 iff @xmath376 is a palindrome .
if @xmath140 with @xmath101 and @xmath350 , @xmath351 , then , from and , @xmath377 thus , if @xmath368 or 3 , or @xmath375 and @xmath376 is a palindrome , @xmath378 since @xmath379 , this a consequence of the definitions in the cases @xmath368 and 3 , of lemma [ lem : qksym ] in the case @xmath375 . similarly , @xmath380 , so that @xmath352 .
conversely , assume that @xmath140 with @xmath375 , @xmath350 , @xmath351 , and @xmath381 . to show that @xmath376 is a palindrome we need to show that @xmath382 , for @xmath383 . using and
then , we see that @xmath384 thus @xmath385 .
but , if @xmath386 and @xmath387 , then , applying , @xmath388 as this holds for all @xmath80 and , by lemma [ lem : qkbound ] and , @xmath389 and @xmath390 are both monic polynomials of degree @xmath391 , it follows first that @xmath392 and then that @xmath393
. thus that @xmath382 for @xmath383 follows by induction on @xmath162 .
the following corollary is immediate from and lemma [ lem : qkbound ] .
[ cor : small ] suppose that @xmath101 , @xmath140 , and @xmath394 . then , as @xmath395 , @xmath396 if @xmath9 is odd , while @xmath397 if @xmath9 is even .
let us denote by @xmath11 the polynomial @xmath178 when @xmath105 has length @xmath398 , @xmath399 , @xmath400 , and all other entries are @xmath138 s .
it is convenient also to define @xmath401 . clearly , as a consequence of the above proposition , @xmath82 for @xmath7 ( that these particular polynomials are in @xmath4
was observed earlier in @xcite ) .
we will write down shortly an explicit formula for @xmath11 in terms of chebyshev polynomials of the 2nd kind . recall that @xmath8 , the chebychev polynomial of the 2nd kind of degree @xmath9 , is defined by @xcite @xmath402 , @xmath403 , and @xmath404 , for @xmath64 . [ lem : cheb ] for @xmath405 , @xmath6 .
this follows easily by induction from and .
we note that , using the standard trigonometric representations for the chebychev polynomials @xcite , for @xmath405 , @xmath406 a similar representation in terms of hyperbolic functions can be given for the polynomial @xmath178 when @xmath105 has length @xmath407 and @xmath408 ; we denote this polynomial by @xmath409 .
clearly , for @xmath7 , @xmath410 by proposition [ prop : eqchar ] , and @xmath409 is an odd function by lemma [ lem : det ] .
the proof of the following lemma , like that of lemma [ lem : cheb ] , is a straightforward induction that we leave to the reader .
[ lem : cheb2 ] @xmath411 , @xmath412 , and @xmath413 , for @xmath7 .
moreover , for @xmath405 and @xmath414 , @xmath415 the following lemma leads , using lemmas [ lem : cheb ] and [ lem : cheb2 ] , to explicit formulae for other polynomials in @xmath4 .
for example , if @xmath416 denotes the polynomial @xmath178 when @xmath105 has length @xmath398 , @xmath399 , @xmath400 , and all other entries are @xmath417 s , then , by lemmas [ lem : cheb ] and [ lem : minus ] , @xmath418 [ lem : minus ] if @xmath140 and @xmath394 , then @xmath419 and @xmath420 .
suppose that @xmath140 and @xmath394 . if @xmath368 , then @xmath421 and @xmath422 . if @xmath369 , defining @xmath423 by @xmath424 , @xmath425 , and @xmath426 , for @xmath427 , @xmath428 by proposition [ prop : eqchar ] , so that @xmath429 .
that @xmath420 comes from lemma [ lem : pksym ] .
we note that proposition [ prop : eqchar ] implies that there are precisely @xmath430 vectors of length @xmath9 in @xmath365 , so that there are between 1 and @xmath430 polynomials of degree @xmath9 in @xmath4 , as conjectured in @xcite .
note , however , that there may be more than one @xmath431 that induce the same polynomial @xmath394 .
for example , @xmath432 for @xmath433 , and , defining @xmath434 and using lemma [ lem : minus ] , also @xmath435 in table [ table1 ] ( cf .
@xcite ) we tabulate all the polynomials in @xmath4 of degree @xmath436 .
.the elements @xmath394 of degree @xmath436 . [ cols=">,<",options="header " , ] if @xmath437 , so that @xmath5 and @xmath438 are polynomial symmetries of @xmath0 in the sense that holds , then also the composition @xmath439 is a polynomial symmetry of @xmath0 in the same sense . but note from table [ table1 ] that , while @xmath440 , none of @xmath441 , @xmath442 , @xmath443 , @xmath444 , or @xmath445 are in @xmath4 .
thus @xmath4 does not contain all polynomial symmetries of @xmath0 , but whether there are polynomial symmetries that are not either in @xmath4 or else compositions of elements of @xmath4 is an open question .
we finish this section by showing in subsection [ subsec : a ] the surprising result that @xmath4 is large enough that we can reconstruct the whole of @xmath2 from the polynomials @xmath394 .
this result will in turn be key to the proof of our main theorem in section [ sec : interior ] .
then in subsection [ subsec : b ] we use that holds for the polynomials in table [ table1 ] to obtain new explicit lower bounds on @xmath2 , including that @xmath446 .
recall from proposition [ prop : eqchar ] that @xmath364 , let @xmath447 , and define @xmath448 the following result seems rather surprising , given that @xmath365 is much smaller than @xmath449 in the sense that there are precisely @xmath430 vectors of length @xmath9 in @xmath365 but @xmath450 in @xmath449 .
[ thm : dense ] @xmath451 , so that @xmath452 is dense in @xmath2 and @xmath453 we will show in proposition [ prop : inc ] below that , for @xmath454 , @xmath455 where , for @xmath7 , @xmath456 denotes the set of those vectors in @xmath365 that have length @xmath457 . since also @xmath458 , for every @xmath405 ( e.g. ( * ? ? ?
* lemma 2.10 ) ) , this implies that @xmath451 , which implies that @xmath452 is dense in @xmath2 since @xmath108 is dense in @xmath48 ( * ? ? ?
* theorem 1 ) , and the result follows .
the key step in the proof of the above theorem is the following refinement of theorem 4.1 in @xcite , which uses our new characterisation , proposition [ prop : eqchar ] , of @xmath4 .
[ prop : inc ] suppose @xmath459 and @xmath140 , for some @xmath101 , and let @xmath460 .
then @xmath461 where @xmath462 .
further , @xmath463 .
the proof modifies ( * ? ? ?
* theorem 4.1 ) where the same result is proved for the special case that @xmath464 , @xmath465 . following that proof ,
suppose that @xmath80 is an eigenvalue of @xmath104 with corresponding eigenvector @xmath466 , let @xmath467 and @xmath468 , so that @xmath469 , and set @xmath470 where @xmath471 marks the entry at position @xmath472
. then @xmath473 is a bi - infinite tridiagonal matrix with zeros on the main diagonal , and with each of the first sub- and super - diagonals a vector in @xmath474 that is periodic with period @xmath475 .
define @xmath476 , the space of bounded , complex - valued sequences @xmath16 , by @xmath477 where @xmath471 marks the entry @xmath478 .
then it is easy to see that @xmath479 , so that @xmath480 .
is in the spectrum of @xmath473 as an operator on @xmath481 , but this is the same as the @xmath15-spectrum from general results on band operators , e.g. @xcite . ] further , by a simple gauge transformation ( * ? ? ?
* lemma 3.2 ) , @xmath482 , where @xmath181 is given by .
we have shown that @xmath483 .
but , choosing in particular @xmath484 and @xmath485 , @xmath486 , we see from proposition [ prop : eqchar ] that @xmath487 , so that @xmath488 .
[ rem : one ] proposition [ prop : inc ] implies that @xmath483 for 16 different vectors @xmath423 , corresponding to different choices of @xmath459 .
some of these vectors correspond to the same polynomial @xmath489 and hence , by , to the same spectrum @xmath490 .
in particular , if @xmath491 , then the choices @xmath492 and @xmath493 lead to the same polynomial by proposition [ prop : eqchar ] and the definition of @xmath4 .
but , if @xmath494 , again by proposition [ prop : eqchar ] and the definition of @xmath4 , the choices @xmath492 and @xmath493 must lead to different polynomials , and neither of these polynomials can be in @xmath4 . on the other hand , as observed already in the proof of proposition [ prop : inc ] , the choices @xmath484 and @xmath495 lead to an @xmath496 and so to a polynomial @xmath428 . thus proposition [ prop : inc ] implies that there are at least three distinct polynomials @xmath489 such that @xmath483 .
figure [ figure1 ] plots @xmath497 and @xmath490 , for the vectors defined by , in the case that @xmath498 and @xmath499 . for other plots of the spectra of finite and periodic feinberg - zee matrices , and the interrelation of these spectra ,
see @xcite . as noted in @xcite , that the polynomials @xmath10 satisfy gives us a tool to compute explicit lower bounds on @xmath2 .
indeed @xcite shows visualisations of @xmath500 , for several @xmath10 and @xmath501 , and visualisations of unions of @xmath500 for @xmath5 varying over some finite subset of @xmath4 . since , by , @xmath502 , it follows from that all these sets are subsets of @xmath2 .
the study in @xcite contains visualisations of subsets of @xmath2 as just described , but no associated analytical calculations .
complementing the study in @xcite we make explicit calculations in this section that illustrate the use of the polynomial symmetries to compute explicit formulae for regions of the complex plane that are subsets of @xmath2 , adding to the already known fact that @xmath502 .
our first lemma and corollary give explicit formulae for @xmath503 when @xmath504 and when @xmath505 .
these formulae are expressed in terms of the function @xmath506 $ ] , where , for @xmath507 , @xmath508 is defined as the largest positive solution of @xmath509 if @xmath510 , this is the unique solution in @xmath511 ( on which interval @xmath512 ) , while if @xmath513 it is the unique solution in @xmath514 ( on which interval @xmath512 ) .
explicitly , for @xmath507 , @xmath515 and @xmath516 , @xmath517 .
[ lem : degree3 ] if @xmath518 , then @xmath519 where @xmath520 , for @xmath265 .
@xmath521 and is even and periodic with period @xmath522 . in the interval
@xmath523 $ ] the only stationary points of @xmath55 are global maxima at @xmath524 and @xmath522 , with @xmath525 , and a global minimum at @xmath299 .
further , @xmath55 is strictly decreasing on @xmath526 $ ] and @xmath527 in @xmath523 $ ] iff @xmath528 or @xmath529 , with equality iff @xmath530 or @xmath531 . if @xmath532 ( with @xmath533 , @xmath534 ) , then @xmath535 .
it is straightforward to show that @xmath536 iff @xmath537 .
the properties of @xmath55 claimed follow easily from the properties of @xmath538 stated above .
[ cor : degree3 ] if @xmath539 , then @xmath540 this is clear from lemma [ lem : degree3 ] and the observation that @xmath541 .
since @xmath542 , it follows from the above lemma and corollary and that @xmath543 . but this implies by , since @xmath544 is also in @xmath4 , that also @xmath545 , where , for @xmath52 , @xmath546 .
in particular , @xmath547 it is easy to check that @xmath548 next note from table [ table1 ] that @xmath10 where @xmath549 factorises as @xmath550 thus , for @xmath551 with @xmath552 , @xmath553 let @xmath554 be the unique positive solution of @xmath555 . clearly @xmath556 if @xmath557 , with @xmath558 , so that @xmath559 we have shown most of the following proposition that extends to a region @xmath560 ( illustrated in figure [ figure2 ] ) the part of the complex plane that is known to consist of interior points of @xmath2 , making explicit implications of the polynomial symmetries @xmath4 of @xmath0 . before @xcite and the current paper
the most that was known explicitly was that @xmath561 .
[ prop : betterbounds ] @xmath562 that @xmath563 is combined with and lemma [ lem : symeasy ] .
it is easy to see that @xmath564 is invariant with respect to the maps @xmath347 and @xmath348 , and so is invariant under the dihedral symmetry group @xmath349 generated by these two maps .
thus to complete the proof it is sufficient to show that @xmath565 for @xmath566 and @xmath567 .
now since , by lemma [ lem : degree3 ] , @xmath527 for @xmath568 , @xmath569 for @xmath570 , @xmath571 .
further , since , by lemma [ lem : degree3 ] , @xmath55 is even and is increasing on @xmath572 $ ] , @xmath573 , for @xmath574 , and @xmath575 , for @xmath576 .
but @xmath577 , so that @xmath578 for @xmath579 , which implies that @xmath580 , so that @xmath581 . similarly , @xmath582 and @xmath583 , so that @xmath584 .
thus @xmath585 for @xmath566 and @xmath586 , while @xmath587 for @xmath566 and @xmath588 . to conclude that @xmath589 , it remains to show that @xmath590 for @xmath591 and @xmath592 .
but it is easy to check that , for these ranges of @xmath593 and @xmath594 , @xmath595 provided @xmath596 .
but this last inequality holds since @xmath597 and @xmath598 so that @xmath599 .
we have just , in proposition [ prop : betterbounds ] , extended to a region @xmath560 the part of the complex plane that is known to consist of interior points of @xmath2 . in this section
we explore the relationship between @xmath2 and its interior further .
we show first of all , using and that @xmath600 for every @xmath101 , that @xmath601 .
next we use this result to show that , for every @xmath101 , all but finitely many points in @xmath602 are interior points of @xmath2 . finally , we prove , using theorem [ thm : dense ] , that @xmath2 is the closure of its interior .
if indeed it can be shown , as conjectured in @xcite , that @xmath49 , then the result will imply the truth of another conjecture in @xcite , that @xmath0 is the closure of its interior .
our technique for establishing that @xmath601 will be to use that @xmath603 , for every @xmath604 , this a particular instance of .
this requires first a study of the real solutions of the equations @xmath605 and their interlacing properties , which we now undertake . from
, @xmath606 this implies that the equation @xmath607 has a solution in @xmath608 .
let @xmath609 denote the largest solution in this interval .
further , if @xmath610 then @xmath611 has a solution in @xmath612 since @xmath613 . for @xmath614 let @xmath615 denote the largest solution to @xmath611 in @xmath616 , which is in the interval @xmath612 if @xmath610 , while an explicit calculation gives that @xmath617 . throughout the following calculations we use the notation @xmath618 from .
[ lem : inc ] for @xmath614 it holds that @xmath11 is strictly increasing on @xmath619 , that @xmath620 , and that @xmath621 for @xmath622 .
explicitly , @xmath623 , so that @xmath624 and these claims are clear for @xmath625 .
suppose now that @xmath610 .
it follows by induction that , for @xmath369 , @xmath618 is strictly decreasing on @xmath626 . for
@xmath627 is strictly decreasing on @xmath628 and , if this statement is true for some @xmath369 , then @xmath629 is strictly decreasing on @xmath630 .
further , @xmath631 since @xmath632 for @xmath633 . as @xmath634 , these observations imply that , on @xmath635 , @xmath11 is strictly increasing , and that @xmath636 .
thus @xmath637 follows from the definitions of @xmath638 , and @xmath621 for @xmath622 .
as observed above it follows from that @xmath639 , for @xmath7 . combining this observation with lemma [ lem : inc ] and the fact that @xmath640
if @xmath641 , we obtain the following corollary .
[ cor : intpts ] for @xmath642 , @xmath643 .
also , @xmath644 . by definition of @xmath609 , @xmath645 as @xmath646 .
thus the above corollary and the following lemma together imply that @xmath601 .
[ lem : inter ] for @xmath647 , @xmath648 . since @xmath649 and @xmath650 , it is easy to see that @xmath651 , so that the claimed result holds for @xmath652 . to see the result for @xmath614
we will show , equivalently , that @xmath653 , for @xmath642 . here
@xmath654 , for @xmath64 , is the smallest solution of @xmath655 in @xmath656 , so that @xmath657 , while @xmath658 , for @xmath659 , denotes the smallest solution of @xmath660 in @xmath661 , so that @xmath662 .
we have shown in the proof of lemma [ lem : inc ] that , for @xmath604 , @xmath663 is strictly decreasing on @xmath664 , and that @xmath665 , for @xmath610 , while @xmath666 so that @xmath667 similarly , for @xmath398 , @xmath668 so that @xmath669 these inequalities imply that , for @xmath614 , @xmath670 , and since @xmath671 is strictly decreasing on this interval and @xmath672 it follows that @xmath673 . since also @xmath674 is strictly decreasing on @xmath675 and @xmath676 since @xmath677 , we see that also @xmath678 .
[ cor : main ] @xmath679 . from corollary [ cor : intpts ] and
lemma [ lem : inter ] it follows that @xmath680 and that @xmath681 , for @xmath604 .
thus @xmath682 for @xmath604 , and so @xmath601 since @xmath645 as @xmath646 .
applying lemma [ lem : symeasy ] we obtain the stated result .
the following lemma follows immediately from corollary [ cor : main ] , , and .
[ lem : pinint ] suppose that @xmath683 , so that @xmath105 has length @xmath9 and @xmath394 has degree @xmath9 .
then all except at most @xmath684 points in @xmath685 are interior points of @xmath2 .
further , if @xmath686 then there exists a sequence @xmath687 such that @xmath688 as @xmath646 . as an example of the above lemma , suppose that @xmath689 . then ( see table [ table1 ] ) @xmath690 and , from , @xmath691 .
there are precisely four points , @xmath692 .
these are not interior points of @xmath2 since they lie on the boundary of @xmath693 .
combining the above lemma with theorem [ thm : dense ] , we obtain the last result of this section .
[ thm : interior ] @xmath2 is the closure of its interior .
suppose @xmath694 .
then , by theorem [ thm : dense ] , @xmath80 is the limit of a sequence @xmath695 , and , by lemma [ lem : pinint ] , for each @xmath9 there exists @xmath696 such that @xmath697 , so that @xmath698 as @xmath121 .
it was shown in @xcite that , for every polynomial symmetry @xmath10 , the corresponding julia set @xmath73 satisfies @xmath699 , where @xmath700 is defined by .
( the argument in @xcite is that @xmath701 by , and that @xmath702 by . )
it was conjectured in @xcite that also the filled julia set @xmath703 , for every @xmath10 . in this section
we will first show by a counterexample that this conjecture is false ; we will exhibit a @xmath10 of degree 18 for which @xmath704 .
however , we have no reason to doubt a modified conjecture , that @xmath95 , for all @xmath10 . and the main result of this section will be to prove that @xmath95 for a large class of @xmath10 , including @xmath705 , for @xmath7 .
our first result is the claimed counterexample .
[ lem : counter ] let @xmath706 , so that ( by proposition [ prop : eqchar ] ) @xmath394 , this polynomial given explicitly by @xmath707 then @xmath708 .
let @xmath709 .
if we can find a @xmath710 that is an attracting fixed point of @xmath5 , then , for all sufficiently small @xmath307 , @xmath711 satisfies @xmath712 and @xmath713 , so that @xmath714 and @xmath715 . calculating in double - precision floating - point arithmetic in matlab we see that @xmath716 appears to be a fixed point of @xmath5 , with @xmath717 so that this fixed point appears to be attracting . to put this on a rigorous footing we work in exact arithmetic to deduce , by the intermediate value theorem , that @xmath718 has a solution @xmath719 , and that @xmath720 .
then , noting that @xmath721 , where @xmath722 and @xmath723 , we see that @xmath724 for @xmath725 . but
this implies that @xmath726 , so that @xmath727 is an attracting fixed point .
numerical results suggest that amongst the polynomials @xmath10 of degree @xmath728 , there is only one other similar counterexample of a polynomial with an attracting fixed point outside the unit disk , the other example of degree 19 .
we turn now to positive results .
part of our argument will be to show , for every @xmath10 , that @xmath729 , via the following lower bounds that follow immediately from lemma [ lem : qkbound ] , and .
[ cor : pkbound ] if @xmath140 , for some @xmath98 , then @xmath730 , for @xmath158 .
if @xmath394 , then @xmath731 , for @xmath158 .
[ cor : pkjul ] let @xmath709 , where @xmath140 , for some @xmath98 .
then @xmath732 . if @xmath10 , then @xmath733 .
let @xmath195 with @xmath734 .
then , by corollary [ cor : pkbound ] , for some neighbourhood @xmath233 of @xmath197 , @xmath735 for @xmath736 .
thus , and by montel s theorem ( * ? ? ? * theorem 14.5 ) , the family @xmath737 is normal at @xmath197 .
so @xmath738 , by .
we have shown that @xmath739 , so that also @xmath740 and @xmath741 . if @xmath10 and @xmath742 then , by corollary [ cor : pkbound ] , @xmath743 so that @xmath744 and so @xmath199 .
thus @xmath745 .
we remark that the bounds in corollary [ cor : pkbound ] appear to be sharp .
in particular , if @xmath746 has length @xmath101 , we see from , , and lemma [ lem : cheb ] that @xmath747 , since @xmath748 @xcite . and we note that , if @xmath705 , for some @xmath405 , then @xmath749 . finally , we recall that we have already noted that , for @xmath709 , with @xmath243 , i.e. , @xmath750 , the julia set is @xmath751 $ ] , so that @xmath752 for this @xmath5 .
the polynomial @xmath750 is an example where @xmath753 so @xmath754 . the next lemma tells us that this does not happen , that @xmath76 is strictly larger than @xmath73 , if @xmath755 .
[ lem : larger ] @xmath756 is non - empty for @xmath10 .
if @xmath10 is even then , by lemma [ lem : det ] and corollary [ cor : small ] , @xmath757 , so that @xmath524 is an attracting fixed point .
clearly @xmath758 ( which is non - empty ) is a subset of @xmath759 .
similarly , by lemma [ lem : det ] and corollary [ cor : small ] , if @xmath5 is odd then @xmath760 and @xmath761 , so that @xmath762 is a rationally neutral fixed point and has a ( non - empty ) attracting region contained in @xmath763 ( * ? ? ?
* section ii.5 ) , this region clearly also in @xmath700 .
the above lemma and imply that @xmath764 is non - empty for all @xmath10 , in particular that @xmath765 if @xmath5 is even .
the main result of this section is the following criterion for the _ whole _ of @xmath763 to be contained in @xmath2 .
[ prop : filled3 ] suppose that @xmath10 , and that the critical points of @xmath5 in @xmath76 have orbits that lie eventually in @xmath766 . then @xmath95 .
choose @xmath767 and @xmath45 with @xmath768 such that @xmath769 $ ] and @xmath770 $ ] , this possible by corollary [ cor : pkjul ] which says that the closed set @xmath771 .
set @xmath772\cup{{\rm i}}[a , b]\cup 1.1{{\overline{\mathbb d}}}$ ] , and choose a simply - connected open set @xmath55 such that @xmath773 , this possible by corollary [ cor : main ] and proposition [ prop : betterbounds ] . by hypothesis ,
the orbits of the critical points in @xmath76 lie eventually in @xmath281 .
thus the lemma follows from proposition [ prop : cd2a ] and . as an example of application of this theorem ,
consider @xmath10 given by ( see table [ table1 ] ) @xmath774 .
this @xmath5 has critical points @xmath524 and @xmath775 .
since @xmath776 it follows from corollary [ cor : pkjul ] that @xmath777 , while @xmath524 is a fixed point .
theorem [ prop : filled3 ] tells us that @xmath76 , visualised in figure [ figure3 ] , is contained in @xmath2 .
we note that , since all the critical points of @xmath5 except the fixed point @xmath524 are in @xmath202 , @xmath76 is not connected ( * ? ? ?
* theorem iii 4.1 ) and , by theorem [ thm : sullivan ] and proposition [ prop : crit ] , @xmath778 , which implies that @xmath93 .
further , recalling the discussion in section [ sec : pre ] , @xmath779 , and , since @xmath76 has more than one component , @xmath763 has infinitely many components ( * ? ? ?
* theorem iv 1.2 ) .
the above example is a particular instance of a more general result .
it is straightforward to see that if @xmath5 is a polynomial with zeros only on the real line , then all the critical points are also on the real line . since , by lemma [ lem : cheb ] , @xmath6 , and all the zeros of the polynomial @xmath780 are real , it follows that all the zeros of @xmath11 are real , so all its critical points are also real , and so the orbits of all the critical points are real .
further , by corollary [ cor : pkjul ] , the orbits of the critical points in @xmath76 stay in @xmath781 . likewise ,
as ( see ) @xmath782 , all the critical points of @xmath416 lie on @xmath783 , and so the orbits of these critical points are real if @xmath457 is even , pure imaginary if @xmath457 is odd .
further , by corollary [ cor : pkjul ] , the orbits of the critical points in @xmath76 stay in @xmath784 . applying theorem [ prop : filled3 ] we obtain : [ cor : kp ] @xmath96 and @xmath785 , for @xmath7 . numerical experiments carried out for the polynomials @xmath10 of degree @xmath786 ( see table [ table1 ] and ( * ?
* table 1 ) ) appear to confirm that these polynomials satisfy the conditions of theorem [ prop : filled3 ] , i.e. , it appears for each polynomial @xmath5 that the orbit of every critical point either diverges to infinity or is eventually in @xmath766 .
the same appears true for the polynomial @xmath10 of degree 18 in lemma [ lem : counter ] for which @xmath704 .
thus it appears , from numerical evidence and theorem [ prop : filled3 ] , that @xmath95 for these examples .
these numerical experiments and corollary [ cor : kp ] motivate a conjecture that @xmath95 for all @xmath10 .
we finish this paper with a note of open problems regarding the spectrum of the feinberg - zee random hopping matrix , particularly problems that the above discussions have highlighted .
we recall first that @xcite made several conjectures regarding @xmath0 .
it was proved in @xcite that @xmath787 , but the following conjectures remain open : of these conjectures , perhaps the first has the larger implications . certainly , if @xmath788 , then we have noted below that we have constructed already convergent sequences of upper ( @xmath325 ) and lower ( @xmath337 ) bounds for @xmath0 that can both be computed by calculating eigenvalues of @xmath42 matrices .
further , if @xmath788 , then the second of the above conjectures follows from theorem [ thm : interior ] .
the last three conjectures in the above list were prompted in large part by plots of @xmath340 in @xcite , the plot of @xmath338 reproduced in figures [ figure2 ] and [ figure3 ] .
it is plausible that these plots , in view of , approximate @xmath2 .
we see no clear route to establishing the third conjecture above . regarding the fourth
, we note that the existence of the set @xmath4 of polynomial symmetries satisfying suggests a self - similar structure to @xmath48 and to @xmath2 and @xmath0 and their boundaries .
further , @xcite has shown that @xmath2 contains the julia sets of all polynomials in @xmath4 , and proposition [ prop : filled3 ] and corollary [ cor : kp ] show that @xmath2 contains the filled julia sets , many of which have fractal boundaries , of the polynomials in an infinite subset of @xmath4 .
this last conjecture follows if @xmath788 , by theorem [ thm : cs ] from @xcite .
further ( see the discussion below ) , it was shown in @xcite that @xmath789 for the only polynomial of degree 2 in @xmath4 , @xmath362 .
the major subject of study and tool for argument in this paper has been hagger s set of polynomial symmetries @xmath4 .
we finish with one final open question raised immediately before section [ subsec : a ] . | in this paper we study the spectrum @xmath0 of the infinite feinberg - zee random hopping matrix , a tridiagonal matrix with zeros on the main diagonal and random @xmath1 s on the first sub- and super - diagonals ; the study of this non - selfadjoint random matrix was initiated in feinberg and zee ( _ phys .
rev .
e _
* 59 * ( 1999 ) , 64336443 ) .
recently hagger ( _ random matrices : theory appl .
_ , * 4 * 1550016 ( 2015 ) ) has shown that the so - called _ periodic part _
@xmath2 of @xmath0 , conjectured to be the whole of @xmath0 and known to include the unit disk , satisfies @xmath3 for an infinite class @xmath4 of monic polynomials @xmath5 . in this paper
we make very explicit the membership of @xmath4 , in particular showing that it includes @xmath6 , for @xmath7 , where @xmath8 is the chebychev polynomial of the second kind of degree @xmath9 .
we also explore implications of these inverse polynomial mappings , for example showing that @xmath2 is the closure of its interior , and contains the filled julia sets of infinitely many @xmath10 , including those of @xmath11 , this partially answering a conjecture of the second author . |
the goal of the paper is to describe all 7-dimensional nilmanifolds admitting an invariant contact structure , see theorem t : main . to achieve this goal , we classify all 7-dimensional real nilpotent lie algebra that admit a contact structure , see corollary c : main .
first , we recall some definitions . for a lie algebra @xmath1
the upper central series is the increasing sequence of ideals defined as follows : @xmath2 \subset c_i(\mathfrak{g})\}.\ ] ] in other words @xmath3 is the inverse image under the canonical mapping of @xmath4 onto @xmath5 of the center of @xmath5 .
the lie algebra @xmath4 is called _ nilpotent _ if there is an integer @xmath6 such that @xmath7 the minimal such @xmath6 is called the _ index of nilpotency _ , @xcite .
a connected lie group is nilpotent if and only if its lie algebra is nilpotent .
a _ nilmanifold _ @xmath8 is a compact homogeneous space of the form @xmath9 , where @xmath10 is a simply connected nilpotent lie group and @xmath11 is a discrete cocompact subgroup in @xmath10 , @xcite .
for example , an @xmath12-dimensional torus @xmath13 is obviously a nilmanifold . if we consider the group @xmath14 of upper triangular matrices having 1s along the diagonal then the quotient @xmath15 is a nilmanifold , called the heisenberg nilmanifold , where @xmath16 is the set of matrices having integral entries .
+ let @xmath17 be a nilmanifold and let @xmath18 be the lie algebra of the lie group @xmath10 .
it is well - known that the exponential map @xmath19 is a global diffeomorphism and the quotient map @xmath20 is the universal covering map .
hence every nilmanifold is the eilenberg - maclane space @xmath21 . by malcev
s theorem a discrete group @xmath11 can be realized as the fundamental group of a nilmanifold if and only if it is a finitely presented nilpotent torsion free group @xcite .
recall that a space @xmath22 is called _ nilpotent _ if the fundamental group @xmath23 is nilpotent and the action of @xmath24 on all homotopy groups @xmath25 is nilpotent .
it is clear that every nilmanifold @xmath9 is a nilpotent space since @xmath26 is nilpotent and all higher homotopy groups @xmath27 are trivial .
hence , it follows from the fundamental theorem of the rational homotopy theory @xcite that two nilmanifolds have the same rational homotopy type if and only if the corresponding chevalley - eilenberg complexes are isomorphic . [
t : ma ] a simply connected nilpotent lie group @xmath10 admits a discrete cocompact subgroup @xmath11 if and only if there exists a basis @xmath28 of the lie algebra @xmath29 of @xmath10 such that the structural constants @xmath30 arising in brackets @xmath31=\sum_k{c_{ij}^ke_k}}\ ] ] are rational numbers for all @xmath32 .
let @xmath4 be a lie algebra with a basis @xmath33 .
denote by @xmath34 the basis for @xmath35 dual to @xmath36 .
we obtain a differential @xmath37 on the exterior algebra @xmath38 by defining it on degree 1 elements as @xmath39)\ ] ] and extending to @xmath38 as a graded bilinear derivation
. then @xmath40=\sum_{l}{c_{ij}^lx_l}}\ ] ] where @xmath41 are the structure constants of @xmath4 , and it follows from duality that @xmath42 hence on generators the differential is expressed as @xmath43 note that the condition @xmath44 is equivalent to the jacobi identity in the lie algebra .
we call the differential graded algebra @xmath45 the _ chevalley - eilenberg _ complex of the lie algebra @xmath46 .
a _ contact structure _ on a manifold @xmath8 of odd dimension @xmath47 is a completely non - integrable @xmath48-dimensional tangent plane distribution @xmath49 . in the coorientable case
the distribution may be defined by a differential 1-form @xmath50 as @xmath51 .
then the non - integrability condition can be expressed by the inequality @xmath52 , i.e. the form @xmath53 is nowhere zero . such differential form @xmath50 is called a _ contact differential form_. a contact structure can be viewed as an equivalence class of contact differential forms , where two forms @xmath50 and @xmath54 are equivalent if and only if @xmath55 where @xmath56 is a nowhere zero smooth function on @xmath8 .
a _ contact manifold _ is a pair @xmath57 where @xmath8 is a smooth manifold of dimension @xmath47 and @xmath58 is a contact structure on @xmath8 , @xcite .
we say that a lie algebra @xmath1 of dimension @xmath47 admits a contact structure if there is an element @xmath58 of degree @xmath59 in the chevalley - eilenberg complex @xmath60 of @xmath1 such that @xmath61 we say that a contact structure ( resp .
a contact form ) on a nilmanifold @xmath62 is left ( right ) _ invariant _ if the contact structure ( resp . a contact form ) is invariant with respect to the left ( resp .
right ) @xmath10-action on @xmath62 .
[ r : invar ] since every invariant diferential form on a nilmanifold @xmath17 is completely determined by its values on the lie algebra @xmath18 of @xmath10 , we conclude that a nilmanifold @xmath17 admits an invariant contact structure if and only if @xmath18 admits a contact structure . note that any invariant contact structure on @xmath62 is coorientable .
clearly , the kernel of invariant contact form gives us an invariant contact structure .
conversely , for any contact form @xmath50 on @xmath62 we can perform the averaging of the lift of @xmath50 to @xmath10 rescaled down by a nowhere zero function with the finite integral and get an invariant contact form @xmath63 . furthermore , @xmath64 and
@xmath50 yield the same contact structure provided @xmath50 defines an invariant contact structure .
in particular , a nilmanifold @xmath8 admits an invariant contact structure if and only if @xmath8 admits an invariant contact form . to find all 7-dimensional nilmanifolds that admit an invariant contact structure we used the classification of 7-dimensional indecomposable nilpotent lie algebras .
many attempts have been done on this topic .
( we shall see below in that the decomposable nilpotent lie algebras do not admit a contact structure . )
there are a few lists available : safiullina ( 1964 , over @xmath65 ) @xcite , romdhani ( 1985 , over @xmath66 and @xmath65 ) @xcite , seeley ( 1988 , over @xmath65 ) @xcite , ancochea and goze ( 1989 , over @xmath65 ) @xcite .
the lists above are obtained by using different invariants .
carles @xcite introduced a new invariant - the weight system , compared the lists of safiullina , romdhani and seeley , and found mistakes and omissions in all of them . later in 1993
seeley incorporated all the previous results and published his list over @xmath65 , @xcite . in 1998
gong used the skjelbred - sund method to classify all 7-dimensional nilpotent lie algebras over @xmath66 , @xcite .
we will use gong s classification with some corrections from the list of magnin , @xcite . to find an invariant contact structure on a nilmanifold
, we first apply remark r : invar .
so , we must check whether the nilpotent lie algebras from the gong s list admit a contact structure . to achieve this goal
, we use the sullivan minimal model theory and corollary c : ce ( see sections [ s : min ] and [ s : main ] ) .
next , for a given 7-dimensional nilpotent lie algebra having a contact structure , we must check whether the corresponding simply connected lie group contains a discrete cocompact subgroup . for this
we use the malcev s criterion ( theorem t : ma ) .
i would like to thank my advisors alexander dranishnikov and yuli rudyak for many useful discussions and comments on this work .
references for the minimal model theory are @xcite . a _ differential graded algebra _ ( dga ) over a field @xmath6 of characteristic @xmath67 is a graded vector space @xmath68 with multiplications @xmath69 which are associative and commutative in graded sense @xmath70 where @xmath71 and @xmath72 denote the degrees of @xmath73 and @xmath74 respectively in the underlying graded vector space .
the dga @xmath75 also possesses a differential @xmath76 which is a graded derivation @xmath77 satisfying @xmath78 a _ morphism _ in the category of the dga s is an algebra homomorphism which commutes with the differential and respects grading . a dga @xmath79 is called a _ model _ for a dga @xmath80 if there exists a dga - morphism @xmath81 that induces an isomorphism on cohomology . if @xmath79 is freely generated in the sense that @xmath82 for a graded vector space @xmath83 then it is called a free model for @xmath84 , where the notation @xmath85means that , as a graded algebra , @xmath86 is the tensor product of a polynomial algebra on even elements @xmath87 and an exterior algebra on odd elements @xmath88 .
a dga @xmath79 is called a _ minimal model _ of @xmath80 if the following conditions are satisfied : +
( i ) @xmath89 is a free model for @xmath80 ; + ( ii ) @xmath37 is indecomposable in the following sense : there exists a basis @xmath90 for some well - ordered index set @xmath91 such that @xmath92 if @xmath93 , and each @xmath94 is expressed in terms of @xmath95 , @xmath96 .
it follows that @xmath97 does not have a linear part . a minimal model of a smooth manifold
@xmath8 is a minimal model of its de rham dga .
we need the following fact , @xcite .
let @xmath9 be a nilmanifold .
then the complex of differential forms on @xmath8 can be identified with the the complex of differential forms on @xmath10 which are right invariant by the elements of @xmath11 .
let @xmath9 be a nilmanifold .
the natural inclusion of the complex of right invariant differential forms on @xmath10 into the complex of the differential forms on @xmath62 @xmath98 induces an isomorphism on the cohomology level .
[ c : ce ] the minimal model of a compact nilmanifold @xmath62 is isomorphic to the chevalley - eilenberg complex of the lie algebra @xmath99 of @xmath100 .
a lie algebra @xmath99 is called _ decomposable _ if it can be represented as a direct sum of its ideals .
[ t : indecom ] let @xmath62 be a nilmanifold of dimension @xmath101 .
if the lie algebra @xmath99 of the lie group @xmath10 is decomposable then the nilmanifold @xmath62 does not admit an invariant contact structure .
let @xmath102 where @xmath103 and @xmath104 are ideals in @xmath99 so that @xmath105=0 $ ] for all @xmath106 .
let @xmath107 be the basis of @xmath103 and @xmath104 , respectively .
denote by @xmath108 the dual basis of @xmath109 and @xmath110 , respectively .
consider the chevalley - eilenberg complex @xmath111 of the lie algebra @xmath99 .
then @xmath112 @xmath113 because @xmath114=0 $ ] for all @xmath115 .
hence , without loss of generality we may assume that @xmath116 is the basis of @xmath117 that is ordered to satisfy the indecomposability condition of the minimal model @xmath111 .
consider an arbitrary @xmath59-form @xmath118 on @xmath99 .
then the derivative @xmath119 will contain no terms with @xmath120 or @xmath121 .
thus , the @xmath48-form @xmath122 vanishes because it is a linear combination of terms each of which is a product of @xmath48 @xmath59-forms from the set @xmath123 of cardinality @xmath124 .
it follows that the nilmanifold @xmath62 can not admit an invariant differential form @xmath58 such that @xmath125 .
now we discuss the existence of contact structures on indecomposable lie algebras .
we have proved that there are no contact structures on decomposable nilpotent lie algebras , see theorem t : indecom above .
all 7-dimensiomal indecomposable nilpotent lie algebras over the real field have been classified , @xcite . over reals
there are 138 non - isomorphic 7-dimensiomal indecomposable nilpotent lie algebras and , in addition , 9 one - parameter families .
for the one - parameter families , a parameter @xmath126 is used to denote a structure constant that may take on arbitrary values ( with some exceptions ) in @xmath127 an invariant @xmath128 is given for each family in which multiple values of @xmath126 yield isomorphic lie algebras , i.e. , if @xmath129 , then the two corresponding lie algebras are isomorphic , and conversely .
all one - parameter families admit a contact structure ( with some restrictions on a parameter @xmath126 ) and only @xmath130 lie algebras admit a contact structure .
we conduct a detailed computation for the algebra ( 1357c ) and the one - parameter family ( 147e ) . here
we use the same notation for the lie algebras as in @xcite , where the lie algebras are listed in accordance with their upper central series dimensions . for instance , the lie algebras that have the upper central series dimensions 1,4,7 are listed as follows : ( 147a ) , ( 147b ) , ( 147c ) , etc . for the sake of brevity
we will drop the sign of the wedge product ( for instance , @xmath131 means @xmath132 ) .
let @xmath18 be a 7-dimensional nilpotent lie algebra with the basis @xmath133 and the corresponding dual basis @xmath134 . consider the chevalley - eilenberg complex @xmath135 of the lie algebra @xmath29 . if one can not find nonzero structural constants @xmath136 such that @xmath137 then @xmath18 does not admit a contact structure because for every @xmath59-form @xmath138 on @xmath18 the @xmath139-form @xmath140 contains at most five distinct elements from the basis @xmath141 and therefore @xmath142 .
for this reason @xmath143 @xmath0-dimensional indecomposable nilpotent lie algebras that are listed in @xcite , pp.@xmath144 do not admit a contact structure and therefore we do not list them in our paper .
thus , if a 7-dimensional simply connected nilpotent lie group @xmath10 admits a discrete cocompact subgroup @xmath145 and the lie algebra @xmath18 of @xmath100 falls into this list of @xmath143 @xmath0-dimensional indecomposable nilpotent lie algebras then the nilmanifold @xmath17 does not admit an invariant contact structure .
now we show that the lie algebra ( 1357c ) and the one - parameter family ( 147e ) admit a contact structure .
first we consider the lie algebra ( 1357c ) .
let @xmath133 be a basis of the lie algebra ( 1357c ) and @xmath134 be the corresponding dual basis .
note that @xmath146 stands for the upper central series dimensions .
the nontrivial lie brackets of the lie algebra ( 1357c ) are defined as follows : + in dimension @xmath147 there are only two non - isomorphic nilpotent lie algebras .
one of them is abelian and therefore does not admit a contact structure .
hence a torus @xmath148 does not admit an invariant contact structure .
another @xmath147-dimensional nilpotent lie algebra has nontrivial lie bracket @xmath149=x_3 $ ] and therefore admits a contact structure @xmath150 .
hence the heisenberg manifold is the only @xmath147-dimensional nilmanifold that admits an invariant contact structure .
j. etnyre .
lectures on open book decompositions and contact structures . in : _
floer homology , gauge theory , and low - dimensional topology _ ,
103141 , clay math .
proc . , 5 , amer .
soc . , providence , ri , 2006 . l. magnin . determination of 7-dimensional indecomposable nilpotent complex lie algebras by adjoining a derivation to 6-dimensianal lie algebras .
represent . theory _ , * 13 * ( 2010 ) , no . 6 , 723 - 753 . | in this paper we give the list of all @xmath0-dimensional nilpotent real lie algebras that admit a contact structure . based on this list
, we describe all @xmath0-dimensional nilmanifolds that admit an invariant contact structure . |
the period - luminosity ( @xmath1-@xmath5 ) relation of classical cepheids is a yardstick in several astrophysical and cosmological problems .
the cepheid distances to external galaxies rely on fiducial @xmath1-@xmath5 relations based on large magellanic cloud ( lmc ) variables and these distance determinations are used to calibrate secondary distance indicators , and in turn to estimate the hubble constant @xmath6 . however , a general consensus on the `` universality '' of the @xmath1-@xmath5 relations , and in particular on their dependence on the cepheid chemical composition has not been reached yet . on the theoretical side ,
it is worth mentioning that the nonlinear convective pulsation models computed by our group ( fiorentino et al .
2007 , hereinafter [ f07 ] ; caputo 2008 , and references therein ) show that the metallicity effect on the predicted @xmath1-@xmath5 relations depends on the adopted photometric band .
the synthetic linear @xmath1-@xmath5 relations , for an increase in the global metal content from @xmath7=0.004 to 0.02 , become on average shallower , with the slope of the optical @xmath1-@xmath8 , @xmath1-@xmath9 and @xmath1-@xmath10 relations decreasing from @xmath1129% , to 15% and to 8% , respectively .
the same change in metallicity causes no significant effect on the near - infrared @xmath1-@xmath5 relations .
moreover , quoted predictions also indicate that the metal - rich pulsators with periods longer than five days present _ fainter _ optical magnitudes than the metal - poor ones .
the extent depends once again on the adopted passband . at even larger metal abundances
@xmath7=0.03 - 0.04 , the pulsation models suggest that the helium content @xmath12 also affects the cepheid properties at periods longer than about ten days .
it was also suggested ( fiorentino et al .
2002 , hereinafter [ f02 ] ; marconi , musella & fiorentino 2005 , hereinafter [ m05 ] ) that the metallicity effect on cepheid distances _ based on @xmath13 and @xmath14 magnitudes _ , is not linear over the entire metallicity range @xmath7=0.004 - 0.04 , but presents a sort of `` turnover '' at roughly solar chemical composition . as a whole , the use of lmc - calibrated @xmath1-@xmath9 and @xmath1-@xmath10 relations to provide distance estimates with an intrinsic error of @xmath150.10 mag is fully justified for cepheids with @xmath16 10 days and/or helium - to - hydrogen enrichment ratio @xmath17 2.0 ( see e.g. f02 and m05 for more details ) . on the other hand , the average correction for cepheids with @xmath18 10 days , high metal abundances ( @xmath19 0.03 ) and @xmath20 3.0 is larger than 0.1 mag . as a consequence , cepheids with @xmath21 20 days and oxygen abundance and @xmath22 one has [ @xmath23=log@xmath24log@xmath25 . ] [ o / h]@xmath26 0.2 , as measured in several spiral galaxies observed by the @xmath4 key projects ( freedman et al .
1994 ; saha et al .
1994 ) , the average metallicity correction varies from about @xmath270.2 mag to @xmath11 + 0.25 mag as the adopted @xmath28 ratio varies from 2 to 3.5 . on the observational side ,
independent investigations suggest either a negligible metallicity effect or that galactic ( metal - rich ) cepheids are somehow _ brighter _ , at fixed period , than lmc ( metal - poor ) variables ( sasselov et al . 1997 ; kennicutt et al .
1998 , 2003 ; kanbur et al . 2003 ; tammann et al .
2003 ; sandage et al .
2004 ; storm et al .
2004 ; groenewegen et al .
2004 ; sakai et al . 2004
; ngeow & kanbur 2004 ; pietrzynski et al .
2007 ) . in the latter case , the empirical metallicity dependence of the cepheid true distance modulus @xmath29 , as usually described by the parameter @xmath30log@xmath7 where @xmath31 is the extent of the metallicity correction and @xmath32log@xmath7=log@xmath33log@xmath34 , spans a large range of _ negative _ values up to @xmath350.4 mag dex@xmath36 , with an average value of approximately @xmath270.25 mag dex@xmath36 ( sakai et al . 2004 and references therein ) . however , spectroscopic iron - to - hydrogen [ fe / h ] measurements of galactic cepheids ( romaniello et al .
2005 ) indicate that the visual @xmath1-@xmath9 relation depends on the metal content , but exclude that the metallicity correction follows the linear relation based on the quoted negative empirical @xmath37-value .
the nonlinear behavior suggested by the pulsation models accounts quite well for the observed trend .
more recently , macri et al . ( 2006 , hereinafter [ m06 ] ) have presented multiband @xmath0 observations of a large cepheid sample in two fields of the galaxy ngc 4258 with different mean chemical compositions ( @xmath38 [ o / h]@xmath110.5 dex ) , and derived a metallicity effect of @xmath350.29 mag dex@xmath36 , also excluding any significant variation in the slope of the @xmath1-@xmath5 relations as a function of the cepheid metal abundance .
their findings agree quite well with the results of a previous investigation by kennicutt et al .
( 1998 ) who found @xmath350.27 mag dex@xmath36 using a sizable sample of cepheids in two fields of m 101 with a difference in mean oxygen abundance of 0.7 dex .
a vanishing metallicity effect between galactic and magellanic cepheids was also found by fouque et al .
( 2007 ) . they collected a sample of 59 calibrating galactic cepheids with distances based on robust indicators : hst and hipparcos trigonometric parallaxes , infrared surface brightness , interferometric baade - wesselink methods and cluster main - sequence fitting . by comparing the slopes of galactic optical - nir pl and wesenheit relations with lmc slopes provided by ogle ( udalski et al .
1999a ) and by persson et al .
( 2004 ) they find no significant difference .
accurate trigonometric parallaxes for ten galactic cepheids have been provided by benedict et al .
( 2007 ) using the fine guide sensor available on board the hst .
they estimated new optical and nir pl relations and they found that their slopes are very similar to the slopes of lmc cepheids .
+ however , mottini et al . ( 2008 ) analyzed a sizable sample of high - resolution , high signal - to - noise spectra collected with [email protected] m eso telescope for galactic ( 32 ) and with uves@vlt for magellanic ( 14 in smc and 22 in lmc ) cepheids .
they found , using individual iron measurements and the same distances adopted by fouque et al .
, that the slope of the @xmath39band pl relation does depend on the metal abundance with a confidence level larger than 90% . in order to overcome current controversy between theory and observations
, we undertook a homogeneous analysis of the ngc 4258 cepheids and a detailed comparison of pulsation predictions with magellanic cloud and galactic cepheids . in
2 we present predicted @xmath1-@xmath5 relations , while in 3 we describe the results of the comparison between theory and observations . the correlation between the cepheid metallicity and the ngc 4258 oxygen abundance gradient
is addressed in 4 and the conclusions close the paper .
the fiducial @xmath1-@xmath5 relations adopted by m06 are based on unreddened @xmath40 magnitudes of the lmc cepheids observed by the ogle ii project ( udalski et al . 1999a , hereinafter [ u99 ] ) and updated on the ogle web siteogle ] .
they are @xmath41 @xmath42 @xmath43 where @xmath1 is the pulsation period in days .
these relations are used to form intrinsic period - color ( @xmath1-@xmath44 ) relations which are compared with the measured colors to determine the @xmath45 , @xmath46 , and @xmath47 reddening values for individual cepheids in ngc 4258 .
then , the absolute lmc - relative distance modulus @xmath31 of each variable is derived by averaging the three values=2.38 value given by m06 is a typo . ]
@xmath48 @xmath49 @xmath50 where @xmath51 is the difference between the observed @xmath14 magnitude and the @xmath52 value from equation ( 3 ) , while the @xmath53 , @xmath54 , and @xmath55 ratios are based on the @xmath56 values from table 6 in schlegel et al .
( 1998 ) for @xmath57=3.1 and the cardelli et al .
( 1989 ) extinction law .
however , since equations ( 1)-(3 ) were derived by adopting @xmath58=4.32 , @xmath57=3.24 and @xmath55=1.96 ( see u99 ) , throughout this paper we adopt @xmath59 @xmath60 @xmath61 together with @xmath62 where @xmath63 is the difference between the observed @xmath13 magnitude and the @xmath64 value from equation ( 2 ) .
the quoted approach is equivalent to the classical method of distance determinations based on the reddening free wesenheit functions .
therefore , we use the computed periods and intensity - averaged @xmath65 magnitudes of our fundamental pulsation models with @xmath7=0.004 to 0.04 , listed in table 1 , to derive the predicted period - wesenheit ( @xmath1-@xmath66 relations based on equations ( 4)-(7 ) , i.e. @xmath67 @xmath68 @xmath69 @xmath70 in our earlier pulsation models , the adopted luminosity for a given mass and chemical composition was fixed according to mass - luminosity ( @xmath71 relations based on canonical ( `` can '' ) evolutionary computations ( castellani , chieffi & straniero 1992 ; bono et al . 2000 ;
girardi et al .
2000 ) . afterwards
, additional models have been computed with higher luminosity levels ( `` over '' ) in order to account for a mild convective core overshooting and/or mass loss before or during the cepheid phase .
the overluminous models were constructed by adopting for the chemical compositions representative of galactic and magellanic cepheids the same abundances ( helium , metal ) and mass values adopted for canonical cepheid models .
this approach allowed us ( bono , castellani , & marconi 2000 ; caputo et al . 2005 ) to constrain the impact that the mass - luminosity relation has on pulsation observables .
the quoted assumptions concerning the adopted chemical compositions are supported by theory and observations .
pulsation models constructed by adopting supersolar iron abundance and helium enhanced compositions ( @xmath72 ) are pulsationally unstable ( fiorentino et al .
moreover , empirical evidence based on he abundance of planetary nebulae suggest a very shallow gradient across the galactic disk ( stanghellini et al .
furthermore , chemical evolution models for both the inner and the outer disk indicate similar helium gradients ( hou et al .
note that current estimates of the helium - to - metal enrichment ratio , @xmath73 , are still affected by large uncertainties . in a recent detailed investigation casagrande et al .
( 2007 ) found , using nearby field k - type dwarf stars , @xmath74 .
the observational scenario is also complicated by the fact that we still lack firm empirical constraints on the linearity of the @xmath73 relation , when moving from the metal - poor to the metal - rich regime , and on the universality of this relation ( peimbert et al .
2003 ; tammann et al .
2008 ) . to account for these uncertainties we constructed sets of cepheid models by adopting , at fixed metal content , different helium abundances , therefore
, the intrinsic error on the zero - point of predicted pl and pw relations include this effect ( fiorentino et al .
2007 ) .
the theoretical wesenheit functions of each pulsator depend on the adopted luminosity , therefore , to avoid any assumption on the @xmath75 relation , we calculated for each model the difference log@xmath76 between the adopted luminosity and the canonical value provided by the bono et al .
( 2000 ) relation @xmath77 where mass and luminosity values are in solar units . then , by a linear interpolation through all the fundamental models with period @xmath78 4 - 80 days , @xmath7=0.004 - 0.04 , without distinction of the helium content at fixed @xmath7 , we derive the linear period - wesenheit ( @xmath1-@xmath2 ) relations listed in table 2 . according to these predicted @xmath1-@xmath2 relations , one can determine the true distance modulus @xmath29 of individual cepheids with known metal content , once the log@xmath76 ratio is fixed .
in this context , it is worth mentioning that the occurrence either of a mild convective core overshooting during hydrogen burning phases or mass - loss before and/or during the pulsation phases yields positive log@xmath76 values . as a consequence , the @xmath1-@xmath2 relations for @xmath5=@xmath79
provide the _ maximum _ value of the cepheid distance .
moreover , we draw attention on the evidence that the metallicity effect on the predicted @xmath1-@xmath2 relations depends on the adopted wesenheit function ( see also caputo , marconi & musella 2000 ; f02 ; f07 ) . in particular , the metallicity dependence of the @xmath1-@xmath80 relation is weak and shows the opposite sign when compared with the other optical @xmath1-@xmath2 relations .
this is warning against the average of the various @xmath29 values based on different @xmath1-@xmath2 relations and , at the same time , provides a plain method to estimate the cepheid metal content ( see below ) . by assuming that the pulsation models listed in table 1 are actual cepheids located at the same distance and with the same reddening , but with different chemical abundances
, we can derive from equations ( 1)-(3 ) their lmc - relative apparent distance moduli @xmath81 , @xmath82 and @xmath83 .
then , by adopting-@xmath44 relations given by equations ( 1)-(3 ) to estimate the reddening values is equivalent to take the differences between two lmc - relative apparent distance moduli and indeed @xmath84 . ]
@xmath85=@xmath45 , @xmath86=@xmath47 and @xmath87=@xmath46 , we use equations ( 4)-(7 ) to determine the four lmc - relative intrinsic values @xmath88 , @xmath89 , @xmath90 , and @xmath91 .
= 0.4 - 1.9 versus the chemical composition , as listed in table 4 .
the lines display the least - squares fit to the data.,width=302 ] by averaging the results over the entire period range ( @xmath78 4 - 80 days ) , without selections between short and long period pulsators , we get the lmc - relative @xmath31 values at @xmath92 listed in table 3 .
as already suggested by the predicted @xmath1-@xmath2 relations given in table 2 , we find that the metallicity effect is _ not constant _ among the different approaches to estimate the lmc - relative distance moduli . on average , we derive @xmath930.59 mag dex@xmath36 , @xmath940.12 mag dex@xmath36 , and @xmath950.35 mag dex@xmath36 , whereas the metallicity dependence of @xmath88 is smaller and seems to depend on the adopted metallicity range .
we find @xmath96 + 0.11 mag dex@xmath36 for @xmath97 0.02 and @xmath980.15 for @xmath19 0.02 , while over the entire metallicity range @xmath7=0.004 - 0.04 , we find @xmath990.03 mag dex@xmath36 .
in summary , the theoretical results suggest that , if the lmc - based @xmath100 , @xmath101 and @xmath102 relations are used to get the distance to cepheids with metal content significantly different from the lmc abundance , then the values of the various @xmath31 formulations _ should not be averaged , but individually considered in order to keep the information provided by their different metal dependence_. note that if the @xmath88 , @xmath89 , and @xmath90 values were averaged to a mean value @xmath103 , the ensuing mean metallicity effect would be @xmath1040.15 mag dex@xmath36 .
this value _ can not _
be used to correct distance estimates based on @xmath3 magnitudes since it might introduce a systematic error up to @xmath105 0.2 mag according to the metallicity range covered by the cepheids . a further relevant result of the present study is the predicted metallicity effect on the _ internal differences _ among the different lmc - relative distance moduli . in particular , these differences revealed to be almost independent of the adopted @xmath75 relation .
data plotted in fig .
1 and listed in table 4 show that all the differences depend on the pulsator chemical composition .
the most metal - sensitive are the @xmath106 , the @xmath107 and the @xmath108 .
we have already discussed in f07 that these differences provide a robust method to estimate the cepheid metal content , since they are independent of both distance and reddening .
we apply the same procedure described above to the smc cepheids collected by the ogle - ii microlensing survey ( udalski et al .
for the sake of uniformity with the ngc 4258 variables ( see later ) , we use fundamental pulsators with @xmath21 6 days ( @xmath11 200 variables ) .
we derive the mean @xmath31 values plotted in fig . 2 and summarized in table 5 , where the results for the lmc cepheids ( u99 ) are also given to validate the current approach .
even though data plotted in fig . 2 present a large scatter , the four lmc - relative @xmath31 formulations do provide different results .
in particular , the @xmath88 gives the shortest lmc - relative distance modulus , in agreement with the theoretical predictions . by using the predicted metallicity effects given in the last row of table 5 and the cepheid spectroscopic measurements [ fe / h]@xmath109 dex and [ fe / h]@xmath110 dex ( see luck et al .
1998 ; romaniello et al .
2005 ) and by assuming @xmath111$]=@xmath38[fe / h ] , we find that the measured @xmath91 , @xmath89 , and @xmath90 values should be _ decreased _ by @xmath11 0.21 , 0.04 and 0.12 mag , whereas the @xmath88 value should be _ increased _ by @xmath11 0.04 mag .
eventually , the discrepancy among the four @xmath31 values is mitigated and the metallicity - corrected results yield that the lmc - relative distance modulus of the smc cepheids is @xmath11 0.55 mag , in close agreement with the difference of 0.50 mag determined from eclipsing binaries in the smc ( hilditch , howarth & harries 2005 ) and in the lmc ( guinan , ribas & fitzpatrick 2004 ) .
the differences @xmath112 among the four @xmath31 values are summarized in table 6 and plotted in fig .
the observed variations between smc and lmc cepheids agree quite well with pulsation predictions for @xmath38[z / x]=@xmath38[fe / h][email protected] , as listed in the last row of table 6 .
in addition , the straight comparison between the observed @xmath112 differences and the predicted values listed in table 4 gives [email protected] and [email protected] , which are consistent with the spectroscopic iron measurements once we assume [ fe / h]=[@xmath115 and @xmath116=0.024 ( grevesse et al . 1996 ) .
we are aware that the solar chemical composition is under revision and that the recent analysis by asplund et al .
( 2004 ) has decreased the solar chemical abundances by roughly a factor of two , yielding @xmath116=0.0165 .
however , we adopted the grevesse et al .
( 1996 ) solar abundances , since they are consistent with the model atmospheres ( castelli , gratton & kurucz 1997a , b ) we use to transform theoretical predictions into the observational plane .
the revised abundances are still debated due to the inconsistency with helioseismic results ( bahcall et al .
2005 ; guzik et al .
2005 ) . following the referee s suggestion ,
it is also worth noting that evolutionary and pulsation models are constructed by adopting the global metallicity ( [ m / h ] ) that is a function of both iron and @xmath117-element abundances .
however , evolutionary prescriptions by salaris et al .
( 1993 ) to estimate the global metallicity were derived using the old solar mixture and we still lack a new relation based on the new solar abundances .
moreover , salaris & weiss ( 1998 ) found that at solar and super - solar metallicities , the metals affects the evolution .
this means that scaled - solar abundances can not be used to replace the @xmath117-enhanced ones of the same total metallicity .
oxygen is an @xmath117-element and intermediate - mass stars with solar abundance typically present solar oxygen abundances ( gratton et al . 2004 ) .
empirical evidence indicates that field lmc giants present a lower @xmath117-enhancement when compared with the galactic ones ( luck et al .
1998 ; hill et al .
2000 ) . however , larger samples are required to constrain the trend in the metal - poor regime ( hill 2004 ; venn et al .
it is worth noting that the new solar abundances imply a change in the nebular oxygen abundance of external galaxies ( zero - point and effective temperature of the ionizing stars ) . to our knowledge
we still lack nebular oxygen abundances in external galaxies accounting for this effect .
@xmath118[fe / h]@xmath119 versus [ fe / h]@xmath119 for galactic cepheids .
the arrow shows the adopted average difference .
same as the top , but for [ fe / h]@xmath118[fe / h]@xmath120 , versus [ fe / h]@xmath120 .
, width=302 ] 6 days versus the three different sets of [ fe / h ] measurements , [ fe / h]@xmath121 : solid circles , [ fe / h]@xmath120 : open circles , and [ fe / h]@xmath119 : open triangles in the andrievsky metallicity scale .
the mean values for smc and lmc variables ( crosses ) are also plotted .
the solid lines are the least square fits to the data , while the dashed lines display the dispersion around the fit .
see text for more details.,width=302 ] in order to verify whether this consistency between the pulsation predictions and the cepheid observed properties also holds up at larger metal abundances , we use the milky way variables with measured iron - to - hydrogen ratios . given the current discrepancy among abundance determinations by different authors , we consider three different sets of measurements : the [ fe / h]@xmath121 values provided by andrievsky and collaborators ( andrievsky et al .
2002a , b , c ; andrievsky et al .
2004 ; luck et al .
2003 ; kovtyukh , wallerstein & andrievsky 2005 ; luck , kovtyukh & andrievsky 2006 ) , the [ fe / h]@xmath119 values by romaniello et al .
( 2005 ) together with lemasle et al .
( 2007 ) , and the [ fe / h]@xmath120 values by yong et al .
( 2006 ) together with fry & carney ( 1997 ) . following yong et al .
( 2006 ) the iron abundances provided by fry & carney were decreased by @xmath270.11 dex .
the arrows plotted in fig . 4 show that the [ fe / h]@xmath120 and the [ fe / h]@xmath119 were normalized to the andrievsky metallicity scale by adding 0.19 and 0.08 dex , respectively .
we also use the @xmath0 magnitudes compiled by berdnikov , dambis & vozyakova ( 2000 ) .
we select the variables with @xmath21 6 days , although the inclusion of first overtone pulsators has no dramatic effects on the _ differences _ among the @xmath31 values .
in fact , adopting log@xmath122=log@xmath123 , one easily derives that the offsets ( first overtone minus fundamental ) are @xmath980.08 ( @xmath106 ) , @xmath270.06 ( @xmath107 ) , @xmath270.03 ( @xmath124 ) , @xmath270.02 ( @xmath125 ) , @xmath270.05 ( @xmath108 ) , and @xmath270.03 mag ( @xmath126 ) .
the results plotted in fig . 5 show that magellanic and galactic cepheids follow reasonably well defined common relations over the metallicity range [ fe / h][email protected] to + 0.3 , with the only exception of ck pup , hq car and tx cyg ( at [ fe / h][email protected] , @xmath270.22 and + 0.20 , respectively ) . eventually , the linear regression through the magellanic and galactic data yields the empirical @xmath112-[fe / h ] calibrations : @xmath127\;\;\;\;\sigma=0.25\ ] ] @xmath128\;\;\;\;\sigma=0.18\ ] ] @xmath129\;\;\;\;\sigma=0.10\ ] ] @xmath130\;\;\;\;\sigma=0.10\ ] ] @xmath131\;\;\;\;\sigma=0.18\ ] ] @xmath132\;\;\;\;\sigma=0.10\ ] ] where the error in parentheses is the error on the coefficients and the sigma gives the sum in quadrature of the uncertainties affecting both the zero - point and the slope of the fit .
these relations are drawn as solid lines in fig . 5 , while the dashed lines display the one @xmath133 statistical uncertainty .
it is worth emphasizing that the _ observed _
@xmath112-[fe / h ] relations based on magellanic and galactic cepheids agree well with the _ theoretical _ ones presented in table 4 , if we assume [ fe / h]=[z / x ] and we adopt @xmath116=0.024 ( grevesse et al . 1996 ) , namely [ fe / h]=log@xmath134)+1.62 .
note that an even better agreement is found if we account for the measured overabundance of @xmath117-elements for subsolar [ fe / h ] ratios , as determined by spectroscopic measurements .
( see , e.g. , fig .
18 in yong et al .
this issue will be discussed in a forthcoming paper .
6 days versus the deprojected radial distance @xmath135 normalized to @xmath136=7.92 .
cepheids located either in the inner or in the outer field are filled and open circles , respectively.,width=302 ] 6 days versus the deprojected radial distance @xmath137 .
symbols are the same as in fig .
6.,width=302 ] the cepheids observed in ngc 4258 belong to two different fields located at different galactocentric distances and whose mean offsets in arcseconds from the nucleus are @xmath138150 ( inner field ) and @xmath139400 ( outer field ) in the east - west direction , while they are @xmath139130 ( inner field ) and @xmath138400 ( outer field ) in the north - south direction .
we apply the same approach already adopted for magellanic and galactic variables to the ngc 4258 cepheids with a variability index @xmath1402 ( `` restricted '' sample in m06 ) , @xmath21 6 days adopted by m06 are 6 and 12 days for the cepheids in the outer and in the inner field , respectively .
we adopt the same period cuts for both fields in order to have homogeneous samples . ] and errors in the mean @xmath0 magnitudes less than 0.05 mag .
the derived lmc - relative distance moduli are plotted in fig .
6 versus the cepheid deprojected galactocentric distance @xmath141( ) normalized to the isophotal radius @xmath136=7.92 , as given values given by m06 adopt @xmath136=7.76 .
however , for consistency with the zaritsky et al .
( 1994 , hereinafter [ z94 ] ) abundance gradient , we normalized them to @xmath136=7.92 .
] by m06 .
data plotted in fig .
6 suggest a correlation between the cepheid distance modulus and its radial distance , with the outer field cepheids yielding larger distance moduli by about 0.2 mag with respect to those in the inner field .
obviously , such a correlation turns into a chemical abundance dependence of the distance modulus if a metallicity gradient is adopted for the cepheids .
as a fact , using for each individual variable the za94 relation based on oxygen abundance measurements of h ii regions @xmath142 where @xmath136 is the isophotal radius equal to 7.92 , the metallicity effect on the the four lmc - relative distance moduli turns out to be consistent with the m06 value , i.e. @xmath1430.29 mag dex@xmath36 ( see fig .
7 ) . versus the oxygen - to - hydrogen ratio [ o / h ] based on the galactic gradient given by za94 and by adopting
log(o / h)@xmath1443.13 .
the crosses mark the results for smc and lmc variables .
solid and dashed lines are the same as in fig . 5 , and have been plotted assuming [ fe / h]=[o / h].,width=302 ] however , we show in fig . 8 that if the individual differences @xmath112 are taken into account , then no clear radial dependence is found , with the cepheids in both fields yielding quite similar mean @xmath112 values .
data plotted in fig .
9 show the differences @xmath112 for the ngc 4258 cepheids versus the oxygen abundance [ o / h]@xmath145=log(o / h)@xmath146log(o / h)@xmath147 , following equation ( 15 ) and by adopting the solar value log(o / h)@xmath1443.13 ( grevesse et al . 1996 ) . in order to make an easy comparison , in this figure we also plotted the mean @xmath112 values for smc and lmc cepheids ( see table 6 ) for [ o / h][email protected]@xmath150.08 dex and for @[email protected] dex ( ferrarese et al . 2000 ) .
the least square fits to the data ( solid lines ) and their dispersions ( dashed lines ) showed in fig . 5
are also plotted in fig . 9 by assuming that the oxygen abundance is a very robust proxy of the iron abundance ( i.e. [ fe / h]=[o / h ] ) .
note that this assumption is fully justified by spectroscopic measurements which yield [ o / fe][email protected] dex over the range [ fe / h][email protected] to + 0.30 dex ( see e.g. luck et al .
moreover , recent spectroscopic measurements based on high resolution , high signal - to - noise spectra of 30 galactic cepheids ( lemasle et al .
2007 ) indicate that oxygen and other @xmath117-elements present radial gradients very similar to the iron gradient .
this means that oxygen is a good proxy of the iron content across the galactic disk .
moreover and even more importantly , empirical evidence indicates that oxygen nebular abundances agree with absorption line abundances ( hill 2004 ) .
although the oxygen abundance of the ngc 4258 cepheids based on the za94 oxygen abundance gradient , is within the range spanned by magellanic and milky way cepheids , the observed @xmath112 values of several variables in the inner field deviate from the `` empirical '' @xmath112-[o / h ] relations provided by magellanic and galactic variables .
this evidence indicates that cepheids in ngc 4258 might have a metal content that is significantly lower than the oxygen abundance based on their radial distance . to make clear this feature , we select the inner field cepheids with @xmath148 ( sample a ) and the outer field cepheids with @xmath149 ( sample b ) for which the mean oxygen abundance suggested by the za94 gradient is [ o / h]@[email protected] and @[email protected] dex , respectively .
we show in table 7 and table 8 that the two samples have different lmc - relative distance moduli but nearly identical mean @xmath112 values , at odds with the behavior of magellanic and metal - rich milky way variables , as listed in the same table 8 . in conclusion
, the observed @xmath112 values would suggest an average lmc - like oxygen abundance [ o / h]@xmath980.4 dex for all the ngc 4258 cepheids .
this result agrees with the za94-based mean value of the outer field , whereas for the inner field the oxygen content provided by the radial distance appears to be 1.3 times larger than the solar value .
we do not found any reason to distrust this intriguing result , since the selection criteria adopted by m06 are very robust , and indeed , we only use objects with errors in mean @xmath150 magnitudes smaller than 0.05 mag .
the adopted @xmath151=6 days should avoid contamination by first overtone cepheids , although the effects of period uncertainties on the @xmath112 differences are quite small and first overtone pulsators should give smaller @xmath112 values than fundamental pulsators with the same period .
however the selection of cepheids with @xmath16 10 days is more difficult , in particular in the inner field ( see m06 ) .
therefore , we performed a new test by removing these short period variables from the sample and we found that the new results are almost identical to those listed in table 7 and table 8 . eventually , it seems plausible to suspect that either the za94 oxygen gradient requires a revision , or the galactic location can not be used as a reliable metallicity parameter for individual cepheids , or a combination of the above effects .
the solid line shows the za94 relation , while the dashed line is based on equation ( 16 ) , and it was estimated by neglecting the h ii region marked by the arrow .
bottom positions of the h ii regions observed by za94 and by ok93 in comparison with the ngc 4258 inner and outer fields.,width=302 ]
in the top panel of fig . 10 , we plot za94 and previous oxygen abundance measurements by oey & kennicutt ( 1993 , hereinafter [ ok93 ] ) versus the fractional isophotal radius with @xmath136=7.92 .
it is quite clear that the variations in abundance among external regions with @xmath1521.2 are greater than the abundance uncertainties and that the lowest value ( 12+log(o / h)=8.4 ) measured by za94 at @xmath137=1.18 strongly affects the slope of equation ( 15 ) . indeed , if we neglect this value the linear fit to all the za94 and the ok93 measurements yields ( dashed line ) a significantly flatter gradient , namely @xmath153 the bottom panel of fig .
10 shows that all the cepheids in the ngc 4258 inner field are located close to h ii regions where , as described by equation ( 16 ) , the oxygen abundance has the almost constant solar value , i.e. 12+log(o / h)[email protected] . on the other hand ,
the variables in the outer field are distant from any observed h ii region , and only marginally close to the h ii region underabundant in oxygen ( 12+log(o / h)=8.4 ) .
even though , we assume a tight star - by - star correlation between oxygen abundance and radial distance , we find that by using the equation ( 16 ) to estimate the individual abundances of ngc 4258 cepheids would yield a mean abundance difference of only @xmath11 0.15 dex between the inner and the outer field .
this would imply that the ngc 4258 is not the right laboratory to constrain the metallicity effect on the lmc - relative distance moduli . on the other hand ,
the assumption of a lower oxygen abundance for the outer field variables would imply that the galactic gradient becomes significantly steeper when moving from the western to the eastern direction .
the solid line is the za94 oxygen gradient , while the dashed line is the best fit line given by equation ( 17 ) .
bottom positions of the h ii regions observed by za94 , by ok93 and by d00 .
, width=302 ] although the occurrence of spatial asymmetric metallicity gradients can not be ruled out ( see , e.g. , kennicutt & garnett 1996 ) , we draw the attention on recent observations specifically addressed to study extragalactic h ii regions which were expected to be metal - rich . as a whole , the new measurements ( see , e.g. , kennicutt , bresolin & garnett 2003 ; bresolin , garnett & kennicutt 2004 ; bresolin et al .
2005 ) yield a significant decrease in the nebular oxygen abundances of regions more metal - rich than lmc , and marginally affect the abundances of metal - poor ones .
therefore , the galactic gradients become significantly shallower than those estimated by previous determinations . in this context
it is worth mentioning that diaz et al .
( 2000 , hereinafter [ d00 ] ) , by performing a more detailed analysis of optical and near - infrared observations of several ngc 4258 regions previously observed by za94 and by ok93 , measured oxygen abundances that are on average a factor of two lower .
data plotted in fig .
11 disclose that , by using the new and more accurate d00 abundances , the ngc 4258 abundance gradient can be ( dashed line ) @xmath154 which implies a lmc - like mean oxygen abundance for both the inner ( sample a : [ o / h][email protected] ) and the outer field cepheids ( sample b : [ o / h][email protected] ) .
this would also imply a reasonable agreement with the predicted correlation between the metal abundance and the @xmath112 values .
in the above sections , we have shown that both the comparison with pulsation models and the most recent h ii abundance measurements suggest a rather constant , lmc - like metal content for the cepheids observed in the two fields of ngc 4258 .
this finding , once confirmed , would prevent any reliable differential determination of the @xmath1-@xmath5 metallicity dependence . as a consequence , the observed difference of @xmath11 0.20 mag in distance modulus between outer and inner field variables might be caused by other observational effects rather than a difference in metal abundance .
-based ) and trgb distances from sakai et al .
( 2004 , [ s04 ] ) , as a function of the nebular oxygen abundance in the za94 scale .
the arrow marks the lmc , while filled and open symbols display the outer and the inner fields in m 101 and in ngc 4258 , respectively .
the solid line was drawn using the s04 value @xmath350.25 mag dex@xmath36 .
same as the top , but with the trgb distances according to rizzi et al .
( 2007 , [ r07 ] ) .
the dashed line was drawn using the value @xmath37=+0.05 mag dex@xmath36 of the predicted @xmath1-@xmath80 relation.,width=302 ] we are facing the evidence that the ngc 4258 results presented by m06 agree quite well with the metallicity effect @xmath350.24 mag dex@xmath36 determined by kennicutt et al .
( 1998 ) from cepheid observations in two fields of m 101 . however , it is worthy mentioning that macri et al .
( 2001 ) brought forward the hypothesis that blended cepheids could be responsible for a large fraction of the difference in distance modulus between the outer and the inner field in m 101 .
we recall that blended cepheids , which are mainly expected in the crowded inner galactic fields , appear brighter than they really are and that their distances are systematically underestimated by @xmath11 6 - 9% ( see mochejska et al .
2000 ) , leading to @xmath29 underestimated by approximately 0.1 - 0.2 mag . moreover , the @xmath350.25 mag dex@xmath36 provided by sakai et al .
( 2004 , hereinafter [ s04 ] ) from the comparison of distances based on cepheid variables and on the tip of the red giant branch ( trgb ) has been recently questioned by rizzi et al .
( 2007 , hereinafter [ r07 ] ) . by adopting the distance determinations listed in table 9 , we plot in the top panel of fig .
12 the s04 difference between the cepheid , based on the lmc @xmath1-@xmath9 and @xmath1-@xmath10 relations , and the trgb distances versus the za94 nebular oxygen abundances . the data(lmc)=18.50 mag . to the s04 original distances we added the current cepheid distances
to ngc 4258 and to smc , and the wlm cepheid distance by pietrzynski et al .
( 2007 ) . ] clearly indicate a trend with metallicity , with the cepheid residual distance modulus decreasing with increasing oxygen abundance and leading to @xmath350.25 mag dex@xmath36 ( solid line ) .
however , data plotted in the bottom panel of this figure show that distance determinations provided by r07 using the trgb method agree within the errors , with cepheid distances , _ with the exception of m 101 and ngc 4258 inner fields _ , over the entire metallicity range .
note that in this case we neglected the metallicity correction .
the agreement becomes even better if we adopt the @xmath37=+0.05 mag dex@xmath36 value of the predicted @xmath1-@xmath80 relation ( dashed line ) .
-@xmath80 relation for galactic cepheids with @xmath4 trigonometric parallaxes .
this relation is used to derive the true distance modulus of cepheids in the lmc ( middle ) and in the outer field of ngc 4258 ( top ) , by neglecting the metallicity correction.,width=302 ] as a final test of the metallicity effect on cepheid distances based on @xmath3 magnitudes we adopted the galactic cepheids with @xmath4 trigonometric parallaxes ( benedict et al .
2007 ) . from the absolute @xmath80 functions of these variables
, we find that they obey to the @xmath1-@xmath80 relation wvi=@xmath157log@xmath1 , as shown by the solid line in the bottom panel of fig . 13 . by using this relation for the cepheids in the lmc and in the outer field of ngc 4258 _ by neglecting the metallicity correction _ , we derive @xmath158(lmc)[email protected] mag and @xmath158(ngc4258)[email protected] mag , which are both only slightly _ larger _ than @xmath29(lmc)[email protected] mag determined by guinan , ribas & fitzpatrick ( 2004 ) from eclipsing binaries ( eb ) and @xmath29(ngc4258)[email protected] mag based on the maser geometric distance measured by herrnstein et al .
but , the variables in the lmc and in the outer field of ngc 4258 _ do _ have a lower metal abundance by @xmath159 dex than the galactic variables and the adoption of the m06 value @xmath350.29 mag dex@xmath36 would _ increase _ the distance moduli to @xmath158(lmc)[email protected] mag and @xmath158(ngc4258)[email protected] mag , causing metallicity corrected cepheid distances which are larger by about 0.15 - 0.17 mag than eb and maser distance determinations .
it goes without saying that by adopting @xmath37=+0.05 mag dex@xmath36 from the predicted @xmath1-@xmath80 relation would further improve the consistency between the cepheid distances and the quoted eb and maser - based determinations .
the theoretical pulsation models suggest that both the sign and the amount of the metallicity dependence of the @xmath1-@xmath2 relations depend on the chosen passbands . in particular , for distances based on @xmath0 magnitudes , the predicted metallicity effect on @xmath29 varies from @xmath1430.61 mag dex@xmath36 ( @xmath1-@xmath160 relation ) to @xmath161 + 0.05 mag dex@xmath36 ( @xmath1-@xmath80 relation ) over the range @xmath7=0.004 - 0.04 .
these predictions are supported by the comparison of smc and milky way cepheids with lmc variables .
accurate @xmath0 photometry of cepheids in two fields of ngc 4258 leads to a systematic difference in the true distance moduli of @xmath11 + 0.2 mag between the outer and the inner field . adopting for individual cepheids the oxygen abundance given by their galactocentric distance and the abundance gradient of z94 ,
one derives a metallicity effect @xmath1430.29 mag dex@xmath36 which is consistent with an earlier @xmath1430.24 mag dex@xmath36 found by kennicutt et al .
( 1998 ) from cepheids in two fields of m 101 .
the comparison with pulsation models as well as with magellanic and galactic variables , indicates a rather small abundance difference between the ngc 4258 inner and outer fields , in agreement with recent nebular oxygen abundances by diaz et al .
( 2000 ) . as a whole
, the two `` direct '' determinations of the metallicity effect which provide negative metallicity dependence @xmath1430.24 and @xmath270.29 mag dex@xmath36 appear undermined by the lack of a significant difference in metal abundance ( ngc 4258 ) or by the possible occurrence of blended cepheids in the inner field ( m 101 ) .
the comparison of @xmath3-based cepheid distances with independent determinations based on the trgb ( external galaxies ) , @xmath4 trigonometric parallaxes ( milky way cepheids ) , eclipsing binaries ( large magellanic cloud ) and water maser ( ngc 4258 ) does not support the negative empirical @xmath37 values .
current results seem to favor the predicted value @xmath161 + 0.05 mag dex@xmath36 .
it is a pleasure to thank a. diaz and l. rizzi for useful discussions on nebular abundance measurements and on distance determinations to external galaxies based on the trgb method .
we also thank n. patat for several insights on the projected distances , f. thevenin and a. walker for a detailed reading of an early draft of this paper .
we acknowledge an anonymous referee for his / her suggestions that improved the content and the readability of the manuscript .
asplund , m. , grevesse , n. , sauval , a. j. , allende prieto , c. , kiselman , d. 2004 , a&a , 417 , 751 andrievsky , s. m. , kovtyukh , v. v. , luck , r. e. 2002a , a&a , 381 , 32 andrievsky , s. m. , bersier , d. , kovtyukh , v. v. 2002b , a&a , 384 , 140 andrievsky , s. m. , kovtyukh , v. v. , luck , r. e. 2002c , a&a , 392 , 491 andrievsky , s. m. , luck , r. e. , martin , p. , lapine , j. r. d. 2004 , a&a , 413 , 159 bahcall , j. n. , serenelli , a. m. , basu , s. , 2005 , apj , 621 , 85 benedict , g. f. , et al .
2007 , aj , 133 , 181 berdnikov , l. n. , dambis , a. k. , vozyakova , o. v. 2000 , a&as , 143 , 211 bono , g. , caputo , f. , cassisi , s. , et al .
2000 , apj , 543 , 955 bresolin , f. ; garnett , d. r. ; kennicutt , r. c. , jr . , 2004 , apj , 615 , 228 bresolin , f. , schaerer , d. , gonzlez delgado , r. m. , stasinska , g. , 2005 , a&a , 441 , 981 caputo , f. 2008 , mem .
sait , in press caputo , f. , marconi , m. , musella , i. 2000 , a&a , 354 , 610 caputo , f. ; bono , g. ; fiorentino , g. ; marconi , m. ; musella , i. , 2005 , apj , 629 , 1021 cardelli , j. a. , clayton , g. c. , mathis , j. s. , 1989 , apj , 345 , 245 casagrande , l. ; flynn , c. , portinari , l. ; girardi , l. ; jimenez , r. 2007 , mnras , 382 , 1516 castellani , v. , chieffi , a. , & straniero , o. 1992 , apjs , 78 , 517 castelli , f. , gratton , r. g. , & kurucz , r. l. 1997a , a&a , 318 , 841 castelli , f. , gratton , r. g. , & kurucz , r. l. 1997b , a&a , 324 , 432 daz , a. i. , castellanos , m. , terlevich , e. , luisa garca - vargas , m. 2000 , mnras , 318 , 462 [ d00 ] ferrarese , l. , et al .
2000 , apjs , 128 , 431 fiorentino , g. , caputo , f. , marconi , m. , musella , i. 2002 , apj , 576 , 402 [ f02 ] fiorentino , g. , marconi , m. , musella , i. , caputo , f. 2007 , a&a , 476 , 863 [ f07 ] fouqu , p. , et al .
2007 , a&a , 476 , 73 freedman , w. l. , hughes , s. m. , madore , b. f. , et al .
1994 , apj , 427 , 628 fry , a. m. , & carney , b. w. 1997 , aj , 113 , 1073 girardi , l. , bressan , a. , bertelli , g. , chiosi , c. 2000 , a&as , 141 , 371 gratton , r. , sneden , c. , & carretta , e. 2004 , ara&a , 42 , 385 grevesse , n. , noels , a. , & sauval , a. j. , 1996 , aspc , 99 , 117 groenewegen , m. a. t. , romaniello , m. , primas , f. , mottini , m. 2004 , a&a , 420 , 655 guinan , e. f. , ribas , i. , fitzpatrick , e. l. 2004 , in variable stars in the local group , iau coll .
d. w. kurtz and k. r. pollard ( san francisco , asp ) , 363 guzik , j. a. , watson , l. s. , cox , a. n. , 2005 , apj , 627 , 1049 herrnstein , j. r. , et al .
1999 , natur , 400 , 539 hilditch , r. w. , howarth , i. d. , harries , t. j. , 2005 , mnras , 357 , 304 hill , v. , francois , p. , spite , m. , primas , f. , spite , f. 2000 , a&a , 364 , l19 hill , v. 2004 , in origin and evolution of the elements ed .
a. mcwilliam and m. rauch ( cambridge : cambridge univ .
press ) , 203 hou , j. l. , prantzos , n. , boissier , s. 2000 , a&a , 362 , 921 kanbur , s. m. , ngeow , c. , nikolaev , s. , tanvir , n. r. , hendry , m. a. 2003 , a&a , 411 , 361 kennicutt , r. c. , jr .
, bresolin , f. , garnett , d. r. 2003 , pasp , 115 , 928 kennicutt , r. c. , jr . , garnett , d. r. 1996 , apj , 456 , 504 kennicutt , r. c. , jr . ,
stetson , p. b. , saha , a. 1998 , apj , 498 , 181 kovtyukh , v. v. , wallerstein , g. , andrievsky , s. m. , 2005 , pasp , 117 , 1173 lemasle , b. , francois , p. , bono , g. , mottini , m. , primas , f. , romaniello , m. 2007 , a&a , 467 , 283 luck , r. e. , gieren , w. p. , andrievsky , s. m. , kovtyukh , v. v. , fouque , p. , p ont , f. , kienzle , f. 2003 , a&a , 401 , 939 luck , r. e. , kovtyukh , v. v. , andrievsky , s. m. 2006 , aj , 132 , 902 luck , r. e. , moffett , t. j. ; barnes , t. g. iii , gieren , w. p. 1998 , aj , 115 , 605 macri , l. m. , stanek , k. z. , bersier , d. , greenhill , l. j. , reid , m. j. 2006 , apj , 652 , 1133 [ m06 ] macri , l. m. et al .
2001 , apj , 549 , 721 marconi , m. , musella , i. , fiorentino , g. , 2005 , apj , 632 , 590 [ m05 ] mochejska , b. j. , macri , l. m. , sasselov , d. d. , stanek , k. z. 2000 , aj , 120 , 810 mottini , m. , et al .
2008 , a&a , accepted ngeow , c .- c .
, & kanbur , s.m .
2004 , mnras , 349 , 1130 oey , m. s. , kennicutt , r. c. jr .
1993 , apj , 411 , 137 [ ok93 ] peimbert , m. ; peimbert , a. ; luridiana , v. ; ruiz , m. t. 2003 , in star formation through time , ed .
e. perez , r. m. gonzalez delgado , g. tenorio - tagle ( san francisco , asp ) , 81 persson , s. e. , madore , b. f. , krzeminski , w. , freedman , w. l. , roth , m. , murphy , d. c. 2004 , aj , 128 , 2239 pietrzynski , g. , et al .
2007 , aj , 134 , 594 rizzi , l. , tully , r. b. , makarov , d. , makarova , l. , dolphin , a. e. , sakai , s. , shaya , e. j. 2007 , apj , 661 , 815 [ r07 ] romaniello , m. , primas , f. , mottini , et al .
2005 , a&a , 429 , 37 saha , a. , labhardt , l. , schwengeler , h. , macchetto , f. d. , panagia , n. , sandage , a. , & tammann , g. a. 1994 , apj , 425 , 14 sakai , s. , ferrarese , l. , kennicutt , r. c. , jr . ,
saha , a. 2004 , apj , 608 , 42 [ s04 ] salaris , m. , chieffi , a. , & straniero , o. 1993 , apj , 414 , 580 salaris , m. ; & weiss , a. 1998 , a&a , 335 , 943 sandage , a. , tammann , g. a. , & reindl , b. 2004 , a&a , 424 , 43 sasselov , d.d . , beaulieu , j. p. , renault , c. 1997 , a&a , 324 , 471 schlegel , d. j. , finkbeiner , d. p. , davis , m. , 1998 , apj , 500 , 525 stanghellini , l. , guerrero , m .- a . , cunha , ka . ,
manchado , a. , villaver , e. 2006 , apj , 651 , 898 storm , j. , carney , b. w. , gieren , w. p. , fouqu , p. , latham , d. w. , fry , a. m. 2004 , a&a , 415 , 531 tammann , g.a .
, sandage , a. , reindl , b. 2003 , a&a , 404 , 423 tammann , g.a . ,
sandage , a. , reindl , b. 2008 , apj , accepted , arxiv:0712.2346 udalski , a. , soszynski , i. , szymanski , m. et al .
1999a , aca , 49 , 223 [ u99 ] udalski , a. , soszynski , i. , szymanski , m. , kubiak , m. , pietrzynski , g. , wozniak , p. , zebrun , k. 1999b , aca , 49 , 437 venn , k. a. , irwin , m. , shetrone , m. d. , tout , c. a. , hill , v. , tolstoy , e. 2004 , aj , 128 , 1177 yong , d. , carney , b. w. , teixera de almeida , m. l. , pohl , b. l. 2006 , aj , 131 , 2256 zaritsky , d. , kennicutt , r. c. , jr . , huchra , j. 1994 , apj , 420 , 87 [ z94 ] llll 0.004 & 0.25 & 3.5 - 11.0 & can , over + 0.008 & 0.25 & 3.5 - 11.0 & can , over + 0.01 & 0.26 & 5.0 - 11.0 & can + 0.02 & 0.25 , 0.26 , 0.28 , 0.31 & 5.0 - 11.0 & can , over + 0.03 & 0.275 , 0,31 , 0,335 & 5.0 - 11.0 & can + 0.04 & 0.25 , 0.29 , 0.33 & 5.0 - 11.0 & can + + @xmath160&@xmath1620.09&@xmath1630.03&@xmath1640.03 & @xmath1650.04 + @xmath80&@xmath1660.13 & @xmath1670.05 & @xmath1680.03 & @xmath1690.04 + @xmath170&@xmath1710.09 & @xmath1720.02&@xmath1730.03 & @xmath1740.04 + @xmath175&@xmath1760.06 & @xmath1770.02&@xmath1780.03 & @xmath1790.04 + lcccccc 0.004 & 0.250 & @xmath272.271&@[email protected]&@[email protected]&@[email protected]&@[email protected] + 0.008 & 0.250 & @xmath271.967&@[email protected]&@[email protected]&@[email protected]&@[email protected] + 0.010 & 0.260 & @xmath271.863&@[email protected]&@[email protected]&@[email protected]&@[email protected] + 0.020 & 0.250 & @xmath271.562&@[email protected]&@[email protected]&@[email protected]&@[email protected] + 0.020 & 0.260 & @xmath271.556&@[email protected]&@[email protected]&@[email protected]&@[email protected] + 0.020 & 0.280 & @xmath271.544&@[email protected]&@[email protected]&@[email protected]&@[email protected] + 0.020 & 0.310 & @xmath271.525&@[email protected]&@[email protected]&@[email protected]&@[email protected] + 0.030 & 0.275 & @xmath271.365&@[email protected]&@[email protected]&@[email protected]&@[email protected] + 0.030 & 0.310 & @xmath271.342&@[email protected]&@[email protected]&@[email protected]&@[email protected] + 0.030 & 0.335 & @xmath271.326&@[email protected]&@[email protected]&@[email protected]&@[email protected] + 0.040 & 0.250 & @xmath271.249&@[email protected]&@[email protected]&@[email protected]&@[email protected] + 0.040 & 0.290 & @xmath271.224&@[email protected]&@[email protected]&@[email protected]&@[email protected] + 0.040 & 0.330 & @xmath271.197&@[email protected]&@[email protected]&@[email protected]&@[email protected] + ccccccc @xmath272.27 & + [email protected] & + [email protected] & + [email protected] & + [email protected] & + [email protected] & + [email protected] + @xmath271.97 & + [email protected] & + [email protected] & + [email protected] & + [email protected] & + [email protected] & + [email protected] + @xmath271.86 & + [email protected] & + [email protected] & + [email protected] & + [email protected] & + [email protected] & @[email protected] + @xmath271.56 & @[email protected] & @[email protected] & @[email protected] & @[email protected] & @[email protected] & @[email protected] + @xmath271.55 & @[email protected] & @[email protected] & @[email protected] & @[email protected] & @[email protected] & @[email protected] + @xmath271.54 & @[email protected] & @[email protected] & @[email protected] & @[email protected] & @[email protected] & @[email protected] + @xmath271.53 & @[email protected] & @[email protected] & @[email protected] & @[email protected] & @[email protected] & @[email protected] + @xmath271.36 & @[email protected] & @[email protected] & @[email protected] & @[email protected] & @[email protected] & @[email protected] + @xmath271.34 & @[email protected] & @[email protected] & @[email protected] & @[email protected] & @[email protected] & @[email protected] + @xmath271.33 & @[email protected] & @[email protected] & @[email protected] & @[email protected] & @[email protected] & @[email protected] + @xmath271.25 & @[email protected] & @[email protected] & @[email protected] & @[email protected] & @[email protected] & @[email protected] + @xmath271.22 & @[email protected] & @[email protected] & @[email protected] & @[email protected] & @[email protected] & @[email protected] + @xmath271.20 & @[email protected] & @[email protected] & @[email protected] & @[email protected] & @[email protected] & @[email protected] + + @xmath180 & @[email protected]&@[email protected]&@[email protected] & @[email protected]&@[email protected]&@[email protected] + @xmath181 & @[email protected]&@[email protected]&@[email protected]&@[email protected]&@[email protected]&@[email protected] + lmc & @xmath270.35 & + [email protected] & @[email protected] & + [email protected] & + [email protected] + smc & @xmath270.70 & + [email protected]&[email protected]&[email protected]&[email protected] + smc@xmath182&@xmath270.70 & @xmath11 + 0.57 & @xmath11 + 0.53 & @xmath11 + 0.54 & @xmath11 + 0.56 + lcccccc lmc & + [email protected]&[email protected]&[email protected]&[email protected]&[email protected]&[email protected] + smc & + [email protected]&[email protected]&[email protected]&[email protected]&[email protected]&[email protected] + smc@xmath27lmc & @xmath11 + 0.30&@xmath11 + 0.22&@xmath11 + 0.12&@xmath11 + 0.07&@xmath11 + 0.18&@xmath11 + 0.10
+ @xmath38[fe / h][email protected] & @xmath11 + 0.21&@xmath11 + 0.16&@xmath11 + 0.08&@xmath11 + 0.05&@xmath11 + 0.13&@xmath11 + 0.08 + a & [email protected] & + [email protected]&[email protected] & [email protected]&[email protected] & [email protected] + b & [email protected] & @[email protected]&[email protected] & [email protected]&[email protected] & [email protected] + a & + 0.13@xmath183&[email protected]&[email protected]&[email protected]&[email protected]&[email protected]&[email protected] + b&@xmath270.37@xmath184&[email protected]&[email protected]&[email protected]&[email protected]&[email protected]&[email protected] + smc&@xmath270.88@xmath183 & + [email protected]&[email protected]&[email protected]&[email protected]&[email protected]&[email protected] + lmc&@xmath270.37@xmath185 & + [email protected]&[email protected]&[email protected]&[email protected]&[email protected]&[email protected] + mw & + 0.15@xmath186&@[email protected]&@[email protected]&@[email protected] & @[email protected]&@[email protected]&@[email protected] + lmc & 8.50 & 18.59@xmath184 & 18.50@xmath187 & 18.57@xmath186 + smc & 7.98 & 18.99@xmath183 & 18.99@xmath188 & 18.98@xmath186 + ic1613 & 7.86 & 24.31@xmath186 & 24.17@xmath189 & 24.38@xmath190 + ic4182 & 8.40 & 28.25@xmath186 & 28.35@xmath186 & 28.23@xmath190 + ngc224 & 8.98 & 24.47@xmath191 & 24.38@xmath190 & 24.37@xmath183 + ngc300 & 8.35 & 26.65@xmath192 & 26.53@xmath190 & 26.48@xmath193 + ngc598 & 8.82 & 24.81@xmath193 & 24.47@xmath191 & 24.71@xmath186 + ngc3109 & 8.06 & 25.52@xmath190 & 25.54@xmath194 & 25.57@xmath190 + ngc3351 & 9.24 & 30.39@xmath195 & 29.92@xmath184 & 29.92@xmath190 + ngc3621 & 8.75 & 29.36@xmath191 & 29.15@xmath186 & 29.26@xmath196 + ngc3031 & 8.75 & 28.03@xmath196 & 27.75@xmath183 & 27.70@xmath193 + ngc4258@xmath197 & 8.93 & 29.46@xmath191 & 29.12@xmath198 & 29.42@xmath186 + ngc4258@xmath199 & 8.53 & 29.46@xmath191 & 29.35@xmath196 & 29.42@xmath186 + ngc5253 & 8.15 & 27.88@xmath191 & 27.63@xmath200 & 99.00@xmath201 + ngc5457@xmath197 & 9.20 & 29.42@xmath191 & 28.93@xmath191 & 29.34@xmath184 + ngc5457@xmath199 & 8.50 & 29.42@xmath191 & 29.24@xmath183 & 29.34@xmath184 + ngc6822 & 8.14 & 23.37@xmath192 & 23.30@xmath192 & 99.00@xmath201 + sexa & 7.49 & 25.67@xmath195 & 25.66@xmath200 & 25.78@xmath186 + sexb & 7.56 & 25.63@xmath193 & 25.63@xmath184 & 25.79@xmath193 + wlm & 7.74 & 24.77@xmath184 & 25.01@xmath187 & 24.93@xmath193 + | we perform a detailed analysis of cepheids in ngc 4258 , magellanic clouds and milky way in order to verify the reliability of the theoretical scenario based on a large set of nonlinear convective pulsation models .
we derive wesenheit functions from the synthetic @xmath0 magnitudes of the pulsators and we show that the sign and the extent of the metallicity effect on the predicted period - wesenheit ( @xmath1-@xmath2 ) relations change according to the adopted passbands . these @xmath1-@xmath2 relations are applied to measured @xmath0 magnitudes of ngc 4258 , magellanic and galactic cepheids available in the literature .
we find that magellanic and galactic cepheids agree with the metallicity dependence of the predicted @xmath1-@xmath2 relations . concerning the ngc 4258 cepheids , the results strongly depend on the adopted metallicity gradient across the galactic disc .
the most recent nebular oxygen abundances support a shallower gradient and provide a metallicity dependence that agrees well with current pulsation predictions .
moreover , the comparison of cepheid distances based on @xmath3 magnitudes with distance estimates based on the revised trgb method for external galaxies , on the @xmath4 trigonometric parallaxes for galactic cepheids , and on eclipsing binaries in the magellanic clouds seems to favor the metallicity correction predicted by pulsation models . the sign and
the extent of the metallicity dependence of the period - wesenheit and of the period - luminosity relations change according to the adopted passbands .
therefore , distances based on different methods and/or bands should not be averaged .
the use of extragalactic cepheids to constrain the metallicity effect requires new accurate and extensive nebular oxygen measurements . |
characteristic polynomials of random matrices have found considerable attention in recent years , one reason being that their correlations seem to reflect the correlations of the eigenvalues @xcite . in this article
we investigate the second - order correlation function of the characteristic polynomial of a sample covariance matrix
. * complex sample covariance matrices .
* let @xmath0 be a distribution on the real line with expectation @xmath1 , variance @xmath2 and finite fourth moment @xmath3 , and for given @xmath4 with @xmath5 , let @xmath6 denote the @xmath7 matrix whose entries @xmath8 are complex random variables whose real and imaginary parts are independent , each with distribution @xmath0 .
let @xmath9 denote the conjugate transpose of @xmath10 .
then the hermitian @xmath11 matrix @xmath12 is called the ( complex ) _ sample covariance matrix _ associated with the distribution @xmath0 . for @xmath13 , let @xmath14 denote the second - order correlation function of the characteristic polynomial of the `` unrescaled '' sample covariance matrix . * real sample covariance matrices . *
let @xmath0 be a distribution on the real line with expectation @xmath1 , variance @xmath15 and finite fourth moment @xmath3 , and for given @xmath4 with @xmath5 , let @xmath6 denote the @xmath7 matrix whose entries @xmath8 are real random variables with distribution @xmath0 .
let @xmath16 denote the transpose of @xmath17 .
then the symmetric @xmath11 matrix @xmath18 is called the ( real ) _ sample covariance matrix _ associated with the distribution @xmath0 . for @xmath13 , let @xmath19 denote the second - order correlation function of the characteristic polynomial of the `` unrescaled '' sample covariance matrix . in the special case
where @xmath0 is the gaussian distribution , the random matrix @xmath10 is also called a ( complex or real ) _ wishart matrix _ , since its distribution is the ( complex or real ) _ wishart distribution _
( see anderson @xcite or muirhead @xcite )
. it will always be clear from the context whether we are considering the complex or real case .
more precisely , we are interested in the asymptotic behavior of the values @xmath20 as @xmath21 , where @xmath22 , @xmath23 for some fixed natural number @xmath24 , and the parameters @xmath25 and @xmath26 are finally used to `` zoom in '' at certain interesting regions of the spectrum of the ( unrescaled ) sample covariance matrix , namely the bulk of the spectrum , the soft edge of the spectrum , and the hard edge of the spectrum .
( see section 3 for details . ) in the complex setting , we will recover , in all the regions previously mentioned , the well - known kernels for the `` correlation functions '' of the eigenvalues , namely the sine kernel , the airy kernel and the bessel kernel .
( see section 3 for details . ) thus , our results indicate that these kernels are _ universal _ in that they arise in connection with quite general sample covariance matrices ( as described at the beginning of this section ) , albeit at the level of the characteristic polynomial .
it is conjectured that this universality also holds at the level of the eigenvalue themselves , but so far this conjecture has only been proven for a restricted class of sample covariance matrices in the bulk of the spectrum ( see ben arous and pch @xcite ) and under stronger moment conditions than ours for the edges of the spectrum ( see soshnikov @xcite and tao and vu @xcite , respectively ) .
similar results will also be obtained in the real setting .
however , the results for the correlation functions of the characteristic polynomial are somewhat more different from the results for the eigenvalues here .
to obtain our results , we follow the strategy proposed by gtze and ksters @xcite in the context of wigner matrices .
first , we obtain an explicit expression for an exponential - type generating function of the second - order correlation function of the characteristic polynomial .
second , we recover the well - known kernels from random matrix theory by asymptotic analysis .
in this section we derive the generating function of the second - order correlation function of the characteristic polynomial . since the derivation is very similar for complex and real sample covariance matrices , we present the details only for the complex setting and restrict ourselves to a few comments for the real setting
. the following well - known representation will prove useful : [ chiral ] let @xmath27 .
then , for @xmath28 , @xmath29 this is an immediate consequence of the matrix factorizations @xmath30 @xmath31 where @xmath32 , @xmath33 , @xmath34 , @xmath35 , and @xmath36 and @xmath37 are invertible . a matrix of the form @xmath38 , where @xmath17 is a ( complex ) random matrix with entries , is also called a ( complex ) _ chiral matrix_.
thus , lemma [ chiral ] establishes a connection between the characteristic polynomial of a sample covariance matrix and that of a chiral matrix .
chiral matrices seem more convenient for our purposes , as they give rise to neater row and column expansions . to illustrate our approach , we start with the first moment of the characteristic polynomial , which has already been determined by forrester and gamburd @xcite by means of a combinatorial argument . fix @xmath39 , and put @xmath40 where @xmath41 is a matrix of size @xmath7 with matrix entries satisfying our standing moment conditions . note that this definition is meaningful even if @xmath42 or @xmath43 ; we then have @xmath44 and @xmath45 .
( in particular , the determinant of the @xmath46 matrix is defined to be @xmath15 . )
let us derive a recursive equation for @xmath47 .
suppose that @xmath48 .
then , doing a row and column expansion about the last row and the last column , we have @xmath49 } \\ ( x^{[\,\cdot\,:m]})^ * & \lambda i_{m-1 } \end{array } \right ) \nonumber \\ & + \sum_{i , j=1}^{n } ( -1)^{i+j-1 } x_{i , m } \overline{x}_{j , m } \det \left ( \begin{array}{cc } \lambda i_{n}^{[i : j ] } & x^{[i : m ] } \\ ( x^{[j : m]})^ * & \lambda i_{m-1 } \end{array } \right ) \
, , \label{double - expansion}\end{aligned}\ ] ] where an upper index @xmath50 $ ] indicates that the @xmath51th row and the @xmath52th column of the corresponding matrix are deleted .
taking expectations , using independence and noting that @xmath53 ( as follows from our standing moment conditions ) , we obtain @xmath54 observe that this is meaningful even if @xmath42 , since the second term vanishes in this case .
hence , recalling the identity @xmath55 we see that the values @xmath47 are completely determined by recursion over @xmath56 . for arbitrary @xmath57 , the laguerre polynomials @xmath58
are defined by @xmath59 ( see equation ( 5.1.6 ) in szeg @xcite ) .
it is well - known that the laguerre polynomials satisfy the relations @xmath60 @xmath61 ( see equations ( 5.1.10 ) and ( 5.1.14 ) in szeg @xcite ) , through which they are completely determined by recursion over @xmath56 . comparing , with ,
, it is easy to see ( by induction on @xmath56 ) that for any @xmath62 , @xmath63 hence , using lemma [ chiral ] , we find that the first moment of the characteristic polynomial of a ( complex ) sample covariance matrix satisfies @xmath64 in fact , this was already proved by forrester and gamburd @xcite by means of a combinatorial argument . equation ( [ firstorder ] ) remains true for real sample covariance matrices ( under the respective moment conditions ) .
we have obtained a recursive equation over @xmath56 for the moments @xmath47 of course , for symmetry reasons , it is clear that it is equally possible to derive a recursive equation over @xmath65 .
indeed , starting with a row and column expansion about the first row and the first column , we obtain @xmath66 instead of , together with the initial condition @xmath67 since the laguerre polynomials satisfy the relation @xmath68 @xmath69 ( see equations ( 5.1.13 ) and ( 5.2.1 ) in szeg @xcite ) , they can also be calculated by recursion over @xmath65 , and ( [ fnm ] ) follows ( by induction on @xmath65 ) .
let us now turn to the second moment of the characteristic polynomial .
similarly as above , we first consider @xmath70 a similar expansion of the determinants as in yields @xmath71 } \\ ( x^{[\,\cdot\,:m]})^ * & \mu i_{m-1 } \end{array } \right ) + \sum_{i , j=1}^{n } ( -1)^{i+j-1 } x_{i , m } \overline{x}_{j , m } \det \left ( \begin{array}{cc } \mu i_{n}^{[i : j ] } & x^{[i : m ] } \\ ( x^{[j : m]})^ * & \mu i_{m-1 } \end{array } \right ) \right ) \\ & \mskip24mu \,\cdot\ , \left ( \nu \det \left ( \begin{array}{cc } \nu i_{n } & x^{[\,\cdot\,:m ] } \\ ( x^{[\,\cdot\,:m]})^ * & \nu i_{m-1 } \end{array } \right ) + \sum_{k , l=1}^{n } ( -1)^{k+l-1 } x_{k , m } \overline{x}_{l , m } \det \left ( \begin{array}{cc } \nu i_{n}^{[k : l ] } & x^{[k : m ] } \\
( x^{[l : m]})^ * & \nu i_{m-1 } \end{array } \right ) \right ) \,.\end{aligned}\ ] ] note that due to our standing moment assumptions , we have @xmath72 hence , expanding the product and taking expectations , we obtain @xmath73 } \\ ( x^{[\sdot : m]})^ * & \mu i_{m-1 } \end{array } \right ) \det \left ( \begin{array}{cc } \nu i_{n } & x^{[\sdot : m ] } \\
( x^{[\sdot : m]})^ * & \nu i_{m-1 } \end{array } \right ) \\ & \mskip24mu \,-\ , \mu \sum_{k=1}^{n } \ee |x_{k , m}|^2 \det \left ( \begin{array}{cc } \mu i_{n } & x^{[\sdot : m ] } \\
( x^{[\sdot : m]})^ * & \mu i_{m-1 } \end{array } \right ) \det \left ( \begin{array}{cc } \nu i_{n-1 } & x^{[k : m ] } \\
( x^{[k : m]})^ * & \nu i_{m-1 } \end{array } \right ) \\ & \mskip24mu \,-\ , \nu \sum_{i=1}^{n } \ee |x_{i , m}|^2 \det \left ( \begin{array}{cc } \mu i_{n-1 } & x^{[i : m ] } \\
( x^{[i : m]})^ * & \mu i_{m-1 } \end{array } \right ) \det \left ( \begin{array}{cc } \nu i_{n } & x^{[\sdot : m ] } \\
( x^{[\sdot : m]})^ * & \nu i_{m-1 } \end{array } \right ) \\ & \mskip24mu \,+\ , \sum_{i } \ee |x_{i , m}|^4 \det \left ( \begin{array}{cc } \mu i_{n-1 } & x^{[i : m ] } \\
( x^{[i : m]})^ * & \mu i_{m-1 } \end{array } \right ) \det \left ( \begin{array}{cc } \nu i_{n-1 } & x^{[i : m ] } \\
( x^{[i : m]})^ * & \nu i_{m-1 } \end{array } \right ) \\ & \mskip24mu \,+\ , \sum_{i \ne k } \ee |x_{i , m}|^2 |x_{k , m}|^2 \det \left ( \begin{array}{cc } \mu i_{n-1 } & x^{[i : m ] } \\
( x^{[i : m]})^ * & \mu i_{m-1 } \end{array } \right ) \det \left ( \begin{array}{cc } \nu i_{n-1 } & x^{[k : m ] } \\ ( x^{[k : m]})^ * & \nu i_{m-1 } \end{array } \right ) \\ & \mskip24mu \,+\ , \sum_{i \ne j } \ee |x_{i , m}|^2 |x_{j , m}|^2 \det \left ( \begin{array}{cc } \mu i_{n}^{[i : j ] } & x^{[i : m ] } \\
( x^{[j : m]})^ * & \mu i_{m-1 } \end{array } \right ) \det \left ( \begin{array}{cc } \nu i_{n}^{[j : i ] } & x^{[j : m ] } \\
( x^{[i : m]})^ * & \nu i_{m-1 } \end{array } \right)\,.\end{aligned}\ ] ] to shorten notation , let us introduce the auxiliary functions @xmath74 } & x^{[\alpha_1:\sdot ] } \\ ( x^{[\alpha_2:\sdot]})^ * & \mu i_m \end{array } \right ) \det \left ( \begin{array}{cc } \nu i_n^{[\alpha_3:\alpha_4 ] } & x^{[\alpha_3:\sdot ] } \\ ( x^{[\alpha_4:\sdot]})^ * & \nu i_m \end{array } \right ) \right ) \,,\ ] ] where @xmath75 ( a dot representing no deletion ) , as well as the abbreviations @xmath76 11_a : = ( 1,1,2,2),\ 11_b : = ( 1,2,1,2),\ 11_c : = ( 1,2,2,1 ) . $ ] ( no other values of @xmath24 will be needed .
moreover , the value @xmath77 will be needed for the real setting only . )
note that @xmath78 .
we then have @xmath79 note that the preceding formula was obtained by starting with an expansion `` in the @xmath56-dimension '' .
of course , it is possible to derive an analogous formula by starting with an expansion `` in the @xmath65-dimension '' ; we then have @xmath80 where @xmath81 is defined by @xmath82 } \\ ( x^{[\sdot:\alpha_2]})^ * & \mu i_m^{[\alpha_2:\alpha_1 ] } \end{array } \right ) \det \left ( \begin{array}{cc } \nu i_n & x^{[\sdot:\alpha_3 ] } \\ ( x^{[\sdot:\alpha_4]})^ * & \nu i_m^{[\alpha_4:\alpha_3 ] } \end{array } \right ) \right ) \,,\ ] ] and the possible values @xmath83 for @xmath24 are the same as before .
here is a list of recursive relations which can be derived using the above arguments : @xmath84 f(n , m,\sdot,01 ) & = ( + \mu ) f(n , m-1,\sdot,\sdot ) \nonumber\\ & - n f(n , m-1,10,\sdot ) & ( n \geq 0 , m \geq 1 ) \
, , \label{fnm - b } \\[+3pt ] f(n , m,\sdot,10 ) & = ( + \nu ) f(n , m-1,\sdot,\sdot ) \nonumber\\ & - n f(n , m-1,01,\sdot ) & ( n \geq 0 , m \geq 1 ) \ , , \label{fnm - c } \allowdisplaybreaks \\[+3pt ] f(n , m,\sdot,11^a ) & = \mu \nu f(n , m-2,\sdot,\sdot ) \nonumber\\ & - \mu n f(n , m-2,01,\sdot ) \nonumber\\ & - \nu
n f(n , m-2,10,\sdot ) \nonumber\\ & + n f(n-1,m-2,\sdot,\sdot ) \nonumber\\ & + n(n-1 ) f(n , m-2,11^a,\sdot ) & ( n \geq 0 , m \geq 2 ) \ , , \label{fnm - d } \\[+3pt ] f(n , m,\sdot,11^c ) & = n f(n-1,m-2,\sdot,\sdot ) \nonumber\\ & + n(n-1 ) f(n , m-2,11^c,\sdot ) & ( n \geq 0 , m \geq 2 ) \ , , \label{fnm - e } \allowdisplaybreaks \\[+3pt ] f(n , m,\sdot,\sdot ) & = \mu \nu f(n-1,m,\sdot,\sdot ) \nonumber\\ & - \mu m f(n-1,m,\sdot,01 ) \nonumber\\ & - \nu m f(n-1,m,\sdot,10 ) \nonumber\\ & + ( 2b+\tfrac12 ) m f(n-1,m-1,\sdot,\sdot ) \nonumber\\ & + m(m-1 ) f(n-1,m,\sdot,11^a ) \nonumber\\ & + m(m-1 ) f(n-1,m,\sdot,11^c ) & ( n \geq 1 , m \geq 0 ) \ , , \label{fnm - f } \\[+3pt ] f(n , m,01,\sdot ) & = ( + \mu ) f(n-1,m,\sdot,\sdot ) \nonumber\\ & - m f(n-1,m,\sdot,10 ) & ( n \geq 1 , m \geq 0 ) \ , , \label{fnm - g } \\[+3pt ] f(n , m,10,\sdot ) & = ( + \nu ) f(n-1,m,\sdot,\sdot ) \nonumber\\ & - m f(n-1,m,\sdot,01 ) & ( n \geq 1 , m \geq 0 ) \ , , \label{fnm - h } \\[+3pt ] f(n , m,11^a,\sdot ) & = \mu \nu f(n-2,m,\sdot,\sdot ) \nonumber\\ & - \mu m f(n-2,m,\sdot,01 ) \nonumber\\ & - \nu m f(n-2,m,\sdot,10 ) \nonumber\\ & + m f(n-2,m-1,\sdot,\sdot ) \nonumber\\ & + m(m-1 ) f(n-2,m,\sdot,11^a ) & ( n \geq 2 , m \geq 0 ) \ , , \label{fnm - i } \\[+3pt ]
f(n , m,11^c,\sdot ) & = m f(n-2,m-1,\sdot,\sdot ) \nonumber\\ & + m(m-1 ) f(n-2,m,\sdot,11^c ) & ( n \geq 2 , m \geq 0 ) \ , .
\label{fnm - j}\end{aligned}\ ] ] ( when evaluating these recursive relations , there may arise some undefined terms with negative arguments , but this poses no problem as these terms are always accompanied by the factor zero .
similar remarks apply to the formulas below . ) together with the initial conditions @xmath85 the equations determine the values @xmath47 completely .
we will now derive a recursive equation involving the values @xmath47 only . by and with @xmath56 replaced by @xmath86
, we have , for @xmath87 , @xmath88 , @xmath89 by ( [ fnm - f ] ) with @xmath90 replaced by @xmath91 , the two summands involving @xmath92 satisfy , for @xmath93 , @xmath88 , @xmath94 whence , for @xmath93 , @xmath88 , @xmath95 plugging this into and rearranging terms , we find that , for @xmath88 , @xmath96 ( for @xmath42 and @xmath97 , follows directly from . )
thus , we have eliminated the terms involving @xmath98 and @xmath99 .
the terms involving @xmath100 and @xmath101 can be eliminated by a similar substitution . using and with @xmath56 replaced by @xmath86 , and with
@xmath90 replaced by @xmath91 and finally with @xmath90 replaced by @xmath91 , we have , for @xmath102 , @xmath88 , @xmath103 plugging this into and rearranging terms , it follows that , for @xmath88 , @xmath104 ( for @xmath42 , follows immediately from . ) by symmetry , we also have , for @xmath102 , @xmath105 hence , the values @xmath47 may be computed recursively starting from and using either or . from now on , for @xmath106 , let @xmath47 denote the second - order correlation function of the characteristic polynomial of the ( complex ) sample covariance matrix as defined in the introduction .
( for the rest of this section , we will usually omit the parameters @xmath107 , which are regarded as fixed . ) then , by lemma [ chiral ] , we have to make the replacements @xmath108 , @xmath109 , @xmath110 in the preceding two equations , thereby obtaining @xmath111 for @xmath88 and @xmath112 for @xmath102 , respectively , along with the initial conditions @xmath113 ( observe that in contrast to the equations for the chiral matrix , the equations for the sample covariance matrix are not fully symmetrical in @xmath65 and @xmath56 . )
let @xmath114 denote the difference of @xmath65 and @xmath56 and assume that @xmath115 .
( the case @xmath116 could be reduced to this case by exchanging the roles of @xmath65 and @xmath56 and by multiplying with the appropriate power of @xmath117 . )
we will determine , for any @xmath118 , the generating function @xmath119 where @xmath120 .
put @xmath121 we will show that for any @xmath118 , the generating function @xmath122 is given by @xmath123 to begin with , observe that @xmath124 defines an analytic function on the unit disc in the complex plane , with power series representation @xmath125 say .
next , differentiating ( [ smartguess ] ) @xmath126 , we obtain @xmath127 since @xmath128 it follows that @xmath129 or ( equivalently ) @xmath130 in terms of the power series coefficients @xmath131 defined by , this translates into the recursive relation , for @xmath88 , @xmath132 where terms with a negative argument are to be regarded as zero . for @xmath43 ,
we clearly have @xmath133 comparing , and , , it is easy to see that the coefficients @xmath131 satisfy the same recursive equations as the values @xmath134 thus , as all values are uniquely determined by these recursive equations , it follows that the generating function is given by , for any @xmath118 .
let us summarize our result for the complex setting as follows : [ complex - scm ] for any @xmath118 , the second - order correlation function @xmath47 of the characteristic polynomial of an unrescaled complex sample covariance matrix satisfies @xmath135 where @xmath136 .
also , let us state the analogous result for the real setting : [ real - scm ] for any @xmath118 , the second - order correlation function @xmath47 of the characteristic polynomial of an unrescaled real sample covariance matrix satisfies @xmath137 where @xmath138 .
as already mentioned at the beginning of this section , the derivation in the real setting is essentially the same as that in the complex setting .
that is why we do not give the full details of the proof of proposition [ real - scm ] , but only mention some noteworthy changes : \(i ) in the real case , we have @xmath139 in particular , as the second moment is not equal to zero anymore , we get an extra term @xmath140 in equation , and an extra term @xmath141 f(n-1,m,\,\cdot\,,11_b)$ ] in equation . also , the factor @xmath142 must be replaced with the factor @xmath3 everywhere .
\(ii ) similar changes arise in equations .
the recursive equation for the @xmath77-terms is the same as that for the @xmath143-terms , except that all occurrences of @xmath143 must be replaced with occurrences of @xmath77 . in fact , it is not hard to see that @xmath144 and @xmath145 for all @xmath62 .
\(iii ) the analogue of reads @xmath146 consequently , the analogue of reads @xmath147 and the analogue of reads @xmath148 similar modifications are required for equations .
\(iv ) defining @xmath124 as in proposition and denoting by @xmath131 the coefficients in the power series representation @xmath149 , it follows by the same arguments as for that for @xmath88 , @xmath150 similarly as in the complex setting , this is the same recursive relation as that for the values @xmath151
from now on , we will always assume that @xmath152 and @xmath153 depend on @xmath154 , where @xmath24 is a fixed natural number .
in particular , @xmath155 and @xmath156 as @xmath21 .
let us begin with the complex setting .
it is well - known from random matrix theory that under the above assumptions , the spectrum of the properly rescaled sample covariance matrix @xmath157 is asymptotically concentrated on the interval @xmath158 $ ] , with distribution given by the _ marenko - pastur density _
@xmath159 moreover , it is well - known from random matrix theory that the following 3 regions deserve particular attention : a. * the bulk of the spectrum * , + the region around a point @xmath160 , b. * the soft edge of the spectrum * , + the region around the point @xmath161 , c. * the hard edge of the spectrum * , + the region around the point @xmath162 .
more precisely , it is widely expected that in these regions , the correlation function of the eigenvalues ( see mehta @xcite or forrester @xcite ) is asymptotically given ( after the appropriate rescaling so that the mean spacing between the eigenvalues is of order @xmath15 ) by a. the sine kernel @xmath163 b. the airy kernel @xmath164 c. the bessel kernel @xmath165 typically , these results were first obtained for wishart matrices ( , gaussian sample covariance matrices ) and then extended to more general sample covariance matrices : a. for the bulk of the spectrum , ben arous and pch @xcite established the emergence of the sine kernel for sample covariance matrices @xmath166 such that the distributions of the entries of the matrix @xmath17 are gaussian convolutions .
b. for the soft edge of the spectrum , soshnikov @xcite established the emergence of the airy kernel by means of the method of moments and a sophisticated comparison with the gaussian case .
c. for the hard edge of the spectrum ,
tao and vu @xcite established the emergence of the bessel kernel ( in the special case where the matrix @xmath17 is a square matrix ) , also via a sophisticated reduction to the gaussian case .
it is the purpose of the following sections to show that the same kernels show up at the level of the characteristic polynomial , for quite general sample covariance matrices .
thus , our results add some support to the universality conjecture that the kernels _ always _ occur , irrespective of the choice of the underlying distribution @xmath0 .
also , our proofs are comparatively simple , and they are based on rather weak moment conditions , considerably weaker ones than those for the above - mentioned results from the literature
. the starting point for our asymptotic analysis will also be the integral representation @xmath167 which follows from by cauchy s formula .
here , @xmath168 is defined as in , @xmath114 , and @xmath169 denotes a counterclockwise circular path of radius @xmath170 , @xmath171 around the origin .
( the precise choice of the radius will be specified in the following sections . ) since @xmath172 where @xmath173 denotes the modified bessel function of order @xmath24 , may be rewritten as @xmath174 we will always use the parametrization @xmath175 , @xmath176 , and we will always assume that @xmath177 is chosen such that @xmath178 on the positive real axis and @xmath179 is continuous otherwise , so @xmath180 $ ] as @xmath126 traverses the path @xmath175 , @xmath176 .
similar remarks apply to the choice of the branch of the modified bessel function . in the course of our asymptotic analysis
, we will use the following classical results about the asymptotic behavior of the modified bessel function : for fixed @xmath115 , @xmath181 and @xmath182 is an absolute constant .
( see chapter 7 in olver @xcite . ) in fact , our results could even be extended to the more general situation where @xmath183 tends to some constant @xmath184 as @xmath21 .
however , this requires a more careful treatment of remainder terms ( particularly the remainder terms originating from the modified bessel functions ) and will therefore not be pursued here .
we will show that in all the above - mentioned cases , given the appropriate specialization of the shift parameters @xmath107 and the radius @xmath185 , the main contribution to the integral comes from a small neighborhood of the point @xmath186 and that we asymptotically end up with the above - mentioned kernels from random matrix theory .
furthermore , by essentially the same proofs , we obtain similar results for real sample covariance matrices . in this setting the correlation function of the characteristic polynomial is asymptotically given ( after the appropriate rescaling ) by a. the `` differentiated '' sine kernel @xmath187 b. the `` differentiated '' airy kernel @xmath188 c. the `` differentiated '' bessel kernel @xmath189 these kernels are obtained from the corresponding kernels for the complex setting by applying the differential operators @xmath190 ( in cases ( i ) and ( ii ) ) and @xmath191 ( in case ( iii ) ) .
( besides that , to obtain the above formulas for the `` differentiated '' kernels , we have to use the differential equation for the airy and bessel function . ) since the derivation of the results for real sample covariance matrices is essentially the same as for complex sample covariance matrices , we will refrain from giving the details of the proofs , but only state the final results .
in order to zoom in around a point in the bulk of the spectrum , we have to make the replacements @xmath192 for some @xmath160 , @xmath193 .
( recall that our sample covariance matrices are unrescaled and that @xmath194 denotes density of the marenko - pastur distribution . )
moreover , we have to multiply by the proper scaling factor in order to obtain a non - degenerate limit .
we then have the following result : [ complex - s ] let @xmath195 denote the second - order correlation function of a complex sample covariance matrix satisfying our standing moment conditions . for any @xmath118 , @xmath160 , @xmath193 ,
setting @xmath196 we have @xmath197 where @xmath198 and @xmath199 is defined as in . by ( [ intrep2 ] ) , we have @xmath200 where we have put @xmath201 for abbreviation . we take @xmath202 .
we reduce the proof of theorem [ complex - s ] to the following two claims : on the one hand , we will show that , for any ( fixed ) @xmath203 , @xmath204 where @xmath205 denotes the restriction of the path @xmath169 to the set @xmath206 , and @xmath207 on the other hand , we will show that for any @xmath208 , there exists some @xmath209 such that for @xmath210 , we have @xmath211 for all @xmath212 large enough . here
, @xmath213 denotes the restriction of the path @xmath169 to the set @xmath214 .
it is easy to see that the main result follows by combining and and by using laplace inversion .
indeed , firstly , and imply that @xmath215 where @xmath216 secondly , by laplace inversion , for @xmath217 , @xmath218 , @xmath219 ( see p.245 in @xcite ) , whence @xmath220 replacing the local shift parameters @xmath107 with @xmath221 , dividing by @xmath194 and noting that @xmath222 yields . to prove and
, it turns out convenient to rewrite the integrand in as @xmath223 where @xmath224 and @xmath225 since @xmath226 and @xmath227 we find that @xmath228 here and in the sequel , we take the convention that the implicit constants in the @xmath229-terms may depend on @xmath230 ( which are regarded as fixed ) . we first prove ( [ claim11 ] ) . in doing so
, we use the notation @xmath231 to denote a bound involving an implicit constant depending also on @xmath232 ( in addition to @xmath230 ) . substituting @xmath233 and @xmath234 on the left - hand side in
, we obtain @xmath235 now , by straightforward taylor expansion , we have the approximations , for @xmath236 , @xmath237 for the second approximation , we have used the observation that for @xmath236 , @xmath238 as well as the asymptotic approximation for the modified bessel function . inserting the preceding approximations into
, it follows that the left - hand side in converges to @xmath239 which proves .
we now turn to the proof of ( [ claim12 ] ) .
it suffices to show that for @xmath210 , @xmath240 \,\cup\ , [ + a / n,+\pi ] } \frac{|f_1(re^{it})||f_2(re^{it})|}{|1-re^{it}|^2 \ , |re^{it}|^{(n+m)/2 } } \ dt \leq \delta \,.\end{aligned}\ ] ] write @xmath241 with @xmath242 , @xmath243 $ ] .
then , @xmath244 @xmath245 @xmath246 from which it follows that @xmath247 also , for @xmath212 sufficiently large , @xmath248 and @xmath249 from which it follows using that latexmath:[\[\begin{aligned } \label{step12c } finally , latexmath:[\[\begin{aligned } \label{step12d } thus , by , , and symmetry , it remains to show that for @xmath210 , @xmath251 to this end , let @xmath252 be a small constant such that @xmath253 for all @xmath254 .
then we have @xmath255 and therefore @xmath256 which entails by choosing @xmath209 large enough .
this completes the proof of theorem [ complex - s ] . with essentially the same proof
, we obtain the following result : [ real - s ] let @xmath195 denote the second - order correlation function of a real sample covariance matrix satisfying our standing moment conditions . for any @xmath118 , @xmath160 , @xmath193 , setting @xmath196 we have @xmath257 where @xmath258 and @xmath259 is defined as in .
in order to zoom in at the soft edge of the spectrum , we have to make the replacements @xmath260 where @xmath193 .
( the factor @xmath261 is for convenience . )
we then have the following result : [ complex - a ] let @xmath195 denote the second - order correlation function of a complex sample covariance matrix satisfying our standing moment conditions . for any @xmath118 , @xmath193 , setting @xmath262 we have @xmath263 where @xmath198 and @xmath264 is defined as in . by ( [ intrep2 ] ) , we have @xmath265 where now @xmath266 this time we take @xmath267 .
similarly as for the bulk of the spectrum , we reduce the proof to the following two claims : on the one hand , we will show that for any ( fixed ) @xmath203 , @xmath268 where @xmath205 denotes the restriction of the path @xmath169 to the set @xmath269 , and @xmath270 on the other hand , we will show that for any @xmath208 , there exists some @xmath209 such that for @xmath210 , we have @xmath271 for all @xmath272 large enough . here
@xmath213 denotes the restriction of the path @xmath169 to the set @xmath273 . using the integral representation for the airy kernel given in @xcite , the theorem
may then be deduced using similar arguments as in the previous section . indeed , first of all , and imply that @xmath274 where @xmath275 substituting @xmath276 , @xmath277 , and shifting the path back to the line @xmath278 ( which is easily justified by cauchy s theorem ) , we obtain @xmath279 making the replacements @xmath280 , @xmath281 and multiplying by @xmath261 , we further obtain @xmath282 by proposition 2.2 in @xcite , the latter expression is equal to @xmath283 , whence . to prove and , we proceed similarly as in the last section .
first of all , we rewrite the integrand as @xmath284 where now @xmath285 and @xmath225 similarly as in the previous section , we find have @xmath286 to prove ( [ claim21 ] ) , we use taylor expansion .
straightforward calculations show that , for @xmath236 , @xmath287 for the second approximation , we have used the observation that for @xmath236 , @xmath288 as well as the asymptotic approximation for the modified bessel function . putting it all together
, the highest - order terms cancel out , and we find that the left - hand side of is asymptotically given by @xmath289 which establishes . it remains to show . to this end
, we will first show that @xmath290 write @xmath241 , where @xmath291 , @xmath243 $ ] .
then , since @xmath292 @xmath293 the proof of is reduced to showing that @xmath294 now , for sufficiently large @xmath212 , @xmath295 @xmath296 where @xmath297 denotes a constant depending only on @xmath24 which need not be the same at each occurrence .
it follows that @xmath298 ( here some uniformly bounded terms have been absorbed into the constant @xmath297 . ) since @xmath299 , the numerator is bounded above by @xmath300 and is proved .
furthermore , the same arguments as those leading to yield latexmath:[\[\begin{aligned } \label{step22b } hence , by , , and symmetry , in order to complete the proof of , it remains to show that for @xmath210 , @xmath302 but this can be proved in the same way as . with essentially the same proof
, we obtain the following result : [ real - a ] let @xmath195 denote the second - order correlation function of a real sample covariance matrix satisfying our standing moment conditions . for any @xmath118 , @xmath193 , setting @xmath262 we have @xmath303 where @xmath258 and @xmath304 is defined as in .
in order to zoom in at the hard edge of the spectrum , we have to make the replacements @xmath305 where @xmath306 .
we then have the following result : [ complex - j ] let @xmath195 denote the second - order correlation function of a complex sample covariance matrix satisfying our standing moment conditions . for any @xmath118 , @xmath306 , setting @xmath307
, we have @xmath308 where @xmath198 and @xmath309 is defined as in . by ( [ intrep2 ] )
, we have @xmath310 where now @xmath311 for the radius @xmath185 we make the same choice as for the bulk of the spectrum , namely @xmath312 .
similarly as in the preceding sections , we reduce the proof to the following two claims : once these two claims are established , the proof of theorem [ complex - j ] is completed using similar arguments as in the preceding section . and imply that @xmath316 where @xmath317 substituting @xmath318 , @xmath319 , and shifting the path back to the line @xmath278 ( which is easily justified by cauchy s theorem ) , we obtain @xmath320 making the replacements @xmath321 , @xmath322 and dividing by @xmath323 , we further obtain @xmath324 the subsequent lemma [ lemma ] states that this is equal to @xmath325 , thereby proving .
to prove ( [ claim31 ] ) , we proceed by taylor expansion once more . a major difference to the previous situations
is given by the fact that , for @xmath236 , @xmath326 remains bounded . since , for @xmath236 , @xmath327 we obtain @xmath328 whence .
we start from the following integral representation for the product of two bessel functions ( see p. 281 in @xcite ) : @xmath332 differentiating with respect to @xmath333 and @xmath334 , we obtain @xmath335 and @xmath336 respectively . multiplying the former equation by @xmath333 and the latter equation by @xmath334 , taking the difference and dividing by @xmath337 yields @xmath338 the assertion now follows by letting @xmath339 , @xmath340 , and @xmath341 .
[ real - j ] let @xmath195 denote the second - order correlation function of a real sample covariance matrix satisfying our standing moment conditions . for any @xmath118 , @xmath306 , setting @xmath307 , we have @xmath342 where @xmath258 and @xmath343 is defined as in . | we investigate the second - order correlation function of the characteristic polynomial of a sample covariance matrix . starting from an explicit formula for a generating function ,
we re - obtain several well - known kernels from random matrix theory . |
at low temperature , hydrogen exhibits the nontrivial structural behavior under the pressure ( @xmath8 ) @xcite , @xcite . below @xmath9
gpa the hexagonal - closed - packed lattice with freely rotating molecules is stable ( phase i ) . at higher pressures , up to @xmath10 gpa , the broken symmetry phase has been observed ( phase ii ) . in the pressure range of @xmath10-@xmath11
gpa the so called phase iii exists .
the experimental measurements have proved , that the all listed phases do not demonstrate the metallic properties . in the pressure range from @xmath12 to @xmath13 gpa ,
the theoretical studies predict the existence of the molecular metallic phase ( the _ cmca _ crystal structure ) @xcite , @xcite , @xcite . above @xmath13 gpa
, the molecular metallic phase transforms to the cs - iv monatomic phase @xcite , @xcite , @xcite , @xcite .
this phase is stable at least up to @xmath14 gpa @xcite . for the extremely high value of the pressure ( @xmath15 gpa ) maksimov and savrasov
have proposed the simple _ fcc _ structure @xcite
. the molecular and monatomic metallic form of the hydrogen can be the superconductor with the high critical temperature ( @xmath16 ) @xcite .
in particular , the calculated values of the critical temperature have been presented in tab .. we notice , that @xmath16 usually has been obtained by using the mcmillan formula @xcite , which represents the weak coupling limit of the more elaborate eliashberg approach @xcite . however , in the case of the metallic hydrogen the electron - phonon interaction is strong , hence the mcmillan expression is inappropriate . for this reason
, we have calculated the critical temperature with the help of the eliashberg equations .
we have considered the case @xmath17 gpa .
additionally , we have studied precisely the properties of the order parameter and the electron effective mass . in the paper we have taken into consideration the eliashberg set in the mixed representation @xcite .
this approach allows one to obtain the stable solutions on the real axis , since the analysis does not involve any principal - part integrals with singular integrands .
.[t1 ] the critical temperature for the selected values of the pressure ; @xmath18 denotes the coulomb pseudopotential . [ cols="^,^,^,^,^",options="header " , ]
the eliashberg equations in the mixed representation have been written in the following form @xcite : @xmath19 } { \sqrt{\omega_m^2z^{2}_{m}+\phi^{2}_{m}}}\phi_{m}\\ \nonumber & + & i\pi\int_{0}^{+\infty}d\omega^{'}\alpha^{2}f\left(\omega^{'}\right ) [ \left[n\left(\omega^{'}\right)+f\left(\omega^{'}-\omega\right)\right]\\ \nonumber & \times&k\left(\omega,-\omega^{'}\right)\phi\left(\omega-\omega^{'}\right)]\\ \nonumber & + & i\pi\int_{0}^{+\infty}d\omega^{'}\alpha^{2}f\left(\omega^{'}\right ) [ \left[n\left(\omega^{'}\right)+f\left(\omega^{'}+\omega\right)\right]\\ \nonumber & \times&k\left(\omega,\omega^{'}\right)\phi\left(\omega+\omega^{'}\right)],\end{aligned}\ ] ] and @xmath20\\ \nonumber & \times&k\left(\omega,-\omega^{'}\right)\left(\omega-\omega^{'}\right)z\left(\omega-\omega^{'}\right)]\\ \nonumber & + & \frac{i\pi}{\omega}\int_{0}^{+\infty}d\omega^{'}\alpha^{2}f\left(\omega^{'}\right ) [ \left[n\left(\omega^{'}\right)+f\left(\omega^{'}+\omega\right)\right]\\ \nonumber & \times&k\left(\omega,\omega^{'}\right)\left(\omega+\omega^{'}\right)z\left(\omega+\omega^{'}\right ) ] , \end{aligned}\ ] ] where : @xmath21 the eliashberg set , eqs .
( and ) , gives the following solutions on the real axis : the order parameter function @xmath22 and the wave function renormalization factor @xmath23 .
the order parameter is defined by the expression : @xmath24 . on the other hand ,
the imaginary axis functions ( @xmath25 and @xmath26 ) should be calculated by using the equations : @xmath27 and @xmath28 where the matsubara energy is given by : @xmath29 and @xmath30 ; @xmath31 is the boltzmann constant .
the pairing kernel for the electron - phonon interaction has the form : @xmath32 the eliashberg function for the molecular hydrogen ( @xmath33 ) has been calculated by l. zhang _ et al .
the maximum phonon energy ( @xmath34 ) is equal to @xmath35 mev .
the function @xmath36 describes the coulomb repulsion between electrons ; @xmath18 denotes the coulomb pseudopotential , @xmath37 is the heaviside unit function and @xmath38 represents the energy cut - off ( @xmath39 ) . for the molecular hydrogen
we assume : @xmath1 .
the symbols @xmath40 and @xmath41 denote the bose and fermi functions respectively .
the eliashberg equations have been solved for @xmath42 matsubara frequencies ( @xmath43 ) by using the numerical method presented in the paper @xcite and @xcite . in the considered case ,
the functions @xmath22 and @xmath23 are stable for @xmath44 k.
, whereas the regular lines correspond to the solutions for @xmath45 .
( b ) the temperature dependence of the order parameter . in the both cases we assume @xmath46 . ] in fig .
we have presented the dependence of the critical temperature on the value of the coulomb pseudopotential .
the exact numerical solutions of the eliashberg equations have been represented by the black circles .
the dotted and dashed lines represent the calculation of the critical temperature by using the mcmillan formula and the more elaborated allen - dynes expression @xcite respectively .
it is easy to see , that mcmillan formula lowers much @xmath16 in the whole range of the coulomb pseudopotential s values , whereas the allen - dynes expression predicts correctly the critical temperature only for very low values of @xmath18 .
for this reason we have modified the classical allen - dynes expression in order to obtain the analytical formula , which reproduces the eliashberg results exactly ( the solid line in fig . ) . in particular ,
we have fitted the selected parameters in the allen - dynes expression ( @xmath47 and @xmath48 ) with the help of @xmath49 numerical values of @xmath50 .
the final result takes the form : @xmath51,\ ] ] where : @xmath52^{\frac{1}{3 } } , \quad { \rm and } \quad f_{2}\equiv 1+\frac{\left(\frac{\sqrt{\omega_{2}}}{\omega_{\rm{ln}}}-1\right)\lambda^{2}}{\lambda^{2}+\lambda^{2}_{2}}.\ ] ] additionally : @xmath53 the parameters @xmath54 , @xmath55 and @xmath56 are equal to @xmath57 , @xmath58 mev and @xmath59 mev , respectively . in fig .
we have shown the order parameter on the real axis for the selected temperatures , and @xmath46 ; the eliashberg function is also plotted .
on the basis of the presented results one can state , that both the real and imaginary part of the function @xmath60 is plainly correlated with the shape of the electron - phonon interaction .
this effect is especially clearly visible for the low values of the temperature .
the full form of the order parameter on the complex plane has been presented in fig .
we have stated , that the @xmath60 values form the distorted spirals which shrink with the growth of the temperature .
basing on fig .
( a ) it is also possible to notice , that the effective electron - electron interaction is attractive ( re@xmath61 > 0 $ ] ) in the range of the frequencies from zero to @xmath62 .
taking into consideration the equation : @xmath63 $ ] we have calculated the dependence of the order parameter on the temperature ( see fig .
next , the value of the ratio @xmath64 can be obtained , where @xmath65 denotes the value of the order parameter close to the zero temperature and @xmath66 . in fig .
we have presented the possible values of @xmath67 for @xmath1 .
it is easy to see , that in the whole range of the considered values of the coulomb pseudopotential , the ratio @xmath67 differs essentially from the bcs prediction ; @xmath68_{\rm bcs}=3.53 $ ] @xcite .
additionally , in the inset in fig .
we have shown the dependence of @xmath65 on @xmath18 . as a function of the coulomb pseudopotential .
the inset shows the open form of @xmath65 . ] ) .
the lines with symbols represent the solutions for @xmath69 , whereas the regular lines correspond to the solutions for @xmath45 .
the inset shows the ratio @xmath70 as a function of the temperature . ]
the second solution of the eliashberg equations on the real axis has been plotted in fig .
( @xmath46 ) .
the obtained results prove , that the function @xmath23 also clearly senses the structure of the electron - phonon interaction .
the shape of the wave function renormalization factor on the complex plane has been presented in fig .. we see , that in contrast to the order parameter , the function @xmath23 weakly depends on the temperature . in the framework of the eliashberg formalism , on the basis of the wave function renormalization factor , the temperature dependence of the electron effective mass ( @xmath71 ) can be calculated .
in particular : @xmath72 $ ] , where the symbol @xmath73 denotes the bare electron mass .
the values of the ratio @xmath70 have been presented in the fig s .
inset . according to the presented data , it is easy to spot , that @xmath71 is high in the full range of the considered temperatures and @xmath6_{\rm max}=2.2 $ ] for @xmath74 .
we notice , that at the critical temperature , the electron effective mass is independent of @xmath18 .
we have calculated the selected thermodynamic properties of the superconducting state in the molecular hydrogen ( @xmath17 gpa ) .
the numerical analysis has been conducted in the framework of the one - band eliashberg formalism for the wide range of the coulomb pseudopotential s values : @xmath1 .
we have proved , that the critical temperature is high even for large values of @xmath18 ( @xmath75_{\rm min}=141 $ ] k ) .
next , it has been shown , that the superconducting phase is characterized by high value of the dimensionless ratio @xmath67 , which differs from the bcs result ; @xmath76 . finally , we have observed , that the electron - phonon interaction strongly enhances the effective electron mass for the temperatures from zero to @xmath16 . in particular , the maximum of @xmath71 is equal to @xmath77 for @xmath78 . in the future
we will analyze the superconducting state in the molecular hydrogen in the framework of the three - gap eliashberg model .
it will be done in order to discuss the effect of the multi - band anisotropy @xcite .
the authors wish to thank k. dziliski for providing excellent working conditions and the financial support ; a.p .
durajski and d. szczniak for the productive scientific discussion that improved the quality of the presented work .
all numerical calculations were based on the eliashberg function sent to us by : * l. zhang * , y. niu , q. li , t. cui , y. wang , * y. ma * , z. he and g. zou for whom we are also very thankful . 99 c. narayana , h. luo , j. orloff , a.l .
ruoff , nature * 393 * , 46 ( 1998 ) .
p. loubeyre , f. occelli , r. letoullec , nature * 416 * , 613 ( 2002 ) .
k. johnson , n.w .
ashcroft , nature * 403 * , 632 ( 2000 ) .
m. stadele , r.m .
martin , phys .
lett . * 84 * , 6070 ( 2000 ) .
pickard , r. needs , nat .
* 3 * , 473 ( 2007 ) .
v. natoli , r.m .
martin , d.m .
ceperley , phys .
lett . * 70 * , 1952 ( 1993 ) .
y. yan , j. gongb , y. liu , phys .
a * 375 * , 1264 ( 2011 ) .
e.g. maksimov , d.y .
savrasov , solid state commun . * 119 * , 569 ( 2001 ) .
n.w . ashcroft , phys .
* 21 * , 1748 ( 1968 ) .
l. zhang , y. niu , q. li , t. cui , y. wang , y. ma , z. he , g. zou , solid state commun . *
141 * , 610 ( 2007 ) .
r. szczniak , m.w .
jarosik , physica b * 406 * , 2235 ( 2011 ) .
barbee , m.l .
cohen , phys .
b * 143 * , 5269 ( 1991 ) .
p. cudazzo , g. profeta , a. sanna , a. floris , a. continenza , s. massidda , e.k.u .
gross , phys .
lett . * 100 * , 257001 ( 2008 ) .
r. szczniak , m.w .
jarosik , solid state commun . * 149 * , 2053 ( 2009 ) .
mcmillan , phys .
rev . * 167 * , 331 ( 1968 ) . for discussion of the eliashberg equations [ originally formulated by g.m .
eliashberg , soviet .
jetp * 11 * , 696 ( 1960 ) ] we refer to : j.p .
carbotte , rev .
phys . * 62 * , 1027 ( 1990 ) .
f. marsiglio , m. schossmann , j.p .
carbotte , phys .
b * 37 * , 4965 ( 1988 ) . ( a )
r. szczniak , acta phys . pol .
a * 109 * , 179 ( 2006 ) ; + ( b ) r. szczniak , solid state commun . * 138 * , 347 ( 2006 ) .
allen , r.c .
dynes , phys .
b * 12 * , 905 ( 1975 ) .
j. bardeen , l.n .
cooper , j.r .
schrieffer , phys . rev . * 108 * , 1175 ( 1957 ) . | the detailed study of the selected thermodynamic properties of the superconducting phase in the molecular hydrogen under the pressure at @xmath0 gpa has been presented . for the increasing value of the coulomb pseudopotential , @xmath1 ,
the following results have been obtained : ( i ) the critical temperature decreases from @xmath2 k to @xmath3 k , ( ii ) the ratio @xmath4 differs noticeably from the bcs value : @xmath5 ; ( iii ) the electron effective mass is large and grows slightly together with the temperature ( @xmath6_{{\rm max } } = 2.2 $ ] for @xmath7 ) .
# 1[#1 ] # 1[#1 ] # 1[#1 ] |
one of the features of polarized dis is that a lot of the present data are in the preasymptotic region ( @xmath4 ) . while in the unpolarized case we can cut the low @xmath2 and @xmath5 data in order to minimize the less known higher twist effects
, it is impossible to perform such a procedure for the present data on the spin - dependent structure functions without losing too much information .
this is especially the case for the hermes , slac and jefferson lab experiments .
so , to confront correctly the qcd predictions with the experimental data and to determine the _ polarized _ parton densities special attention must be paid to higher twist ( powers in @xmath6 ) corrections to the nucleon structure functions .
the latter are non - perturbative effects and can not be calculated without using models .
that is why a _ model independent _ extraction of the dynamical higher twists from the experimental data is important not only for a better determination of the polarized parton densities but also because it would lead to interesting tests of the non - perturbative qcd regime and , in particular , of the quark - hadron duality . before one can properly obtain information on the higher twist contribution ,
it is important to take into account in the analysis the so - called target mass corrections ( tmcs ) arising from purely kinematic effects associated with finite values of the quantity @xmath7 .
these are also powers in @xmath6 corrections , however formally related to the twist - two operators and therefore , unlike the higher twist ones , can be calculated without using models . in this note
we present numerical results which illustrate the main features of the tmcs to the spin structure function @xmath0 valid in the preasymptotic dis region .
we consider that their knowledge is useful and important in the qcd analyses of the present and future data on polarized dis at moderate energies .
let us now discuss the numerical results on the target mass corrections to the spin structure function @xmath0 . in our analysis we will mainly concentrate on their features in the preasymptotic dis region ( the invariant mass @xmath19 and moderate values of @xmath20 ) . in the calculations of the target
mass effects we have used for @xmath17 our recent nlo results from @xmath21 fit presented in @xcite . for the further discussion it is useful to rewrite ( [ g1tmc ] ) in the form : @xmath22 where @xmath23 the questions we address here are _
i ) _ how large are the target mass effects in the preasymptotic dis region and _ ii ) _ how fast the series ( [ g1tmc ] ) converges , or in other words how large is the difference between @xmath24 calculated by ( [ g1tmctot ] ) and @xmath25 obtained when the tmcs are calculated only up to the first order in @xmath26 ( the first two terms in ( [ g1tmc ] ) or equivalently , in ( [ g1tmc1 ] ) ) . we will focus on the proton target , but in the end of our discussion we will mention also our results concerning the neutron spin structure function @xmath27 .
note that it is of interest from a theoretical point of view to examine the convergence of the tmcs irrespective to the size of the dynamical ht . for practical purposes , however , there is no sense to use the exact formula ( [ g1tmctot ] ) in the analysis of the experimental data if the dynamical ht are not negligible to those of the tmcs . in this case
the target mass and dynamical higher twist corrections have to be taken into account up to the same finite order in @xmath28 .
the target mass corrections at fixed order in @xmath26 in the expansion ( [ g1tmc1 ] ) are illustrated in fig .
1 where the notations @xmath29 are used .
the tmcs are given outside the resonance region , _
i.e. _ , @xmath30 , where @xmath31 , at different fixed values of @xmath32 ( black curves ) , @xmath33 ( red curves ) and @xmath34 ( green curves ) .
one can see from fig .
1 that the series ( [ g1tmc ] ) , or equivalently ( [ g1tmc1 ] ) , is an alternating series and the size of tmcs of higher then the first order in @xmath26 are small , especially the target mass corrections of third order in @xmath26 , @xmath35 , which are smaller than @xmath36 . = 3.8 in in order to estimate the target mass effects in the preasymptotic dis region we demonstrate also the relative changes of @xmath37 taking into account tmcs up to order @xmath38 , @xmath39 and all orders of @xmath40 ( eq .
[ g1tmctot ] ) .
these effects are presented for @xmath41 ( fig .
2 ) , @xmath42 and @xmath43 ( fig .
3 ) , respectively .
the notations used in these figures are as follows : @xmath44 \nonumber g_1^{{\rm tmc},q^6}&=&g_1(m=0 ) + \sum _ { j=1}^{3 } \left ( { m^2\over q^2}\right ) ^jg_1^{(j)\rm tmc},\\[3 mm ] g_1^{\rm tmc , tot}&=&g_1^{\rm tmc}({\rm eq.~ 3 } ) .
\label{g1q2q6}\end{aligned}\ ] ] note that to calculate @xmath45 we have used the prescription given in @xcite that @xmath46 in the rhs of eq . [ g1tmctot ] vanishes for @xmath47 , where @xmath48 .
= 3.3 in in figures 2 and 3 the dot , dashed and solid curves correspond to the change of @xmath0 proton when the first , third and all orders in @xmath26 , respectively , are taken into account . as seen from these figures , the sign of the target mass corrections depends on the @xmath49 region and the maximum change of the magnitude of @xmath0 due to the first order tmcs is 3.6% at @xmath50 , 6% at @xmath51 and 28.6% at @xmath52 are generally small was first established in @xcite . ] .
the tm effects are large in the large @xmath49 and small @xmath2 region . however , because of the cut @xmath30 , the large @xmath49 region at small @xmath2 is outside of the preasymptotic dis region and their effects are much smaller than those in the resonance one .
one can see also from figs . 2 and 3 that taking into account a first order tmcs to @xmath0 a good approximation of the exact equation ( [ g1tmc ] ) is already achieved .
the deviation of this approximation of @xmath0 from @xmath45 , is not more than 1.5% for @xmath50 , 0.8% for @xmath51 and 1.8% for @xmath53 .
the dashed curves , which correspond to @xmath54 ( the tmcs are taken into account up to a third order in @xmath40 ) , practically coincide with @xmath55 ( solid curves ) , especially in the cases of @xmath51 and @xmath53 .
this observation is a consequence of the fact we have already discussed above that the series of the target mass corrections ( [ g1tmc ] ) is an alternative sign one .
= 3.0 in = 3.0 in in fig .
4 we compare the tmcs to neutron spin structure function @xmath27 with those to @xmath37 . as seen from fig .
4 its magnitude is much smaller than that one in the proton case .
this is due to the fact that the negative @xmath56 crosses zero at @xmath57 and becomes positive at @xmath49 higher than 0.4 @xcite . as a result ,
the values of the moments of @xmath56 in the rhs of ( [ g1tmc ] ) are much smaller than those of @xmath37 .
the main conclusion about the proton tmcs , namely , that the first order tmcs is a good approximation of their total account , holds for the neutron target too .
= 3.2 in finally , we would like to compare the size of the tmcs with that of the higher twist power corrections to the spin structure function @xmath0 .
taking into account only the first order corrections in @xmath6 , @xmath0 has the following form : @xmath58 where @xmath59 are the dynamical higher twist ( @xmath60 and @xmath61 ) corrections to @xmath0 , which are related to multi - parton correlations in the nucleon .
the latter are non - perturbative effects and can not be calculated without using models . in fig .
5 we compare the first order tmcs , @xmath62 ( see eq .
[ notation ] ) , with the values of the ht corrections , @xmath63 , extracted in a model independent way from the world data on polarized dis @xcite .
one can see from fig . 5
that : _ i ) _ in the neutron case the target mass contribution in @xmath0 is negligible compared to the higher twist effects and _ ii ) _ in the proton case the tmcs are essential at @xmath64 while for @xmath65 their magnitude is much smaller than that of the ht corrections .
note also the different shapes of tm and ht corrections .
= 3.0 in = 3.0 in as a result of this analysis we conclude that in the preasymptotic dis region the higher twist contribution in @xmath0 becomes more and more important , so the tmcs can not be trusted alone .
we have studied the target mass effects in polarized dis scattering which would be important in preasymptotic region .
it was demonstrated that accounting for the first order target mass corrections to @xmath0 a very good approximation of the exact formula ( [ g1tmctot ] ) is achieved .
it was also shown that the size of the tmcs in the preasymptotic dis region is small ( @xmath66 ) except for @xmath67 where their effect is a partially suppressed by the large values of @xmath2 due to the cut @xmath68 .
compared to the higher twist effects the contribution of target mass corrections to the spin structure function @xmath0 in the preasymptotic dis region is insignificant except for @xmath64 in the proton case . nevertheless , to extract correctly the unknown high twist effects from the experimental data , the calculable target mass corrections to the spin structure functions should be taken into account .
we would like to thank e. leader for useful discussions .
this research was supported by the uk royal society and the jinr - bulgaria collaborative grants , and by the rfbr ( no 05 - 01 - 00992 , 05 - 02 - 17748 , 06 - 02 - 16215 ) . | -5 mm the target mass effects in polarized dis have been studied .
it was demonstrated that taking into account the first order target mass corrections to @xmath0 a very good approximation of the exact formula is achieved .
it was also shown that their magnitude in the preasymptotic dis region is small except for @xmath1 , where their large effect is partially suppressed by the large values of @xmath2 due to the cut @xmath3 .
the difference between the size of the target mass and higher twist corrections is illustrated .
1.0 cm keywords : polarized dis , structure functions , quantum chromodynamics + pacs numbers : 13.60.hb , 12.38.-t , 14.20.dh = -1.5 cm = -15 truemm * target mass effects in polarized + deep inelastic scattering * * aleksander v. sidorov * + _ bogoliubov theoretical laboratory + joint institute for nuclear research , 141980 dubna , russia _ 0.5 cm * dimiter b. stamenov + _ institute for nuclear research and nuclear energy + bulgarian academy of sciences + blvd .
tsarigradsko chaussee 72 , sofia 1784 , bulgaria _ * + 3.0 cm |
the standard symplectic structure on @xmath1 is : @xmath2 by * darboux s theorem * every symplectic form is locally diffeomorphic to this one , so it is crucial to understand its properties .
let @xmath3 be the standard ( closed ) ball of radius @xmath4 ( so @xmath5 is proportional to a @xmath6-dimensional area ) , and let @xmath7 be a * symplectic embedding * ( i.e. a smooth embedding onto @xmath8 that preserves @xmath9 ) . since @xmath10 is a volume form , every symplectic embedding preserves volume .
* gromov s question : * _ what can one say about the images @xmath8 of a symplectic ball ?
how are the symplectic and volume preserving cases different ?
_ the following result can be easily proven using moser s homotopy method .
* volume preserving embeddings : * _ if @xmath11 is diffeomorphic to a ball and @xmath12 then there is a volume preserving diffeomorphism @xmath13 . _
the analogous statement is not true for symplectic diffeomorphisms called _ symplectomorphisms _ for short since the boundary of @xmath14 has symplectic invariants given by the characteristic foliation . is spanned by the unique null direction of @xmath15 , i.e. by the vectors @xmath16 such that @xmath17 for all @xmath18 . ]
nevertheless one might ask if one can * fully fill * @xmath14 by a symplectic ball .
this means that for every @xmath19 there is a ball @xmath20 and a symplectic embedding @xmath21 such that @xmath22 .
let @xmath23 be the cylinder @xmath24 we provide @xmath23 with the symplectic structure @xmath9 .
thus its first two coordinates lie in symplectically paired directions . in particular
, @xmath9 has non zero integral over each disc @xmath25 .
* gromov s nonsqueezing theorem : * _ there is a symplectic embedding @xmath26 if and only if @xmath27 .
_ this nonsqueezing property is fundamental . to a first approximation , it is true that a diffeomorphism with this nonsqueezing property for all balls and all symplectic cylinders must preserve the symplectic structure .
( more precise versions of this statement were proven by eliashberg @xcite and ekeland
hofer @xcite . )
it follows also that symplectomorphisms are very different from volume preserving embeddings .
for example , the largest ball that embeds symplectically in the polydisc @xmath28 is @xmath29 which has volume just half that of the polydisc . in their paper @xcite , ekeland and
hofer formalized the idea of a _
symplectic capacity _ , which is a measurement @xmath30 of the size of a symplectic manifold @xmath31 ( possibly open or with boundary ) with the following properties : \(i ) _ @xmath32 is a function with values in @xmath33 $ ] defined on some class of @xmath34-dimensional symplectic manifolds is said to be symplectic if it is provided with a symplectic form @xmath35 . here
@xmath35 is any closed @xmath6-form such that @xmath36 never vanishes .
a fundamental result due to darboux is that every symplectic form is locally diffeomorphic to euclidean space @xmath1 with its standard structure @xmath9 .
hence the importance of understanding the symplectic properties of subsets of @xmath1 . ]
@xmath37 that contains the closures of all open subsets of euclidean space ; _
\(ii ) ( monotonicity ) _ if there is a symplectic embedding @xmath38 then @xmath39 _ ; \(iii ) ( scaling ) _ @xmath40 for all @xmath41 _ ; \(iv ) ( normalization ) _
@xmath42 and @xmath43_. the monotonicity property implies that @xmath32 is a symplectic invariant , that is , it takes the same value on symplectomorphic sets , while the scaling property implies that it scales like a @xmath6-dimensional invariant .
one can satisfy the first three properties by considering an appropriate power of the volume ; for example one could consider the function @xmath44 .
however , this function @xmath32 does not satisfy the second half of the normalization axiom .
indeed , the requirement that a cylinder has finite capacity is what makes this an interesting definition .
here are some examples that illustrate some of the variety of possible definitions .
\(i ) * the gromov width .
* @xmath45 the nonsqueezing theorem implies that this does satisfy the axioms .
moreover , it satisfies the following strong normalization condition : @xmath46 \(ii ) * the hofer
zehnder capacity @xcite*. every smooth function @xmath47 on a closed symplectic manifold generates a @xmath48-parameter subgroup @xmath49 , of the group of symplectomorphisms of @xmath37 by the following procedure ; define the vector field @xmath50 , the _ symplectic gradient _ of @xmath51 , by requiring that @xmath52 and then define @xmath53 to be the flow generated by @xmath50 with @xmath54 .
the identity @xmath55 ( where @xmath56 denotes the lie derivative ) shows that @xmath57 for all @xmath58 .
in other words , the flow preserves the symplectic form .
a point @xmath59 for which there are times @xmath60 such that @xmath61 but @xmath62 is said to be a nontrivial periodic orbit of @xmath53 of period @xmath63 .
further we define @xmath64 to be the set of functions @xmath47 with the following properties : @xmath65 @xmath66 for all @xmath67 , @xmath65 there is an open subset of @xmath68 on which @xmath69 ; @xmath65 @xmath51 is constant outside a compact subset of the interior of @xmath68 ; @xmath65 every nontrivial periodic orbit of @xmath51 has period @xmath70 .
then we define the hofer
zehnder capacity @xmath71 as follows : @xmath72 c_hz(u , ) = _ hh ( _ xm h(x ) ) @xmath72 this function @xmath71 obviously satisfies the first three conditions for capacities .
moreover as @xmath73 increases the flow of @xmath74 moves faster so that the periods of the periodic orbits decrease . therefore in order to prove that a set such as @xmath75 has finite capacity one needs a mechanism to prove that periodic orbits for @xmath53 must exist under suitable circumstances ( for example , if @xmath76 where @xmath77 satisfies the first three conditions to be in @xmath64 and @xmath73 is sufficiently large . ) in the original papers this mechanism involved subtle arguments in variational analysis
; however one can also prove such results using @xmath78-holomorphic methods .
\(iii ) * the displacement energy*. if @xmath79,$ ] is a smooth family of functions then one can define a path of vector fields @xmath80 by requiring that @xmath81,\ ] ] and then integrate this family of vector fields to get a flow @xmath82 $ ] . as before ,
each diffeomorphism @xmath53 preserves @xmath35 , and the family is called a hamiltonian path , or hamiltonian isotopy .
define the length @xmath83 of such a path by setting @xmath84 ( thus this is a measure of the total variation of @xmath85 . )
if @xmath86 then a hamiltonian isotopy @xmath53 of @xmath68 is said to disjoin @xmath87 in @xmath68 if @xmath88 the _ displacement energy _ of a subset @xmath31 of a symplectic manifold @xmath37 is defined to be : @xmath89 where we take the infimum of an empty set of real numbers to be @xmath90 .
it is a deep fact that the displacement energy of a ball is always positive , no matter what @xmath68 is . using this ,
it is easy to see that this energy defines a capacity on the set of all subsets of a given symplectic manifold @xmath37 . interestingly enough , as explained in @xcite there are relations between the displacement energy of a ball and the nonsqueezing theorem .
although the displacement energy of a set @xmath31 in principle depends on the choice of ambient manifold @xmath37 , i am not aware of any examples of sets @xmath87 that have different displacement energies when considered as subsets of two different manifolds @xmath91 and @xmath92 , except in the trivial case when @xmath87 is displaceable in @xmath93 but not in @xmath94 . as a refinement of this question one could consider two different symplectic embeddings
@xmath95 and @xmath96 of one set
@xmath31 into another @xmath37 . are there any examples where @xmath97 this will not happen if the two embeddings are hamiltonian isotopic ( i.e. there is a hamiltonian isotopy @xmath53 of @xmath68 such that @xmath98 ) .
but there are cases in which nonisotopic embeddings are known to exist ; in other words , the space of embeddings @xmath99 need not be connected . such examples ( of embeddings of one polydisc @xmath100 into another ) were constructed in @xcite .
this leads into the whole question of what is the topology of such embedding spaces .
rather little is known about such questions .
the above three capacities are by now classical invariants .
there are many other more recently introduced functions that share some of the properties of capacities .
for example , biran and cornea @xcite are interested in exploring the size " of lagrangian submanifolds @xmath101 of @xmath37 .
a lagrangian submanifold is a submanifold of dimension @xmath102 on which the symplectic form vanishes identically .
a basic example is the subspace @xmath103 of @xmath104 spanned by the coordinates @xmath105 .
lagrangian submanifolds are fundamental objects of study in symplectic geometry .
let us write @xmath106 for the intersection @xmath107 .
then the _ width _ @xmath108 of a lagrangian submanifold @xmath109 is defined to be the supremum of @xmath110 for which there is a symplectic embedding @xmath111 such that @xmath112 .
although some calculations can be made , it is not even known whether every compact lagrangian submanifold of @xmath1 must have finite width . because so little
is known about these general embedding questions , we shall now return to considering balls and ellipsoids .
as we shall see , their embedding properties are very closely related , at least in dimension @xmath0 .
we shall first consider the problem of embedding several equal balls , and then discuss ellipsoids in
[ s : ell ] .
* question : * _ how much of the volume of @xmath113 can be filled by @xmath114 equal symplectic balls ? _ [ gromov @xcite ] : _ there are obstructions when @xmath115 : _
if @xmath116 then @xmath117 ; in other words , the volume of the image of @xmath118 is @xmath119 [ mcduff and polterovich @xcite ] : _ for any @xmath120 the @xmath34-dimensional ball @xmath113 can be fully filled by @xmath121 equal balls . _ one can explicitly see these embeddings in the following way .
* describing embeddings by toric models : * consider the ( moment ) map @xmath122 then @xmath123 is the ( standard ) triangle @xmath124 thus the inverse image of the closed triangle can obviously be fully filled by a ball .
somewhat surprisingly , traynor @xcite showed that it is possible to fully fill the inverse image of the _ interior _ of any standard triangle .
hence , we can obtain full fillings of @xmath125 by @xmath126 equal balls by cutting a standard triangle into @xmath126 standard pieces as in figure [ fig : jap2 ] .
standard triangles into a triangle of size @xmath48.,width=240 ] however , when @xmath127 the story is different : [ mcduff polterovich @xcite ] : _ in dimension @xmath0 there are obstructions to full fillings by @xmath114 balls when _ @xmath128 ; [ biran @xcite ] _ @xmath125 can be fully filled by @xmath114 equal balls for all @xmath129 . _
thus , with enough balls , the obstructions disappear . in later work @xcite
, biran showed that for all ( closed ) rational is called _ rational _ if @xmath130\in h^2(m;{{\mathbb q}})$ ] . ]
symplectic @xmath0-manifolds @xmath37 there is an integer @xmath131 such that @xmath37 can be fully filled by @xmath114 balls for all @xmath132 .
this is rather surprising .
note that @xmath133 depends on @xmath35 and can diverge to infinity even when @xmath68 is fixed .
for example , in the case when @xmath134 and @xmath135 for some area form @xmath136 on @xmath137 of total area @xmath138 , the nonsqueezing theorem shows that one can not embed a ball larger than @xmath139 . since the volume of @xmath140 is @xmath141 while that of @xmath139 is @xmath142
, we find that @xmath143 .
thus , it is not at all clear whether one can remove the rationality condition on @xmath144 $ ] .
it is also not clear whether there is an analogous result in higher dimensions .
[ s : ell ] from now on we shall work in @xmath0 ( real ) dimensions , and shall write @xmath145 to mean that there is a symplectic embedding of @xmath146 into @xmath20 .
define the ellipsoid @xmath147 by setting : @xmath148 consider the embedding capacity function @xmath32 for @xmath149 : @xmath150 note that @xmath151 because @xmath152 . in @xcite ,
cieliebak , hofer , latschev and schlenk describe a wide variety of symplectic embedding problems in quantitative terms , formulating many interesting questions , but providing relatively few answers .
( see also schlenk @xcite . ) in particular , they define the function @xmath153 described above , but could say rather little about its properties .
the first most significant open question concerned the value of @xmath154 .
this was in fact first calculated by opshtein @xcite , though he did not make this explicit in his paper .
as we shall see , the function @xmath153 turns out to be surprisingly interesting .
it is the first of the capacity functions of @xcite to be calculated .
note also that although the nonsqueezing theorem stated above is valid in all dimensions , we only understand the embedding of ellipsoids in @xmath0-dimensions . as is shown by the work of guth @xcite and hind
kerman @xcite , very interesting new phenomena appear in higher dimensions .
* description of the fibonacci stairs : * let @xmath162 be the odd terms in the sequence of fibonacci numbers ; set @xmath163 so that @xmath164 and @xmath165 here we set @xmath166 for convenience , so that @xmath167
. then the claim is that for all @xmath168 we have @xmath169 , \;\;\mbox { and } \;\ ; c(x)=\sqrt { a_{n+1 } } \mbox { on } [ b_n , a_{n+1}].\ ] ] since @xmath170 this gives a continuous graph on the interval @xmath157 .
for example , @xmath171 note that the function @xmath172 is continuous and nonincreasing .
( this holds because , for all @xmath173 and @xmath19 , we have @xmath174 . )
it follows that the function @xmath32 is determined on the interval @xmath175 $ ] by its values at the points @xmath176 . as we describe in more detail below
, the first and third parts of theorem [ thm : main ] are fully proven , while the proof of the second statement is almost complete .
the proofs are based on the two propositions [ prop : wgt ] and [ prop:2 ] stated below .
the first relates the ellipsoidal embedding problem to a ball packing problem , while the second gives a recipe for solving ball packing problems .
[ prop : wgt ] for each rational @xmath177 there is a finite weight expansion @xmath178 such that @xmath147 embeds symplectically in the interior of @xmath179 if and only if the disjoint union of balls @xmath180 embeds symplectically in the interior of @xmath179 .
the second main proposition solves the ball packing problem .
write @xmath196 , @xmath197 .
let @xmath198 we shall refer to the first two conditions above as diophantine conditions .
the third condition @xmath199 is more algebraic , requiring that tuple @xmath200 can be reduced to @xmath201 by repeated cremona moves .
such a move takes @xmath202 to @xmath203 , where @xmath203 is obtained by first transforming @xmath200 to @xmath204 and then reordering the new @xmath205 ( discarding zeros ) so that they do not increase .
it was shown in li li @xcite ( using seiberg witten theory ) that @xmath206 is the set of homology classes @xmath207 represented by symplectic exceptional divisors in the @xmath114-fold blow up of @xmath208 .
the class of a line @xmath209 $ ] and by @xmath210 the class of the @xmath211th exceptional divisor . for more about blow
ups see [ ss : proofs ] .
observe also that this is a place where symplectic geometry shows how flexible it is in comparison with algebraic geometry .
in algebraic ( or complex ) geometry , one would want to describe the homology classes that can be represented by holomorphically embedded exceptional divisors for a generic " complex structure on the blow up .
this question is far from being understood . ]
it follows that the intersection number @xmath212 of any two elements in @xmath213 is nonnegative .
* example : * @xmath216 has the single element @xmath217 ( corresponding to @xmath218 ) and @xmath219
. therefore @xmath220
. however @xmath221 also contains @xmath222 corresponding to @xmath223 .
thus @xmath224 .
the existence of the fibonacci staircase in theorem [ thm : main ] is based on the somewhat surprising discovery that there are elements of @xmath213 related to the weight expansions of the odd fibonacci numbers .
it is not hard to show that the elements @xmath230 and @xmath231 satisfy the diophantine conditions needed to be in @xmath213 .
moreover , there is an easy inductive proof that @xmath231 satisfies the third condition since @xmath231 reduces to @xmath232 under five cremona moves .
however , the corresponding behavior of @xmath230 is much more complicated and takes much more effort to analyze .
since the intersection of any two elements of @xmath213 is nonnegative , proposition [ prop : fib ] part ( ii ) implies that for any @xmath242 we have @xmath243 where @xmath244 .
thus @xmath245 it follows that @xmath246 .
so far this has been proved for @xmath255 and for the ( even ) convergents to @xmath256 .
the proof is quite hard because there are fake " elements of @xmath213 .
more precisely , by looking at ratios of even fibonacci numbers one can construct tuples @xmath200 that satisfy the diophantine conditions to be in @xmath213 but that fail condition @xmath199 ; in fact they have negative intersection with @xmath257 , that is @xmath258 there are also infinitely many other elements of @xmath213 , this time constructed from the even fibonacci numbers , for which @xmath259 on some interval @xmath260 $ ] .
( these form what might be called a _ ghost staircase _ because it lies below the graph of @xmath153 . ) thus one can not understand @xmath153 on this interval by easy estimates using the diophantine conditions .
on the other hand , this approach is sufficient to show that there are no obstructions on the interval @xmath261 and that there are finitely many obstructions on the interval @xmath262 .
full details of the proof will be given in @xcite .
we saw above that @xmath264 .
hence @xmath265 only if there is @xmath266 such that @xmath267 . but
@xmath268 , and @xmath269 so @xmath270 always .
this was biran s argument in @xcite .
more complicated versions of this argument also show there are no constraints for @xmath271 @xmath272 is the positive root of @xmath273 .
the sharpest constraints come from @xmath202 where @xmath274 .
since @xmath275 and @xmath276 we need @xmath277 .
but also @xmath278 implies @xmath279 or @xmath280 . the embedded contact homology theory of hutchings and taubes @xcite is not yet fully developed .
but , according to hutchings , the index calculations in this theory should provide a series of obstructions to symplectically embedding @xmath147 into @xmath281 of the following nature .
let @xmath177 be irrational .
for each pair of integers @xmath282 , consider the triangle @xmath283 we define @xmath284 where @xmath285 is the smallest positive integer such that @xmath286 where @xmath114 is the number of integer points on the slant edge of @xmath287 .
further , set @xmath288 since @xmath289 is nondecreasing , we may extend it to rational @xmath5 by defining @xmath290 as mentioned above , claim ii is not yet fully proven , but it is expected to hold .
further , it seems probable that these two functions actually coincide .
as evidence for this , we have the following results that were explained to me by hutchings . [
le : ech1 ] suppose that @xmath5 is rational , define @xmath292 as above and suppose that @xmath293 where @xmath294 is the number of integral points on the slant edge of @xmath63 .
assume that @xmath295 ( resp .
@xmath296 ) is the integral point on the slant edge with smallest ( resp .
largest ) @xmath59-coordinate .
then there is @xmath19 such that @xmath297 to prove the statement for @xmath298 it suffices by continuity to consider irrational @xmath299 of the form @xmath300 .
then , for small enough @xmath19 , the triangle @xmath301 contains @xmath302 fewer integral points than @xmath63 .
therefore @xmath303 , which proves the first statement .
similarly , the second statement holds because if @xmath304 is irrational and @xmath19 is sufficiently small , the triangle @xmath305 contains @xmath302 fewer integral points than @xmath63 .
consider the triangle @xmath307 with vertices @xmath308 , @xmath309 and @xmath310 , where @xmath311 is the @xmath102th odd fibonacci number . because @xmath312 are mutually prime and satisfy the identities @xmath313 we find that @xmath314 therefore , because @xmath315 , we can apply lemma [ le : ech1 ] with @xmath115 . hence for some @xmath19 we have @xmath316,}\ ] ] and @xmath317 in particular , @xmath318 .
the arguments that prove claim ii should also prove that @xmath319 has the same _ scaling property _ as @xmath32 , namely the function @xmath320 should be nonincreasing .
if so , one can conclude that @xmath321 for @xmath322 $ ] . to see this ,
observe first that lemma [ le : ech2 ] would imply that the two functions agree at @xmath323 , and second that @xmath153 is the smallest nondecreasing function with the scaling property that takes the given values at @xmath324 .
this holds because when @xmath326 is large the number of integral points in @xmath327 is estimated by its area @xmath328 with error @xmath329 .
therefore @xmath330 as @xmath331 .
since @xmath332 in this case , the result follows . in @xcite ,
hardy and littlewood consider rather different asymptotic questions about the number of lattice points in @xmath287 , looking at asymptotical behavior for fixed @xmath5 rather than comparing the triangles for a given @xmath5 with those for @xmath333 as we do .
interestingly enough the golden ratio @xmath256 also plays a prominent role in their work . given positive numbers @xmath334 with @xmath335 , denote by @xmath336 the list ( with repetitions ) of all nonzero numbers of the form @xmath337 , where @xmath338 are nonnegative integers , put in nondecreasing order .
thus @xmath339 .
write @xmath340 if for all @xmath294 the @xmath114 entry of @xmath336 is no greater than the @xmath114th entry of @xmath341 .
then it is easy to check that @xmath342 thus , if @xmath343 , it would follow that @xmath344 as pointed out by hofer , it is an easy step from here to wonder whether @xmath345 note that the embedding results of guth @xcite and hind kerman @xcite imply that the analogous result in dimensions @xmath346 does _ not _ hold .
however , embedded contact homology also does not exist in dimensions @xmath346 .
the question of what the obstructions actually are in higher dimensions is very interesting . to blow up a point in complex geometry
you remove the point and replace it by the set of all ( complex ) lines through that point . in two complex dimensions , this set is a complex line @xmath347 called the * exceptional divisor*. its normal bundle @xmath348 has chern class @xmath349 . in this context
the basic fact is that @xmath350 is biholomorphic to @xmath351 . on the other hand , in symplectic geometry the picture of blowing up
is somewhat different .
darboux s theorem shows that a neighborhood of a point @xmath352 in @xmath353 can be identified with a neighborhood of @xmath354 in @xmath355 .
therefore one can blow up as before ( with respect to a compatible complex structure ) to get a manifold @xmath356 with blow down map @xmath357 .
but one must put a symplectic form on @xmath356 , and the pullback form @xmath358 vanishes on @xmath146 .
so the symplectic structure @xmath359 on @xmath356 must have the form @xmath360 near @xmath146 , where @xmath361 and @xmath362 nbhd@xmath363 . in this context the basic fact is the following : this means that to blow up symplectically with weight @xmath5 one removes an embedded copy of the open ball @xmath367 and then collapses @xmath368 to @xmath369 by the hopf map ( along the leaves of the characteristic foliation ) .
proposition [ prop:2 ] is proved by using this description of symplectic blowing up to convert the ball packing problem into a question about the existence of symplectic forms on the @xmath114-fold blow up of @xmath208 .
the latter problem is solved using the theory of @xmath78-holomorphic curves as in @xcite .
if @xmath370 the image of @xmath147 under the moment map @xmath371 is the triangle @xmath372 as illustrated on the left of figure [ fig:4 ] , this corresponds to a singular variety , with singular points of orders @xmath373 .
( in contrast , the triangles in figure [ fig : jap2 ] are standard , i.e. equivalent under an integral affine transformation to a triangle with vertices @xmath374 , and so represent the smooth variety @xmath208 . ) we can cut this singular triangle into standard triangles of different sizes . figure [ fig:4 ] also shows that the triangle decomposition on the left is structurally the same as that giving the weights for @xmath375 .
one needs to remove the top right corner of the rectangle and then collapse its top and right side .
as we explain in more detail in @xcite , it also corresponds to a joint resolution of the two singularities of the toric variety corresponding to the _ complement _ of the triangle in the positive quadrant .
figure [ fig:5 ] illustrates another key point .
the cuts needed to resolve the two singularities in the shaded polytope ( the result of removing the singular triangle ) are illustrated on the right .
note that they are _ parallel _ to the cuts that decompose the singular triangle on the left of figure [ fig:4 ] into standard triangles .
( these cuts resolve the singularities because the matrices whose rows are adjacent normal vectors all have determinant @xmath376 . )
the information contained in the weight expansion of @xmath5 can be summarized in a diagram called the _
riemenschneider staircase_. this usually arises in the context of resolving singularities so that the interesting numbers are the coefficients of the hirzebruch jung continued fraction expansions of the two related fractions @xmath377 and @xmath378 ( where we assume @xmath379 ) ; cf .
fulton @xcite .
however one can also construct the diagram from the multiplicities @xmath380 of the entries in the weight expansion @xmath194 .
we illustrate this by the two examples in figure [ fig : jap5 ] .
see popescu - pampu @xcite for an extended discussion of these combinatorics ; and craw
reid @xcite for an example of the use of this diagram in current algebraic geometry .
also compare ( * ? ? ?
* remark 3.12(ii ) ) .
$ ] while ( ii ) illustrates @xmath381 $ ] .
the labels @xmath382 are the smaller numbers by the dots ; the vertical and horizontal numbers give the components of the hirzebruch
jung continued fractions for @xmath383 and @xmath384 .
the arrows give the direction in which one should read the numbers.,width=240 ] observe first that if @xmath373 are relatively prime with @xmath385 then exactly one of the numbers @xmath386 is greater than two .
thus , without loss of generality , we may assume that @xmath383 so that @xmath387 .
given positive integers @xmath388 define @xmath389 = n_0 + \frac1{n_1 + \frac 1{n_2 + \dots + \frac 1{n_k}}}.\ ] ] it is not hard to see that the entries of the weight expansion @xmath390 have multipicities @xmath388 .
moreover @xmath391 for all @xmath211 . in other words ,
the vector @xmath392 has the form @xmath393 where @xmath394 and @xmath395 .
we shall call these renormalized weights @xmath396 the _ labels _ of @xmath397 .
the staircase is constructed from the upper right , proceeding to the left and down , by first placing @xmath398 horizontal dots , then @xmath399 vertical dots ( starting in the row of the first dots ) , then @xmath400 horizontal dots ( starting in the row of the second dots ) , and so on .
the first dot is white , and the others are black .
one can recover the labels from the staircase by starting with @xmath48 at the bottom left and moving back up the staircase according to the rule : the coefficients of the hirzebruch
jung continued fraction for @xmath377 are one more than the number of black dots in each row ( read downwards ) .
if one moves the white dot , putting it in a new row instead of a new column , and constructs the labels as described above , then one gets the labels for the associated fraction @xmath378 .
correspondingly the coefficients of the hirzebruch
jung continued fraction for @xmath378 are one more than the number of blacks dots in each column ( read from right to left ) .
thus figure [ fig : jap5 ] ( i ) gives @xmath401 and @xmath402 while figure [ fig : jap5 ] ( ii ) gives @xmath403 and @xmath404 | as has been known since the time of gromov s nonsqueezing theorem , symplectic embedding questions lie at the heart of symplectic geometry . after surveying some of the most important ways of measuring the size of a symplectic set , these notes discuss some recent developments concerning the question of when a @xmath0-dimensional ellipsoid can be symplectically embedded in a ball .
this problem turns out to have unexpected relations to the properties of continued fractions and of exceptional curves in blow ups of the complex projective plane .
it is also related to questions of lattice packing of planar triangles .
= 1 |
the discovery of new superconductivity in mgb@xmath0 took the world by surprise @xcite .
early studies based on polycrystalline samples lead to a two gap model @xcite .
on the other hand the anisotropy in @xmath5 suggested an anisotropic @xmath6-wave model @xcite .
further it is clear that the simple two gap model can not describe the strong temperature dependence of @xmath7 observed in single crystal mgb@xmath0 @xcite . for this
we need an order parameter @xmath8 of oblate shape @xcite . on the other hand
, it is well known that @xmath9 and @xmath10 for single crystal experiments @xcite , which contradicts the ginzburg - landau phenomenology .
further , both magnetic penetration depth data and @xmath11 data suggest a prolate order parameter as in @xcite . indeed , an earlier stm study suggested a prolate order parameter as well @xcite .
is the order parameter prolate or oblate ?
the answer is that we need both .
we suggest , that the oblate order parameter is attached to the @xmath2-band while the prolate order parameter to the @xmath3-band . in the following
we shall describe salient features of single crystal mgb@xmath0 within the present model .
we just point out two features in the upper critical field which are outside of the ginzburg - landau phenomenology .
\(a ) the strong temperature dependence of the anisotropy parameter @xmath7 @xcite . ( b )
second the deviation from the effective mass model @xmath12 @xcite .
for example @xmath13 , where @xmath14 and @xmath15 can describe the temperature dependence of @xmath16 , as has been shown in ref . @xcite . also , by fitting the experimental data we obtained @xmath17 cm / sec and @xmath18 .
we point out that these values are very consistent with the ones for the @xmath2-band @xcite . by fitting experimental data for @xmath19
@xcite close to @xmath20 we can deduce the ratio @xmath21 , implying @xmath22 , for example . this result is bigger than that estimated in our previous analysis of @xmath23 @xcite .
next order correction in @xmath24 leads to the increase of the ratio @xmath25 for the same values of the parameter @xmath26 . in the future we are going to elaborate our study of the upper critical field behavior in mgb@xmath0 by taking into account impurity scattering , what we believe will improve the agreement with experiment
> from @xmath27-axis oriented films and single crystals , the superfluid density @xmath28 and the lower critical fields @xmath29 and @xmath30 have been extracted reliably @xcite .
it is clear this time that we need a prolate order parameter to fit these data .
we choose @xmath31 with @xmath32 .
these values give a good fit of penetration depth data @xcite .
the magnetic penetration depths are related to the lower critical fields via the formulas @xmath33 where @xmath34 and @xmath35 are the magnetic penetration depth with the supercurrent in the ab - plane and in parallel to the c - axis , respectively .
they are related to the superfluid density via @xmath36 and @xmath37 .
taking @xmath38 , which fits @xmath39 from ref .
@xcite , we calculate @xmath40 and obtain the temperature dependences of @xmath29 and @xmath30 from eqs .
( [ eq2 ] ) and ( [ eq3 ] ) , neglecting the temperature dependence of the logarithms . the result is shown in fig .
[ fig2 ] along with the experimental results from ref . @xcite . from these fits
we obtain @xmath41 mt and @xmath42 mt . from this the ratio of the relevant fermi velocities can be estimated as @xmath43 . in other words ,
the corresponding fermi surface is more isotropic and further @xmath44 .
this strongly suggests that the prolate order parameter we are considering has to be associated with the @xmath3-band .
( dashed line ) and @xmath30 ( solid line ) along with the corresponding experimental data from ref.@xcite . ,
we have seen so far that we need two energy gaps of different shape in order to describe @xmath5 and @xmath45 .
the temperature dependence of @xmath16 indicates that an oblate order parameter ( @xmath46 ) dominates the behavior at high magnetic field .
also we need @xmath47 .
this suggests the cylindrical fermi surface associated with the @xmath2-band as the carrier of this order parameter . on the other hand , for @xmath45
we need a prolate order parameter ( @xmath31 ) .
also , the anisotropy in @xmath45 suggests the @xmath3-band as the carrier of this order parameter . in other words , if we assume that the high field properties are controlled by the oblate order parameter attached to the @xmath2-band , while the low field properties are due to the prolate order parameter attached to the @xmath3-band , we have a consistent picture for superconductivity in mgb@xmath0 . | the recently discovered superconductivity in mgb@xmath0 has created the world sensation . in spite of the relatively high superconducting transition temperature @xmath1 39 k ,
the superconductivity is understood in terms of rare two gap superconductor with energy gaps attached to the @xmath2 and @xmath3-band . however
, this simple model can not describe the temperature dependent anisotropy in @xmath4 or the temperature dependence of the anisotropic magnetic penetration depth . here
we propose a model with two anisotropic energy gaps with different shapes .
indeed the present model describes a number of pecularities of mgb@xmath0 which have been revealed only recently through single crystal mgb@xmath0 . |
a well - known conjecture in the theory of random walk on a group asserts that for the random walk on the symmetric group generated by all permutations of a common cycle structure , the mixing time of the walk depends only on the number of fixed points .
given measure @xmath5 on @xmath0 and integer @xmath6 , denote @xmath7 its @xmath8-fold convolution , which is the law of the random walk starting from the identity in @xmath0 after @xmath8 steps from @xmath5 .
we measure convergence to uniformity in the total variation metric , which for probabilities @xmath5 and @xmath9 on @xmath0 is given by @xmath10 a precise statement of the conjecture is as follows .
[ main_conjecture ] for each @xmath11 let @xmath12 be a conjugacy class of the symmetric group @xmath0 having @xmath13 fixed points , and let @xmath14 denote the uniform probability measure on @xmath12 .
let @xmath15 be a sequence of positive integer steps and for each @xmath16 let @xmath17 be uniform measure on the coset of the alternating group @xmath18 supporting the measure @xmath19 .
for any @xmath20 , if eventually @xmath21 , then @xmath22 if eventually @xmath23 , then @xmath24 this conjecture aims to generalize the mixing time analysis of diaconis and shahshahani @xcite for the random transposition walk .
the formal conjecture seems to have first appeared in @xcite .
when @xmath1 grows with @xmath25 like a constant times @xmath25 the conjecture is known to be false because the walk mixes too rapidly .
this is the work of a number of authors , but first @xcite and for later results see @xcite and references therein . whenever @xmath3 the conjecture is expected to hold , however , in part because the proposed lower bound on mixing time follows from standard techniques in the field .
we give the second moment method proof in appendix [ lower_bound_appendix ] .
since the initial work of diaconis and shahshahani , conjecture [ main_conjecture ] has received quite a bit of attention , see @xcite , @xcite , @xcite , @xcite , @xcite , @xcite , @xcite for cases of conjugacy classes with finite numbers of non - fixed points .
a discussion of ongoing work of schlage - puchta towards the general conjecture is contained in @xcite .
see @xcite and @xcite for broader perspective .
recently berestycki , schramm and zeitouni @xcite have established the conjecture for any set of @xmath1 cycles with @xmath1 fixed as @xmath4 .
moreover , they assert that their analysis will go through to treat the case of any conjugacy class having bounded total length of non - trivial cycles , and may cover the case when the total cycle length grows like @xmath26 , although they state that they have not checked carefully the uniformity in @xmath1 with respect to @xmath25 .
the purpose of this article is to prove conjecture [ main_conjecture ] for the random @xmath1 cycle walk in the full range @xmath3 .
[ main_theorem ] conjecture [ main_conjecture ] holds when @xmath27 is the conjugacy class of all @xmath1 cycles with @xmath1 permitted to be any function of @xmath25 that satisfies @xmath28 as @xmath4 . if @xmath29 then the conclusion of conjecture [ main_conjecture ] remains valid when @xmath30 is any function satisfying @xmath31 . when @xmath29 the window @xmath32 is essentially best possible . as remarked above , the lower bound
was already known .
prior to @xcite , which is purely combinatorial , all approaches to conjecture [ main_conjecture ] have followed the work of diaconis and shahshahani in passing through bounds for characters on the symmetric group .
we return to this previous approach ; in particular , we give an asymptotic evaluation of character ratios at a @xmath1 cycle for many representations , in the range @xmath33 .
the asymptotic formula appears to be new when @xmath34 , see @xcite for the smaller cases .
a precise statement of our result appears in section [ s_n_char_theory_section ] , after we introduce the necessary notation .
the basis for our argument is an old formula of frobenius @xcite , which gives the value of a character of the symmetric group evaluated at a @xmath1 cycle as a contour integral of a certain rational function , characterized by the cycle and the representation . in the special case of a transposition
this formula was already used by diaconis and shahshahani .
previous authors in attempting to extend the result of @xcite had also used frobenius formula , but they had attempted to estimate the sum of residues of the function directly , which entails significant difficulties since nearby residues of the function are unstable once the cycle length @xmath1 becomes somewhat large .
we avoid these difficulties by estimating the integral itself rather than the residues in most situations .
in doing so , a certain regularity of the character values becomes evident .
for instance , while nearby residues appear irregular , by grouping clumps of poles inside a common integral we are able to show that the ` amortized ' contribution of any pole is slowly varying . it may be initially surprising that our analysis becomes greatly simplified as @xmath1 grows .
for instance , once @xmath1 is larger than a sufficiently large constant times @xmath35 , essentially trivial bounds for the contour integral suffice , and the greatest part of the analysis goes into showing that our method can handle the handful of small cases which had been treated previously using the character approach .
a similar feature occurs in the related paper @xcite where a contour integral is also used to bound character ratios at rotations on the orthogonal group .
it seems that the contour method works best when there is significant oscillation in the sum of residues , but is difficult to use in the cases where the residues are generally positive . in appendix [ contour_appendix ]
we show that frobenius formula has a natural generalization to other conjugacy classes on the symmetric group , but with increasing complexity , since one contour integration enters for each non - trivial cycle .
the analysis here applies more broadly than to just the class of @xmath1 cycles , but we have not pushed the method to its limit because it seems that some new ideas are needed to obtain the full conjecture [ main_conjecture ] when the number of small cycles in the cycle decomposition becomes large . from our point of view , the classes containing @xmath36 2-cycles would appear to pose the greatest difficulty .
we remark that , as typical in the character ratio approach , our upper bound is a consequence of a corresponding upper bound for the walk in @xmath37 , so that our result gives a broader set of markov chains on @xmath0 to which one can apply the comparison techniques of @xcite . regarding notation , it will occasionally be convenient to use the vinogradov notation @xmath38 , with the same meaning as @xmath39 .
the implicit constants in notation of both types should be assumed to vary from line to line .
i am grateful to persi diaconis and john jiang for stimulating discussions and to k. soundararajan for drawing my attention to the paper @xcite .
i thank the referees for correcting a number of minor errors in the original submission .
we recall some basic facts regarding the character theory of a finite group . a good reference for these
is @xcite .
a conjugation - invariant measure @xmath5 on finite group @xmath40 is a class function , which means that it has a ` fourier expansion ' expressing @xmath5 as a linear combination of the irreducible characters @xmath41 of @xmath40
. it will be convenient to normalize the fourier coefficients by setting @xmath42 by orthogonality of characters , the fourier expansion takes the form @xmath43 since , writing @xmath44 for the conjugacy class of @xmath45 , @xmath46 in this setting , the plancherel identity is @xmath47 note the somewhat non - standard factor of the inverse of the dimension @xmath48 in the fourier coefficients .
the advantage of this choice is that the fourier map satisfies the familiar property of carrying convolution to pointwise multiplication : for conjugation - invariant measures @xmath49 and
@xmath50 @xmath51 this is because the regular representation in @xmath52 splits as the direct sum over irreducible representations , @xmath53 and convolution by @xmath54 acts as a scalar multiple of the identity on each representation space @xmath55 , the scalar of proportionality being @xmath56 . when @xmath57 is the symmetric group on @xmath25 letters , the irreducible representations are indexed by partitions of @xmath25 , with the partition @xmath58 corresponding to the trivial representation , and partition @xmath59 corresponding to the sign .
given a conjugacy class @xmath27 on @xmath0 and integer @xmath60 , the measure @xmath61 is conjugation invariant , and supported on permutations of a fixed sign , odd if both @xmath27 and @xmath8 are odd , and otherwise even .
we let @xmath62 be the uniform measure on permutations of this sign : @xmath63 with fourier coefficients @xmath64 the total variation distance between @xmath61 and @xmath62 is equal to @xmath65 thus cauchy - schwarz , plancherel , and the convolution identity give the upper bound @xmath66 the main ingredient in the proof of the upper bound of theorem [ main_theorem ] will thus be the following estimate for character ratios .
[ char_ratio_prop ] let @xmath27 be the class of @xmath1 cycles on @xmath0 .
there exists a constant @xmath67 and a constant @xmath68 such that uniformly in @xmath25 , partition @xmath69 and all @xmath70 @xmath71 for the deduction of theorem [ main_theorem ] we will use the following technical result on the dimensions of the irreducible representations of @xmath0 . [ dimension_prop ] for a sufficiently large fixed constant @xmath72 , @xmath73 with the estimate uniform in @xmath25 .
we use the fact that , apart from the trivial and sign representations , the lowest dimensional irreducible representation of @xmath0 has dimension @xmath74 ( so @xmath0 is essentially quasi - random ) .
let @xmath75 as as above , and let @xmath76 .
then for @xmath77 the application of cauchy - schwarz above gives @xmath78 the proof of proposition [ char_ratio_prop ] will require some more detailed information regarding the characters of @xmath0 evaluated at a cycle . we discuss this in the next section .
proposition [ dimension_prop ] is deduced from a very useful approximate dimension formula of larsen - shalev @xcite .
the rather technical proof of this proposition is given in appendix [ dimension_bound_appendix ] .
the irreducible representations of @xmath0 are indexed by partitions of @xmath25 . given a partition @xmath79 the dual partition @xmath80 is found by reflecting the diagram of @xmath81 along its diagonal .
we write @xmath82 for the character of representation @xmath83 , and @xmath84 for the dimension .
let @xmath85 .
then the dimension is given by ( see e.g. @xcite ) @xmath86 setting apart those terms that pertain to @xmath87 , we find @xmath88 the product is bounded by 1 , and @xmath89 , so that we have the bound employed by diaconis - shahshahani @xmath90 the trivial representation is @xmath91 while the sign representation is @xmath92 , both one - dimensional .
@xmath93 corresponds to the standard representation , which is the irreducible @xmath94-dimensional sub - representation of the representation @xmath95 in @xmath96 given by @xmath97 this representation and its dual are the lowest dimensional non - trivial irreducible representations of @xmath0 .
given the character @xmath82 , the character of the dual representation is given by @xmath98 the characters corresponding to partitions with long first piece have relatively simple interpretations .
for instance , if we write @xmath99 for the number of fixed points and @xmath100 for the number of 2-cycles in permutation @xmath101 then ( @xcite , ex .
4.15 ) @xmath102 it is now immediate that @xmath103 also , @xmath104 we use these formulas in the proof of the lower bound of conjecture [ main_conjecture ] .
many properties of partitions are most readily evident in frobenius notation , and we will use this notation to state our main technical theorem . in frobenius notation
we identify the partion @xmath69 by drawing the diagonal , say of length @xmath105 , and measuring the legs that extend horizontally to the right and vertically below the diagonal : @xmath106 @xmath107 notice that @xmath108 in appendix [ dimension_bound_appendix ] we consider the quantity @xmath109 , which is the number of boxes contained neither in the square formed by the diagonal nor in the first row .
the notation used here is summarized in figure [ frobenius_figure ] .
we now state our asymptotic evaluation of the character ratio .
[ character_ratio_theorem ] let @xmath25 be large , let @xmath110 , and let @xmath27 be the class of @xmath1 cycles on @xmath0 .
let @xmath111 be a partition of @xmath25 with frobenius notation @xmath112 . 1 . ( long first row )
let @xmath113 , let @xmath114 and suppose that @xmath115 .
then @xmath116\right)\right ) .\end{aligned}\ ] ] if @xmath117 then the error term is actually 0
( large @xmath1 , short first row and column ) let @xmath118 .
there exists @xmath119 such that , for all @xmath25 sufficiently large , for all @xmath1 with @xmath120 and for all @xmath69 such that @xmath121 , @xmath122 3 .
( asymptotic expansion ) let @xmath123 and suppose now that @xmath124 .
we have the approximate formula @xmath125 this theorem is the main technical result of the paper . actually it gives more than we will need , and we apply the detailed statement of part c. only in the case when the cycle length @xmath1 is relatively short , @xmath126 .
the cruder bounds of parts a. and b. suffice when the cycle length @xmath1 is larger .
when @xmath27 is the class of @xmath1 cycles frobenius @xcite ( see @xcite , p.52 ex .
4.17 b ) proved a famous formula that expresses the character ratio of a given representation at @xmath27 as the ` residue at @xmath127 ' of a meromorphic function depending upon the representation . in frobenius notation , and using the falling power @xmath128 the formula is @xmath129 where the integration has winding number 1 around each ( finite ) pole of the integrand .
[ note that our @xmath130 correspond to @xmath131 of @xcite , and replace @xmath132 there with our @xmath133 . ]
our proof of theorem [ character_ratio_theorem ] is an asymptotic estimation of this integral .
we first record several properties of @xmath134 in the following lemma .
[ f_properties ] let @xmath135 be a partition of @xmath25 in frobenius notation . 1 .
each of @xmath136 and @xmath137 has at most @xmath138 poles . 2 . if @xmath139 then @xmath140 is holomorphic
. denote by @xmath141 the partition of @xmath142 found by deleting the first row of @xmath81 . 1 .
we have @xmath143 2 . if @xmath144 and if @xmath145 then @xmath140 has a single simple pole at @xmath146 .
the first item follows from the bound for the diagonal @xmath147 , since both @xmath148 and @xmath149 are products of @xmath105 terms . for 2 . , first observe that @xmath150 and @xmath151 are strictly decreasing so that the poles of @xmath152 are all simple , as are the poles of @xmath137 .
since @xmath153 , @xmath154 , so @xmath155 .
thus the poles of @xmath156 are all simple , and are all cancelled by the factor of @xmath157 . for 3 . , observe that deleting the first row of the diagram for @xmath81 shifts the diagonal down one square . thus , if @xmath158 then @xmath159 and @xmath160 in this case , accounting for the lost factor from @xmath161 , @xmath162 if instead @xmath163 then @xmath164 , which implies @xmath165 , and @xmath166 thus ( [ factor_drop ] ) still holds , the new factor accounting for the introduction of @xmath167 .
thus , in either case , @xmath168 for 4 . , the condition @xmath144 and @xmath169 implies that @xmath170 .
thus , for all @xmath171 , @xmath172 , so that @xmath173 , and @xmath174 is a simple pole of @xmath156 .
furthermore , it is not cancelled by @xmath175 , so that @xmath176 is a simple pole of @xmath140 .
it is the only pole , since @xmath177 has @xmath178 , hence is holomorphic .
parts a. and b. of theorem [ character_ratio_theorem ] are more easily proven .
we give these proofs immediately , and then prove several lemmas before proving part c. in this part , the first row @xmath87 of partition @xmath81 is significantly larger than the remainder of the partition , of size @xmath179 .
we extract a residue contribution from the pole corresponding to @xmath87 , and bound the remainder of the character ratio by approximating it with a character ratio on @xmath180 .
observe that @xmath181 . recall that we assume @xmath182 , and that this part of the theorem is the asymptotic @xmath183\right)\right),\end{aligned}\ ] ] with the error equal to 0 if @xmath184 .
the poles of @xmath134 are among the points @xmath185 , @xmath186 , @xmath187 , some of which may be cancelled by the numerator .
the condition @xmath188 guarantees that @xmath189 since @xmath190 for @xmath191 , @xmath192 , it follows that @xmath176 is the furthest pole from zero of the function @xmath193 , and that this is a simple pole of @xmath140 .
set @xmath194 and notice @xmath195 so that @xmath176 is outside the loop @xmath196 , while all other poles are inside .
thus @xmath197 the residue term is equal to the main term of the theorem , so it remains to bound the integral .
recall the notation @xmath198 , and that @xmath199 thus we express the integral of ( [ error_integral ] ) as @xmath200 when @xmath201 , @xmath202 is holomorphic , and so ( [ residue_extracted ] ) is zero , which proves the latter claim of the theorem .
thus we may now assume that @xmath203 . on the contour @xmath196 ,
@xmath204 so that @xmath205 thus ( [ residue_extracted ] ) is equal to ( @xmath206 is the @xmath1 cycle class on @xmath180 ) @xmath207 we bound the first term by @xmath208 , since all character ratios are bounded by 1 . to bound the integral , note that @xmath209 and @xmath210 each have at most @xmath211 poles
. on the contour @xmath196 , @xmath212 and the terms in @xmath210 are bounded similarly .
it follows that on @xmath196 , @xmath213 meanwhile , also on @xmath214 , @xmath215 putting these bounds together , we deduce a bound for the second term of ( [ contracted_evaluation ] ) of @xmath216 \right)\right).\end{aligned}\ ] ] to complete the proof of the theorem , use @xmath217 and note that the term @xmath208 from the character ratio in ( [ contracted_evaluation ] ) is trivially absorbed into this error term .
choose @xmath218 with @xmath219 .
since the poles of @xmath220 are among @xmath185 and @xmath221 , @xmath222 , choosing @xmath223 sufficiently small , @xmath224 and @xmath225 guarantees that the contour @xmath226 contains all poles of @xmath227 .
thus @xmath228 write @xmath229 . since @xmath230 , for @xmath196 , @xmath231 with @xmath232 indicating a quantity that may be made arbitrarily small with a sufficiently small choice of @xmath233 .
also , @xmath234 and @xmath235 are each composed of at most @xmath236 factors , each of size at most @xmath237 .
we deduce that for @xmath196 , @xmath238 the error term requiring that first @xmath233 be small , and then that @xmath25 be large .
the length of the contour is @xmath239 . since @xmath240
, it follows that the bound @xmath241 holds for the character ratio for all @xmath25 sufficiently large .
we now turn to part c. of theorem [ character_ratio_theorem ] .
we will again estimate the integral @xmath242 but the idea now will be to evaluate clumps of poles together . with an appropriate choice of contour , at any given point
the poles that are sufficiently far away make an essentially constant contribution to the integrand , and so may be safely removed , simplifying the integral .
denote by @xmath243 the multi - set ( that is , set counted with multiplicities ) of poles of @xmath136 and @xmath137 .
one trivial fact concerning the distribution of poles in @xmath244 is the following bound .
[ trivial_upper_bound ] for any @xmath245 , @xmath246 indeed , this follows from the fact that @xmath247 a useful consequence is that for @xmath248 real , @xmath249 , we can always find a real point @xmath132 nearby @xmath248 with large distance from all poles in @xmath244 .
let @xmath250 and let @xmath251 be a parameter .
we say that a real number @xmath132 is @xmath252__-well - spaced _
_ for @xmath248 with respect to @xmath244 if the bound is satisfied @xmath253 the following lemma says that if @xmath254 is sufficiently large , then there are always many points @xmath132 nearby @xmath248 that are well - spaced for @xmath248 with respect to @xmath244 . [ point_spacing ]
let @xmath255 and let @xmath256 .
let @xmath248 be real with @xmath257 .
then there exists a real @xmath132 with @xmath258 which is @xmath259-well - spaced for @xmath248 with respect to @xmath244 .
let @xmath260 be the interval @xmath261 .
by lemma [ trivial_upper_bound ] , the interval @xmath260 contains at most @xmath262 poles of @xmath244 .
deleting the segment of radius @xmath1 around each pole in @xmath263 removes a set of total length at most @xmath264 leaving @xmath265 with @xmath266 .
now @xmath267 applying the cardinality bound of lemma [ trivial_upper_bound ] a second time ( recall @xmath268 ) we obtain a bound for ( [ average_of_p_to_y ] ) of @xmath269 it follows that a typical point @xmath270 is @xmath271-well - spaced for @xmath248 .
recall that @xmath272 .
we now assume , as we may , that @xmath25 is sufficiently large to guarantee @xmath273 and we choose a sequence of well - spaced points at which we partition our integral . to find our sequence of points , first set @xmath274 and @xmath275 for those @xmath276 such that @xmath277 .
in each interval @xmath278 apply lemma [ point_spacing ] to find @xmath279 , a @xmath280-well - spaced point for @xmath281 . also find a point @xmath282 in each interval @xmath283 $ ] which is @xmath280-well - spaced for @xmath284 .
thus we have the disjoint intervals @xmath285 , \quad j = 1 , 2 , 3 , ... \ ] ] which together cover @xmath286 let @xmath287 ( resp .
@xmath288 ) be the number of poles of @xmath244 contained in @xmath289 , ( resp .
@xmath290 ) . around each interval
@xmath291 @xmath192 we draw a rectangular box @xmath292 and similarly around @xmath293 . if @xmath294 then the box may be discarded .
we also draw a large box @xmath295 containing the origin , with endpoints at @xmath296 . a permissible contour with which to apply integral formula ( [ contour_short_form_again ] )
is given by @xmath297 each box being positively oriented , see figure [ contour_figure ] .
we use a somewhat shortened version of this contour . in what follows
we treat only the integrals around the boxes @xmath298 and @xmath295 , and we drop superscripts ( so we write e.g. @xmath299 for @xmath287 , @xmath300 for @xmath289 etc ) .
the argument may be carried out symmetrically for @xmath301 .
the main proposition is as follows .
let @xmath192 .
there exists a piecewise linear contour @xmath302 , having total length @xmath303 such that @xmath302 has winding number 1 around each pole @xmath304 and winding number 0 around all poles @xmath305 . for @xmath306 , @xmath140 satisfies @xmath307 recall that @xmath308 is a box containing @xmath299 poles and surrounding interval @xmath309 , with corners at @xmath310 recall also that @xmath311 and @xmath312 , so that @xmath313 . to form @xmath302
we shorten the contour @xmath314 . at each pole
@xmath315 consider the segment @xmath316.\ ] ] if there exists a subinterval @xmath317 of interval @xmath300 that is not covered by @xmath318 then such @xmath317 may be discarded from the box @xmath308 by drawing vertical segments at the endpoints of @xmath317 and deleting the parts of @xmath308 vertically above and below @xmath317 .
deleting subintervals in this way if it reduces the total perimeter , or allowing them to remain when the perimeter is increased , we arrive at a contour @xmath302 which is the union of some rectangles , and has total length at most @xmath319 the first term bounding the result of doing nothing , and the second bounding the length of a contour with an individual box around every pole . using @xmath320 , we
have the bound @xmath321 we now prove the formula ( [ approximate_formula ] ) .
observe that by lemma [ trivial_upper_bound ] , the number of poles @xmath299 in @xmath322 satisfies @xmath323 , so that each point of @xmath302 has distance @xmath324 from @xmath281 .
thus @xmath325 in particular , it follows that @xmath326 we prove @xmath327 and @xmath328 inserted in ( [ first_approximation_j ] ) these combine to prove the formula ( [ approximate_formula ] ) , ( use @xmath329 ) .
we first consider ( [ distant_poles_bound ] ) . this estimate is equivalent to @xmath330 the sign determined according as @xmath331 is a pole from @xmath148 or @xmath149 .
say a pole @xmath332 is ` good ' if @xmath333 and ` bad ' otherwise . for @xmath334 , @xmath335
if @xmath25 is sufficiently large , which implies that for good @xmath331 , @xmath336 .
since the total number of poles is @xmath324 , the product over good poles evidently satisfies the bound @xmath337 since it is not contained in the box @xmath308 , a bad pole @xmath331 necessarily lives in the complement @xmath338 $ ] , and either satisfies @xmath339 and @xmath340 or else @xmath341 , which entails @xmath342 . since @xmath343 , and @xmath344
( at least if @xmath25 is sufficiently large ) , it follows that in the second case , @xmath345 .
thus , for @xmath346 , the fact that @xmath347 implies @xmath348 by invoking the @xmath271 well - spaced property of @xmath349 and @xmath350 .
it follows @xmath351 proving ( [ distant_poles_bound ] ) . to prove ( [ near_poles_bound ] ) we consider two cases .
when either @xmath352 or @xmath353 , observe that for each @xmath354 , @xmath355 also , since @xmath356 , @xmath357 thus , invoking the well - spaced property of @xmath349 and @xmath350 , @xmath358 so that if @xmath359 , @xmath360 and similarly when @xmath353 .
when @xmath361 then @xmath362 for each @xmath363 , so that in this case the bound @xmath364 is immediate from @xmath365 .
thus ( [ near_poles_bound ] ) holds in either case .
we now bound integration around the box @xmath295 .
[ cb_0_bound ] assume @xmath255 .
we have @xmath366 since @xmath295 is a box with corners at @xmath367 , the integral has length @xmath368 , so it will suffice to prove @xmath369 recall that @xmath370 recall also that @xmath371 and @xmath372 .
thus , on @xmath373 we have @xmath374 .
hence , using @xmath375 , @xmath376 the last bound requiring @xmath377 , which plainly holds for @xmath255 .
since @xmath378 when @xmath375 , we obtain the required bound for @xmath379 by checking that @xmath380 to do so , for each @xmath381 write the corresponding factor of @xmath234 or @xmath235 as @xmath382 for a given @xmath383 , say the pole @xmath384 is ` good ' for @xmath385 if @xmath386 . since the total number of poles in @xmath244 is @xmath324 , the part of the product in @xmath193 contributed by good poles is bounded by @xmath387 , so we may consider only the bad poles .
bad poles exist only if @xmath385 is on one of the two vertical sides of the box , that is @xmath388 .
notice that , e.g. @xmath389 ( similarly if @xmath331 is near @xmath390 ) , and so bad poles satisfy @xmath391 . by the well - spaced property of @xmath392 ,
when @xmath393 we have @xmath394 it follows that @xmath395 we now have all of the estimates that we need to complete the proof of theorem [ character_ratio_theorem ] .
we use the following integral formula .
[ exact_integral ] let @xmath396 be a sequence of poles and let @xmath397 be an arbitrary complex number . then @xmath398 where the contour is such that it has winding number one about each pole .
the integral is independent of the appropriate contour , so take the contour to be a large loop containing the poles .
we may write the integrand as @xmath399 differentiating the integrand with respect to @xmath400 results in a rational function having total degree @xmath401 .
it follows that @xmath402 for each @xmath276 , since the integral can be made made arbitrarily small by taking the contour to be sufficiently large .
taking @xmath403 for each @xmath276 , we find that the residue at 0 is @xmath404 .
write @xmath405 the contribution from @xmath373 is bounded in lemma [ cb_0_bound ] by @xmath406 which may be absorbed into the error term of the theorem . for each @xmath192
, write the contribution from the @xmath287 poles inside @xmath407 as @xmath408 dz.\ ] ] for the main term of this integral , lemma [ exact_integral ] gives the evaluation @xmath409 the error resulting since each of the @xmath1 terms of @xmath410 is equal to @xmath411 we use that @xmath412 here . had we considered @xmath413 rather than @xmath407 this main term would involve a sum over the @xmath151 , with an appropriate sign factor .
the integral of the error term over @xmath302 is bounded trivially by using @xmath414 , which gives @xmath415 the ratio of the error from @xmath302 to the corresponding main term is @xmath416 and since each @xmath150 in the sum above is within constants of @xmath281 , this proves the theorem . note that in the claimed formula of the theorem , the sums over @xmath150 and @xmath417 contain terms for which @xmath418 and @xmath419 , which do not appear in our evaluation .
however , only a bound , rather than an asymptotic , is claimed in this range , so that the theorem is valid with the extra terms discarded .
we now prove the upper bound on mixing time from theorem [ main_theorem ] by proving proposition [ char_ratio_prop ] .
recall that this proposition is a bound for character ratios at class @xmath27 of @xmath1 cycles , @xmath420 uniformly for @xmath1 less than a fixed constant times @xmath421 and for all @xmath422 . before proving the estimate
, we collect together several observations .
first note that the character ratio bound is trivial for the one - dimensional representations @xmath423 .
also , exchanging @xmath81 with its dual @xmath80 leaves the dimension @xmath424 unchanged , and at most changes the sign of @xmath425 , so we will assume without loss of generality that @xmath81 has @xmath426 .
we set @xmath427 .
[ the_target_lemma ] to prove the character ratio bound ( [ target_char_ratio_bound ] ) , it suffices to prove that for all @xmath25 larger than the maximum of @xmath1 and a fixed constant , and for a sufficiently large @xmath428 , that for all non - trivial @xmath81 with @xmath429 , @xmath430 by the bound ( [ diaconis_shahshahni_dim_bound ] ) for the dimension , we have ( use @xmath431 ) @xmath432 so that a bound of @xmath433 is sufficient . on the other hand , for all @xmath81 , @xmath434 , so that a bound of @xmath435 also suffices .
our proof splits into two cases depending upon whether @xmath436
. the essential tool in both cases will be the evaluation of the character ratio from theorem [ character_ratio_theorem ] .
for large @xmath1 we will use only parts a. and b. of that theorem .
recall that part a. had a main term equal to @xmath437 [ mt_lemma ] assume @xmath438 .
we have the bound @xmath439 recall @xmath440 .
we estimate @xmath441 when @xmath442 we have @xmath443 .
let @xmath444 , and note that @xmath445 .
thus @xmath446 so that part b of theorem [ character_ratio_theorem ] guarantees that there exists @xmath447 , such that , for @xmath25 larger than a fixed constant , for all @xmath448 , @xmath449 thus the second condition of lemma [ the_target_lemma ] is satisfied .
so we may suppose that @xmath450 and appeal to part a. of theorem [ character_ratio_theorem ] .
suppose that @xmath451 is sufficiently small so that that @xmath452 . if @xmath117 then the error term of part a. of theorem [ character_ratio_theorem ] is zero so that the previous lemma implies @xmath453 thus the first criterion of lemma [ the_target_lemma ] is satisfied with @xmath454 .
assume now that @xmath203 .
let now @xmath451 sufficiently small so that we may choose @xmath455 in part a , that is , @xmath456 , and also assume that @xmath457 for some fixed @xmath458 .
then part a. gives @xmath459 with @xmath460\right ) \leq 2^{-k } \ ; ( n \text { large}).\ ] ] since @xmath461 we deduce that @xmath462 since @xmath463 and @xmath464 we deduce @xmath465 so that the first criteria of lemma [ the_target_lemma ] is satisfied .
when @xmath466 we make essential use of the asymptotic evaluation of the character ratio proved in part c. of theorem [ character_ratio_theorem ] .
the next lemma shows that we may restrict attention to only the main term of that evaluation .
[ condition_lemma ] let @xmath467 .
we have the bound @xmath468 + o\left(\frac{e^{-k}(\log n)^4}{n^{\frac{1}{4}}}\right ) .
\end{aligned}\ ] ] in particular , if @xmath469 and @xmath25 is sufficiently large , and if @xmath470 then after changing constants , the same estimate holds for @xmath471 , so that the condition of lemma [ the_target_lemma ] is satisfied . to deduce the second statement from the first , note that @xmath472 which , for @xmath473 may be absorbed into the rhs of the second statement by increasing the value of @xmath474 . regarding the error of @xmath475 , it is no loss of generality to assume that @xmath476 so that this error term has relative size @xmath477 . again
, the logarithm of this error may be absorbed by increasing @xmath474 . to prove the first statement , part c. of theorem [ character_ratio_theorem ] gives @xmath478 for @xmath467 , the last error term is plainly @xmath479 .
split the sum over @xmath150 according to @xmath480 .
thus the sum over @xmath150 is bounded by @xmath481 in the last sum , @xmath482 is minimized at @xmath483 . thus @xmath484 handling the sum over @xmath151 in the same way proves the lemma .
recall that we require @xmath426 and that @xmath440 .
thinking of @xmath179 as fixed , set @xmath485 . since @xmath486 , an upper bound for @xmath487
is given by the solution to the following optimization problem .
let @xmath488 be real variables and let @xmath489 .
@xmath490 let @xmath491 it is easily checked by varying parameters that an optimal solution of this problem has @xmath492 , and @xmath493 , @xmath494 for @xmath495 , which yields the maximum @xmath496 @xmath497 being the solution of the continuous analogue of the optimization problem @xmath498 which is less constrained .
[ calculus_lemma ] let @xmath489 and let @xmath499 .
assume @xmath500 .
there exists a @xmath501 such that if @xmath502 then @xmath503 for all @xmath504 we have @xmath505 set @xmath506 .
we may assume that @xmath507 .
then the first bound for the maximum reduces to @xmath508 and we require the statement @xmath509 write the left hand side as @xmath510 .
then it is readily checked with calculus that this is decreasing in @xmath1 , hence maximized at @xmath511 .
since @xmath512 it follows that the left is bounded by the right for @xmath451 sufficiently small . for the second statement , we use the second bound of the maximum , so that we need to show @xmath513 . the worst case is @xmath511 , which reduces to the true inequality @xmath514 for @xmath515 .
we can now prove the case @xmath516 of proposition [ char_ratio_prop ] .
as usual , assume @xmath426 . as in the proof of the proposition in the case @xmath517 , we write @xmath518 . for @xmath519
we appeal to part a. of theorem [ character_ratio_theorem ] , with the main term @xmath520 from lemma [ mt_lemma ] .
this gives @xmath521 in this range , @xmath522 , so that , since @xmath489 , the error term is negligible , and the condition of the first criterion lemma , lemma [ the_target_lemma ] , is satisfied .
now suppose @xmath469 .
let @xmath523 be the constant from lemma [ calculus_lemma ] .
let @xmath524 be a constant , such that , for @xmath525 , @xmath526 . notice that this implies @xmath527 . for @xmath528 ,
lemma [ calculus_lemma ] implies that @xmath529 so that this quantity is bounded by the first term in the maximum of the small @xmath1 criterion lemma , lemma [ condition_lemma ] .
thus we may assume that @xmath530 .
in this case , using that @xmath531 , the second bound of lemma [ calculus_lemma ] implies that for a sufficiently large constant @xmath474 @xmath532 thus again this is bounded by the first term in the maximum of lemma [ condition_lemma ] .
in this appendix we show a proof of the lower bound in conjecture [ main_conjecture ] for any conjugacy class @xmath27 on @xmath0 having @xmath533 non - fixed points .
let the number of 2-cycles in @xmath27 be @xmath534 .
the proof goes by comparing the distribution of the number of fixed points in a randomly chosen permutation , chosen either according to uniform measure or to @xmath535 .
note that expectation against uniform measure at step @xmath8 is the same as calculating the @xmath37 inner product with @xmath536 . by the formulas in ( [ small_chars ] ) , @xmath537 is one less than the number of fixed points in @xmath101 , and @xmath538 thus according to uniform measure @xmath539 has mean 0 and variance @xmath540 . with respect to @xmath535 we readily check that for any @xmath69 , ( see ( [ fourier_expansion ] ) and ( [ convolution ] ) ) @xmath541 thus , @xmath542 and similarly the second moment generates contributions from @xmath543 , @xmath539 , @xmath544 and @xmath545 , @xmath546 obviously @xmath547 also , since @xmath548 , @xmath549 for @xmath550 , @xmath20 we deduce @xmath551 let @xmath552 applying chebyshev s inequality , first with measure @xmath62 and then with measure @xmath553 , we deduce that @xmath554 which completes the proof . note that if @xmath29 then we may take @xmath555 on the scale of @xmath556 .
recall the dimension formula ( @xmath557 ) @xmath558 in our character ratio bounds we set apart the terms pertaining to @xmath87 to find @xmath88 of course , one could extract more rows from the partition @xmath81 , and , since @xmath559 , removing columns is also possible .
this makes plausible the following useful result of larsen - shalev concerning dimensions .
[ larsen_shalev_theorem ] write @xmath69 in frobenius notation with diagonal of length @xmath105 , and set @xmath560 , @xmath561
. then @xmath562 where @xmath563 the error term holds as @xmath4 and is independent of the partition @xmath564 . using this theorem , we now prove proposition [ dimension_prop ] . in this proof
we suppress the @xmath565 on @xmath150 and @xmath151 .
also , we write @xmath566 to indicate strictly decreasing length @xmath105 vectors of non - negative natural numbers , and given such a vector @xmath567 we write @xmath568 .
recall that we intend to show that for a sufficiently large fixed @xmath428 , uniformly in @xmath25 we have the estimate @xmath569 in view of theorem [ larsen_shalev_theorem ] , it suffices to prove instead that , for a possibly larger value of @xmath474 , @xmath570 by stirling s approximation , we reduce to showing that ( the @xmath565 on the sum indicates that if either @xmath571 or @xmath572 , its contribution is excluded ) @xmath573 use that @xmath574 to cancel the @xmath575 term . thus we seek a bound of @xmath576 for @xmath577 by symmetry we may assume that @xmath426 .
set @xmath578 , and @xmath579 .
thus @xmath109 is the number of boxes of partition @xmath81 that are neither contained in the first row , nor contained in the square determined by the diagonal .
we consider separately the cases @xmath580 and @xmath581 .
first consider @xmath580 .
evidently @xmath582 so that , for fixed @xmath583 satisfying @xmath584 , the sum inside the exponential of ( [ ab_sum ] ) is bounded by @xmath585 thinking of @xmath586 as fixed , we wish to bound the number of possible strings @xmath587 , @xmath588 . to do this ,
observe that the numbers @xmath589 together with @xmath590 are a pair of partitions of combined length @xmath109 . from the well - known bound for @xmath591
the number of partitions of an integer @xmath252 , it follows that the number of possible choices for @xmath592 , @xmath593 ( or , equivalently @xmath594 ) is at most @xmath595 we can now bound the contribution to ( [ ab_sum ] ) from @xmath596 .
it is bounded by @xmath597 exchanging the order of summation , this has the bound @xmath598 when @xmath581 we bound the sum inside the exponential of ( [ ab_sum ] ) crudely .
notice that @xmath599 , so that @xmath600 .
thus the sum inside the exponential of ( [ ab_sum ] ) is bounded by @xmath601 the total number of such possible @xmath602 is at most the number of partitions of @xmath25 , which is @xmath603 .
it follows that the contribution of @xmath604 to ( [ ab_sum ] ) is at most @xmath605 combining this bound with the bound for small @xmath109 proves the proposition .
we conclude by deriving the extension of frobenius character formula to character ratios of arbitrary conjugacy classes on @xmath0 .
this derivation is modeled upon the derivation of frobenius formula from @xcite , p. 118 .
let @xmath613 be two partitions of @xmath25 with @xmath614 indexing an @xmath615 dimensional representation of @xmath0 , and @xmath616 indexing a conjugacy class .
following @xcite , we start from the classical fact that the character @xmath617 is equal to the coefficient of the monomial @xmath618 in @xmath619 the key calculation of @xcite is that , for any @xmath620 and for any @xmath5 as above , @xmath621 the calculation is as follows .
expanding @xmath109 , @xmath622\\ & = \frac{m!}{\mu ! } \det\left[\mu_i^{\underline{\sigma_j-1}}\right ] = \frac{m!}{\mu!}\delta(\mu).\end{aligned}\ ] ] in particular , this gives the dimension formula , @xmath623 and , when @xmath624 is the class of a single @xmath1-cycle , @xmath625 here , a term @xmath171 with @xmath626 is to be omitted .
thus @xmath627 expanding the vandermonde , one may check by inspection that the sum of ( [ k_cycle_ratio ] ) is equal to the sum of the ( finite ) residues of the meromorphic function @xmath628 this last fact is the version of frobenius formula derived in @xcite .
the equivalence between this formulation from @xcite and the integral formula used throughout the present paper , @xmath629 follows from the identity @xmath630 see e.g. @xcite pp . 5152 .
now consider a conjugacy class with non - trivial cycles @xmath631 so that @xmath632 . the appropriate generalization of ( [ k_cycle_formula ] ) and ( [ k_cycle_ratio ] ) are @xmath633 again with the convention that a term with negative coordinate is to be omitted .
it follows @xmath634 the formula ( [ general_formula ] ) follows , as does ( [ k_cycle_formula ] ) , by expanding @xmath635 let @xmath69 index an irrep . of @xmath0 , and let @xmath636 , @xmath637 denote the non - trivial cycles in conjugacy class @xmath27
. then @xmath638 with @xmath639 and each @xmath640 given by @xmath641 the contours @xmath642 are to be chosen such that @xmath643 has winding number 1 about each pole of @xmath644 , and in general , @xmath645 has winding number 1 about each pole of @xmath646 and also , viewed as loops in a single complex plane , encloses @xmath647 for all @xmath648 .
the case @xmath651 is frobenius formula .
suppose that @xmath652 and that the formula holds in case @xmath653 .
let @xmath654 to indicate @xmath655 with @xmath656 omitted and similarly @xmath657 . write @xmath658 invoking the inductive assumption , the rhs becomes @xmath659_{z_1 = \mu_{i_1}}\\ & \qquad \times \left(\frac{1}{2\pi i}\right)^{r-1 } \oint\cdots \oint_{{{\mathcal{c}}}_2 \times ... \times { { \mathcal{c}}}_{r } } g_{\mu - k_1 { { \mathbf{e}}}_{i_1}}^{\hat{{{\mathbf{k}}}}_1 } ( \hat{{{\mathbf{z}}}}_1 ) dz_2 ... dz_{r}. \notag\end{aligned}\ ] ] since the poles of @xmath662 are at @xmath663 , for any fixed @xmath664 , for any choice of contour @xmath643 having winding number 1 about the poles of @xmath665 and winding number zero about @xmath666 , @xmath667_{z_1 = \mu_{i_1}}g_{\mu - k_1 { { \mathbf{e}}}_{i_1}}^{\hat{{{\mathbf{k}}}}_1 } ( \hat{{{\mathbf{z}}}}_1 ) .\ ] ] in view of the restriction on @xmath668 , the identity follows . | we study the random walk on the symmetric group @xmath0 generated by the conjugacy class of cycles of length @xmath1 .
we show that the convergence to uniform measure of this walk has a cut - off in total variation distance after @xmath2 steps , uniformly in @xmath3 as @xmath4 .
the analysis follows from a new asymptotic estimation of the characters of the symmetric group evaluated at cycles . |
wu and vos @xcite introduce a parameter - free distribution estimation framework and utilize the kullback leibler ( kl ) divergence as a loss function .
they show that the kl risk of a distribution estimator obtained from an i.i.d .
sample decomposes in a fashion parallel to the mean squared error decomposition for a parameter estimator , and that an estimator is distribution unbiased , or simply unbiased , if and only if its distribution mean is equal to the true distribution .
distribution unbiasedness can be defined without using any parameterization .
we call this approach parameter - free even though there may be applications where it is desirable to use a particular parameterization .
when the distributions are , in fact , parametrically indexed , distribution unbiasedness handles multiple parameters simultaneously and is consistent under reparametrization .
wu and vos @xcite also show that the mle for distributions in the exponential family is always distribution unbiased .
the kl expectation and variance functions @xmath3 and @xmath4 are defined by minimizing over the space of all distributions .
these functions completely describe an estimator in terms of its kl divergence around any distribution . in this paper
, we introduce distribution expectation and variance functions @xmath0 and @xmath1 that are defined by minimizing over a smaller space of distributions . for exponential and mixture families , the expected kl risk is a function only of these quantities . even though the focus of this paper is on parametric exponential families ,
our approach is parameter - free in that the definitions and results are provided without regard to the parameterization of the family .
there are three advantages to this approach : one , the lack of invariance of bias across parameter transformations is avoided ; two , we can allow for estimators taking values outside of the exponential family ; three , the case where the true distribution does not belong to the family is easily addressed .
section [ sec : kullback - leibler - risk ,- variance , ] introduces the distribution expectation and variance functions and shows how these are a generalization of the mean and expectation functions for mean square error .
exponential families and their extension are discussed in section [ sec : exponential - family ] .
the fundamental properties of the distribution mean and variance functions allow using the ideas of rao
blackwell @xcite to show that the mle is the unique uniformly minimum distribution variance unbiased estimator ( umv@xmath2ue ) .
this result is proved in section [ sec : rao - blackwell - and - the ] .
three examples are given in section [ sec : examples ] and section [ sec : discussions ] contains further remarks .
the parametric version of the rao
blackwell theorem can be proved using a pythagorean relationship that holds for mean square error ( mse ) and the expectation operator . to prove the distribution version of the rao
blackwell theorem , we use a similar relationship that holds for kl risk and the kl expectation along with a second pythagorean relationship that holds in exponential families for kl divergence and the kl projection .
basic properties of the expectation operator for real - valued random variables used in the proof can be extended to distribution - valued random variables .
we begin with the property that the expectation minimizes the mse . for (
real - valued ) random variable @xmath5 and @xmath6 we can define the average behavior of @xmath5 relative to @xmath7 using the risk function @xmath8,\ ] ] where @xmath9 is a loss function , that is , a nonnegative convex function on @xmath10 . when @xmath11<\infty$ ] for some @xmath7 , we define @xmath12\ ] ] and @xmath13\ ] ] if the minimum exists , in which case , @xmath14.\ ] ] when @xmath15 , that is , risk is mse , we have @xmath16 = \int y\,\mathrm{d}r_{0}\defeq ey , \\
\label{eq : vequiv}v_{l}y&\defeq&\inf_{b\in\mathbb{r}}e\bigl[l(y , b)\bigr ] = e \bigl[l(y , ey)\bigr]\defeq vy.\end{aligned}\ ] ] note that we use the loss function as subscript to indicate expectation and variance defined in terms of an argmin and infimum of the loss function , while expectations and variances without a subscript are defined in terms of an integral , or in terms of a sum if the sample space is discrete .
the middle equality signs in equations ( [ eq : eequiv ] ) and ( [ eq : vequiv ] ) are well - known results for @xmath17 and @xmath18 .
these two values completely characterize the risk because of the relationship @xmath19=\l ( e_{l}y , a ) + \v_{l}y \quad\quad\forall a \in\mathbb{r}.\ ] ] in particular , the mse for a random variable @xmath5 is completely determined by knowing its expectation @xmath20 and variance @xmath21 .
note that ( [ eq : mse1 ] ) holds for any distribution function such that @xmath20 and @xmath21 exist . for general loss functions
@xmath9 , the argmin @xmath22 and min @xmath23 do not characterize the risk ; that is , @xmath8-d ( e_{d}y , a ) \ ] ] will be a function of @xmath7 .
the expectation and variance also have the following conditional properties @xmath24 , \\
\label{eq : mse3}\v y & = & \v\e [ y|x ] + \e\bigl[\v(y|x)\bigr].\end{aligned}\ ] ] in the next section , we consider random variables that take values on a space of distributions @xmath25 and show that when the kl divergence is used to compare distributions , equations ( [ eq : eequiv ] ) through ( [ eq : mse3 ] ) hold for kl risk .
let @xmath26 be a sample space equipped with a @xmath27-finite measure @xmath28 .
when @xmath29 is finite or countable , @xmath28 is usually the counting measure .
when @xmath30 and @xmath29 contains an open set of @xmath31 for some @xmath32 , then @xmath28 is usually the lebesgue measure on @xmath31 . requiring @xmath29 to contain an open set implies that the dimension of @xmath29 is @xmath9 .
let @xmath25 be the collection of all probability measures @xmath33 on @xmath26 that are absolutely continuous with respect to @xmath28 , that is , @xmath34 implies @xmath35 for all @xmath36 .
this is denoted as @xmath37 .
note that we allow the support of @xmath33 to be a proper subset of @xmath29 .
let @xmath38 ( in bold font ) be a random quantity whose values are distributions in @xmath25
. the density of the distribution @xmath33 with respect to @xmath28 will be denoted by @xmath39 ( in lower case ) , and the corresponding random variable by @xmath40 ( in bold font lower case ) . following definition 2.1 in @xcite
, @xmath38 is an @xmath25-valued random variable if @xmath41 is a real - valued random variable for all @xmath36 .
we are considering the problem of estimating a distribution so for this paper @xmath42 is any estimator of an unknown distribution @xmath43 where @xmath44 is an i.i.d .
sample from @xmath45 .
a random distribution is a mapping from @xmath46 to @xmath25 .
let @xmath47 be another random quantity that is jointly distributed with @xmath38 .
[ the : ptilde]for every @xmath48 , @xmath49 $ ] is a probability measure that is absolutely continuous with respect to @xmath28 , that is , @xmath50 , is unique up to measure zero ( @xmath28 ) , and has a density @xmath51 \quad\quad \mbox{for } y\in \mathbb{x}.\ ] ] in addition , when @xmath52 is replaced with the random variable @xmath47 , @xmath53 $ ] is an @xmath25-valued random variable .
for all @xmath52 it is easily seen that @xmath54 is a probability measure because @xmath54 is countably additive and @xmath55 , where @xmath56 is the empty set .
the remaining claims of the theorem can be established by noting that equation ( [ eq : ptilde ] ) can be written as @xmath57 where @xmath58 is the conditional distribution of @xmath59 given @xmath52 . since @xmath60=\int_{\mathbb{x}^{n}}\int _ { a}r_{\mathbf{x}}(y)\,\mathrm{d}\lambda(y)r_{0}^{n } ( \mathbf{x}|s)\,\mathrm{d}\lambda^{n}(\mathbf{x}),\ ] ]
the set @xmath61 is arbitrary , and the integrals can be interchanged , we see that @xmath62 is the density for @xmath54 and @xmath50 for each @xmath52 so @xmath63 is an @xmath25-valued random variable . for @xmath25-valued random variable @xmath64 and @xmath65
we can define the average behavior of @xmath64 relative to @xmath33 using the risk function @xmath66,\ ] ] where @xmath9 is a loss function , that is , a nonnegative convex function on @xmath67 . note that the expectation used to define the risk is with respect to some distribution @xmath43 ; @xmath45 will be fixed but arbitrary other than constraints to ensure that the quantities in the expressions below exist and that the support of @xmath45 is @xmath29 . for any function @xmath9 such that @xmath68<\infty$ ] for some @xmath33 , we define @xmath69\ ] ] and @xmath70\ ] ] if the minimum exists , in which case , @xmath71.\ ] ] for kl risk , that is , when @xmath72 , we have @xmath73 = \int\r_{\mathbf{x}}(y)r_{0}^{n}(\mathbf{x})\ , \mathrm{d}\lambda^{n}(\mathbf{x})\defeq e\r , \\
\label{eq : vequiv-1}v_{d}\r&\defeq&\inf_{r_{1}\in\mathcal{r}}e\bigl[d ( \r , r_{1})\bigr ] = ed(\r , e\r)\defeq v\r.\end{aligned}\ ] ] the middle equalities in equations ( [ eq : eequiv-1 ] ) and ( [ eq : vequiv-1 ] ) are established in wu and vos @xcite . since these are equal when @xmath74 is the kl divergence and we consider no other divergence functions on @xmath67 , we will simply write @xmath75 and @xmath76 for the kl mean and variance .
furthermore , @xmath77 and @xmath78 completely characterize the average behavior of the @xmath25-valued random variable @xmath64 relative to any distribution @xmath79 because of the relationship @xmath80=d ( \e\r , r ) + \v\r \quad\quad\forall r\in\mathcal{r}.\ ] ] this means the kl risk for an @xmath25-valued random variable @xmath64 , having any distribution function , is completely determined by knowing its argmin , @xmath81 , and minimum , @xmath82 . when @xmath83 , equation ( [ eq : klr1 ] ) gives the decomposition of the kl risk in terms of bias and variance .
the relationship in ( [ eq : klr1 ] ) will not hold for general nonnegative convex functions @xmath9 .
in this paper we only consider kl divergence @xmath84 .
furthermore , a conditional expectation on @xmath25-valued random variables can be defined so that the following conditional properties hold @xmath85 , \\ \label{eq : klr3}v\r & = & v\e[\r|\s]+\e\bigl[v(\r|\s)\bigr],\end{aligned}\ ] ] where @xmath86 could be @xmath25-valued but could also be real or other valued since values of @xmath86 will only be used to generate sub sigma fields .
let @xmath43 have support @xmath29 and let @xmath64 be an @xmath25-valued random variable such that the kl mean @xmath87 and the kl variance @xmath88 exist and are finite .
then for any @xmath79 the mean divergence between @xmath64 and @xmath33 depends only on the kl mean @xmath77 and kl variance @xmath78 .
furthermore , the kl mean and kl variance satisfy the classical conditional equalities ( [ eq : klr2 ] ) and ( [ eq : klr3 ] ) . equation ( [ eq : klr1 ] ) follows from the definition of kl variance and theorem 5.2 in @xcite who show that the expected kl loss @xmath89 $ ] from an @xmath25-valued random variable @xmath38 to a distribution @xmath79 decomposes as @xmath90=\e\bigl[d(\r,\e\mathbf{r})\bigr]+d(\e\r , r).\ ] ] equation ( [ eq : klr2 ] ) follows from the fact that the kl means @xmath87 and @xmath91 $ ] have densities with respect to @xmath28 and the order of integration can be interchanged .
the steps are the same as those that establish @xmath92 $ ] for @xmath93-valued random variables @xmath94 and @xmath5 .
we rewrite ( [ eq : klr1 ] ) as @xmath95-d(\e\r , r)=\v\r.\ ] ] note that both expectations ( with domain @xmath93-valued random variables and with domain @xmath25-valued random variables ) and the variance depend on the data generation distribution @xmath45 , which can be any point in @xmath25 with support @xmath29 .
if this equation holds for random sample @xmath96 then it also applies to the conditional distribution of @xmath96 given @xmath97 @xmath98-\d\bigl(e[\r|s],r\bigr)=\v(\r|s).\ ] ] substituting @xmath47 into the equation above and taking expectation gives @xmath99-\e\bigl[\d\bigl(\e[\r|\s],r\bigr)\bigr]=\e\bigl[\v(\r|\s ) \bigr].\ ] ] substituting @xmath91 $ ] into @xmath64 in ( [ eq : v1 ] ) and using @xmath100=\e\r$ ] gives @xmath101,r\bigr)\bigr]-d(\e\r , r)=\v\bigl(\e[\r|\s ] \bigr).\ ] ] adding ( [ eq : v3 ] ) to ( [ eq : v4 ] ) and substituting from ( [ eq : v1 ] ) proves ( [ eq : klr3 ] ) .
the random variable @xmath64 is a distribution function defined on the sample space and it will be useful to relate @xmath64 to a statistic @xmath102 .
we define @xmath103 and when we consider only one statistic we write @xmath104 .
the @xmath31-valued random variable @xmath105 describes the behavior of the @xmath25-valued random variable @xmath64 and the mean of @xmath105 can be obtained from the kl mean .
[ thm : expectation - property - on ] for any statistic @xmath102 such that @xmath106 a.e .
, the mean of @xmath102 under @xmath87 equals the mean of @xmath31-valued random variable @xmath105 @xmath107.\label{eq : klexpectationproperty}\ ] ] the density for @xmath87 can be written as @xmath108 so that @xmath109\end{aligned}\ ] ] because the order of integration can be switched .
we typically are interested in a subfamily of distributions @xmath111 and we describe a distribution in terms of the kl risk @xmath112 $ ] for @xmath113 .
we add the regularity condition that the support of each distribution in @xmath110 is @xmath29 .
equation ( [ eq : klr1 ] ) shows that @xmath87 and @xmath88 give the kl risk for any @xmath113 .
however , generally @xmath114 even if @xmath64 takes values only in @xmath110 .
we consider whether an expectation can be defined that takes values in @xmath110 and so that ( [ eq : klr1 ] ) holds
. we will define this expectation as a minimum over @xmath110 .
we define @xmath115\ ] ] and @xmath116\ ] ] if the minimum exists , in which case @xmath117.\ ] ] equation ( [ eq : klr1 ] ) now becomes @xmath118=d\bigl(\ed\r , p\bigr)+\vd\r+\delta\bigl(\e\r,\ed\r , p\bigr ) \quad\quad\forall p\in \mathcal{p},\ ] ] where @xmath119 if @xmath120 vanishes for all @xmath113 then the argmin @xmath121 and the min @xmath122 completely characterize @xmath64 in terms of kl risk .
when @xmath120 is small these functions can be used to approximate the kl risk of @xmath64 .
we will show the term @xmath120 vanishes when @xmath110 is an exponential family .
the relationship between the expectations @xmath77 and @xmath121 can be expressed by using the kl projection onto @xmath110 @xmath123 by equation ( [ eq : klr1 ] ) , @xmath124 for any @xmath125 we have that @xmath126 since @xmath111 .
these results are summarized in the following theorem .
let @xmath43 such that the support of @xmath45 is @xmath29 and let @xmath64 be an @xmath25-valued random variable such that the distribution mean @xmath121 and the distribution variance @xmath122 exist and are finite .
then for any @xmath113 the mean divergence between @xmath64 and @xmath127 is given by ( [ eq : klp0 ] ) .
the term @xmath120 measures the extent to which the kl mean , distribution mean , and @xmath127 depart from forming a dual pythagorean triangle .
the kl variance is less than or equal to the distribution variance , @xmath126 , and the distribution mean is the kl projection of the kl mean onto @xmath110 , @xmath128 .
wu and vos @xcite show that @xmath129 for all @xmath113 an exponential family .
for mixture families @xmath130 .
hence , @xmath120 vanishes when @xmath110 is either an exponential family or mixture family .
while we do nt know how to write @xmath0 as an integral and the expectation property ( [ eq : klexpectationproperty ] ) does not hold for @xmath0 in general , we show equations ( [ eq : klr2 ] ) and ( [ eq : klr3 ] ) hold with @xmath3 replaced with @xmath0 and @xmath4 replaced with @xmath1 when @xmath110 is either an exponential or mixture family . furthermore , the expectation property will hold for @xmath0 when @xmath110 is an exponential family and @xmath102 is the canonical statistic .
for a general subspace @xmath111 the distribution mean @xmath121 and distribution variance @xmath122 do not characterize @xmath112 $ ] for @xmath113 .
however , when @xmath110 is an exponential family these quantities do characterize @xmath112 $ ] and the classical equalities relating conditional mean and variance hold .
a standard reference for exponential families is brown @xcite , but the approach we take here is slightly different since our emphasis is on the distributions without regard to any particular parameterization
. an exponential family @xmath110 will be defined by selecting a point @xmath131 and statistic @xmath132 taking values in @xmath31 .
the defining property of an exponential family is that for any @xmath113 the log of the density of @xmath127 with respect to @xmath133 is a linear combination of @xmath132 and the constant function .
we start with some definitions and basic properties .
@xmath110 is an _ exponential family on _
@xmath29 if there exists @xmath131 such that the support of @xmath133 is @xmath29 and a function @xmath134 such that for any @xmath113 @xmath135 the distribution @xmath133 is called a _ base point _ and @xmath102 is called the _ canonical statistic _ of @xmath110
. the _ canonical parameter space _ is @xmath136 without loss of generality , we can choose a base point @xmath133 such that @xmath137 .
we ll refer to exponential families using base points that belong to the family .
let @xmath110 be an exponential family with base point @xmath133 , canonical statistic @xmath102 , and set @xmath138 .
the _ cumulant function _ has domain @xmath139 and is defined as @xmath140 the density with respect to @xmath133 for any @xmath113 is @xmath141 the family @xmath110 is _ regular _ if @xmath142 is open and @xmath110 is _ full _ if @xmath143 . by the factorization theorem ,
@xmath102 is sufficient
. it will often be useful to restrict the choice of @xmath102 so that it is complete for the full exponential family @xmath110 .
a statistic @xmath102 is _ complete _ for @xmath110 if @xmath144 the following theorem shows that the projection operator on @xmath110 behaves like the expectation operator on @xmath25 ( theorem [ thm : expectation - property - on ] ) and will be used to show that the classical conditional expectation equation holds for @xmath0 .
[ thm : expmu ] if @xmath145 is the kl projection onto @xmath110 , where @xmath110 is an exponential family having canonical statistic @xmath102 and @xmath146 , then for any @xmath79 such that @xmath147 , @xmath148 where @xmath149 is the mean parameter space of @xmath110 .
this result follows from the relationship between the natural and expectation parameters for an exponential family @xmath110 .
let @xmath150 for some @xmath151 .
then the natural parameter @xmath152 of this distribution satisfies @xmath153\ ] ] and since @xmath154 parameterizes @xmath110 , @xmath155.\ ] ] the result now follows for exponential family @xmath110 by simple calculation @xmath156 where @xmath157 by ( [ eq : dual1 - 1 ] ) .
[ cor : pythagorean - property - for ] let @xmath110 be an exponential family and let @xmath79 such that @xmath158 exists . for all @xmath113 @xmath159 this is a well - known result .
see , for example , @xcite or @xcite .
we define an extended projection @xmath160 to be any distribution in @xmath25 such that expectation and pythagorean properties hold and it belongs to the `` boundary '' of @xmath110 ; that is , @xmath161 note that @xmath158 satisfies these three equalities , and that the last two equalities imply @xmath162 the extended projection allows us to define the extended mle in the next section .
the pythagorean property allows us to improve @xmath25-valued random variables by the projection @xmath145 or , more generally , by @xmath163 . if @xmath164 exists a.e .
, then @xmath165\ge e\bigl[d(\overline{\pi}\r , p)\bigr]\ ] ] with equality holding if and only if @xmath166 a.e . replacing @xmath33 with @xmath64 in equation ( [ eq : extended pythagorean ] ) and taking expectations shows @xmath165=e\bigl[d(\r,\overline{\pi}\r)\bigr]+e\bigl[d(\overline{\pi } \r , p)\bigr ] \quad\quad\forall p\in\mathcal{p}\ ] ] and the result follows from the fact that @xmath167\ge0 $ ] with equality holding if and only if @xmath168 a.e . for exponential families , the distribution expectation and variance
have the same properties as the kl expectation and variance .
one distinction is that the expectation property of @xmath3 holds for any statistic while for @xmath0 the expectation property holds only for the canonical statistic @xmath102 .
let @xmath43 have support @xmath29 and let @xmath64 be an @xmath25-valued random variable such that the distribution mean @xmath121 exists and the distribution variance @xmath122 is finite .
then for any @xmath113 , where @xmath110 is an exponential family , the mean kl divergence between @xmath64 and @xmath127 depends only on the distribution mean and distribution variance @xmath169=d\bigl(\ed\r , p\bigr)+\vd\r \quad\quad\forall p\in\mathcal{p}.\ ] ] assuming the conditional expectations and variances exist , the distribution mean and distribution variance satisfy the classical conditional equalities @xmath170 , \\ \label{eq : klp3}\vd\r & = & \vd\ed[\r|\s]+\e\bigl[\vd(\r|\s)\bigr],\end{aligned}\ ] ] where @xmath47 is a real - valued random vector . furthermore , the expectation property holds for the canonical statistic @xmath102 @xmath171.\ ] ] by corollary [ cor : pythagorean - property - for ] and equation ( [ eq : klproj ] ) the correction term ( [ eq : klcorrection ] ) vanishes showing that equation ( [ eq : klp1 ] ) holds . equation ( [ eq : klp4 ] ) follows from @xmath172 and the expectation property on @xmath25 ( [ eq : klexpectationproperty ] ) .
equation ( [ eq : mued ] ) follows from the ( extended ) projection property for exponential families ( [ eq : klproj ] ) and ( [ eq : mupi ] ) and the relationship between @xmath173 and @xmath0 ( [ eq : klprojection ] ) .
now equation ( [ eq : klp2 ] ) follows from @xmath174\bigr ) & = & \e\bigl[\mu\bigl(\ed[\r|\s]\bigr)\bigr ] \\ & = & \e\bigl[\mu\bigl(\e[\r|\s]\bigr)\bigr ] \\ & = & \mu\bigl(\e\e[\r|\s]\bigr ) \\ & = & \mu(\e\r ) \\ & = & \mu\bigl(\ed\r\bigr),\end{aligned}\ ] ] where the first equality follows from ( [ eq : klp4 ] ) , the second and fifth equalities follow from ( [ eq : mued ] ) , the third equality follows from the expectation property of the kl mean on @xmath25 , and the fourth equality follows from the conditional expectation property that holds on @xmath25 ( [ eq : klr2 ] ) .
equation ( [ eq : klp3 ] ) follows again the same steps that justified ( [ eq : klr3 ] ) .
we rewrite ( [ eq : klp1 ] ) as @xmath175-d\bigl(\ed\r , r\bigr)=\vd\r.\ ] ] if this equation holds for random sample @xmath96 then it also applies to the conditional distribution of @xmath96 given @xmath97 @xmath176-d\bigl(\ed[\r|s],r\bigr)=\vd(\r|s).\ ] ] substituting @xmath86 into the equation above and taking expectation gives @xmath177-\e\bigl[d\bigl(\ed[\r|\s],r\bigr)\bigr]=\e\bigl[\vd(\r|\s ) \bigr].\ ] ] substituting @xmath178 $ ] into @xmath64 in ( [ eq : v1 - 1 ] ) and using @xmath179=\ed\r$ ] gives @xmath180,r\bigr)\bigr]-d\bigl(\ed\r , r\bigr)=\vd\ed[\r|\s].\ ] ] adding ( [ eq : v3 - 1 ] ) to ( [ eq : v4 - 1 ] ) and substituting from ( [ eq : v1 - 1 ] ) proves ( [ eq : klp3 ] ) .
an immediate corollary to the characterization theorem on @xmath110 ( equations ( [ eq : klp1 ] ) , ( [ eq : klp2 ] ) , and ( [ eq : klp3 ] ) ) is that for any random distribution @xmath64 and any statistic @xmath86 , the random distribution @xmath178 $ ] will have the same distribution mean and have distribution variance less than or equal to that of @xmath64 .
if @xmath182 is sufficient then @xmath183 $ ] is an estimator and if @xmath102 is also complete @xmath183 $ ] will have smaller variance than @xmath64 unless they are equal with probability one .
this conditional expectation is enough to establish a rao
blackwell result for distribution estimators if these were restricted to @xmath110 .
however , since we are allowing @xmath25-valued estimators we also need to project the distributions onto @xmath110 using @xmath163 . for an exponential family @xmath184 having mean parameter @xmath185 and discrete sample space we typically have that @xmath186 while @xmath187 where @xmath188 is the closure of @xmath189 . in this case , the mle does not always exist
. however , the characterization theorem applies to @xmath25-valued estimators so we can define an estimator that equals the mle @xmath190 when it exists and as a distribution @xmath191 such that @xmath192 and @xmath193 if @xmath194 .
the _ extended mle _ as distribution estimator is @xmath195 unbiasedness of @xmath196 follows from the following theorem .
[ thm : unbiasedestimatorsinexp ] let @xmath110 be an exponential family with complete sufficient statistic @xmath102 and let @xmath64 be a @xmath25-valued random variable .
the estimator @xmath64 is distribution unbiased for @xmath197 if and only if @xmath198)=t$ ] a.e .
we must show @xmath199 for all @xmath137 if and only if @xmath200)=t$ ] a.e .
for all @xmath137 .
consider the following equivalencies each of which holds for all @xmath137 : @xmath201\bigr ) & = & \mu(p_{0 } ) \\
\iff\quad e\bigl[\mu\bigl(e[\r|t]\bigr)\bigr ] & = & \mu(p_{0 } ) \\
\iff\quad e\bigl[\mu\bigl(e[\r|t]\bigr)\bigr ] & = & e(t).\end{aligned}\ ] ] the first equivalence follows because the expectation of @xmath102 parameterizes @xmath110 , the second equivalence follows from the projection property for exponential families , the third equivalence follows from the conditional expectation defined for the kl mean , the fourth equivalence follows from the expectation property for the kl mean , and the fifth equivalence follows from the definition of the function @xmath202 . clearly , @xmath200)=t$ ] a.e . implies the last equality .
since @xmath102 is complete and the last equality holds for all @xmath137 , this implies @xmath203\bigr)=t\quad\quad \mbox{a.e.}\ ] ] [ thm : optimality - of - the ] let @xmath96 be i.i.d . from a distribution @xmath43
such that the support of @xmath45 is @xmath29 .
let @xmath110 be an exponential family with complete sufficient statistic @xmath102 such that @xmath204 . if @xmath205 is the mle or an extended mle that exists a.e . , then @xmath205 is distribution unbiased for the @xmath206 and it is the unique uniformly minimum distribution variance estimator among all @xmath25-valued estimators that are distribution unbiased for @xmath206 and for which the extended projection @xmath164 exists a.e .
uniqueness and uniform minimum distribution variance follow from the projection property for @xmath25-valued random variables , the characterization theorem on @xmath110 described above , and the unbiasedness from theorem [ thm : unbiasedestimatorsinexp ] .
we consider the number of events or `` successes '' in @xmath207 trials .
the sample space is @xmath208 under the assumptions that these trials are independent and each trial has the same success probability @xmath209 , the distribution of @xmath94 belongs to the @xmath207-binomial family @xmath210 the mle for the parameter @xmath154 is @xmath211 for @xmath212 but is undefined otherwise .
the extended mle ( it will correspond in a natural way to the extended mle distribution estimator ) is @xmath211 for all @xmath213 and it is unbiased for @xmath154 . however , it is not unbiased for other parameterizations such as the odds @xmath214 , or the log odds @xmath215 . when viewed as a distribution , that is , @xmath216 , equivalently , @xmath217 or @xmath218 ( where we allow the odds @xmath219 and log odds @xmath220 to take values in the extended reals ) , the mle is the unique uniformly minimum distribution variance unbiased estimator .
as is common practice , we have used the same notation @xmath221 for both the mle and the extended mle .
estimators , whether real - valued or distribution - valued , are functions with domain @xmath29 . for the @xmath207-binomial family an estimator is given by a sequence of @xmath222 values , real numbers for @xmath221 and probability distributions for @xmath223 . for @xmath221 , we have the sequence @xmath224 let @xmath225 be a distribution in @xmath110 . if probabilities of @xmath225 are used to assign weights to the values in ( [ eq : mlereal ] ) , then the real number that is closest to the weighted values of ( [ eq : mlereal ] ) is @xmath226 .
that is , @xmath227 by the rao blackwell theorem ,
for any other sequence of @xmath222 real numbers @xmath228 that satisfy @xmath229 the realized minimum will be greater than the minimum obtained using the values in ( [ eq : mlereal ] ) unless the sequences are equal , @xmath230 for @xmath231 . a distribution estimator @xmath223 obtained from the real valued estimator given in ( [ eq : mlereal ] )
can be defined as @xmath232 where @xmath233 is the indicator function for its subscript ; that is , the degenerate distribution putting all mass on 0 or 1 . since @xmath234 it is easily checked that @xmath235 which means that the sequence in ( [ mledist ] ) is the extended mle @xmath196 .
hence , @xmath236 .
again , we let @xmath225 be any distribution in @xmath110 .
if @xmath225 is used to assign weights to the distributions in ( [ mledist ] ) , then the distribution in @xmath110 that is closest to the weighted average of the distributions in ( [ mledist ] ) is @xmath225 .
that is , @xmath237.\ ] ] by the distribution version of the rao blackwell theorem ( theorem [ thm : optimality - of - the ] ) for any estimator @xmath238 , expressed as a distribution estimator , @xmath239 that satisfies @xmath240,\ ] ] the realized minimum will be greater than that of the mle ( [ mledist ] ) unless the two sequences of functions ( [ mledist ] ) and ( [ eq : otherdist ] ) are equal .
theorem [ thm : optimality - of - the ] provides a stronger result than this since the distributions need not belong to @xmath110 . in the class of all distribution unbiased estimators of the form @xmath241
for which the extended projections @xmath163 exists , the mle ( [ mledist ] ) has smallest distribution variance . in the hardy
weinberg model estimators that do not belong to the family @xmath110 have been suggested .
we consider the details in section [ sub : hardy - weinberg - model ] .
the simplex represents the trinomial model space on @xmath242 for @xmath243 , while the solid curve is the hw model space on @xmath244 , @xmath245 , and @xmath246 for @xmath209 .
the open circles represent the ( extended ) mle under the trinomial model @xmath247 , and the solid dots are the ( extended ) mle under the hw model @xmath248 .
the dashed curve shows the kl mean of the hw mle for each value of @xmath154 . ]
the choice of the @xmath207-binomial model @xmath110 was based on the assumptions that the data represented independent and identical trials .
if either of these assumptions were grossly violated , the binomial model would not be appropriate .
however , this model can be used when these assumptions hold approximately in the sense that there is a distribution @xmath197 in @xmath110 that is close to the data generation distribution @xmath45 , that is , @xmath249 is small . in this case
, the mle is the unique umv@xmath2u estimator for @xmath133 . for a single pair of alleles a and a , which occur with probabilities @xmath154 and @xmath250 for @xmath251 , the hardy weinberg ( hw ) model defines the relative frequency of genotypes aa , aa , and aa to be @xmath244 , @xmath245 , and @xmath246 .
for this example , we can take @xmath25 to be the collection of trinomial models with probabilities @xmath242 for @xmath243 which can be represented by the simplex in 2-dimensional space .
see figure [ fig : hardy - weinberg - model ] for the simplex .
the open circles in figure [ fig : hardy - weinberg - model ] are the extended mle @xmath247 for the trinomial with @xmath252 trials , where @xmath253 and @xmath254 are the counts for aa and aa .
the solid curve in the simplex is the hw model @xmath255 which is a one dimensional exponential family with canonical sufficient statistic @xmath256 and canonical natural parameter @xmath257 . chow and fong @xcite
find the umvu for @xmath258 and @xmath259 using @xmath260+e_{\theta}\bigl[\bigl(\hat{\pi}_{3}-(1- \theta)^{2}\bigr)^{2}\bigr]\ ] ] as squared - error loss .
they show the umvu is inadmissible by exhibiting a dominating estimator . both the umvu and the dominating estimator take values outside the hw model . in terms of distribution estimators , these are @xmath25-valued estimators .
the extended mle for the hw model is @xmath248 while the extended distribution mle is @xmath223 where @xmath133 is the degenerate distribution putting all its mass on @xmath261 ( the lower left vertex ) and @xmath262 is the degenerate distribution putting all its mass on @xmath263 ( the lower right vertex ) .
the extended hw mle is represented by the solid dots in figure [ fig : hardy - weinberg - model ] . among the difficulties with the umvu estimator and the dominating estimator
is that there are other ways to define squared - error loss ( using one bin or two other bins ) .
these are avoided by using kl divergence .
since @xmath110 is an exponential family the extended mle is the umv@xmath2u for all @xmath110-valued estimators but also for all @xmath25-valued estimators since the projection exists for all points in the simplex other than the two lower vertices which satisfy the extended projection . as a comparison ,
the kl mean , represented by the dashed curve in figure [ fig : hardy - weinberg - model ] , lives outside the model so the extended mle is nt kl unbiased .
this is due to the curvature in the exponential family .
the poisson family of distributions is @xmath264 where @xmath265 .
let @xmath96 be a simple random sample from a poisson distribution @xmath266 .
the sum @xmath267 is a complete sufficient statistic of the family .
although the poisson family is typically parametrized by a single parameter , we consider estimates for the probability @xmath268 for some @xmath269 . a crude but unbiased estimator
is @xmath270 given the sum @xmath271 , @xmath272 is distributed as a binomial(@xmath271 , @xmath273 ) random variable , the rao
blackwell theorem shows that @xmath274=\cases{\displaystyle { s_{n } \choose i } \biggl(\frac{1}{n } \biggr)^{i } \biggl(1-\frac{1}{n } \biggr)^{s_{n}-i } & \quad if $ i\le s_{n } $ , \vspace*{2pt } \cr 0 & \quad otherwise , } \ ] ] is an unbiased estimator of @xmath275 .
since @xmath276 depends on the complete sufficient statistic @xmath271 only , it must be the unique mvue of @xmath275 .
using the criterion of distribution unbiasedness , these anomalous estimators do not arise . since @xmath271 is the canonical statistic ,
the mle @xmath277 is the unique umvu estimator for @xmath28 and the extended distribution mle @xmath278 is the umv@xmath2u estimator for @xmath279 where @xmath278 is @xmath280 when @xmath281 . to show how the umvu estimator can fail completely
, lehmann @xcite considers the parameter @xmath282 for @xmath283 . in this case , the unique umvu estimator is @xmath284 .
since the sample consists of nonnegative integers this estimator is represented by the following sequence of real numbers @xmath285 parametric unbiasedness means that if the poisson distribution that assigns probability @xmath286 to @xmath287 is used to assign probability to the terms in the sequence then @xmath288 .
that is , the parameter is the real number that is closest to this sequence in terms of mean square error . in addition , the weighted average of the above sequence is @xmath289 . by focusing on distributions rather than the parameters that name the distributions
these problems are avoided .
the mle , as a distribution estimator , is represented by the following sequence of probability distributions @xmath290 distribution unbiasedness means that if the poisson distribution @xmath279 is used to assign probability to the terms in the sequence then @xmath291.\ ] ] that is , the distribution that generates the data is the distribution in the exponential family that is closest to this sequence in terms of kl risk .
any other sequence of distributions with this property will have greater distribution variance .
the distribution version of the rao blackwell theorem [ thm : optimality - of - the ] has been developed by analogy with important properties of mean square error for the parametric version . in particular , we have used a pythagorean - type property for two asymmetric distribution - like functions : the kl divergence @xmath292 and its expectation @xmath293 $ ] . for exponential family @xmath125
we have @xmath294 while for all @xmath25 @xmath295=\e\bigl[d(\r,\e\r)\bigr]+\e\bigl[d(\e\r , r)\bigr]\ ] ] so that the expectation operator @xmath173 defined on @xmath25-valued random variables for the kl risk plays the role of the projection operator @xmath145 for the kl divergence .
each operator is a map from a more complicated space to a simpler space , @xmath173 from @xmath25-valued random variables to a distribution in @xmath25 and @xmath145 from distributions in @xmath25 to a distribution in @xmath110 , that preserve the kl risk and kl divergence , respectively .
the restriction to exponential families is essentially required by the criterion of having a sufficient statistic of fixed dimension for all sample sizes @xmath207 .
specifically , the darmois koopman pitman theorem which follows from independent works of darmois @xcite , koopman @xcite and
pitman @xcite shows that when only continuous distributions are considered , the family of distributions of the sample has a sufficient statistic of dimension less than @xmath207 if and only if the population distribution belong to the exponential family .
denny @xcite shows that for a family of discrete distributions , if there is a sufficient statistic for the sample , then either the family is an exponential family or the sufficient statistic is equivalent to the order statistics .
the mle is parameter - invariant which means that the same distribution is named by the parametric ml estimate regardless of the parameter chosen to index the family .
one approach to studying parameter - invariant quantities is to use differential geometry ( e.g. , amari @xcite or kass and vos @xcite ) .
the parameter - invariant approach does not work well for parameter - dependent quantities such as bias and variance of parametric estimators .
our approach allows for the definition of parameter - free versions of bias and variance .
furthermore , the distribution version of the rao
blackwell provides two extensions : ( 1 ) minimum variance is taken over a larger class of estimators that includes estimators that are not required to take values in the model space @xmath110 , ( 2 ) the true distribution need not belong to @xmath110 .
the fact that the mle is the unique uniformly minimum distribution variance unbiased estimator for exponential families distinguishes the mle from other estimators .
this is in contrast to asymptotic methods applied to mse that can be used to show superior properties of the mle but , being asymptotic results , do not apply uniquely to the mle .
asymptotically , mse and kl risk are the same and the mse can be viewed as an approximation to kl risk for large @xmath207 .
the distribution version of the rao
blackwell theorem [ thm : optimality - of - the ] provides support for fisher s claim of the superiority of the mle even in small samples .
we thank the associate editor and external reviewers for their insightful comments and suggestions which have made great improvement on this paper . | we employ a parameter - free distribution estimation framework where estimators are random distributions and utilize the kullback
leibler ( kl ) divergence as a loss function .
wu and vos [ _ j .
statist .
plann . inference _ * 142 * ( 2012 ) 15251536 ] show that when an estimator obtained from an i.i.d .
sample is viewed as a random distribution , the kl risk of the estimator decomposes in a fashion parallel to the mean squared error decomposition when the estimator is a real - valued random variable . in this paper , we explore how conditional versions of distribution expectation ( @xmath0 ) can be defined so that a distribution version of the rao
blackwell theorem holds .
we define distributional expectation and variance ( @xmath1 ) that also provide a decomposition of kl risk in exponential and mixture families . for exponential families , we show that the maximum likelihood estimator ( viewed as a random distribution ) is distribution unbiased and is the unique uniformly minimum distribution variance unbiased ( umv@xmath2u ) estimator .
furthermore , we show that the mle is robust against model specification in that if the true distribution does not belong to the exponential family , the mle is umv@xmath2u for the kl projection of the true distribution onto the exponential families provided these two distribution have the same expectation for the canonical statistic . to allow for estimators taking values outside of the exponential family , we include results for kl projection and define an extended projection to accommodate the non - existence of the mle for families having discrete sample space .
illustrative examples are provided . ./style / arxiv - general.cfg |
collisionless simulations of dark matter halos in a @xmath4cdm cosmology consistently predict a density profile that diverges towards the centre ( e.g. * ? ? ? * ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) over a wide mass range .
this profile typically rolls over to a steep slope at large radii where it is broadly consistent with the rotation curves of many late - type galaxies ( e.g. * ? ? ?
* ; * ? ? ?
* ) . however , the inner cusp ( e.g. @xmath5 , * ? ? ?
* ) is _ inconsistent _ with observations of rotation profiles in dwarf galaxies both within the local group and beyond ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) . instead
, these dwarf galaxies are better fit using a constant density within the central kpc or so .
two classes of solutions have been proposed to resolve this discrepancy .
the first suggests that coupling to baryons can reshape the dark matter cusp into a core ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
these solutions have met with some criticism ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
in particular , @xcite claimed it was not energetically feasible .
however , the sub - structure present within massive halos can induce collisionless scattering of dark matter particles within the centre , flattening the inner cusp without the need of baryons ( e.g. * ? ? ?
alternative models to @xmath4cdm have been also been suggested , such as ` warm ' dark matter ( e.g. * ? ? ? * ) or ( non - gravitational ) dark matter self - interactions ( e.g. * ? ? ?
+ @xcite argued that cyclic star formation bursts within the centres of dark matter halos could provide the necessary energy to re - shape the dark matter cusp into a dark matter core . in this model , feedback from type ii supernovae ( sne ) periodically drives bulk gas motions in the star forming regions in the central kpc of dwarf galaxies .
since gas can dominate the central gravitational field , these bulk motions can significantly perturb the central gravitational potential on timescales comparable to dark matter orbital times leading to an efficient transfer of kinetic energy to the central dark matter distribution and the formation of a significant dark matter core .
a notable feature of the @xcite framework was that the gas need not be expelled or even pushed very far outside the core - forming region ( as opposed to the impulsive blow - out scheme e.g. * ? ? ? * ; * ? ? ? * ) .
@xcite gave an analytical description of how changing the dominate potential leads to irreversible heating of the dark matter orbits .
+ the mechanism was first verified in a cosmological simulation by @xcite . a required feature for simulations to flatten cusps
is realistically clustered star formation so as to ensure a star formation history that significantly fluctuates in both space and time .
several simulations since have reproduced these results over a range of simulated galaxy masses ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* see @xcite @xcite ) .
the process also has consequences for other collisionless components such as stars and star clusters .
initially centrally concentrated populations ( such as younger stars ) are pushed outward over time , leading to older populations having larger orbits on average .
this results in a broad match to observations of the stellar content of dwarf galaxies ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
. it also has interesting consequences for multiple populations in galactic globular clusters @xcite .
+ @xcite made the first analytical estimate of the energy difference between cored and cuspy dark matter halos of the same mass , finding that sne alone would have difficulty providing the energy required to form dark matter cores given the average stellar masses of dwarf galaxies .
though there are other sources of energy available from star formation , such as stellar winds and the ionizing flux from young stars ( e.g. * ? ? ?
* ) , it is not clear that this added energy would be sufficient to remove the discrepancy or couple as well as sne - driven bulk gas motions .
however , the @xcite results are difficult to reconcile with the core forming simulations which typically employ feedback only from sne ( and with realistic efficiencies ) .
+ in this work we re - examine the energy requirements and show that the formation of dark matter cores is energetically plausible in a @xmath4cdm cosmology .
the layout of the paper is as follows .
[ sec : amb ] deals with the ambiguity inherent in the definition of a core size , and argues for a new , general way to define the core radius . in [ sec : cfeb ] we discuss the energy required to form dark matter cores in halos that were initially cuspy using assumptions that are more consistent with our understanding of how the process operates . in [ sec : results ] we show predictions for core sizes in dark matter halos and examine the implications for observed systems such as fornax .
many observed cores in dwarf galaxies are inferred from the rotation profiles of the stars and gas by finding the best - fit dark - matter density model ( e.g. * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ) .
commonly , one would use either a cored isothermal sphere , whose density profile goes as @xmath6 at large radii , or that proposed by @xcite where the density goes as @xmath7 at large radii , @xmath8 both satisfy the requirement that the density profile transitions to a flat core within a scale radius . without probing the rotation profile to large radii it is difficult to determine which density profile is the best fit .
the cored isothermal profile predicts a constant velocity at large radii while the @xcite profile looks like the standard galaxy rotation curve ( e.g. * ? ? ?
however , the common practice of defining the core size as the scale radius leads to large variations in core radii dependent on the chosen functional form . + to avoid this ambiguity , we define the core radius as _ the radius , @xmath9 , at which the logarithmic slope of the density profile satisfies _ : @xmath10 we will use this definition from now on when referring to the core size . characterizing cores this way
is consistent with the density profile slopes found in simulations of dwarf galaxies ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) and with the mass profile slopes observed in local group dwarfs ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
furthermore , the simulated galaxies always transition to a density profile with a strongly negative slope ( @xmath11 ) beyond a few kpc where the effects of baryonic feedback are limited .
this definition of the core radius lessens the reliance on an assumed dark matter density distribution , and it can be determined directly from the velocity profile .
it is also straightforward to relate @xmath12 ( for any given density profile ) to @xmath9 .
+ to demonstrate the robustness of our new core definition , and to illustrate the ambiguity with using the scale radius , we show in figure [ fig : kdn ] the rotational velocity profile for the low surface brightness ( lsb ) galaxy ngc 959 from @xcite .
these authors assumed a cored isothermal density profile , and the best fit of this profile to the data is shown as the solid line .
although it is perfectly sufficient to match the profile in the inner 12kpc , it does not roll over sufficiently to match the slope in the outer part of a cosmological halo .
we have also fit three other density profiles .
the first , shown as the dotted line in the figure , is ( e.g. * ? ? ?
* fixed in the figure ) : @xmath13 the second , shown as the dashed line , is a pseudo - isothermal profile , where we let the power index vary : @xmath14
lastly , we show the ( * ? ? ?
[ eq : rhoburk ] ) profile as the dash - dot line .
+ the differences between the three rotation profile fits are only evident at radii past the furthest observable point , where the predicted rotation velocity rolls over .
we have also done this for the eight other lsbs from @xcite ( see the appendix ) , and the resulting scale radii and core radii are shown in table [ tab : kdnm ] .
it is evident that velocity rotation data alone are not sufficient to distinguish the best fit dark matter density profile .
each fitting function requires a different scale radius , the common definition of the core size .
however , using @xmath15 to define the core radius leads to consistent core sizes , independent of which of the four density profiles was used , so long as the core location is comfortably within the range of data points .
+ there are some lsbs in the @xcite sample that do not show consistent @xmath9 values , particularly ugc 4325 and ddo 64 .
these are galaxies where the core size is beyond the edge of the data , so a fitting algorithm is only able to place lower bounds on @xmath9 and is certainly not able to constrain the parameters of a non - linear density function .
cccc|ccc|ccc|cc & 2&0.4&0.23&2.10@xmath16&0.42@xmath17&0.24&1.60@xmath18&0.68@xmath19&0.25&0.73@xmath20 & 0.27 + ngc 7137&2&0.6&0.35&2.01@xmath21&0.61@xmath22&0.35&1.41@xmath23&1.14@xmath24&0.36&1.09@xmath25&0.40 + ugc 11820&2&1.1&0.64&3.84@xmath26&1.15@xmath27&0.70&2.76&0.89@xmath28&0.50&2.11@xmath29&0.77 + ugc 128&2&2.3&1.33&1.97@xmath30&2.48@xmath31&1.43&1.32@xmath32&5.01@xmath33&1.48&4.50@xmath34&1.64 + ugc 191&2&1.7&0.98&1.64@xmath35&1.84@xmath36&1.11&1.48@xmath37&3.37@xmath38&1.14&3.53@xmath39&1.28 + ugc 1551&2&1.3&0.75&1.89@xmath40&1.32@xmath41&0.77&1.49@xmath42&2.39@xmath43&0.81&2.40@xmath44&0.87 + ugc 4325&2&4.1&2.37&1.74@xmath45&2.94&1.74&2.05@xmath46&2.52&1.08&9.52&3.46 + ddo 64&2&3.3&1.91&9.51@xmath47&7.44@xmath48&5.49&8.83@xmath49&7.44@xmath50&6.20&8.76&3.18 + f583 - 1&2&2.5&1.44&2.16@xmath51&2.55@xmath52&1.46&1.87@xmath53&3.57@xmath54&1.51&4.49@xmath55&1.63 [ tab : kdnm ]
in this section we estimate the energy required to build a core in an initially cuspy density profile .
we assume that each cuspy halo begins with an ` nfw ' profile @xcite .
if we assume the halos are virialized , we can estimate the energy as the difference between the work required to form the cusp and the work required to form the core , as was done in @xcite : @xmath56 where @xmath57 and @xmath58 are the cusp and core potential energies , respectively .
@xmath59 @xmath60 implies the cusped dark matter halo had to gain energy to redistribute mass to form a core . if the simplifying assumption is made that both halos are spherically symmetric , we can write the potential energy integral as @xcite : @xmath61 for a density profile that goes as @xmath62 at large @xmath63 , the integral converges for @xmath64 ; all of the cored profiles considered ( if @xmath65 in eqn . [ eq : rhocore ] ) and the nfw profile lead to well defined values for @xmath66 .
+ taking the upper limit in equation to infinity leads to a very large estimate for the energy ( quite apart from the ambiguity of embedding the halo in the cosmic background ) and is not appropriate given that we are considering _ redistributing _ mass within the central part of an existing halo .
@xcite chose the halo virial radius @xmath67 as the upper limit for the potential energy integral , and normalised the mass of their cored density profile , @xmath68 to that of their initial nfw halo at @xmath67 .
however , simply normalising both halos to have the same mass at @xmath67 does not remove the requirement that equation be carried to infinity . in order for the potential energy integral to be valid with a finite upper limit @xmath69 , the density _ and _ mass profiles for both halos must match at that radius . the choice of @xcite to match the mass at @xmath67 and to truncate the integral for @xmath66 at that radius builds in an implicit discontinuity in the density at that radius .
furthermore , moving the cusp mass to near the virial radius of the halo leads to a core formation energy budget from equations and which will be significantly overestimated compared with the more restricted redistribution considered here ( @xmath69@xmath70 @xmath67 ) .
+ the motivation for constraining the mass and density profiles of the pre- and post - core halos outside @xmath69 to be the same is simply that we expect that strong coupling of star formation feedback to dark matter is only effective in , and only modifies , the central parts of the halo .
the outer parts of the halo will remain largely untouched and reflect the collisionless formation process .
this picture is supported by simulations ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
previous simulations @xcite suggest that for a @xmath7110@xmath72 halo @xmath69 is about 23 times the core radius . +
our method proceeds as follows : * since we do not have any robust predictions of what the core profile should be , we adopt equation as the density profile when computing @xmath58 .
( we have verified that using a different density profile , such as equation or equation , does not change our results . )
* we use the relations given by @xcite and @xcite to set the nfw profile parameters at redshift zero as a function of virial halo mass , @xmath0 ( see * ? ? ?
we then have @xmath73 and @xmath74 as functions of @xmath75 . *
we find the set of parameters @xmath76 , @xmath77 , and @xmath9 for the cored density profile that _ simultaneously _
satisfy @xmath73 and @xmath74 from the previous step .
* we set @xmath78 , where @xmath79 is given by our new definition .
figure [ fig : ill ] illustrates the profile and mass matching .
* we compute the potential energy difference in eqn .
the preceding steps give @xmath59 in terms of @xmath9 and @xmath0 .
in figure [ fig : halocomp ] we show our calculation for the amount of energy required to turn a dark matter cusp into a core as a function of halo mass .
since we may write approximately : @xmath80 for the cored profile , we see , as expected , that the parameter that most influences the energy required to convert a cusp into a core is @xmath81 . a larger core moves more mass to larger radii and thus requires more energy to redistribute it from the cusp . on the other hand , if we allow @xmath69 to scale as a fixed fraction of @xmath67 ( we chose @xmath69/@xmath67=1/6 ) , we find that @xmath59 scales as @xmath0@xmath82 .
this scaling is shown by the solid black line in figure [ fig : halocomp ] and lies within the range of solutions calculated by @xcite ( the gray shaded region in the figure ) . a more realistic estimate for @xmath69 , as a small multiple of @xmath9 ( here 3 ) , leads to a greatly reduced energy requirement and a much shallower dependence on halo mass .
this is shown by the grey lines in figure [ fig : halocomp ] for several core sizes .
our calculations suggest the energy scaling with halo mass is much weaker when @xmath9 ( and hence @xmath69 ) is fixed . in the @xmath83@xmath3 halo mass range ,
@xmath59 scales as @xmath0@xmath84 for a 1kpc core and flattens to @xmath0@xmath85 for a 100pc core .
this flattening is related to how the mass interior to @xmath69 depends on the halo concentration . in the functional form we adopted
the concentration is inversely proportional to @xmath0 @xcite .
varying the ratio @xmath69/@xmath9 creates a modest vertical shift in the @xmath59 curves .
+ in figure [ fig : halocomp ] we also plot the stellar mass to halo mass ( @xmath86-@xmath0 ) relations from @xcite ( dashed ) , @xcite ( dotted ) , and @xcite ( dash - dot ) at @xmath87 .
we have converted @xmath86 to total sne energy using a @xcite imf , assuming that every star above 8 contributes @xmath88ergs to the energy budget .
our results show that _ there is no conflict between the amount of energy available to form a dark matter core and the @xmath4cdm framework _ , regardless of halo mass .
the main question is how large the dark matter core is likely to be .
+ in reality , not all of the sne energy will couple to the cusp dark matter through the induced perturbations to the gravitational potential .
this inefficiency can be represented by scaling the @xmath86-@xmath0 relations shown in figure [ fig : halocomp ] by some coupling efficiency factor , @xmath89 , ( for example , the 40% value used by * ? ? ?
. only for @xmath90 would there be no possibility of creating a core larger than 250pc in the 10@xmath9110@xmath72 halo range .
the implication is that most , if not all , dwarf galaxies should host dark matter cores if their central star formation rate was high enough to generate 10@xmath9210@xmath93 in stars .
+ in the previous section we showed that stellar feedback , of which we have only considered sne , can provide enough energy to the dense star forming gas in the centres of dwarf galaxies to transform the dark matter cusps predicted by numerical experiments to the flattened dark matter cores inferred from observations ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
this is contrary to previous work ( e.g. * ? ? ?
* ) and results from our adoption of a more consistent mass normalisation criterion .
this was motivated in part by the results of hydrodynamic simulations and by the observation that the most efficient injection of star formation feedback energy will occur at the deepest point in the galactic potential , which implies a physical limit to the radius to which cusp dark matter can be redistributed . in our calculations
, we have parameterized this limit as the radius @xmath69 .
our results suggest significant dark matter cores can be formed with less than 1% of the total sne energy available over the star forming history of a typical dwarf galaxy .
+ so far we have been concerned with whether a dark matter core can form in an initially cuspy halo due to star formation feedback .
our new calculations allow us to ask a more interesting question : what is the typical core size for a given halo mass ?
in figure [ fig : halocore ] we show the predicted core size as a function of halo mass using the three @xmath86-@xmath0 relations shown in figure [ fig : halocomp ] and 100% of the associated sne energy .
although figure [ fig : halocore ] reinforces our main conclusion that dark matter cores can form from cusps in even the smallest dwarf galaxy halos , it also emphasizes the uncertainty in trying to make detailed predictions .
the variation in the low mass slope of the @xmath86-@xmath0 relation between the @xcite , @xcite , and @xcite results corresponds directly to a factor of 24 variation in predicted @xmath9 as a function of halo mass .
furthermore , these relations only give the average stellar mass as a function of halo mass .
+ figure [ fig : halocore ] is further complicated because we have not yet considered the efficiency with which the energy from stellar feedback is transferred to dark matter particle orbits .
the coupling efficiency will likely vary in time with changing conditions in the centre of the halo , and will depend on the star formation rate , gas accretion rate , gas heating rate , and the dark matter particle orbits ; the value of @xmath89 used in @xcite is merely a time - averaged value over the dwarf galaxy star formation and gas accretion histories . finally , our results show that the energy required to form the core , @xmath59 , has only a weak dependence on @xmath0 for a given @xmath9 .
+ if , indeed , our ansatz that matter redistribution happens only within @xmath69 is correct , then the total halo mass @xmath0 has only a modest influence on the final core sizes in dwarf galaxies .
many observations of dwarf galaxies in the local universe can only measure dynamical masses out to a few kpc ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) and so can not accurately determine the halo virial mass .
this leads us to figure [ fig : stellarcore ] where we have plotted the range of core sizes as a function of stellar mass required to produce a core of that size , _ without _ adopting a specific coupling efficiency in dark matter core formation .
each curve represents the track in @xmath9-@xmath86 space for each halo mass , and is determined by the relations between @xmath0 , @xmath67 , and @xmath12 ( or concentration parameter ) .
the endpoints of each curve represent the upper and lower limits that we imposed when computing the energy requirements from equation .
+ in this figure , the efficiency @xmath89 is equivalent to sliding a given dwarf galaxy along the @xmath86 axis . to illustrate this
, we show the local group dwarf fornax , whose stellar mass was taken from @xcite and whose dark matter core size was measured by @xcite .
fornax lies significantly below the predicted core size for its stellar mass at 100% efficiency , but is consistent with our core predictions if only a few per cent of the total sne budget contributed to the formation of the dark matter core in the host halo .
fornax s host halo could range anywhere from 10@xmath9110@xmath94 while @xmath89 could range anywhere from 0.11% .
in this paper we presented a new estimate for the energy requirement for galaxies to form dark matter cores in their host halos .
we assume former cusp material remains within a characteristic radius , @xmath69 , at which both the density and mass profiles of the cored dark matter halo and the original cusped halo match .
this significantly reduces the amount of energy required compared to the estimate of @xcite without having to require , for example , that cores form only at high redshift ( e.g. * ? ? ?
our results suggest that dark matter cores are present in most , if not all , dark matter halos that experienced clustered star formation within their centres @xcite . in the case of the local group dwarf galaxy fornax ,
if it lives in a @xmath3 halo then less than 1% of the energy released by all supernovae over its star formation history would be required to build a 1 kpc core .
+ our new calculations alleviate another potential problem raised by @xcite : the accretion of additional high phase - space density collisionless particles ( cuspy satellites ) was shown by @xcite to tend to preserve the steepest density cusp in the merger .
however , if all large halos merging in the collisionless mass assembly process are already cored , the dark matter density profile of the merger remnant should be cored as well .
while the smallest accretors , common at early times , may be cuspy and rebuild a central cusp , subsequent star formation feedback should be enough to destroy the dark matter cusp in a short timemyr . ] .
+ with the apparent tension between @xmath4cdm and the presence of cores in dwarf galaxies relieved , we can now focus on making firm predictions of the sizes of dark matter cores as a function of stellar mass ( figure [ fig : stellarcore ] ) . for example , simulations of dark matter core formation in dwarf galaxies can test the efficiency and energy output of star formation prescriptions . the resulting density profiles and other properties
can be directly compared to observations of local dwarfs ( e.g. * ? ? ?
* ; * ? ? ?
it is also important to note that we have neglected all other forms of star formation feedback aside from supernovae ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
however , these extra sources of energy can simply be approximated by tuning @xmath89 , the efficiency with which the energy is transferred to the dark matter , using the energy from supernovae as a baseline .
+ the ability for a given halo to form a core will depend on the coupling between the depth of the central halo potential and the star formation rate which limits the amount of available feedback energy .
the bursty star formation at the centres of dwarf galaxies should be able to drive @xmath89 high enough to form a sizeable core @xcite at high redshift @xcite .
however , low mass halos may not be able to form cores if their galactic potential is shallow enough that each burst of star formation results in completely efficient gas expulsion ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
+ the authors wish to thank nserc for support and r. kuzio de naray for providing the observed rotation profile data .
natexlab#1#1 , j. j. , simon , j. d. , fabricius , m. h. , et al .
2014 , apj , 789 , 63 , o. , kravtsov , a. v. , leitner , s. n. , & gnedin , n. y. 2013 , apj , 770 , 25 , n. c. , agnello , a. , & evans , n. w. 2013 , mnras , 429 , l89 , n. c. , zavala , j. , & de boer , t. j. l. 2014 , apjl , 782 , l39 , p. s. , conroy , c. , & wechsler , r. h. 2010 , apj , 717 , 379 , j. , & tremaine , s. 2008 , galactic dynamics : second edition ( princeton university press ) , a. 1981 , aj , 86 , 1825 , g. l. , & norman , m. l. 1998 , apj , 495 , 80 , j. s. , kolatt , t. s. , sigad , y. , et al .
2001 , mnras , 321 , 559 , a. 1995 , apjl , 447 , l25 , s. , carignan , c. , & freeman , k. c. 2000 , aj , 120 , 3027 , j. j. , williams , b. f. , seth , a. c. , et al .
2009 , apjs , 183 , 67 , w. 2005 , mnras , 360 , 892 , j. , berentzen , i. , & shlosman , i. 2009 , apj , 697 , 293 , j. , & carlberg , r. g. 1991 , apj , 378 , 496 , a. , shlosman , i. , & hoffman , y. 2001 , apj , 560 , 636 , r. a. , & primack , j. r. 1994 , apjl , 427 , l1 , g. , wilkinson , m. i. , wyse , r. f. g. , et al .
2007 , apj , 663 , 948 , f. , brook , c. , mayer , l. , et al .
2010 , nature , 463 , 203 , f. , zolotov , a. , pontzen , a. , et al .
2012 , mnras , 422 , 1231 , c. j. , & dalcanton , j. j. 2000 , phrvd , 62 , 063511 , p. f. , quataert , e. , & murray , n. 2011 , mnras , 417 , 950 , j. r. , & sellwood , j. a. 2009 , apj , 691 , 1300 , b. w. , wadsley , j. , benincasa , s. m. , & couchman , h. m. p. 2014 , mnras , 442 , 3013 , a. , kravtsov , a. v. , bullock , j. s. , & primack , j. r. 2001 , apj , 554 , 903 , a. 2010 , adast , 2010 , p. 2001 , mnras , 322 , 231 , r. , mcgaugh , s. s. , & de blok , w. j. g. 2008 , apj , 676 , 920 , r. , mcgaugh , s. s. , de blok , w. j. g. , & bosma , a. 2006 , apjs , 165 , 461 , c. , schaerer , d. , goldader , j. d. , et al . 1999 ,
apjs , 123 , 3 , c .- p .
, & boylan - kolchin , m. 2004 , phrvl , 93 , 021301 , a. v. , dutton , a. a. , van den bosch , f. c. , et al .
2007 , mnras , 378 , 55 , p. , shen , s. , & governato , f. 2014 , apjl , 789 , l17 , s. , couchman , h. m. p. , & wadsley , j. 2006 , nature , 442 , 539 , s. , wadsley , j. , & couchman , h. m. p. 2008
, science , 319 , 174 , a. j. , wadsley , j. , couchman , h. m. p. , & mashchenko , s. 2012 , apjl , 755 , l35 , a. j. , wadsley , j. , couchman , h. m. p. , & sills , a. 2014 , mnras , 439 , 2043 , a. w. 2012 , aj , 144 , 4 , b. 1994 , nature , 370 , 629 , b. p. , somerville , r. s. , maulbetsch , c. , et al .
2010 , apj , 710 , 903 , j. f. , eke , v. r. , & frenk , c. s. 1996 , mnras , 283 , l72 , j. f. , frenk , c. s. , & white , s. d. m. 1995 , mnras , 275 , 720 . 1996 , apj , 462 , 563 . 1997 , apj , 490 , 493 , s .- h . , de blok , w. j. g. , brinks , e. , walter , f. , & kennicutt , jr . , r. c. 2011 , aj , 141 , 193 , j. , pontzen , a. , walker , m. g. , & koposov , s. e. 2012 , apjl , 759 , l42 , a. , & governato , f. 2012 , mnras , 421 , 3464 . 2014 , nature , 506 , 171 , j. i. , & gilmore , g. 2005 , mnras , 356 , 107 , v. c. , ford , w. k. j. , & .
thonnard , n. 1980 , apj , 238 , 471 , v. c. , thonnard , n. , & ford , jr . , w. k. 1978 , apjl , 225 , l107 , j. a. 2003 , apj , 587 , 638 , s. , madau , p. , conroy , c. , governato , f. , & mayer , l. 2014 , apj , 792 , 99 , s. , wadsley , j. , & stinson , g. 2010 , mnras , 407 , 1581 , d. n. , & steinhardt , p. j. 2000 , phrvl , 84 , 3760 , j. , potter , d. , moore , b. , et al .
2009 , mnras , 398 , l21 , a. 2003 , ijmpd , 12 , 1157 , r. , pontzen , a. , dubois , y. , & read , j. i. 2013 , mnras , 429 , 3068 , m. g. , & pearrubia , j. 2011 , apj , 742 , 20 , m. d. , & katz , n. 2002 , apj , 580 , 627 , l. m. 2000 , apjs , 131 , 39 , a. , brooks , a. m. , willman , b. , et al .
2012 , apj , 761 , 71
figure [ fig : kdnm ] repeats the analysis shown in figure [ fig : kdn ] for all nine low surface brightness galaxies observed by @xcite . [
fig : kdnm ]
we determine the best fit by finding the minimum in the square of the velocity residual , @xmath95 : @xmath96 where @xmath97 is the velocity residual at each measured radius and @xmath98 is the measured velocity error from @xcite .
if the density models used in the text were linear in the parameters @xmath12 , @xmath99 , and @xmath77 , the uncertainties could be derived by using the @xmath100 probability distribution to determine confidence levels .
since this is not the case , the probability distribution of @xmath95 can not simply be approximated to that of @xmath100 . instead , we compute an uncertainty contour given by : @xmath101 where @xmath102 is the number of free parameters
two for equations and and one for equation since the central density is fixed . in figure
[ fig : kdnc ] , these contours are shown by the solid black and gray lines for equations and , respectively , for the sample of nine galaxies from @xcite .
velocity data points that showed large variations from the expected velocity profile were ignored when computing the error contours ; these are shown as open circles in figure [ fig : kdnm ] .
the location of the minimum @xmath95 for each are given by the filled triangles . for most of these galaxies , minimizing the residuals gives relatively tight constraints on the scale radius and power index for the density profiles .
however , ugc 4325 and ddo 64 do not have closed contours , which is unsurprising given that constraining @xmath12 requires the velocity profile to begin to turn over traits that are not exhibited in the velocity data for these galaxies .
instead , we can only define acceptable lower limits for the density profile parameters . | we present a new analytic estimate for the energy required to create a constant density core within a dark matter halo .
our new estimate , based on more realistic assumptions , leads to a required energy that is orders of magnitude lower than is claimed in earlier work .
we define a core size based on the logarithmic slope of the dark matter density profile so that it is insensitive to the functional form used to fit observed data .
the energy required to form a core depends sensitively on the radial scale over which dark matter within the cusp is redistributed within the halo .
simulations indicate that within a region of comparable size to the active star forming regions of the central galaxy that inhabits the halo , dark matter particles have their orbits radially increased by a factor of 23 during core formation .
thus the inner properties of the dark matter halo , such as halo concentration , and final core size , set the energy requirements . as a result ,
the energy cost increases slowly with halo mass as @xmath0@xmath1 for core sizes @xmath21kpc .
we use the expected star formation history for a given dark matter halo mass to predict dwarf galaxy core sizes .
we find that supernovae alone would create well over 4kpc cores in @xmath3 dwarf galaxies _ if _ 100% of the energy were transferred to dark matter particle orbits .
we can directly constrain the efficiency factor by studying galaxies with known stellar content and core size , such as fornax .
we find that the efficiency of coupling between stellar feedback and dark matter orbital energy need only be at the 1% level or less to explain fornax s 1 kpc core . |
heavy quark effective theory ( hqet ) @xcite is a framework which is widely used to study the spectra and transition amplitudes of heavy hadrons containing one heavy quark . in hqet , the expansion is performed in terms of @xmath8 , where @xmath9 is the mass of the heavy quark involved . at the leading order of @xmath8 ,
the hqet lagrangian respects the heavy quark flavor - spin symmetry , therefore heavy hadrons form a series of degenerate doublets . the two members in a doublet share the same quantum number @xmath10 , the angular momentum of the light components .
the @xmath11 @xmath12-wave doublet @xmath5 is conventionally denoted as @xmath13 and the @xmath14 @xmath15-wave doublets @xmath16 are conventionally denoted as @xmath17 .
we denote the @xmath18 @xmath7-wave doubtlets @xmath19 as @xmath20 .
shifman - vainshtein - zakharov ( svz ) sum rules @xcite is a nonperturbative approach used to determine hadronic parameters such as the hadron mass .
the vacuum expectation value of the @xmath21 product of two interpolating currents is considered in this approach . after performing the operator product expansion ( ope ) , one obtains sum rules which relate the hadronic parameters to expressions containing vacuum condensates parameterizing the qcd nonperturbative effect . in the late 1980s ,
light - cone qcd sum rules ( lcqsr ) @xcite was developed to calculate various hadronic transition form factors .
now the ope of the @xmath21 product of two interpolating currents sandwiched between the vacuum and an hadronic state is performed near the light - cone rather than at a small distance as in the conventional svz sum rules .
the @xmath1 coupling constants @xmath22 and @xmath23 were calculated with lcqsr in full qcd in ref . @xcite .
the couplings @xmath24 , @xmath25 , @xmath26 , and @xmath27 were calculated in full qcd in ref .
their values in the limit @xmath28 are also discussed in this paper .
the @xmath1 coupling constants between the three doublets @xmath13 , @xmath12 , @xmath21 and within the two doublets @xmath13 , @xmath12 are systematically studied with lcqsr at the leading order of hqet in ref .
the @xmath0 coupling constants between the @xmath12-wave and @xmath15-wave heavy mesons have been studied using qcd sum rules or / and lcqsr in ref .
the @xmath0 coupling constants between @xmath29/@xmath30 and @xmath13/@xmath12/@xmath21 are calculated with lcqsr at the leading order of hqet in ref .
@xcite . the radiative decay between @xmath13 , @xmath12 , and @xmath21 are studied using the light - cone qcd sum rule at the leading order of hqet in ref .
@xcite . in ref .
@xcite , the radiative decays of @xmath31 and @xmath32 are studied using lcqsr approach . in this work
, we use lcqsr to calculate the @xmath0 , @xmath1 , @xmath2 , and @xmath3 coupling constants between the doublets @xmath29 and @xmath13/@xmath12 . because of the covariant derivative in the interpolating currents of the @xmath29 doublet , the contribution from the 3-particle light - cone distribution amplitudes of @xmath0 , @xmath1 , @xmath2 , and @xmath3 have to be included .
we work in hqet to differentiate the two states with the same @xmath33 value and yet quite different decay widths .
the interpolating currents @xmath34 adopted in our work have been properly constructed in ref .
they satisfy @xmath35 in the limit @xmath28 . here
@xmath36 is the polarization tensor for the spin @xmath37 state , @xmath38 is the velocity of the heavy quark , @xmath39 , @xmath40 denotes symmetrizing the indices and subtracting the trace terms separately in the sets @xmath41 and @xmath42 .
we shall perform the calculation at the leading order of hqet . according to ref .
@xcite , the interpolating currents for the doublets @xmath5 , @xmath6 , and @xmath4 read as @xmath43 where @xmath44 is the heavy quark field in hqet , @xmath45 , @xmath46 , @xmath47 , and @xmath48 is the velocity of the heavy quark .
we consider the @xmath0 decay of @xmath49 to @xmath50 to illustrate our calculation . here
the subscript of @xmath29(@xmath13 ) indicates the spin of the meson involved .
owing to the conservation of the angular momentum of the light components in the limit @xmath28 , there is only one independent @xmath0 coupling constant between doublets @xmath29 and @xmath13 .
we denote it as @xmath51 where @xmath52 and the number following it indicate the orbital and total angular momentum @xmath53 of the final @xmath0 meson respectively .
@xmath51 can be defined in terms of the decay amplitude @xmath54 as @xmath55g^{p1}_{m_2h_1\pi}\,,\end{aligned}\ ] ] where @xmath56 and @xmath57 denote the polarization tensors of the initial and final heavy mesons respectively , @xmath58 is the momentum of the @xmath0 meson .
the transversal tensor are defined as @xmath59 , @xmath60 , and @xmath47 .
@xmath61 for the charged and neutral @xmath0 meson , respectively . to obtain the sum rules for the coupling constants @xmath62 , we consider the correlation function @xmath63g^{p1}_{m_2h_1\pi}(\omega,\omega')\,,\end{aligned}\ ] ] where @xmath64 , @xmath65 . at the leading order of hqet
, the heavy quark propagator reads as @xmath66 the correlation function can now be expressed as @xmath67 \langle\pi(q)|q(x)\bar{q}(0)|0\rangle\biggl\rbrace\,.\end{aligned}\ ] ] it can be further calculated using the light - cone wave functions of the @xmath0 meson . to our approximation , we need the 2- and 3-particle light - cone wave functions .
their definitions are collected in appendix [ appendixlcda ] . at the hadron level , @xmath68 in ( [ eq : corhadpi ] ) has the following pole terms @xmath69 where @xmath70 , @xmath71 , @xmath72 , etc .
are the overlap amplitudes of their interpolating currents with the heavy mesons .
@xmath73 can now be expressed by the @xmath0 meson light - cone wave functions .
after the wick rotation and the double borel transformation with @xmath2 and @xmath74 , the single - pole terms in ( [ eq : poletermpi ] ) are eliminated .
we arrive at @xmath75t^2f_1(\frac{\omega_c}{t})\nonumber\\ & & \mathrel{\phantom{= } } -\frac{4m_\pi^2}{m_u+m_d}\left[6\mathcal{t}^{[1,0]}(u_0 ) + 6\phi_p(\bar{u}_0 ) -6(u\phi_p)(\bar{u}_0 ) + \phi_\sigma(\bar{u}_0)\right]tf_0(\frac{\omega_c}{t})\nonumber\\ & & \mathrel{\phantom{= } } + 3m_\pi^2\left[\mathbb{a}'(\bar{u}_0 ) -(u\mathbb{a})'(\bar{u}_0 ) + 8\mathbb{b}^{[2]}(\bar{u}_0 ) -8\mathbb{b}^{[1]}(\bar{u}_0 ) + 8(u\mathbb{b})^{[1]}(\bar{u}_0 ) -16\mathcal{v}^{[0,0]}_\perp(u_0)-16\mathcal{a}^{[0,0]}_\parallel(u_0)\right]\biggl\rbrace\,,\end{aligned}\ ] ] where @xmath76 is the continuum subtraction factor , and @xmath77 is the continuum threshold , @xmath78 , @xmath79 , and @xmath80 .
@xmath81 and @xmath82 are the two borel parameters .
we have employed the borel transformation @xmath83 to obtain ( [ eq : sumrulepi1 ] ) . in the above expressions ,
we have used the functions @xmath84}(\bar{u}_0)$ ] and @xmath85}(u_0)$ ] which are defined in appendix [ appendixf ] .
the @xmath0 coupling constant between doublets @xmath29 and @xmath86 can be defined similarly : @xmath87g^{d2}_{m_1t_1\pi}\,,\nonumber \\ \mathcal { m}(m_2\rightarrow m_1+\pi ) & = & 2i\eta_{\alpha_1\alpha_2}\left[\epsilon^{*\alpha_1}_tq^{\alpha_2}_t-\frac{1}{3}g_t^{\alpha_1\alpha_2}(\epsilon^*\cdot q_t)\right]g^{p1}_{m_2m_1\pi}\nonumber\\ & & + i\eta_{\alpha_1\alpha_2}\left\{q^{\alpha_1}_tq^{\alpha_2}_t(\epsilon^*\cdot q_t ) -\frac{q_t^2}{5}\left[2\epsilon^{*\alpha_1}_tq^{\alpha_2}_t+g_t^{\alpha_1\alpha_2}(\epsilon^*\cdot q_t)\right]\right\}g^{f3}_{m_2m_1\pi}\,.\end{aligned}\ ] ] here the vector notations in the levi - civita tensor come from an index contraction between the levi - civita tensor and the vectors , for example , @xmath88 .
the levi - civita tensor is defined as @xmath89 .
the sum rules for the coupling constants in eq .
( [ piamplitude ] ) are @xmath90tf_0(\frac{\omega_c}{t } ) + \frac{4m_\pi^2}{m_u+m_d}\left[6\mathcal{t}^{[0,0]}(u_0 ) -\phi_\sigma(\bar{u}_0 ) + ( u\phi_\sigma)(\bar{u}_0)\right]\nonumber\\ & & \mathrel{\phantom{= } } -3m_\pi^2\left[\mathbb{a}(\bar{u}_0 ) -(u\mathbb{a})(\bar{u}_0 ) + 16\mathcal{a}^{[-1,0]}_\parallel(u_0 ) + 16\mathcal{a}^{[-1,0]}_\perp(u_0)\right]\frac{1}{t}\biggl\rbrace\,,\nonumber \\ & & g^{s0}_{m_1t_1\pi}f_{+,\frac{3}{2}}f_{-,\frac{3}{2}}e^{-\frac{\bar{\lambda } _ { -,3/2}+\bar{\lambda } _ { + , 3/2}}{t}}\nonumber \\ & & = \frac{f_\pi}{48}(u(1-u)\phi_\pi)^{(3)}(\bar{u}_0)t^4f_3(\frac{\omega_c}{t})+\frac{f_\pi m_\pi^2}{24m_{ud } } \left[\frac{m_\pi^2-m_{ud}^2}{6m_\pi^2}\phi_\sigma^{(2)}(\bar{u}_0 ) + ( u(1-u)\phi_p)^{(2)}(\bar{u}_0)+(1-\alpha_2\mathcal{t})^{[3,0]}(u_0)\right.\nonumber \\ & & \mathrel{\phantom{=}}\left .
-(\alpha_3\mathcal{t})^{[3,1]}(u_0)\right]t^3f_2(\frac{\omega_c}{t})-\frac{f_\pi m_\pi^2}{12 } \left[\frac{3}{8}\mathbb{a}'(\bar{u}_0)+\frac{1}{16}(u(1-u)\mathbb{a})^{(3)}(\bar{u}_0 ) + \mathbb{b}^{[1]}(\bar{u}_0)+(u\mathbb{b}-\frac{1}{2}\mathbb{b})(\bar{u}_0)\right.\nonumber \\ & & \mathrel{\phantom{= } } -\frac{1}{2}(u(1-u)\mathbb{b})'(\bar{u}_0)+(u(1-u)\phi_\pi)'(\bar{u}_0 ) + ( 2\mathcal{a}_\parallel+2\mathcal{a}_\perp+\mathcal{v}_\parallel+\mathcal{v}_\perp)^{[1,0]}(u_0 ) + ( ( 1-\alpha_2)(\mathcal{a}_\parallel+\mathcal{v}_\perp))^{[2,0]}(u_0)\nonumber \\ & & \mathrel{\phantom{=}}\left . -(\alpha_3(\mathcal{a}_\parallel+\mathcal{v}_\perp))^{[2,1]}(u_0)\right]t^2f_1(\frac{\omega_c}{t } ) -\frac{f_\pi m_\pi^4}{6m_{ud}}\left[(u(1-u)\phi_p)(\bar{u}_0 ) + \frac{m_\pi^2-m_{ud}^2}{6m_\pi^2}\phi_\sigma(\bar{u}_0 ) + ( ( 1-\alpha_2)\mathcal{t})^{[1,0]}(u_0)\right.\nonumber \\ & & \mathrel{\phantom{=}}\left .
-(\alpha_3\mathcal{t})^{[1,1]}(u_0)\right]tf_0(\frac{\omega_c}{t } ) + \frac{f_\pi m_\pi^4}{3}\left[\frac{1}{16}(u(1-u)\mathbb{a})'(\bar{u}_0 ) + ( u\mathbb{b}-\frac{1}{2}\mathbb{b})^{[2]}(\bar{u}_0 ) -\frac{1}{2}(u(1-u)\mathbb{b})^{[1]}(\bar{u}_0)\right.\nonumber \\ & & \mathrel{\phantom{=}}\left .
+ \mathbb{b}^{[3]}(\bar{u}_0 ) + ( \mathcal{a}_\parallel+\mathcal{a}_\perp-\mathcal{v}_\parallel-\mathcal{v}_\perp)^{[-1,0]}(u_0 ) -((1-\alpha_2)(\mathcal{a}_\parallel+\mathcal{v}_\perp))^{[0,0]}(u_0 ) + ( \alpha_3(\mathcal{a}_\parallel+\mathcal{v}_\perp))^{[0,1]}(u_0)\right]\,,\nonumber \\ & & g^{d2}_{m_1t_1\pi}f_{+,\frac{3}{2}}f_{-,\frac{3}{2}}e^{-\frac{\bar{\lambda } _ { -,3/2}+\bar{\lambda } _ { + , 3/2}}{t}}\nonumber \\ & & = -\frac{f_\pi}{8}(u(1-u)\phi_\pi)'(\bar{u}_0)t^2f_1(\frac{\omega_c}{t } ) -\frac{f_\pi m_\pi^2}{4m_{ud}}\left[(u(1-u)\phi_p)(\bar{u}_0 ) + \frac{m_\pi^2-m_{ud}^2}{6m_\pi^2}\phi_\sigma(\bar{u}_0 ) + ( ( 1-\alpha_2)\mathcal{t})^{[1,0]}(u_0)\right.\nonumber \\ & & \mathrel{\phantom{=}}\left .
-(\alpha_3\mathcal{t})^{[1,1]}(u_0)\right]tf_0(\frac{\omega_c}{t } ) + \frac{f_\pi m_\pi^2}{2}\left[\frac{1}{16}(u(1-u)\mathbb{a})'(\bar{u}_0 ) -\frac{1}{2}(u(1-u)\mathbb{b})^{[1]}(\bar{u}_0 ) + \mathbb{b}^{[3]}(\bar{u}_0)+((u-\frac{1}{2})\mathbb{b})^{[2]}(\bar{u}_0)\right.\nonumber \\ & & \mathrel{\phantom{=}}\left .
+ ( \mathcal{a}_\parallel+\mathcal{a}_\perp+2\mathcal{v}_\parallel+2\mathcal{v}_\perp)^{[-1,0]}(u_0 ) -((1-\alpha_2)(\mathcal{a}_\parallel+\mathcal{v}_\perp))^{[0,0]}(u_0 ) + ( \alpha_3(\mathcal{a}_\parallel+\mathcal{v}_\perp))^{[0,0]}(u_0)\right]\,,\nonumber \\ & & g^{p1}_{m_2m_1\pi}f_{-,\frac{3}{2}}^2e^{-\frac{2\bar{\lambda } _ { -,3/2}}{t}}\nonumber \\ & & = \frac{f_\pi m_{ud}}{80\sqrt{6}}(u(1-u)\phi_\pi)^{(2)}(\bar{u}_0)t^3f_2(\frac{\omega_c}{t } ) -\frac{f_\pi m_\pi^2}{240\sqrt{6}m_{ud } } \left[\frac{m_\pi^2-m_{ud}^2}{m_\pi^2}(u(1-u)\phi_\sigma)^{(2)}(\bar{u}_0 ) + 6((1-\alpha_2)\mathcal{t})^{[2,0]}(u_0)\right.\nonumber \\ & & \mathrel{\phantom{=}}\left .
-6(\alpha_3\mathcal{t})^{[2,1]}(u_0)\right]t^2f_1(\frac{\omega_c}{t } ) + \frac{f_\pi m_\pi^2}{20\sqrt{6}}\left[-\frac{5}{8}\mathbb{a}(\bar{u}_0 ) -\frac{1}{16}(u(1-u)\mathbb{a})^{(2)}(\bar{u}_0)-(u(1-u)\phi_\pi)(\bar{u}_0)\right.\nonumber \\ & & \mathrel{\phantom{=}}\left .
-4(\mathcal{a}_\parallel+\mathcal{a}_\perp)^{[0,0]}(u_0 ) + ( ( 1-\alpha_2)(\mathcal{a}_\parallel+\mathcal{a}_\perp))^{[1,0]}(u_0 ) -(\alpha_3(\mathcal{a}_\parallel+\mathcal{a}_\perp))^{[1,1]}(u_0 ) -5(\mathcal{v}_\parallel+\mathcal{v}_\perp)^{[0,0]}(u_0)\right]\times\nonumber \\ & & \mathrel{\phantom{=}}tf_0(\frac{\omega_c}{t } ) + \frac{f_\pi m_\pi^4}{60\sqrt{6}m_{ud}}\left[\frac{m_\pi^2-m_{ud}^2}{m_\pi^2}(u(1-u)\phi_\sigma)(\bar{u}_0 ) -6((1-\alpha_2)\mathcal{t})^{[0,0]}(u_0 ) + 6(\alpha_3\mathcal{t})^{[0,1]}(u_0)\right ] + \frac{f_\pi m_\pi^4}{5\sqrt{6}}\times \nonumber \\ & & \mathrel{\phantom{= } } \left[\frac{1}{16}(u(1-u)\mathbb{a})(\bar{u}_0 ) -(\mathcal{a}_\parallel+\mathcal{a}_\perp)^{[-2,0]}(u_0)+((1-\alpha_2)(\mathcal{a}_\parallel+\mathcal{a}_\perp))^{[-1,0]}(u_0 ) -(\alpha_3(\mathcal{a}_\parallel+\mathcal{a}_\perp))^{[-1,1]}(u_0)\right]\frac{1}{t}\,,\nonumber \\ & & g^{f3}_{m_2m_1\pi}f_{-,\frac{3}{2}}^2e^{-\frac{2\bar{\lambda } _ { -,3/2}}{t}}\nonumber \\ & & = -\frac{\sqrt{6}f_\pi}{4}(u(1-u)\phi_\pi)(\bar{u}_0)tf_0(\frac{\omega_c}{t } ) +
\frac{\sqrt{6}f_\pi m_\pi^2}{2m_{ud}}\left[\frac{m_\pi^2-m_{ud}^2}{6m_\pi^2}(u(1-u)\phi_\sigma)(\bar{u}_0 ) -((1-\alpha_2)\mathcal{t})^{[0,0]}(u_0)\right.\nonumber \\ & & \mathrel{\phantom{=}}\left .
+ ( \alpha_3\mathcal{t})^{[0,1]}(u_0)\right ] + \sqrt{6}f_\pi m_\pi^2\left[\frac{1}{16}(u(1-u)\mathbb{a})(\bar{u}_0 ) -(\mathcal{a}_\parallel+\mathcal{a}_\perp)^{[-2,0]}(u_0 ) + ( ( 1-\alpha_2)(\mathcal{a}_\parallel+\mathcal{a}_\perp))^{[-1,0]}(u_0)\right.\nonumber \\ & & \mathrel{\phantom{=}}\left . -(\alpha_3(\mathcal{a}_\parallel+\mathcal{a}_\perp))^{[-1,1]}(u_0)\right]\frac{1}{t}\,,\end{aligned}\ ] ] with @xmath91 .
the @xmath0 coupling constants of the other decay channels are defined as @xmath92g^{d2}_{m_1s_1\pi}\,,\nonumber \\
\mathcal { m}(m_2\rightarrow s_0+\pi ) & = & i\eta_{\alpha_1\alpha_2}\left[q_t^{\alpha_1}q_t^{\alpha_2}-\frac{1}{3}g_t^{\alpha_1\alpha_2}q_t^2\right]g^{d2}_{m_2s_0\pi}\,,\nonumber \\
\mathcal { m}(m_1\rightarrow t_2+\pi ) & = & 2ii\epsilon^*_{\beta_1\beta_2}\varepsilon^{\beta_1\eta qv}q_t^{\beta_2}g^{d2}_{m_1t_2\pi}\,,\nonumber \\
\mathcal { m}(m_2\rightarrow t_1+\pi ) & = & 2ii\eta_{\alpha_1\alpha_2}\varepsilon^{\alpha_1\epsilon^ * qv}q_t^{\alpha_2}g^{d2}_{m_2t_1\pi}\,,\nonumber \\ \mathcal { m}(m_2\rightarrow t_2+\pi ) & = & 2i\eta_{\alpha_1\alpha_2}\epsilon^*_{\beta_1\beta_2 } \left[g_t^{\alpha_1\beta_1}g_t^{\alpha_2\beta_2}-\frac{1}{3}g_t^{\alpha_1\alpha_2}g_t^{\beta_1\beta_2}\right]g^{s0}_{m_2t_2\pi}\nonumber \\ & & + i\eta_{\alpha_1\alpha_2}\epsilon^*_{\beta_1\beta_2 } \left\{q_t^{\alpha_1}q_t^{\alpha_2}g_t^{\beta_1\beta_2}+q_t^{\beta_1}q_t^{\beta_2}g_t^{\alpha_1\alpha_2}\right.\nonumber \\ & & \left.-3q_t^{\alpha_1}q_t^{\beta_1}g_t^{\alpha_2\beta_2 } -q_t^2\left[\frac{2}{3}g_t^{\alpha_1\alpha_2}g_t^{\beta_1\beta_2}-g_t^{\alpha_1\beta_1}g_t^{\alpha_2\beta_2}\right]\right\}g^{d2}_{m_2t_2\pi}\,,\nonumber \\ \mathcal { m}(m_1\rightarrow m_1+\pi ) & = & ii\varepsilon^{\eta\epsilon^ * qv } g^{p1}_{m_1m_1\pi}\,,\nonumber \\ \mathcal { m}(m_2\rightarrow m_1+\pi ) & = & 2i\eta_{\alpha_1\alpha_2}\left[\epsilon^{*\alpha_1}_tq_t^{\alpha_2 } -\frac{1}{3}g_t^{\alpha_1\alpha_2}(\epsilon^*\cdot q_t)\right]g^{p1}_{m_2m_1\pi}\nonumber \\ & & + i\eta_{\alpha_1\alpha_2 } \left\{q_t^{\alpha_1}q_t^{\alpha_2}(\epsilon^*\cdot q_t ) -\frac{q_t^2}{5}\left[2\epsilon_t^{*\alpha_1}q_t^{\alpha_2}+(\epsilon^*\cdot q_t)g_t^{\alpha_1\alpha_2}\right]\right\}g^{f3}_{m_2m_1\pi}\,,\nonumber \\
\mathcal { m}(m_2\rightarrow m_2+\pi ) & = & 4ii\eta_{\alpha_1\alpha_2}\eta^*_{\beta_1\beta_2}\varepsilon^{\alpha_1\beta_1qv}g_t^{\alpha_2\beta_2 } g^{p1}_{m_2m_2\pi}\nonumber \\ & & + 4ii\eta_{\alpha_1\alpha_2}\eta^*_{\beta_1\beta_2}\varepsilon^{\alpha_1\beta_1qv } \left[q_t^{\alpha_2}q_t^{\beta_2}-\frac{q_t^2}{5}g_t^{\alpha_2\beta_2}\right]g^{f3}_{m_2m_2\pi}.\end{aligned}\ ] ] we notice that @xmath93 these relations are consistent with the hqet expectation at the leading order . to get the numerical values of the six independent coupling constants , we need the following mass parameters @xmath94 s and @xmath95 s , the overlap amplitudes of these interpolating currents : [ cols="^,^,^,^,^ " , ]
we have calculated the @xmath0 , @xmath1 , @xmath2 , and @xmath3 couplings with heavy mesons at the leading order of hqet within the framework of lcqsr .
most of the sum rules obtained are stable with the variations of the borel parameter @xmath21 and the continuum threshold @xmath77 . for the other sum rules , we can not find a stable working interval of @xmath21 .
the extracted vector meson heavy meson coupling constants may be helpful in the study of the interaction between two @xmath96 mesons .
some possible sources of the errors in our calculation include the inherent inaccuracy of lcqsr : the omission of the higher twist terms in the ope near the light - cone , the variation of the coupling constant with the continuum threshold @xmath77 and the borel parameter @xmath21 in the working interval , the omission of the higher conformal partial waves in the light - cone distribution amplitudes of the involved mesons or photon , and the uncertainty in the parameters that appear in these light - cone distribution amplitudes .
the uncertainty in @xmath95 s and @xmath94 s is another source of errors .
it s also understood that the @xmath8 correction may turn out to be quite large for the charm mesons while such a correction is under control for the bottom ssytem . with these extracted coupling constants
, we have made a very rough estimate of the decay widths of the @xmath4 @xmath7-wave heavy mesons .
their dominant decay modes are @xmath97 and @xmath98 .
the former mode is of @xmath12-wave while the latter decay occurs in the @xmath15-wave but with a larger phase space .
the total width of the @xmath7-wave heavy mesons is roughly several tens of mev .
therefore the @xmath4 @xmath7-wave heavy mesons are not expected to be extremely broad , which is helpful in the future experimental search .
this project is supported by the national natural science foundation of china under grants no .
10625521 , no . 10721063 , and the ministry of science and technology of china ( 2009cb825200 ) . 99 b. grinstein , nucl
b339 * , 253 ( 1990 ) ; e. eichten and b. hill , phys . lett . *
b234 * , 511 ( 1990 ) ; a. f. falk , h. georgi , b. grinstein , and m. b. wise , nucl . phys . *
b343 * , 1 ( 1990 ) .
shifman , a.i .
vainshtein , and v.i .
zakharov , nucl .
* b174 * , 385 , 448 , 519 ( 1979 ) .
balitsky , v.m .
braun , and a.v .
kolesnichenko , nucl .
b312 * , 509 ( 1989 ) ; v.m .
braun and i.e. filyanov , z. phys .
* c44 * , 157 ( 1989 ) ; v.l .
chernyak and i.r .
zhitnitsky , nucl . phys . * b345 * , 137 ( 1990 ) .
t. m. aliev , d. a. demir , e.iltan , and n. k. pak , phys .
* d53 * , 355 ( 1996 ) .
z. h. li , w. liu , and h. y. liu , phys . lett . *
b659 * , 598 ( 2008 ) .
s. l. zhu and y. b. dai , phys . rev .
* d58 * , 094033 ( 1998 ) ; p. z. huang , l. zhang , and s. l. zhu , phys . rev .
* d80 * , 014023 ( 2009 ) .
p. colangelo , g. nardulli , a. deandrea , n. di bartolomeo , raoul gatto , and f. feruglio , phys .
b339 * , 151 ( 1994 ) ; p. colangelo , f. de fazio , g. nardulli , n. di bartolomeo , and raoul gatto , phys . rev . *
d52 * , 6422 ( 1995 ) ; p. colangelo and f. de fazio , eur .
j. * c4 * , 503 ( 1998 ) .
y. b. dai and s. l. zhu , phys . rev . * d58 * , 074009 ( 1998 ) ; euro .
j. * c6 * , 307 ( 1999 ) ; s. l. zhu and y. b. dai , phys . lett . *
b429 * , 72 ( 1998 ) .
w. wei , x. liu , and s. l. zhu , phys . rev . *
d75 * , 014013 ( 2007 ) .
s. l. zhu and y. b. dai , phys . rev .
* d59 * , 114015 ( 1999 ) .
p. colangelo , f. de fazio , and a. ozpineci , phys . rev . *
d72 * , 074004 ( 2005 ) .
y. b. dai , c. s. huang , m. q. huang , and c. liu , phys . lett . *
b390 * , 350 ( 1997 ) ; y. b. dai , c. s. huang , and m. q. huang , phys . rev . *
d55 * , 5719 ( 1997 ) .
e. bagen , p. ball , v. m. braun , and h. g. dosch , phys . lett . *
b278 * , 457 ( 1992 ) ; m. neubert , phys . rev . * d45 * , 2451(1992 ) ; d. j. broadhurst and a. g. grozin , phys . lett . * b274 * , 421 ( 1992 ) .
s. l. zhu and y. b. dai , mod .
. lett . * a14 * , 2367 ( 1999 ) .
p. ball , jhep * 9901 * , 010 ( 1999 ) ; p. ball , v. m. braun , and a. lenz , jhep * 0605 * , 004 ( 2006 ) .
i. i. balitsky , v. m. braun , and a. v. kolesnichenko , nucl . phys . *
b312 * , 509 ( 1989 ) ; v. m. braun and i. e. filyanov , z. phys . *
c44 * , 157 ( 1989 ) ; v. m. belyaev , v. m. braun , a. khodjamirian , and r. ruckl , phys . rev .
* d51 * , 6177 ( 1995 ) . p. ball and v. m. braun , phys
* d58 * , 094016 ( 1998 ) ; p. ball and r. zwicky , phys .
rev . * d71 * , 014029 ( 2005 ) .
p. ball and g. w. jones , jhep * 0703 * , 069 ( 2007 ) ; p. ball , v. m. braun , and a. lenz , jhep * 0708 * , 090 ( 2007 ) . p. ball , v. m. braun , and n. kivel , nucl .
b649 * , 263 ( 2003 ) .
i. i. balitsky , v. m. braun , and a. v. kolesnichenko , sov .
. phys . * 48 * , 348 ( 1988 ) , [ yad . fiz . * 48 * , 547 ( 1988 ) ] ; nucl . phys .
* b312 * , 509 ( 1989 ) . c. amsler _
et al_. ( particle data group ) , phys .
b667 * , 1 ( 2008 ) .
s. godfrey and n. isgur , phys . rev .
* d32 * , 189 ( 1985 ) .
the 2-particle distribution amplitudes of the @xmath0 meson are defined as @xcite @xmath99 where @xmath100 , @xmath101 is the leading twist-2 distribution amplitude , @xmath102 are of twist-3 .
all the above distribution amplitudes @xmath103 are normalized to unity : @xmath104 .
there is one 3-particle distribution amplitudes of twist-3 , defined as @xcite @xmath105 where we used the following notation for the integral defining the 3-particle distribution amplitude : @xmath106 here @xmath107 is the set of three momentum fractions @xmath108 , @xmath109 , and @xmath110 .
the integration measure is @xmath111 the 3-particle distribution amplitudes of twist-4 are @xmath112 where @xmath113 is the dual field @xmath114 .
the definitions of the distribution amplitudes of the @xmath1 meson used in the text read as @xcite @xmath115\,,\nonumber \\
\langle 0|\bar u(z ) \gamma_{\mu } \gamma_{5 } d(-z)|\rho^-(p,\lambda)\rangle & = & \frac{1}{2}f_{\rho } m_{\rho } \varepsilon_{\mu}^{\phantom{\mu}\nu \alpha \beta } e^{(\lambda)}_{\perp \nu } p_{\alpha } z_{\beta } \int_{0}^{1 } \!du\ , e^{i \xi p \cdot z } g^{(a)}_{\perp}(u , \mu^{2})\,,\nonumber
\\ \langle 0|\bar u(z ) \sigma_{\mu \nu } d(-z)|\rho^-(p,\lambda)\rangle & = & i f_{\rho}^{t } \left [ ( e^{(\lambda)}_{\perp \mu}p_\nu - e^{(\lambda)}_{\perp \nu}p_\mu ) \int_{0}^{1 } \!du\ , e^{i \xi p \cdot z } \varphi_{\perp}(u , \mu^{2 } ) \right . \nonumber \\ & & + ( p_\mu z_\nu - p_\nu z_\mu ) \frac{e^{(\lambda ) } \cdot z}{(p \cdot z)^{2 } } m_{\rho}^{2 } \int_{0}^{1 } \!du\ , e^{i \xi p \cdot z } h_\parallel^{(t ) } ( u , \mu^{2 } ) \nonumber \\ & & \left.+ \frac{1}{2 } ( e^{(\lambda)}_{\perp \mu } z_\nu -e^{(\lambda)}_{\perp \nu } z_\mu ) \frac{m_{\rho}^{2}}{p \cdot z } \int_{0}^{1 } \!du\ , e^{i \xi p \cdot z } h_{3}(u , \mu^{2 } ) \right]\,,\nonumber \\
\langle 0|\bar u(z)d(-z)|\rho^-(p,\lambda)\rangle & = & -if_{\rho}^{t}(e^{(\lambda)}z)m_\rho^2 \int_{0}^{1 } \!du\ , e^{i \xi p \cdot z } h_\parallel^{(s ) } ( u , \mu^{2})\,.\end{aligned}\ ] ] the distribution amplitude @xmath116 and @xmath117 are of twist-2 , @xmath118 , @xmath119 , @xmath120 and @xmath121 are twist-3 and @xmath122 , @xmath123 are twist-4 . all functions @xmath124 are normalized to satisfy @xmath125 .
the 3-particle distribution amplitudes of the @xmath1 meson are defined as @xcite @xmath126{\cal a}(v , pz ) \nonumber\\ & & { } + f_\rho m_\rho^3\frac{e^{(\lambda)}\cdot z}{pz } [ p_\mu g^\perp_{\alpha\nu}-p_\nu g^\perp_{\alpha\mu } ] \widetilde\phi(v , pz ) \nonumber\\ & & { } + f_\rho m_\rho^3\frac{e^{(\lambda)}\cdot z}{(pz)^2 } p_\alpha [ p_\mu z_\nu - p_\nu z_\mu ] \widetilde\psi(v , pz)\,,\nonumber \\ \langle 0|\bar u(z ) g g_{\mu\nu}i\gamma_\alpha d(-z)|\rho^-(p)\rangle & = & f_\rho m_\rho p_\alpha[p_\nu e^{(\lambda)}_{\perp\mu } - p_\mu e^{(\lambda)}_{\perp\nu}{\cal v}(v , pz ) \nonumber\\ & & { } + f_\rho m_\rho^3\frac{e^{(\lambda)}\cdot z}{pz } [ p_\mu g^\perp_{\alpha\nu } - p_\nu g^\perp_{\alpha\mu } ] \phi(v , pz ) \nonumber\\ & & { } + f_\rho m_\rho^3\frac{e^{(\lambda)}\cdot z}{(pz)^2 } p_\alpha [ p_\mu z_\nu - p_\nu z_\mu ] \psi(v , pz)\,,\nonumber \\
\langle 0|\bar u(z ) \sigma_{\alpha\beta } gg_{\mu\nu}(vz ) d(-z)|\rho^-(p,\lambda)\rangle & = & f_{\rho}^t m_{\rho}^3 \frac{e^{(\lambda)}\cdot z } { 2 ( p \cdot z ) } [ p_\alpha p_\mu g^\perp_{\beta\nu } -p_\beta p_\mu g^\perp_{\alpha\nu } -p_\alpha p_\nu g^\perp_{\beta\mu } + p_\beta p_\nu g^\perp_{\alpha\mu } ] { \cal t}(v , pz ) \nonumber\\ & & + f_{\rho}^t m_{\rho}^2 [ p_\alpha e^{(\lambda)}_{\perp\mu}g^\perp_{\beta\nu } -p_\beta e^{(\lambda)}_{\perp\mu}g^\perp_{\alpha\nu } -p_\alpha e^{(\lambda)}_{\perp\nu}g^\perp_{\beta\mu } + p_\beta e^{(\lambda)}_{\perp\nu}g^\perp_{\alpha\mu } ] t_1^{(4)}(v , pz ) \nonumber\\ & & + f_{\rho}^t m_{\rho}^2 [ p_\mu e^{(\lambda)}_{\perp\alpha}g^\perp_{\beta\nu } -p_\mu e^{(\lambda)}_{\perp\beta}g^\perp_{\alpha\nu } -p_\nu e^{(\lambda)}_{\perp\alpha}g^\perp_{\beta\mu } + p_\nu e^{(\lambda)}_{\perp\beta}g^\perp_{\alpha\mu } ] t_2^{(4)}(v , pz ) \nonumber\\ & & + \frac{f_{\rho}^t m_{\rho}^2}{pz } [ p_\alpha p_\mu e^{(\lambda)}_{\perp\beta}z_\nu -p_\beta p_\mu e^{(\lambda)}_{\perp\alpha}z_\nu -p_\alpha p_\nu e^{(\lambda)}_{\perp\beta}z_\mu + p_\beta p_\nu e^{(\lambda)}_{\perp\alpha}z_\mu ] t_3^{(4)}(v , pz ) \nonumber\\ & & + \frac{f_{\rho}^t m_{\rho}^2}{pz } [ p_\alpha p_\mu e^{(\lambda)}_{\perp\nu}z_\beta -p_\beta p_\mu e^{(\lambda)}_{\perp\nu}z_\alpha -p_\alpha p_\nu e^{(\lambda)}_{\perp\mu}z_\beta + p_\beta p_\nu e^{(\lambda)}_{\perp\mu}z_\alpha ] t_4^{(4)}(v , pz)\,,\nonumber \\ \langle 0|\bar u(z ) gg_{\mu\nu}(vz ) d(-z)|\rho^-(p,\lambda)\rangle & = & i f_{\rho}^t m_{\rho}^2 [ e^{(\lambda)}_{\perp\mu}p_\nu - e^{(\lambda)}_{\perp\nu}p_\mu ] s(v , pz)\ , , \nonumber\\ \langle 0|\bar u(z ) ig\widetilde g_{\mu\nu}(vz)\gamma_5 d(-z)|\rho^-(p,\lambda)\rangle & = & i f_{\rho}^t m_{\rho}^2 [ e^{(\lambda)}_{\perp\mu}p_\nu - e^{(\lambda)}_{\perp\nu}p_\mu ] \widetilde s(v , pz)\,.\end{aligned}\ ] ] the distribution amplitudes @xmath127 , @xmath128 and @xmath129 are of twist-3 and the other 3-particle distribution amplitudes are of twist-4 .
the definitions of the 2-particle distribution amplitudes of the photon read as@xcite @xmath130 the 3-particle distribution amplitudes of the photon are defined as @xcite @xmath131 { \cal a}(v , qz)\,,\nonumber \\
\langle 0 |\bar q(z ) g g_{\mu\nu}(vz)i\gamma_\alpha q(-z)|\gamma^{(\lambda)}(q ) \rangle & = & e_q \,e\,f_{3\gamma}\ , q_\alpha[q_\nu e^{(\lambda)}_{\perp\mu } - q_\mu e^{(\lambda)}_{\perp\nu } ] { \cal v}(v , qz)\,,\nonumber \\
\langle 0 | \bar q(z)g{g}_{\mu\nu}(vz ) q(-z ) | \gamma^{(\lambda)}(q ) \rangle & = & ie_q \,e\,\langle\bar q q\rangle [ q_\nu e^{(\lambda)}_{\perp\mu}-q_\mu e^{(\lambda)}_{\perp\nu } ] s(v , qz)\ , , \nonumber\\ \langle 0| \bar
q(z)g\tilde{g}_{\mu\nu}(vz)i\gamma_5 q(-z ) |\gamma^{(\lambda)}(q ) \rangle & = & ie_{q}\,e\,\langle\bar q q\rangle [ q_\nu e^{(\lambda)}_{\perp\mu } -q_\mu e^{(\lambda)}_{\perp\nu } ] \widetilde{s}(v , qz)\,,\nonumber\\ \langle 0 | \bar q(z)\sigma_{\alpha \beta}g{g}_{\mu\nu}(vz ) q(-z ) |\gamma^{(\lambda)}(q ) \rangle&= & e_{q}\,e\,\langle\bar q q\rangle [ q_\alpha e^{(\lambda)}_{\perp\mu}g^\perp_{\beta\nu } -q_\beta e^{(\lambda)}_{\perp\mu}g^\perp_{\alpha\nu } -q_\alpha e^{(\lambda)}_{\perp\nu}g^\perp_{\beta\mu } + q_\beta e^{(\lambda)}_{\perp\nu}g^\perp_{\alpha\mu } ] t_1(v , qz ) \nonumber\\ & & + e_{q}\,e\,\langle\bar q q\rangle [ q_\mu e^{(\lambda)}_{\perp\alpha}g^\perp_{\beta\nu } -q_\mu e^{(\lambda)}_{\perp\beta}g^\perp_{\alpha\nu } -q_\nu e^{(\lambda)}_{\perp\alpha}g^\perp_{\beta\mu } + q_\nu e^{(\lambda)}_{\perp\beta}g^\perp_{\alpha\mu } ] t_2(v , qz ) \nonumber \\ & & + \frac{e_q\,e\,\langle\bar q q\rangle}{qz } [ q_\alpha q_\mu e^{(\lambda)}_{\perp\beta}z_\nu -q_\beta q_\mu e^{(\lambda)}_{\perp\alpha}z_\nu -q_\alpha q_\nu e^{(\lambda)}_{\perp\beta}z_\mu + q_\beta q_\nu e^{(\lambda)}_{\perp\alpha}z_\mu ] t_3(v , qz ) \nonumber\\ & & + \frac{e_q\,e\,\langle\bar q q\rangle}{qz } [ q_\alpha q_\mu e^{(\lambda)}_{\perp\nu}z_\beta -q_\beta q_\mu e^{(\lambda)}_{\perp\nu}z_\alpha -q_\alpha q_\nu e^{(\lambda)}_{\perp\mu}z_\beta + q_\beta q_\nu e^{(\lambda)}_{\perp\mu}z_\alpha ] t_4(v , qz)\,.\hspace*{1cm}\nonumber\\\end{aligned}\ ] ] here @xmath132 is the leading twist-2 distribution amplitude , @xmath133 , @xmath134 , @xmath127 , and @xmath128 are of twist-3 , @xmath135 , @xmath12 , @xmath136 , and @xmath137 are of twist-4 . the photon distribution amplitudes are @xmath138,\nonumber\\ h_\gamma(u ) & = & -10(1 + 2k^+)c_2^{1/2}(2u-1),\nonumber\\ \psi^v(u ) & = & 5\left(3(2u-1)^2 - 1\right)+\frac{3}{64}\left(15\omega^v_\gamma-5\omega^a_\gamma\right ) \left(3 - 30(2u-1)^2 + 35(2u-1)^4\right),\nonumber\\
\psi^a(u ) & = & \left(1-(2u-1)^2\right)\left(5(2u-1)^2 - 1\right)\frac{5}{2 } \left(1+\frac{9}{16}\omega^v_\gamma-\frac{3}{16}\omega^a_\gamma\right),\nonumber \\ { \mathcal v}(\alpha_q , \alpha_{\bar{q } } , \alpha_g ) & = & 540\omega_\gamma^v(\alpha_q-\alpha_{\bar{q}})\alpha_q\alpha_{\bar{q } } \alpha_g^2,\nonumber\\ { \mathcal a}(\alpha_q , \alpha_{\bar{q } } , \alpha_g ) & = & 360\alpha_q \alpha_{\bar{q}}\alpha_g^2\left[1+\omega_\gamma^a\frac{1}{2}(7\alpha_g-3)\right],\nonumber \\ s(\alpha_q , \alpha_{\bar{q } } , \alpha_g ) & = & 30\alpha_g^2\left[\right.(k+k^+)(1-\alpha_g)+(\zeta_1+\zeta_1^+)(1-\alpha_g)(1 - 2\alpha_g)\nonumber\\ & & + \zeta_2[3(\alpha_{\bar{q}}-\alpha_q)^2-\alpha_g(1-\alpha_g)]\left.\right],\nonumber\\ \widetilde{s}(\alpha_q , \alpha_{\bar{q } } , \alpha_g ) & = & -30\alpha_g^2\left[\right.(k - k^+)(1-\alpha_g)+(\zeta_1-\zeta_1^+)(1-\alpha_g)(1 - 2\alpha_g)\nonumber\\ & & + \zeta_2[3(\alpha_{\bar{q}}-\alpha_q)^2-\alpha_g(1-\alpha_g)]\left.\right],\nonumber\\ t_1(\alpha_q , \alpha_{\bar{q } } , \alpha_g ) & = & -120(3\zeta_2+\zeta_2^+)(\alpha_{\bar{q}}-\alpha_q)\alpha_q\alpha_{\bar{q}}\alpha_g,\nonumber\\ t_2(\alpha_q , \alpha_{\bar{q } } , \alpha_g ) & = & 30\alpha_g^2(\alpha_{\bar{q}}-\alpha_q)\left[(k - k^+)+(\zeta_1-\zeta_1^+)(1 - 2\alpha_g)+\zeta_2(3 - 4\alpha_g)\right],\nonumber\\ t_3(\alpha_q , \alpha_{\bar{q } } , \alpha_g ) & = & -120(3\zeta_2-\zeta_2^+)(\alpha_{\bar{q}}-\alpha_q)\alpha_q\alpha_{\bar{q}}\alpha_g,\nonumber\\ t_4(\alpha_q , \alpha_{\bar{q } } , \alpha_g ) & = & 30\alpha_g^2(\alpha_{\bar{q}}-\alpha_q)\left[(k+k^+)+(\zeta_1+\zeta_1^+)(1 - 2\alpha_g)+\zeta_2(3 - 4\alpha_g)\right].\nonumber\\\end{aligned}\ ] ]
the functions @xmath139}$ ] and @xmath140}$ ] used in the text are defined as @xmath141}(\bar{u}_0)&\equiv&\int_0^{\bar{u}_0}\cdots \int_0^{x_3}\int_0^{x_2}\mathcal { f}(x_1)dx_1dx_2\cdots dx_n\,,\nonumber \\ \mathcal { f}^{[0,0]}(u_0)&\equiv&\int_0^{u_0}\int_{u_0-\alpha_2}^{1-\alpha_2}\frac{\mathcal { f}(1-\alpha_2-\alpha_3,\alpha_2,\alpha_3)}{\alpha_3}d\alpha_3d\alpha_2\,,\nonumber \\ \mathcal { f}^{[0,1]}(u_0)&\equiv&\int_0^{u_0}\int_{u_0-\alpha_2}^{1-\alpha_2}\frac{(u_0-\alpha_2)\mathcal { f}(1-\alpha_2-\alpha_3,\alpha_2,\alpha_3)}{\alpha_3 ^ 2}d\alpha_3d\alpha_2\,,\nonumber \\ \mathcal { f}^{[1,0]}(u_0)&\equiv&\int_0^{u_0}\frac{\mathcal { f}(1-u_0,\alpha_2,u_0-\alpha_2)}{u_0-\alpha_2}d\alpha_2 -\int_0^{1-u_0}\frac{\mathcal { f}(1-u_0-\alpha_3,u_0,\alpha_3)}{\alpha_3}d\alpha_3\,,\nonumber \\ \mathcal { f}^{[1,1]}(u_0 ) & \equiv&\int_0^{u_0}\frac{\mathcal { f}(1-u_0,\alpha_2,u_0-\alpha_2)}{u_0-\alpha_2}d\alpha_2 -\int_0^{u_0}\int_{u_0-\alpha_2}^{1-\alpha_2}\frac{\mathcal { f}(1-\alpha_2-\alpha_3,\alpha_2,\alpha_3)}{\alpha_3 ^ 2}d\alpha_3d\alpha_2\,,\nonumber \\ \mathcal { f}^{[2,0]}(u_0 ) & \equiv&\frac{\mathcal { f}(0,u_0,1-u_0)}{1-u_0}+\frac{\mathcal { f}(1-u_0-\alpha_3,u_0,\alpha_3)}{\alpha_3}\biggl\lvert_{\alpha_3=0}\nonumber\\ & & + \int_0^{u_0}d\alpha_2\frac{\partial[\mathcal { f}(1-\alpha_2-\alpha_3,\alpha_2,\alpha_3)/\alpha_3]}{\partial\alpha_3}\biggl\lvert_{\alpha_3=u_0-\alpha_2}\nonumber\\ & & -\int_0^{1-u_0}d\alpha_3\frac{\partial[\mathcal { f}(1-\alpha_2-\alpha_3,\alpha_2,\alpha_3)]/\partial\alpha_2}{\alpha_3}\biggl\lvert_{\alpha_2=u_0}\,,\nonumber \\ \mathcal { f}^{[2,1]}(u_0 ) & \equiv&\frac{\mathcal { f}(1-u_0-\alpha_3,u_0,\alpha_3)}{\alpha_3}\biggl\lvert_{\alpha_3=0 } + \int_0^{u_0}d\alpha_2\frac{\partial[\mathcal { f}(1-\alpha_2-\alpha_3,\alpha_2,\alpha_3)/\alpha_3]}{\partial\alpha_3}\biggl\lvert_{\alpha_3=u_0-\alpha_2}\nonumber\\ & & + \int_0^{u_0}d\alpha_2\frac{\mathcal { f}(1-u_0,\alpha_2,u_0-\alpha_2)}{(u_0-\alpha_2)^2 } -\int_0^{1-u_0}d\alpha_3\frac{\mathcal { f}(1-u_0-\alpha_3,u_0,\alpha_3)}{\alpha_3 ^ 2}\,,\nonumber \\ \mathcal { f}^{[3,0]}(u_0 ) & \equiv&\frac{\partial[\mathcal { f}(1-\alpha_2-\alpha_3,\alpha_2,\alpha_3)/\partial\alpha_2]}{\alpha_3}\biggl\lvert_{\alpha_2=u_0,\alpha_3=0 } + \frac{\partial[\mathcal { f}(1-\alpha_2-\alpha_3,\alpha_2,\alpha_3)/\alpha_3]}{\partial\alpha_3}\biggl\lvert_{\alpha_2=u_0,\alpha_3=0}\nonumber\\ & & + \frac{\partial[\mathcal { f}(0,1-\alpha_3,\alpha_3)/\alpha_3]}{\partial\alpha_3}\biggl\lvert_{\alpha_3=1-u_0 } + \frac{\partial[\mathcal { f}(u_0-\alpha_2,\alpha_2,1-u_0)/(1-u_0)]}{\partial\alpha_2}\biggl\lvert_{\alpha_2=u_0}\nonumber\\ & & + \int_0^{u_0}d\alpha_2\frac{\partial^2[\mathcal { f}(1-\alpha_2-\alpha_3,\alpha_2,\alpha_3)/\alpha_3]}{\partial\alpha_3 ^ 2}\biggl\lvert_{\alpha_3=u_0-\alpha_2}\nonumber\\ & & -\int_0^{1-u_0}d\alpha_3\frac{\partial^2[\mathcal { f}(1-\alpha_2-\alpha_3,\alpha_2,\alpha_3)]/\partial\alpha_2 ^ 2}{\alpha_3}\biggl\lvert_{\alpha_2=u_0}\,,\nonumber
\\ \mathcal { f}^{[3,1]}(u_0 ) & \equiv&\frac{\partial[\mathcal { f}(1-\alpha_2-\alpha_3,\alpha_2,\alpha_3)/\partial\alpha_2]}{\alpha_3}\biggl\lvert_{\alpha_2=u_0,\alpha_3=0 } + \frac{\partial[\mathcal { f}(1-\alpha_2-\alpha_3,\alpha_2,\alpha_3)/\alpha_3]}{\partial\alpha_3}\biggl\lvert_{\alpha_2=u_0,\alpha_3=0}\nonumber\\ & & + \frac{\mathcal { f}(1-\alpha_2-\alpha_3,\alpha_2,\alpha_3)}{\alpha_3 ^ 2}\biggl\lvert_{\alpha_2=u_0,\alpha_3=0 } + \frac{\mathcal { f}(1-\alpha_2-\alpha_3,\alpha_2,\alpha_3)}{\alpha_3 ^ 2}\biggl\lvert_{\alpha_2=u_0,\alpha_3=1-u_0}\nonumber\\ & & + \int_0^{u_0}d\alpha_2\frac{\partial^2[\mathcal { f}(1-\alpha_2-\alpha_3,\alpha_2,\alpha_3)/\alpha_3]}{\partial\alpha_3 ^ 2}\biggl\lvert_{\alpha_3=u_0-\alpha_2}\nonumber\\ & & + \int_0^{u_0}d\alpha_2\frac{\partial[\mathcal { f}(1-\alpha_2-\alpha_3,\alpha_2,\alpha_3)/\alpha_3 ^ 2]}{\partial\alpha_3}\biggl\lvert_{\alpha_3=u_0-\alpha_2}\nonumber\\ & & -\int_0^{1-u_0}d\alpha_3\frac{\partial\mathcal { f}(1-\alpha_2-\alpha_3,\alpha_2,\alpha_3)/\partial\alpha_2}{\alpha_3 ^ 2}\biggl\lvert_{\alpha_2=u_0}\,,\nonumber
\\ \mathcal { f}^{[-1,0]}(u_0 ) & \equiv&\int_0^{u_0}\int_0^{u_0-\alpha_2}\mathcal { f}(1-\alpha_2-\alpha_3,\alpha_2,\alpha_3)d\alpha_3d\alpha_2\nonumber\\ & & + \int_0^{u_0}\int_{u_0-\alpha_2}^{1-\alpha_2}\frac{(u_0-\alpha_2)\mathcal { f}(1-\alpha_2-\alpha_3,\alpha_2,\alpha_3)}{\alpha_3}d\alpha_3d\alpha_2\,,\nonumber \\ \mathcal { f}^{[-1,1]}(u_0 ) & \equiv&\frac{1}{2}\int_0^{u_0}\int_0^{u_0-\alpha_2}\mathcal { f}(1-\alpha_2-\alpha_3,\alpha_2,\alpha_3)d\alpha_3d\alpha_2\nonumber\\ & & + \frac{1}{2}\int_0^{u_0}\int_{u_0-\alpha_2}^{1-\alpha_2}\frac{(u_0-\alpha_2)^2\mathcal { f}(1-\alpha_2-\alpha_3,\alpha_2,\alpha_3)}{\alpha_3 ^ 2}d\alpha_3d\alpha_2\,,\nonumber \\ \mathcal { f}^{[-2,0]}(u_0 ) & \equiv&\int_0^{u_0}\int_0^{u_0-\alpha_2}\int_0^{\alpha_3}\mathcal { f}(1-\alpha_2-x,\alpha_2,x)dxd\alpha_3d\alpha_2\nonumber\\ & & + \frac{1}{2}\int_0^{u_0}\int_0^{u_0-\alpha_2}\alpha_3\mathcal { f}(1-\alpha_2-\alpha_3,\alpha_2,\alpha_3)d\alpha_3d\alpha_2\nonumber\\ & & + \frac{1}{2}\int_0^{u_0}\int_{u_0-\alpha_2}^{1-\alpha_2}\frac{(u_0-\alpha_2)^2}{\alpha_3}\mathcal { f}(1-\alpha_2-\alpha_3,\alpha_2,\alpha_3)d\alpha_3d\alpha_2\,,\nonumber \\ \mathcal { f}^{[-3,0]}(u_0 ) & \equiv&\int_0^{u_0}\int_0^{u_0-\alpha_2}\int_0^{\alpha_3}\int_0^{x_2}\mathcal { f}(1-\alpha_2-x_1,\alpha_2,x_1)dx_1dx_2d\alpha_3d\alpha_2\nonumber\\ & & + \frac{1}{2}\int_0^{u_0}\int_0^{u_0-\alpha_2}\int_0^{\alpha_3}x\mathcal { f}(1-\alpha_2-x,\alpha_2,x)dxd\alpha_3d\alpha_2\nonumber\\ & & + \frac{1}{6}\int_0^{u_0}\int_0^{u_0-\alpha_2}\alpha_3 ^ 2\mathcal { f}(1-\alpha_2-\alpha_3,\alpha_2,\alpha_3)d\alpha_3d\alpha_2\nonumber\\ & & + \frac{1}{6}\int_0^{u_0}\int_{u_0-\alpha_2}^{1-\alpha_2}\frac{(u_0-\alpha_2)^3}{\alpha_3}\mathcal { f}(1-\alpha_2-\alpha_3,\alpha_2,\alpha_3)d\alpha_3d\alpha_2\,.\end{aligned}\ ] ]
the single - pion decay widths can be expressed in terms of the 3-momentum @xmath142 of the final pion : @xmath143 | we calculate the @xmath0 , @xmath1 , @xmath2 , and @xmath3 coupling constants between the heavy meson doublets @xmath4 and @xmath5/@xmath6 within the framework of the light - cone qcd sum rule at the leading order of heavy quark effective theory .
most of the sum rules are stable with the variations of the borel parameter and the continuum threshold .
then we calculate the strong and electromagnetic decay widths of the @xmath4 @xmath7-wave heavy mesons .
their total widths are around several tens of mev , which is helpful in the future experimental search . |
alloying elements with complementary qualities in order to tailor their physical behavior for specific technological purposes has been a major route in material development since the antiquity .
cluster research is no exception to this trend .
after decades of systematic studies of the size and structural dependence of the most wide variety of properties of monoelement particles , the interest has actually been moving progressively over the past years towards investigations on finite - size binary alloys.@xcite the magnetism of transition - metal ( tm ) clusters opens numerous possibilities and challenges in this context .
for example , one would like to understand how to modify the magnetic characteristics of clusters , in particular the saturation magnetization and the magnetic anisotropy energy ( mae ) , as it has been done in solids .
this would indeed allow one to design new nanostructured materials from a microscopic perspective .
nevertheless , it also true that controlling composition , system size , and magnetic behavior sets serious difficulties for both experiment and theory .
pure tm clusters such as fe@xmath9 , co@xmath9 and ni@xmath9 show spin moments , orbital moments , and maes that are enhanced with respect to the corresponding periodic solids.@xcite still , the possibilities of optimizing the cluster magnetic behavior by simply tuning the system size have been rather disappointing , particularly concerning the mae , which remains relatively small despite being orders of magnitude larger than in solids@xcite due to the rather weak spin - orbit ( so ) coupling in the @xmath10 atoms .
this is one of the motivations for alloying @xmath10 tms with @xmath11 and @xmath12 elements which , being heavier , are subject to stronger so interactions . in this context
it is useful to recall that large nanoparticles and three dimensional solids of these elements are non - magnetic .
however , at very small sizes the @xmath11 and @xmath12 clusters often develop a finite spontaneous low - temperature magnetization , due to the reduction of local coordination and the resulting @xmath7-band narrowing.@xcite the first experimental observation of this important finite - size effect has been made by cox _
et al . _ by performing stern - gerlach - deflection measurements on rh@xmath9 clusters . in this work
the average magnetic moments per atom @xmath5 = @xmath13@xmath14 have been experimentally determined for @xmath15@xmath16 atoms.@xcite in view of these contrasting features one expects that @xmath10-@xmath11 and @xmath10-@xmath12 alloy clusters should show very interesting structural , electronic and magnetic behaviors . the purpose of this paper is to investigate the ground - state properties of the small ferh clusters in the framework of hohenberg - kohn - sham s density functional theory.@xcite besides the general interest of the problem from the perspective of @xmath10-@xmath11 nanomagnetism , these clusters are particularly appealing because of the remarkable phase diagram of ferh bulk alloys.@xcite in the case of @xmath17 the magnetic order at normal pressure and low temperatures is antiferromagnetic ( af ) .
as the temperature increases this @xmath18 phase undergoes a first order transition to a ferromagnetic ( fm ) state , the @xmath19 phase , which is accompanied by a change in lattice parameter .
the corresponding transition temperature @xmath20 increases rapidly with increasing external pressure p , eventually displacing the fm @xmath19 phase completely for @xmath21 gpa ( @xmath22 for @xmath17 at normal pressure ) .
moreover , @xmath20 decreases very rapidly with decreasing rh content . at low pressures the fm @xmath19 phase undergoes a fm to paramagnetic ( pm ) transition at ( @xmath23).@xcite in addition ,
the properties of @xmath24-ferh bulk alloys have been the subject of first principles and model theoretical investigations.@xcite in particular these show that the relative stability of the fm and af solutions depends strongly on the interatomic distances .
such remarkable condensed - matter effects enhance the appeal of small ferh particles as specific example of @xmath10-@xmath11 nanoscale alloy .
investigations of their magnetic properties as a function of size , composition , and structure are therefore of fundamental importance .
the remainder of the paper is organized as follows . in sec .
[ sec : teo ] the main details of the theoretical background and computational procedure are presented .
this includes in particular a description of the strategy used for exploring the cluster energy landscape as a function of geometrical conformation and chemical order .
the results of our calculations for ferh clusters having @xmath8 atoms are reported in secs .
[ sec : stmagn ] and [ sec : comp ] .
first , we focus on the interplay between structure , chemical order and magnetism in the most stable geometries for different cluster sizes .
second , we analyze the concentration dependence of the cohesive energy , the local and average magnetic moments , and the spin - polarized electronic structure . finally , we conclude in sec .
[ sec : concl ] with a summary of the main trends and an outlook to future extensions .
the calculations reported in this work have been performed in the framework of hohenberg - kohn - sham s density functional theory,@xcite as implemented in the vienna ab initio simulation package ( vasp).@xcite the exchange and correlation energy is described by using both the spin - polarized local density approximation ( lda ) and perdew and wang s generalized - gradient approximation ( gga).@xcite the vasp solves the spin - polarized kohn - sham equations in an augmented plane - wave basis set , taking into account the core electrons within the projector augmented wave ( paw ) method.@xcite this is an efficient frozen - core all - electron approach which allows to incorporate the proper nodes of the kohn - sham orbitals in the core region and the resulting effects on the electronic structure , total energy and interatomic forces .
the @xmath25 and @xmath10 orbitals of fe , and the @xmath26 and @xmath11 orbitals of rh are treated as valence states .
the wave functions are expanded in a plane wave basis set with the kinetic energy cut - off @xmath27 ev . in order to improve the convergence of the solution of the selfconsistent ks equations the discrete energy levels are broadened by using a gaussian smearing @xmath28 ev .
the validity of the present choice of computational parameters has been verified.@xcite the paw sphere radii for fe and rh are @xmath29 and @xmath30 , respectively .
a simple cubic supercell is considered with the usual periodic boundary conditions .
the linear size of the cell is @xmath31 = @xmath32@xmath33 , so that any pair of images of the clusters are well separated and the interaction between them is negligible . since we are interested in finite systems , the reciprocal space summations are restricted to the @xmath34 point .
although the potential advantages of alloying magnetic @xmath10 elements with highly - polarizable @xmath11 or @xmath12 elements can be grasped straightforwardly , the problem involves a number of serious practical challenges . different growth or synthesis conditions can lead to different chemical orders , which can be governed not just by energetic reasons but by kinetic processes as well .
for instance , one may have to deal with segregated clusters having either a @xmath11 core and a @xmath10 outer shell or vice versa .
post - synthesis manipulations can induce different degrees of intermixing , including for example surface diffusion or disordered alloys .
moreover , the inter atomic distances are also expected to depend strongly on size and composition .
typical tm - cluster bond - lengths are in fact @xmath32@xmath35% smaller than in the corresponding bulk crystals . taking into account that itinerant @xmath36-electron magnetism is most sensitive to the local and chemical environments of the atoms,@xcite it is clear that controlling the distribution of the elements within the cluster is crucial for understanding magnetic nanoalloys .
systematic theoretical studies of binary - metal clusters are hindered by the diversity of geometrical conformations , ordered and disorder arrangements , as well as segregation tendencies that have to be taken into account .
this poses a serious challenge to both first - principles and model approaches . in order to determine the interplay between cluster structure , chemical order and magnetism in ferh clusters we have performed a comprehensive set of electronic calculations for clusters having @xmath37 atoms . in the present paper we focus on the most stable cluster structure and magnetic configuration , which are determined by exploring the ground - state energy landscape.@xcite
this is a formidable task , since one needs to consider a large , most possibly complete and unbiased set of initial structures .
such a thorough geometry optimization must include not only the representative cluster geometries or topologies , but also all relevant chemical orders .
this requires taking into account all distributions of the fe and rh atoms for any given size and composition .
these two aspects of the problem of determining the structure of nanoalloys are discussed in more detail in the following .
the different cluster topologies are sampled by generating all possible graphs for @xmath38 atoms as described in ref .
( see also ref . ) . for each graph or adjacency matrix
it is important to verify that it can be represented by a true structure in @xmath39 dimensions .
a graph is acceptable as a cluster structure , only if a set of atomic coordinates @xmath40 with @xmath41 exists , such that the interatomic distances @xmath42 satisfy the conditions @xmath43 if the sites @xmath44 and @xmath45 are connected in the graph ( i.e. , if the adjacency matrix element @xmath46 ) and @xmath47 otherwise ( i.e. , if @xmath48 ) . here
@xmath49 refers to the nearest neighbor ( nn ) distance , which at this stage can be regarded as the unit of length , assuming for simplicity that it is the same for all clusters .
notice that for @xmath50 all graphs are possible cluster structures .
for example , for @xmath51 , the different structures are the tetrahedron , rhombus , square , star , triangular racket and linear chain.@xcite however , for @xmath52 there are graphs , i.e. , topologies , which can not be realized in practice .
for instance , it is not possible to have five atoms being nns from each other in a three dimensional space .
consequently , for @xmath52 there are less real structures than mathematical graphs .
the total number of graphs ( structures ) is 21 ( 20 ) , 112 ( 104 ) , and 853 ( 647 ) for @xmath53 , and @xmath54 , respectively.@xcite for clusters having @xmath55 atoms all these topologies have indeed been taken as starting points of our structural relaxations . out of this large number of different initial configurations the unconstrained relaxations using vasp lead to only a few geometries , which can be regarded as stable or metastable isomers . for larger clusters ( @xmath3 and
@xmath4 ) we do not aim at performing a full global optimization .
our purpose here is to explore the interplay between magnetism and chemical order as a function of composition for a few topologies that are representative of open and close - packed structures .
taking into account our results for smaller sizes , and the available information on the structure of pure fe@xmath9 and rh@xmath9 clusters , we have restricted the set of starting topologies for the unconstrained relaxation of ferh heptamers and octamers to the following : bicapped trigonal bipyramid , capped octahedra , and pentagonal bipyramid for @xmath3 , and tricapped trigonal bipyramid , bicapped octahedra , capped pentagonal bipyramid and cube for @xmath56 .
although , the choice of topologies for @xmath3 and @xmath4 is quite restricted , it includes compact as well as more open structures .
therefore , it is expected to shed light on the dependence of the magnetic properties on the chemical order and composition .
the dependence on concentration is investigated systematically for each topology of fe@xmath57rh@xmath58 by varying @xmath59 and for each size @xmath1 , including the pure fe@xmath9 and rh@xmath9 limits .
moreover , we take into account all possible non - equivalent distributions of the @xmath59 fe and @xmath60 rh atoms within the cluster . in this way ,
any _ a priori _ assumption on the chemical order is avoided .
obviously , such an exhaustive combinatorial search increasingly complicates the computational task as we increase the cluster size , and as we move away from pure clusters towards alloys with equal concentrations . finally , in order to perform the actual density - functional calculations we set for simplicity all nn distances in the starting cluster geometry equal to the fe bulk value@xcite @xmath61 .
subsequently , a fully unconstrained geometry optimization is performed from first principles by using the vasp.@xcite the atomic positions are fully relaxed by means of conjugate gradient or quasi - newtonian methods , without imposing any symmetry constraints , until all the force components are smaller than the threshold @xmath62 mev / .
the convergence criteria are set to @xmath63 ev / for the energy gradient , and @xmath64 for the atomic displacements.@xcite the same procedure applies to all considered clusters regardless of composition , chemical order , or total magnetic moment . notice that the diversity of geometrical structures and atomic arrangements often yields many local minima on the ground - state energy surface , which complicates significantly the location of the lowest - energy configuration .
lattice structure and magnetic behavior are intimately related in tms , particularly in weak ferromagnets such as fe and its alloys.@xcite on the one side , the optimum structure and chemical order depend on the actual magnetic state of the cluster as given by the average magnetic moment per atom @xmath5 and the magnetic order . on the other side ,
the magnetic behavior is known to be different for different structures and concentrations .
therefore , in order to rigorously determine the ground - state magnetic properties of ferh clusters , we have varied systematically the value of the total spin polarization of the cluster @xmath65 by performing fixed spin - moment ( fsm ) calculations in the whole physically relevant range .
let us recall that @xmath66 where @xmath67(@xmath68 ) represents the number of electrons in the majority ( minority ) states .
in practice we start from the non - magnetic state ( @xmath69 ) and increase @xmath65 until the local spin moments are fully saturated , i.e. , until the fe moments in the paw sphere reach @xmath70 and the rh moments @xmath71 ( typically , @xmath72 ) .
the above described global geometry optimizations are performed independently for each value of @xmath65 .
these fsm study provides a wealth of information on the isomerization energies , the spin - excitation energies , and their interplay .
these are particularly interesting for a subtle magnetic alloy such as ferh , and would therefore deserve to be analyzed in some more detail . in the present paper we shall focus on the ground - state properties by determining for each considered fe@xmath57rh@xmath58 the most stable structural and magnetic configuration corresponding to energy minimum as a function of @xmath65 and of the atomic positions.@xcite
once the optimization with respect to structural and magnetic degrees of freedom is achieved , we derive the binding energy per atom @xmath73/ n$ ] in the usual way by referring the total energy @xmath74 to the corresponding energy of @xmath59 fe and @xmath60 rh isolated atoms .
moreover , for each stationary point of the total energy surface ( i.e. , for each relaxed structure having a nearly vanishing @xmath75 ) we determine the vibrational frequencies from the diagonalization of the dynamical matrix .
the latter is calculated from finite differences of the analytic gradients of the total energy . in this way we can rule out saddle points to which the local optimization procedure happens to converge on some occasions .
only configurations which correspond to true minima are discussed in the following . finally , a number of electronic and magnetic properties for example , the magnetic energy @xmath76 , the local magnetic moments @xmath77 integrated within the wigner - seitz ( ws ) or bader atomic cells of atom @xmath44,@xcite and the spin polarized density of electronic states ( dos ) @xmath78 are derived from the self - consistent spin - polarized density and kohn - sham spectrum .
[ sec:2 ] in this section we discuss the ground - state structure , chemical order , binding energy , and magnetic moments of fe@xmath57rh@xmath58 clusters having @xmath79 atoms .
the main emphasis is here on understanding how the various electronic , structural and magnetic properties depend on the chemical composition of the alloy .
first , each cluster size @xmath80 is analyzed separately , since a strong dependence on @xmath80 is expected in the small size , non - scalable regime .
comparisons between the various @xmath80 are stressed by means of cross - references between different subsections .
in addition the main trends as a function of size and concentration are summarized in sec .
[ sec : comp ] . despite being the simplest possible systems , dimers allow to infer very useful trends on the relative strength , charge transfers and magnetic order in the various types of bonds which are found in ferh alloy clusters .
the results summarized in table [ tab:2 ] show that the ferh bond yields the highest cohesive energy , followed by the rh@xmath81 bond , the fe@xmath81 bond being the weakest .
the particular strength of the heterogeneous bond is confirmed by the fact that the corresponding vibrational frequency is the highest .
the bond length , however , follows the trend of the atomic radius which , being larger for rh , gives @xmath82 .
quantitatively , the binding energy per atom @xmath83 ev obtained for fe@xmath81 within the gga is smaller than the lda result @xmath84 ev @xcite although it still remains larger than the experimental value @xmath85 ev reported in ref . .
the calculated vibrational frequency @xmath86 is consistent with previous experimental results [ @xmath87 from ref . and
@xmath88 from ref . ] . our result
for @xmath89 and @xmath5 of rh@xmath81 coincide with previous gga calculations by b. v. reddy _
et al._@xcite these are however larger than the experimental values @xmath90 ev derived from knudsen effusion,@xcite @xmath91 ev derived from resonance raman in ar matrices@xcite and @xmath92 ev derived from the resonant two - photon ionization.@xcite the calculated vibrational frequency of @xmath93 should be compared with the experimental value @xmath94 reported in ref . . .(color
online)[tab:2 ] structural , electronic and magnetic properties of ferh dimers .
results are given for the binding energy @xmath89 ( in ev ) , the magnetic stabilization energy @xmath95 = @xmath96 ( in ev ) , the average interatomic distance @xmath97 ( in ) between atoms @xmath24 and @xmath98 ( @xmath99 or rh ) , the average spin moment per atom @xmath100 ( in @xmath6 ) , the local spin moment @xmath101 ( in @xmath6 ) at the fe or rh atoms , and the vibrational frequency @xmath102 ( in @xmath103 ) . [ cols="^,^,^,^,^,^,^,^,^ " , ] in table [ tab:8 ] results for ferh octamers are reported .
the general trends concerning the composition dependence of the binding energy , chemical order , as well as the average and local magnetic moments are very similar to smaller clusters .
the most stable structure that we obtain for @xmath104 is a bco having @xmath105 ev , an average magnetic moment @xmath106 , and a relatively short average bond - length @xmath107 ( see table [ tab:8 ] ) .
a similar structure is also found in previous spin - polarized lda calculations , where @xmath108 ev and @xmath106 were obtained.@xcite we have repeated these calculation for the bco structure with our computational parameters and atomic reference energies and found @xmath109 ev . the discrepancies between lda and gga results reflect the importance of exchange and correlation to the binding energy . in the other extreme , for pure rh@xmath110
, the structure that we find with the considered starting topologies is a regular cube having @xmath111 ev , an average magnetic moment @xmath112 , and all bond lengths equal to @xmath113 .
these results are in good agreement with previous calculations by bae _
et al._.@xcite it is interesting to observe that the substitution of a single rh atom by fe in ferh@xmath114 results in a compact topology , which is more stable than the relatively open ( relaxed ) cube - like structures derived from pure rh@xmath110 .
the same trend holds for higher fe content ( i.e. , fe@xmath57rh@xmath115 with m @xmath116 ) .
the dominant structure for non - vanishing fe content is a bco with slight distortions . only for fe@xmath117rh@xmath118
we find a different topology , namely , a distorted cpbp .
the typical isomerization energies between the bco and the tctbp are @xmath119@xmath120 mev per atom .
the average magnetic moments in the lowest lying isomers are either the same or very similar .
the magnetic properties of heptamers and octamers follow qualitatively the behavior observed in smaller clusters . in most cases the average magnetic moment per atom @xmath5 and the magnetic energy @xmath121 increase with fe concentration .
the only exception is the pure rh heptamer , for which @xmath122 is somewhat larger than in ferh@xmath123 .
this is not due to af - like coupling between fe impurity moment and the remaining rh atoms but rather to a reduction of the rh local moments in ferh@xmath123 ( @xmath124@xmath125 in rh@xmath114 , while @xmath126@xmath127 in ferh@xmath123 ) .
remarkably , the rh local moments in rh@xmath114 are the largest among all the heptamers
. they amount to 87% of the total moment , which stresses the importance of the local @xmath7-electron contributions .
also in rh@xmath110 one finds quite large local moments , which are actually larger than the rh moments in most fe doped clusters .
this shows that for these sizes the fe atoms do not necessarily increase the rh moments by simple proximity effects ( see tables [ tab:7 ] and [ tab:8 ] ) .
nevertheless , a different behavior is expected for larger @xmath80 , where pure rh clusters are no longer magnetic on their own .
the local fe moments are strongly enhanced with respect to pure fe@xmath9 ( @xmath128 in fe@xmath114 or fe@xmath110 ) reaching values up to @xmath129 , particularly when the fe atoms are in a rh rich environment .
as in the smaller clusters , this is a consequence of a charge transfer from the fe to the rh atoms , which increases the number of polarizable fe @xmath7-holes . notice that some kind of interaction between the fe atoms seems to favor this effect , since the largest @xmath130 are found for clusters having 2 or 3 fe atoms rather than for the single fe impurity .
large fe moments are also found in bulk ferh alloys.@xcite to conclude this section it is interesting to compare the cluster results with available experiments and calculations for macroscopic alloys.@xcite band structure calculations for the periodic fe@xmath131rh@xmath131 alloy having a cscl structure yield an antiferromagnetic ( af ) ground state , which is more stable than the ferromagnetic solution.@xcite this is qualitatively in agreement with experiments showing af order when the rh concentration is above or equal to 50%.@xcite in contrast our results for small clusters show a fm - like order for all rh concentrations , even for the pure rh clusters .
this is a consequence of the reduction of local coordination number and the associated effective @xmath7-band narrowing , which renders the stoner criterion far easier to satisfy , and which tends to stabilize the high - spin states with respect to the low - spin af states .
in fact , even in the bulk calculations on ferh , the energies of the af and fm states are not very different , and a coexistence of both solutions is found over a wide range of volumes.@xcite moreover , experiment shows an af to fm transition with increasing temperature , which is accompanied by an enhanced thermal expansion.@xcite recent _ ab initio _ calculations have revealed the importance of competing fm and af exchange interactions in stoichiometric @xmath24-ferh.@xcite moreover , neutron diffraction experiments@xcite on fe@xmath132rh@xmath133 for @xmath134 and calculations@xcite for @xmath135 show that the fe moments @xmath130 are significantly enhanced with respect to @xmath130 in pure @xmath24-fe , particularly in the fm state where it reaches values of about @xmath136.@xcite these bulk results are remarkably similar to the trends found in fe@xmath57rh@xmath58 clusters over a wide range of compositions .
as in the clusters , the induced rh moments @xmath137 play an important role in the stability of the fm phase .
bulk experiments@xcite on fe@xmath132rh@xmath133 yield @xmath138 for @xmath134 which is comparable to , though somewhat smaller than the present cluster results .
the main purpose of this section is to focus on the dependence of the electronic and magnetic properties of fe@xmath57rh@xmath58 clusters as a function of size and composition . in the following we present and discuss results for the binding energy , average and local spin moments , and electronic densities of states for @xmath139 . in fig .
[ fig : eb ] the binding energy per atom @xmath89 is given as a function of the number of fe atoms @xmath59 .
besides the expected monotonic increase of @xmath89 with increasing @xmath80 , an interesting concentration dependence is observed .
for very small sizes ( @xmath50 ) @xmath89 is maximal for @xmath140 or @xmath141 , despite the fact that @xmath89 is always larger for pure rh than pure fe clusters .
this indicates that in these cases the bonding resulting from ferh pairs is stronger than rhrh bonds . only for @xmath142 ,
when the number of weaker fefe bonds dominates , one observes that @xmath89 decreases with increasing @xmath59 . for larger sizes ( @xmath52 )
the strength of rhrh and ferh bonds becomes very similar , so that the maximum in @xmath89 is replaced by a range of fe concentrations @xmath143 where @xmath89 depends very weakly on @xmath59 . of fe@xmath57rh@xmath58 clusters as a function of the number of fe atoms .
the lines connecting the points for each @xmath144 are a guide to the eye .
, width=302,height=196 ] in fig .
[ fig : moma ] the average magnetic moments @xmath5 of fe@xmath57rh@xmath58 are shown as a function of @xmath59 for @xmath8 .
as already discussed in previous sections , @xmath5 increases monotonously , with the number of fe atoms .
this is an expected consequence of the larger fe local moments and the underlying fm - like magnetic order .
the average slope of the curves tends to increase with decreasing @xmath80 , since the change in concentration per fe substitution is more important the smaller the size is .
the typical increase in @xmath5 per fe substitution is about ( @xmath145)@xmath6 per fe substitution .
notice , moreover , the enhancement of the magnetic moments of the pure clusters in particular for fe@xmath9 ( @xmath146 ) , which go well beyond @xmath147 , the value corresponding to a saturated @xmath7-band in the @xmath148 configuration . in contrast , the moments of pure rh@xmath9 are far from saturated except for @xmath149 and @xmath54 ( see fig . [
fig : moma ] for @xmath150 ) . in this context
it is important to recall that a thorough global optimization , for example , by considering a large number of initial topologies , could affect the quantitative values of the magnetic moments for @xmath3 and @xmath4 . of fe@xmath57rh@xmath58 clusters as a function of number of fe atoms .
the symbols corresponding to each size are the same as in fig .
[ fig : eb ] .
the lines connecting the points for each @xmath144 are a guide to the eye .
, width=302,height=196 ] at the fe and rh atoms as a function of the number fe atoms @xmath59.,width=302,height=196 ] the local magnetic moments in the paw sphere of the fe and rh atoms provide further insight on the interplay between @xmath10 and @xmath11 magnetism in fe@xmath57rh@xmath58 . in fig . [
fig : moml ] @xmath130 and @xmath137 are shown as a function of @xmath59 for @xmath151@xmath4 .
the fe moments are essentially given by the saturated @xmath7-orbital contribution .
for pure fe clusters the actual values of @xmath130 within the paw sphere are somewhat lower than @xmath152 due to a partial spill - off of the spin - polarized density .
notice that the fe moments increase as we replace fe by rh atoms showing some weak oscillations as a function of @xmath59 .
the increase is rather weak for a single rh impurity in fe@xmath153rh but becomes stronger reaching a more or less constant value as soon as the cluster contains 2 or more rh ( @xmath154 , see fig .
[ fig : moml ] ) .
this effect can be traced back to a @xmath7 electron charge transfer from fe to rh which , together with the extremely low coordination number , yields a full polarization of the larger number of fe @xmath7 holes . on the other side the rh moments are not saturated and therefore are more sensitive to size , structure and composition .
the values of @xmath137 are in the range of @xmath155@xmath156 showing some oscillations as a function of @xmath59 .
no systematic enhancement of @xmath137 with increasing fe content is observed .
this behavior could be related to charge transfers effects leading to changes in the number of rh @xmath7 electrons as a function of @xmath59 .
finally , it is interesting to analyze the role played by magnetism in defining the cluster structure by comparing magnetic and non - magnetic calculations . for the smallest ferh clusters ( @xmath157 and @xmath158 ) the magnetic energy @xmath159 gained upon magnetization is higher in the first excited isomer than in the most stable structure .
this implies that the contribution of magnetism to the structural stability is not crucial , since the non - magnetic calculations yield the same ordering , at least concerning the two best structures .
this suggests that for the smallest sizes the kinetic or bonding energy dominates the structural stability , which also explains that the two most stable isomers have different topologies .
the situation changes for large clusters . for @xmath52
one finds a number of ferh clusters for which the optimal structure is actually stabilized by magnetism .
for example , in fe@xmath160rh , fe@xmath118rh@xmath81 , and ferh@xmath160 the energy ordering of the two most stable isomers would be reversed if magnetism were neglected .
it should be noted that in these cases the structures differ only in the chemical order , not in the topology which is a tbp . in the ferh
hexamers the energy differences between the low - lying isomers are more important and only in one case , fe@xmath160rh@xmath81 , magnetism appears to be crucial for stabilizing the actual optimal structure . a similar strong interplay between structure , chemical order and magnetism is expected for larger ferh clusters .
in the previous sections the structure and spin moments of ferh alloy clusters have been discussed as a function of 3@xmath7/4@xmath7 concentration . although these properties are intimately related to the size and composition dependence of electronic structure , it is in general very difficult to achieve a physical transparent correlation between global and microscopic behaviors .
nevertheless , it is very interesting to analyze , at least for some representative examples , how the electronic structure depends on the composition of magnetic nanoalloys .
to this aim we report in fig .
[ fig : dos ] the spin - polarized @xmath7-electron density of states ( dos ) of representative ferh octamers having the relaxed structures illustrated in table [ tab:8 ] .
results for pure fe@xmath110 and rh@xmath110 are also shown for the sake of comparison .
in all the clusters , the dominant peaks in the relevant energy range near @xmath161 correspond either to the fe-@xmath10 or to the rh-@xmath11 states .
the valence spectrum is largely dominated by these @xmath7-electron contributions .
in fact the total dos and the @xmath7-projected dos are difficult to tell apart .
first of all , let us consider the dos of the pure clusters .
our results for rh@xmath110 with a cube structure are similar to those of previous studies.@xcite they show the dominant @xmath7-electron contribution near @xmath161 , with the characteristic ferromagnetic exchange splitting between the minority and majority spin states . in fig .
[ fig : dos ] we also included the dos for rh@xmath110 with a bco structure , since it allows us to illustrate the differences in the electronic structure of compact and open geometries .
moreover , the dos of pure rh@xmath110 with bco structure is very useful in order to demonstrate the dependence of dos on fe content , since the structures of fe@xmath57rh@xmath115 with @xmath162 are similar to the bco .
both fe@xmath110 and rh@xmath110 show relatively narrow @xmath7-bands which dominate the single - particle energy spectrum in the range @xmath163 .
the spin polarization of the dos clearly reflects the ferromagnetic order in the cluster .
putting aside the exchange splitting , the peak structure in the up and down dos @xmath164 are comparable .
there are even qualitative similarities between the two elements . however , looking in more detail , one observes that the effective @xmath7-band width in fe@xmath110 ( about 4 ev ) is somewhat smaller than in rh@xmath110 ( about 5 ev ) . moreover , in rh@xmath110 the dos at @xmath161 is non - vanishing for both spin directions and the finite - size gaps are very small ( see fig . [
fig : dos ] ) .
in contrast , the majority @xmath7-dos is fully occupied in fe@xmath110 , with the highest majority state lying about 0.5 ev below @xmath161 . in addition
there is an appreciable gap ( about 0.1 ev ) in the corresponding minority spectrum .
these qualitative differences are of course consistent with the fact that fe@xmath110 is a strong ferromagnet with saturated moments , while rh@xmath110 should be regarded as a weak unsaturated ferromagnet .
the trends as a function of concentration reflect the crossover between the previous contrasting behaviors . for low fe concentration ( e.g. , fe@xmath81rh@xmath123 )
we still find states with both spin directions close to @xmath161 .
the magnetic moments are not saturated , although the fermi energy tends to approach the top of the majority band .
moreover , the majority - spin states close to @xmath161 have dominantly rh character .
small fe doping does not reduce the @xmath7-band width significantly . notice the rather important change in the shape of the dos in fe@xmath81rh@xmath123 as compared to the dos in rh@xmath110 .
this is a consequence of the change in topology from cubic to bicapped octahedron ( bco ) . for equal concentrations ( fe@xmath160rh@xmath160 ) the first signs of @xmath7-band narrowing and enhanced exchange splitting start to become apparent .
the spin - up states ( majority band ) which in fe@xmath81rh@xmath123 contribute to the dos at @xmath161 now move to lower energies ( 0.3 ev below @xmath161 ) so that the majority band is saturated .
only spin - down ( minority ) states are found around @xmath161 , although there is a significant gap in @xmath165 ( see fig .
[ fig : dos ] ) . in the majority band rh dominates over fe at the higher energies ( closer to @xmath161 ) , while fe dominates in the bottom of the band . in the minority band
the participation of rh ( fe ) is stronger ( weaker ) below @xmath161 and weaker ( stronger ) above @xmath161 .
this is consistent with the fact that the rh local moments are smaller than the fe moments .
finally , in the fe rich limit ( e.g. , fe@xmath123rh@xmath81 ) , the majority - band width becomes as narrow as in fe@xmath110 , while the minority band is still comparable to rh@xmath110 .
the exchange splitting is large , the majority band saturated and only minority states are found close to @xmath161 . as in fe@xmath110 , @xmath165 shows a clear gap at @xmath161 ( see fig .
[ fig : dos ] ) .
however , the rh contribution to the minority states below @xmath161 remains above average despite the relative small rh content .
the fe contribution largely dominates the unoccupied minority - spin dos , in agreement with the larger local fe moments .
the structural , electronic and magnetic properties of small fe@xmath57rh@xmath58 clusters having @xmath166 atoms have been investigated systematically in the framework of a generalized gradient approximation to density - functional theory .
for very small sizes ( @xmath50 atoms ) the binding energy @xmath89 shows a non - monotonous dependence on concentration , which implies that the ferh bonds are stronger than the homogeneous ones .
however , for larger sizes the ferh and rhrh bond strengths become comparable , so that @xmath89 depends weakly on concentration for high rh content .
the magnetic order of the clusters having the most stable structures is found to be fm - like .
moreover , the average magnetic moment per atom @xmath5 increases monotonously , which is almost linear over a wide range of concentration with fe content .
consequently , the energy gain @xmath121 associated to magnetism also increases with the number of fe atoms .
the largest part of the spin polarization ( about 90% ) can be traced back to the local @xmath7 magnetic moments within the paw sphere of the atoms .
the @xmath167 and @xmath168 spin polarizations are almost negligible in general .
a remarkable enhancement of the local fe moments is observed as result of rh doping .
this is a consequence of the increase in the number of fe @xmath7 holes , due to charge transfer from fe to rh , combined with the extremely reduced local coordination .
the rh local moments are important already in the pure clusters ( @xmath8 ) .
therefore , they are not significantly enhanced by fe doping .
however , the overall stability of magnetism , as measured by the total energy gained by the onset of spin polarization , is found to increase with increasing fe content .
a further interesting aspect , particularly for future studies is the role of spin - orbit ( so ) interactions on the magnetism of nanoalloys .
we have performed some representative calculations by taking into account spin - orbit coupling ( soc ) in order to explore their effect on the ground - state structure , chemical order and spin moments .
for example in fe@xmath123 , fe@xmath118rh@xmath118 and rh@xmath123 we find that the changes in the ground - state energy resulting from so interaction are typically of the order of 0.2 ev for the whole cluster .
this is often comparable to or larger than the energy differences between the low - lying isomers .
the so energies are very similar for different structures , so that the ground - state structures remain essentially the same as in the scalar relativistic ( sr ) calculations . the changes in the bond lengths and in the spin moments resulting from soc are also very small ( e.g. , @xmath169 0.01@xmath6 and @xmath170 0.001 in fe@xmath118rh@xmath118 ) . as a result ,
the conclusions drawn from our sr calculations on the relative stability and local spin moments seem to be unaffected by the spin - orbit contributions .
ferh clusters are expected to develop a variety of further interesting behaviors , which still remain to be explored .
for instance , larger cluster should show a more complex dependence of the magnetic order as a function of concentration .
in particular for large rh content one should observe a transition from fm - like to af - like order with increasing cluster size , in agreement with the af phase found in solids for more than 50% rh concentration .
moreover , the metamagnetic transition observed in bulk ferh alloys also puts forward the possibility of similar interesting phenomena in nanoalloys as a function of temperature .
finally , the contributions of orbital magnetism and magnetic anisotropy deserve to be explored in detail as a function of composition and chemical order , even for the smallest sizes , particularly because of their implications for potential applications.@xcite it is pleasure to thank dr .
j. l. ricardo - chvez and dr .
l. daz - snchez for helpful discussions and useful comments .
computer resources from its ( kassel ) and csc ( frankfurt ) are gratefully acknowleged .
j. bansmann , s. baker , c. binns , j. blackman , j .-
buecher , j. dorantes - dvila , v. dupuis , l. favre , d. kechrakos , a. kleibert , k .- h .
meiwes - broer , g. m. pastor , a. perez , o. toulemonde , k.n .
trohidou , j. tuaillon , and y. xie , surf .
sci . rep . * 56 * , 189 ( 2005 ) .
j. p. bucher , d. c. douglass , and l. a. bloomfield phys .
lett . * 66 * , 3052 ( 1991 ) ; d. c. douglass , j. p. bucher , and l. a. bloomfield , phys . rev .
b * 45 * , 6341 ( 1992 ) ; d. c. douglass , a. j. cox , j. p. bucher , and l. a. bloomfield , _ ibid .
_ * 47 * , 12874 ( 1993 ) . a number of tests have been performed in order to assess the numerical accuracy of the calculations .
increasing the cut - off energy e@xmath171 = 268 ev and supercell size @xmath31 = 12 to e@xmath171 = 500 ev and @xmath31 = 22 in rh@xmath160 increases the computation time by a factor @xmath158@xmath54 .
this yields a total energy differences of 1.75 mev and 0.25 mev , respectively . in the above calculations the changes in average bond length ( bond angle )
amounts to 10@xmath172 ( 10@xmath173 degrees ) .
these difference are not significant for our physical conclusions .
in fact , typical isomerization energies in these clusters are an order of magnitude larger , of the order of @xmath32@xmath120 mev .
we also found that the total energy is nearly independent of the choice of the smearing parameter @xmath174 , provided it is not too large ( @xmath174 @xmath175 0.05 ev ) .
values from @xmath174 = 0.01 to 0.1 ev have been checked .
therefore , we judge that our set of standard parameters ( e@xmath171 = 268 ev , supercell size @xmath31 from 10 to 22 , and @xmath174 = 0.02 ev ) offers a sufficiently good accuracy at a reasonable computational costs .
m. muoz - navia , j. dorantes - dvila , d. zitoun , c. amiens , b. chaudret , m .- j .
casanove , p. lecante , n. jaouen , a. rogalev , m. respaud , and g. m. pastor , faraday discuss . * 138 * , 181 ( 2008 ) ; m. muoz - navia , j. dorantes - dvila , m. respaud , and g. m. pastor , eur . j. phys .
d * 52 * , 171 ( 2009 ) .
g. l. gutsev , s. n. khanna , and p. jena , phys .
b * 62 * , 1604 ( 2000 ) ; s. chrtien and d. r. salahub , phys . rev .
b * 66 * , 155425 ( 2002 ) ; g. rollmann and p. entel , computing letters ( cole ) , vol . 1 , no . 4 , 288 - 296 ( 2004 ) .
m. castro , and d. r. salahub , phys .
b * 49 * , 11842 ( 1994 ) ; l. wang , q ge , chem .
. lett . * 366 * , 368 - 376 ( 2002 ) ; c. h. chien , e. blaisten - barojas , and m. r. pederson , phys . rev . a * 58 * , 2196 ( 1998 ) ; j. l. chen , c. s. wang , k. a. jackson , and m. r. pederson , phys .
b * 44 * , 6558 ( 1991 ) .
d. m. cox , d. j. trevor , r. l. whetten , e. a. rohlfing , and a. kaldor , phys .
b * 32 * , 7290 ( 1985 ) .
m. castro , c. jamorski , and d. r. salahub , chem .
* 271 * , 133 ( 1997 ) ; s. e. weber and p. jena , chem .
lett . * 281 * , 401 ( 1997 ) ; b. v. reddy , s. k. nayak , s. n. khanna , b. k. rao , and p. jena , j. phys .
a * 102 * , 1748 ( 1998 ) .
the optimal fe@xmath118rh@xmath81 structure has 6 ferh and 3 fefe nn pairs , while the first isomer has 5 ferh , 3 fefe , and 1 rhrh nn pairs .
the optimal fe@xmath81rh@xmath118 structure has 6 ferh and 3 rhrh nn pairs , while the first isomer has 6 ferh , 2 rhrh , and 1 fefe nn pairs . | the structural , electronic and magnetic properties of small @xmath0 clusters having @xmath1 atoms are studied in the framework of a generalized - gradient approximation to density - functional theory .
the correlation between structure , chemical order , and magnetic behavior is analyzed as a function of size and composition . for @xmath2
a thorough sampling of all cluster topologies has been performed , while for @xmath3 and @xmath4 only a few representative topologies are considered . in all cases
the entire concentration range is systematically investigated .
all the clusters show ferromagnetic - like order in the optimized structures . as a result ,
the average magnetic moment per atom @xmath5 increases monotonously , which is almost linear over a wide range of concentration with fe content .
a remarkable enhancement of the local fe moments beyond 3 @xmath6 is observed as result of rh doping .
this is a consequence of the increase in the number of fe @xmath7 holes , due to charge transfer from fe to rh , combined with the extremely reduced local coordination .
the rh local moments , which are important already in the pure clusters ( @xmath8 ) are not significantly enhanced by fe doping .
however , the overall stability of magnetism , as measured by the energy gained upon spin polarization , increases when rh is replaced by fe .
the composition dependence of the electronic structure and the influence of spin - orbit interactions on the cluster stability are discussed . |
we acknowledge enlightening discussions with c. urbina , d. esteve and j. m. van ruitenbeek . we have enjoyed fruitful interaction with j. c. cuevas and we thank him for providing us with his computer codes .
this work was supported by the deutsche forschungsgemeinschaft through sfb 195 , sfb 513 and the alfried krupp von bohlen und halbach - stiftung .
a. avishai , a. golub , a. d. zaikin , europhys . lett . * 54 * , 640 . n. agrat , a. levy yeyati , and j. m. van ruitenbeek , physics reports * 377 * , 81 ( 2003 ) .
r. landauer , phil .
magazine * 21 * , 863 ( 1970 ) a. i. yanson and j. m. van ruitenbeek , phys .
79 * , 2157 ( 1997 ) .
m. hfner , p. konrad , f. pauly , e. scheer , and j. c. cuevas , submitted to phys .
b ( rapid communication ) a. i. yanson , _ atomic chains and electronic shells : quantum mechanisms for the formation of nanowires _ , phd thesis , university of leiden , the netherlands , 2000 . | we report low - temperature measurements of the electrical transport properties of few - atom contacts of the superconducting metal zinc , arranged with lithographically fabricated mechanically controllable breakjunctions ( mcb ) .
the conductance histogram shows several narrow peaks , not regularly observed for multivalent metals . the first peak - corresponding to the single - atom contact -
is located slightly below one conductance quantum and is split into two subpeaks , possibly indicating two preferred configurations of the single - atom contact .
the current - voltage characteristics in the superconducting state show nonlinearities due to multiple andreev reflections ( mar ) .
in addition we find nonlinearities in the current - voltage characteristics that we attribute to electronic - interaction effects due to the granular structure of the film .
some of these latter findings are in qualitative agreement with a prediction by avishai et al .
@xcite . for revealing and understanding the electronic transport properties of atomic scale circuits , single- or few - atom contacts are used as model systems . in this limit
the electronic conduction can be regarded as a fully quantum - mechanical scattering and wave - matching problem , i.e. the conductance can be attributed to independent electronic modes , nick - named `` conduction channels '' .
an atomic - size contact between two metallic electrodes can accommodate only a small number of conduction channels .
the transport properties are thus fully described by a set @xmath0 of transmission coefficients which depends both on the chemical properties of the atoms forming the contact and on their geometrical arrangement .
experimentally , contacts consisting of even a single atom have been obtained using both scanning - tunnel microscopy and breakjunction techniques @xcite .
the total transmission @xmath1=@xmath2 of a contact is deduced from its conductance @xmath3 measured in the normal state , using the landauer formula @xmath4 where @xmath5 is the conductance quantum @xcite .
experiments on a large ensemble of metallic contacts have demonstrated the _ statistical _ tendency of atomic - size contacts to adopt element - specific preferred values of conductance .
the actual preferred values depend not only on the metal under investigation but also on the experimental conditions .
however , for many metals , and in particular simple ones ( like na , au ... ) which in bulk form good free electron metals , the smallest contacts have a conductance @xmath3 close to @xmath6 ( ref .
@xcite ) . statistical examinations of al point contacts at low temperatures yield preferred values of conductance at @xmath7 and @xmath8 ( ref .
@xcite ) , indicating that single - atom contacts of al have a typical conductance slightly below the conductance quantum .
however , it has been shown by the analysis of current - voltage ( iv ) characteristics in the superconducting state that al single - atom contacts accommodate in general three transport channels , the transmissions of which add up to a value around 1 ( ref .
@xcite ) . these findings are in agreement with measurements of the shot noise @xcite , conductance fluctuations @xcite , thermopower @xcite and the supercurrent @xcite in such contacts .
a quantum chemical model @xcite which links the @xmath9 to the chemical valence and the atomic arrangement of the region around the central atom of the constriction , has been put forward .
this model predicts that monovalent metals would transmit a single channel while sp - like metals as e.g. al or pb up to four channels due to the contributions of the ( unoccupied ) p orbitals .
so far al appears to be the only metal with the property to show peaks in the histogram close to multiples of @xmath6 although this does not correspond to quantized transmission values .
according to simplistic arguments and ignoring the band structure , divalent elements such as the alkaline - earth elements or the iib subgroup elements cd , zn and hg , should be insulating , since they have a completely filled outer s - shell .
this apparent discrepancy to the experimental findings indicates that additional orbitals besides the s - orbitals contribute to the electronic conductance .
tight - binding calculations reveal that in the case of zn the 4p orbitals are the most relevant ones for a correct description of the bulk band structure ( ref .
@xcite ) . with these considerations , single - atom contacts of zn
are expected to have very similar transport properties as al single - atom contacts , i.e. up to four channels with a total conductance of about 1 @xmath6 @xcite .
the distribution of the transmission coefficients , however , may deviate , since it depends on the exact atomic configuration .
since zn crystallizes in a hexagonal , slightly distorted hcp structure while al is a fcc crystal , the actual atomic arrangement of the point contacts might be different .
moreover , due to the `` loss '' of many neighbors , a single zn atom in a contact may behave more atomic - like . here , we report the fabrication and low - temperature electronic transport properties of lithographic mcbs of zn .
this work represents the first study of atomic - size contacts of zn .
the measurements in the normal state demonstrates the tendency of the atomic - size contacts to adopt preferred conductance values .
the analysis of the current - voltage characteristics reveals superconducting properties of the nanowires as well as indications of electron - electron interactions .
we show data of four , nominally identical samples , labeled no .
1 to 4 .
the samples are fabricated along the lines of ref .
@xcite .
our samples are 2@xmath10 m long , 200 nm thick suspended zn nanobridges , with a @xmath11@xmath12@xmath13 constriction in the center ( cf .
fig .
[ amg_f1 ] ) .
the thickness of the bronze substrates is @xmath14 mm , the distance between the counter - supports of the bending mechanism is 10 mm . with these sample parameters and the geometry of the mechanics we achieve reduction ratios ( between the motion of the pushing rod and the displacement of the bridge arms ) of the order of 5000:1 ( see below ) .
electron micrograph of sample no . 1 under an inclination angle of 70@xmath15 , taken after measurement .
the poly - crystalline structure of the zn film is clearly visible . a remaining spacing of the bridge arms of several nm and a fine nanowire between them is observed . ]
special care has to be given to the evaporation of zn .
because of its low boiling temperature and high vapor pressure , zn desorbs easily from the substrate if no crystallization seeds are provided .
if zn is evaporated at room temperature , it grows in islands of micron - size with the hexagonal axis preferably perpendicular to the substrate plane .
since this structure is not suitable for the formation of atomic - size contacts , we take special care to reduce the grain size .
we first deposit a 1 nm thick seed layer of ge onto the substrate which is cooled by liquid nitrogen to a temperature of @xmath16 . at
lower substrate temperatures the different thermal expansions of the metallic substrate , the zn itself , the polyimide sacrificial layer and the polymer lithographic mask provoke cracks in either the mask or the zn layer on warming up to room temperature .
the metal is then evaporated from a home - made boron - nitride crucible heated by a tungsten foil at a pressure of @xmath17 mbar and a rate of @xmath18 nm / s at the same substrate temperature .
the parameters are chosen empirically such that the grain size of the zn layer does not exceed @xmath19 nm and the adhesion is good enough for the subsequent lift - off process . finally , the bridge is underetched in an isotropic oxygen plasma which reduces the height of the polyimide layer by about 300 to 500 nm . at the narrow constriction
the nanobridge is now unsupported .
the resistance of the nanobridge does not increase during the etching process , indicating that reaction between the oxygen ions and zn is weak .
the bridge is then mounted on a three point bending mechanism on a stick which is immersed into liquid helium and can be pumped down to reach temperatures of 1.5 k. the residual resistivity ratio of the bridges is typically rrr = @xmath20 from which we deduce an elastic mean free path of 13 nm . on an unpatterned thin film of the same thickness we measured the critical temperature @xmath21k and a perpendicular critical field of @xmath22mt .
@xmath23 is slightly reduced with respect to the bulk critical temperature of @xmath24 k while @xmath25 is enhanced , as usual in thin films ( @xmath26 mt ) ( ref .
@xcite ) . as a first step
we have determined the preferred conductance values of few - atom contacts when opening or closing continuously and have constructed the corresponding conductance histograms .
a screw with 500@xmath10 m pitch , driven by a dc - motor on top of the cryostat through an ( exchangeable ) reduction gear box , controls the motion of the pushing rod that bends the substrate .
the relative displacement of the pushing rod can be controlled to a precision of @xmath27 m , which due to the geometry of the bending mechanism results in a relative motion of the two anchor points of the bridge of around 0.9pm .
this was verified using the exponential dependence of the conductance on the inter - electrode distance in the tunnel regime .
the absolute precision is restricted due to hysteresis of the bending mechanism ( torsion of the pushing rod etc . ) and roughness of the screw pitch to about 100 nm for the pushing rod ( corresponding to 0.02 nm for the bridge ) .
we note that the stability of the setup allows us to halt at any point along the breaking or closing curves , even on the last plateau when the contact is formed by a single atom , without any observable drift of the measured conductance .
the resistance of the sample is recorded by a 4-point resistance bridge with a measuring current of 1 na and a resolution of @xmath28 in a range from 200 @xmath29 to 200 k@xmath29 . due to the mentioned texture of the evaporated film
there are only few sliding planes parallel to the substrate which would allow for a stretching of the film .
consequently it often happens that the two electrodes remain separated by a gap of several nm after the first breaking .
therefore the first opening of the contact has to be performed very slowly .
typical opening speeds for the systematic measurements are 155 nm / s for the pushing rod , corresponding to a displacement of the bridge anchor points of 36 pm / s .
the output of the resistance bridge is recorded by an oscilloscope and then transferred to a computer for calculating the histogram .
the conductance @xmath3 decreases in steps of the order of 0.2 to 1@xmath6 , their exact sequence changing from opening to opening ( see panel e of fig .
[ amg_f2 ] ) .
since the mechanical deformation is different for opening and closing , we construct separate histograms for both experiments .
fig .
[ amg_f2]a displays the conductance histograms calculated from @xmath30 subsequent opening and closing sequences recorded during two weeks on sample no . 1 which has been kept at low temperatures .
similar histograms were achieved with a lower motion speed of the bridge arms of 17 pm / s .
after an accidental warming up to @xmath31 k no step curves could be observed any more .
a ( b and c ) : conductance histograms for opening and closing the contacts , recorded on sample no . 1 ( 4 ) at @xmath32 k. the histograms have been calculated from @xmath30 individual opening and closing curves , respectively .
d ( e ) examples for typical closing ( opening ) curves . ]
several peaks appear for both motion directions at similar positions of @xmath33 , 1.4@xmath6 and @xmath34 .
interestingly the first peak is split into two substructures at @xmath35 and 0.9 @xmath6 , a feature which has not been observed for other materials .
this splitting has been observed in several independent measurements carried out with two different samples ( nos . 1 and 4 ) .
although the splitting is observed in all our measurements , the absolute values of the peak positions and the relative peak heights vary slightly from experiment to experiment ( see fig . [ amg_f2]b and c ) .
a possible explanation of the splitting would be two different configurations of the central atom e.g. different crystalline orientations or a short contact with only one plane with a single atom in cross - section vs. a dimer contact with two single atoms in series .
the latter configuration has been found in molecular dynamics calculations to be typical for au contacts just before breaking @xcite .
this geometrical difference gives rise to different conductances @xcite .
in similar experiments by yanson et al .
@xcite on `` classical '' breakjunctions made of notched zn wires , no splitting of the first peak is found , but a shoulder that is compatible with our findings .
the peaks at higher conductance values are in reasonable agreement with our findings as well .
it is well known that details of the histograms do depend on the experimental conditions , as one can easily see by comparing different published results of e.g. the most extensively studied metal gold . in the experiment with zn by yanson et al .
higher voltages have been applied , a slightly higher temperature and faster opening speeds have been used and only opening traces were recorded . in our experiment ,
the main difference between opening and closing consists in the peak heights .
the low - conductance peaks are less well pronounced when closing . in the closing traces ( see panel d ) we often observe that the first contact has a conductance well above @xmath6 .
we attribute this to the formation of contacts larger than a single - atom contact since the apex atoms from the previous opening sequence might have relaxed back to positions deeper in the tip .
conductance histograms of multivalent metals in general do not show well pronounced multiple peaks @xcite .
an exception of this general observation was so far al , as mentioned already in the introduction .
the histogram of zn shows similarities to that of al : several peaks occur with an average spacing between the peaks of about 1.3 @xmath6 .
one could argue whether this similarity in the preferred conductance values would be mirrored in the quantum transport properties , i.e. the conduction channels .
therefore we performed measurements in the superconducting state for the determination of the channel ensemble .
for this purpose extremely stable contacts and thus a more complex bending mechanism is required .
the samples are mounted onto a three - point bending mechanism thermally anchored to the mixing chamber of a dilution refrigerator and that contains now a differential screw close to the sample holder .
the pushing rod of the bending mechanism is driven by a dc motor at room temperature via a mechanical feedthrough and several gear boxes with a reduction ratio of 1:2,400,000 between the turning speed of the motor and the axis entering the differential screw .
the differential screw transforms one turn of the axis into a motion of the pushing rod of 100 @xmath36 m .
the sample geometry is the same as in the previous measurements , giving again rice to a reduction ratio of 1:5000 .
taking into account the reduction of the gear boxes , the differential screw and the geometry of the sample itself , one turn of the motor yields a displacement of the bridge arms of the sample of about 8 fm .
therefore , stable tunnel or few - atom wide contacts are achievable , but no histograms can be recorded within a reasonable time . for further experimental details see refs . @xcite and @xcite .
contrary to the experiments described in @xcite , only lossy coaxial wires and twisted - pair wires in capillaries were used for high - frequency filtering .
this restricts the electronic temperature of the samples to about 160 mk .
fig .
[ amg_f3 ] depicts a current - voltage sweep ( iv ) measured at @xmath37mk when sample no .
2 was in the close tunnel regime with a conductance of @xmath38 . in the inset
, the differential conductance is shown .
the tunnel contact was achieved by carefully closing the contact right before the so - called `` jump to contact '' , i.e. the strong conductance increase signalling hte formation of an atomic bond between the two bridge arms @xcite .
apparently , the spectrum deviates from the expected behavior for bcs superconductors , as has already been observed in earlier studies of tunnel contacts of zn fabricated by different methods @xcite .
we deduce a superconducting gap @xmath39v , somewhat higher than the values found in the literature , which vary typically between @xmath40ev and @xmath41ev , depending on the measuring method , crystallographic orientation and other parameters @xcite .
after these characterization measurements , we close the bridge again to form single- or few - atom contacts .
this is done by reducing the bending of the substrate until the sample achieves a conductance of @xmath42 and opening it again carefully . just before break typically conductances of @xmath43
are observed .
current - voltage characteristic ( iv ) of sample no .
2 when in the near tunneling regime at @xmath44 mk .
from the slope at voltages higher than the superconducting gap we deduce a conductance of @xmath45 .
the superconducting gap is @xmath46ev .
inset : numeric derivative of the iv . ] in fig .
[ amg_f4 ] we depict a typical set of ivs recorded on contacts on last plateaus of sample no .
2 before breaking of the wire , measured at temperatures around 160 mk .
the typical nonlinearities due to multiple andreev reflection ( mar ) are observable @xcite . however , they appear rounded compared to the theoretical prediction and the observations for contacts of superconductors with a density of states obtained from @xmath47 vs. @xmath48 that is closer to the bcs shape .
this fact limits the precision with which the channel ensemble can be determined .
the lines shown in fig . [ amg_f4 ] are fits to the theory of mar taking into account 1 , 2 or 4 channels for curves a , b , and c ( all measured on sample no .
2 ) .
because of the rounded mar structures , only the number of channels can be estimated , but no meaningful value for the transmissions can be given for contacts with more than one channel because several channel transmissions combinations give fits with the same quality .
the fact that the determination of the number of channels is more robust than the precision with which the individual transmissions can be deduced has already been reported for few - atom contacts of nb , a metal that also shows a dos deviating from the bcs shape @xcite .
the robustness of the determination of the number of channels is due to the super - linear dependence of the excess current ( the extrapolated intersection of the high - bias current with the current axis ) on the transmission values @xcite .
interestingly , contacts with a dominating single channel , accompanied by a second , but much smaller one , similar to the observations for au @xcite , are achievable .
these ivs are routinely observed at the end of a last plateau after an abrupt change of the iv including an increase of the conductance , and thus an re - arrangement of the atomic - size contact .
once such a contact has been established , it remains stable for an elongation of the bridge of 1 - 2 . for al samples , contacts with a dominating channel with the contribution of one or two smaller channels
could be observed as well @xcite .
however , for al these contacts evolved continuously from a situation with three channels of similar size and occurred only very rarely @xcite .
current - voltage characteristics of few - atom contacts of sample no .
2 . the measuring temperature was @xmath49 mk .
the symbols are experimental data , while the fits are calculated with the theory of mar from refs .
[ ] for @xmath50v and @xmath51 mk .
the transmission coefficients are : curve ( a ) @xmath52 ; ( b ) @xmath53 ; ( c ) @xmath54 ] the origin of these single - channel like contacts and their possible correlation with the observed splitting of the first maximum in the conductance histogram has to be clarified further .
therefore , measurements on contacts , fabricated from whisker breakjunctions have been carried out .
the results of these investigations together with a detailed theoretical analysis will be reported elsewhere @xcite .
we expect the superconducting dos to be closer to the bcs dos for those samples and consequently allowing to determine the transmission values precisely .
another observation hampers the quantitative determination of @xmath55 . as can be seen in curve c of fig .
[ amg_f4 ] , the current onsets corresponding to mar processes involving two or more electrons seem to be shifted to higher voltages with respect to the current onsets predicted by the theory and observed for contacts of other superconductors @xcite .
we attribute this observation to the onset of coulomb blockade @xcite due to the granularity of the film .
our findings are in qualitative agreement with a prediction by avishai et al . @xcite .
the authors have calculated the iv characteristics of an interacting quantum dot between two superconductors and found a decrease of the subgap current and a shift of the higher - order mar processes to higher voltages . however , since no quantitative theory for our experimental realization and resulting transmission coefficients are available , we can not perform a quantitative comparison with this suggestion .
cron et al .
@xcite have placed an al breakjunction in a high - impedance environment in order to study the dynamic coulomb blockade @xcite of a single junction .
they found a reduction of the low - bias conductance in reasonable agreement with a theoretical model when the contact was driven normalconducting by applying a small magnetic field .
however , no effect was detected in the superconducting state . in order to test the interpretation of coulomb blockade ,
we record ivs of tunnel contacts for various magnetic fields , applied perpendicular to the film plane and for temperatures below and above @xmath56 .
fig .
[ amg_f5 ] displays a typical set of ivs , recorded on sample no . 3 for a contact with a conductance at high bias voltages of about @xmath57 .
at the lowest measuring temperature and without magnetic field the conductance is reduced for small voltages @xmath58v . for larger voltages a strong increase is observed .
around @xmath59v the conductance decreases again . at voltages above @xmath60v
the iv is roughly linear .
the small gap in the iv at low voltage is suppressed by applying a magnetic field of only 2 mt , which is smaller than the critical field of the unpatterned film but could correspond to the critical field of the lithographiclly patterned film .
we therefore attribute this gap to superconductivity .
the remaining nonlinearities would than have a different origin .
the conductance at intermediate voltages is progressively suppressed by increasing the field strength .
however , the nonlinearities remain up to fields of @xmath61 t , gradually evolving from an enhanced conductance at zero bias to a slightly reduced one .
current - voltage characteristics of a tunnel contact of sample no .
3 with a conductance of @xmath62 for the magnetic fields and temperatures indicated in the legend .
the magnetic field was applied perpendicular to the film plane . for the curves at @xmath63 and @xmath64 t data for increasing and decreasing voltage
are shown in order to demonstrate the reproducibility .
the nonlinearities observed at low @xmath65 and low @xmath66 vanish only at @xmath67k . ]
a possible explanation of the occurrence of such nonlinear ivs would be a strong enhancement of the critical field of a contact with high transmission values @xmath68 .
a drastically enhanced critical field was observed previously for atomic - size contacts of pb prepared with the help of an stm @xcite in a magnetic field applied parallel to the current direction .
it was attributed to the incomplete penetration of the magnetic field in narrow structures . in our previous measurements on al contacts and au contacts with al
leads fabricated with the same technique and field direction as the present experiments on zn contacts , all signatures of superconductivity vanished at magnetic field values smaller than or equal to the bulk critical field of al . however , in those samples the film structure was optically smooth .
accordingly , the granularity of the zn film could be the reason for the unusually high critical , the nonlinearities only vanish at temperatures @xmath69k , well above the critical temperature , which we estimate to be about 750 mk , since the signatures of mar for atomic - contacts realized with the same sample disappear at this temperature .
differential conductance , calculated numerically from the ivs of fig . 5 .
for clarity , the reproducibility is only shown for the curves recorded at @xmath63 and @xmath70 t . ] therefore and in accordance with the observed shift of the mar peaks we attribute the nonlinearities to electron - interaction effects , such as coulomb blockade , resonant tunneling or quantum interference effects .
the interplay between electron - electron interaction effects and superconductivity in atomic - size structures has so far not been explored theoretically .
consequently , a detailed interpretation of the observed ivs below @xmath56 including the magnetic - field behavior is not possible . taking the typical voltage scale of the nonlinearities of @xmath71v as a measure for the charging energy @xmath72
, we deduce a capacity @xmath73ff corresponding to the capacity of a sphere with a diameter of 7 @xmath36 m .
this value is about a factor 10 larger than the typical grain size of the zn film .
thus , we interpret our observations to indicate the onset of the interaction effect . due to the low - impedance environment of the circuit and the effective screening of the surrounding metal
all interaction effects would be expected to be damped .
the numerical derivatives @xmath74 of the ivs of fig.[amg_f5 ] , depicted in fig .
[ amg_f6 ] show small , but reproducible fluctuations superimposed over the larger scale variations .
the fluctuations have a typical voltage period of @xmath75v , appear at the same voltage values for both sweep directions ( but different for opposite sign of the voltage ) and are correlated from curve to curve .
these findings underline at first that the fluctuations are not due to experimental noise and second that the contact has not been changed by varying temperature and magnetic field .
there are several possible origins of the fluctuations , including ( universal ) conductance fluctuations ( cf ) or again the onset of coulomb blockade due to the granular structure of the film .
the analysis in terms of cf yields a typical coherence length of @xmath76 nm roughly corresponding to the grain size of the evaporated film . a gradual transition from cf to coulomb - blockade behavior
has recently been observed in carbon nanotubes @xcite .
however , the physical mechanism giving rise to this transition is not yet clear . in conclusion ,
a single - atom contact of zn is likely to have a conductance of @xmath77 .
the histograms for both closing and opening the contacts indicate the importance of different contact geometries . in the iv characteristics in the superconducting state we find indications of interaction effects , perhaps due to the granular structure of the evaporated films .
however , due to the lack of systematic measurements and theoretical investigations the interpretation of the features being due to a coulomb - blockade type mechanism remains somewhat speculative . |
our initial samples , which we prepare as described in @xcite , consist of two noninteracting particles in the ground state of our trap . at the end of the preparation
the optical trap depth is @xmath48 ( relative error of @xmath9 : @xmath49 ) and thus there are 4 bound states in the potential . at this trap depth tunneling from the two lowest levels
is completely suppressed . for the measurement of
the interaction - induced tunneling we ramp to a trap depth of @xmath50 with a ramp speed of @xmath51 ms@xmath52 where tunneling occurs on experimentally accessible time scales . in @xcite
we have estimated the probability of exciting particles when performing a single ramp of the optical potential at this speed to be consistent with @xmath53 . to deduce the tunneling time constant we fit the measured particle number with a @xmath54-square minimization algorithm ( levenberg - marquardt ) considering the statistical error of the mean particle number .
+ to tune the strength of the interaction we apply magnetic offset fields ranging from @xmath55 g to @xmath56 g . at two distinct magnetic field values close to the cir ( @xmath57 g and @xmath58 g at @xmath59 ) we observe molecule formation resonances with a width of @xmath60 g ( fwhm ) .
we attribute these resonances to a coupling between the relative motion of the two atoms and the center - of mass motion of a molecular bound state due to the anharmonicity of the trapping potential @xcite .
we prevent formation of molecules by ramping across those resonances with sufficiently high speed ( @xmath61 g / ms ) .
thus we can neglect these resonances for our experiments . + to determine the energy of the bound states of the potential with the wkb ( wentzel - kramers - brillouin ) approximation one has to solve the following implicit equation for the energy e : @xmath62 with @xmath63 a positive integer number and @xmath64 the solutions of @xmath65 .
the tunneling time constant @xmath41 of a particle with energy e is calculated by @xmath66 with the transmission coefficient @xmath67 ) and the knock - frequency @xmath68 . for the energy range studied in the experiment
we observe that only one of the two particles tunnels through the barrier .
this particle has an energy identical to the kinetic energy of the relative motion of the two - particle system .
hence we can determine the kinetic energy of the relative motion of the two - particle system by extracting the energy of the tunneled particle from the tunneling time constant @xmath69 . to map @xmath69 onto energies
we calculate tunneling time constants for a set of energies . by matching these calculated tunneling time constants to the measured ones we determine the kinetic energy of the relative motion of the two - particle systems .
a more detailed description of this method will be given in @xcite .
+ the potential consists of a cigar - shaped cylindrically symmetric optical potential @xmath70 created by a tightly focused laser beam and a linear magnetic potential @xmath71 .
the longitudinal part of the potential is given by where @xmath72 is the initial depth at the center of the optical dipole trap , @xmath9 the optical trap depth as a fraction of the initial depth , @xmath73 the rayleigh range , @xmath74 nm the wavelength of the trapping light , @xmath75 the magnetic field gradient and @xmath13 the magnetic moment of the atoms .
+ the optical potential is created by a tightly focused laser beam with a power of @xmath76w .
we assume that the beam has a gaussian shape , neglecting any aberrations that might be introduced by the optical setup . to deduce the shape of the potential we directly measure the structure of the energy levels in the optical potential by periodically modulating the power or position of the trapping beam . for this
we use noninteracting atoms and thus the observed excitation frequencies correspond to the energy differences between the different trap levels . at an optical trap depth of @xmath77 we find level spacings of , and for the lowest trap states in longitudinal direction where the indices label the energy levels in the trap . to obtain the waist we vary @xmath78 and @xmath79 within its error @xmath80 to minimize @xmath81 ^ 2 $ ] with @xmath82 being the energy of the calculated bound states of the varied optical potential .
the bound state energies are calculated using a wkb calculation .
we find a minimum deviation for @xmath83 and @xmath84w resulting in an initial depth at the center of the optical trap of @xmath85@xmath86k . for these parameters
two of the three frequencies match the calculated bound states within their errors and all three match within @xmath87 .
after fixing the parameters for the optical potential we have to determine the strength of the linear magnetic potential which is created by a magnetic field gradient @xmath75 . from a levitation measurement
we obtain @xmath88g / cm .
to obtain a more accurate parametrization of the potential barrier we use the tunneling measurement of two identical fermions at a trap depth of @xmath50 as a calibration .
we perform a wkb calculation of the tunneling rate and modify the value of @xmath75 in the calculation such that the resulting tunneling time constant agrees with the experimentally observed tunneling time constant of @xmath89ms . from this
we obtain @xmath90g / cm . using this magnetic field gradient
the calculation is consistent with the deterministic preparation @xcite of ( 2 , 4 , 6 , 8 , 10 ) particles at @xmath9=(0.6575 , 0.7025 , 0.7475 , 0.7863 , 0.8200 ) in the experiment . + we measure the perpendicular trap frequency by exciting noninteracting particles in the pure optical potential @xmath91 . in perpendicular direction
we observe two resonances at @xmath92khz and @xmath93khz for @xmath77 which we attribute to a slight ellipticity of the trap .
following @xcite we only consider one cir which we calculate from the mean perpendicular frequency of @xmath94khz .
+ at @xmath50 , where we perform the tunneling measurements , the trap frequency of the perpendicular confinement has to be scaled by @xmath95 and is given by @xmath6khz . at this depth the mean longitudinal trap frequency of the pure optical potential is given by @xmath5khz which is calculated from the excitation frequencies @xmath96 and @xmath97 . | we study a system of two distinguishable fermions in a 1d harmonic potential .
this system has the exceptional property that there is an analytic solution for arbitrary values of the interparticle interaction .
we tune the interaction strength and compare the measured properties of the system to the theoretical prediction . for diverging interaction strength ,
the energy and square modulus of the wave function for two distinguishable particles are the same as for a system of two noninteracting identical fermions .
this is referred to as fermionization .
we have observed this by directly comparing two distinguishable fermions with diverging interaction strength with two identical fermions in the same potential .
we observe good agreement between experiment and theory . by adding more particles our system can be used as a quantum simulator for more complex systems where no theoretical solution is available .
a powerful tool for solving complex quantum systems is to map their properties onto systems with simpler solutions . for interacting bosons in one dimension
there is a one - to - one correspondence of the energy and the square modulus of the wave function @xmath0 to a system of identical fermions @xcite .
as one consequence the local pair correlation @xmath1 of an interacting 1d bose gas vanishes for diverging interaction strength just like in a gas of noninteracting identical fermions .
thus , a large decrease of @xmath1 in a repulsively interacting 1d bose gas is strong evidence for the existence of fermionization @xcite .
+ the many - body properties of such 1d bosonic systems have been studied in @xcite .
however , the essential property of a such a gas namely the fermionization @xcite is already present in a system of two interacting particles , regardless of the particles being identical bosons or distinguishable fermions @xcite .
this two - particle problem is of significant interest because it is the main building block of all 1d quantum systems with short - range interactions .
it is also one of the few quantum mechanical systems for which an analytic solution exists .
in contrast to measurements of bulk properties such as compressibility and collective oscillations or measurements of local pair correlations @xcite , we access the energy and the square modulus of the wave function of the fundamental two - particle system . we directly observe fermionization of two distinguishable fermions by comparing two distinguishable fermions with two identical fermions in the same potential . in optical lattices
the energy of similar two - particle systems has been measured for large but not diverging interaction strength @xcite .
we realize such a two - particle system with tunable interaction using two fermionic @xmath2li atoms in the ground state of a potential created by an optical dipole trap and a magnetic field gradient [ figs.1(a ) and 1(b ) ] .
we can prepare this state with a fidelity of @xmath3 @xcite .
the energy of such two particles interacting via contact interaction which is fully described by one parameter , the 1d coupling strength @xmath4 can be analytically calculated for a harmonically trapped 1d system @xcite .
[ tb ] the problem can be separated into center - of - mass and relative motion because of the harmonic trapping potential and because the interaction term only depends on the relative distance between the two particles .
[ tb ] then the solution can be written as a product of the center - of - mass and the relative wave function . the latter is shown in fig.2(a ) for different values of the coupling strength . for diverging coupling strength
the square modulus of the wave function of a system of two distinguishable fermions is the same as for two noninteracting identical fermions .
this is the point where fermionization occurs .
+ in our setup the particles are confined in a three - dimensional cigar - shaped potential with an aspect ratio of about 1:10 , which can be harmonically approximated with trap frequencies of @xmath5khz in longitudinal direction and @xmath6khz in perpendicular direction .
it has been shown in @xcite that the energy of two interacting particles in the ground state of such a potential is well described by the 1d solution given in @xcite .
hence we treat our system in this 1d framework .
the combined optical and magnetic potential in one - dimensional form reads : @xmath7 where @xmath8k is the initial depth of the optical potential , @xmath9 is the optical trap depth in units of the initial trap depth , @xmath10 is the rayleigh range of the optical trapping beam with minimal waist @xmath11 m and wavelength @xmath12 nm , @xmath13 is the magnetic moment of the atoms and @xmath14 g / cm is the strength of the magnetic field gradient .
the determination of the trap parameters is described in the supplemental material .
+ the 1d coupling constant @xmath4 can be calculated from the 3d scattering length @xmath15 and depends strongly on the confining potential , which is characterized by the harmonic oscillator length @xmath16 @xcite , where @xmath17 is the reduced planck constant and @xmath18 the reduced mass of two @xmath2li atoms with mass @xmath19 .
the coupling constant is given by @xmath20 with @xmath21@xmath22@xmath23=@xmath24 ... and @xmath25 the riemann zeta function .
the value of @xmath4 can be changed by tuning the 3d scattering length via a magnetic feshbach resonance @xcite .
when @xmath15 approaches the extension of the confining harmonic oscillator potential @xmath26 , a confinement - induced resonance ( cir ) occurs for @xmath27 @xcite .
fig.4(b ) shows @xmath28 for two distinguishable atoms in the two lowest @xmath2li hyperfine states , @xmath29f@xmath30,@xmath31=-@xmath32 and @xmath29f@xmath30,@xmath31=@xmath32 labeled @xmath33 and @xmath34 as a function of the magnetic offset field @xcite . for two identical fermions s - wave scattering
is forbidden and thus @xmath35 for all values of the magnetic offset field .
+ to determine the energy of the two - particle system in state @xmath36 we modify the trapping potential such that there is a potential barrier of fixed height through which the particles can tunnel out of the trap ( see fig.1(c ) and supplemental material ) . in the presence of repulsive interactions
the energy of the system is increased according to the blue curve in fig.2(b ) .
this decreases the effective height of the barrier and the particles tunnel faster .
we allow the particles to tunnel out of the trap for different durations and record the number of particles remaining in the trap . by choosing an adequate barrier height
we ensure that the time scale for tunneling is smaller than the lifetime of our samples in the ground state ( about @xmath37s ) . additionally obtaining meaningful tunneling time constants
requires the timescale of the tunneling to be much larger than the inverse longitudinal trap frequencies of @xmath38ms . by averaging over many experimental realizations we obtain the expectation value of the particle number in the potential for different hold times [ fig.(3 ) ] .
[ tb ] by performing this measurement for various values of the coupling strength we can determine the dependence of the system s energy on @xmath28 .
+ we find that for the observed range of interaction energies which are on the order of @xmath39 only one particle leaves the potential even for long hold times . in a simple picture this can be explained as follows : if one particle tunnels through the barrier the interaction energy is released as kinetic energy , which leaves the other particle in the unperturbed ground state of the potential .
this state has a tunneling time scale much larger than the duration of the experiment .
thus we can fit exponentials of the form @xmath40 to the mean particle number to deduce the tunneling time constant @xmath41 for different magnetic fields .
the mean numbers of tunneled ( @xmath42 ) and remaining particles ( @xmath43 ) are expected to be unity . however , due to the finite preparation fidelity they are slightly lower . in fig.(4 ) we show the determined tunneling time constants of a system of two interacting fermions for different interaction energies as a function of the magnetic field .
[ tb ] we observe a decrease in the tunneling time constant over two orders of magnitude for increasing magnetic field due to the gain in interaction energy caused by the cir . + for a direct comparison of the properties of the two interacting distinguishable fermions with those of two identical fermions we perform the same measurement with two fermions in state @xmath44 in the same potential [ fig.1(c ) ] .
the results of these reference measurements are shown in fig.3 and fig.4 ( green points ) .
as the identical fermions are noninteracting we find no dependence of the tunneling time constant on the magnetic field in this measurement .
+ comparing the results of the two systems we find that the tunneling time constant for the interacting system decreases monotonically with increasing magnetic field and crosses the magnetic field independent tunneling time constant of the two identical fermions .
thus there is one magnetic field value where the tunneling time constants of both systems are equal . at this point
both systems must have the same energy .
for a 1d system with given energy there is only one unique solution for the square modulus of the wave function .
therefore , right at the observed crossing point of the tunneling time constants the energy and the square modulus of the wave function @xmath45 of the two interacting distinguishable fermions and the two noninteracting identical fermions must be equal . hence , exactly at this crossing point the system of two distinguishable fermions is fermionized .
as predicted by theory @xcite we find the position of the fermionization at the magnetic field value where @xmath28 diverges due to the confinement - induced resonance . + for magnetic field values below the cir we have realized the two - particle limit of a tonks - girardeau gas @xcite .
above the cir we have created a super - tonks state consisting of two particles .
the super - tonks state is a strongly correlated metastable state above the attractive ground state branch [ see fig.2(b ) ] . in a system with particle numbers @xmath46 inelastic three - body collisions
lead to a fast decay of the metastable super - tonks - girardeau gas @xcite .
in contrast , our two - particle super - tonks state is stable against collisional losses since there is no third particle available to undergo an inelastic three - body event .
+ to determine the energy of the two interacting fermions from the measured tunneling time constants we use a wkb calculation ( see supplemental material ) .
this requires knowledge of the potential shape .
the parameters of the optical potential are determined by precise measurements of the level spacings in the potential .
the final parameter to determine the barrier height is fixed by the measured tunneling time constant of two identical fermions ( see supplemental material ) .
the energies obtained from the tunneling time constants of two distinguishable fermions are shown in fig.5 .
+ [ tb ] we compare these energies to the analytic theory for a harmonic potential @xcite ( see fig.2 ) .
this theory needs two input parameters , the coupling strength and the level spacing . for the coupling strength we use @xmath28 of our system shown in fig.4(b ) . for the level spacing we use the energy difference @xmath47hz between the ground and first excited state of the potential which we calculate using the wkb method . with this approximation
the energy obtained from the tunneling measurements and the energy obtained from the analytic theory @xcite are the same at the cir . for the tonks regime
we find excellent agreement of the experimentally determined energy with the theoretical prediction for a harmonic trap . above the cir
the harmonic theory is not applicable because the second excited state is not bound in our potential .
additionally , we expect deviations for larger energies due to the limited validity of the wkb approximation for energies close to the continuum threshold . a more precise description could be achieved by adapting the theory described in @xcite to our nonharmonic potential and by using a more accurate theory for the tunneling process @xcite . in summary
, we have measured the interaction energy of two distinguishable fermions as a function of the interaction strength and identified the point of fermionization .
the good agreement between our results and theoretical predictions shows that our experiment has the capability to simulate strongly correlated few - body quantum systems . using the experimental methods
established in this work it is straightforward to extend our studies to more complex systems . simply adding a third particle either in one of the present spin states @xcite or a different spin state @xcite allows us to study a highly nontrivial system where no analytical solution exists . in a few - body system with defined particle number and attractive interaction
we could investigate pairing phenomena and thus work towards studying superfluidity in finite systems .
this has already been investigated in the context of nuclear physics @xcite . by dynamically changing the shape of the trapping potential we could simulate a vast amount of different time - dependent quantum systems .
a feasible experiment would be to periodically modulate the strength of the magnetic field gradient .
this would allow us to study ionizationlike excitations in the strong - field regime @xcite which have been studied in ultrafast physics @xcite .
we thank h.j .
pirner for valuable theoretical input .
we thank t. busch , m. rontani and a. saenz for inspiring discussions and j. ullrich and his group for their support .
this work was supported by the imprs - qd ( g.z . and
a.n.w . ) , the helmholtz alliance ha216/emmi , and the heidelberg center for quantum dynamics . 30 natexlab#1#1bibnamefont # 1#1bibfnamefont # 1#1citenamefont # 1#1url # 1`#1`urlprefix[2]#2 [ 2][]#2 , * * , ( ) . , , , * * , ( ) . , , , * * , ( ) . , , , , , , , * * , ( ) . , , ,
* * , ( ) . ,
* * , ( ) . , , , , , * * , ( ) .
, , , , , , * * , ( ) . , , , , , , * * , ( ) . , , , , * * , ( ) .
, , , , * * , ( ) . , * * , ( ) . , * * , ( ) . , , , , , , * * , ( ) . , , , ,
* * , ( ) . , , , * * , ( ) .
, , , , , , , , * * , ( ) . , , , , , , , , , , , * * , ( ) . , * * , ( ) . , * * , ( ) .
, , , * * , ( ) . , , , , ,
* * , ( ) . , * * , ( ) . , * * , ( ) . ,
* * , ( ) . , * * , ( ) . , , , * * , ( ) . , , ,
, * * , ( ) . , ph.d .
thesis , ( ) .
, , , , , * * , ( ) . |
the shapley supercluster ( ssc ) has been investigated by numerous authors since its discovery in 1930 ( quintana et al .
1995 ; hereafter paper i ) .
it lies in the general direction of the dipole anisotropy of the cosmic microwave background ( cmb ) , and is located at 130 beyond the hydra - centaurus supercluster ( @xmath7 away from us ) .
it consists of many clusters and groups of galaxies in the redshift range @xmath8 .
the central cluster a3558 has also been measured with a rosat pspc observation by bardelli et al .
( 1996 ) who derive a total mass of @xmath9 within an abell radius of 2 .
several other x - ray clusters form part of the shapley supercluster ( pierre et al .
1994 ) .
the shapley supercluster is recognised as one of the most massive concentrations of galaxies in the local universe ( scaramella et al.1989 ; raychaudhury 1989 ) , so it is of particular interest to consider its effect on the dynamics of the local group . in paper
i it was estimated that for @xmath10 and @xmath11 the gravitational pull of the supercluster may account for up to 25% of the peculiar velocity of the local group required to explain the dipole anisotropy of the cmb radiation , in which case the mass of the supercluster would be dominated by inter - cluster dark matter .
previous studies of the shapley supercluster ( paper i , quintana et al . 1997
; hereafter paper ii ) have concentrated on the various rich abell galaxy clusters in the region , but this might give a very biased view of the supercluster . as was noted in paper i ,
`` the galaxy distribution inside the supercluster must be confirmed by the detection in redshift space of bridges or clouds of galaxies connecting the different clusters '' .
we are continuing this project , using data from wide - field multi - fibre spectrographs to measure many more galaxy redshifts and get a more complete picture of the composition of the supercluster .
our main aims are first to define the real topology of the ssc : in paper i it was shown that the ssc is significantly flattened , but the real extent of the concentration is not well defined .
secondly we will analyse the individual x - ray clusters that are true members of the shapley supercluster in order to estimate the cluster masses , and investigate suspected sub - structure .
additional observations are planned before we present a full analysis of the dynamics ( proust et al .
1998 in preparation ) . in paper ii we presented data from the mefos spectrograph on the european southern observatory 3.60 m telescope .
this has 30 fibres in a 1 deg diameter field , so the observations were again mainly concentrated on the known clusters , determining for several of them if they were members of the supercluster or not . in this paper
we present new data obtained with the flair - ii ( parker & watson 1995 ; parker 1997 ) multi - fibre spectrograph on the uk schmidt telescope at the anglo - australian observatory .
this has 90 fibres in a @xmath12 deg@xmath1 field and has allowed us to measure a much more uniform distribution of galaxies in the direction of the ssc , avoiding the previous bias in favour of the rich clusters .
our data reveal the existence of a sheet of galaxies connecting the main parts of the supercluster .
we describe the sample and observations in section [ sec_obs ] .
we present the results along with previous measurements in section [ sec_res ] and discuss the significance of the measurements in section [ sec_discuss ] .
[ sec_sel ] although a large body of galaxy velocity data is available in the literature for the ssc , the existing samples of redshifts in each cluster are highly incomplete , even at the bright end of the luminosity function
. we have therefore started a campaign to obtain complete samples down to the same magnitude below @xmath13 for each cluster .
each selected cluster has a projected diameter of 2.5 to 3.0 degrees , so the flair - ii system on the ukst with a @xmath12 deg@xmath1 field is an ideal facility for this project .
the very wide field also permits us to probe the regions between the dominant clusters neglected in previous observations . in this paper
we emphasise our results from these regions .
we selected targets from red eso / src sky survey plates scanned by the mama machine at paris observatory ( as described in paper ii ; see also infante et al .
the fields observed ( listed in table [ tab_obs ] ) were the standard survey fields nearest to the centre of the cluster ( 13:25:00 @xmath1431:00:00 b1950 ) .
these covered an area of 77 deg@xmath1 allowing us to probe the limits of the ssc out to radii as large as 8 deg .
we defined a sample of galaxies to a limit of @xmath0 , corresponding ( assuming a mean @xmath15 ) to @xmath16 , the nominal galaxy limiting magnitude of the flair - ii system .
this corresponds to an absolute magnitude of @xmath17 at the shapley distance of 200 .
this gave samples of 6001000 galaxies per field .
we then removed any galaxies with published measurements in the ned database or measured by h. quintana and r. carrasco ( private communication , 1997 ) : 46 galaxies for f382 , 81 for f383 and 200 for f444 . for each observing run
we then selected random subsamples of about 110 targets per field from the unobserved galaxies .
when preparing each field for observation at the telescope we made a further selection of 80 targets to observe ( 10 fibres being reserved for measurement of the sky background ) .
this final selection was essentially random , but we did reject any galaxies too close ( less than about 1 arcminute ) to another target already chosen or a bright star .
.journal of flair observations [ cols="<,>,^,>,>,>,>,>",options="header " , ] note : n@xmath18 is the number of galaxies with measured redshifts in each field .
we observed a total of 3 fields with the flair - ii spectrograph in 1997 may and two more in 1998 april .
the details of the observations are given in table [ tab_obs ] . in 1997 , out of 6 allocated nights we were only able to observe 3 flair fields successfully due to poor weather and the first of these was repeated over 3 nights .
field f444 was observed in particularly poor weather resulting in a much lower number of measured redshifts . in 1998
we again had poor weather , and were only able to observe two fields in an allocation of 8 half - nights .
the data were reduced as in drinkwater et al .
( 1996 ) using the dofibers package in iraf ( tody 1993 ) .
we measured the radial velocities with the rvsao package ( kurtz & mink 1998 ) contributed to iraf .
redshifts were measured for absorption - featured spectra using the cross - correlation task xcsao in rvsao .
we decided to adopt as the absorption velocity the one associated with the minimum error from the cross - correlation against the templates . in the great majority of cases ,
this coincided also with the maximum r parameter of tonry & davis ( 1979 ) .
the redshifts for the emission line objects were determined using the emsao task in rvsao .
emsao finds emission lines automatically , computes redshifts for each identified line and combines them into a single radial velocity with error .
spectra showing both absorption and emission features were generally measured with the two tasks xcsao and emsao and the result with the lower error used . in two spectra with very poor signal ( 13:05:19.9 @xmath1433:00:31 and 13:23:22.9 @xmath1436:47:09 )
the emission lines were measured manually and a conservative error of 150 assigned .
we measured velocities successfully for 306 galaxies in the sample : these are presented in table [ tab_data ] .
we have compared the distributions of the galaxies we measured to the input samples to check they are fair samples .
there is no significant difference in the distributions of the coordinates but there is a small difference in the magnitude distributions in the sense that the measured sample does not have as many of the faintest galaxies as the input sample .
this is to be expected as these would have the lowest signal in the flair - ii spectra , but this should not affect our study of the spatial distribution significantly .
previous studies of the ssc have covered a very large region of sky , but we will limit our analysis in this paper to the region of sky we observed with flair - ii : the three uk schmidt fields in table [ tab_obs ] . in some cases
we will further restrict our analysis to the two southern fields f382 and f383 where our observations were much more complete .
the distribution of these fields and the galaxies we observed is shown in fig .
[ fig_sky1 ] .
we also show any previously observed galaxies and the known abell clusters .
we present the resulting distribution of galaxies towards the shapley supercluster in fig .
[ fig_cone1 ] as cone diagrams and in fig .
[ fig_histo ] as the histogram of all velocities up to @xmath19 .
the importance of the ssc in this region of the sky is demonstrated by the fact the fully three quarters of the galaxies we measured belong to the ssc with velocities in the range 758018300 . in all the plots
the new data are indicated by different symbols to emphasise their impact ( this can also be seen by comparing these figures with the equivalent ones in paper ii ) .
it can be seen that by probing large regions of the ssc away from the rich abell clusters , we have revealed additional structure which we discuss in the following sections .
first in agreement with previous work we also note the presence of a foreground wall of 269 galaxies ( hydra - centaurus region ) at @xmath20 with @xmath21 in the range @xmath22 .
this distribution can be related with the nearby cluster a3627 associated with the `` great attractor '' ( kraan - korteweg et al .
1996 ) .
the previous observations reported in papers i and ii concentrated on the abell clusters , clarifying the location of many of them . we reproduce a list of the main clusters in the ssc region in table [ tab_clus ] for reference and plot their positions in fig .
[ fig_sky1 ] . as noted above
, our new measurements concentrate on galaxies outside the rich clusters in this field .
in particular we observed virtually no galaxies in foreground or background clusters .
we compare the distribution of the ssc galaxies to the abell clusters in two velocity slices in figs .
[ fig_skya ] and [ fig_skyb ] . in the near side of the ssc ( @xmath23 :
[ fig_skya ] ) we detected several new galaxies in the clusters a3571 and a3572 .
this region has a very extended velocity structure with several galaxies in the higher range ( fig .
[ fig_skyb ] ) . at the velocity of the main part of the ssc ( @xmath24 : fig .
[ fig_skyb ] ) we have found additional galaxies in many of the clusters , especially the poorer ones like as726 , as731 and a3564 .
the main conclusion however is that the clusters are seen as peaks in a sheet - like distribution rather than isolated objects .
the main impact of our new data is to revise our knowledge of the large - scale structure of the ssc by measuring a large number of galaxies away from the rich abell clusters previously studied .
the majority of the galaxies we observed were part of the ssc , so our principal result is to show that the ssc is bigger than previously thought with an additional 230 galaxies in the velocity range @xmath25 compared to 492 previously known in our survey area .
looking at the cone diagrams ( fig .
[ fig_cone1 ] ) and the velocity histogram in fig .
[ fig_histo ] our first new observation is that the ssc is clearly separated into two components in velocity space , the nearer one at @xmath26 ( @xmath27 ) to the east of the main concentration at @xmath28 ( @xmath29 ) .
the two regions contain 200 and 522 galaxies respectively .
some evidence for this separation was noted in the velocity distribution in paper ii , but it is much clearer with our new data .
secondly , it can be see from the declination cone diagram in fig .
[ fig_cone1 ] as well as the sky plots in figs .
[ fig_skya ] and [ fig_skyb ] that the southern part of the ssc consists of two large sheets of galaxies of which the previously measured abell clusters represent the peaks of maximum density . to consider the significance of this extended distribution of galaxies
it is helpful to define an inter - cluster sample consisting of galaxies in the southern fields ( f382 and f383 ) outside the known abell clusters in the ssc velocity range .
we eliminated all galaxies within a 0.5 degree radius ( about 1 abell radius ) of all the clusters shown in figs .
[ fig_skya ] and [ fig_skyb ] . very few of the previously - measured galaxies remain in the sample . in fig .
[ fig_histo2 ] we plot a histogram of the galaxy velocities in this inter - cluster sample compared to the predicted @xmath30 distribution of galaxies .
the predicted distribution was based on the number counts of metcalfe et al .
( 1991 ) normalised to the area of the southern sample after removing clusters ( 44 deg@xmath1 ) and corrected for completeness ( 304 out of a possible 1194 galaxies measured in total ) .
we also show the histogram ( shaded ) and predictions ( dashed ) for the previously - measured galaxies in the same field ( 128 out of a possible 1194 ) .
the histogram shows that even for the inter - cluster galaxies there is a large overdensity in the ssc region ( @xmath31 ) : we measure 161 galaxies compared to 74 expected .
this is an overdensity of @xmath2 detected at the 10 sigma level .
this is averaged over the whole ssc velocity range ; the overdensity in individual 1000 bins peaks at about 7 . by comparison the previous data ( 42 galaxies , 33 expected ) gave an overdensity of 1.3 detected at only 1.5 sigma .
the overdensity for the whole ssc including the abell clusters is , of course , much larger still .
these new observations mean that we must modify the conclusions of paper i about the overall shape of the ssc . in paper
i it was concluded from the velocity distribution of the clusters that the ssc was very elongated and either inclined towards us or rotating .
the ssc extends as far as our measurements to the south , so we find it is not elongated or flattened .
we now suggest that it is more complex still , being composed of the known abell clusters embedded in two sheets of galaxies of much larger extent .
our new observations of galaxies towards the shapely supercluster have , by surveying a large area away from known clusters , revealed substantial new large structures in the region .
the cluster is part of a much large structure than was apparent from the previous observations , extending uniformly in two sheets over the whole region we surveyed to the south of the core of the ssc .
we detected an additional 230 members of the ssc in our whole survey area , representing a 50% increase on the previous total of 492 ssc galaxies .
our measurements to the north of the cluster were much less complete ( only one field in poor weather ) so we can not exclude the possibility that these sheets of galaxies extend equally to the north .
recent results by bardelli , zucca & zamorani ( 1999 ) support this possibility : they have measured galaxies in 18 small ( 40 arcmin ) inter - cluster fields north of the core of the ssc and also find an overdensity at the ssc velocity . in paper i the effect of the ssc on the dynamics of the local group was estimated .
it was found that the mass in the cluster could account for at least 25% of the motion of the local group with respect to the cosmic microwave background .
our new data suggest that the ssc is at least 50% more massive with a significant part of the extra mass in the closer sub - region .
the ssc therefore has a more important effect on the local group that previously thought , although we defer a detailed calculation until we have additional data ( proust et al . 1999 , in preparation ) .
we wish to thank roberto de propris for kindly providing the software to calculate the predicted galaxy @xmath30 distributions and we are grateful of the staff of the ukst and aao for their assistance in the observations .
this research was partially supported by the cooperative programme ecos/ conicyt c96u04 and hc acknowledges support from a presidential chair in science .
mjd acknowledges receipt of an afcop travel grant and a french embassy fellowship in support of visits to paris observatory where some of this work was carried out .
this research 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 .
bardelli , s. , zucca , e. , malizia , a. , zamorani , g. , scaramella , r. , vettolani , g. , 1996 , a&a , 305 , 435 crrrr ra ( b1950 ) dec & @xmath32 & name & dist & ref + & ( ) + + 12:47:12 @xmath1436:29:00 & - & 3527 & 6 & n + 12:51:18 @xmath1426:07:00 & - & 1633 & 6 & n + 12:51:36 @xmath1428:45:00 & 15854 & 3528 & 4 & i + 12:52:54 @xmath1430:05:00 & 15960 & 3530 & 4 & i + 12:54:24 @xmath1432:39:00 & 22454 & 3531 & 5 & ii + 12:54:36 @xmath1430:06:00 & 16100 & 3532 & 4 & i + 12:56:18 @xmath1426:21:00 & 22994 & 1648 & - & n + 12:57:00 @xmath1433:24:00 & 14330 & s718 & 4 & ii + 12:58:18 @xmath1432:10:00 & 5007 & 3537 & - & n + 13:05:54 @xmath1434:18:00 & 27551 & 3542.1 & 4 & ii + 13:05:54 @xmath1434:18:00 & 39393 & 3542.2 & 4 & ii + 13:08:36 @xmath1433:49:00 & 22245 & 3545 & 6 & ii + 13:11:36 @xmath1429:11:00 & 22634 & 3549 & 5 & ii + 13:12:24 @xmath1433:23:00 & 17688 & s726 & 5 & i + 13:16:06 @xmath1431:33:00 & 14818 & 3552 & 6 & i + 13:16:24 @xmath1436:55:00 & 15110 & 3553 & 4 & i + 13:16:42 @xmath1433:13:00 & 14570 & 3554 & 4 & ii + 13:18:00 @xmath1428:43:00 & 14810 & 3555 & 4 & i + 13:18:42 @xmath1435:32:00 & 14960 & s729 & 5 & i + 13:20:12 @xmath1434:37:00 & 15140 & s731 & 4 & ii + 13:21:18 @xmath1431:24:00 & 14130 & 3556.1 & 4 & b + 13:21:18 @xmath1431:24:00 & 15066 & 3556.2 & 4 & b + 13:22:06 @xmath1428:37:00 & 14270 & 3557.1 & 5 & ii + 13:22:06 @xmath1428:37:00 & 23084 & 3557.2 & 5 & ii + 13:23:00 @xmath1436:59:00 & 20806 & s733 & - & n + 13:24:06 @xmath1426:51:00 & 10368 & 1736.1 & 3 & i + 13:24:06 @xmath1426:51:00 & 13357 & 1736.2 & 3 & i + 13:25:06 @xmath1431:14:00 & 14403 & 3558 & 3 & b + 13:26:58 @xmath1431:21:00 & 14841 & 1327 & - &
n + 13:27:06 @xmath1429:16:00 & 13812 & 3559 & 4 & i + 13:28:38 @xmath1431:33:43 & 12868 & 1329 & - & i + 13:29:00 @xmath1432:58:00 & 13864 & 3560 & 3 & i + 13:30:42 @xmath1431:25:00 & 14492 & 3562 & 3 & b + 13:31:30 @xmath1434:58:00 & 14930 & 3564 & 3 & i + 13:33:48 @xmath1433:43:00 & 3268 & 3565 & - & n + 13:36:06 @xmath1435:18:00 & 15469 & 3566 & 4 & ii + 13:39:24 @xmath1426:01:00 & 32048 & 1771 & - & n + 13:43:54 @xmath1437:40:00 & 11152 & 3570 & - & n +
13:44:36 @xmath1432:37:00 & 11730 & 3571 & 2 & i + 13:45:18 @xmath1433:08:00 & 12142 & 3572 & - & n + 13:46:18 @xmath1430:03:00 & 4227 & 3574 & - & n + 13:49:42 @xmath1432:38:00 & 11242 & 3575 & 4 & ii + 13:51:30 @xmath1427:36:00 & 14870 & 3577 & 4 & ii + 13:54:42 @xmath1424:29:00 & 11152 & 3578 & 3 & ii + | we present new radial velocities for 306 bright ( @xmath0 ) galaxies in a 77 deg@xmath1 region of the shapley supercluster measured with the flair - ii spectrograph on the uk schmidt telescope .
the galaxies we measured were uniformly distributed over the survey area in contrast to previous samples which were concentrated in several rich abell clusters .
most of the galaxies ( 230 ) were members of the shapley supercluster : they trace out two previously unknown sheets of galaxies linking the abell clusters of the supercluster . in a 44 deg@xmath1 area of the supercluster
excluding the abell clusters , these sheets alone represent an overdensity of a factor of @xmath2 compared to a uniform galaxy distribution .
the supercluster is not flattened in the declination direction as was suggested in previous papers . within our survey area
the new galaxies contribute an additional 50% to the known contents of the shapley supercluster with a corresponding increase in its contribution to the motion of the local group .
= 2em = 17.5 cm = 24.6 cm = -2.0 cm = -0.7 cm = -0.7 cm @xmath3 school of physics , university of new south wales , sydney 2052 , australia , [email protected] + @xmath1 daec , observatoire de meudon , 92195 meudon cedex , france , [email protected] + @xmath4 anglo - australian observatory , coonabarabran , nsw 2357 , australia , [email protected] + @xmath5 departamento de astronomia y astrofisica , pontificia universidad catlica de chile + casilla 104 , santiago 22 , chile , [email protected] + @xmath6 observatoire de nice , 06304 nice cedex 4 france , [email protected] + * keywords : * galaxies : clusters ; distances and redshifts large - scale structure of the universe |
the weak lensing of cosmic microwave background ( cmb ) anisotropy angular power spectrum is now well understood in the literature @xcite .
the modifications result in a smoothing of the acoustic peak structure at large angular scales and an increase in power below a few arcminute angular scales corresponding to the the damping tail of cmb anisotropies @xcite .
the angular power spectrum of the lensing potential out to the last scattering surface can be established with quadratic estimators that probe lensing non - gaussianity at the 4-point level of a cmb map @xcite .
such a reconstruction of the lensing potential is helpful for cmb b - mode studies of polarization @xcite , especially in the context of searching for the signature of the primordial tensor modes @xcite .
this is due to the fact that in addition to inflationary gravitational waves , the b - modes of cmb polarization also contains a signal generated by lensing of scalar e - modes with a peak in power at a few arcminute angular scales @xcite .
the lensing reconstruction has now been applied to the existing wilkinson microwave anisotropy probe ( wmap ) data leading to a @xmath2 2@xmath3 to 3@xmath3 detection of gravitational lensing in the cmb through a correlation between the reconstructed lensing potential and tracers of the large - scale structure such as radio galaxies @xcite . in parallel with the progress on lensing studies with the cmb
, the search for primordial non - gaussianity using the cmb bispectrum with constraints on the non - gaussianity parameter @xmath0 is now an active topic in cosmology @xcite .
the 5-year wmap data is consistent with @xmath4 at the 95% confidence level for the local model @xcite , though a non - zero detection of primordial non - gaussianity at the same 95% confidence level with @xmath5 is claimed elsewhere using the wmap 3-year data @xcite .
this result , if correct , has significant cosmological implications since the expected value under standard inflationary models is @xmath6 @xcite , though alternative models of inflation , such as the ekpyrotic cosmology @xcite , generally predict a large primordial non - gaussianity with @xmath0 at few tens .
just as the cmb power spectrum is modified by lensing from potential fluctuations of the intervening large - scale structure @xcite , the cmb bispectrum will also be modified by gravitational lensing .
the correlation between the projected lensing potential and cmb secondary effects , such as the integrated sachs - wolfe ( isw ) effect or the sunyaev - zeldovich ( sz ) effect , leads to a non - gaussian signal at the three point level @xcite .
these secondary non - gaussianities are expected even if the primordial perturbations are gaussian and impact existing primordial non - gaussianity parameter measurements by introducing a small , but unavoidable , bias @xcite . beyond secondary non - gaussianities , weak lensing by the intervening large - scale structure
maps the intrinsically non - gaussian cmb sky to a different anisotropy pattern when observed today .
thus the bispectrum one reconstructs with a cmb map , assuming it to be of the expected form at the last - scattering surface due to primordial non - gaussian perturbations , will result in a biased estimate of the primordial non - gaussianity parameter .
the existing estimator can be modified to account for lensing modifications and to obtain a bias - free estimate of the non - gaussianity , but at the expense of factorizability that has allowed fast computation of the bispectrum in existing analyses @xcite . since lensing modifies the anisotropy pattern by smoothing the fluctuations , a change in the minimum detectable non - gaussianity parameter @xmath0 for a given experiment
is expected to be different from the existing values in the literature @xcite . in this paper
, we present a general derivation of the lensed cmb primary bispectrum and quantify above statements on the changes imposed by lensing for the detection of primordial non - gaussianity .
we find that the non - gaussianity parameter measured with wmap , ignoring lensing , will result in an estimate of @xmath0 for the local model that is biased low by about 6% , when measurements are extended out to @xmath7 .
furthermore , with lensing , the minimum detectable level of @xmath0 with planck is increased by roughly 40% from less than 5 to about 7 and the cosmic variance limit of @xmath0 is increased from 3 to 5 .
this paper is organized as follows : we first discuss the cmb primary bispectrum of the local type in section ii .
some basic ingredients related to the lensing calculation is presented in section iii .
we derive the lensing effect on the bispectrum , under both flat - sky and all - sky formulations , in section iv .
we discuss our results and conclude with a summary in section v. in illustrating our results we make use of the standard flat @xmath8cdm cosmological model consistent with wmap with @xmath9 , @xmath10 , @xmath11 , @xmath12 and @xmath13 .
the cmb temperature perturbation on the sky , @xmath14 , is decomposed into its multipole moments @xmath15 the angular power spectrum and bispectrum of cmb anisotropies are defined in the usual way , respectively , as @xmath16 where , for the bispectrum , we have introduced the wigner-3@xmath17 symbol ( see the appendix of ref .
@xcite for some useful properties of this symbol ) .
the cmb bispectrum is generated by a coupling of the local - type with a quadratic correction to the newtonian curvature such that @xmath18 \
, \label{phi}\ ] ] where @xmath19 is the linear and gaussian perturbation and @xmath20 in the non - gaussianity parameter , which is taken to be scale independent @xcite . in fourier space
, we can decompose equation ( [ phi ] ) as @xmath21 with @xmath22 where @xmath23 is the linear power spectrum , defined as @xmath24 the multipole moments of the anisotropy can be written as @xmath25 where @xmath26 from above is the primordial curvature perturbation in the fourier space , and @xmath27 is the radiation transfer function . with @xmath28 and @xmath29 , the moments can be separated into two components with @xmath30 . the cmb angular power spectrum can be defined using the transfer function and the power spectrum of dominant linear fluctuations as @xmath31 using the definition of the angular bispectrum ( equation [ eqn : pscmb ] ) , the primordial temperature anisotropy bispectrum can be written as @xmath32 \
, , \end{aligned}\ ] ] which can be simplified to @xcite @xmath33 where @xmath34 and @xmath35 when illustrating our results , we make use of a modified code of cmbfast @xcite for the standard flat @xmath8cdm cosmological model to fully calculate radiation transfer functions when generating the cmb primary bispectrum .
the effects of weak lensing can be encapsulated , under the born approximation , in the radial projection of the gravitational potential ( @xmath36 ) , given as @xcite @xmath37 where @xmath38 is the line - of - sight comoving distance ( or look - back time ) to a redshift @xmath39 from the observer with last scattering surface at @xmath40 , and @xmath41 is the comoving angular diameter distance . in a spatially flat universe , @xmath42 . here
, we ignore the time - delay effect as it is small compared to the geometric lensing effect captured by equation ( [ eqn : lenspotential ] ) @xcite .
the calculation related to the cmb bispectrum described below requires the angular power spectrum of lensing potential @xmath43 , which can be decomposed into the multiple moments as @xmath44 with the lensing power spectrum defined using @xmath45 to obtain @xcite @xmath46 ^ 2 \ , , \label{eqn : clphi}\ ] ] where @xmath47 here @xmath48 describes the radial evolution of potential fluctuations .
modifications to the cmb anisotropies , generated at higher order in lensing potential fluctuations , are at the level of at most 5% relative to those due to the lensing potential angular power spectrum @xcite .
the bispectrum of lensing potentials , due to the non - linear evolution of density perturbations , also modifies the cmb primary bispectrum , but these changes can also be ignored since the lensing potential bispectrum is at the order @xmath49 , while changes we describe are first order in the angular power spectrum of the lensing potential . using the limber approximation , equation ( [ eqn : clphi ] ) can be further simplified , but we do not make use of this approximation in numerical calculations illustrated here since the flat - sky form of the potential power spectrum is known to bias lensing results of the power spectrum by about 10% at all multipoles
we first give a treatment of the lensing of the cmb bispectrum assuming a flat - sky approximation and discuss a derivation under the spherical sky later .
weak lensing deflects the path of background photons resulting in a remapping of the observed anisotropy pattern on the sky . following an approach similar to @xcite , we write the lensed temperature anisotropy as @xmath50 { \nonumber\\ } & \approx & { { \theta}}({{\bf \hat{n } } } ) + \nabla_i { \phi}({{\bf \hat{n } } } ) \nabla^i { { \theta}}({{\bf \hat{n } } } ) { \nonumber\\ } & & + { 1 \over 2 } \nabla_i { \phi}({{\bf \hat{n } } } ) \nabla_j { \phi}({{\bf \hat{n } } } ) \nabla^{i}\nabla^{j } { { \theta}}({{\bf \hat{n } } } ) + \ldots \ ; \label{eqn : pert}\end{aligned}\ ] ] here , @xmath51 is the unlensed cmb temperature anisotropy , @xmath52 is the lensed anisotropy , and @xmath53 is the lensing deflection angle for the cmb photons .
taking the fourier transform , as appropriate for a flat - sky , we write the lensed temperature anisotropy in fourier space as @xmath54 where @xmath55 to the second order in lensing potential in the perturbative expansion .
the observed angular power spectrum of cmb anisotropies under weak lensing is discussed in @xcite .
the resulting power spectrum consists of both the unlensed intensity and a perturbative correction related to the lensing effect . making use of the expansion and after some straight forward calculations , we obtain the lensed anisotropy power spectrum as @xmath56 ^ 2 , \label{eqn : ttflat}\ ] ] where @xmath57 here
, @xmath58 describes the variance of the deflection angle . for @xmath8cdm cosmology , @xmath59 .
this derivation makes use of the flat - sky approximation to describe the lensing effect on cmb anisotropy power spectrum .
when the expressions derived in the previous section for @xmath60 and @xmath61 under the exact spherical - sky treatment are used in equation ( [ eqn : ttflat ] ) , the lensed cmb power spectrum can be derived with a less bias than using , say , the flat - sky result for @xmath61 in the same expression @xcite . keeping the flat - sky approximation
, we can define the angular bispectrum as @xmath62 and following the approach similar to the lensed angular power spectrum that led to equation ( [ eqn : ttflat ] ) , the lensed bispectrum can be expressed as @xmath63 + { \int { d^2 { \bf l'}\over ( 2\pi)^2}}c^{\phi}_{l ' } { \nonumber\\ } & & \quad \times \big[b^{{{\theta}}}_{({{\mathbf{l}}}_1,{{\mathbf{l}}}_2-{{\mathbf{l}}}',{{\mathbf{l}}}_3+{{\mathbf{l}}}')}({{\mathbf{l}}}_2-{{\mathbf{l}}}')\cdot { { \mathbf{l}}}'({{\mathbf{l}}}_1+{{\mathbf{l}}}_2-{{\mathbf{l}}}')\cdot { { \mathbf{l } } } ' { \nonumber\\ } & & \quad \quad + b^{{{\theta}}}_{({{\mathbf{l}}}_1-{{\mathbf{l}}}',{{\mathbf{l}}}_2+{{\mathbf{l}}}',{{\mathbf{l}}}_3)}({{\mathbf{l}}}_1-{{\mathbf{l}}}')\cdot { { \mathbf{l}}}'({{\mathbf{l}}}_3+{{\mathbf{l}}}_1-{{\mathbf{l}}}')\cdot { { \mathbf{l } } } ' { \nonumber\\ } & & \quad \quad + b^{{{\theta}}}_{({{\mathbf{l}}}_1+{{\mathbf{l}}}',{{\mathbf{l}}}_2,{{\mathbf{l}}}_3-{{\mathbf{l}}}')}({{\mathbf{l}}}_3-{{\mathbf{l}}}')\cdot { { \mathbf{l}}}'({{\mathbf{l}}}_3+{{\mathbf{l}}}_2-{{\mathbf{l}}}')\cdot { { \mathbf{l } } } ' \big ] \ , \nonumber.\end{aligned}\ ] ] note that we have identified the flat - sky bispectrum as @xmath64 to distinguish from the all - sky bispectrum @xmath65 .
the two are related through @xmath66 the derivation related to the lensing of the cmb bispectrum under the more appropriate spherical sky can be obtained by replacing the fourier components with spherical harmonic multipole moments . in this case , the lensed field can be represented as @xcite @xmath67 \ , , \nonumber\end{aligned}\ ] ] where , the integrals over the spherical harmonics were replaced , in the last step , by the geometrical factors @xmath68 using equation ( [ eqn : pscmb ] ) , the lensed cmb temperature bispectrum can then be expressed as @xmath69 leading to @xmath70 \ , . \nonumber\end{aligned}\ ] ] noting that the wigner-3@xmath17 symbol obeys the identity @xmath71 we can re - write the lensed bispectrum as @xmath72 where @xmath73 the sum of the geometric term @xmath74 over @xmath75 yields @xcite @xmath76 hence , with summation over @xmath77 leading to a factor @xmath78 , the expression for @xmath79 simplifies to @xmath80 ( equation [ eqn : f ] ) as a function of @xmath81 and @xmath82 for the primary bispectrum ( top panel ) and the lensed primary bispectrum ( bottom panel ) .
we take @xmath83.,title="fig : " ] ( equation [ eqn : f ] ) as a function of @xmath81 and @xmath82 for the primary bispectrum ( top panel ) and the lensed primary bispectrum ( bottom panel ) .
we take @xmath83.,title="fig : " ] as a function of @xmath84 and @xmath82 ( same as the difference between bottom and top panels of figure 4 ) . ] in order to evaluate @xmath85 , we first re - express @xmath86 as @xmath87 where @xcite @xmath88 { \left ( \begin{array}{ccc } \ ! l\ ! & { l^{\prime}}\ ! & { l^{\prime \prime}}\ ! \\ \ ! 0\ ! & 0\ ! & 0\ ! \end{array } \right ) } \,.\ ] ] then the expression for @xmath85 can be re - written as @xmath89 where , in the last step , we have introduced the wigner-6@xmath17 symbol @xcite .
the values of the wigner-6@xmath17 symbol can be computed numerically with a fast and efficient recursive algorithm @xcite . finally , substituting the expressions for @xmath79 and @xmath85 in equation ( [ eq : lensedbis ] ) and including all permutations in a single expression , we can write the lensed bispectrum as @xmath90 b_{l_1 l_2 l_3}^{{{\theta}}}{\nonumber\\ } & & + \sum_{lpq } c_l^{\phi \phi}\bigg [ f_{l_2 l p}f_{l_3 l q } ( -1)^{n } { \left\ { \begin{array}{ccc } l_1 & l_2 & l_3 \\
l & q & p \end{array } \right\ } } { b}_{l_1 p q}^{{{\theta } } } { \nonumber\\ } & & \hspace{17 mm } f_{l_3 l p}f_{l_1 l q } ( -1)^{n } { \left\ { \begin{array}{ccc } l_1 & l_2 & l_3 \\
p & l & q \end{array } \right\ } } { b}_{p l_2 q}^{{{\theta } } } { \nonumber\\ } & & \hspace{17 mm } f_{l_1 l p}f_{l_2 l q } ( -1)^{n } { \left\ { \begin{array}{ccc } l_1 & l_2 & l_3 \\
q & p & l \end{array } \right\ } } { b}_{p q l_3}^{{{\theta } } } \bigg]\nonumber\end{aligned}\ ] ] where @xmath91 and @xmath92 .
we illustrate the modification to the equilateral configurations of the bispectrum in figure 1 , where we plot @xmath93 as a function of the multipole @xmath94 .
the primary cmb bispectrum assumes @xmath95 and is calculated with the full radiation transfer function @xmath27 .
the lensing description makes use of the all - sky treatment to calculate both @xmath61 and the lensed bispectrum .
the flat - sky expression gives a result consistent with the all - sky expression at better than 5% at all multipoles if the all - sky expression for @xmath61 is used in both calculations .
the difference is at the level of 10% if the two expression make use of the two separate calculations of @xmath61 , as in the case of the angular power spectrum @xcite . in the case of the equilateral configurations of the bispectrum ,
the lensing effect can be best described as a smoothing and a decrease of the amplitude of non - gaussianity power in the equilateral configuration of the bispectrum when @xmath96 and a subsequent increase in the bispectrum amplitude at small angular scales . in figure 2 , we show two squeezed configurations ( @xmath97 ) of the bispectrum , with the short length fixed at either 10 ( left panel ) or 100 ( right panel ) , as a function of the multipole @xmath94 of one side with the third side fixed at either @xmath98 or @xmath99 . without lensing , a comparison of figure 1 and figure 2 reveals a well - known result in the literature that the local type of the cmb bispectrum is dominated by squeezed configurations with one small side and two large sides for the bispectrum mode shape in the multipole space . with lensing , the amplitude of the squeezed configurations is significantly reduced when two of the sides have lengths @xmath100 in the multipole space .
this can be again described as a smoothing effect with lensing by the intervening large - scale structure ; removing or `` washing out '' the primordial non - gaussian signature in the cmb map at angular scales below a few arcminutes .
thus , when lensed by the large - scale structure , the primordial non - gaussian cmb sky appears more gaussian at arcminute scales when studying the non - gaussianity at the three - point level . at same angular scales ,
however , the cmb sky appears more non - gaussian due to lensing at the four - point level probed by the trispectrum @xcite .
the removal of the non - gaussianity is associated with the squeezed configurations , which dominate the overall signal - to - noise ratio for the detection of the primary bispectrum without lensing .
although non - gaussianity is reduced for the squeezed configurations , lensing leads to an increase in the amplitude of the bispectrum for equilateral configurations where @xmath101 .
this increase , however , is insignificant in terms of the overall signal - to - noise ratio as the contribution to the cumulative signal - to - noise coming from these configurations is lower , owing to the higher variances associated with foregrounds and instrumental noise at these angular scales . to further quantify this statement , we plot , in figure 3 , the signal - to - noise ratio calculated as @xmath102 where the noise variance calculation involves all contributions to the angular power spectrum with @xmath103 where we include the lensed cmb power spectrum ( @xmath104 ) , secondary anisotropies ( @xmath105 ) , and the noise power spectrum ( @xmath106 ) for both wmap and planck . for secondaries ,
we include the sz power spectrum calculated with the halo model @xcite and make use of the noise calculations from ref .
@xcite for wmap and planck .
for the case involving an experiment limited by the cosmic variance , we set @xmath107 . in the left panel of figure 3 ,
we plot @xmath108 as a function of @xmath109 , while in the right panel we plot the cumulative signal - to - noise ratio out to @xmath109 in the x - axis .
in the case where lensing is not included , signal - to - noise ratio estimates for the bispectrum detection are consistent with previous calculations in the literature @xcite .
with lensing , however , the signal - to - noise ratios are changed .
as can be seen from the left panel of figure 3 , there is an overall reduction in the signal - to - noise ratio when @xmath110 .
this difference comes from the previously described decrease in the amplitude of the non - gaussianity in squeezed configurations of the bispectrum with lensing imposed .
to further understand the differences in the signal - to - noise ratio of the lensed primary bispectrum , we plot , in figure 4 , the quantity @xmath111 with two separate estimates for @xmath112 and @xmath113 to estimate this quantity without and with lensing , respectively .
note that the overall signal - to - noise ratio comes from integrating this quantity over the variables @xmath84 and @xmath82 and we include the factor @xmath114 to account for the logarithmic scaling .
a comparison of the two panels in figure 4 reveals an overall decrease in the amplitude at @xmath115 in the case with the lensed primary bispectrum relative to the primary bispectrum alone . in figure 5
, we plot the difference of the two as a contour plot to show that lensing results in an overall decrease in the signal - to - noise ratio in the squeezed configurations when @xmath116 and @xmath117 , while there is an increase in the signal - to - noise ratio when @xmath118 for all values of @xmath119 .
these plots demonstrates the same trends described with respect to figure 2 involving a decrease in the amplitude of the squeezed configurations of the lensed bispectrum .
while there is a slight increase in the lensed bispectrum amplitude at @xmath120 , such small angular scales are not probed by planck . even in the cosmic variance limit , unfortunately , this small increase is insignificant given that at these same angular scales secondary anisotropies dominate the bispectrum noise variance . in terms of the cumulative signal - to - noise ratio values shown in figure 3 , the minimum @xmath121 to detect the bispectrum with planck and a cosmic variance limited experiment is increased by about 30% to 40% from @xmath122 for planck to @xmath123 . the cosmic variance - limited detection threshold for @xmath124
is increased from 3 to 5 . while the difference in cumulative signal - to - noise ratio seems insignificant for an experiment like wmap , weak lensing could impact existing measurements of the non - gaussianity parameter @xmath0 @xcite . to understand the lensing bias introduced to @xmath0 , we follow the discussion in ref .
@xcite and note that the current estimators of the non - gaussianity parameter use @xmath125 @xcite , with @xmath126 where @xmath127 is the primary bispectrum , and @xmath128 is the covariance matrix for bispectrum measurements involving triplets of @xmath129 and @xmath130 .
this estimator is the optimal estimator for non - gaussianity measurements , but given the complications associated with estimating the covariance , existing studies make use of a sub - optimal estimator which approximates the covariance with variance @xmath131 , and introduces a linear term to equation ( [ eqn : s ] ) to minimize the variance of @xmath132 @xcite .
note that @xmath133 is the overall normalization factor that can be calculated from equation ( [ eqn : s ] ) by replacing @xmath134 with @xmath112 . while weak lensing modifies the observed bispectrum @xmath135
, existing measurements make the assumption that @xmath136 .
this results in a biased estimate of @xmath132 from the true value of the non - gaussianity parameter @xmath137 .
the fractional difference of this bias @xmath138 can be calculated through the covariance between the lensed and unlensed cmb primary bispectrum @xmath139 $ ] , where we have simply written the variance as @xmath140 .
we plot @xmath141 as a function of @xmath94 to which non - gaussianity parameter measurements are performed in figure 6 .
existing measurements with wmap data probe out to @xmath142 and we find that existing estimates of @xmath0 are biased by @xmath2 6% . for planck ,
if lensing is ignored , the bias is at the level of 30% .
this bias is not the same fractional difference in the signal - to - noise ratio that one can infer from figure 3 since the fractional difference in the signal - to - noise ratio with and without lensing involves a ratio of the form @xmath143 $ ] .
note that in future an unbiased estimate of @xmath0 can be obtained by replacing @xmath144 in equation ( [ eqn : s ] ) with the lensed bispectrum @xmath145 and recalculating the normalization factor @xmath133 with the lensed primary bispectrum .
unfortunately , while without lensing the cmb primary bispectrum of the local model factorizes into two separate integrals with @xmath146 and @xmath147 ( described in section ii ) , this factorizability is no longer preserved when lensed and impacts an easy estimation of the non - gaussianity parameter with the existing estimator @xcite . for planck and other cmb experiments that can probe down to small angular scales for primordial non - gaussianity measurements
, it will be necessary to implement an estimator that accounts for the lensing effect . to summarize the main results of this paper ,
we have discussed the primary cmb bispectrum generated at the last scattering surface , but observed today after it is weak lensed by the intervening large - scale structure .
unfortunately , as we have found , weak lensing leads to an overall decrease in the amplitude of non - gaussianity with the biggest change on the squeezed configurations of the bispectrum that dominate the overall signal - to - noise ratio when studying the primordial non - gaussianity parameter . for an experiment such as the wilkinson microwave anisotropy probe ( wmap ) ,
the modifications imposed by lensing results in an estimate of @xmath124 of the local model that is biased low by about 6% . for a high resolution experiment such as planck , the lensing modification to the bispectrum must be accounted for when attempting to estimate the primordial non - gaussianity .
the minimum detectable value of @xmath124 for a cosmic variance limited experiment is @xmath2 5 .
we thank participants of the recent non - gaussianity workshop at the perimeter institute for helpful discussions and alex amblard for help with figure 4 .
ps thanks tommaso stasi .
this work was supported by nsf career ast-0645427 .
we acknowledge the use of cmbfast by uros seljak and matias zaldarriaga @xcite .
see , e.g. , u. seljak and m. zaldarriaga , phys .
lett . * 82 * , 2636 ( 1999 ) ; phys .
d * 60 * , 043504 ( 1999 ) ; m. zaldarriaga and u. seljak , phys .
rev . d * 59 * , 123507 ( 1999 ) ; m. zaldarriaga , phys .
d * 62 * , 063510 ( 2000 ) [ arxiv : astro - ph/9910498 ] ; m. h. kesden , a. cooray and m. kamionkowski , phys .
d * 66 * , 083007 ( 2002 ) [ arxiv : astro - ph/0208325 ] ; c. vale , a. amblard and m. j. white , new astron . *
10 * , 1 ( 2004 ) [ arxiv : astro - ph/0402004 ] ; a. amblard , c. vale and m. j. white , new astron . * 9 * , 687 ( 2004 ) [ arxiv : astro - ph/0403075 ] ; s. das and p. bode , arxiv:0711.3793 [ astro - ph ] . for a recent review
see , a. lewis and a. challinor , phys .
rept . *
429 * , 1 ( 2006 ) [ arxiv : astro - ph/0601594 ] .
w. hu , phys .
rev . d * 62 * 043007 ( 2000 ) [ arxiv : astro - ph/0001303 ] w. hu and t. okamoto , astrophys . j. * 574 * , 566 ( 2002 ) [ arxiv : astro - ph/0111606 ] ; m. kesden , a. cooray , and m. kamionkowski , phys . rev .
d * 67 * , 123507 ( 2003 ) [ arxiv : astro - ph/0302536 ] ; c. m. hirata and u. seljak , phys .
d * 68 * , 083002 ( 2003 ) [ arxiv : astro - ph/0306354 ] .
m. kamionkowski , a. kosowsky , and a. stebbins , phys .
78 * , 2058 ( 1997 ) [ arxiv : astro - ph/9609132 ] ; u. seljak and m. zaldarriaga , phys .
lett . * 78 * , 2054 ( 1997 ) [ arxiv : astro - ph/9609169 ] .
m. kesden , a. cooray , and m. kamionkowski , phys .
* 89 * , 011304 ( 2002 ) [ arxiv : astro - ph/0202434 ] ; l. knox and y .- s .
song , phys .
* 89 * , 011303 [ arxiv : astro - ph/0202286 ] ; u. seljak and c. hirata , phys . rev .
d * 69 * , 043005 ( 2004 ) [ arxiv : astro - ph/0310163 ] .
e. i. buchbinder , j. khoury and b. a. ovrut , arxiv:0710.5172 [ hep - th ] . j. l. lehners and p.
j. steinhardt , arxiv:0712.3779 [ hep - th ] .
d. m. goldberg and d. n. spergel , phys .
d * 59 * , 103002 ( 1999 ) [ arxiv : astro - ph/9811251 ] .
a. r. cooray and w. hu , astrophys .
j. * 534 * , 533 ( 2000 ) [ arxiv : astro - ph/9910397 ] .
d. babich and e. pierpaoli , arxiv:0803.1161 [ astro - ph ] .
e. komatsu , d. n. spergel and b. d. wandelt , astrophys .
j. * 634 * , 14 ( 2005 ) [ arxiv : astro - ph/0305189 ] .
u. seljak and m. zaldarriaga , astrophys .
j. * 469 * , 437 ( 1996 ) [ arxiv : astro - ph/9603033 ] .
r. d. blandford et al . , mon . not .
soc . * 251 * , 600 ( 1991 ) ; j. miralda - escud , astrophys .
j. * 380 * , 1 ( 1991 ) ; n. kaiser , astrophys .
j. * 388 * , 277 ( 1992 ) ; m. bartelmann and p. schneider , astron .
astrophys . * 259 * , 413 ( 1992 ) . for reviews ,
see , m. bartelmann and p. schneider , phys .
rept . * 340 * , 291 ( 2001 ) ; p.
schneider gravitational lensing : strong , weak & micro , lecture notes of the 33rd saas - fee advanced course , ( berlin : springer - verlag ) .
w. hu and a. cooray , phys .
d * 63 * , 023504 ( 2001 ) [ arxiv : astro - ph/0008001 ] .
a. r. cooray , new astron .
* 9 * , 173 ( 2004 ) [ arxiv : astro - ph/0309301 ] ; a. challinor and a. lewis , phys .
d * 71 * , 103010 ( 2005 ) [ arxiv : astro - ph/0502425 ] .
w. hu , phys .
d * 64 * , 083005 ( 2001 ) .
a. cooray and r. k. sheth , phys .
rept . * 372 * , 1 ( 2002 ) [ arxiv : astro - ph/0206508 ] ; a. cooray , phys .
d * 62 * , 103506 ( 2000 ) [ arxiv : astro - ph/0005287 ] .
p. creminelli , a. nicolis , l senatore , m. tegmark and m. zaldarriaga , journal of cosmology and astro - particle physics * 5 * , 4 ( 2006 ) , [ arxiv : astro - ph/0509029 ] | the cosmic microwave background ( cmb ) bispectrum is a well - known probe of the non - gaussianity of primordial perturbations .
just as the intervening large - scale structure modifies the cmb angular power spectrum through weak gravitational lensing , the cmb primary bispectrum generated at the last scattering surface is also modified by lensing .
we discuss the lensing modification to the cmb bispectrum and show that lensing leads to an overall decrease in the amplitude of the primary bispectrum at multipoles of interest between 100 and 2000 through additional smoothing introduced by lensing . since weak lensing is not accounted for in current estimators of the primordial non - gaussianity parameter , the existing measurements of @xmath0 of the local model with wmap out to @xmath1 is biased low by about 6% . for a high resolution experiment such as planck , the lensing modification to the bispectrum must be properly included when attempting to estimate the primordial non - gaussianity or the bias will be at the level of 30% . for planck ,
weak lensing increases the minimum detectable value for the non - gaussianity parameter of the local type @xmath0 to 7 from the previous estimate of about 5 without lensing .
the minimum detectable value of @xmath0 for a cosmic variance limited experiment is also increased from less than 3 to @xmath2 5 . |
a _ coloring _ of graph @xmath8 is a function @xmath9 where @xmath10 for all @xmath11 .
we consider only finite graphs .
the _ clique size _ of graph @xmath0 is the number @xmath12 of vertices in a largest complete subgraph of @xmath13 every coloring of graph @xmath0 has at least @xmath12 colors .
an _ interval _ is a convex set of real numbers .
graph @xmath8 is an _ interval graph _ if there is a function that assigns to each @xmath14 an interval @xmath15 so that @xmath16 such a function @xmath17 is an _ interval representation .
_ every interval graph @xmath0 has a coloring of just @xmath12 colors .
one can be constructed by ordering vertices by interval left end and coloring by first fit . in some applications
, however , we begin assigning colors before the whole graph is seen .
a coloring algorithm is _ online _ if , given graph @xmath0 on vertices @xmath18 , it assigns ( irrevocably ) for each @xmath19 a color to @xmath20 that depends only on the subgraph of @xmath0 induced by seen vertices @xmath21 for example , the _ first - fit _ algorithm produces coloring @xmath22 as follows : for @xmath23 from 1 to @xmath24 , let @xmath25 be the least positive integer available , i.e. , not a member of @xmath26 the coloring produced by this algorithm depends on vertex order .
for example , figures [ f01 ] ] and [ f02 ] ] show first - fit colorings of the 4-vertex path
. first - fit may be the first algorithm that comes to mind when the goal is to use few colors .
how wasteful is it in the worst case ?
let @xmath27 be the maximum over all vertex orders of the number of colors it uses on graph @xmath0 .
for example , @xmath28 kierstead @xcite showed that @xmath29 for every interval graph @xmath0 .
this was improved to @xmath30 by kierstead and qin @xcite , then @xmath31 by pemmaraju , raman , and varadarajan @xcite .
brightwell , kierstead , and trotter ( unpublished ) and narayanaswamy and subhash babu @xcite observed that the technique in @xcite yields @xmath32 .
these results provide bounds above the worst - case performance ratio of the first - fit algorithm on interval graphs , @xmath33 so @xmath34 we want to know the exact value of @xmath35 .
this motivates a search for lower bounds .
already @xmath36 due to the example @xmath37 variations on this example yield improvements , and this is the structure of the present paper : we recapitulate results of earlier authors in section [ s2 ] , generalize their constructions in section [ s3 ] , and tune the new construction to get @xmath38 in section [ rseq ] . in section [ nocap ] we show that this lower bound is the best that our approach allows .
a few more historical notes on related problems appear in chapter 1 of the dissertation of smith @xcite .
figure [ f02 ] is a certificate of a bound below @xmath39 pertinent features include total number of colors , clique size , and the set of colors neighboring each vertex , especially colors below its own .
compare this to a brick wall with bricks arranged in courses .
we may be concerned with its height , with its density , and , for stability , that each brick be supported by something below .
gyrfs and lehel @xcite defined a wall to be essentially a graph and vertex order .
a linear order is overprecise for our purpose .
more suitable is a partial order .
for example , the first - fit algorithm uses 3 colors on the 4-vertex path whenever the path s end vertices precede its inner vertices .
colors can be determined during the building of a wall .
suppose for all @xmath23 that the levels of the neighbors of each vertex in level @xmath23 include @xmath40 , but not @xmath23 ( as in figures [ f01 ] , [ f02 ] , and [ f03 ] ) .
when we present such a wall to the first - fit algorithm from bottom to top , the algorithm gives color 1 to the first level , color 2 to the second level , and so on .
first - fit colors are determined by this structural condition known as support .
we get large bounds below @xmath35 from tall walls with small cliques and supported vertices .
let @xmath41 denote the set of neighbors of vertex @xmath42 , and @xmath43 : = n(v ) \cup \{v\}.$ ] when @xmath22 is a function on domain @xmath44 , and @xmath45 we denote the image of @xmath46 by @xmath47 pair @xmath48 is an _
@xmath49-wall _ if : 1 .
@xmath0 is an interval graph on vertex set @xmath44 with a given interval representation ; 2 .
[ wc2 ] @xmath50 is a coloring of @xmath0 ; 3 .
[ wc3 ] @xmath51 ) \supseteq \{1,\ldots , f(v)\}$ ] for all @xmath14 ( _ support _ ) ; and 4 .
@xmath52 for example , figure [ f01 ] depicts a @xmath53-wall .
figure [ f02 ] depicts a @xmath54-wall .
figure [ f03 ] depicts a @xmath55-wall .
we may refer to an @xmath49-wall as a _
wall_. a wall is _ nonempty _ when @xmath44 is nonempty .
the _ clique size _ of a wall is the clique size of its graph .
a _ wall on @xmath56 colors _ is a wall with @xmath57 if a nonempty @xmath49-wall exists , then @xmath58 given @xmath49-wall @xmath48 where @xmath44 is the vertex set of @xmath0 , let @xmath18 be an enumeration of @xmath44 that increases in color , i.e. , @xmath59 whenever @xmath60 then the first - fit algorithm ( applied to this order ) assigns color @xmath61 to @xmath42 for each @xmath62 in particular , @xmath63 so @xmath64 by figure [ f03 ] . ] a given interval graph has many interval representations .
for example , if the transformation @xmath65 is applied to each interval of a representation , the graph structure is unchanged .
the same is true of @xmath66 and of every invertible affine transformation .
if we use intervals of finite length in our representations , we can move them by affine transformation into any given interval of positive length while preserving graph structure .
this is what we mean by _ squeezing _ an interval representation into an interval .
consider the sequence @xmath67 of walls depicted in figure [ f04 ] .
@xmath68 is a 1-vertex wall .
@xmath69 is the wall of figure [ f03 ] .
@xmath70 is the wall of figure [ f04_5 ] . for @xmath71 wall @xmath72
is made as in figure [ f04 ] by placing 4 new vertices in levels above 4 copies of @xmath73 .
the copies , whose tops are represented in figure [ f04 ] by wedge - like cones , have their interval representations squeezed into narrow intervals away from the junctions of the intervals of the 4 new top vertices , so that the clique size of the graph of @xmath72 is just 1 greater than that of the graph of @xmath74 wall @xmath72 has @xmath75 levels , and its graph has clique size @xmath76 , so @xmath77 is made by placing 4 intervals above 4 narrow copies of @xmath73 .
the construction yields @xmath78 we generalize this arrangement of new vertices above previously constructed walls ( as in figure [ f04 ] ) by defining a wall - like object called a cap .
this is wall @xmath79 let @xmath0 be a ( finite ) nonempty interval graph on vertex set @xmath44 with interval representation @xmath80 .
let @xmath81 be a coloring of @xmath0 .
( caps are built from the top down .
it may not be clear at first how many colors are needed .
so we use 0 for the top level and build the cap downward using negative integers . ) the remaining conditions for caps ( yet to be defined precisely ) are illustrated in figure [ f05 ] .
for each vertex @xmath42 : 1 .
empty space is reserved for a previously constructed wall @xmath82 .
this void is bounded horizontally by some interval @xmath83 vertically the void begins at some level @xmath84 below @xmath61 and extends downward without bound .
few cap vertices @xmath85 in levels above @xmath84 have intervals @xmath86 that meet @xmath87 3 .
each level between @xmath61 and @xmath84 contains a ( supporting ) cap vertex that neighbors @xmath42 . here
caps are defined precisely .
let @xmath88 suppose for each @xmath14 there is an interval @xmath89 of positive length ( reserving horizontal space for a wall ) where @xmath90.$ ] we denote the colors of vertices @xmath85 whose intervals @xmath86 meet @xmath91 by @xmath92 that is , @xmath93 .
a cap is used to build a sequence of walls that tends to @xmath49 in the ratio of number of colors to clique size of the graph .
space is reserved for a wall @xmath82 whose vertices will support @xmath42 when the cap is extended to a wall .
a wedge - shaped cone at the top of the void indicates where the top of @xmath82 will go ( but does not indicate how tall @xmath82 is ) .
vertex @xmath42 is supported by cap vertices in levels from its own ( @xmath61 ) down to the top of the void ( @xmath84 ) . above the void
is a sparse region . at most @xmath94 of the levels from cap top to cone top
have a vertex @xmath85 where @xmath86 meets @xmath87 ] if for each @xmath14 there is a color @xmath95 where : @xmath96 & \subseteq&f(n[v ] ) & \text{(\emph{support } ) ; } \end{array}\ ] ] and if @xmath97 for some @xmath14 , then @xmath98 is an
_ @xmath49-cap_. [ tcap ] let @xmath88 if an @xmath49-cap exists , then @xmath99 ; specifically , there is a constant @xmath100 so that for all @xmath101 there is a wall of clique size @xmath23 with at least @xmath102 colors . before the proof
, here are some lemmas .
[ cliquewall ] for @xmath103 there is a wall on @xmath23 colors with clique size @xmath104 consider a wall on @xmath23 vertices with constant interval representation @xmath105 $ ] .
[ walldrop ] if there is a wall on @xmath106 colors whose graph has no @xmath23-clique , then there is a wall on @xmath107 colors whose graph has no @xmath23-clique .
delete all vertices of the highest color . alternatively , delete all vertices of the lowest color and renumber colors .
let @xmath108 ( the depth of the deepest cone top ) .
let @xmath109 argue by induction on @xmath23 .
when @xmath110 , lemma [ cliquewall ] provides a wall .
this concludes the base case .
now suppose @xmath111 for the inductive step .
let @xmath42 be a cap vertex .
note that @xmath112 . by the emptiness condition , @xmath113.$ ] by the sparseness condition ,
@xmath114| \ge 1.$ ] and @xmath115 so @xmath116 and we may invoke the inductive hypothesis to obtain a wall @xmath82 of clique size @xmath117 with at least @xmath118 colors .
we squeeze the interval representation of the graph of wall @xmath82 into interval @xmath87 by the sparseness condition , this addition creates no clique on more than @xmath23 vertices . in the cap
, @xmath42 has neighbors of colors @xmath119 .
we extend @xmath22 to @xmath82 by resuming the sequence of colors .
we paused at @xmath120 .
we continue by recoloring @xmath82 from its top down with colors @xmath121 the sequence lasts until levels of @xmath82 are exhausted .
the sequence extends to @xmath122 or farther .
note that @xmath123 does not depend on @xmath42 .
after this is done for all cap vertices , each vertex has wall support down to color @xmath124 and some particular cap vertex @xmath42 with @xmath125 has neighbors colored @xmath126 vertices beyond ( that is , of color at most @xmath127 ) are dropped as in the proof of lemma [ walldrop ] .
now the colors used are @xmath128 walls are supposed to have positive colors , so we translate the palette , adding @xmath129 to each color used .
the result is a wall of clique size at most @xmath23 with at least @xmath102 colors .
we may increase the clique size to @xmath23 by placing a wall of lemma [ cliquewall ] on the side .
witsenhausen @xcite proved @xmath130 as did chrobak and lusarek @xcite , who drew a picture much like the 4-cap of figure [ f07 ] .
each wedge - shaped cone marks the top of a void reserved for a previously constructed wall .
a wall is placed in each void when the cap is extended to a wall by proposition [ tcap ] .
on the left are drawn tall gray bars , each corresponding to a hatched region ( sparse and empty combined ) of figure [ f05 ] .
a number @xmath131 inside a box indicates @xmath131 vertices with a common interval and @xmath131 consecutive colors . ] with figure [ f05 ] to remind us of cap conditions , let us verify that figure [ f07 ] depicts a 4-cap .
each box has a cone beneath it .
each cone has a void of positive width below .
when a vertical line segment is drawn from the top of the cap to the top of a cone , at most 25% of the ink covers boxes .
each box is supported by other boxes down to its cone .
so @xmath132 by proposition [ tcap ] .
we have verified that figure [ f07 ] depicts a 4-cap .
now let us see how this 4-cap was invented , toward improving on it .
how do we arrange boxes and cones to meet the cap conditions ?
the plan roughly is this : * a cap can not be empty , so start with a box . *
each box needs a cone beneath it , placed low due to the sparseness condition .
* when a cone is too low to support its box alone , add supporting boxes .
a cone divides the lower space into left and right , so add a left supporter , a right supporter , or both . *
keep the left and right halves of the cap similar . here
we execute the plan .
1 . twin boxes of height 1 ( _ @xmath53-tall _ ) go at the top .
one is marked ( a ) in figure [ f07 ] .
the rest of the cap is the same left and right .
we describe the right .
a cone marked ( b ) in figure [ f07 ] goes below the 1-tall box at depth @xmath133 , suited to the target ratio @xmath134 that is , we place the cone high in order to keep small the number of levels in which the 1-tall box above it requires support ; but not so high as to violate the sparseness condition .
in particular , if the cone is positioned @xmath135 units from the cap top , then the sparseness condition implies @xmath136 so we choose @xmath137 .
the 1-tall box still requires support in 2 levels .
this is addressed immediately .
3 . a 2-tall box is placed to fill the vertical gap between the bottom of box ( a ) and its cone ( b ) .
we are filling the gap to satisfy the cap support condition for intervals of box ( a ) .
a cone goes below box ( c ) at proper depth ( @xmath138 ) .
a gap of 4 colors remains between the new cone and the 2-tall box it supports .
5 . a 1-tall box ( and cone )
fits on the inside , leaving a gap of only 3 on the outside .
the idea is to distribute the support of box ( c ) between two others , boxes ( e ) and ( f ) , left and right of the cone under box ( c ) .
just left of box ( c ) , there is only 1 unit of `` weight '' above ( i.e. , box ( a ) ) , and yet we are at depth 8 , yielding the opportunity to add another unit of weight and still satisfy the sparseness condition .
a 3-tall box fills the gap .
a cone goes below .
a 5-gap remains ( between cone and 3-tall box ) .
a 1-tall box fits inside .
a 4-tall box fills the gap .
a cone goes below .
a 5-gap remains .
a 1-tall box fits inside .
a 4-tall box fills the gap . 13 .
a cone goes below .
a 1-gap remains .
a 1-tall box on the outside finishes the construction .
can we improve this result by the same procedure with @xmath139 ?
let us record the sequence @xmath140 of box heights along the outer strand , which for @xmath141 is @xmath142 .
the procedure halts because the last number does not exceed its predecessor . for @xmath139
the sequence is @xmath143 we can not complete the construction if the sequence increases forever .
does it ?
crucially , sequence @xmath144 obeys a linear recurrence .
this is depicted in figure [ f08 ] .
the recurrence follows from the procedure by a comparison of depths of cones under two consecutive boxes ( @xmath145-tall and @xmath146-tall ) of the outer strand .
when a box @xmath147 is placed on the inside to support the @xmath145-tall box , @xmath147 has no supporting box of its own .
so the cone under @xmath147 is at the same depth @xmath148 as the cone under the @xmath145-tall box . to satisfy the sparseness condition for @xmath149 the total height of boxes above the cone under @xmath147 must be at most @xmath150 and in the procedure we choose equality . above @xmath147
is a @xmath151-tall box .
so @xmath147 is @xmath152-tall .
similarly a @xmath153-tall box supports the @xmath146-tall box from the inside . comparing cone depths , @xmath154 so heights of boxes of the outer strand are given by the sequence @xmath155 for @xmath156 what determines the eventual behavior of such sequences ?
the discriminant @xmath157 of the characteristic polynomial @xmath158 of the recurrence is negative when @xmath141 and positive when @xmath159 as we will see when we refine the construction , such distinctions are important to sequence behavior .
meanwhile we may improve the bound @xmath132 by considering noninteger @xmath49 as lusarek @xcite did when he obtained @xmath160 observe that if each number in figure [ f07 ] is doubled , say , the result is again a cap .
therefore we may use ( temporarily ) rational numbers instead of integers for box heights and depths .
@xmath161 has a root @xmath162 the best we can do with the rational generalization of the method of chrobak and lusarek is @xmath163 .
( of course @xmath164 is irrational , but a sequence of rationals increasing to @xmath164 establishes @xmath165 )
in the previous section , our attempt to obtain a lower bound @xmath38 failed due to an unhalting procedure . yet
a certain resource was not fully exploited : the cap grew only 1 box inward .
that is , boxes of the outer strand had 1 or 2 supporting boxes , but inner boxes ( e ) , ( h ) , and ( k ) of figure [ f07 ] had no supporters .
we will enable more growth inward by a construction that generalizes the one of chrobak and lusarek .
let us reformulate the cap conditions as they apply to our construction .
1 . intervals are bundled into boxes , where a box of height @xmath131 represents identical intervals in @xmath131 consecutive levels .
distinct boxes meet pairwise in sets of area 0 .
intervals bundled in a common box share a cone .
emptiness _ : each box has a cone beneath it , representing a void for an older wall .
sparseness _ : each vertical line segment from cap top to cone top ( note : cones have positive width , so for a given cone , there are infinitely many such line segments ) meets boxes in portion at most @xmath94 of its length .
if @xmath135 is the distance of cone top from cap top , and @xmath166 ( for _ weight above _ the cone ) is the total length of the part that meets boxes , then the sparseness condition is @xmath167 3 . _
support _ : each box is supported by at most two others below it that cover the vertical extent from its bottom to its cone top .
depths of box features ( i.e. , tops and bottoms ) need not be integral , but are rational . they can be made integral by scaling later .
this defines a _
rational cap_. our construction is less direct than that of chrobak and lusarek in two ways .
first , there is no rational cap with @xmath139 ( see section [ nocap ] ) .
instead we construct a sequence of rational caps with @xmath168 second , each rational cap is made in a finite sequence of quasi - cap constructions .
the construction of the @xmath169-cap of chrobak and lusarek ( figure [ f07 ] ) begins with twin 1-tall boxes at the top , one supporting the other .
apart from the difference in vertical position of the twin 1-tall boxes , the left and right sides are mirror images .
consequently , the lower 1-tall box has at most one supporter .
this supporter ( if any ) is the _ key box _ , or the box in _
key position_. in figure [ f07 ] the key box is a 2-tall box labeled ( c ) .
such a rational cap ( with twin 1-tall boxes , symmetry , and at most 1 supporter of the lower twin ) is _ normal .
_ in a normal rational cap , the purpose of a key box is to support a 1-tall box , so the top of the key box meets the bottom of the 1-tall box .
a _ quasicap _ is a normal rational cap , except that a quasicap always has a key box , and the top of the key box may fall short , leaving a vertical gap ( as in figures [ f09 ] and [ f10 ] ) .
further , we require of a quasicap that apart from the twin 1-tall boxes , no box has its top higher than the top of the key box . -tall box in the key position ( bold ) . when @xmath170 there is a gap between this box and the 1-tall box it should support
the hatched portion appears in the next figure . ]
-tall box in the key position . pasted into place as shown
are distorted copies of the hatched portion of the rational cap from the previous figure .
( the main text defines the distortion and explains why the result is a quasicap . )
the goal is a quasicap with an @xmath171-tall box in the key position ( that is , a rational cap ) .
the sequence of boxes in the outer strand depends on our choice @xmath172 ( the size of our attempt to improve ) .
if @xmath172 is too large , then the sequence fails to terminate . ]
our goal is to make a rational cap from a sequence of quasicaps where the vertical gap shrinks to nothing .
figures [ f09 ] and [ f10 ] illustrate a gap - reducing step . from a quasicap with a @xmath173-tall key box
we make one with a @xmath174-tall key box .
outer - strand box heights of the new quasicap depend on @xmath173 and @xmath172 .
a linear recurrence defines those heights : @xmath175 for @xmath156 in the special case @xmath176 this definition agrees with the sequence of outer - strand box heights in the construction of chrobak and lusarek . that construction can be regarded as an instance of ours in the case of a single gap - reducing step from a trivial initial quasicap ( with a 1-tall box in the key position ) .
[ p3 ] fix some rational @xmath177 suppose there exists a quasicap with key box height @xmath178 ( henceforth a _ @xmath173-quasicap _ ) .
suppose also @xmath179 if there exist rational @xmath180 and integer @xmath181 so that the sequence @xmath182 defined above satisfies @xmath183 then a @xmath174-quasicap exists .
this result , proved in the many lemmas of this section , is a way to produce rational caps , provided that improvements @xmath172 can be made until the gap shrinks to 0 ( or equivalently , the key box grows to height @xmath184 ) .
here is the construction of proposition [ p3 ] .
the gap - reducing step is from an old quasicap ( figure [ f09 ] ) to a
_ new object _
( figure [ f10 ] ) which we will show to be a quasicap . 1 .
as in figure [ f10 ] , lay out an outer strand of boxes of heights @xmath185 down the right side .
the @xmath186-tall box has its bottom at depth @xmath49 , and for @xmath187 the @xmath188-tall box has its bottom at depth @xmath189 place the cone of the @xmath188-tall box at depth @xmath190 for @xmath191 place the cone of the @xmath192-tall box at depth @xmath193 2 .
paste into place under the outer strand ( see figure [ f10 ] ) copies @xmath194 of part of the old quasicap ( from figure [ f09 ] ) .
the part copied is only the right side of the old quasicap , excluding the twin 1-tall boxes
. these are not exact copies , but are stretched or squeezed by an amount calculated to satisfy the sparseness condition .
that the sparseness condition is satisfied is verified at the end of this section .
the stretching , squeezing , and moving of a given copy @xmath24 can be expressed in a cartesian coordinate system by an invertible affine plane transformation @xmath195 where 1 .
the horizontal component @xmath196 of the transformation , itself an invertible affine transformation , is strictly decreasing for @xmath197 and strictly increasing for @xmath198 , and 2 .
the vertical component @xmath199 is the invertible affine transformation @xmath200 for @xmath201 with @xmath202 where @xmath203 is depth measured from cap top .
+ function @xmath196 is continuous ; it keeps each box in one piece .
it is strict ; it preserves cap features .
when it decreases , features are reversed horizontally .
when it increases , they are not reversed .
we consider @xmath196 to be defined by figure [ f10 ] : the copy should be squeezed into an interval so narrow that the copy avoids colliding with cones or other parts of the cap ; and the image of the key box should be in place to support the @xmath204-high box of the outer strand .
copy @xmath24 is the image of the old quasicap part under transformation @xmath205 we claim without proof these facts about transformation @xmath206 : 1 .
the image of a box is a box , and 2 .
when two boxes meet in a set of area 0 , so do their images .
figure [ f10 ] suggests that @xmath207 let us prove it .
the hypothesis of proposition [ p3 ] implies @xmath208 .
@xmath209 [ imagebottom ] transformation @xmath199 maps the bottom of the old key box to depth @xmath210 in the new object for @xmath211 . in the old quasicap , the key box bottom is at depth @xmath49 .
therefore under transformation @xmath212 the new bottom is at depth @xmath213 the following lemma will be used twice in the proof of the lemma after it .
[ shiftscale ] for @xmath214 @xmath215 @xmath216 [ imagetop ] transformation @xmath199 maps the top of the old key box ( depth @xmath217 in the old quasicap ) to depth @xmath218 in the new object for @xmath219 by induction on @xmath220 in case @xmath221 , @xmath222 in case @xmath223 @xmath224 [ suppouter ] the @xmath204-tall box of the outer strand of the new object is supported for @xmath219 support of the @xmath204-tall box requires coverage of the vertical extent from its bottom ( depth @xmath225 ) to its cone top ( depth @xmath210 ) .
the neighboring @xmath226-tall box covers the vertical extent from depth @xmath227 to depth @xmath228 the image under @xmath199 of the old key box covers the rest , i.e. , the vertical extent from depth @xmath229 ( lemma [ imagetop ] ) to depth @xmath210 ( lemma [ imagebottom ] ) .
[ suppouterlast ] the @xmath192-tall box of the outer strand of the new object is supported . like the @xmath230- and @xmath231-tall boxes , and unlike the @xmath188-tall box for @xmath232 the @xmath192-tall box has only one supporting box .
the image under transformation @xmath233 of the old key box has its top at depth @xmath234 by lemma [ imagetop ] ( level with the bottom of the @xmath192-tall box by definition of the new object ) and its bottom at depth @xmath235 ( the depth of the cone under the @xmath192-tall box by definition of the new object ) .
[ nooverlap ] distinct boxes of the new object meet pairwise in sets of area 0 . for most box pairs , the conclusion is immediate : * 2 boxes from the outer strand : by definition of the new object * 2 boxes from a common copy : because the old object is a quasicap * 2 boxes from distinct copies : by definition of @xmath236 . in the remaining case ,
one box is from the outer strand , and the other is from copy @xmath237 .
no box top of copy @xmath24 is higher than the top of the key box of the copy , by definition of quasicap . in case
@xmath198 , this means no box of copy @xmath24 extends above the bottom of the @xmath192-tall box , as a consequence of the proof of lemma [ suppouterlast ] .
so no box of the copy meets the @xmath192-tall box in a set of positive area .
boxes of copy @xmath24 are kept away from the remaining boxes of the outer strand by definition of @xmath238 in cases @xmath201 no box of copy @xmath24 extends above the bottom of the @xmath226-tall box , as a consequence of the proof of lemma [ suppouter ] .
boxes of copy @xmath24 do not meet the @xmath188- or @xmath204-tall boxes , which are higher than the @xmath226-tall box by definition of the new object .
they are kept away from the remaining boxes of the outer strand by definition of @xmath196 . except for the twin 1-tall boxes ,
no box of the new object has its top higher than that of the key box .
this is a consequence of the proof of lemma [ nooverlap ] .
[ sparseouter ] the sparseness condition holds for each box of the outer strand .
the @xmath188-tall box of the outer strand , @xmath239 , has its cone placed at depth @xmath190 , and there are no other boxes between cap top and cone top .
the @xmath192-tall box has its cone at depth @xmath240 [ sparse ] the sparseness condition holds for each box in copy @xmath241 . the goal is to prove the sparseness condition ( as in the definition of rational cap ) for a box @xmath242 in copy @xmath24 .
box @xmath242 is the image under @xmath199 of a box @xmath147 in the old quasicap . in the old quasicap ,
let @xmath135 be the distance from cap top to cone top of @xmath147 , and let @xmath166 be the weight above its cone . in the new structure ,
let @xmath243 be the distance from cap top to cone top of @xmath242 , and let @xmath244 be the weight above its cone .
the old structure is a quasicap .
sparseness holds for @xmath147 , so @xmath245 .
now @xmath246\end{aligned}\ ] ] because @xmath247 now we compute weight above the cone of @xmath242 . first observe the effect of @xmath199 on a single box .
suppose a box in the old quasicap has top at depth @xmath203 and bottom at depth @xmath248 .
in the new structure , its image under @xmath199 has top at depth @xmath249 and bottom at depth @xmath250 the difference is @xmath251 that is , the image under @xmath199 is @xmath252 times taller than the original .
the weight @xmath244 above the cone of @xmath242 in the new structure is similarly related to @xmath166 , for all the boxes of the old quasicap except the twin ones appear in the new structure , stretched by the same factor .
there is also in the new structure a new box of the outer strand contributing weight above .
there are two kinds of box @xmath147 in the old quasicap to consider : those under the 1-tall box ( the _ 1 side _ in figure [ f09 ] ) , and those not ( the _ key side _ in figure [ f09 ] ) . in case @xmath147
is on the 1 side , all the weight above its cone except for 1 unit ( the twin ) is mapped into the new structure , scaled by a factor of @xmath253 and then a @xmath204-tall box is added .
that is , the weight above the cone of @xmath242 in the new structure is @xmath254 in case @xmath147 is on the key side , all the weight above its cone is mapped into the new structure , scaled by a factor of @xmath253 and then a @xmath255-tall box is added .
that is , the weight above the cone of @xmath242 in the new structure is @xmath256 same as when @xmath147 is on the 1 side . in both cases , @xmath257 sparseness holds .
[ sparsen ] the sparseness condition holds for each box in copy @xmath258 .
refer to figure [ f10 ] and compare copy @xmath258 to copy @xmath259 as @xmath260 distances from cap top to cone top are equal in corresponding cones of the two copies . now consider weight above .
it is the same in the two copies except for contributions of outer - strand boxes .
each cone of copy @xmath258 either has a @xmath192-tall outer - strand box above , while the corresponding cone of copy @xmath261 has a @xmath262-tall box ; or it has no outer - strand box above , while the corresponding cone has a @xmath263-tall box . because @xmath264 , in both cases there is no more weight above cones of copy @xmath258 than above corresponding cones of copy @xmath265 the proof of proposition [ p3 ] is complete .
new lower bounds on @xmath35 are available .
for example , let @xmath266 to start a sequence of quasicaps , we construct a @xmath53-quasicap .
this is easy .
we put a 1-tall box in the key position and a cone directly beneath it .
if @xmath267 , then sequence @xmath268 begins @xmath269 so by proposition [ p3 ] there exists a @xmath270-quasicap . if @xmath271 , then sequence @xmath268 begins @xmath272 so by proposition [ p3 ] there exists a @xmath273-quasicap
this is a @xmath274-cap , so by proposition [ tcap ] we have @xmath275 in the next section we show that the construction works for all @xmath276
sequence @xmath188 of section [ s3 ] is a linear homogeneous recursive sequence .
so is its sequence of differences .
the ordinary power series generating function @xmath277 of the difference sequence is ( cf .
chapter 4 of @xcite ) @xmath278 where @xmath279 is a polynomial whose coefficients are those of the recurrence , and @xmath280 is a polynomial of lesser degree related to boundary values of the sequence .
specifically , @xmath281\ ] ] and @xmath282 form gives names @xmath283 and @xmath284 to the ( complex ) roots of @xmath285 they affect the asymptotic behavior of the sequence .
when they are distinct , @xmath286 and @xmath287 for some complex numbers @xmath288 using , @xmath289 [ roots ] when @xmath290 we can assume @xmath291 and @xmath292 the first conclusion follows from the intermediate value theorem and : @xmath293 for the second conclusion , compare cubic terms of and : @xmath294 for the rest of the section we restrict @xmath49 and @xmath173 : @xmath295 @xmath296 so proposition [ roots ] applies .
this simplifies our calculations .
we may ignore @xmath297 in the asymptotic analysis because the corresponding term of vanishes .
[ realg ] if @xmath298 then @xmath299 and @xmath284 are real , and we can assume @xmath300 by the intermediate value theorem . using , @xmath301 and @xmath302 [ real ] if @xmath299 and @xmath284 are real , and @xmath303 then @xmath304 when @xmath305 in this case @xmath306 and the desired conclusion follows when @xmath307 . comparing two expressions of @xmath308 , @xmath309 & = & a(1-x/\beta)(1-x/\gamma ) + \label{coeffs}\\ & & b(1-x/\alpha)(1-x/\gamma ) + \notag \\ & & c(1-x/\alpha)(1-x/\beta ) .
\notag\end{aligned}\ ] ] substituting @xmath284 for @xmath310 @xmath311 = c(1-\gamma/\alpha)(1-\gamma/\beta).\ ] ] because @xmath312 we have @xmath307 when @xmath313 [ decreasing ] if @xmath298 then @xmath284 decreases strictly in @xmath49 , and so does @xmath314 suppose @xmath315 using for @xmath316 @xmath317 has roots @xmath318 by proposition [ realg ] .
when @xmath319 we have @xmath320 and @xmath321 adding the last two inequalities , @xmath322 specifically , @xmath323 . because @xmath324 for @xmath325\cup [ \beta_1,1],$ ] it follows that @xmath326 that is , @xmath284 decreases in @xmath327 the second conclusion follows .
[ pos ] if @xmath328 and @xmath329 then @xmath330 using , @xmath331 by proposition [ realg ] and the surrounding discussion , @xmath332 or @xmath333 the latter contradicts the hypothesis @xmath334 .
so @xmath335 [ last ] if @xmath336 then @xmath337 proposition [ decreasing ] implies that @xmath284 is least when @xmath139 .
evaluate @xmath338 there ( using ) to obtain a lower bound for @xmath284 : @xmath339 so @xmath340 and @xmath337 the discriminant @xmath161 of @xmath279 is ( cf .
95 - 102 of @xcite ) @xmath341 and @xmath342 if and only if @xmath343\text{im}[\beta]\ne 0.$ ] [ imaginary ] if @xmath344 then @xmath345 if @xmath346 then @xmath161 increases in @xmath49 : @xmath347 so @xmath161 is greatest when @xmath159 @xmath348 [ complex ] if @xmath342 and @xmath349 then @xmath350 for some @xmath351 in this case , is equivalent to : @xmath352,\ ] ] because formula for @xmath353 and @xmath354 is symmetric , and complex conjugation is a field automorphism .
clearly @xmath355 as @xmath356 the behavior of the other term @xmath357 $ ] can be understood with the help of the euler formula @xmath358 where the _ modulus _ @xmath359 of complex number @xmath360 is the distance in the complex plane of @xmath361 from @xmath362 , and the _ argument _
@xmath363 of @xmath361 is an angle in the complex plane from the positive real axis to the ray emanating from @xmath362 to @xmath361 .
complex multiplication is multiplicative in modulus and additive in argument . because @xmath364 as @xmath365 the desired conclusion holds if the ray emanating from @xmath362 to @xmath366 is near the negative real axis for some @xmath24 large enough that the second term @xmath357 $ ] of the sum is dominant .
indeed , @xmath357 $ ] oscillates in sign .
because @xmath367 eventually @xmath368 say , while @xmath357<-0.1 $ ] infinitely often ( cf .
exercise ii.1.11 of @xcite ) .
[ fivelter ] @xmath369 fix some rational @xmath49 with @xmath370 let @xmath371 be the set of @xmath173 such that a @xmath173-quasicap exists . clearly @xmath372 recall that @xmath284 depends on @xmath49 and @xmath373 though @xmath49 is fixed , @xmath173 is free , so @xmath284 is a function of @xmath373 let @xmath374 then @xmath375 is closed ( in the usual topology on the reals ) because it is the continuous image of a compact set .
propositions [ decreasing ] and [ pos ] imply @xmath376 fix some rational @xmath172 with @xmath377 so that @xmath378 is an integer .
propositions [ roots ] , [ realg ] , and [ real ] imply @xmath379 for all @xmath380.$ ] invoking proposition [ p3 ] many ( i.e. , @xmath381 ) times yields @xmath382 propositions [ p3 ] and [ last ] imply @xmath383 because @xmath384 propositions [ p3 ] , [ imaginary ] , and [ complex ] imply @xmath385 that is , a rational @xmath49-cap exists .
proposition [ tcap ] implies @xmath369 for some bad pairs @xmath386 no positive @xmath172 is small enough for proposition [ p3 ] . in the present section
, we worked to avoid such sequences while getting as close as possible .
the curve @xmath387 with @xmath388 in the @xmath389-plane ( see figure [ extreme ] ) is an important part of the boundary of the bad region .
what is the least @xmath49 on this curve ?
we aim to express @xmath49 as a function of @xmath173 and optimize .
the curve is defined @xmath390 so @xmath391 now @xmath392 and the minimum value of @xmath49 on the curve occurs when @xmath393 and @xmath159 the @xmath49-minimal bad sequence @xmath188 is the one with @xmath394 .
if @xmath395 then @xmath188 is the fibonacci sequence . ( 80,98)(-35,-48 ) ( 10,20)(-10,0)(10,-20 ) ( -40,-40)(1,0)93 ( -40,-40)(0,1)83 ( -40,-40)(0,40)3(-1,0)1 ( -40,-40)(40,0)3(0,-1)1 ( -46,40)(0,0)@xmath396 ( -46,0)(0,0)@xmath397 ( -46,-40)(0,0)@xmath398 ( -42,47)(0,0)@xmath173 ( -40,-44)(0,0)@xmath399 ( 0,-44)(0,0)@xmath400 ( 40,-44)(0,0)@xmath401 ( 56,-44)(0,0)@xmath49 ( -7,15)(0,0)@xmath402 ( 18,15)(0,0)@xmath403 ( 42,0)(0,0)fibonacci sequence ( 23,0)(-1,0)20 ( 0,0 ) ( -40,0)(1,0)40 ( 0,-40)(0,1)40 ( 14,-11)(0,0)bad ( 22,-15)(0,0)sequences
caps of section [ s3 ] are like ordered binary trees ( see figure [ f12 ] ) : 1 .
a cap has at least one box .
each box but one ( the _ top _ box ) supports a unique box .
each box has at most 1 left supporter and at most 1 right supporter .
4 . each box is a member of the set @xmath147 of boxes that are either the top box or a supporter of a member of @xmath147 .
we name each box of binary cap @xmath354 by a binary word , as in figure [ f12 ] .
box @xmath166 may have a high supporter , its _ high child _
@xmath404 ; and a low supporter , its _ low child _ @xmath405 .
accordingly @xmath166 is the _ parent _ of @xmath404 and @xmath405 .
indeed , we may describe the whole binary cap @xmath354 by a function @xmath406 where @xmath354 has a box @xmath407 if and only if @xmath408 . the word of length @xmath362 is denoted @xmath409 we assume @xmath408 whenever @xmath410 or @xmath411 .
binary caps are nonempty , and in particular , @xmath412 let @xmath413 let @xmath414 denote the depth of the top of the box @xmath166 .
let @xmath415 denote the weight above the cone of box @xmath166 minus the height of box @xmath166 itself .
let @xmath416 denote the weight above the cone of the parent of @xmath166 .
let @xmath417 ( the _ penalty _ of @xmath166 ) be @xmath418 let @xmath419 where @xmath420 ( the fibonacci recurrence ) for @xmath421 suppose @xmath443 is a binary @xmath442-cap .
the top box @xmath444 is @xmath362-hard as @xmath445 and @xmath446 .
since caps are finite , there is a hard box @xmath166 with maximum word length . for a contradiction
, we show that there exists @xmath447 such that @xmath448 is hard and @xmath449 ( so @xmath448 is in @xmath443 ) . | first - fit is the online graph coloring algorithm that considers vertices one at a time in some order and assigns each vertex the least positive integer not used already on a neighbor .
the maximum number of colors used by first - fit on graph @xmath0 over all vertex orders is denoted @xmath1 the exact value of @xmath2 over interval graphs @xmath0 is unknown .
pemmaraju , raman , and varadarajan ( 2004 ) proved @xmath3 , and this can be improved to @xmath4 witsenhausen ( 1976 ) and chrobak and lusarek ( 1988 ) showed @xmath5 , and lusarek ( 1993 ) improved this to @xmath6 we prove @xmath7 * key words : * graph coloring , online algorithm , first - fit , interval graph . |
jordan s inequality , @xmath3 has numerous extensions and generalizations .
( to simplify the notation we agree that for @xmath4 the value of @xmath5 is 1 ) . an exhaustive review of the literature is available in @xcite .
one of the many types of generalizations is a sequence of the two sided inequalities .
the first of them is the inequality @xmath6 by ling zhu @xcite .
it is followed by @xmath7 by @xcite ; finally ling zhu verifies @xcite the general @xmath8 where @xmath9 are given by a recursion formula .
see also ( * ? ? ?
* section 3.7 ) . later ( [ eq : j - n ] ) was extended in @xcite to @xmath10 for @xmath11 .
the proofs of inequalities ( [ eq : j-1])([eq : j - r ] ) are based on a monotone version of the lhospital rule : _
if @xmath12 and @xmath13 is increasing on @xmath14 , so is the quotient @xmath15_. the two sided inequalities ( [ eq : j - n ] ) are closely related to the infinite expansion of @xcite @xmath16 where @xmath17 and @xmath18 .
see also ( * ? ? ?
* section 2.3 ) . identity ( [ eq : j - series ] ) is verified by a straightforward rearrangement of a double power series . by direct integration
one achieves also @xmath19 similar ideas were developed for bessel functions in a paper by baricz and wu @xcite and are summarized in @xcite . in @xcite
the author uses the notation @xmath20 , normalized so that @xmath21 . without normalization the result of baricz and wu
are formulated as @xmath22 for all @xmath23 and for @xmath24 , where @xmath25 and @xmath26 denotes the first positive zero of @xmath27 .
the aim of this study is to develop similar ideas for the the trigonometric functions which have singularities , namely , the @xmath28 and @xmath29 functions .
our attention was drawn to this subject in an attempt to refine the inequality of becker and stark @xcite @xmath30 our main result ( theorem 1 ) is that the function @xmath31 admits a sequence of upper and lower bounds and these inequalities are closely related to a laurent series expansion in powers of @xmath1 .
the same approach will be applied also to @xmath32 and @xmath29 . in spite of the similarity
, there are several essential differences between our result and those of ling zhu and for @xmath5 and of baricz for the bessel functions .
+ ( a ) bessel functions and , consequently @xmath33 , are solutions of simple linear second order differential equations .
this is a useful tool in @xcite .
unfortunately , we do not know any simple linear differential equation for the tan function . on the contrary ,
the results for @xmath34 are influenced by its poles at @xmath2 .
+ ( b ) we start our study with the infinite series and the upper and lower bounds appear as consequences of this alternating series .
we do not use the monotone lhospital rule .
our method uses partial fractions expansions . +
in analogy with ( [ eq : j - r ] ) and ( [ eq : baricz ] ) we also expand @xmath35 into a laurent series of @xmath36 , with @xmath37 .
this expansion has different properties from that which corresponds to @xmath38 .
in this section we present an infinite series expansion for the tan function and two sets of upper and lower bounds , one with an even number of terms and one with odd number of terms .
[ thm : tan ] ( a ) we have the infinite expansion @xmath39 , \ ] ] where @xmath40 } \binom{2p-2m-1}{p-1 } \zeta(2 m ) - \binom{2p-1}{p-1 } \right ] , \qquad p=1,2,\ldots \ .\ ] ] all the coefficients @xmath41 are positive . the infinite series in ( [ eq : tan - laurent ] ) converges for @xmath42 . ' '' '' + ( b ) for all even number @xmath43 and @xmath44
, we have the pair of inequalities @xmath45 \qquad\qquad\qquad\qquad \\ & \qquad\qquad\qquad\qquad \le \ \frac { \tan ( \pi x / 2 ) } { \pi x / 2 } \
\le \ \frac { 8}{\pi^2 } \left [ \frac{1 } { 1 - x^2 } + \sum _ { k=0 } ^{2n } ( -1)^k \frac { t_{k+1 } } { 4^{k+1 } } ( 1-x^2)^k \right ] , \end{aligned}\ ] ] where @xmath46 and @xmath47
. the left hand side inequality of ( [ eq : tan-2n ] ) is strict for @xmath4 .
+ ( c ) ' '' '' for all odd number @xmath48 and @xmath44 , we have @xmath49 \ \le \ \frac { \tan ( \pi x / 2 ) } { \pi x / 2 } \qquad\qquad \\ & \qquad\qquad \le \ \frac { 8}{\pi^2 } \left [ \frac{1 } { 1 - x^2 } + \sum _ { k=0 } ^{2n } ( -1)^k \frac { t_{k+1 } } { 4^{k+1 } } ( 1-x^2)^k - h_{2n+2 } ( 1-x^2)^{2n+1 } \right ] , \end{aligned}\ ] ] where @xmath50 and @xmath51 . the right hand side inequality of ( [ eq : tan-2n+1 ] ) is strict for @xmath4 .
* examples . *
the first coefficients are @xmath52 , @xmath53 and @xmath54 .
accordingly , @xmath55 the first two special cases of ( [ eq : tan-2n ] ) ( for @xmath56 ) and ( [ eq : tan-2n+1 ] ) ( for @xmath57 ) are @xmath58 and @xmath59 * proof of theorem [ thm : tan ] * for sake of convenience we divide the proof into several steps . _
( i ) the infinite expansion for tan . _
our starting point is the partial fraction expansion @xmath60 which holds in every domain which contains no poles .
it follows that @xmath61 .\ ] ] each term of the series may be expanded into a geometric series , provided that @xmath62 for @xmath63 , i.e. , for @xmath64 .
this yields @xmath65 \\ = & \frac { 8}{\pi^2 } \left [ \frac{1 } { 1 - x^2 } + \sum _ { k=0 } ^\infty \left ( \sum _ { n=1 } ^\infty \frac{1 } { n^{k+1 } ( n+1)^{k+1 } } \right ) \frac { ( -1)^k } { 4^{k+1 } } ( 1-x^2)^k \right ] .
\end{aligned}\ ] ] the expansion ( [ eq : tan - laurent ] ) will follow if we show that the sums @xmath66 are given explicitly by ( [ eq : tp - zeta ] ) for @xmath67 .
the proof is based on the following identity ( * ? ? ? * ex .
39 , p. 246 ) : _ if @xmath68 , then for every _ @xmath69 , @xmath70 ( [ eq : xm - yk ] ) may be verified by mathematical induction on @xmath71 and @xmath72 .
we take in ( [ eq : xm - yk ] ) @xmath73 , @xmath74 , @xmath75 , and conclude that @xmath76 .\ ] ] to calculate @xmath41 we sum ( [ eq : x=-1/n ] ) for @xmath77 and separate @xmath78 from @xmath79 : @xmath80 \\ & = \binom{2p-2 } { p-1 } \sum_{n=1}^\infty \left [ -\frac{1}{n } + \frac{1}{n+1 } \right ] + \sum_{j=2}^p \binom{2p - j-1 } { p-1 } \big [ ( -1)^{j } \zeta(j ) + \big ( \zeta(j ) - 1 \big ) \big ] \\ & = \binom{2p-2}{p-1 } \cdot ( -1 ) + \sum _ { j = 2 } ^ { p } \binom{2p - j-1}{p-1 } \big [ \big ( ( -1)^{j } + 1 \big ) \zeta(j ) - 1 \big ) \big ] \\ & = 2 \sum _ { 2 \le 2 m \le p } \binom{2p-2m-1}{p-1 } \zeta(2 m ) - \sum_{j=1}^{p } \binom{2p - j-1}{p-1 } .
\end{aligned}\ ] ] since @xmath81 , formula ( [ eq : tp - zeta ] ) is verified and part ( a ) of the theorem follows .
we remark that similar sums @xmath82 are mentioned in ( * ? ? ? * , ( 373 ) ) for @xmath83 and @xmath84 . _
( ii ) the inequalities . _
let us write the series ( [ eq : tan - laurent ] ) as @xmath85 , \ ] ] with the remainder term @xmath86 for each @xmath87 , @xmath88 , the series @xmath89 and its derivative series @xmath90 are alternating series with smaller and smaller terms .
indeed , @xmath91 thus @xmath92 and similarly @xmath93 thus , by properties of alternating series , @xmath94 and @xmath95 are positive functions for @xmath88 . consequently @xmath94 increases for @xmath88 , and is bounded from below and from above by @xmath96 the lower bound of @xmath94 , @xmath97 may be written as a finite sum : for @xmath4 , ( [ eq : tan - laurent ] ) yields @xmath98 , \ ] ] so @xmath99 .
\end{aligned}\ ] ] consequently @xmath100 is bounded for @xmath101 and for @xmath102 , respectively , by @xmath103 and parts ( b ) and ( c ) of the theorem follow .
the remainder term satisfies @xmath104 hence it decreases rapidly , in particular near the poles @xmath105 .
for example , for @xmath106 and @xmath107 we have @xmath108 .
the bounds in ( [ eq : tan-2n+1 ] ) and ( [ eq : tan-2n ] ) are also very precise near the poles @xmath2 . indeed ,
the difference between the @xmath71-th upper and lower bounds ( @xmath109 odd or even ) is less then @xmath110 .
it is interestiong to compare the convergence of the series ( [ eq : tan - laurent ] ) with that of the taylor series ( * ? ? ?
* section 3:14 ) @xmath111 where @xmath112 denotes the dirichlet lambda function . since @xmath113
, the remainder of the series ( [ eq : tan - lambda ] ) after @xmath109 terms is @xmath114 here @xmath115 is uniformly small for large values of @xmath109 . on the other hand ,
we saw above that the remainder of the series ( [ eq : tan - laurent ] ) after @xmath109 terms is @xmath116 . after extracting @xmath109-th root from the two remainders for large @xmath109
, we compare , up to @xmath117 , @xmath118 with @xmath119 .
the consequence is that after @xmath120 terms , for large @xmath109 , the taylor series ( [ eq : tan - lambda ] ) yields a smaller remainder for @xmath121 , while for @xmath122 , the series ( [ eq : tan - laurent ] ) has a smaller remainder .
@xmath123 due to ( [ eq : j - r ] ) and ( [ eq : baricz ] ) it is also natural to expand @xmath124 into power series of @xmath36 .
formally we follow ( [ eq : tan - fractions ] ) and ( [ eq : tan ( pi x/2 ) ] ) : @xmath125 .
\\ = & \frac { 8}{\pi^2 } \sum _ { k=0 } ^\infty ( -1)^k \frac { \tilde{t}_{k+1}(r ) } { 4^{k+1 } } ( r^2-x^2)^k \ , , \end{aligned}\ ] ] with @xmath126 note that @xmath127 differs from @xmath41 by it s range of summation and it is not defined for @xmath38 .
equations ( [ eq : tan - r2-x2 ) ] ) and ( [ eq : tan ( pi x/2 ) ] ) are superficially similar
. however , they describe different phenomena .
( [ eq : tan ( pi x/2 ) ] ) describes the function @xmath128 in the interval @xmath129 , including its singularities at @xmath2 .
expansion ( [ eq : tan - r2-x2 ) ] ) is valid when @xmath130 for all @xmath131 .
assuming @xmath132 , this holds when @xmath133 , i.e. , @xmath134 which includes the points @xmath135 and their neighborhoods .
note that in the complex plane @xmath136 is the interior of a lemniscate with focal points @xmath137 .
inequalities analogous to ( [ eq : tan-2n ] ) and ( [ eq : tan-2n+1 ] ) may be deduced from expansion ( [ eq : tan - r2-x2 ) ] ) .
we do not attempt to calculate the sums ( [ eq : t - p - r ] ) explicitly as functions of @xmath138 .
rather we shall obtain a recursive formula for the coefficients @xmath127 . the function @xmath139 satisfies the first order , quadratic riccati differential equation @xmath140 .
it easy to see that @xmath141 satisfies @xmath142 for @xmath143 .
if we let @xmath144 and @xmath145 , then @xmath146 satisfies @xmath147 for @xmath148 .
substitution of @xmath149 , @xmath150 and comparison of coefficients lead to the following quadratic recursion formula for @xmath151 , @xmath37 : @xmath152
expansions and inequalities of similar type are available for the @xmath32 and @xmath29 functions . in this section we present the results for the secant function .
we remind , in analogy with inequalities ( [ eq : beckerstark ] ) and ( [ eq : chen - qi - cot ] ) , that @xmath153 to prove ( [ eq : sec - beckerstark ] ) , it is sufficient to show that the even function @xmath154 increases for @xmath88 .
this indeed holds , since by ( [ eq : beckerstark ] ) @xmath155 \ge 0 .\ ] ] the constants in ( [ eq : sec - beckerstark ] ) follow now by the limit values for @xmath4 and as @xmath156 .
our aim is to generalize the right hand side of ( [ eq : sec - beckerstark ] ) into an infinite sequence of inequalities .
[ thm : sec ] ( a ) we have the infinite expansion @xmath157 , \ ] ] where @xmath158 \eta(2 m ) \right]\ ] ] for @xmath159 and @xmath160 denotes the dirichlet eta function .
all the coefficients @xmath161 are positive . ' '' '' + ( b ) for all even number @xmath43 and @xmath162 , we have the pair of inequalities @xmath163 \le \ \sec ( \pi x /2 ) \\ & \qquad\qquad \le \ \frac{4}{\pi } \left [ \frac{1 } { 1 - x^2 } - \sum _ { k=0 } ^{2n-1 } ( -1)^k \frac { s_{k+1 } } { 4^{k+1 } } ( 1-x^2)^k - j_{2n+1 } ( 1-x^2)^{2n } \right ] , \end{aligned}\ ] ] where @xmath164 and @xmath165 .
the right hand side inequality of ( [ eq : sec-2n ] ) is strict for @xmath4 .
+ ( c ) ' '' '' for all odd number @xmath48 and @xmath166 , we have @xmath167 \\ &
\qquad\qquad \le \ \sec ( \pi x /2 ) \le \
\frac{4}{\pi } \left [ \frac{1 } { 1 - x^2 } - \sum _ { k=0 } ^{2n+1 } ( -1)^k \frac { s_{k+1 } } { 4^{k+1 } } ( 1-x^2)^k \right ] , \end{aligned}\ ] ] where @xmath168 and @xmath169 . the left hand side inequality of ( [ eq : sec-2n+1 ] ) is strict for @xmath4 .
* examples . *
the first coefficients are @xmath170 , @xmath171 and @xmath172 .
accordingly , @xmath173 for @xmath56 we get on @xmath166 @xmath174 \le \sec ( \pi x / 2 ) \le \frac { 4}{\pi } \left [ \frac{1 } { 1 - x^2 } - 1 + \frac{\pi}{4 } \right]\ ] ] and for @xmath175 , @xmath176 \le \sec ( \pi x / 2 ) \le \frac { 4}{\pi } \left [ \frac{1 } { 1 - x^2 } - \frac{1}{4 } + \frac{\pi^2 -6}{96 } ( 1 - x^2 ) \right ] \ .\ ] ] * proof . * we start with the partial fraction expansion @xmath177 .
then for @xmath166 , @xmath178 \\ & = \frac { 4}{\pi } \left [ \frac{1 } { 1 - x^2 } - \sum _ { k=0 } ^\infty \left ( \sum _ { n=1 } ^\infty \frac { ( -1)^{n-1 } ( 2n+1 ) } { n^{k+1 } ( n+1)^{k+1 } } \right ) ( -1)^k \frac { ( 1-x^2)^k } { 4^{k+1 } } \right ] \\ & = \frac { 4}{\pi } \left [ \frac{1 } { 1 - x^2 } - \sum _ { k=0 } ^\infty ( -1)^k \frac { s_{k+1 } } { 4^{k+1 } } ( 1-x^2)^k \right ] , \end{aligned}\ ] ] where @xmath179 for @xmath180 . the coefficients @xmath161 may be presented explicitly by finite sums of the dirichlet @xmath160 function .
let us multiply ( [ eq : x=-1/n ] ) by @xmath181 and separate @xmath78 from @xmath182 : @xmath183 \\ & + \sum_{j=2}^p \binom{2p - j-1 } { p-1 } \left [ 2\frac { ( -1)^j } { n^{j-1 } } + \frac { ( -1)^j } { n^j } + 2\frac { 1 } { ( n+1)^{j-1 } } - \frac { 1 } { ( n+1)^j } \right ] . \end{aligned}\ ] ] we multiply the last equation by @xmath184 and sum it for @xmath185 . with the dirichlet eta function
@xmath186 we get @xmath187 \\ & + \sum _ { j=2 } ^p \binom{2p - j-1 } { p-1 } \big [ 2(-1)^j \eta(j-1 ) + ( -1)^j \eta(j ) + 2 \big(-\eta(j-1 ) + 1 \big ) - \big(-\eta(j ) + 1 \big ) \big ] \\ & = -\binom{2p-2 } { p-1 } + \sum _ { j=2 } ^p \binom{2p - j-1 } { p-1 } + \sum _ { j=2 } ^p \binom{2p - j-1 } { p-1 } \big [ 2 \big ( ( -1)^j - 1 \big ) \eta(j-1 ) + \big ( ( -1)^j + 1 \big ) \eta(j ) \big ] \\ & = - \binom{2p-2 } { p-1 } + \binom{2p-2 } { p } + \sum _ { 2 \le 2 m \le p } 2 \big [ \binom{2p-2m-1 } { p-1 } - 2 \binom{2p-2m-2 } { p-1 } \big ] \eta(2 m ) \\ & = \binom{2p-2 } { p-2 } - \binom{2p-2 } { p-1 } + 2 \sum _ { 2 \le 2 m \le p } \big [ \binom{2p-2m-2 } { p-2 } - \binom{2p-2m-2 } { p-1 } \big ] \eta(2 m ) .
\end{aligned}\ ] ] here we used the facts that @xmath188 and that @xmath189 thus , ( [ eq : sp - eta ] ) is proved . the inequalities ( [ eq : sec-2n ] ) and ( [ eq : sec-2n+1 ] ) are verified as in theorem [ thm : tan ] .
@xmath123 the coefficients @xmath41 and @xmath161 are easily related . in the ( [ eq : sec - derivative ] ) we substitute @xmath190 and @xmath191 , respectively , by the series ( [ eq : tan - laurent ] ) and ( [ eq : sec - laurent ] ) and compare powers of @xmath1 , the result is @xmath192
we outline some other results which may be verified by similar arguments . for the cotangent function ,
let us recall the inequality ( * ? ? ?
* eq . ( 17 ) ) @xmath193 it is possible generalize the left hand side of ( [ eq : chen - qi - cot ] ) into the infinite expansion @xmath194 \ , \ ] ] @xmath195 , \quad p \ge 1 .
\end{aligned}\ ] ] all @xmath196-s are positive .
the infinite series in ( [ eq : cot - laurent ] ) converges for @xmath197 .
for example , @xmath198 .
inequalities in the style of theorems [ thm : tan ] and [ thm : sec ] may be proved for @xmath199 .
( [ eq : cot - laurent ] ) is obtained from the expansion @xmath200 . for the calculation of @xmath196 in ( [ eq : cp - zeta ] )
, it is useful to take in ( [ eq : xm - yk ] ) , @xmath73 , @xmath201 and sum the result for @xmath202 . for the summation of the series
, one has to use the binomial identity @xmath203 ( * ? ? ?
* , eq . 5.20 ) .
@xmath123 for the cosecant function we remind , in analogy with ( [ eq : beckerstark ] ) , ( [ eq : sec - beckerstark ] ) and ( [ eq : chen - qi - cot ] ) , the double inequality ( * ? ? ?
* eq . ( 19 ) ) @xmath204 the right hand side of ( [ eq : chen - qi - cosec ] ) may be generalized into the expansion @xmath205 , \ ] ] where @xmath206 .
\end{aligned}\ ] ] here @xmath207 .
expansion ( [ eq : cosec - laurent ] ) follows from the partial fraction expansion @xmath208 .
@xmath209 is calculated similarly to @xmath196 .
@xmath123 if we substitute the series ( [ eq : cot - laurent ] ) and ( [ eq : cosec - laurent ] ) into the identity @xmath210\ ] ] and compare powers of @xmath211 , it follows that @xmath212 and @xmath213 are related by @xmath214
finally we show that our methods are applicable also to bessel functions and they provide a short proof of ( [ eq : baricz ] ) , based on other principles than those of baricz and wu in @xcite .
an expansion of the form @xmath215 holds for all @xmath87 in the complex plane . to see this ,
put @xmath144 , i.e. , @xmath216 . since @xmath217 is an even analytic function in the whole complex plane
, it follows that @xmath218 is analytic for all @xmath219 and may be expanded as @xmath220 . consequently ( [ r2-x2-expansion ] ) holds for all @xmath87 . to calculate the @xmath221-s ,
let us apply the identity ( * ? ? ?
9.1.30 ) @xmath222 to expansion ( [ r2-x2-expansion ] ) .
since @xmath223 , we get @xmath224 for @xmath225 we obtain that @xmath226 . consequently , @xmath227 to prove the inequalities ( [ eq : baricz ] ) , we have to estimate the remainder of the expansion ( [ baritz - expansion ] ) , namely @xmath228 let @xmath229 denote the first positive zero of @xmath230 .
it is well known that @xmath231 and that @xmath232 in @xmath233 .
see figure 1 .
if @xmath236 satisfies @xmath237 , all the coefficients in the tail series @xmath238 are positive for all @xmath23 , hence @xmath239 .
so , in the left hand side of ( [ eq : baricz ] ) , @xmath240 ; and on the right hand side of ( [ eq : baricz ] ) , @xmath241 @xmath242 may be written as a finite sum .
as @xmath243 , we get from ( [ baritz - expansion ] ) for @xmath4 that @xmath244 thus , @xmath245 | the function @xmath0 is expanded into a laurent series of @xmath1 , where the coefficients are given explicitly as combinations of zeta function of even integers .
this is used to achieve a sequence of upper and lower bounds which are very precise even at the poles at @xmath2 .
similar results are obtained for other trigonometric functions with poles . * keywords * : trigonometric inequalities , jordan inequality , laurent series , zeta function , dirichlet functions + * mathematics subject classification ( 2010 ) * : 26d05 |
accurate measurements of the machine tune and of the chromaticity are of importance for the operation of fast ramping , high intensity ion synchrotrons . in such machines
the tune spread @xmath0 at injection energy due to space charge and chromaticity can reach values as large as 0.5 . in order to limit the incoherent particle tunes to the resonance free region the machine tune has be controlled with a precision better than @xmath1 . in the gsi heavy - ion synchrotron sis-18 there are currently two betatron tune measurement systems installed .
the frequency resolution requirements of the systems during acceleration are specified as @xmath2 , but they provide much higher resolution ( @xmath3 ) on injection and extraction plateaus .
the tune , orbit and position measurement system ( topos ) is primarily a digital position measurement system which calculates the tune from the measured position @xcite .
the baseband q measurement system ( bbq ) conceived at cern performs a tune measurement based on the concept of diode based bunch envelope detection @xcite .
the bbq system provides a higher measurement sensitivity than the topos system .
passive tune measurements require high sensitivity ( schottky ) pick - ups , low noise electronics and long averaging time to achieve reasonable signal - to - noise ratio .
for fast tune measurements with the topos and bbq systems , which both use standard pick - ups , the beam has to be excited externally in order to measure the transverse beam signals . for low intensities the theory of transverse signals from bunched beams and the tune measurement principles from schottky or externally excited signals are well known @xcite .
in intense , low energy bunches the transverse signals and the tune spectra can be modified significantly by the transverse space charge force and by ring impedances .
previously , the effect of space charge on head - tail modes had been the subject of several analytical and simulation studies @xcite .
recently , the modification of transverse signals from high intensity bunches was observed in the sis-18 for periodically excited beams @xcite and for initially kicked bunches @xcite , where the modified spectra was explained in terms of the space charge induced head - tail mode shifts .
this contribution aims to complement the previous studies and extract the relevant intensity parameters from tune spectra measurements using the topos and bbq tune measurement systems . presents the frequency content of transverse bunched beam signals briefly .
presents the space charge and image current effects on tune measurements and respective theoretical models .
report on the experimental conditions and compares the characteristics of two installations as well as the various excitation methods . presents the experimental results in comparison with the theoretical estimates of various high intensity effects
theoretical and experimental work related to transverse schottky signals and beam transfer functions ( btfs ) for bunched beam at low intensities can be found in the existing literature @xcite .
the transverse signal of a beam is generated by the beam s dipole moment @xmath4 where @xmath5 is the horizontal / vertical offset and @xmath6 the current of the @xmath7 particle at the position of a pick - up ( pu ) in the ring .
the sum extends over all @xmath8 particles in the detector .
the schottky noise power spectrum as a function of the frequency @xmath9 is defined as @xmath10 where @xmath11 is the fourier transformed pu signal . for a beam excited by an external amplitude spectrum @xmath12 , the transverse response function is defined as @xmath13 @xcite . if the transverse signal from a low intensity bunch is sampled with the revolution period @xmath14 , then the positive frequency spectrum consists of one set of equidistant lines @xmath15 usually defined as baseband tune spectrum , where @xmath16 is the fractional part of the machine tune , @xmath17 are the synchrotron satellites and @xmath18 is the synchrotron tune . for a single particle performing betatron and synchrotron oscillations , the relative amplitudes of the satellites are @xcite @xmath19 where @xmath20 is the chromatic phase , @xmath21 is the chromaticity
, @xmath22 is the longitudinal oscillation amplitude of the particle and @xmath23 the frequency slip factor .
@xmath24 are the bessel functions of order @xmath25 . in bunched beams ,
the relative height and width of the lines depends on the bunch distribution .
the relative height is also affected by the characteristics of the external noise excitation or by the initial transverse perturbation applied to the bunch . in the absence of transverse nonlinear field components the width of each satellite with @xmath26
is determined by the synchrotron tune spread @xmath27 . an example tune spectrum obtained from a simulation code for a gaussian bunch distribution
is shown in .
baseband tune spectrum from a simulation code for a low intensity bunch.,width=321 ]
at high beam intensities , the transverse space charge force together with the coherent force caused by the beam pipe impedance will affect the motion of the beam particles and also the tune spectrum .
the space charge force induces an incoherent tune shift @xmath28 for a symmetric beam profile of homogeneous density where @xmath29 is the tune shift , @xmath30 the bunch peak current , @xmath31 the particle charge and @xmath32 the total energy .
the relativistic parameters are @xmath33 and @xmath34 , the ring radius is @xmath35 and the emittance of the rms equivalent k - v distribution is @xmath36 . in the case of an elliptic transverse cross - section the emittance @xmath36 in should be replaced by @xmath37 for the vertical plane the procedure is the same , with @xmath38 replaced by @xmath39 .
the image currents and image charges induced in the beam pipe , assumed here to be perfectly conducting , cause a purely imaginary horizontal impedance @xmath40 and real coherent tune shift @xmath41 for a round beam profile with radius @xmath42 and pipe radius @xmath43 the coherent tune shift is smaller by @xmath44 than the space charge tune shift .
therefore the contribution of the pipe is especially important for thick beams @xmath45 at low or medium energies . in the presence of incoherent space charge , represented by the tune shift @xmath46 , or pipe effects , represented by the real coherent tune shift @xmath47
, the shift of the synchrotron satellites in bunches can be reproduced rather well by @xcite @xmath48 where the sign @xmath49 is used for @xmath50 . for k=0
one obtains @xmath51 .
the above expression represents the head - tail eigenmodes for an airbag bunch distribution in a barrier potential @xcite with the eigenfunctions @xmath52 where @xmath53 is the local transverse bunch offset , @xmath54 is the chromatic phase , @xmath55 is the full bunch length and @xmath23 is slip factor .
the head - tail mode frequencies obtained from are shown in . in ref .
@xcite the analytic solution for the eigenvalues is obtained from a simplified approach , where the transverse space charge force is assumed to be constant for all particles .
this assumption is correct if there are only dipolar oscillations . in ref .
@xcite it is has been pointed out , that in the presence of space charge there is an additional envelope oscillation amplitude . for the negative-@xmath25
eigenmodes the envelope contribution dominates and therefore those modes disappear from the tune spectrum . in ref .
@xcite has been successfully compared to schottky spectra obtained from 3d self - consistent simulations for realistic bunch distributions in rf buckets .
analytic and numerical solutions for gaussian and other bunch distribution valid for @xmath56 were presented in @xcite . in an rf bucket the synchrotron tune
@xmath18 is a function of the synchrotron oscillation amplitude @xmath57 . for short bunches
@xmath18 corresponds to the small - amplitude synchrotron tune @xmath58 where @xmath59 is the rf voltage amplitude and @xmath60 is the rf harmonic number .
head - tail mode frequencies as a function of the space charge parameter using .
the red curves represent the result obtained for @xmath61.,width=321 ] for head - tail modes the space charge parameter is defined as a ratio of the space - charge tune shift ( ) to the small - amplitude synchrotron tune , @xmath62 and the coherent intensity parameter as , @xmath63 an important parameter for head - tail bunch oscillations in long bunches is the effective synchrotron frequency which will be different from the small - amplitude synchrotron frequency in short bunches . for an elliptic bunch distribution ( parabolic bunch profile ) with the bunch half - length @xmath64 ( rms bunch length @xmath65 ) ,
one obtains the approximate analytic expression for the longitudinal dipole tune @xcite , @xmath66 using @xmath67 instead of @xmath68 in eq .
[ eq : htmodes ] shows a much better agreement with the simulation spectra for long bunches in rf buckets ( see ref .
@xcite ) .
simulation result for @xmath69.,width=321 ] simulation result for @xmath70.,width=321 ] for gaussian bunches with a bunching factor @xmath71 ( @xmath72 , @xmath73 is the dc current ) , the transverse tune spectra obtained from patric simulations @xcite for different space charge factors and thin beams ( @xmath74 ) are shown in figures [ fig : patric_single_q0 ] , [ fig : patric_single_q2 ] and [ fig : patric_single_q9 ] .
the dotted vertical lines indicate the positions of the head - tail tune shifts obtained from with @xmath75 . for the low-@xmath25 satellites
there is a good agreement between and the simulation results .
lines with @xmath76 can only barely be identified in the simulation spectra .
the positions of the satellites for @xmath77 together with the predicted head - tail tune shifts from are shown in .
the error bars indicate the obtained widths of the peaks in the tune spectra .
head - tail tune shifts and their width obtained from the simulations for @xmath78 .
the error bars indicate the widths of the lines.,width=321 ] it is important to notice that the simulations for moderate space charge parameters ( @xmath79 ) require a 2.5d self - consistent space charge solver .
the theoretical studies rely on the solution of the mhl - schnauer equation @xcite , which assumes a constant space charge tune shift for all transverse particle amplitudes .
in contrast to the self - consistent results , patric simulation studies using the mhl - schnauer equation gave tune spectra with pronounced , thin satellites also for large @xmath25 . in order to account for the intrinsic damping of head - tail modes @xcite , which is the main cause of the peak widths obtained from the simulations , a self - consistent treatment is required . for thick beams ( here @xmath80 , which corresponds to the conditions at injection in the sis-18 ) the positions of the synchrotron satellites obtained from the simulation
are indicated in .
again , the error bars indicate the widths of the peaks . from the plot we notice an increase in the spacing between the @xmath81 satellites , relative to the analytic expression .
also the peak width for @xmath82 does not shrink with increasing @xmath83 .
head - tail tune shifts and their width obtained from the simulations for @xmath84 .
the error bars indicate the widths of the peaks.,width=321 ] the tune spectra obtained for @xmath85 , @xmath86 and @xmath70 are shown in , and .
one can observe that for thick beams ( here @xmath87 ) the @xmath88 peak remains very broad up to @xmath70 .
tune spectrum obtained from the simulation for @xmath85 and @xmath89.,width=321 ] tune spectrum obtained from the simulation for @xmath86 and @xmath89.,width=321 ] tune spectrum obtained from the simulation for @xmath70 and @xmath89.,width=321 ] this observation is consistent with a very simplified picture for the upper @xmath83 threshold for the intrinsic landau damping of head - tail modes .
for a gaussian bunch profile , the maximum incoherent tune shift , including the modulation due to the synchrotron oscillation is @xmath90 where @xmath46 is determined by eq .
[ eq : sctune ] .
the minimum space charge tune shift is ( see refs .
@xcite ) @xmath91 where @xmath92 is determined from the average of the space charge tune shift along a synchrotron oscillation with the amplitude @xmath93 . for a parabolic bunch we obtain @xmath94 . for a gaussian bunch and @xmath95
, we obtain @xmath96 .
each band of the incoherent transverse spectrum has a lower boundary determined by the maximum tune shift @xmath97 and an upper boundary determined by @xmath98 .
landau damping , in its very approximate treatment , requires an overlap of the coherent peak with the incoherent band .
the head - tail tune for low @xmath83 can be approximated as @xmath99 the distance between the coherent peak and the upper boundary of the incoherent band for fixed @xmath25 is @xmath100 for large @xmath83 the head - tail modes with positive @xmath25 converge towards @xmath101 .
for a given @xmath25 the mode is still inside the incoherent band if @xmath102 holds . in order to illustrate the above analysis ,
the incoherent band for @xmath103 is shown in ( shaded area ) .
for @xmath74 the coherent head - tail mode frequency crosses the upper boundary of the band at @xmath104 . for @xmath80
the @xmath103 head - tail mode remains inside the incoherent band until @xmath105 . for the @xmath88 modes the above analysis leads to thresholds of @xmath106 ( thin beams ) and @xmath107 for @xmath80 .
head - tail mode frequencies as a function of the space charge parameter .
the grey shaded area indicates the incoherent band for a gaussian bunch and @xmath103 .
the red curves represent the head - tail mode frequencies obtained for @xmath80.,width=321 ]
in this section , a brief description of the transverse beam excitation mechanisms as well as the two different tune measurement systems , topos and bbq , in the sis-18 is given .
further the experimental set - up , typical beam parameters and uncertainty analysis of the measured beam parameters is discussed .
the electronics used for beam excitation consist of a signal generator connected to two @xmath108 w amplifiers which feed power to @xmath109 terminated stripline exciters as shown in .
excitation types such as band limited noise and frequency sweep are utilized at various power levels to induce coherent oscillations .
band limited noise is a traditionally used beam excitation system for slow extraction in the sis-18 .
the rf signal is mixed with direct digital synthesis ( dds ) generated fractional tune frequency , resulting in rf harmonics and their respective tune sidebands .
this signal is further modulated by a pseudo - random sequence resulting in a finite band around the tune frequency .
the width of this band is controlled by the frequency of the pseudo - random sequence .
typical bandwidth of band limited exciter is @xmath110 of the tune frequency .
there are two main advantages of this system ; first it is an easily tunable excitation source available during the whole acceleration ramp and second , the band limited nature of this noise results in an efficient excitation of the beam in comparison to white noise excitation .
the main drawback is the difficulty in correlating the resultant tune spectrum with the excitation signal
. frequency sweep ( chirp / harmonic excitation ) using a network analyzer for btf measurements is an established method primarily for beam stability analysis @xcite .
however , using this method for tune measurements during acceleration is not trivial , and thus the method is not suitable for tune measurements during the whole ramp cycle .
nevertheless , this method offers advantages compared to the previous excitation method for careful interpretation of tune spectrum in storage mode , e.g. , injection plateau or extraction flat top .
thus frequency sweep is used during measurements at injection plateau to compare and understand the dependence of tune spectra on the type of excitation . following the beam excitation , the signals from each of the 12 shoe - box type bpms @xcite at sis-18 pass through a high dynamic range ( 90 db ) and broadband ( 100 mhz ) amplifier chain from the synchrotron tunnel to the electronics room , where the signals are digitized using fast 14 bit adcs at 125 msa / s .
bunch - by - bunch position is calculated from these signals using fpgas in real time and displayed in the control room .
the spatial resolution is @xmath1110.5 mm in bunch - by - bunch mode .
further details can be found in @xcite .
hence , topos is a versatile system which provides accurate bunch - by - bunch position and longitudinal beam profile .
this information is analyzed to extract non trivial parameters like betatron tune , synchrotron tune , beam intensity evolution etc .
the bbq system is a fully analog system and its front end is divided into two distinct parts ; a diode based peak detector and an analog signal processing chain consisting of input differential amplifier and a variable gain filter chain of 1 mhz bandwidth .
the simple schematic of bbq system configuration at sis-18 is shown in and the detailed principle of operation can be found in ref .
@xcite .
the sensitivity of bbq has been measured to be @xmath11110 - 15 db higher than that of topos under the present configuration .
the main reason for the difference is the relative bandwidth of the two systems and their tune detection principles . in bbq
, the tune signal is obtained using analog electronics ( diode based peak detectors and differential amplifier ) immediately after the bpm plates , while in topos position calculation is done after passing the whole bunch signal through a wide bandwidth amplifier chain . even though the bunches are integrated to calculate position in topos which serves as a low pass filter ,
the net signal - to - noise ratio is still below bbq .
operations similar to bbq could also be performed digitally in topos to obtain higher sensitivity but would require higher computation and development costs .
topos can provide individual tune spectra of any of the four bunches in the machine , while bbq system provides `` averaged '' tune spectra of all the bunches .
both the systems have been benchmarked against each other .
tune spectra shown in are mostly from the bbq system while topos is primarily used for time domain analysis , nevertheless this will be pointed out when necessary .
experiments were carried out using n@xmath112 and u@xmath113 ion beams at the sis-18 injection energy of 11.4mev / u .
the data were taken during 600ms long plateaus .
at injection energy space charge effects are usually strongest .
four bunches are formed from the initially coasting beam during adiabatic rf capture .
the experiment was repeated for different injection currents . at each intensity level
several measurements were performed with different types and levels of beam excitation in both planes .
tune measurements were done simultaneously using the topos and bbq systems .
the beam current and the transverse beam profile are measured using the beam current transformer @xcite and the ionization profile monitor ( ipm ) @xcite respectively .
an examplary transverse beam profile is shown in .
the dipole synchrotron tune @xmath114 is deduced using the residual longitudinal dipole fluctuations of the bunches .
@xmath67 has been used as an effective synchrotron tune for all experimental results and will be referred as @xmath18 from hereon .
the momentum spread is obtained from longitudinal schottky measurements @xcite .
important beam parameters during the experiment are given in and .
it is important to note that all the parameters required for analytical determination of @xmath115 are recorded during the experiments .
.beam parameters during the @xmath116 experiment [ cols="^,^,^",options="header " , ] [ tab : beam parameters n ] from and one can estimate that in the measurements the head - tail space charge and image current parameters were in the range @xmath79 and @xmath117 for the horizontal and vertical planes . ions .
the dotted lines shows the normal distribution for the rms width obtained by evaluation of beam profile around its centre.,width=321 ]
tune spectra measurements at different currents are presented and interpreted in comparison with the predictions of .
the effect of different excitation types and power on the transverse tune spectra is studied .
transverse impedances in both planes are obtained from the coherent tune shifts .
incoherent tune shifts are obtained from the relative frequency shift of head - tail modes in accordance with .
chromaticity is measured using head - tail eigenmodes and the obtained relative height of the observed peaks for different chromaticities is analyzed . ions and beam parameters given in ( see text ) .
the dashed lines indicate the head - tail tune shifts from .,width=321 ] shows the horizontal tune spectra obtained with the bbq system using band - width limited noise at different intensities .
( a ) shows the horizontal tune spectrum at low intensity . here
the @xmath118 peaks are almost equidistant , which is expected for low intensity bunches .
the space charge parameter obtained using the beam parameters and is @xmath119 .
the vertical lines indicate the positions of the synchrotron satellites obtained from ( with @xmath75 ) .
( b ) shows the tune spectrum at moderate intensity ( @xmath120 ) .
the @xmath121 peaks can both still be identified .
( c ) shows the tune spectra at larger intensity ( @xmath122 ) .
an additional peak appears between the @xmath123 and @xmath124 peaks which can be attributed to the mixing product of diode detectors ( since at this intensity @xmath125 acts across the diodes pushing it into the non - linear regime ) .
the @xmath77 peaks can be identified very well , whereas the amplitudes of the lines for negative @xmath25 already start to decrease ( see ) . in the horizontal plane
the effect of the pipe impedance and the corresponding coherent tune shift can usually be neglected because of the larger pipe diameter .
ions and beam parameters given in .
the dashed lines indicate the head - tail tune shifts from .,width=321 ] shows the vertical tune spectrum obtained by the bbq system with band limited noise excitation for @xmath116 beams with @xmath83 values larger than @xmath126 . here
the negative modes @xmath127 could not be resolved anymore . in the vertical plane
the coherent tune shift is larger due to the smaller sis-18 beam pipe diameter ( @xmath128 ) .
the shift of the @xmath123 peak due to the effect of the pipe impedance is clearly visible in . in the measurements
the width of the peaks is determined by the cumulative effect of non - linear synchrotron motion , non - linearities of the optical elements , closed orbit distortion , tune fluctuation during the measurement interval as well as due to the intrinsic landau damping ( ) . from the comparison to the simulations we conclude that the intrinsic landau damping is an important contribution to the width of the @xmath82 peaks .
the coherent tune shift @xmath129 can be obtained by measuring shift of the @xmath123 line as a function of the peak bunch current as shown in .
the transverse impedance is obtained by a linear least square error fit of the measured shifts in both planes to and .
the impedance values are obtained in the horizontal and vertical planes at injection energy are @xmath130 m@xmath131 and @xmath132 m@xmath131 respectively , which agrees very well with the expected values for the average beam pipe radii of the sis-18 . mode .
the fft resolution is @xmath133 and is always kept higher than the mode width.,width=321 ] beam intensities together with the analytical curves from using the space charge tune shift estimated from the beam parameters in .
the dotted line corresponds to the incoherent tune shift .
the error bars for the vertical plane correspond to the width of measured modes ( see text ) . in the horizontal plane
the error bars are estimated by the propagation of uncertainties ( see text).,width=321 ] ion beam at various current levels ( ) .,width=321 ] ion beam at various current levels ( ) .,width=321 ] ion beam at various current levels ( ) .,width=321 ] figures [ fig : inc_shift1 ] , [ fig : inc_shift2 ] , [ fig : inc_shift3 ] and [ fig : inc_shift4 ] show the measured positions of the peaks in the tune spectra for different intensities . in comparison the analytical curves ( solid lines ) obtained from for the head - tail tune shifts
are plotted using @xmath83 estimated from the beam parameters in and for each intensity .
the error bars in the vertical plane ( @xmath134 ) correspond to the @xmath135 db width of the measured peak due to accumulation of various effects ( see ) . in the horizontal plane , error bars ( @xmath136 )
are estimated by propagation of parameter uncertainties mentioned in . in this subsection
, we introduce another space charge parameter @xmath137 which is the measured space charge parameter using the following method .
it is not to be confused with @xmath83 which is predicted for a given set of beam parameters by .
the incoherent space charge tune shift can be determined directly from the tune spectra by measuring the separation between the @xmath123 and @xmath88 peaks , i.e. @xmath138 ) and fitting it with the parameter @xmath83 in the predictions from .
the value of @xmath83 for the best fit is denoted as @xmath137 .
@xmath139 is obtained by rearranging for @xmath140 while @xmath141 .
the linearized absolute error on measured @xmath137 ( @xmath142 ) is given by @xmath143 @xmath144 is given by either the width of the @xmath145 lines or by the frequency resolution of the system . in a typical tune spectra measurement using data from 4000 turns @xmath146 .
the absolute error is a non - linear function of @xmath83 in accordance to the .
it is possible to define the upper limit of @xmath83 where this method is still adequate based on the system resolution and .
if we define a criterion that , @xmath147 to resolve the head - tail modes .
this gives the limit to be @xmath148 where the measurement error is still within the defined criterion .
shows a plot of the predicted space charge tune shifts ( @xmath83 ) versus the ones measured from the tune spectra using the above procedure ( @xmath137 ) .
for @xmath149 the space charge tune shifts measured from the tune spectra are systematically lower by a factor @xmath150 than the predicted shifts .
it is shown by the dotted line in which is obtained by total least squares fit of the measured data points . for larger @xmath83
the factor decreases to @xmath151 .
thus the method for measuring the incoherent tune shift based on head - tail tune shifts is found to be satisfactory only in the range @xmath149 .
a possible explanation is the effect of the pipe impedance .
similar observations are made by the results of self - consistent simulations in , where for @xmath152 the separation of the @xmath123 and @xmath88 peaks observed is underestimated by ( see ) . , [ fig : inc_shift2 ] , [ fig : inc_shift3 ] and [ fig : inc_shift4 ] , a plot of predicted @xmath83 using against measured @xmath83 using the distance between modes @xmath123 and @xmath88 is obtained.,width=321 ] presents the tune spectra obtained from bbq system at various excitation power levels of band limited noise .
the beam is excited with @xmath153 and @xmath154 mw / hz power spectral density on a bandwidth of @xmath155 khz .
signal - to - noise ratio ( snr ) increases with excitation power whereas the spectral position of various modes is independent of excitation power . particles with beam excited at three different excitation amplitudes .
the beam is excited with @xmath153 and @xmath154 mw / hz power ( blue , red and black respectively ) .
the dotted lines mark the relative positions of the head - tail modes .
the frequencies of various head - tail modes are unaffected.,width=321 ] beam excitation using two other excitation types i.e. frequency sweep and white noise is also performed to study the effect of excitation type on the tune spectra .
shows the tune spectra under same beam conditions for different types of beam excitation obtained from the bbq system .
the frequencies of various modes in the tune spectra are independent of the type of excitation .
the signal - to - noise ratio ( snr ) is optimum for band limited noise due to long averaging time compared to `` one shot '' spectra from sweep excitation .
shows the 2-d contour plot for frequency sweep excitation in vertical plane obtained from the topos system , where various head - tail modes are individually excited as the excitation frequency crosses them .
frequency sweep excitation allows resolving the transverse center of mass along the bunch for various modes which helps in identifying each head - tail mode in time domain .
this serves as a direct cross - check for the spectral information and leaves no ambiguity in identification of the order ( k ) of the modes . shows the corresponding transverse center of mass along the bunch for @xmath156 and @xmath126 at the excited time instances .
this method works only with sweep excitation and requires high signal - to - noise ratio in the time domain , which amounts to higher beam current or high excitation power . at @xmath157 particles with band limited noise ( rf noise ) , frequency sweep ( chirp noise ) and white noise ( top to bottom ) .
the dotted lines mark the relative positions of the head - tail modes.,width=321 ] with @xmath158 particles .
the various modes get individually excited as the sweep frequency ( marked by the arrow ) coincides with the mode frequency .
the @xmath81 modes are marked .
symmetric sidebands due to coherent synchrotron oscillations are visible around the sweep frequency spaced with @xmath18 and are clearly distinguished from the head - tail modes.,width=321 ] and @xmath126 modes corresponding to .
frequency sweep is used to excite the beam which enables to resolve the distinct head - tail modes temporally .
the rms bunch length ( @xmath159 ) is indicated.,width=321 ] as highlighted in the previous section , the frequency sweep allows to resolve the different head - tail modes both spectrally and temporally .
this procedure can be used for the precise determination of the chromaticity by fitting the analytical expression for the head - tail eigenfunction to the measured bunch offset , with the chromaticity ( @xmath21 ) as the fit parameter as shown in for @xmath156 and @xmath126 .
the measured chromaticity is independent of the order of head - tail eigenfunction used to estimate it .
the fitting method is shown in ; the head - tail eigenfunction from is multiplied with the beam charge profile @xmath160 and corrected for the beam offset @xmath161 at the bpm where the signal is measured .
@xmath162 the fit error @xmath163 is reduced as a function of independent variables ; chromaticity @xmath21 and head - tail mode amplitude @xmath164 .
the fit error gives the goodness of the fit .
it is used to determine the error bars on measured chromaticity .
this method has been utilized for the determination of chromaticity at sis-18 as shown in .
the set and the measured chromaticity can be fitted by linear least squares to obtain the form @xmath165 as shown by red dashed line in
. also shows a coherent tune shift due to change in sextupole strength which is used to adjust the chromaticity .
this is due to uncorrected orbit distortions during these measurements .
these chromaticity measurements agree with the previous chromaticity measurements using conventional methods @xcite .
it is also possible to determine the relative response amplitude of each head - tail mode to the beam excitation both in time and in frequency domain with topos .
shows the tune spectrum obtained with sweep excitation for different chromaticity values .
the beam parameters are kept the same ( @xmath166 ) . the spectral position and and relative amplitude of each head - tail mode peak
are confirmed using the time domain information ( see ) . in the single particle response amplitudes for different @xmath25 ( )
are plotted as a function of the chromaticity .
the measured relative amplitudes are indicated by the colored symbols .
the comparison indicates that the simple single particle result ( ) describes quite well the dependence of the relative height of the peaks obtained from the topos measurement .
and @xmath88 modes with the head - tail phase shift as the fit parameter .
this method has been used for the precise determination of the chromaticity.,width=321 ] .
all the modes are relative to k=1 mode , which has a significant amplitude during the whole measurement range.,width=321 ]
in this section we will discuss the application of our results to tune measurements in the sis-18 and in the projected sis-100 , as part of the fair project at gsi @xcite .
as shown in the previous section the relative amplitudes of the synchrotron satellites in the tune spectra are primarily a function of chromaticity and possibly the excitation mechanism . in order to determine the coherent tune with high precision the position of the @xmath123 mode has to be measured .
depending on the machine settings , if the relative height of the @xmath123 peak with respect to the other modes is small , then the @xmath123 mode may not be visible at all . to estimate the bare tune frequency in this case ,
the information of space charge parameter , coherent tune shift and chromaticity are all simultaneously required with good precision .
another important point is the tune measurement during acceleration .
the space charge parameter for @xmath167 stored ions in the sis-18 from injection to extraction reduces only by @xmath168 as shown in .
the dynamic shift of head - tail modes during acceleration is shown in obtained from topos system under same conditions .
the asymmetry of @xmath169 modes around @xmath123 mode can only be understood in view of the space charge effects predicted by .
thus , a correct estimate of this parameter plays an important role in understanding the tune spectra not only during dedicated experiments on injection plateau , but also during regular operations .
the measurement time required to resolve the various head - tail modes ( @xmath170 ) is a complex function of @xmath18 , @xmath83 , beam intensity and excitation power . to give some typical numbers for sis-18 ; on a measurement time of 600 ms on the injection plateau , if one spectrum is obtained in @xmath171 ms ( @xmath172 turns ) , an improvement of factor @xmath173 in snr by averaging 30 spectra .
following the calculations in , @xmath148 can be resolved under typical injection operations . however , the constraints on measurement time are much higher during acceleration , where the tune / revolution frequency increases due to acceleration .
this allows the measurement of a single spectrum typically only over @xmath174 turns ( depends on ramp rate as well ) .
there are no averaging possibilities since the tune is moving during acceleration due to dynamic changes in machine settings as seen in .
in addition , the synchrotron tune reduces with acceleration making it practically very difficult to resolve the fine structure of the head - tail modes for @xmath175 . with @xmath176 particles .
the estimated space charge parameter reduces by @xmath168 in this typical case from injection at 11.4 mev / u to 300 mev / u at extraction.,width=321 ] and @xmath124 modes around @xmath123 is maintained throughout the ramp which depends primarily on the space charge parameter @xmath177.,width=321 ]
two complimentary tune measurement systems , topos and bbq , installed in the sis-18 are presented .
analytical as well as simulation models predict a characteristic modification of the tune spectra due to space charge and image current effects in intense bunches .
the position of the synchrotron satellites corresponds to the head - tail tune shifts and depends on the incoherent and the coherent tune shifts .
the modification of the tune spectra for different bunch intensities has been observed in the sis-18 at injection energy , using the topos and bbq systems . from the measured spectra
the coherent and incoherent tune shift for bunched beams in sis-18 at injection energies were obtained experimentally using the analytic expression for the head - tail tune shifts .
head - tail modes were individually excited and identified in time domain and correlated with the spectral information .
a novel method for determination of chromaticity based on gated excitation of individual head tail mode is shown .
the dependence of the relative amplitudes of various head - tail modes on chromaticity is also studied .
the systems were compared against each other as well as with different kinds of excitation mechanisms and their respective powers .
these measurements give a clear interpretation of tune spectra at all stages during acceleration under typical operating conditions . the understanding to tune
spectra provides an important input to new developments related to planned transverse feedback systems for sis-18 and sis-100 .
the measurement systems also open new possibilities for detailed beam investigations as demonstrated in this contribution .
we thank marek gasior of cern bi group for the help in installation of bbq system and many helpful discussions on the subject .
gsi operations team is also acknowledged for setting up the machine .
we also thank klaus - peter ningel from the gsi - rf group who helped setting up the amplitude ramp for the adiabatic bunching .
shows a typical longitudinal profile of the bunch . if @xmath179 is the amplitude at each time instance @xmath180 .
the bunching factor is calculated by the .
@xmath181 where n @xmath182 is the number of samples in one rf period ( at injection ) .
the topos system samples the bunch at 125 msa / s , thus the difference between adjacent samples is 8ns .
the calculation of @xmath46 from has a dependence on measured current , transverse beam profiles , longitudinal beam profiles and the twiss parameters .
the measurement uncertainty on each of these measured parameters at gsi sis-18 were commented in the detailed analysis in @xcite . even though some parameters and the associated uncertainties are correlated ,
any correlations are neglected in present analysis .
uncertainties in each measured parameter are propagated to find the error bars on the calculated and measured incoherent tune shifts .
where @xmath185 mm is the wire spacing and @xmath186 is the adc resolution of the ipm and b is defined as @xmath187 . if the error bars are derived from @xmath188 measurements , the measured profile is given by @xmath189 for each tune measurement at the given intensity and excitation power , 5 - 8 transverse beam profiles were measured , and the relative error is obtained @xmath110 using and . the relative systematic error ( bias ) in transverse beam width measurements is @xmath190 @xcite and ignored in this analysis .
the uncertainty in the injected current is dominated by fluctuations in the source and the relative uncertainty is estimated to be @xmath110 based on 5 - 8 measurements at the same intensity settings for each measurement point .
bunch length and bunching factor vary by @xmath191 due to long term beam losses only under high intensity beam conditions . the maximum relative bias in the lattice parameter @xmath192 is assumed to be @xmath110 at the ipm location .
taking all the relative errors , uncertainty propagation using familiar gives relative error for estimated incoherent tune shifts @xmath193 .
@xmath194 tune measurements done by averaging over long intervals contribute to the width of modes due to long term beam losses .
beam losses lead to change in coherent tune especially in the vertical plane where the image current effects are larger
. this has been highlighted at appropriate sections in the text .
99 p. kowina et al . , digital baseband tune determination " , proc . of biw10 , santa fe , u.s.a ( 2010 ) m. gasior , high sensitivity tune measurement using direct diode detection " , proc . of biw12 , virginia , u.s.a ( 2012 ) s. chattopadhyay , some fundamental aspects of fluctuations and coherence in charged particle beams in storage rings " , super proton synchrotron division , cern 84 - 11 , ( 1984 ) d. boussard ,
schottky noise and beam transfer function diagnostics `` , cern accelerator school ; fifth advanced accelerator physics course '' , ( 1995 ) m.blaskiewicz , phys . rev .
beams * 1 * , 044201 ( 1998 ) o. boine - frankenheim , v. kornilov , phys . rev .
beams , * 12 * , 114201 ( 2009 ) a. burov , phys . rev .
beams * 12 * , 044202 ( 2009 ) v. balbekov , phys . rev .
beams * 12 * , 124402 ( 2009 ) v. kornilov , o. boine - frankenheim , phys . rev .
beams,*13 * , 114201 ( 2010 ) r. singh et al .
, proc . of biw12 , newport news , u.s.a ( 2012 ) v. kornilov , o. boine - frankenheim , phys
beams,*15 * , 114201 ( 2012 ) d. mhl , part .
50 , 177 ( 1995 ) f. j. sacherer , transverse space - charge effects in circular particle accelerators " , ucrl-18454 ( 1968 ) t. linnecar and w. scandale , proc . of pac
1981 , p. 2147
( 1981 ) j. borer , proc . of pac 1979 ( 1979 ) k. schindl , space charge " , joint us - cern - japan - russia school on particle accelerators " ( 1998 ) o. boine - frankenheim , t. shukla , phys .
st accel .
beams * 8 * , 034201 ( 2005 ) p. forck , lecture notes on beam intrumentation " ( 2004 ) h. reeg et al .
, proc . of dipac ,
grenoble , france ( 2001 ) t. giacomini , proc . of biw04 , knoxville ,
u.s.a ( 2004 ) f. caspers ,
schottky signals",cas 2008 , dourdon , france ( 2008 ) s. paret , phd thesis , tu darmstadt , germany(2009 ) g. franchetti et al . ,
phys . rev .
beams * 13 * , 114203 ( 2010 ) o. boine - frankenheim , the fair accelerators : highlights and challenges " , proc . of ipac 2010 , japan ( 2010 ) | two different tune measurement systems have been installed in the gsi heavy - ion synchrotron sis-18 .
tune spectra are obtained with high accuracy using these fast and sensitive systems .
besides the machine tune , the spectra contain information about the intensity dependent coherent tune shift and the incoherent space charge tune shift .
the space charge tune shift is derived from a fit of the observed shifted positions of the synchrotron satellites to an analytic expression for the head - tail eigenmodes with space charge .
furthermore , the chromaticity is extracted from the measured head - tail mode structure .
the results of the measurements provide experimental evidence of the importance of space charge effects and head - tail modes for the interpretation of transverse beam signals at high intensity . |
there is considerable interest in estimating age and transit times of elements in a physical system or , equivalently , of individuals in a population , in disciplines as diverse as population dynamics and demography @xcite , chemical engineering @xcite , and hydrology and geophysics ( e.g. , among many others @xcite ) .
it has in fact become increasingly clear that the age , survival time , and the total time spent by each element in a system may provide additional key insights into specific aspects of a system s behavior .
this viewpoint , which can be considered as a time - integrated lagrangian perspective , has been especially emphasized in groundwater systems @xcite and in the hydrological response of watersheds @xcite , using both theoretical @xcite and field approaches @xcite .
recent discussions in the literature about the role of internal variability and external forcing ( e.g. , rainfall ) on the properties of age distributions @xcite , as well as the differences between age and survival time distributions , their degree of statistical dependence , and their symmetry under time reversal ( e.g. , @xcite ) have made evident that a comprehensive theory of age and related concepts is still missing . toward this goal , in this contribution we focus on the linkage between age and survival time distribution in both transient and steady - state conditions . differently from what was assumed in @xcite
, we show that age and survival time are in general statistically dependent quantities ( the only case of independence being the one of time- and age - independent loss ( or input ) in steady state ) .
the theoretical framework afforded by the evolution equation of the joint distributions of age and survival also provides a means to easily understand the time symmetries between age and survival , and the derivation of the general properties of the transit - time distribution .
we should warn the readers unfamiliar with the previously cited literature that , perhaps because of the contributions from many disciplines , the terminology which identifies these variables is hardly unified .
for example , apart from the age , the definition of which seems uncontroversial ( @xmath0 in what follows ) , the survival time ( here indicated as @xmath1 ) is often also indicated as life expectancy , while input and output rates are often also called birth and death functions .
the variable with possibly the most appellations is the so - called transit time ( @xmath2 ) , the sum of age and survival time , which is also indicated as travel time , life span , total life time , and sojourn time .
as long as the mathematical formalism is clear and the notation kept consistent , as we have hopefully done here , we trust that these different names will not confuse the readers .
the paper is organized as follows .
the evolution equation for the joint distribution of age and survival time is introduced and solved in section [ sec : joint ] , with boundary conditions given by the survival time distribution at birth and the related age distribution at death .
the solution is used to derive the transit time distribution by a simple integration in section [ sec : transit ] .
the steady state conditions are discussed in section [ sec : steadystate ]
. finally , we present some applications in section [ sec : applications ] with the purpose of showing some interesting details of the theory . while most of these applications have a close connection to hydrological and fluid mechanic systems , they are by necessity highly idealized to allow us to focus on the novel theoretical results , avoiding the additional complications that more realistic applications with random external forcing ( e.g , rainfall ) and spatial heterogeneities would add .
the transit time ( @xmath2 ) of an element of a system is the sum of the time spent since the entrance / birth , called the age ( @xmath0 ) , and the time that it will spend before exit / death , called the survival time ( @xmath1 ) . at a given time @xmath3
, each element is characterized by a certain age and survival ( and thus transit ) time , which globally can be described by the joint distribution @xmath4 . in words
, @xmath5 represents the ( infinitesimal ) amount of elements ( e.g. , a mass or population number having age between @xmath0 and @xmath6 and survival time @xmath1 and @xmath7 at time @xmath3 .
the balance equation for the joint distribution @xmath4 can be obtained considering that , as the system evolves in time , @xmath4 is conserved along the lines orthogonal to the bisector in the @xmath8 plane , which are characterized by having constant @xmath2 . based on these considerations ,
one can readily write @xmath9 the equation is controlled by the boundary conditions @xmath10 , which is the survival time distribution at input / birth , and @xmath11 , which is the age distribution at output / death .
an example of the evolution of the joint distribution is shown in figure [ fig : translationbcs ] , showing how @xmath12 is simply the input boundary condition on the @xmath1 axis , translating in time along lines of constant @xmath2 until it crosses the @xmath0 axis where @xmath13 .
the figure also clearly shows how the two boundary conditions can not be independent , as will be seen more precisely later .
it is interesting that the contribution of input and output to the system is entirely felt through the boundary conditions . in more general cases
, elements could also enter with age different from zero ( immigration ) or exit with a non - zero survival time ( emigration ) , in which case equation ( [ eq : jointeq ] ) should also contain corresponding source and sink terms ; these generalization however will not be pursued in this paper .
more formally , moving along the characteristic curves , defined by @xmath14 , @xmath15 and @xmath16 , which are obviously also lines of constant transit time , it is possible to re - express equation ( [ eq : jointeq ] ) as @xmath17 so that the solution is then @xmath18 or equivalently , @xmath19 from this it is evident how the joint distribution at a given time is simply the time shift of the boundary conditions and that , if time is reversed , the whole process is flipped , with the age playing the role of the survival time and vice versa .
this type of time symmetry will appear frequently in the following .
results from a simple translation of the boundary conditions . ] by integrating equation ( [ eq : jointeq ] ) over @xmath1 , one obtains the mkendrick - von foerster ( mkvf ) equation @xcite , describing the dynamics of an age - structured population equation , @xmath20 where @xmath21 is the age distribution ( mass over time ( age ) ) , quantifying the amount of substance having age @xmath0 at time @xmath3 . the sink term , @xmath22 , is the age distribution at output / death , previously introduced as a boundary condition for ( [ eq : jointeq ] ) .
it can be written as @xmath23 where @xmath24 is the age and mass specific output rate . with initial condition
@xmath25 and boundary condition @xmath26 , where @xmath27 is the input / birth rate , the solution of equation ( [ eq.mkvf ] ) is @xcite @xmath28 on the other hand , by integrating ( [ eq : jointeq ] ) over @xmath0 , a corresponding equation for the survival time distribution is obtained , @xmath29 where @xmath30 quantifies the amount of substance having survival @xmath1 at time @xmath3 . the source term @xmath31 is the survival time distribution at input / birth , which can be expressed as @xmath32 with @xmath33 being the survival time and mass specific birth rate .
the boundary condition is the overall output @xmath34 and the initial condition is @xmath35 .
as for the mkvf equation , the solution is obtained with the method of characteristics as @xmath36 by integrating again either equation ( [ eq.mkvf ] ) over @xmath0 or equation ( [ eq.mkvfsurv ] ) over @xmath1 , the familiar form of the balance equation is obtained , @xmath37 with @xmath38 and where the input and output can also be written as @xmath39 and @xmath40 . it should be noted that , when the solution of ( [ eq : jointeq ] ) is available , the age and survival distributions , @xmath41 and @xmath42 , and the evolution of @xmath43 can be directly obtained by integrating the joint distribution @xmath4 , without need to go through the corresponding equations ( [ eq.mkvf ] ) , ( [ eq.mkvfsurv ] ) , and ( [ eq : balance ] ) . looking at these equations , it is also worth noting again the symmetry of the problem with respect to the time reversal , upon which age and survival time exchange their roles , with the output becoming the input and the age - specific loss function playing the part of the survival - specific birth function and vice versa .
it is easy to see , in fact , that with these substitutions and @xmath44 , equation ( [ eq.mkvf ] ) and ( [ eq.mkvfsurv ] ) are interchangeable .
it is possible at this point to establish a relationship between the age - specific output and the survival - specific input and , in turn , discuss the conditional distributions between age and survival times .
the latter will be essential in deriving the residence time statistics in section [ sec : steadystate ] . to this purpose ,
we begin by returning to equation ( [ eq : jointsol ] ) which immediately furnishes the relationship between the boundary conditions by setting either @xmath45 or @xmath13 ( see figure [ fig : translationbcs ] ) , @xmath46 these equations , when expressed in terms of their probability density functions ( pdfs ) normalized to have area one , @xmath47 become a relationship already obtained by @xcite , @xmath48 the subscripts @xmath49 and @xmath50 , in particular , refer to the variables age at death and survival time at birth , respectively .
they will be used explicitly when it is necessary to refer to them as random variables to distinguish them from age @xmath0 and survival time @xmath1 of the entire population , as in section [ sec : meanvalues ] . from ( [ eq : relbc ] ) , a relationship between the birth and loss function is then obtained from the definition of @xmath51 and @xmath52 , in equations ( [ eq : defloss ] ) and ( [ eq : defbirth ] ) , in which the respective solutions for @xmath41 and @xmath42 can be substituted from ( [ eq : mkvfsol ] ) and ( [ eq : mkvfsurvsol ] ) , giving @xmath53 this clearly shows that the age- and survival - specific birth and loss functions are not independent .
coming back to equation ( [ eq : jointsol2 ] ) , the joint distribution can be expressed in terms of the conditional distribution between age and survival time . to this purpose , it is more useful to consider the pdfs , instead of distributions , and define @xmath54 where @xmath55 and @xmath56 are the marginal pdfs of age and survival time , respectively . focusing , as an example , on the conditional pdf of survival given age
, an expression can be derived combining ( [ eq : conditional ] ) and ( [ eq : jointsol2 ] ) , @xmath57 substituting in ( [ eq : conddistr ] ) the solutions for @xmath51 and @xmath58 given by ( @xmath59 ) and ( [ eq : defloss ] ) , one readily obtains @xmath60 only when this expression is equal to the marginal distribution of age , are the age and survival time statistically independent . thus comparing with the marginal @xmath61 one sees that , in general , marginal and conditional probability distributions are different
this implies that ( when elements are sampled at random in the system ) age and survival are typically statistically dependent variables .
this remains true , in transient conditions , even when the loss and birth functions are constant , because the two distributions remain different for @xmath62 and @xmath63 .
further considerations on ( [ eq : condsol ] ) will be given , for steady state conditions in section [ sec : steadystate ] .
as already said , the transit time , @xmath2 , is the total time spent by an element in the system , given by the sum of age and survival time , @xmath64 its distribution can thus be obtained as the distribution of the sum of the two random variables , the age and survival time , e.g. @xcite , @xmath65 which can also be written as @xmath66 as shown in figure [ fig : linestconst ] , equation ( [ eq : intjoint2 ] ) is the integral along lines of constant @xmath2 , which are orthogonal to the bisector on the plane @xmath67 .
using equation ( [ eq : jointsol ] ) one also has @xmath68 which tells us that we can know the transit time distribution at time @xmath3 by summing up the amount leaving the system with age @xmath69 within the time window @xmath70 .
only when @xmath0 and @xmath1 are statistically independent is equation ( [ eq : transdistr ] ) a simple convolution integral .
an alternative expression can be similarly obtained from equation ( [ eq : jointsol ] ) , @xmath71 which instead looks back to the time window ( @xmath72,@xmath3 ) and sums all the elements entering with survival time @xmath73 . .
the boundary conditions @xmath74 and @xmath75 are also indicated on the @xmath0 and @xmath1 axes , respectively . ]
it may be useful to note that the transit time distribution refers in general to any element of the control volume ( or population ) . if one instead only focuses on the elements entering the system ( or the newborns ) , their transit time @xmath2 is equal to their survival time @xmath1 , because for them @xmath45 , and their distribution is @xmath76 .
analogously , focusing on the elements leaving the system ( or dying ) , @xmath2 equals the age @xmath0 and their residence time distribution is the equal to @xmath77 .
several of the previous relationships assume an interesting , simplified form at steady state , a necessary condition for which is that @xmath51 and @xmath52 are time independent .
in such a case , the balance equation for a steady state system is simply @xmath78 and from equation ( [ eq : jointsol ] ) it follows that @xmath79 meaning that not only the overall input equals the overall output , but also that an input with a fixed age @xmath0 must be balanced by an output of equal survival time @xmath1 . as a result , the joint distribution @xmath80 is constant , and equal to @xmath81 along lines of constant @xmath2 . in steady state conditions , age and survival time distributions
are derived by taking @xmath82 in ( [ eq : mkvfsol ] ) and ( [ eq : mkvfsurvsol ] ) , that is @xmath83 and @xmath84 in particular , from ( [ eq : jointsol ] ) , the integral of @xmath80 over @xmath0 is the same as the integral over @xmath1 , suggesting that @xmath85 in turn , it follows immediately that , at steady state , birth and loss functions must be equal , @xmath86 we note that the so - called survivor function @xcite , defined as the exceedance probability of survival at steady state , can be obtained by dividing either the age distribution or the survival distribution by the input @xmath87 ( or by the output @xmath88 ) , @xmath89 so that , as well known @xcite , @xmath90 the transit - time distribution at steady state can also be easily obtained by solving the integral in equation ( [ eq : transdistr ] ) and substituting equation ( [ eq : agesteady ] ) , @xmath91 or , normalized as a pdf , @xmath92 finally , regarding the conditional probabilities , from equation ( [ eq : conddistr ] ) , @xmath93 and @xmath94 because of ( [ eq : equallossbirth ] ) , the two are obviously equal .
comparing for example the distribution ( [ eq : ageconddistrsteady ] ) with the corresponding marginal pdf @xmath95 it becomes clear that only for constant @xmath96 , are ( [ eq : survconddistrsteady ] ) and ( [ eq : ageconddistrsteady ] ) equal to their marginal distributions , thereby implying that age and survival time are statistically independent .
this is essentially due to the rescaling ( or memoryless ) property of the resulting exponential distributions @xcite , which is the form taken by all these distributions in this special case ( see section 5.1 ) . in general , however , when the input and loss functions depend respectively on age and survival time , the two variables are statistically dependent , as will be shown in detail in the applications . in steady state ,
because of ( [ eq : equaldistr ] ) , the age distribution and survival distribution have same mean @xmath97 the mean age at death and mean survival time at birth are also equal , @xmath98 while the mean transit time is then @xmath99 with regard to the mean transit time , by definition , @xmath100 where the last equality has been obtained by multiplying and dividing by the output @xmath88 .
now , remembering that @xmath101 where @xmath102 is the variance of the respective variable , then one obtains the exact relationship @xmath103 thus , in general , @xmath104 , so that @xmath105 represents an upper bound for both mean age and mean survival time . in particular , @xmath106 equals @xmath105 only when the loss function is a dirac delta function , for which the variance of @xmath2 is zero , as in the case of a plug - flow system .
in addition , substituting expression ( [ eq : ageand transit ] ) into the equation above , an exact link between @xmath107 and @xmath106 is also obtained as @xmath108 the same condition was obtained in a somewhat different way by @xcite . only in the case of @xmath109 and @xmath110
constant , then @xmath111 and @xmath112 , and @xmath113 , while for a dirac delta loss function , @xmath114 .
the manner in which input and loss functions depend on age and survival time plays a key role in determining whether the mean of the age and survival time , @xmath107 and @xmath115 , are greater or lower than the mean age at death and the mean survival time at birth , @xmath116 and @xmath117 . for example , as already discussed by @xcite and @xcite , in the case of a loss function @xmath109 which selects preferably young elements leaving older element to age in the system , the resulting mean age at death is lower than the mean age in the system , i.e. , @xmath118 . on the contrary
, the case @xmath119 is true whenever older elements tend to be chosen by @xmath109 , leaving young ones to keep the mean age low in the system , compared to the mean age at death ( see section [ app4 ] ) .
we present four examples to illustrate the previously discussed theory .
the first is a simple steady state system with constant birth and loss functions .
the second consists of a plug - flow system in which all the elements have the same transit time .
the third application is characterized by a periodicity of the age - independent loss function , while the fourth one focuses on the role of age - dependence in the loss function .
this simplest case , which represents a well - mixed system , serves as a point of reference for the more complex cases presented later .
it is characterized by constant and equal birth and loss function , @xmath120 so that the balance equation gives @xmath121 while the normalized age and survival time distributions are @xmath122 with mean @xmath123 ( see figure [ fig : linstead ] ) .
it is easy to show that the same exponential function results from ( [ eq : survconddistrsteady ] ) and ( [ eq : ageconddistrsteady ] ) for the conditional pdfs , so that in these special case age and survival are statistical independent and their joint distribution is simply the product of the two distributions . the age distribution at death ( survival time distribution at birth )
is simply obtained by multiplying @xmath124 by @xmath125 ( @xmath126 by @xmath125 ) , @xmath127 which in its normalized form is equal to the age distribution , @xmath128 , and similarly with @xmath129 and @xmath130 .
the mean values are @xmath131 the transit time distribution ( fig .
[ fig : linstead ] ) is calculated from equation ( [ eq : transdistrsteady ] ) , @xmath132 which , written as a pdf , becomes @xmath133 an erlang-2 distribution for the sum of two independent , exponentially distributed random variables ( see figure [ fig : linstead ] ) , with mean @xmath134 @xcite . .
the distributions are calculated for @xmath135 and @xmath136 . ]
while in the previous example age and survival time were statistically independent , this application takes the opposite extreme of complete ( i.e. , deterministic ) dependence . to this purpose , we assume that the survival distribution at birth is a dirac delta on a specific survival time @xmath137 , modulated sinusoidally in time , @xmath138 with @xmath139 . because all the elements exit after the prescribed time @xmath137 , this example is representative of a time - dependent plug - flow system . at the outlet ,
the age distribution at death @xmath51 is the time shifted @xmath52 , i.e. , @xmath140 while inside the system , the age distribution is given by @xmath141 where @xmath142 is the heaviside function @xcite .
the survival time distribution is @xmath143 while the joint distribution is given by @xmath144 see figure [ fig : deltaapp ] .
the transit time distribution can be calculated from ( [ eq : transdistr2 ] ) , @xmath145 and the conditional distribution , @xmath146 which reflects a deterministic relationship between age and survival , imposed by the survival distribution at birth ( see figure [ fig : deltaapp ] ) .
imposed as a dirac delta modulated by the sinusoidal amplitude in ( [ eq : impsurv ] ) , with @xmath147 , @xmath148 and @xmath149 .
the marginal distributions of age and survival time are also plotted on the corresponding vertical planes . ]
we now consider an extension of the first application , in which the input is constant but the loss function , although still independent of age , is now time periodic , @xmath150 from a hydrological point of view , this case is representative of a system with rainfall homogeneously distributed during the year but with seasonally modulated potential evapotranspiration and negligible other losses , @xmath151 the system is still well - mixed , although the transient conditions bring about additional complications that result in statistical dependence between age and survival . considering @xmath152 so that the system has forgotten the initial conditions and has settled on a periodic steady state , the age distribution is from ( [ eq : mkvfsol ] ) @xmath153 while the joint distribution can be obtained through equations ( [ eq : jointsol2 ] ) , @xmath154 the joint distribution is plotted in figure ( [ fig : jointappperiod ] ) for different days of the year corresponding to the four seasons .
they show that the systems has low values of age and survival for the season with high losses ( summer ) , while when the losses diminish ( winter ) , the age and survival start increasing again .
also visible is the asymmetry with respect to the bisector indicating statistical dependence between age and survival induced by the time - varying conditions .
the age , survival time and transit time distributions are plotted in figure [ fig : pdfsperiod ] .
the joint distribution integrated with respect to @xmath0 recovers the survival time distribution @xmath42 .
however , the integral does not appear to be elementary and here it was only solved numerically . finally , regarding the mean values ( figure [ fig : applperiod ] ) , the loss function being age - independent , the mean age at death is equal to the mean age , @xmath155 the mean survival time at birth @xmath156 was computed analytically , while the mean survival time @xmath115 was computed numerically , and they appeared to be equal . )
, calculated at four different times of the year .
@xmath157 d ( a ) , @xmath158 d ( b ) , @xmath159 ( c ) , @xmath160 d ( d ) .
the parameters @xmath161 , @xmath162 , @xmath163 and @xmath164 . ] ) , calculated at four different times of the year .
@xmath157 d ( a ) , @xmath158 d ( b ) , @xmath159 ( c ) , @xmath160 d ( d ) .
the parameters are the same as in figure [ fig : jointappperiod ] . ] )
.the parameters are the same as in figure [ fig : jointappperiod ] . ] in this last example , we analyze the role of the loss function under conditions of steady state . as shown in section [ sec : steadystate ] , in steady state , the age and survival time distributions are the same , and therefore similar considerations also apply to a survival - time dependence of the birth function .
we consider the age - dependent loss function @xmath165 with @xmath166 .
for @xmath167 , @xmath109 is a decreasing function , thus selecting younger elements for output , while it is an increasing function of age for @xmath168 with preference for older elements .
the age distributions is obtained from ( [ eq : agesteady ] ) @xmath169 whereas the age distributions at death is @xmath170 for @xmath167 , the age pdf at death is a stretched exponential distribution @xcite and , as can be verified through ( [ eq : relageedeath ] ) , the mean age at death @xmath106 is lower than the mean age @xmath107 . when @xmath168 , the age pdf at death is a weibull distribution @xcite and the mean age at death @xmath106 is greater than the mean age @xmath107 . in the limiting case @xmath171 , @xmath109 is a constant and the well - mixed system of section [ sec : linesteady ] is recovered .
in addition , the plug - flow system of section [ sec : appdelta ] is recovered when taking @xmath172 . for different values of @xmath173 , the age , age at death and transit time distributions
are plotted in figure [ fig : distrappl2 ] , while the joint distributions are plotted in figure [ fig : applossjoints ] .
the mean values are shown in figure [ fig : medieofc ] , where the mean age at death @xmath106 is lower than @xmath107 when @xmath167 , while it is larger than @xmath107 when @xmath168 , and tends asymptotically to @xmath105 for @xmath172 .
as shown in equation ( [ eq : reltransdeath ] ) , @xmath105 serves as an upper bound for both mean age and mean age at death . ) with parameter @xmath174 ( a ) , @xmath171 ( b ) , @xmath175 ( c ) and @xmath176 ( d ) . ]
( [ eq : imposedlossufunction2 ] ) , for @xmath171 ( a ) , @xmath175 ( b ) and @xmath176 ( c ) . ] , given the age dependent loss function ( [ eq : imposedlossufunction2 ] ) . ]
our main results regard the evolution equation of the joint distribution , equation ( [ eq : jointeq ] ) , which in turn allowed us to obtain the corresponding evolution equation for age ( mkvf ) and survival time , given by equations ( [ eq.mkvf ] ) and ( [ eq.mkvfsurv ] ) , respectively .
the theory naturally led us to consider the conditional distributions of age and survival , equations ( [ eq : conditional ] ) , ( [ eq : conddistr ] ) , and ( [ eq : condsol ] ) , which helped us clarify some of the statements in the literature about the statistical dependence of these two quantities @xcite .
we also obtained general relationships for the transit time distribution , equations ( [ eq : transdistr ] ) and ( [ eq : transdistr2 ] ) , and discussed the simplifications induced by the steady state conditions ; see equations ( [ eq : equaldistr ] ) , ( [ eq : equalcond ] ) , ( [ eq : equallossbirth ] ) , ( [ eq : transdistrsteadynorm ] ) and ( [ eq : ageconddistrsteady ] ) .
furthermore , we derived exact relationships among the means ( [ eq : reltransdeath ] ) and ( [ eq : relageedeath ] ) , although the latter was already known to @xcite .
the present theory is spatially implicit in that it considers , globally , entire populations or finite amounts of substance in a control volume @xcite .
interesting future work will consist of connecting it with the spatially explicitly formulation pioneered by @xcite ( and further developed by @xcite ) , as done in the case of the mkvf equation by @xcite and @xcite .
it also seems promising to explore the use of nonlinear formulations ( see @xcite and related contributions ) , in which mortality and birth functions also may depend on the total amount @xmath177 , that is @xmath178 .
such dependence of birth and mortality on global quantities may be linked to the nonlocal nature of the pressure equation in fluid mechanics @xcite .
indeed hydrological systems are known to behave such that the loss function at a point is non - local @xcite .
this is often tacitly assumed in spatially implicit , event - based formulations of rainfall - runoff , which use closure assumptions that depend on the total system storage ( _ bartlett et al .
, 2015 _ , in preparation ) .
finally , for realistic hydrologic applications it is necessary to include the effect of stochasticity from the external environment on the input and output functions @xcite . in this case ,
even the mean quantities become random variables with statistical distributions that may be of great theoretical and practical interest .
birkel , c. , d. tetzlaff , s. m. dunn , and c. soulsby ( 2011 ) , using time domain and geographic source tracers to conceptualize streamflow generation processes in lumped rainfall - runoff models , _ water resources research _ , _
47_(2 ) .
cornaton , f. , and p. perrochet ( 2006 ) , groundwater age , life expectancy and transit time distributions in advective dispersive systems : 1 .
generalized reservoir theory , _ advances in water resources _ , _ 29_(9 ) , 12671291 .
ginn , t. r. ( 1999 ) , on the distribution of multicomponent mixtures over generalized exposure time in subsurface flow and reactive transport : foundations , and formulations for groundwater age , chemical heterogeneity , and biodegradation , _ water resources research _ , _
35_(5 ) , 13951407 .
hrachowitz , m. , c. soulsby , d. tetzlaff , j. dawson , s. dunn , and i. malcolm ( 2009 ) , using long - term data sets to understand transit times in contrasting headwater catchments , _ journal of hydrology _ , _
367_(3 ) , 237248 .
maoszewski , p. , and a. zuber ( 1982 ) , determining the turnover time of groundwater systems with the aid of environmental tracers : 1 .
models and their applicability , _ journal of hydrology _ , _
57_(3 ) , 207231 .
mcdonnell , j. j. , and k. beven ( 2014 ) , debates the future of hydrological sciences : a ( common ) path forward ?
a call to action aimed at understanding velocities , celerities and residence time distributions of the headwater hydrograph , _ water resources research_. | although the concepts of age , survival and transit time have been widely used in many fields , including population dynamics , chemical engineering , and hydrology , a comprehensive mathematical framework is still missing . here
we discuss several relationships among these quantities by starting from the evolution equation for the joint distribution of age and survival , from which the equations for age and survival time readily follow .
it also becomes apparent how the statistical dependence between age and survival is directly related to either the age - dependence of the loss function or the survival - time dependence of the input function .
the solution of the joint distribution equation also allows us to obtain the relationships between the age at exit ( or death ) and the survival time at input ( or birth ) , as well as to stress the symmetries of the various distributions under time reversal .
the transit time is then obtained as a sum of the age and survival time , and its properties are discussed along with the general relationships between their mean values .
the special case of steady state case is analyzed in detail .
some examples , inspired by hydrologic applications , are presented to illustrate the theory with the specific results . |
in magnetic materials , in addition to the ordinary hall effect originating from the deflection of moving charge carriers by the lorentz force in a magnetic field , there is an anomalous term which is proportional to the spontaneous magnetization @xmath7 .
although this anomalous hall effect ( ahe ) has been recognized for more than a century,@xcite it is still a long - standing topic in condensed matter physics which attracted great interest due to the fundamental physics and great potential applications .
it is now widely accepted that there are three mechanisms responsible for the ahe.@xcite karplus and luttinger first proposed a model related to the band structure of ferromagnetic metals with the spin - orbit interaction ( soi ) to explain the ahe ( intrinsic kl mechanism).@xcite in modern language , it has been reinterpreted as a manifestation of berry - phase effects on occupied electronic bloch states.@xcite for this mechanism , the anomalous hall resistivity ( @xmath2 ) scales quadratically with longitudinal resistivity ( @xmath3 ) .
the ahe can also be induced by asymmetric scattering of conduction electrons which are subject to soi .
these extrinsic mechanisms for the ahe involve skew - scattering and side - jump scattering ones.@xcite for the former , the @xmath2 is linearly proportional to @xmath3 .
by contrast , the latter gives @xmath8 ( @xmath9 2 ) , similar to the intrinsic mechanism .
recently , the ahe of materials with frustrated structure and/or non - collinear magnetism , such as spin - ice like ferromagnetic nd@xmath1mo@xmath1o@xmath10,@xcite pr@xmath1ir@xmath1o@xmath10 with chiral spin - liquid state,@xcite and non - collinear antiferromagnetic mn@xmath0sn,@xcite have attracted much attention .
these systems exhibit large ahe like in the traditional ferromagnet possibly due to the effects of berry - phase originating from the non - coplanar spin sites with spin chirality .
these studies deepen our understanding of the relation between topological nature of the quantum states of matter and its transport properties , however , the effects of frustrated magnetism on the ahe , i.e. , the interaction between localized spins in the frustrated lattice and itinerant electrons , are still not fully solved because of limited materials systems and theoretical calculations until now .
ferromagnetic fe@xmath0sn@xmath1 with geometrically frustrated kagome bilayer of fe is a good candidate to study this issue .
previous neutron powder diffraction ( npd ) and mssbauer spectra show that the curie temperature of fe@xmath0sn@xmath1 polycrystal is about 640 k and it may be a frustrated ferromagnet with a temperature - dependent non - collinear spin structure.@xcite more surprisedly , fe@xmath0sn@xmath1 polycrystal exhibits ahe with unusual scaling relationship between the anomalous hall coefficient @xmath11 and the longitudinal resistivity @xmath3 , @xmath12 , i.e. , @xmath9 3.15 when the @xmath7 is almost unchanged with temperature.@xcite this large scaling index can not be explained by either one of three mechanisms introduced above . moreover
, theoretical calculation suggests that fe@xmath0sn@xmath1 could be a model system to exhibit flat - band ferromagnetism , leading to the fractional quantum hall states at high temperatures ( maybe even room temperatures).@xcite these results motivate us to study the ahe of fe@xmath0sn@xmath1 in detail . in this work ,
we report the results of transport and magnetic properties of fe@xmath0sn@xmath1 single crystal . distinctly different from @xmath9 3.15 in polycrystal ,
the quadratically scaling behavior between anomalous hall resistivity @xmath2 and @xmath3 are observed in fe@xmath0sn@xmath1 single crystal and our analysis implies that the main contribution of observed ahe should originate from the intrinsic kl mechanism .
the large hall conductivity might be intimately related to the topologically nontrivial bands in fe@xmath0sn@xmath1 , originating from the ferromagnetism with geometrically frustrated kagome bilayer of fe .
single crystals of fe@xmath0sn@xmath1 were grown by the self - flux method with fe : sn = 4 : 96 molar ratio .
fe ( purity 99.98 % ) and sn ( purity 99.99 % ) grains were put into alumina crucible and sealed in quartz ampoule under partial argon atmosphere .
the sealed quartz ampoule was heated to and soaked at 1323 k for 4 h , then cooled down to 1043 k with 6 k / h .
at this temperature , the ampoule was taken out from the furnace and decanted with a centrifuge to separate fe@xmath0sn@xmath1 crystals from sn flux .
x - ray diffraction ( xrd ) of a single crystal was performed using a bruker d8 x - ray machine with cu @xmath13 radiation ( @xmath14 0.15418 nm ) at room temperature .
magnetization and electrical transport measurements were carried out by using quantum design ppms-9 and mpms - xl5 .
the longitudinal and hall electrical resistivity were performed using a four - probe method on single crystals cutting into rectangular shape with the dimensions of [email protected]@xmath150.4 mm@xmath16 .
the current flows in the @xmath6 plane of hexagonal structure . for the measurement of hall resistivity @xmath17 , in order to effectively get rid of the longitudinal resistivity contribution due to voltage probe misalignment , the hall resistivity was obtained by the difference of transverse resistance measured at the positive and negative fields , i.e. , @xmath18/2 $ ] .
sn@xmath1 ( left ) and kagome layer of fe atoms ( right ) .
the small red and big green balls represent fe and sn atoms , respectively .
( b ) xrd pattern of a fe@xmath0sn@xmath1 single crystal .
inset : photo of typical fe@xmath0sn@xmath1 single crystals . ] as shown in fig .
1(a ) , the crystal structure of fe@xmath0sn@xmath1 is composed of fe - sn bilayer and sn layer stacking along the @xmath5-axis direction alternatively . in fe - sn bilayer ,
the fe atoms form two kagome layers with two kinds of equilateral triangles which have different fe - fe bond lengths ( 0.2732 and 0.2582 nm for orange and blue ones , respectively ) .
the sn atoms occupy the centers of the hexagons in the kagome layers .
moreover , the triangles of fe atoms with small side length in two layers connect each other and make up the octahedra of fe atoms . on the other hand , there is another sn layers in between two fe - sn bilayer .
the sn atoms in this layer form a two - dimensional network of six - member ring connected each other by edge - sharing , which is same as in graphene .
1(b ) shows the xrd pattern of a fe@xmath0sn@xmath1 single crystal .
all of peaks can be indexed by the indices of ( 0
0 l ) lattice planes .
it indicates that the crystal surface is parallel to the @xmath6 plane and perpendicular to the @xmath5-axis .
the inset of fig .
1(b ) shows a photograph of typical fe@xmath0sn@xmath1 crystals on a 1 mm grid paper .
it can be seen that these crystals have hexagonal shape with small thickness , consistent with the layered structure and the rhombohedral symmetry of fe@xmath0sn@xmath1 with space group @xmath19 . of fe@xmath0sn@xmath1 single crystal at zero field and 9 t along the @xmath5 axis .
( b ) magnetoresistance of @xmath20 at 5 and 400 k for @xmath21 .
( c ) temperature dependence of magnetic susceptibility @xmath22 for fe@xmath0sn@xmath1 single crystal in the high - temperature region ( 300 - 850 k ) with zfc mode at @xmath23 0.1 t for @xmath24 and @xmath21 .
inset : corresponding @xmath25 vs. @xmath26 .
( d ) temperature dependence of @xmath22 with zfc and fc modes at @xmath23 0.01 t for both field directions .
inset : @xmath22 vs. @xmath26 with zfc mode at @xmath23 2 t for both field directions . ]
electrical resistivity @xmath27 in the @xmath6 plane as a function of temperature for fe@xmath0sn@xmath1 single crystal is shown in fig .
the @xmath27 at zero field exhibits metallic behavior with much smaller absolute value than that of polycrystal,@xcite partially due to the absence of grain boundary ( gb ) effects . the residual resistivity ratio ( rrr ) that @xmath28 is about 21 , indicating the high quality of sample . at @xmath23 9 t ,
the metallic behavior of @xmath27 is almost unchanged ( fig .
2(a ) ) . but the signs of magnetoresistance ( mr @xmath29 ) at low and high temperatures are opposite ( fig .
2(b)).the mr at 5 k is positive and reaches about 80 % at @xmath23 9 t. in contrast , there is a negative mr with small value ( @xmath30 2 % at @xmath23 9 t ) at 400 k. the positive mr at low temperature should be originate from the lorenz force caused by magnetic field on the carrier motion . on the other hand , the negative mr becomes obvious at high temperature near @xmath31 , because of the suppression of spin scattering in ferromagnetic fe@xmath0sn@xmath1 by applying a magnetic field .
moreover , it can be seen that there is a slope change of @xmath20 at @xmath32 0.7 t , because the magnetization of fe@xmath0sn@xmath1 becomes saturated when field further increases . in other words ,
once the magnetic moment is saturated , further increase of field will have much weaker influence on the alignment of spin , the change of mr with field would also become smaller
. high - temperature magnetic susceptibility @xmath33 and @xmath34 with zero - field - cooling ( zfc ) mode at @xmath23 0.1 t for @xmath24 and @xmath21 are shown in fig .
there are sharp increases for both @xmath33 and @xmath34 , corresponding to the ferromagnetic transition in fe@xmath0sn@xmath1 .
the curies temperature @xmath31 determined from the peaks of @xmath35 and @xmath36 is about 641 and 643 k , respectively ( inset of fig .
it is very close to the @xmath31 of fe@xmath0sn@xmath1 polycrystal.@xcite surprisedly , when temperature is below @xmath31 , the @xmath22 start to decrease with temperature for both field directions at beginning and then increase when @xmath37 ( @xmath30 423 k ) .
this behavior is also observed in fe@xmath0sn@xmath1 polycrystal with slightly different @xmath38 and it was ascribed to the rotation of the fe moments from the @xmath5 axis to the @xmath6 plane with decreasing temperature , consistent with previous results of npd and mssbauer spectra.@xcite when @xmath39 100 k , the zfc and field - cooling ( fc ) @xmath22 curves for both field directions at @xmath23 0.01 t ( fig .
2(d ) ) still almost overlap with only slight differences between each other .
it is different from the obviously branching behavior between zfc and fc @xmath22 curves of fe@xmath0sn@xmath1 polycrystal in which there is a spin glass ( sg ) state appearing below 80 k.@xcite the much weaker branching behavior in single crystal than in polycrystal implies that there may be a frustrated state rather than sg state because of the spin disorder at gbs in polycrystal might have significant influence on magnetic properties at low temperature . on the other hand ,
the @xmath34 exhibits more obviously downturn behavior , which could be ascribed to the accelerated process of spin reorientation from the @xmath5 axis to the @xmath6 plane.@xcite when field increases to 2 t , both @xmath33 and @xmath34 exhibit typical ferromagnetic behavior in the whole temperature range ( inset of fig .
2(d ) ) , similar to previous results.@xcite it implies that the spin reorientation process is suppressed by the high magnetic field .
and ( b ) hall resistivity @xmath40 as a function of magnetic induction @xmath41 for fe@xmath0sn@xmath1 single crystal at various temperatures when @xmath21 .
the blue dashed line in ( b ) is the linear fit of @xmath40 at high-@xmath41 region when @xmath42 350 k. ] fig .
3(a ) exhibits the effective field dependence of magnetization @xmath43 at various temperatures between 5 and 350 k for @xmath21 .
here @xmath44 , where @xmath45 is the demagnetizing factor and the calculated value of @xmath45 is 0.65 . because of @xmath46 , the shape of @xmath43 curves is typical for ferromagnets ,
i.e. , a rapid increase at low field region with a saturation in higher magnetic fields because most of fe moments have been aligned along the field direction .
the saturation magnetization @xmath47 increases with decreasing temperature , well consistent with the temperature dependence of @xmath34 curve at @xmath23 2 t ( inset of fig .
2(d ) ) as well as the trend of magnetic moment obtained from npd.@xcite but it is in sharp contrast to the results of @xmath43 for fe@xmath0sn@xmath1 polycrystal in which the @xmath47 becomes smaller with decreasing temperature . on the other hand , the @xmath47 at the lowest measuring temperature ( 5 k ) for @xmath21 is about 2.1 @xmath48/fe , in good agreement with those values in the literature.@xcite .
hall resistivity @xmath40 as a function of magnetic induction @xmath41 for fe@xmath0sn@xmath1 single crystal at various temperatures is shown in fig .
3(b ) . here
the @xmath40 increases quickly to certain saturated values at low @xmath41 region . with increasing @xmath41 further ,
the @xmath40 decreases slightly and the @xmath41 dependence of @xmath40 is almost linear , i.e. , the @xmath50 is constant . on the other hand ,
when lowering temperature , the saturated value of @xmath40 becomes smaller gradually .
these behaviors are qualitatively consistent with the results in fe@xmath0sn@xmath1 polycrystal , however , the saturated values at each temperature are much larger than those for polycrystal.@xcite this difference can be partially ascribed to the effects of gb scattering on @xmath40 . moreover , at 5 k , the inflecting behavior of @xmath40 becomes very small and the high-@xmath41 value is negative , suggesting that the dominant carrier in fe@xmath0sn@xmath1 is electron .
it should be noted that the saturated value of @xmath40 for fe@xmath0sn@xmath1 single crystal at 350 k is enormous , much larger than that of typical itinerant ferromagnets , such as fe and ni.@xcite there is a similarity between the shapes of @xmath40 and @xmath43 curves .
it clearly indicates that there is an ahe in itinerant ferromagnet fe@xmath0sn@xmath1 . in general ,
the hall resistivity @xmath17 in the ferromagnets is made up of two parts,@xcite @xmath51 where @xmath52 and @xmath2 is the ordinary and anomalous hall resistivity , respectively .
@xmath53 is the ordinary hall coefficient from which apparent carrier concentration and type can be determined ( @xmath54 ) , and @xmath11 is the anomalous hall coefficient . as shown in fig .
3(b ) , the values of @xmath53 and @xmath2 in principle can be determined from the linear fit of @xmath40 curve at the saturation region .
the slope and @xmath55-axis intercept is corresponding to the @xmath53 and @xmath2 , respectively .
the @xmath11 can be obtained by using the formula @xmath56 with the @xmath47 taken from the @xmath43 curves at @xmath57 2 t ( fig .
the @xmath58 is negative at various temperature , in contrast to the positive @xmath59 ( fig . 4(a ) and
the former confirms that the electron - type carrier is dominant .
the @xmath58 shows weak temperature - dependence above 100 k. when @xmath39 100 k , the absolute value of @xmath53 becomes larger . using the values of @xmath53 , the @xmath60
is evaluated ( inset of fig .
it is insensitive to the temperature ( @xmath30 1.8@xmath1510@xmath61 @xmath62 , corresponding to 1.47 carriers per fe@xmath0sn@xmath1 ) when @xmath63 150 k and increases slightly with decreasing temperature down to 100 k. then the @xmath64 reduces sharply with further lowering temperature below 100 k. the strong variation of @xmath64 below 100 k is possibly due to the influence of spin reorientation on fermi surface , leading to the change of carrier concentration . on the other hand ,
the @xmath59 decreases with decreasing temperature monotonically and approaches to almost zero at low temperature ( fig .
the value of @xmath11 at 0.5@xmath65 ( @xmath30 320 k ) is about 1@xmath1510@xmath66 @xmath67 cm g@xmath68 , somewhat larger than fe@xmath0sn@xmath1 polycrystal.@xcite importantly , at same reduced temperature , fe@xmath0sn@xmath1 single crystal has about two orders of magnitude large @xmath11 when compared to the conventional itinerant ferromagnets , such as pure fe and ni.@xcite and @xmath59 from @xmath69 curves using eq . (
inset of ( a ) : derived @xmath64 from @xmath58 .
( c ) plot of log@xmath70 vs. log@xmath27 .
the blue and green solid lines are the fits using the relation @xmath71 below and above 100 k , respectively .
( d ) anomalous hall conductivity @xmath72 as a function of temperature .
inset : temperature dependence of @xmath73 . ] fig .
4(c ) shows the scaling behavior of @xmath2 vs. @xmath3 . there is a jump at about 100 k corresponding to the temperature where the process of spin reorientation from the @xmath5 axis to the @xmath6 plane
is almost completed , as observed in the @xmath22 curves . using the formula @xmath74 ,
the fits for @xmath75 100 k and @xmath39 100 k give @xmath76 1.92(2 ) and @xmath77 1.94(5 ) , respectively . both nearly quadratic relationships between @xmath2 and @xmath3
clearly indicates that the intrinsic kl or extrinsic side - jump mechanism dominates the ahe in fe@xmath0sn@xmath1 single crystal rather than extrinsic skew - scattering mechanism which will give the linear relationship between @xmath2 and @xmath3 .
by contrast , the fe@xmath0sn@xmath1 polycrystal exhibits the scaling behavior with much larger @xmath78 (= 3.15),@xcite similar to those has also been observed in some inhomogeneous systems , such as fe / cr multilayers , co - ag granular films.@xcite zhang@xcite proposed that in the magnetic multilayered systems , the @xmath2 will depend on the ratio of relaxation times or scattering potentials in magnetic and nonmagnetic layers in the long mean - free path limit where the mean - free path @xmath79 is much larger than the layer thickness , and the scaling behavior with @xmath9 2 will breakdown .
this is one of possible explanations for the large @xmath78 in fe@xmath0sn@xmath1 polycrystal when regarding the spin - disordered gbs and the gains of fe@xmath0sn@xmath1 as the nonmagnetic and magnetic layers , and the @xmath79 is comparable or larger than the thickness of gb layer or gain size . by contrast , although the @xmath79 of single crystal is larger than that of polycrystal , fe@xmath0sn@xmath1 single crystal can be regarded as a uniform magnet .
it is more homogeneous than polycrystal without any heterostructure like gbs and the crystal size should be much larger than the @xmath79 . thus , the local limit is valid and the anomalous hall conductivity ( ahc ) @xmath4 ( @xmath80 ) is independent of the scattering potential or the relaxation time , i.e. , @xmath9 2 for fe@xmath0sn@xmath1 single crystal .
temperature dependence of @xmath4 is shown in fig .
above 100 k , the @xmath4 is about -170 @xmath81 cm@xmath68 , insensitive to temperature .
when @xmath26 is below 100 k , it increases quickly to about -400
@xmath81 cm@xmath68 .
this variation is also reflected in the scaling relationship between @xmath2 and @xmath3 , leading to the jump of prefactor @xmath83 significantly because the @xmath7 is almost unchanged with temperature between 5 and 350 k. theoretically , the intrinsic contribution of @xmath84 is of the order of @xmath85 , where @xmath86 is the electronic charge , @xmath87 is the plank constant and @xmath88 is the lattice constant.@xcite taking @xmath89 12.8 @xmath90 approximately , the @xmath84 is @xmath30 302 @xmath81 cm@xmath68 , close to the value of @xmath91 at 5 k ( @xmath30 402 @xmath81 cm@xmath68 ) . by contrast
, the extrinsic side - jump contribution of @xmath92 has been shown to be on the order of @xmath93 , where @xmath94 and @xmath95 is the soi and fermi energy , respectively.@xcite since the @xmath96 is usually less than 10@xmath97 for the metallic ferromagnets , the extrinsic side - jump contribution should be small and the ahe of fe@xmath0sn@xmath1 is dominated by the intrinsic berry - phase kl contribution .
it would be interesting to carry out the theoretical calculation to verify this analysis , as did for bcc fe and srruo@xmath0.@xcite the dramatic increase of the @xmath4 at 100 k is unclear at present .
since the large contribution to @xmath98 comes from the enhanced berry curvature near the avoided crossings of band dispersions at @xmath95 which only occur in very small regions of @xmath99 space , the @xmath98 is very sensitive to the topology of fermi surface.@xcite the subtle changes of fermi surface could happen due to the spin - reorientation transition below 100 k , leading to the variation of @xmath4 .
the decrease of @xmath60 also reflects this change of fermi surface to some extent ( inset of fig .
4(a ) ) . for the intrinsic ahc ,
the @xmath84 is proportional to the @xmath7,@xcite the scaling coefficient @xmath100 should be constant and independent of temperature . as shown in the inset of fig .
4(d ) , when @xmath26 is away from 100 k , the @xmath101 is nearly constant , confirming the intrinsic kl mechanism is dominant contribution of ahe . the sudden change of @xmath101 at @xmath30 100 k is related to the evolution of @xmath84 .
the value of @xmath101 is comparable with those in the traditional itinerant ferromagnets , such as fe and ni ( @xmath102 0.01 - 0.2 v@xmath68).@xcite thus , even there is rather large @xmath11 in fe@xmath0sn@xmath1 single crystal in contrast to fe and ni , the larger @xmath3 in the former leads to similar values of @xmath101 in these materials .
previous theoretical studies predicted that an integer or fractional quantum hall ( iqh or fqh ) states can appear at high temperature in the ferromagnet with the two - dimensional geometrically frustrated kagome lattice.@xcite the iqh is corresponding to the fully filled flatband with a nonzero chern number and the fqh state results from the 1/3 or 1/2 filling of that band .
tang et al . proposed that fe@xmath0sn@xmath1 is one of possible materials realizing these qh states.@xcite present study indicates that instead of the appearance of qh states , the intrinsic ahe exists in fe@xmath0sn@xmath1 .
the metallic behavior and the different filling of the flatband in fe@xmath0sn@xmath1 from theoretically required 1/3 or 1/2 filling could be related to the absence of iqh or fqh states . moreover
, the finite interlayer interaction between kagome bilayer of fe could also have some influences on the band gap @xmath103 , electron - electron interaction @xmath104 and band width @xmath105 and the condition of @xmath106 favoring the fqh state may not be satisfied . because the ahe can be regarded as the unquantized version of the qh effect once the fermi energy is not lying in the gap,@xcite both of the ahe observed in experiment and
the qh states predicted by theory reflect the nontrivial topology of bands for the ferromagnetic system with geometrically frustrated kagome lattice.@xcite
in summary , we successfully grow the fe@xmath0sn@xmath1 single crystal using the self - flux method .
the fe@xmath0sn@xmath1 contains geometrically frustrated kagome bilayers of fe and has a ferromagnetic transition at about 640 k followed by another spin - reorientation transition from @xmath5 axis to @xmath6 plane at the lower temperature .
the transport and magnetic measurement indicates that fe@xmath0sn@xmath1 single crystal exhibits a scaling relationship between @xmath2 and @xmath3 , @xmath107 .
the detailed analysis suggests that the ahe should be dominated by the intrinsic kl mechanism and there is a large @xmath4 .
further theoretical calculation is highly expected in order to clarify whether such large @xmath4 is related to the topologically nontrivial bands in fe@xmath0sn@xmath1 , which originates from the ferromagnetism with geometrically frustrated kagome bilayer of fe . moreover , although the qh states are absent in the undoped fe@xmath0sn@xmath1 , adjusting the band filling by doping could be a way to realize the theoretically predicted qh states . on the other hand ,
the @xmath4 exhibits a sudden change at about 100 k , corresponding to the ending temperature of spin - reorientation transition .
this jump of @xmath4 is tentatively explained by the evolution of fermi surface .
this work was supported by the ministry of science and technology of china ( 2012cb921701 , 2016yfa0300504 ) , the fundamental research funds for the central universities , and the research funds of renmin university of china ( ruc ) ( 15xnlf06 , 15xnlq07 ) , and the national natural science foundation of china ( grant no .
11574394 ) . | the anomalous hall effect is investigated for ferromagnetic fe@xmath0sn@xmath1 single crystal with geometrically frustrated kagome bilayer of fe .
the scaling behavior between anomalous hall resistivity @xmath2 and longitudinal resistivity @xmath3 is quadratic and further analysis implies that the ahe in fe@xmath0sn@xmath1 single crystal should be dominated by the intrinsic karplus - luttinger mechanism rather than extrinsic skew - scattering or side - jump mechanisms .
moreover , there is a sudden jump of anomalous hall conductivity @xmath4 appearing at about 100 k where the spin - reorientation transition from the @xmath5 axis to the @xmath6 plane is completed .
this change of @xmath4 might be related to the evolution of fermi surface induced by the spin - reorientation transition . |
current standard sets of parton distribution function ( pdf ) do not include uncertainties @xcite . in practice ,
as long as the pdf s are used to calculate observables that themselves have large experimental uncertainties this shortcoming is obviously not a problem . in the past
the precision of the hadron collider data was such that there was no ostensible need for the pdf uncertainties , as was testified by the good agreement between the theory and measurements .
however , the need for pdf uncertainties became apparent with the measurement of the one jet inclusive transverse energy at the tevatron @xcite . at large transverse jet energies the data was significantly above the theoretical prediction , a possible signal for new physics .
the deviation was ultimately `` fixed '' by changing the pdf s in such a manner that they still were consistent with the observables used to determine the pdf @xcite .
this is a reflection of the significant pdf uncertainties for this observable . knowing the uncertainties on the pdf s would have cleared the situation immediately .
note that once the data is used in the pdf fit , it can not be used for other purposes . specifically , setting limits on possible physics beyond the standard model . in that case
, one should fit the pdf s and the new physics simultaneously .
the technique presented in this paper is well suited for this sort of problem .
the spread between different sets of pdf s is often associated with pdf uncertainties .
currently , this is what is used for the determination of the pdf uncertainty on the @xmath0-boson mass at the tevatron .
it is not possible to argue that this spread is an accurate representation of all experimental and theoretical pdf uncertainties . for the next planned high luminosity run at fermilab , assuming an integrated luminosity of @xmath1 , the expected 40 mev uncertainty on the @xmath0-boson mass is dominated by a 30 mev production model uncertainty .
the latter uncertainty itself is dominated by the pdf uncertainty , estimated to be 25 mev @xcite .
this determination of the pdf uncertainty is currently nothing more than an educated guess .
it is made by ruling out existing pdf s using the lepton charge asymmetry in @xmath0-boson decay events .
the spread of the remaining pdf s determines the uncertainty on the extracted @xmath0-boson mass . because the pdf uncertainty seems to be the dominant source of uncertainty in the determination of the @xmath0-boson mass ,
such a procedure must be replaced by a more rigorous quantitative approach .
the method described in this paper is well suited for this purpose . in this paper , using the framework of statistical inference @xcite , we illustrate a method that can be used for many purposes .
first of all , it is easy to propagate the pdf uncertainties to a new observable without the need to calculate the derivative of the observable with respect to the different pdf parameters .
secondly , it is straightforward to assess the compatibility of new data with the current fit and determine whether the new data should be included in the fit .
finally , the new data can be included in the fit without redoing the whole fit .
this method is significantly different from more traditional approaches to fit the pdf s to the data .
it is very flexible and beside solving the problems already mentioned , it offers additional advantages .
first , the experimental uncertainties and the probability density distributions for the fitted parameters do not have to be gaussian distributed .
however , such a generalization would require a significant increase in computer resources .
second , once a fit has been made to all the data sets , a specific data set can be easily excluded from the fit .
such an option is important in order to be able to investigate the effect of the different data sets .
this is particularly useful in the case of incompatible new data . in that case
one can easily investigate the origin of the incompatibility .
finally , because it is not necessary to redo a global fit in order to include a new data set , experimenters can include their own new data into the pdf s during the analysis phase .
the outline for the rest of the paper is as follows . in sec .
[ sect : method ] , we describe the inference method . the flexibility and simplicity of the method is illustrated in sec .
[ sect : expanding ] , by applying it to the cdf one jet inclusive transverse jet energy distribution @xcite and the cdf lepton charge asymmetry data @xcite . in sec .
[ sect : conclusions ] we draw our conclusions and outline future improvements and extensions to our method .
statistical inference requires an initial probability density distribution for the pdf parameters
. this initial distribution can be rather arbitrary , in particular it can be solely based on theoretical considerations .
once enough experimental data are used to constrain the probability density distribution of the parameters the initial choices become irrelevant . obviously , the initial choice does play a role at intermediate stages .
the initial distribution can also be the result of a former fit to other data .
the data that we will use later in this paper do not constrain the pdf s enough by themselves to consider using an initial distribution based only on theory .
the final answer would depend too much on our initial guess .
we therefore decided to use the results of ref .
@xcite . in this work
the probability density distribution was assumed to be gaussian distributed and was constrained using deep inelastic scattering ( dis ) data .
all the experimental uncertainties , including correlations , were included in the fit , but no theoretical uncertainties were considered . the fact that no tevatron data were used allows us to illustrate the method with tevatron data .
we briefly summarize ref .
@xcite in the appendix . in sec .
2.1 we explain the propagation of the uncertainty to new observables .
2.2 shows how the compatibility of new data with the pdf can be estimated . finally , in sec .
2.3 we demonstrate how the effect of new data can be included in the pdf s by updating the probability density distribution of the pdf parameters .
we now assume that the pdf s are parametrized at an initial factorization scale @xmath2 , with @xmath3 parameters , @xmath4 and that the probability density distribution is given by @xmath5 . note that @xmath5 does not have to be a gaussian distribution
. by definition @xmath5 is normalized to unity , @xmath6 where the integration is performed over the full multi - dimensional parameter space and @xmath7 . to calculate the parameter space integrals we use a monte - carlo ( mc ) integration approach with importance sampling .
we generate @xmath8 random sets of parameters @xmath9 distributed according to @xmath10 .
this choice should minimize the mc uncertainty for most of the integrals we are interested in . for reference
we also generate one set at the central values of the @xmath9 , the @xmath11 .
the number of parameter sets to be used depends on the quality of the data .
the smaller the experimental uncertainty is compared to the pdf uncertainty , the more pdf s we need .
we must ensure a sufficient fraction of pdf s span the region of interest ( i.e. close to the data ) .
for the purposes of this paper , we found that @xmath12 is adequate .
clearly , to each of the @xmath8 sets of parameters @xmath9 correspond a set of unique pdf s .
each of these pdf sets have to be evolved using the altarelli - parisi evolution equations .
we used the cteq package to do this evolution @xcite .
we now can evaluate any integral @xmath13 over the parameter space as a finite sum @xcite @xmath14 with @xmath15 is the @xmath16-th random set of @xmath9 .
the function @xmath17 represents an integrable function of the pdf parameters .
the uncertainty on the integral @xmath13 due to the mc integration is given by , @xmath18 for any quantity , @xmath19 , that depends on the pdf parameters @xmath20 ( for example an observable , one of the flavor pdf s or for that matter one of the parameter itself ) , the theory prediction is given by its average value , @xmath21 , and its uncertainty , @xmath22 : @xmath23 note that @xmath21 is not necessarily equal to the value of @xmath24 evaluated at the central value of the @xmath9 .
however , this is how observables are evaluated if one has only access to pdf s without uncertainties .
given @xmath25 , another quantity calculable from the @xmath9 , the covariance of @xmath19 and @xmath25 is given by the usual expression : @xmath26 the correlation between @xmath19 and @xmath25 is given by @xmath27 .
for example , this can be used to calculate the correlation between two experimental observables , between an observable and one of the pdf parameters , or between an observable and a specific flavor pdf at a fixed bjorken-@xmath28 . using eq . [
eq : deltai ] , the mc uncertainty on the average and ( co)variance is given by @xmath29 the mc technique presented in this sub - section , gives a simple way to propagate uncertainties to a new observable , without the need for calculating the derivatives of the observable with respect to the parameters .
we will assume that one or several new experiments , not used in the determination of the initial probability density distribution , have measured a set of @xmath30 observables @xmath31 .
the experimental uncertainties , including the systematic uncertainties , are summarized by the @xmath32 experimental covariance matrix @xmath33 .
note that the correlations between experiments are easily incorporated .
here however , we have to assume that the new experiments are not correlated with any of the experiments used in the determination of @xmath34 .
the probability density distribution of @xmath35 is given by @xmath36 where @xmath37 is the conditional probability density distribution ( often referred to as likelihood function ) .
this distribution quantifies the probability of measuring the specific set of experimental values \{@xmath38 } given the set of pdf parameters @xmath9 . in pdf sets without uncertainties
, @xmath10 is a delta function and @xmath39 . instead of dealing with the probability density distribution of eq .
[ eq : pxe ] , one often quotes the confidence level to determine the agreement between the data and the model .
the confidence level is defined as the probability that a repeat of the given experiment(s ) would observe a worse agreement with the model . the confidence level of \{@xmath38 }
is given by @xmath40 where @xmath41 is the confidence level of \{@xmath38 } given @xmath9 .
if @xmath42 is larger than an agreed value , the data are considered consistent with the pdf and can be included in the fit .
if it is smaller , the data are inconsistent and we have to determine the source of discrepancy . for non - gaussian uncertainties the calculation of the confidence level might be ambiguous . in this paper
we assume that the uncertainties are gaussian .
the conditional probability density distribution and confidence level are then given by @xmath43 where @xmath44 is the chi - squared of the new data .
the theory prediction for the @xmath45-th experimental observable , @xmath46 , is calculated using the pdf set given by the parameters @xmath9 .
the matrix @xmath47 is the inverse of the total covariance matrix , @xmath48 , which in turn is given by the sum of the experimental , @xmath33 , and theoretical , @xmath49 , covariance matrix .
we assume that there is no correlation between the experimental and theoretical uncertainties .
we will use a minimal value of 0.27% on the confidence level , corresponding to a three sigma deviation , as a measure of compatibility of the data with the theory .
if the new data are consistent with the theory prediction then the maximum of the distribution of the @xmath50 should be close to @xmath30 ( within the expected @xmath51 uncertainty ) .
the standard deviation of @xmath50 , @xmath52 , tells us something about the relative size of the pdf uncertainty compared to the size of the data uncertainty .
the larger the value of @xmath52 is compared to @xmath51 , the more the data will be useful in constraining the pdf s . note that if there are several uncorrelated experiments , the total @xmath50 is equal to the sum of the @xmath50 of the individual experiments and the conditional probability is equal to the product of the individual conditional probabilities .
once we have decided that the new data are compatible with the initial pdf s , we can constrain the pdf s further .
we do this within the formalism of statistical inference , using bayes theorem .
the idea is to update the probability density distribution taking into account the new data .
this new probability density distribution is in fact the conditional probability density distribution for the @xmath9 considering the new data \{@xmath38 } and is given directly by bayes theorem @xmath53 where @xmath54 , defined in eq .
[ eq : pxe ] , acts as a normalization factor such that @xmath55 is normalized to one . because @xmath56 is normalized to unity , we can replace @xmath37 in eq .
[ eq : pnew ] simply by @xmath57 .
this factor acts as a new weight on each of the pdf s .
we can now replace @xmath10 by @xmath56 in the expression for the average , standard deviation and covariance given in sec .
2.1 and obtain predictions that include the effect of the new data . with the mc integration technique described before , these quantities can be estimated by weighted sums over the @xmath8 pdf sets @xmath58 where the weights are given by @xmath59 note that for the calculation of the monte - carlo uncertainty of the weighted sums , the correlation between the numerator and denominator in eq .
[ eq : newmusig ] has to be taken into account properly .
our strategy is very flexible .
once the theory predictions @xmath60 using the @xmath8 pdf sets are known for each of the experiments , it is trivial to include or exclude the effect of one of the experiments on the probability density distribution .
if the different experiments are uncorrelated then all what is needed is the @xmath50 of each individual experiments for all the pdf sets . in that case , each experiment is compressed into @xmath8 @xmath50 values .
one other advantage is that all the needed @xmath61 can be calculated beforehand in a systematic manner , whereas standard chi - squared or maximum likelihood fits require many evaluations of @xmath62 during the fit as the parameters are changed in order to find the extremum .
these methods are not very flexible , as a new fit is required each time an experiment is added or removed .
the new probability density distribution of the pdf parameters is gaussian if the following three conditions are met .
first , the initial probability density distribution , @xmath10 , must be gaussian .
second , all the uncertainties on the data points must be gaussian distributed ( that includes systematic and theoretical uncertainties ) .
finally , the theory predictions , @xmath63 , must be linear in @xmath9 in the region of interest .
this last requirement is fulfilled once the pdf uncertainties are small enough . for the studies in this paper
all three requirements are fulfilled .
the new probability density distribution can therefore be characterized by the average value of the parameters and their covariance matrix , which can be calculated , together with their mc integration uncertainty , using eq .
[ eq : newmusig ] .
once the new values of the average and the covariance matrix have been calculated , a new set of pdf parameters can be generated according to the new distribution and used to make further prediction instead of using the initial set of pdf with the weights . an alternative way to generate a pdf set distributed according to @xmath56 is to unweight the now weighted initial pdf set .
the simplest way to unweight the pdf sets is to use a rejection algorithm .
that is , define @xmath64 as the largest of the @xmath8 weights given in eq .
next generate for each pdf set a uniform stochastic number , @xmath65 , between zero and one .
if the weight @xmath66 is larger or equal to @xmath67 we keep pdf set @xmath45 , otherwise it is discarded .
the surviving pdf s are now distributed according to @xmath56 .
the number of surviving pdf s is on average given by @xmath68 .
we can now apply all the techniques of the previous sub - sections , using the new unweighted pdf set .
the mc integration uncertainties are easily estimated using the expected number of surviving pdf s . in the extreme case
that @xmath64 is close to one and only a few pdf survive the unweighting procedure , the number of initial pdf s must be increased .
the other extreme occurs when all the weights are approximately equal , i.e. @xmath69 . in that case
the new data puts hardly any additional constraints on the pdf .
the @xmath50 is only used to calculate the weight of a particular pdf , so that the new probability density distribution of the pdf parameters can be determined .
we do not perform a chi - squared fit . however , if the new probability density distribution of the parameters is gaussian distributed then our method is equivalent to a chi - squared fit . in that case
the average value of the parameters correspond to the maximum of the probability density distribution .
the minimum chi - squared can be estimated ( with mc uncertainties ) from the average @xmath50 calculated with the new probability density distribution .
indeed , by definition this average must be equal to the minimum chi - squared , @xmath70 , plus the known number of parameter .
note that the variance of the @xmath50 must itself be equal to twice the number of parameters . to obtain the overall minimum chi - squared , the value of the minimum chi - squared of the initial fit must be added to @xmath70 .
as long as the confidence level of the new data that were included in the fit is sufficiently high , the overall minimum chi - squared obtained is guaranteed to be in accordance with expectations was within expectations . ] .
the viability of the method described in sec . 2 is studied using two cdf measurements . in sec .
3.1 the one jet inclusive transverse energy distribution is considered , while the lepton charge asymmetry in @xmath0-boson decay is examined in sec .
3.2 . the statistical , systematic , and theoretical uncertainties on the observables will be taken into account .
the cdf results on the one jet inclusive transverse energy distribution @xcite demonstrated the weakness of the current standard pdf sets due to the absence of uncertainties on the pdf parameters .
the observables are the inclusive jet cross section at different transverse energies ,
@xmath71 @xmath72 we first have to construct the experimental covariance matrix , @xmath73 , using the information contained in ref .
the paper lists the statistical uncertainty at the different experimental points , @xmath74 , together with eight independent sources of systematic uncertainties , @xmath75 .
hence , the experimental measurements , @xmath76 are given by @xmath77 where as before , @xmath78 is the theoretical prediction for the observable calculated with the set of parameters @xmath9 .
the @xmath79 and @xmath80 are independent random variables normally distributed with zero average and unit standard deviation .
note that some of the systematic uncertainties given in ref .
@xcite are asymmetric . in those cases
we symmetrized the uncertainty using the average deviation from zero . from eq .
[ eq : xie ] we can construct the experimental covariance matrix @xmath81 we also need to estimate the theoretical uncertainty . in eq . [
eq : xie ] no theoretical uncertainties were taken into account .
we consider two types of uncertainties : the uncertainty due to the numerical monte carlo integration over the final state particle phase space , @xmath82 , and the renormalization / factorization scale , @xmath83 , uncertainty , @xmath84 .
the theoretical prediction in eq .
[ eq : xie ] must then be replaced by @xmath85 from which we can derive the theoretical covariance matrix @xmath86 here we assume that there is no bin to bin correlation in the mc uncertainty . on the other hand ,
we take the correlation of the scale uncertainty fully into account .
both @xmath87 and @xmath88 are evaluated at the central values of the pdf parameters , assuming that the variation is small .
we evaluate the scale uncertainty in a very straightforward manner . as the central prediction the renormalization and factorization scale
are taken to be equal to half the transverse energy of the leading jet in the event , @xmath89 . to estimate the uncertainty we make another theoretical prediction now choosing as a scale @xmath90 .
the `` one - sigma '' uncertainty is defined as @xmath91 as we will see later in this section the theoretical uncertainties are small compared to the other uncertainties
. therefore this crude estimate suffices for the purposes of this paper . in the future
a more detailed study of the theoretical uncertainty is required .
the scale uncertainty is often associated with the theoretical uncertainty due to the truncation of the perturbative series .
however , it is important to realize this is only a part of the full theoretical uncertainty . in fig .
[ fig3]@xmath92 we present results for the single inclusive jet cross section as a function of the jet transverse energy .
both data and theoretical predictions are divided by the average prediction of the initial pdf s .
the nlo predictions are calculated using the jetrad prediction @xcite .
the inner ( outer ) error bar on the experimental points represent the diagonal part of the experimental ( total ) covariance matrix .
the dotted lines represent the initial one - sigma pdf uncertainties .
the solid lines are the theory predictions calculated with the new pdf s ( i.e. , the new probability density distribution ) .
the plot is somewhat misleading because of the large point - to - point correlation of the uncertainties .
the confidence level of 50% is very high , indicating a good agreement between the prediction and the data .
this leads us to the conclusion that the one jet inclusive transverse energy distribution is statistically in agreement with the nlo theoretical expectation based on the initial probability density distribution of the pdf parameters .
no indication of new physics is present .
note that the prediction using the initial pdf differs quite a bit from the more traditional fits such as mrsd0 , see the dashed line in fig .
[ fig3]@xmath92 . having no uncertainties on the traditional
fits it is hard to draw any quantitative conclusion from this observation .
the larger value of the jet cross section calculated using the initial pdf set at high transverse energies compared to mrsd0 was anticipated in ref .
@xcite and can probably be traced back to the larger @xmath93 and @xmath94 quark distribution at the reference scale @xmath2 and moderate @xmath95 .
this difference in turn was partially attributed to the different way of treating target mass and fermi motion corrections . given the confidence level of 50%
the one jet inclusive data can be included in the fit . using eq .
[ chisqnew ] we calculate for each pdf set @xmath45 the corresponding @xmath96 .
this gives us the 100 weights @xmath66 ( conditional probabilities ) defined in eq .
using eq .
[ eq : newmusig ] , we can calculate the effects of including the cdf data into the fit .
the results are shown in figs .
[ fig3]@xmath92 and [ fig3]@xmath97 . as can be seen in fig .
[ fig3]@xmath92 the effect is that the central value is pulled closer to the data and the pdf uncertainty is reduced substantially .
two of the fourteen pdf parameters are affected the most .
as expected these are the strong coupling constant @xmath98 and the gluon pdf coefficient @xmath99 , which controls the high @xmath28 behavior ( the gluon pdf is proportional to @xmath100 at the initial scale ) . in fig .
[ fig3]@xmath97 we show the correlation between these two parameters before and after the inclusion of the cdf data .
as can be seen the impact on @xmath99 is very significant .
similarly , the uncertainty on @xmath101 is reduced substantially and the correlation between the two parameters is also changed .
this indicates that the one jet inclusive transverse energy distribution in itself has a major impact on the uncertainty of @xmath101 and the determination of the gluon pdf .
note that we do not address the issue of the parametrization uncertainty .
other choices of how to parameterize the initial pdf s will change the results . to obtain a value and uncertainty of @xmath98 which is on the same footing as the one obtained from @xmath102-colliders
, one needs to address this issue .
[ ryw ] our second example is the lepton charge asymmetry in @xmath0-boson decay at the tevatron .
as already explained , this observable is important for the reduction of the pdf uncertainties in the @xmath0-boson mass extraction at hadron colliders .
the asymmetry is given by @xmath103 where @xmath104 and @xmath105 are respectively the number of positrons and electrons at the pseudo - rapidity @xmath106 . in fig .
[ fig : asym]@xmath92 , we show the preliminary cdf data of run 1@xmath97 ( solid points ) for the asymmetry , along with the nlo predictions ( dotted lines ) including the pdf uncertainties , relative to the theory average prediction using the initial pdf s . for the nlo calculations
the dyrad prediction @xcite was used .
the inner error bars on the experimental points are the statistical uncertainties ; the systematic uncertainties are small and we can safely neglect them .
the outer error bars are the diagonal of the total covariance matrix . in this case
, the theoretical uncertainty is dominated by the phase space monte carlo integration uncertainty ; we took its bin to bin correlation into account .
similar to the one jet inclusive transverse energy case , the scale uncertainty is defined by the difference between the theoretical prediction calculated using two scales , @xmath107 and @xmath108 . as is clear from fig .
[ fig : asym]@xmath92 , there is a good agreement between the data and the nlo prediction , except for the last experimental point at the highest pseudo - rapidity . the confidence level including the last point is well below our threshold of 0.27% . in order to be able to include the data into the pdf fit we decided to simply exclude this data point from our analysis . without the highest pseudo - rapidity point
we obtain a reasonable confidence level of 4% .
it is not as good as in the single inclusive jet case even though the plots appear to indicate otherwise .
the reason for this is the absence of significant point - to - point correlation for the charge asymmetry uncertainties .
we can now include the lepton charge asymmetry data into the fit by updating the probability density distribution with bayes theorem , as described in the previous section .
in fig . [
fig : asym]@xmath92 the prediction obtained with the new probability density distribution are shown by the solid lines . as expected , the data are pulling the theory down and reducing the pdf uncertainties .
it is difficult to correlate the change in the asymmetry to a change in a particular pdf parameter . on the other hand
, it is well known that the lepton asymmetry can be approximately related to the following asymmetry of the ratio of up quark ( @xmath94 ) and down quark ( @xmath93 ) distribution function @xmath109 the bjorken-@xmath28 are given by @xmath110 where @xmath111 is the mass of the @xmath0-boson , @xmath112 the center of mass of the collider , and @xmath113 the @xmath0-boson rapidity .
the pdf s were evaluated with the factorization scale equal to @xmath111 .
the ratio @xmath114 is approximately the @xmath0-boson asymmetry and obviously is sensitive to the slope of the @xmath94/@xmath93 ratio . in fig .
[ ryw]@xmath97 we show the ratio @xmath114 calculated with both the initial and the new probability density distributions .
as can be seen , the change is very similar to the change experienced by the lepton charge asymmetry itself .
the change in @xmath114 can be traced to a simultaneous increase in the anti - up quark distribution and decrease in the anti - down quark distribution at low @xmath28 .
current standard sets of pdf do not include uncertainties .
it is clear that we can not continue to discount them .
already current measurements at the tevatron have highlighted the importance of these uncertainties for the search of physics beyond the standard model .
furthermore , the potential of future hadron colliders to measure @xmath115 and the @xmath0-boson mass is impressive , but can not be disentangled from pdf uncertainties .
the physics at the lhc will also undoubtedly require a good understanding of the pdf uncertainties . on a more general level , if we want to quantitatively test the framework of perturbative qcd over a very large range of parton collision energies the issue of pdf uncertainties can not be sidestepped . in this paper
we have illustrated a method , based on statistical inference , that can be used to easily propagate uncertainties to new observables , assess the compatibility of new data , and if the latter is good to include the effect of the new data on the pdf without having to redo the whole fit .
the method is versatile and modular : an experiment can be included in or excluded from the new pdf fit without any additional work .
the statistical and systematic uncertainties with the full point - to - point correlation matrix can be included as well as the theoretical uncertainties .
none of the uncertainties are required to be gaussian distributed .
one remaining problem is the uncertainty associated with the choice of parametrization of the input pdf .
this is a difficult problem that does not have a clear answer yet and will require a compromise between the number of parameters and the smoothness of the pdf .
we plan to address this question in another paper
. the next phase would then be to obtain a large number of initial pdf s sets based on theoretical consideration only , in the spirit of the inference method and bayes theorem .
the dis and tevatron data could then be used to constraint the range of these pdf s resulting in a set of pdf s which would include both experimental and theoretical uncertainties .
for our initial pdf parameter probability density distribution we use the results of ref .
there a chi - squared fit was performed to dis data from bcdms @xcite , nmc @xcite , h1 @xcite and zeus @xcite .
both statistical uncertainties and experimental systematic uncertainties with point - to - point correlations were included in the fit , assuming gaussian distributions . however , _ no _ theoretical uncertainties were considered .
it is important to include the correlation of the systematic uncertainties because the latter usually dominate in dis data . simply adding them in quadrature to the statistical uncertainty
would result in a overestimation of the uncertainty .
a standard parametrization at @xmath116 is used with 14 ( = @xmath3 ) parameters : @xmath117 , @xmath118 , @xmath119 , @xmath120 , and @xmath121 are parametrized using the functional form @xmath122 whereas @xmath123 is parametrized as @xmath124 . here
@xmath28 is the bjorken-@xmath28 .
parton number and momentum conservation constraints are imposed .
the full covariance matrix of the parameters , @xmath125 , is extracted at the same time as the value of the parameters that minimize the chi - squared .
the uncertainties on the parameters were assumed to be gaussian , such that the fitted values also correspond to the average values of the parameters , @xmath126 .
the probability density distribution is then given by @xmath127 where @xmath128 is the difference between the total chi - squared of the experimental data used in the fit and the minimum chi - squared ( 1256 for 1061 data points ) with the pdf s fixed by the set of parameters @xmath9 .
the matrix @xmath129 is the inverse of the covariance matrix @xmath125 .
the @xmath130 is the determinant of the covariance matrix .
all the calculations were done in the @xmath131-scheme . comparison with
mrs and cteq sets showed a good overall agreement with a few exceptions .
one example is the difference in the gluon distribution function at large values of @xmath28 .
the cteq and mrs distribution are somewhat above the result of ref .
this difference was attributed to the fact that prompt photon data were included in the cteq and mrs fits .
note that the direct photon data have large scale uncertainty , and it might be misleading to include them in a fit without taking the theoretical uncertainty into account . also , it is important to keep in mind that it is misleading to compare specific pdf s , as the correlations between different pdf parameters are large .
99 h. l. lai et al . , phys .
d55 ( 1997 ) 1280 ; + a. d. martin et al . , dtp-96 - 102 , dec 1996 , hep - ph/9612449 . the cdf collaboration , f. abe et al . ,
lett . 77 ( 1996 ) 438 .
j. huston et al . , phys .
77 ( 1996 ) 444 . the cdf ii collaboration , f. abe et al .
, the cdf ii detector technical design report , fermilab - pub-96/390-e .
r. m. barnett et al .
d54 , 1 ( 1996 ) .
numerical recipes , w. h. press et al . , cambridge university press .
the cdf collaboration , phys .
lett . 74 ( 1995 ) 850 ; + r. g. wagner , for the cdf collaboration , published proceedings 5th international conference on physics beyond the standard model , balholm , norway , april 29-may 4 , 1997 .
s. alekhin , ifve-96 - 79 ( sep 1996 ) , hep - ph/9611213 .
the program was obtained from the cteq collaboration www - site `` http://www.phys.psu.edu/@xmath132cteq/ '' .
w. t. giele , e. w. n. glover and d. a. kosower , nucl .
* b403 * ( 1993 ) 633 .
the bcdms collaboration , a. c. benvenuti et al .
223b ( 1989 ) 485 ; phys .
lett 237b ( 1990 ) 592 .
the nm collaboration , m. arneodo et al .
b483 ( 1997 ) 3 .
the h1 collaboration , s. aid et al .
470b ( 1996 ) 3 . the zeus collaboration , m. derrick et al . ,
c72 ( 1996 ) 399 . | .1 in standard parton distribution function sets do not have rigorously quantified uncertainties . in recent years
it has become apparent that these uncertainties play an important role in the interpretation of hadron collider data . in this paper , using the framework of statistical inference , we illustrate a technique that can be used to efficiently propagate the uncertainties to new observables , assess the compatibility of new data with an initial fit , and , in case the compatibility is good , include the new data in the fit
. # 1 # 2 # 3 _ phys .
lett . _ * # 1 * ( # 3 ) # 2 # 1 # 2 # 3 _ nucl .
phys . _ * # 1 * ( # 3 ) # 2 # 1 # 2 # 3 _ z. phys . _ * # 1 * ( # 3 ) # 2 # 1 # 2 # 3 _ phys .
rev .
lett . _ * # 1 * ( # 3 ) # 2 # 1 # 2 # 3 _ phys .
rev . _
* # 1 * ( # 3 ) # 2 # 1 # 2 # 3 _ mod .
phys .
lett . _ * # 1 * ( # 3 ) # 2 # 1 # 2 # 3 _ rev .
mod .
phys . _ * # 1 * ( # 3 ) # 2 = cmbx10 = cmr10 = cmti10 = cmbx10 scaled1 = cmr10 scaled1 = cmti10 scaled1 = cmbx9 = cmr9 = cmti9 = cmbx8 = cmr8 = cmti8 = cmr7 -0.5 cm 0.1 cm implications of hadron collider observables .4 cm on parton distribution function uncertainties 1.4 cm walter t. giele and stephane keller + 0.2 cm _ fermilab , ms 106 + batavia , il 60510 , usa _ |
gamma rays emitted from radioactive isotopes , which are explosively synthesized at a supernova ( sn ) explosion ( truran , arnett , & cameron 1967 ; bodansky , clayton , & fowler 1968 ; woosley , arnett , & clayton 1973 ) , play an important and unique role in emission from a supernova , not only at @xmath0-ray energies but also at the lower energies : from x - rays to even optical or near - infrared ( nir ) band . up to a few years after the explosion , the decay chain @xmath5ni @xmath6 @xmath5co @xmath6 @xmath5fe ( clayton , colgate , & fishman 1969 ) dominates the high energy radiation input to a supernova , with minor contributions from other radioactivities such as @xmath7ni ( @xmath6 @xmath7co @xmath6 @xmath7fe : clayton 1974 ) .
the decay produces @xmath0-ray lines with average energy per decay @xmath8 mev ( @xmath5ni decay with the e - folding time 8.8 days ) or @xmath9 mev ( @xmath5co decay with the e - folding time 113.7 days ) .
these line @xmath0-rays are degraded in their ways through the sn ejecta , predominantly by compton scatterings ( see e.g. , cass & lehoucq 1990 for a review ) .
non - thermal electrons produced at the scatterings and other processes ( pair production and photoelectric absorption ) heat the ejecta , yielding thermal emissions at optical and nir wavelengths .
the @xmath0-ray emissions from supernovae provide a unique tool to study the amount and distribution of predominant radioactive isotopes @xmath5co ( therefore @xmath5ni produced at the explosion ) .
density structure in the supernova ejecta could also be inferred by modeling line profiles that are affected by compton scatterings .
for example , unexpectedly early detections of the high energy emission from sn 1987a ( dotani et al .
1987 ; sunyaev et al . 1987 ; matz et al .
1988 ) revealed the important role of rayleigh - taylor instability and mixing in the sn ejecta ( e.g. , chevalier 1976 ; hachisu et al .
the example highlights the very importance of modeling and analyzing @xmath0-ray emission from supernovae .
except the very nearby sn 1987a , unfortunately there are to date only a few other examples of possible detection of @xmath0-rays from the @xmath5ni decay chain from supernovae : one marginal detection ( sn ia 1991 t : lichti et a. 1994 ; morris et al .
1997 ) and two upper limits ( sne ia 1986 g and 1998bu : matz & share 1990 ; leising et al .
1999 ) . however , now that the _ international gamma - ray astrophysics laboratory ( integral ) _ has been launched and some new gamma ray observatories are being planed ( e.g. , takahashi , kamae , & makishima 2001 ) , it should be important to make prediction of @xmath0-ray emissions ( spectra and light curves ) , taking into account recent development of supernova researches , i.e. , multi - dimensionality ( see e.g. , maeda , mazzali , & nomoto 2006b and references therein ) .
most of studies on @xmath0-ray transport in sn ejecta have been restricted to one - dimensional , spherical models . lately , a few 3d @xmath0-ray transport computations for sne have become available : hflich ( 2002 ) developed a 3d @xmath0-ray transport computational code and applied it to a 3d hydrodynamic model of a type ia supernova ( khoklov 2000 ) .
hungerford , fryer , & warren ( 2003 ) also developed a 3d @xmath0-ray transfer code .
they computed @xmath0-ray spectra for 3d asymmetric ( bipolar ) type ii supernova models , and later for 3d single lobe explosion models ( hungerford , fryer , & rockefeller 2005 ) .
given higher density and smaller @xmath5ni mass for core - collapse supernovae than sne ia , there is no doubt that core - collapse supernovae are more difficult to detect in the high energy range ( e.g. , timmes & woosley 1997 ) .
for example , hflich , wheeler , & khoklov ( 1998 ) predicted the detection limit of sne ia by _ integral _ @xmath10 mpc ( @xmath11 per year ) , while hungerford et al .
( 2003 ) estimated that for sne ii ( more or less similar to sn 1987a ) @xmath12 kpc ( @xmath13 per 100 years ) .
accordingly , the number of theoretical prediction of hard x - ray and @xmath0-ray emission from core - collapse supernovae is still small to date , except models for the very nearby sn 1987a ( e.g. , mccray , shull , & sutherland 1987 ; woosley et al .
1987 ; shibazaki & ebisuzaki 1988 ; kumagai et al .
1989 ) . despite this , in view of proposed large asymmetry in core - collapse supernovae ( see e.g. , maeda & nomoto ( 2003b ) and references therein ) and its possible direct relation to high energy emissions , theoretical prediction of high energy emission for a variety of core - collapse supernova models is important in order to uncover the still - unclarified nature of the explosion and understand trends that may be possible to observe with current and future instruments
. in this respect , potentially interesting targets among core collapse supernovae are very energetic supernovae , often called `` hypernovae '' ( iwamoto et al . 1998 ) .
a prototypical hypernova is sn 1998bw discovered in association with a gamma ray burst grb980425 ( galama et al . 1998 ) .
its broad absorption features in optical spectra around maximum brightness and very bright peak magnitude led iwamoto et al .
( 1998 ) , assuming spherical symmetry , to conclude that the kinetic energy @xmath14 ergs @xmath15 , the ejecta mass @xmath16 , the main sequence mass @xmath17 , and the mass of newly synthesized radioactivity @xmath5ni @xmath18(@xmath5ni ) @xmath19 ( see also woosley , eastman , & schmidt 1999 ) . following this and motivated by the deviation between the spherical hypernova model prediction and observations after @xmath20 days , maeda et al .
( 2006ab ) have presented comprehensive study comparing various observations of sn 1998bw with theoretical expectations from jet - like aspherical explosion models of maeda et al .
( 2002 ) , using multi - dimensional radiation transport calculations .
they found that an aspherical model with @xmath21 provides a good reproduction of optical emission from the explosion to @xmath22 year consistently .
this paper follows the analysis of optical emission from aspherical hypernovae applied to sn 1998bw by maeda et al .
( 2006b ) .
in this paper we present theoretical predictions of high energy emission from the same set of aspherical hypernova models as presented in maeda et al .
( 2002 , 2006ab ) . because the models are intrinsically aspherical , we make use of fully 3d hard x- and @xmath0-ray transport computations . in 2 , details of computational method are presented with a brief summary of the input models . in 3 , overall synthetic spectra are presented .
4 focuses on line profiles . in 5 , light curves of some lines are presented .
in addition to modeling high energy emission from sn 1998bw - like hypernovae , also interesting is applicability of a gray effective absorptive opacity for @xmath0-ray transport , which is often used as an approximation in computations of optical spectra and light curves of supernovae to save computational time ( e.g. , sutherland & wheeler 1984 ) .
although it has been examined in 1d spherically symmetric cases and concluded that the approximation is good if an appropriate value is used for the effective @xmath0-ray opacity , it has not been examined yet if this is also the case for aspherical models . in 6 , applicability of the gray absorptive @xmath0-ray opacity for ( jet - like ) aspherical models is examined by comparing results with detailed @xmath0-ray transport and with gray transport . finally , in 7 conclusions and discussion , including an estimate of detectability of the high energy emission , are presented .
we have developed a fully 3d , energy - dependent , and time - dependent gamma ray transport computational code .
it has been developed following the individual packet method using a monte carlo scheme as suggested by lucy ( 2005 ) .
the code follows gamma ray transport in sn ejecta discretised in 3d cartesian grids ( @xmath23 : linearly discretised ) and in time steps ( @xmath24 : logarithmically discretised ) . for the ejecta dynamics , we assume homologous expansion , which should be a good approximation for sne ia / ib / ic .
the density at time interval ( @xmath25 ) is assumed homogeneous in each spatial zone with the value at time @xmath26 .
the transport is solved in the sn rest frame .
the expansion of the ejecta is taken into account as follows .
first , cross sections for interactions with sn materials given in the comoving frame are transformed into the rest frame ( castor 1972 ) . the fate of a photon
is then determined in the rest frame .
if a packet survives as high energy photons after the interaction , the new direction and the energy are given at the comoving frame , depending on the specific interaction taking place ( see below ) .
the direction and the energy are then converted to the sn rest frame ( castor 1972 ) .
gamma ray lines from the decay chains @xmath5ni @xmath6 @xmath5co @xmath6 @xmath5fe and @xmath7ni @xmath6 @xmath7co @xmath6 @xmath7fe are included .
the numbers of lines included in the computation are 6 ( @xmath5ni decay ) , 24 ( @xmath5co ) , 4 ( @xmath7ni ) , and 3 ( @xmath7co ) ( lederer & shirley 1978 ; ambwani & sutherland 1988 ) . in the present study we focus on the @xmath5ni @xmath6 @xmath5co @xmath6 @xmath5fe chain , which includes 812 kev ( @xmath5ni ) , 847 kev ( @xmath5co ) , and 1238 kev ( @xmath5co ) lines . for the interactions of gamma rays with sn materials , we consider pair production , compton scattering , and photoelectric absorption . for pair production cross sections , a simplified fitting formula to hubbell ( 1969 )
is adopted from ambwani & sutherland ( 1988 ) . for compton scattering , the klein - nishina cross section is used . for photoelectric absorption ,
cross sections compiled by hflich , khoklov , & mller ( 1992 ) from veigele ( 1973 ) are used .
if an interaction takes place , the fate of the gamma ray packet is chosen randomly in proportion to the cross section of each possible interaction .
if the pair production is the fate , an electron and a positron are created . for the electron path , the electron deposits the entire energy to the ejecta . for the positron path ,
the positron annihilates with an ambient electron , producing two @xmath0-rays ( assuming no positronium formation ) .
the positron kinetic energy is deposited to the thermal pool .
these processes are assumed to take place _ is situ_. according to this prescription , a packet either becomes the 511 kev @xmath0-ray lines or is absorbed .
the fate is selected randomly in proportion to the branching probability depending on the initial photon energy before the pair production .
if the compton scattering is the fate , the polar angle of the scattering relative to the incoming photon direction in the comoving frame is randomly sampled according to the klein nishina distribution using a standard monte carlo rejection technique ( the kahn s method ) . the azimuthal angle is randomly selected in the comoving frame .
the packet is now either a gamma - ray packet or a non - thermal electron packet ( assumed to deposit the entire energy to the sn ejecta immediately ) , determined randomly according to the energy flowing into each branch .
if the photoelectric absorption is the fate , the entire energy of the packet is absorbed .
possible x - ray fluorescence photons are not followed in the present simulations , but these photons are below the low energy cut - off , making negligible effects on the energy range examined in this paper . for a time advancing scheme , the code has two options : the one is the exact time advancing , taking into account the time interval at each photon flight . the other is no time advancing , assuming the flight time through the ejecta is negligible . in the present work , we use the exact one for computations of @xmath0-ray light curves ( and for computations of optical emission in maeda et al .
2006b ) , while use the no - time advancing for computations of spectra .
this is in order to optimize the number of photons escaping at a given time interval in the spectral computations , since the computation of spectra with fine energy bins already needs a large number of packets and large computational time . for @xmath0-rays ,
the negligible time delay approximation is a good one .
we have confirmed this for the present models , by performing the fully time - dependent computation but with the number of packets entering each time interval smaller than used in the standard no - time advancing spectrum computations . in view of the recent investigation by milne et al .
( 2004 ) that not all the published 1d gamma ray transport codes give mutually consistent results , we test the capability of our new code by computing gamma ray spectra based on the ( spherical ) sn ia model w7 ( nomoto , thielemann , & yokoi 1984 ) , for which many previous studies are available for comparison . we have compared our synthetic gamma ray spectra at 25 and 50 days with figure 5 of milne et al .
we found an excellent agreement between our results and the spectra resulting from majority of previous codes , e.g. , of hungerford et al .
( 2003 ) . in this study ,
the ejecta are mapped onto @xmath27 cartesian grids . for spectrum synthesis ,
@xmath28 photon packets with equal initial energy content are used . in escaping the eject ,
these packets are binned into 10 angular zones with equal solid angle from @xmath29 to @xmath30 ( here @xmath31 is the polar angle from the @xmath1-axis ) and into 3000 energy bins up to 3 mev with equal energy interval 1 kev . for @xmath0-ray light curve computations ,
@xmath32 photon packets are used . in escaping the ejecta ,
the packets are binned into 36 time step logarithmically spaced from day 5 to day 300 , as well as into the angular and energy bins .
input models for the @xmath0-ray transport computations are taken from the aspherical model a and the spherical model f from maeda et al .
( 2006b ) .
model a is a result of a jet - like explosion with the initial energy input at the collapsing core injected more in the @xmath1-axis than in the @xmath33-axis ( see maeda et al .
2002 for details ) . in maeda
et al . ( 2006ab ) , we examined optical light curves ( @xmath34 days ) as well as expected optical spectral characteristics in both early ( @xmath35 days ) and late ( @xmath36 days ) phases . the structure of model a at homologous expansion phases is shown in figure 1 . in the present work
we examine the following models : ( @xmath37 , @xmath38 , @xmath18(@xmath5ni)/@xmath39 ) @xmath40 and @xmath41 for model a , and @xmath42 and @xmath43 for model f ( hereafter @xmath44 is the ejecta mass , @xmath38 is the kinetic energy of the expansion in @xmath45 ergs , and @xmath18(@xmath5ni ) is the mass of @xmath5ni synthesized at the explosion ) .
maeda et al .
( 2006b ) concluded that optical properties of sn 1998bw are explained consistently by model a with the energy @xmath21 and with @xmath18(@xmath5ni ) @xmath46 , so that we regard model a representing a prototypical hypernova .
as seen in figure 1 , model a is characterized by concentration of @xmath5ni distribution along the @xmath1-axis , which is a consequence of the explosive nucleosynthesis in jet - like aspherical explosions ( nagataki 2000 ; maeda et al .
2002 ; maeda & nomoto 2003b ) .
because @xmath5ni is the source of @xmath0-rays , the distribution will affect the @xmath0-ray transport and resulting hard x- and @xmath0-ray emissions sensitively .
figure 2 shows the distribution of @xmath5ni along the line of sight for our models .
the amount of @xmath5ni is integrated in the plane perpendicular to the line of sight within the constant line - of - sight velocity interval .
such that , figure 2 shows a profile of @xmath0-ray lines from the decay of @xmath5ni or @xmath5co in optically thin limit .
figure 2 shows that model a yields the @xmath5ni at high velocities if it is viewed from the polar ( @xmath1 ) direction . on the other hand ,
if it is viewed from the equatorial ( @xmath33 ) direction , it yields the distribution sharply peaked at zero and low velocities .
note that model a with @xmath47 shows @xmath5ni at velocities higher than in model f with @xmath48 .
figures 3 and 4 show synthetic @xmath0- and hard x - ray spectra at 25 days after the explosion . figures 5 and 6 show the ones at 50 days after the explosion .
thanks to the large @xmath49 as compared to ( normal ) sne ii and to the absence of a massive hydrogen layer , the emergence of the high energy emission is much earlier for the present models ( in the order of a month ) than sne ii ( in the order of a year ) . an aspherical model yields the emergence of the high energy emission earlier than a spherical model with the same energy : at 25 days , model a with @xmath50 already shows high energy appearance although model f with @xmath50 does not show up . indeed , the former , despite the energy @xmath50 , gives the date of emergence comparable to model f with @xmath48 .
the time scale of the @xmath0-ray emission will be further discussed in 5 .
model a is characterized by the large line - to - continuum ratio , especially at early on ( figure 3 ) . because the continuum is formed by degradation of line photons by compton scatterings , the large line - to - continuum ratio is realized for the plasma with low optical depth for compton scatterings . indeed
, @xmath0-ray transport computations typically predict larger the ratio at more advanced epochs ( see e.g. , sumiyenov et al .
it is also seen by comparing model f with different energies ( @xmath51 ) at day 50 ( figure 5 ) : the larger energy leads to smaller optical depth , and therefore to the larger line - to - continuum ratio . at 25 days ,
model a with either @xmath47 or @xmath52 yields the ratio larger than model f with @xmath48 , which is attributed to the existence of high velocity @xmath5ni at low optical depth in model a ( figures 1 & 2 ) . at 50 days , the effect becomes less significant ( i.e. , the ratio becomes comparable for model a ( @xmath50 or @xmath53 ) and for model f with @xmath48 ) , but still visible as compared with model f with @xmath50 ( figure 5 ) .
another feature is seen in hard x - ray spectra .
model a has the hard x - ray cut - off at energy higher than model f ( figures 4 and 6 ) .
the cut - off is formed by photoelectric absorption , which is dependent on metal content ( e.g. , grebenev & sunyaev 1987 ) . at 25 days ,
only the emission near the surface ( along the @xmath1-axis for model a ) can escape out of the ejecta .
the cut - off is therefore determined by metal content near the surface . in model
f , the surface layer is dominated by intermediate mass elements ( i.e. , a co core of the progenitor star ) . on the other hand , in model
a the emitting region contains a large fraction of fe - peak elements , most noticeably @xmath5ni ( or co , fe ) .
therefore , the photoelectric cut - off should be at the energy higher in model a than model f. at more advanced epochs , an observer looks into deeper regions .
then the contribution from the deeper region becomes bigger and bigger , yielding increase of the cut - off energy in model f since @xmath5ni is centrally concentrated ( figure 2 ) .
this effect is also seen in model a , but to the smaller extent .
although the mass of @xmath5ni within a given velocity interval increases toward the center ( except the inner most region where the @xmath5ni fraction is very small ) also in model a along the @xmath1-axis , the increase is less dramatic than model f ( e.g. , compare the masses of @xmath5ni at 20,000 and 10,000 km s@xmath4 for model a in figure 2 ) .
therefore , the metal content of the emitting region does not temporally evolve very much in model a. in the above discussion on the photoelectric absorption , one would expect to use the @xmath5ni distribution along the `` @xmath33-axis '' , not the @xmath1-axis , for the observer at the @xmath33-axis .
however , it is not the case . as discussed by hungerford et al .
( 2002 , 2005 ) , even for the observer at the @xmath33-axis , @xmath5ni contributing the most of the emission is that at the @xmath1-axis , i.e. , an observer at the @xmath33-axis looks at the emitting polar @xmath5ni blobs sideways ( see also maeda et al .
to clarify this , in figure 7 we show last scattering points of hard x- and @xmath0-rays .
also shown are contours of the optical depth for observers at @xmath54- and @xmath55-directions .
the observer at the @xmath1-axis looks at the emitting blob moving toward the observer , while the observer on the @xmath33-plane looks at a pair of the emitting blobs sideways .
this is further discussed in 4 and 5 , but here we point out one difference related to effects of the viewing angle in our model from hungerford et al .
their models do not yield large difference in the absolute flux , while our models do show boost of high energy luminosity toward the @xmath1-axis .
a similar behavior is also seen in optical emission in our models ( maeda et al .
the mechanism of the boost should be the same as the one for optical photons .
initially at high density only the sector - shaped region along the @xmath1-axis can yield escaping photons ( fig .
the cross sectional area of this photosphere is larger toward the @xmath1-axis than the @xmath33-axis , making the luminosity boost toward the @xmath1-axis .
as time goes by , the ejecta optical depth decreases , and therefore the photosphere moves to cover the equatorial region .
the difference becomes less and less significant , as seen by comparing figures 3 and 5 .
the different behavior is probably due to the different degree of the penetration of @xmath5ni into the outer layers .
hungerford et al . ( 2002 , 2005 ) presented sn ii models with a massive hydrogen envelope , which does not exist in our model sne ib / c .
the existence of the hydrogen envelope generally yields less aspherical distribution of @xmath5ni and the emitting region in sne ii than sne ib / c ( e.g. , wang et al .
2001 ) . in any case , as hungerford et al .
( 2005 ) suggested , the high energy emission can very sensitively depending on the type and degree of asymmetry , therefore transport calculations are very important .
we now turn our attention to line profiles .
figure 8 shows line profiles of the @xmath5ni 812 kev and @xmath5co 847 kev lines at 25 days . figure 9 shows the @xmath5co 1238 kev line .
figures 10 and 11 show the same lines at 50 days . at these early epochs ,
the ejecta are not thin to high energy photons ( figure 7 ) , and therefore the line profiles are different from the ones expected in optically thin limit ( figure 2 ) . at day 25 ,
an observer only sees the region near the surface ( figure 7 ) . for model a , the region is further concentrated along the @xmath1-axis ( the top of the `` @xmath5ni bubble '' in figure 1 ) .
figures 8 and 9 show that the line profiles are very asymmetric , showing only the emission at the blue coming from the region moving toward an observer , _ except _ an observer at the @xmath33-axis . for an observer at the @xmath33-axis , even the emission at the rest wavelength
is seen , which is especially evident for the more energetic , therefore less optically thick , model ( model a with @xmath47 ) .
the emitting region is now at the outer edge(s ) of the @xmath5ni distribution , and the density is relatively small there ( figure 1 ) .
the line - of - sight , connecting the emitting region(s ) and an observer at the @xmath33-direction passes only through the low density region , including the region moving away from the observer . on the other hand , the line - of - sight for an observer at the @xmath1-axis passes through the dense , central region .
for example , this behavior is seen in figure 7 by comparing the regions , having the optical depth @xmath56 less than 10 at day 25 , for the observers at @xmath1- and @xmath33-axes .
for the observer at @xmath54 direction , the region contains only the emitting blob moving toward the observer .
for the observer at the @xmath55 direction , on the other hand , a pair of the emitting blobs along the @xmath1-axis , seen from the side by the observer , are included in the regions with @xmath57 . at 50 days , the line becomes broad redward as a photon even from the far side escape out of the ejecta more easily because of decreasing density .
still , only the blue part is seen for an observer at the @xmath1-direction ( figures 10 & 11 ) .
it is in contrast to model f with @xmath48 , which now show the emission at the rest wave length .
it is interesting to see that the line profiles of model a ( either @xmath50 or @xmath53 ) viewed at the @xmath1-direction resemble to the @xmath5ni distribution in the hemisphere moving toward the observer ( figure 2 ) .
this suggests that the photons from the far hemisphere are almost totally blocked by the high density central region , while the @xmath5ni - rich region itself is nearly optically thin .
it is indeed the case as seen in figure 7 . for model a viewed at the @xmath33-direction and for model f ,
the line profile is explained by continuously decreasing escape probability .
these arguments imply that , assuming that we have a temporal series of @xmath0-ray observations , the line profiles for model a viewed at the @xmath1-direction should show the time interval within which the line profiles are almost fixed , while model a viewed at the @xmath33-direction and model f should show the continuous changes in the line profile until the entire ejecta become optically thin .
this could be an interesting observational target , since the temporally `` fixed '' line profile , if observed , suggests that the viewing angle is close to the pole . in this case , the line shape directly traces the distribution of @xmath5ni .
in addition , line profiles , sensitively dependent on the viewing angle as well as the degree of asphericity , could be used as a tracer of the distribution of @xmath5ni and density once @xmath0-ray observations at @xmath58 mev with sufficient sensitivity become possible .
the detectability will be discussed in 7 .
figures 12 shows light curves of the @xmath5ni 812 , @xmath5co 847 , and @xmath5co 1238 kev lines at the distance 10 mpc . for the 1238 kev line , the flux is computed by integrating a spectrum in 1150 1340 kev range . for the 812 and 847 kev lines , the sum of the fluxes
is computed first by integrating a spectrum in the 800 920 kev range , then the individual contribution is computed with the decay probability at the corresponding epochs assuming that the escape fractions for these two lines are equal .
the assumption of the equal escape fractions should be a good approximation , since these two lines are close in the energy , yielding almost identical compton cross sections ( see also milne et al .
2004 ) . model a shows the emergence of the @xmath0-rays earlier , even though the energy is smaller , than model f. this is due to the large amount of @xmath5ni at low density and high velocity regions .
later on , the @xmath0-rays become stronger for model f with @xmath48 than for model a ( @xmath59 ) .
three effects are responsible : ( 1 ) the contribution of the dense central region stopping the photons becomes large .
( 2 ) the overall optical depth @xmath60 is larger for model a than model f. ( 3 ) model f with @xmath48 has a bit large @xmath18(@xmath5ni ) as compared to model a. because of this temporal behavior , the enhancement of @xmath0-ray flux is more noticeable for the @xmath5ni line ( with the e - folding time 8.8 days ) than the @xmath5co lines ( with the e - folding time 113.7days ) .
as discussed in 3 , the effect of the viewing angle is seen at early epochs .
model a emits more toward the @xmath1-direction than the @xmath33-direction .
this behavior is qualitatively similar to that in optical emission ( maeda et al .
the effect virtually vanishes around @xmath20 days for model a with @xmath47 , which is consistent with a rough estimate that the ejecta become optically thin to compton scatterings at @xmath61 days ( see e.g. , maeda et al .
2003a : see also figure 7 ) , where @xmath62 is the optical depth to compton scatterings at day 1 . in the above estimate ,
the cross section is assumed to be @xmath63 of the thomson scattering cross section , and composition @xmath64 .
in previous sections , we presented model predictions for high energy spectra and light curves .
however , @xmath0-ray transport in the sn ejecta has another role which is as important as the high energy emission itself : the @xmath0-rays give heating of the ejecta , i.e. , non - thermal electrons scattered off from ions by compton scatterings rapidly pass their energies to thermal particles through ionization , excitation , and scatterings with thermal electrons .
the thermal energy is then converted to optical photons . in this way
, the energy lost from the high energy photons determines the optical luminosity from a supernova . in this section ,
we examine how much energy is stored in the ejecta . in maeda
( 2006b ) , we made use of the detailed 3d high energy photon transport to compute optical light curves and nebular spectra of models a and f ( and other models ) .
on the other hand , a gray effective absorptive assumption for @xmath0-ray transport has been frequently used for computations of @xmath0-ray deposition and optical light curves .
sometimes it has been used , although in spherically symmetric models , to compute optical light curves for ( hypothetical ) non - spherically symmetric supernovae ( e.g. , tominaga et al . 2005 ; folatelli et al .
it is therefore important to examine applicability of the assumption for aspherical supernovae .
figure 13 shows synthetic optical light curves for models a and f. results with the detailed 3d transport and with the simplified gray absorptive @xmath0-ray opacity ( with various effective opacity @xmath65 ) are compared .
figure 13 shows that the gray absorptive approximation is actually good for computations of @xmath0-ray deposition , thus for computations of optical light curves , if an appropriate value is used for the effective opacity . with the value @xmath66
@xmath3 g@xmath4 , we obtain optical light curves for both the spherical model f and the aspherical model a almost identical to those obtained with the detailed @xmath0-ray transport computations . at late epochs @xmath67 days ( depending on models )
the ejecta become optically thin to @xmath0-rays . at these epochs ,
@xmath0-rays suffer at most only one compton scattering before escaping the ejecta . in this idealized situation ,
the `` effective '' @xmath0-ray opacity can be computed by taking an appropriate average of the klein - nishina cross section for various scattering angles and @xmath0-ray line energies . in this way ,
sutherland & wheeler ( 1984 ) yielded @xmath68 @xmath3 g@xmath4 assuming the typical line energy @xmath69 mev .
this value is dependent on the line list .
we have also performed the estimate of the effective opacity in the optically thin limit under the condition that the averaged absorbed energy per photon flight is equal for the detailed case and for the effective absorption case , and found that for the @xmath5co lines , @xmath70 ( z / a ) @xmath3 g@xmath4 .
this yields @xmath71 @xmath3 g@xmath4 for the sne ib / c composition .
indeed , figure 13 implies that this value is probably even better than @xmath72 @xmath3 g@xmath4 in the late phase , although the difference is small .
we emphasize that in this situation , the effective opacity is independent from geometry of the ejecta .
it is confirmed , although not generally , by the fact that the @xmath66 @xmath3 g@xmath4 reproduces the light curve obtained by the detailed computations for both models a and f. somewhat surprising is that ( 1 ) the same value @xmath66 @xmath3 g@xmath4 reproduces the detailed transport computations rather well at early epochs , and ( 2 ) this applies not only to the spherical model , but also to the aspherical model . because multiple scatterings take place , now the effective @xmath0-ray opacity can be geometry- and time - dependent , and can be different from the value at the late phases .
for example , sutherland & wheeler suggested @xmath73 @xmath3 g@xmath4 taking into account multiple scatterings .
frannson ( 1990 ) gave the value @xmath73
@xmath3 g@xmath4 with @xmath64 from a series of monte carlo simulations .
colgate , petschek , & kriese ( 1980 ) gave the value @xmath74
@xmath3 g@xmath4 .
these authors used different ejecta models .
the small differences among these studies ( although there are some exceptions , e.g. , 0.07 @xmath3 g@xmath4 by woosley , taam , & weaver ( 1986 ) ) suggest that the effects of density and @xmath5ni distribution on the total deposition rate is small ( in spherically symmetric case ) .
the present study suggests that this is also the case even in asymmetric cases .
in this paper , we first reported the development of a new code for 3d computations of hard x - ray and @xmath0-ray emission resulting from radioactive decays in supernovae .
the code is useful for many problems . among them
is applicability of a gray absorptive approximation in supernova ejecta without spherically symmetry .
we have shown , although only for some specific models , this is actually a good approximation .
this approximation has been extensively used even in analyzing supernovae with hypothetical asymmetry ( e.g. , tominaga et al .
2005 ; folatelli et al .
2006 ) , and we have confirmed the applicability using the 3d computations . with the code , we have presented predictions of hard x - ray and @xmath0-ray emission in aspherical hypernova models .
the models were verified by fitting optical properties of the prototypical hypernova sn 1998bw ( maeda et al .
2006ab ) , therefore we regard the model prediction being based on `` realistic '' hypernovae .
irrespective of the observer s direction , the aspherical models yield ( 1 ) early emergence of the high energy emission , and therefore the large peak flux , ( 2 ) large line - to - continuum ratio , and ( 3 ) large cut - off energy in the hard x - ray band , as compared to spherical models . these are qualitatively similar to what are expected from extensive mixing of @xmath5ni ( see e.g. , cass & lehoucq 1990 ) by e.g. , rayleigh - taylor instability ( e.g. , hachisu et al .
1990 ) . however , degree of the @xmath5ni penetration is more extreme , and therefore the effects of @xmath5ni mixing is more noticeable , in our models than spherical mixing cases .
for example , for model a , the cut - off energy ( by photoelectric absorptions ) is already at @xmath20 kev at the emergence of the @xmath0-rays , and it does not temporally evolve very much , because the mass fraction of @xmath5ni ( co ) does not increase very much toward the center along the @xmath1-axis
. this constant cut - off energy would be useful to distinguish the aspherical models from spherical ones , once we have a sequence of hard x - ray observations .
another interesting result is that line profiles are sensitively dependent on asphericity , as well as observer s direction .
lines should show larger amount of blueshift for an observer closer to the polar ( @xmath1 ) direction .
for an observer at equatorial ( @xmath33 ) direction , lines show emission at the rest wavelength even at very early epochs .
these are not expected in spherically symmetric models .
although analysis of line profiles will need future observatories with very high sensitivity ( e.g. , takahashi et al . 2001 ) and/or a supernova very nearby , once it becomes possible then it will be extremely useful to trace asphericity and energy , therefore the explosion mechanism , of supernovae .
for example , a possibly interesting observational strategy is implied . in model
a the @xmath5ni region becomes optically thin at a rather early phase when the inner regions are still optically thick .
it will lead to the following temporal evolution for an observer at the @xmath1-direction .
first , a line becomes redder and redder with time .
then at some epoch it stops changing the shape very much , presenting closely the @xmath5ni distribution in the hemisphere moving toward the observer . finally , the line from the other hemisphere appears .
this evolution is unique as compared with spherical models or for an observer at @xmath33-direction , either of which should show continuous reddening until entire ejecta become optically thin . because we show that there are features of the aspherical models , some of which depend on the degree of asphericity but not on the observer s direction , while the others depend on both the asphericity and the direction , combination of various analyses will be helpful to distinguish models and observer s directions .
a problem is , of course , how many hypernovae are expected to be able to be observed in current and future observatories . for the aspherical model a with @xmath47 , the expected maximum line fluxes are about a half of those of typical sn ia models .
the less energetic model ( @xmath50 ) yields even smaller fluxes .
this is because of smaller amount of @xmath5ni and larger ratio @xmath75 ( and therefore later emergence of the radioactive decay lines ) than the sn ia models .
even worse , the occurring rate is much smaller for hypernovae than sne ia . in this respect
, there is no doubt that sne ia are most promising targets in radioactive @xmath0-rays .
however , hypernovae could still be an interesting targets in @xmath0-rays among core - collapse supernovae .
taking sn 1987a as a typical sn ii , its peak 847 kev line flux was @xmath76 photons @xmath77 s@xmath4 , yielding @xmath78 @xmath77 s@xmath4 if it were at 10 mpc .
our aspherical hypernova models predict the peak flux more than two orders of magnitude larger than this .
therefore , for a given sensitivity of instruments , it is visible at least an order of magnitude farther than a typical sn ii .
low mass sne ib / c should also be discussed .
although the nature of the sne ib / c seems very heterogeneous , let us assume that @xmath79 , @xmath80 , and @xmath18(@xmath5ni ) @xmath81 .
these values give the @xmath0-ray escaping time scale comparable to the present models ( see 5 ) , resulting in the estimate of the peak line flux @xmath82 times smaller than the present models .
for the sensitivity of the _ integral _ , we estimate that the maximum distances within which @xmath0-rays are detectable are @xmath83 kpc , @xmath84 mpc , and @xmath85 mpc , for sne ii , ib / c , and hypernovae , respectively ( here we assume the line width 10 , 20 , and 40 kev for sne ii , ib / c , and hypernovae , respectively ) .
the latter two , especially the detection limit for hypernovae , cover some star burst galaxies like m82 ( @xmath86mpc ) .
since sn 1987a , the nearest sne for each type are sn ii 2004dj ( @xmath11 mpc ) , sn ic 1994i ( @xmath87 mpc ) , and a hypernova ( but weaker than sn 1998bw ) sn ic 2002ap ( @xmath10 mpc ) .
the prototypical hypernova sn 1998bw was at @xmath88 mpc .
therefore , the sensitivity of the _ integral _ is unfortunately not enough to detect these core - collapse supernovae , and we will have to wait for next generation hard x- and @xmath0-ray telescopes . if these telescopes will be designed to archive sensitivity better than the _ integral _ by two orders of magnitudes ( see e.g. , takahashi et al .
2001 ) , then the maximum distances are 5 mpc , 40 mpc , and 70 mpc for sne ii , ibc , and hypernovae , respectively . the detection limits for sne
ib / c and hypernovae then cover some clusters of galaxies like virgo ( @xmath89mpc ) , fornax ( @xmath89mpc ) , and even hydra ( @xmath90mpc ) .
this sensitivity will lead to comprehensive study of @xmath0-ray emission , and therefore of explosive nucleosynthesis , in core collapse supernovae .
observed occurring rate is @xmath91 and @xmath92 per year in an average galaxy for sne ii and sn ib / c , respectively .
the hypernova rate is rather uncertain .
podsiadlowski et al . (
2004 ) gave a conservative estimate of the rate @xmath93 per year .
these values give , by multiplying the rate and the volume detectable , the observed probability @xmath58 ( sne ii ) : @xmath94 ( sne ib / c , but not hypernovae ) : @xmath84 ( hypernovae ) for a given detector .
the difference between usual sne ib / c and hypernovae may ( probably ) be even smaller , since the estimate of sne ib / c flux above is probably the upper limit ( i.e. , @xmath79 seems to be almost the lower limit ) and that the hypernova rate is possibly the underestimate . in sum ,
as long as the detectability is concerned , a hypernova is worse than ( usual ) sne ib / c , but potentially better than sne ii .
another question is if any asphericity similar to that derived for the prototypical hypernova sn 1998bw ( maeda et al .
2006ab ) exists in ( usual ) sne ib / c .
if it does , as suggested by e.g. , wang et al . ( 2001 ) , then most of the properties shown in this paper should also apply for hard x- and @xmath0-ray emission from these sne ib / c ( since the time scale is similar to hypernovae : see 5 ) .
one should of course take into account the fact that the expansion velocity and @xmath18(@xmath5ni ) are smaller than hypernovae , and ultimately needs direct computations based on a variety of explosion models . | hard x- and @xmath0-ray spectra and light curves resulting from radioactive decays are computed for aspherical ( jet - like ) and energetic supernova models ( representing a prototypical hypernova sn 1998bw ) , using a 3d energy- and time - dependent monte carlo scheme .
the emission is characterized by ( 1 ) early emergence of high energy emission , ( 2 ) large line - to - continuum ratio , and ( 3 ) large cut - off energy by photoelectric absorptions in hard x - ray energies .
these three properties are not sensitively dependent on the observer s direction .
on the other hand , fluxes and line profiles depend sensitively on the observer s direction , showing larger luminosity and larger degree of blueshift for an observer closer to the polar ( @xmath1 ) direction .
strategies to derive the degree of asphericity and the observer s direction from ( future ) observations are suggested on the basis of these features , and an estimate on detectability of the high energy emission by the _
integral _ and future observatories is presented .
also presented is examination on applicability of a gray effective @xmath0-ray opacity for computing the energy deposition rate in the aspherical sn ejecta .
the 3d detailed computations show that the effective @xmath0-ray opacity @xmath2 @xmath3 g@xmath4 reproduces the detailed energy - dependent transport for both spherical and aspherical ( jet - like ) geometry . |
let @xmath0 be a homotopy on torus and @xmath7 . r.geoghegan and a. nicas in @xcite developed an one - parameter theory and defined the one - parameter trace @xmath8 of @xmath2 .
the element @xmath8 is a 1-chain in @xmath9 , where the structure of the bimodule @xmath10 is given in section 2 .
this 1-chain gives information about the fixed points of @xmath2 , that is , using @xmath8 is possible to define the one - parameter nielsen number @xmath4 of @xmath2 and the one - parameter lefschetz class @xmath1 of @xmath2 .
n(f ) is the number of non - zero c - components in @xmath8 and @xmath1 is the image of @xmath8 in @xmath11 by homomorfism @xmath12 , induced by inclusion @xmath13 , where @xmath14 is the semicentralizer of an element @xmath15 which represents the semiconjugacy class @xmath16 .
the precise definition is given in @xcite .
the main purpose this paper is show that for which homotopy on torus then @xmath17 , where @xmath5 is on of the two generators in @xmath11 . in @xcite r.b.s.brooks et al .
showed that if @xmath18 is any map on a k - dimensional torus @xmath19 then @xmath20 , where @xmath21 is the nielsen number and @xmath22 the lefschetz number of @xmath23 . in some sense
our result is a version of this result for one - parameter case when @xmath24 .
this paper is organized into five sections , besides this one . in section 2
contain a review of one - parameter fixed point theory . in section 3
we presented some results of semiconjugacy classes on torus . in section 4
have the proof of the main result which is the theorem [ maintheorem ] .
let @xmath19 be a finite connected cw complex and @xmath25 a cellular homotopy .
we consider @xmath26 $ ] with the usual cw structure and orientation of cells , and @xmath27 with the product cw structure , where its cells are given the product orientation .
pick a basepoint @xmath28 , and a basepath @xmath29 in @xmath19 from @xmath30 to @xmath31 .
we identify @xmath32 with @xmath33 via the isomorphism induced by projection @xmath34 .
we write @xmath35 for the homomorphism ; @xmath36 we choose a lift @xmath37 in the universal cover , @xmath38 , of @xmath19 for each cell @xmath39 and we orient @xmath37 compatibly with @xmath39 . let @xmath40 be the lift of the basepath @xmath29 which starts at in the basepoint @xmath41 and @xmath42 the unique lift of @xmath2 satisfying @xmath43
. we can regard @xmath44 as a right @xmath45 chain complex as follows : if @xmath46 is a loop at @xmath30 which lifts to a path @xmath47 starting at @xmath48 then @xmath49^{-1 } = h_{[w]}(\tilde{e})$ ] , where @xmath50 } $ ] is the covering transformation sending @xmath48 to @xmath51 .
the homotopy @xmath52 induces a chain homotopy @xmath53 given by @xmath54 , for each cell @xmath55 .
this chain homotopy satisfies ; @xmath56 and the boundary operator @xmath57 satisfies ; @xmath58 .
define endomorphism of , @xmath59 , by @xmath60 , @xmath61 , @xmath62 and @xmath63 .
we consider trace@xmath64 .
this is a hochschild 1-chain whose boundary is : trace@xmath65 we denote by @xmath66 the subset of @xmath67 consisting of semiconjugacy classes associated to fixed points of @xmath68 or @xmath69 .
the one - parameter trace of homotopy @xmath2 is : @xmath70 @xmath71 the @xmath72component of @xmath8 is denoted by @xmath73 @xmath74 we call it the fixed point index of @xmath2 corresponding to semiconjugacy class @xmath75
. a fixed point index @xmath76 of @xmath2 is zero if the all cycle in @xmath76 is homologous to zero . given a cellular homotopy @xmath25 the one - parameter nielsen number , @xmath4 , of @xmath2 is the number of nonzero fixed point indices .
the one - parameter lefschetz class , @xmath1 , of @xmath2 is defined by ; @xmath77 where @xmath78 is induced by the inclusion @xmath79 . from @xcite we have the following theorems .
[ invariance ] let @xmath80 be cellular ; if @xmath2 is homotopic to @xmath81 relative to @xmath82 then @xmath83 .
if @xmath84 then every map homotopic to @xmath2 relative to @xmath85 has a fixed point not in the same fixed point class as any fixed point in @xmath85 .
in particular , if @xmath68 and @xmath69 are fixed point free , every map homotopic to @xmath2 relative to @xmath85 has a fixed point .
[ nielsen - number ] every map homotopic to @xmath2 relative to @xmath85 has at least @xmath4 fixed point classes other than the fixed point classes which meet @xmath85 .
in particular , if @xmath68 and @xmath69 are fixed point free maps , then every map homotopic to @xmath2 relative to @xmath85 has at least @xmath4 path components . for
a complete description of the one - parameter fixed point theory see @xcite .
in this subsection we describe some results about the semiconjugacy classes in the torus related to a homotopy @xmath0 .
we will consider the homomorphism @xmath86 given above .
we take @xmath87 \in t$ ] and @xmath88 , where @xmath89 and @xmath90 .
thus , given the homomorphism @xmath35 we have @xmath91 and @xmath92 .
therefore , @xmath93 , for all @xmath94 .
we denote this homomorphism by the matrix : @xmath95 = \left ( \begin{array}{cc } b_{1 } & b_{3 } \\
b_{2 } & b_{4 } \\
\end{array } \right ) \ ] ] [ prop - conjugacy - classes ] two elements @xmath96 and @xmath97 in @xmath81 belong to the same semiconjugacy class , if and only if there are integers @xmath98 satisfying the following equations : @xmath99 if there is @xmath100 satisfying @xmath101 then we obtain the equation of the proposition .
the other direction is analogous .
we take the isomorphism @xmath102 such that @xmath103 . by proposition
[ prop - conjugacy - classes ] two elements @xmath96 and @xmath97 in @xmath81 belong to the same semiconjugacy class , if and only if there is @xmath104 satisfying : @xmath105-i)z = \theta(g_{2}g_{1}^{-1})$ ] , where @xmath106 is the identity matrix .
if determinant of the matrix @xmath105-i ) $ ] is zero then will have an infinite amount of elements in a semiconjugacy class .
[ corol-2 ] for each @xmath107 the semicentralizer @xmath108 is isomorphic to the kernel of @xmath109-i$ ] .
[ lemma-1 ] the 1-chain , @xmath110 , is a cycle if and only if the element @xmath111 belongs to the kernel of @xmath109-i$ ] . if @xmath112 is a cycle , then @xmath113 @xmath114 @xmath115 @xmath116 @xmath117 .
this implies @xmath118 and @xmath119 .
the other direction is analogous .
[ corollary-1 ] if the matrix of the homomorphism @xmath120 is given by @xmath95 = \left ( \begin{array}{cc } 1 & b_{3 } \\ 0 & b_{4 }
\\ \end{array } \right ) \ ] ] with @xmath121 or @xmath122 , then the 1-chain , @xmath112 , is a cycle if and only if @xmath123 .
by definition given a 2-chain @xmath124 then @xmath125 [ prop-1 ] the 1-chain , @xmath126 , is homologous to the 1-chain , @xmath127 , for all @xmath128 .
note that for @xmath129 and @xmath130 the proposition is true .
we suppose that for some @xmath131 , the 1-chain @xmath132 is homologous to the 1-chain @xmath133 , we will write @xmath132 @xmath134 . considering to the 2-chain @xmath135 we have @xmath136 therefore @xmath137 using induction , we obtain the result .
the case in which @xmath138 is analogous .
the proof of following results can be found in @xcite .
[ prop-2 ] in the case @xmath139 and @xmath140 each 1-cycle @xmath141 is homologous to a 1-cycle the following form : @xmath142 .
[ prop-3 ] each 1-cycle @xmath143 is not trivial , that is , is not homologous to zero . [ corollary-2 ]
let @xmath144 be a cycle .
if the cycles @xmath145 and @xmath146 are in different semiconjugacy classes for @xmath147 , @xmath148 , then @xmath149is a nontrivial cycle .
each cycle @xmath150 projects to the class @xmath151 $ ] that is one of the two generators of @xmath152
let @xmath0 be a homotopy on torus @xmath153 . we can choose a homotopy @xmath68 homotopic to @xmath2 with @xmath68 transverse the projection @xmath159 .
therefore , @xmath160 is transverse , that is , @xmath161 is finite . since @xmath162 then for each @xmath163 we can deform @xmath68 to a homotopy @xmath69 such that @xmath164 for each @xmath165 $ ] and @xmath69 has no fixed points in @xmath166 .
in fact , take @xmath167 defined by @xmath168 thus , taking @xmath177 we have a homotopy between @xmath68 and a map @xmath156 where @xmath156 satisfying the hypothesis of the theorem . note that we can choose @xmath178 such that @xmath179 $ ] because @xmath160 is contained in @xmath180 .
thus , @xmath181 is transverse .
let @xmath182 be the subset of @xmath183 consisting of those circles of fixed points which are not in the same fixed point class as any fixed point of @xmath68 or @xmath69 . from @xcite @xmath183
consists of oriented arcs and circles . from proposition [ prop - h ]
if @xmath0 is a homotopy and @xmath184 the projection then we can choose @xmath2 such that @xmath183 is transverse the projection @xmath159 .
thus , @xmath182 is a closed oriented 1-manifold in the interior of @xmath185 .
let @xmath186 be space of all paths @xmath187 in @xmath185 from the graph @xmath188 of @xmath2 to the graph @xmath189 of @xmath159 with the compact - open topology , that is , maps @xmath190 \to t \times
i \times t$ ] such that @xmath191 and @xmath192 .
let @xmath193 be isolated circles in @xmath194 , oriented by the natural orientations , and @xmath195 then @xmath196 determines a family of circles @xmath197 in @xmath186 via constant paths , i.e. each oriented isolated circle of fixed points @xmath198 of @xmath2 determines an oriented circle @xmath199 defined by @xmath200 where @xmath201 is the constant path at @xmath202 , that is , @xmath203 for each @xmath204 $ ] .
therefore , we can write @xmath205 \in h_{1}(e_{f})$ ] . since @xmath206 is transverse
then @xmath207 for all @xmath208 , see @xcite . from @xcite
we have ; from @xcite , section @xmath211 , given @xmath0 a homotopy then we can to deform @xmath2 to a homotopy @xmath81 such that in each fixed point class of @xmath81 has an unique circle , and this circle is transverse the projection .
let us suppose that the homomorphism @xmath120 is induced by a homotopy @xmath2 satisfies @xmath212-i)=0 $ ] .
using the notation above we can suppose which @xmath120 is given by @xmath95 = \left ( \begin{array}{cc } 1 & b_{3 } \\ 0 & b_{4 }
\\ \end{array } \right ) , \ ] ] and @xmath109 \neq i \equiv $ ] ( identity ) , that is , @xmath213 and @xmath140 , with @xmath121 or @xmath214 .
this is done choosing a base @xmath215 for @xmath216 , where @xmath30 is a eigenvector of @xmath109 $ ] associated to @xmath130 .
note that if @xmath109 = i$ ] then @xmath217 because any @xmath2 can be deformed to a fixed point free map .
for example , take the homotopy @xmath0 defined by ; @xmath218 with @xmath219 any irrational number between @xmath220 and @xmath130 .
we will have @xmath221 = [ \phi ] = i$ ] , but @xmath2 is a fixed point free map .
thus @xmath217 , which implies @xmath222 .
therefore , henceforth we suppose @xmath109 \neq i.$ ] since @xmath153 is a polyhedron then @xmath153 is a regular cw - complex .
thus , for any cellular decomposition of the torus the entries of matrices of the operators @xmath223 and @xmath224 will be composed by elements @xmath225 because the incindence number of a 2-cell in a 1-cell is @xmath226 and the the incindence of one 1-cell in one 0-cell is @xmath226 , see chapter ii of @xcite .
therefore chosen an orientation to each cell in a decomposition cellular to the torus then the one - parameter trace @xmath8 will be the form of the following matrix : @xmath227 \otimes [ \tilde{d}_{0 } ] & 0 \\ 0 & [ \tilde{\partial}_{2 } ] \otimes [ \tilde{d}_{1 } ] \\
\end{array } \right ) \ ] ] where @xmath228_{ij } , [ \tilde{\partial_{2}}]_{kl } \in \{0 , \pm 1 , \pm u,\pm v , \}.$ ] thus , we can write @xmath229 or only @xmath230 or @xmath231 , where @xmath232 , @xmath233 and @xmath234
. we will suppose which @xmath8 is write like above .
the case with @xmath230 or @xmath231 the proof is analogous . from lemma 4.1 of @xcite the element @xmath235 is homologous to zero . by proposition [ prop - h ]
we can suppose that @xmath2 has no fixed points in @xmath236 .
in this situation @xmath8 will be a 1-cycle in @xmath237 .
thus , by proposition [ prop-2 ] , the sum @xmath238 can not be appear in one - parameter trace @xmath8 of @xmath2 .
therefore , in this case the trace @xmath8 has the form : @xmath239 from section 2 the one - parameter lefschetz class is the image of @xmath8 in @xmath11 by induced of inclusion @xmath241 .
thus , each element @xmath242 is sending in @xmath11 in the class @xmath151 $ ] , that is , the image of @xmath8 in @xmath11 will be @xmath243 = n[u ] = n(f)[u]\ ] ] take @xmath244 $ ] , which is one of the two generators of @xmath11 .
if we consider left action instead right action in the covering space we will obtain @xmath245 $ ] .
therefore , @xmath246 in this case we have @xmath212-i ) = l(f_{|t } ) \neq 0 $ ] .
therefore , by corollary [ corol-2 ] , for each element @xmath107 the semicentralizer , @xmath108 , of @xmath247 in @xmath81 is trivial .
thus , @xmath248 for each semiconjugacy class @xmath16 , that is , @xmath249 which implies @xmath250 in this case we have @xmath251 we have other interpretation in case ii .
note that by definition of @xmath8 in section 2 we are not considering in trace @xmath8 the semiconjugacy classes represented by fixed point classes which meet @xmath236 .
if we consider all fixed points classes then the trace @xmath8 has the form : @xmath252 because in this situation can not be appear circles in @xmath183 , but only arcs join @xmath253 to @xmath254 . by proposition [ prop-2
] @xmath8 can not be a 1-cycle .
since for each @xmath255 the map @xmath256 can be deformed to a map with @xmath257 fixed points , then from theorem 3.3 of @xcite we will have @xmath258-i),$ ] i.e. in this case @xmath183 will be compose by @xmath212-i)=l(f_{|t})$ ] arcs join @xmath259 to @xmath260 | the main result this paper states that if @xmath0 is a homotopy on torus then the one - parameter lefschetz class @xmath1 of @xmath2 is given by @xmath3 , where @xmath4 is the one - parameter nielsen number of @xmath2 and @xmath5 is one of the two generators in @xmath6 . |
electron - positron ( ep ) pair plasmas are ubiquitous and play important roles in many astrophysical situations such as the early universe @xcite , van allen radiation belts , and near the polar cap of fast rotating neutron stars @xcite , black holes @xcite , pulsars @xcite , quasars @xcite , active galactic nuclei @xcite , accretion disks @xcite as well as in laboratories @xcite . in black holes , pulsars and quasars such ep plasmas are emitted in the form of ultra - relativistic winds or collimated jets by some of their most energetic objects . because of the intrinsic and complete symmetry between the positively charged ( e.g. , positrons or positive ions ) and negatively charged ( e.g. , electrons or negative ions ) particles , the dynamics of pair plasmas become significantly different from that of electron - ion plasmas or from a purely electronic beam . over the last few years , a number of experiments have been performed to create ep plasmas ( see , e.g. , refs . ) . in such experiments ,
it has been observed that because of the fast annihilation and the formation of positronium atoms , the identification of collective modes in ep plasmas is practically impossible . to resolve this issue and to identify the collective modes properly ,
a number of experiments have been proposed to create pair - ion plasmas ( see , e.g. , ref . ) . however ,
most recently , ion free high - density ( @xmath8 @xmath9 ) neutral ep plasmas with unique characteristics have been produced in the laboratory by using a compact laser driven setup @xcite .
it has been reported that because of their unique characteristics together with the charge neutrality , small divergence as well as high average lorentz factor , such ep plasmas can exhibit collective behaviors and thereby opening up the possibility of studying the collective dynamics of ep plasmas in a controlled laboratory environment .
the collective oscillations of ep plasmas and associated wave dynamics together with the formation of solitary waves and shocks have been extensively studied over the past two decades ( see , e.g. , refs . ) . in other works ,
the modulational instability ( mi ) and the nonlinear evolution of electrostatic and electromagnetic wave envelopes have also been studied in ep plasmas ( see , e.g. , refs . ) .
however , most of these works are based on hydrodynamic models . using the kinetic theory approach , i.e. , using vlasov - poisson equations ,
the effects of landau damping on various kinds of wave modes ( linear theory ) as well as on electrostatic and electromagnetic solitary waves [ governed by the korteweg - de vries ( kdv ) equation ] have also been studied in ep plasmas @xcite and in some other environments ( see , e.g. refs . and references therein ) . however , to our knowledge , no theory for the explanation of electrostatic wave modulation and the effects of nonlinear landau damping on mi and nonlinear evolution have been reported in nonextensive ep plasmas .
many spacecraft observations ( e.g. , in the earth s bow - shock , upper ionosphere of mars , the vicinity of the moon etc . ) @xcite and laboratory experiments @xcite confirm the presence of nonthermal and superthermal particles which do not follow the maxwellian distribution , but show some deviation from the thermodynamic equilibrium .
the presence of such nonthermal particles has also been confirmed in astrophysical environments @xcite .
several models for phase - space plasma distributions with nonthermal or superthermal wings or other deviations from maxwellian distribution have been proposed in recent years .
one of such distributions , which was first reported by renyi @xcite and subsequently proposed by tsallis @xcite , is the boltzmann - gibbs - shannon ( bgs ) entropy in which the degree of nonextensivity of the plasma particles is characterized by the entropic index @xmath0 .
the distribution function with @xmath10 characterizes the system with more superthermal particles ( superextensivity ) , while the distribution with @xmath11 indicates plasmas containing a large number low - speed particles ( subextensivity ) compared to the maxwellian one @xmath12 . such @xmath0-nonextensive distribution has been widely considered in a number of works to investigate various collective plasma modes and nonlinear coherent structures ( see , e.g. , refs . ) .
it is to be noted that the @xmath13 distribution function and the @xmath0-nonextensive distribution function in tsallis statistics are somewhat equivalent in the sense that in both these cases , the spectrum of the distribution functions show similar behaviors .
in fact , there is a formal transformation @xmath14 which can provide the missing links between these two velocity distribution functions @xcite on the other hand , it is well known that waves in plasmas can undergo collisionless damping when they resonantly interact with trapped and/or free particles , i.e. , when the particle s velocity approaches the wave phase velocity or group velocity .
such collisionless damping was first theoretically predicted by landau @xcite , and later confirmed experimentally by malmberg _
_ @xcite . motivated by these inventions , ott
_ et al . _
@xcite first theoretically investigated the effects of linear landau damping on the nonlinear propagation of ion - acoustic solitons in electron - ion plasmas through the description of kdv equations and on the assumption that particle s trapping time is much longer than that of landau damping .
later , ikezi _ et al . _
@xcite , based on their experimental investigation , have emphasized the importance of nonlinear landau damping in plasmas .
accordingly , ichikawa @xcite investigated the the effects of nonlinear landau damping due to resonant particles having the group velocity of the wave , on the modulation of electrostatic wave envelopes in electron - ion plasmas . in this work
, he assumed that the typical time scale is much longer than the bouncing period of particles trapped in the potential trough .
it has been shown that the nonlinear wave - particle resonance leads to the modification of the nonlinear schrdinger ( nls ) equation with a nonlocal nonlinear term proportional to a cauchy principal value integral as well as the local ( cubic ) nonlinear term . furthermore , in ref .
it was reported that , in contrast to the ordinary nls equation , the nonlinear resonance always leads to mi of wave envelopes against a plane wave perturbation regardless of the sign of @xmath15 positive or negative , where @xmath16 is the coefficient of the group velocity dispersion and @xmath17 is the cubic nonlinear term of the nls equation . in this work ,
our aim is to revisit and extend the work of ichikawa @xcite in an unmagnetized collisionless ep pair plasma in the context of tsallis @xmath0-nonextensive statistics . starting from a set of vlasov - poisson equations and using the reductive perturbation technique ( rpt ) , we show that the nls equation is not only modified by the nonlinear resonant effects but also by the nonextensive parameter @xmath0 which contributes to both the dispersive and nonlinear ( local and nonlocal ) terms .
it is found that the nonextensivity significantly modifies the wave frequency , the group velocity , the nonlinear frequency shift and the energy transfer rate for the modulated wave packets as well as the amplitude of solitary wave solutions of the nls equation .
this section mainly focuses on the derivation of the nls equation . though , the relevant analysis is almost the same as in ref .
, we , however , review the analysis and present the subsequent derivations and explanations for the expressions of various plasma modes in more details .
we consider the nonlinear propagation of electrostatic wave packets along the @xmath18-axis in a collisionless electron - positron pair plasma .
the basic equations for the dynamics of ep plasmas are given by the following vlasov and poisson equations @xmath19 @xmath20 where @xmath21 is the distribution function with its unperturbed part @xmath22 , @xmath23 is the particle velocity and @xmath3 is the electrostatic potential .
also , the particle charge and mass are given by @xmath24 , @xmath25 for electrons and @xmath26 , @xmath27 for positrons .
we assume that the equilibrium state ( at @xmath28 ) is spatially uniform field - free ep plasma and that the perturbation from the equilibrium state is purely electrostatic .
furthermore , we consider the equilibrium distribution of electrons and positrons @xmath22 to be the @xmath0-distribution function as in the tsallis nonextensive statistics @xcite . it is to be mentioned that due to the resonance of plasma particles having the group velocity of the wave , the distribution functions become singular , and so the direct application of rpt to the vlasov - poisson equations and may not determine uniquely the contributions of the resonant particles . in order to treat these singularities
properly we introduce the multiple space - time scales as @xcite @xmath29 we assume that the amplitude of the carrier wave ( with wave number @xmath30 and the wave frequency @xmath31 ) is infinitesimally small , and so for @xmath32 a slight deviation @xmath33 $ ] from the uniform equilibrium value will occur .
thus , we expand @xmath34 } \\ & \phi(x , t)=\sum_{n=1}^{\infty}\epsilon^{n } \sum_{l=-\infty}^{\infty}\phi^{(n)}_l(\eta,\sigma,\zeta ) \exp{[il(kx-\omega t ) ] } , \label{expansion } \end{split}\ ] ] where @xmath35 is a small @xmath36 positive parameter measuring the weakness of the wave amplitude and @xmath37 , @xmath38 for reality conditions to hold . here , the asterisk denotes the complex conjugate quantity .
we note that the expansion for @xmath39 is valid only in the non - resonance region where @xmath40 is satisfied . in the resonance region
where @xmath41 , the above expansion may not be appropriate to apply to the vlasov equations .
so , the components @xmath42 and @xmath43 are further expanded into multi - scale fourier - laplace integrals as @xmath44 \\ & \phi^{(n)}_l(\eta,\sigma,\zeta)= \frac { 1}{(2\pi)^2 } \int_c d\omega \int_{-\infty}^{\infty}dk\tilde{\phi}^{(n)}_l ( k,\omega,\zeta ) \exp[i(k\eta-\omega\sigma ) ] , \label{fourier - lap - int } \end{split}\ ] ] where the contour @xmath45 is parallel to the real axis and lies above the coordinate of convergence .
we first derive the evolution equation of electrostatic wave envelopes for arbitrary species of particles with unperturbed arbitrary velocity distribution function , and then we extend our analysis in neutral ep plasmas in the context of tsallis @xmath0-nonextensive statistics @xcite .
so , we substitute the stretched coordinates and the expansions into eqs . and to obtain , respectively , as @xmath46 @xmath47 where we have used the symbol @xmath48 to denote the equality in the weak sense and disregarded the terms which contain @xmath49 and @xmath50 in eq . .
in the subsequent analysis , we determine the contributions of the resonant particles having the wave group velocity by solving the @xmath51-evolution of the components @xmath42 and @xmath43 as an initial value problem with the initial condition @xmath52 in the multiple space - time scheme corresponding to that on the distribution function @xmath53 equating the coefficients of @xmath35 from eqs . and for @xmath55 , we obtain ( for details see appendix [ appendix1 ] ) the following linear dispersion law : @xmath56 together with the linear landau damping rate given by @xmath57 where @xmath58 denotes the cauchy principal value , and we have made the analytical continuation of the integral over @xmath23 along the real axis passing infinitesimally above and under the pole at @xmath59 with the constraint of weakly damped waves and @xmath60 from eqs . and , we equate the components for @xmath62 and @xmath63 , and then use the dispersion relation to obtain the following conditions ( see appendix [ appendix2 ] ) @xmath64 here , we examine the second order terms with @xmath66 and @xmath67 .
thus , we have from eq . @xmath68 while the first order terms with @xmath63 and @xmath67 of eq . yields
@xmath69 substituting the fourier - laplace integrals given by eq . into eq . and solving it as an initial value problem we obtain @xmath70 where we have used an arbitrary constant as @xmath71 . next , a substitution of eq .
gives @xmath72 using the initial condition and taking fourier inversion of eqs .
and we obtain for @xmath73 the following zeroth order components : @xmath74 the vanishing of @xmath75 and hence of @xmath76 in fact eliminates the contributions of the modes associated with the singularities of the integrals in eq . .
thus , in this way the initial condition uniquely defines the present problem to investigate nonlinear auto modulation of @xmath77 modes . we proceed to examine the the second order , first harmonic modes with @xmath66 and @xmath79 .
thus , from eqs . and
we obtain the following compatibility condition ( see for details appendix [ appendix3 ] ) @xmath80 where @xmath81 is given by @xmath82^{-1 } \left[1 + 4\pi\sum e_\alpha \int_c\frac{v}{(\omega - kv)^2}g_\alpha dv\right ] , \label{lambda}\ ] ] we find that this expression of @xmath81 is exactly the same as the group velocity @xmath83 obtained by differentiating the dispersion relation with respect to @xmath30 .
equation shows that the @xmath84 variation of @xmath85 can be related to a new coordinate defined by @xmath86 such that @xmath87 this indicates that the coordinate @xmath4 in eq .
establishes a clear relationship between the reductive perturbation theory and the multiple space - time expansion method . for the second order quantities with @xmath88
, we obtain from eqs . and the following expressions ( see appendix [ appendix4 ] ) @xmath89 , \label{f - alpha2-n2-l2}\ ] ] @xmath90 where @xmath91 note that in eq .
the contributions of the resonance terms at the phase velocity @xmath92 are neglected because our basic assumption [ eq . ] for ep plasma oscillations is that the linear landau damping is a higher - order effect than second order . from eqs .
and we consider the terms corresponding to @xmath94 , @xmath67 , and use the relations and to obtain a set of reduced equations which , after use of the fourier - laplace transforms with respect to @xmath95 and @xmath51 and the initial condition , yield @xmath96 h(k,\omega ) , \label{ff - alpha - n2-l0-final}\ ] ] @xmath97 where the relevant details and the expressions for @xmath98 and @xmath99 are given in appendix [ appendix5 ] .
considering the terms for @xmath94 and @xmath79 , we obtain from eqs . and a set of reduced equations , which after few steps ( see appendix [ appendix6 ] ) , result into the following modified nonlinear schrdinger equation for the small but finite amplitude perturbation @xmath101 as @xmath102 the coefficients of the dispersion ( group velocity ) , cubic nonlinear ( local , nonlocal nonlinear terms , respectively , are @xmath103 and @xmath104 , given by @xmath105 and @xmath106 , where @xmath107 @xmath108 @xmath109 @xmath110 with @xmath111 dv .
\label{b}\ ] ] furthermore , the coefficient @xmath112 of eq . , representing the linear landau damping rate associated with the resonant particles having the phase velocity of the carrier wave ,
is given by @xmath113 in eq .
, the coefficient @xmath16 appears due to the group velocity dispersion of the wave envelope . the most significant contribution of the resonant particles having the wave group velocity is the appearance of the nonlocal nonlinear term @xmath114 .
this resonance contribution also modifies the local nonlinear coefficient @xmath17 , which appears due to carrier wave self - interactions originating from the zeroth harmonic modes ( or slow modes ) .
before we proceed to the modulation of electrostatic wave envelopes in @xmath0-nonextensive ep plasmas , we verify some important conservation laws that are associated with the ordinary nls equation ( here , ordinary means in absence of any nonlocal nonlinearity in the nls equation ) .
we will show that the nonlocal nonlinear term in eq . violates these conservation laws .
we note that in absence of the landau damping effects , i.e. , for @xmath115 , the nls equation possesses an infinite number of conservation laws .
the first three conserving quantities are namely , the mass @xmath116 , the momentum @xmath117 and the energy @xmath118 .
the similar quantities , however , for the nls equation with landau damping satisfy the following equations : @xmath119 @xmath120 @xmath121 where the terms @xmath122 and @xmath104 ( the cauchy principal value integrals ) are due to the linear and nonlinear landau damping effects .
next , in eq . using the fact that the integral over @xmath4 is a convolution of the functions @xmath123 $ ] and @xmath124 , where @xmath125 , and noting that the fourier inverse transform of @xmath126 , we have @xmath127 so , performing the integral over @xmath128 as a fourier transform of @xmath129 we obtain @xmath130 where ` hat ' denotes the fourier transform with respect to @xmath4 or @xmath128 .
furthermore , using @xmath131 we obtain from eq .
@xmath132 from this equation we observe that if the linear landau damping is a higher - order effect than @xmath133 , the term @xmath122 can be neglected , and so the the left - hand side of eq .
represents the rate of change of the wave energy .
also , the integral on the right - hand side is positive definite , and to be shown later that for long wavelength ep plasma oscillations , @xmath104 is always positive for @xmath134 and @xmath11 , and negative for @xmath135 .
furthermore , it has been shown in ref . and will be shown in this work also that the relevant results may not be valid for @xmath135 .
thus , in both the superextensive and subextensive ep plasmas we have the inequality ( the equality holds for @xmath136 ) @xmath137 implying that an initial perturbation ( e.g. , in the form a soliton ) will decay to zero with time @xmath5 , and hence a steady state solution with @xmath138 of the nls equation may not exist in presence of the nonlinear landau damping term in nonextensive ep plasmas .
the nls equation in sec . [ section - nls - equation ] has been derived in a general way to describe the evolution of electrostatic waves in plasmas with arbitrary species @xmath139 and with arbitrary velocity distribution function for equilibrium plasma state . however , the main purpose of the present work is to investigate the dispersion properties of carrier wave modes , the linear landau damping rate associated with the resonant particles with the phase velocity , the modulational instability and nonlinear landau damping due to resonant particles having the group velocity as well as nonlinear evolution of wave envelopes in ep plasmas with @xmath0-nonextensive stationary states . in the latter ,
the @xmath0-distribution function in one space dimension is given by @xcite @xmath140 ^{1/(q_\alpha-1 ) } , \label{q - distribution}\ ] ] where @xmath141 and @xmath142 are respectively the mass and temperature of @xmath139-species particles ( @xmath143 , respectively , stand for electrons and positrons ) and @xmath144 is the boltzmann constant . the spectral index @xmath145 gives a measure that determines the slope of the energy spectrum of nonthermal particles .
it also measures the deviation from the thermal equilibrium state @xmath146 .
we consider a fully symmetric and charge - neutral ep plasma in which @xmath147 and @xmath148 .
such assumptions are justified with the experimental works in pure pair plasmas with particles having the same dynamics @xcite .
also , in the creation of a pure ep plasma , the whole system reaches a common thermal state with equal particle temperature .
so , the spectral index for electrons and positrons are taken as the same , i.e. , @xmath149 @xcite .
the normalization constant @xmath150 is given by @xmath151 , where @xmath152 is the particle s thermal velocity and @xmath153 is defined by @xmath154 note that in the extensive limit @xmath155 , the distribution function reduces to that of the standard maxwell - boltzmann distribution @xmath156 .
however , in the present work the case of @xmath155 may not be recovered directly because of simplifications of some expressions in the subsequent analysis , e.g. , for long - wavelength perturbations . the distribution function with @xmath10 represents the superextensivity , i.e. , presence of more ( compared to the maxwellian ) particles with velocities larger than their thermal velocities ( superthermal particles ) , whereas the case of subextensive distribution is represented by @xmath11 , meaning that there is a large number of particles with velocities lower than their thermal velocities .
furthermore , because of long - range nature of coulomb forces between the plasma particles and the presence of many superthermal particles in astrophysical environments @xcite , a @xmath0-distribution with @xmath10 is strongly suggested in real plasma systems or superthermal plasmas . for more about the behaviors of the distribution function @xmath22 for different values of @xmath0 ,
the readers are referred to , e.g. , ref . .
in what follows , we consider the nonlinear propagation of high - frequency @xmath157 and long wavelength [ @xmath158 , where @xmath159 is the debye wave number and @xmath160 is the equilibrium plasma number density ] oscillations in @xmath0-nonextensive plasmas whose phase velocity greatly exceeds the thermal velocities of electrons and positrons , i.e. , @xmath161 . in this case , the linear dispersion laws and the coefficients of the modified nls equation can be simplified . to this end
, we evaluate the cauchy principal value integral in eq . with the following expansion for the integrand @xcite @xmath162 where the integration limits are taken as @xmath163 for @xmath10 and @xmath164 for @xmath11 .
these limits are considered due to the fact that for @xmath11 , the distribution function has a thermal cutoff , which limits the velocity of particles to a maximum value , i.e. , @xmath165 .
however , such cutoff is absent when @xmath10 . in this case
, the velocity of particles remains unbounded ( for details , see , e.g. , ref . ) .
next , to evaluate the integrals in eq .
we note that the @xmath0-distribution function @xmath166 is an even function of @xmath23 , while @xmath167 is an odd function of @xmath23 .
thus , one can evaluate the integrals as follows : @xmath168 we have evaluated the above integrals by parts and the average value of @xmath169 as @xmath170 we mention that for @xmath11 , the above integrals , in which the limits are @xmath171 , are obtained by reducing the integrals to beta functions of the form @xmath172 with @xmath173 and finally to gamma functions using the relation @xmath174 .
similar integrals can also be evaluated for @xmath175 in which the limits are @xmath176 by using the relation @xmath177 and the above relation between beta and gamma functions . however , in each of these cases of superextensive and subextensive plasmas , we will obtain the same results by means of eq .
except the factors @xmath178 for @xmath175 and @xmath179 for @xmath11 . from eq .
we also note that the values of @xmath180 are further restricted to the region @xmath134 , because otherwise , the average value of @xmath169 may diverge .
in particular , in the limit @xmath155 , eq . reduces to the well known energy equipartition relation @xmath181 .
thus , our results may be valid for both the superextensive @xmath182 and subextensive @xmath183 distributions of electrons and positrons in plasmas .
now , we use the results as in eqs . and , and in the region of small wave numbers @xmath184 to obtain from eq .
the following dispersion relation for electrostatic carrier waves in a nonextensive ep plasma : @xmath185 , \label{dispersion - reduced}\ ] ] together with the linear landau damping rate , obtained from eq .
, as @xmath186^{(2-q)/(q-1 ) } , \label{landau - damping - linear - nonextensive}\ ] ] where @xmath187 is the plasma oscillation frequency in a charge neutral ep plasma . in the same way
, the group velocity expression reduces to @xmath188 .
\label{lambda - grop - velocity}\ ] ] the dispersion relation exactly agrees with that obtained in ref . up to the second order correction term @xmath189 .
however , we observe some interesting features by retaining the term @xmath190 in eq . which were overlooked in ref .
these interesting features , as can be seen from fig .
[ fig1 ] , are that the carrier wave dispersion curve and hence the group velocity of the wave envelope can turn over through the @xmath191-axis , going to zero values and then to negative values .
these reduction of the wave frequency and the group velocity occurs in the superextensive sub - region @xmath192 . beyond this region ,
i.e. , for @xmath193 , both @xmath194 and @xmath81 increase with increasing values of @xmath191 .
furthermore , it is observed that as one goes from the superextensive sub - regime @xmath195 to subextensive one with @xmath11 , the wave frequency and hence the group velocity are seen to get reduced .
this is expected since more the superthermal particles the larger is the phase velocity , in agreement with the previous results @xcite . however , some disagreements are also there for the linear landau damping rate @xmath196 . here
, we mention that though its analytic expression is the same as obtained in ref .
, however , the features we observe here are quite distinct .
in fact , the possibility of growing instability , as shown in ref . , can not be made in the superextensive or subextensive plasma regions . from the mathematical expression of
@xmath196 one can check that it is always negative regardless of the values of @xmath0 and @xmath191 in @xmath134 ( or @xmath11 ) and @xmath197 respectively .
these are clear from fig .
physically , since the phase velocity of the carrier wave is assumed to be larger than the particle s velocity , the wave modes may lose energy to the particles instead of gaining energy from them , and thus be damped . from fig .
[ fig2 ] , we also find that there are two subregions of @xmath191 , namely @xmath198 and @xmath199 . in the former , the damping rate increases , while in the latter the same decreases with increasing values of @xmath191 ( see the left panel ) .
furthermore , the damping rate is seen to be higher in plasmas with more superthermal particles and with wavelength in @xmath200
. however , the same can be true with higher the number of low - speed particles ( or with increasing values of @xmath0 ) in the regime @xmath199 .
these are also clear from the right panel of fig .
[ fig2 ] . here , as @xmath0 increases ,
the damping rate increases , however , it gets reduced at long - wavelength perturbations ( see the dashed line ) .
[ subplot ( a ) ; eq . ] and the group velocity @xmath201 [ subplot ( b ) ; eq . ] of the wave envelope
are plotted against the carrier wave number @xmath202 for different values of the nonextensive parameters as in the figure .
it is seen that the carrier wave dispersion curve can turn over with the group velocity of the wave envelope going to zero and then to negative values ( see the solid and dashed lines ) in the subregion @xmath203 of the superextensive region @xmath134 . in the region @xmath204 ,
the values of @xmath201 greater than unity are inadmissible , while in the other regime @xmath193 , both the frequency and the group velocity assume only the positive values .
, width=672,height=240 ] [ the linear landau damping rate given by eq . ]
is plotted against @xmath202 [ left panel ( a ) ] and the nonextensive parameter @xmath0 [ right panel ( b ) ] for different values of @xmath0 and @xmath191 as in the figure . from the left panel
it is seen that in one subregion @xmath198 , the damping rate increases with @xmath191 , while in the other @xmath199 it reduces with @xmath191 .
furthermore , lower the percentage of superthermal particles , the higher is the landau damping rate .
the right panel shows that the damping rate becomes weaker in the limit of long - wavelength @xmath205 oscillations .
, width=672,height=240 ] next , we calculate the various terms ( in the limit @xmath206 ) which appear in the coefficients of the nls equation for the @xmath0 distribution as follows : @xmath207 @xmath208 , \label{beta - coeff}\ ] ] @xmath209 @xmath210 @xmath211,\label{delta - coeff}\ ] ] @xmath212 , \label{gamma - coeff}\ ] ] @xmath213 , \label{c - coeff}\ ] ] @xmath214 .
\label{d - coeff}\ ] ] thus , the coefficient @xmath16 of the nls equation reduces to @xmath215 .
\label{p - reduced}\ ] ] to reduce the other coefficients , namely @xmath17 and @xmath104 , we first obtain the expressions for @xmath216 and @xmath217 in the limit @xmath206 .
so , we calculate , respectively , the resonant and non - resonant contributions to @xmath216 as @xmath218 @xmath219 thus , it turns out that the resonant contribution to the nonlinear coupling coefficient @xmath17 being smaller can be disregarded to obtain @xmath220 however , the contribution from the group velocity resonance through @xmath217 gives rise the coefficient @xmath104 as @xmath221 note that the above expressions for @xmath103 and @xmath104 are obtained for @xmath134 .
the similar expressions for @xmath11 can also be obtained by replacing the factor @xmath222 by @xmath223 under the square root in the expressions for @xmath224 and @xmath17 . in sec .
[ sec - mi ] , we will find that though the condition for the modulational instability does not depend on the sign of @xmath15 , but on the presence of @xmath104 , however , the sign of @xmath15 are important for determining the values of the frequency shift and the energy transfer rate ( in particular , their maximum values ) as well as their values in some particular cases , namely when the wave intensity exceeds or smaller than a critical value which depends on @xmath103 and the wave number of modulation . on the other hand , in sec .
[ sec - conservation - laws ] we have seen that whether a steady state solution of the nls equation exists or not depends on the coefficient @xmath104 . furthermore , in sec . [ sec - soliton - sol ] we will also examine the effect of @xmath104 on solitary wave solution of the nls equation .
thus , it is useful to investigate the properties of @xmath103 and @xmath104 .
inspecting on the expressions for @xmath103 and @xmath104 , which explicitly depend on the nonextensive parameter @xmath0 and the nondimensional wave number @xmath225 , we find that for superextensive plasmas with @xmath226 , we have @xmath227 .
however , when @xmath11 , i.e. , for subextensive plasmas , we have @xmath228 @xmath229 in @xmath230 @xmath231 and for @xmath232 . also ,
@xmath233 in both the superextensive and subextensive regimes , i.e. , @xmath226 and @xmath11 with @xmath232 .
furthermore , @xmath16 , which has singularity at @xmath234 , is positive for @xmath193 , i.e. , for @xmath235 ( superextensive ) and @xmath11 ( subextensive ) .
however , in the superextensive subregion @xmath236 , @xmath16 can be negative or positive depending on the rage of values of @xmath191 in @xmath232 ( see the left panel of fig .
[ fig3 ] ) .
the regions for @xmath16 and @xmath15 are shown as contour plots in the @xmath237 planes in fig .
we find that in the superextensive regime , @xmath15 is negative for @xmath238 , however , it can be positive or negative in the sub - regime @xmath6 depending on the values of @xmath191 in @xmath232 ( see the middle panel ) . from right panel ( third from left )
we find that in subextensive plasmas , @xmath239 for @xmath240 and for @xmath232 except in some small regions of @xmath241 and @xmath191 in which @xmath242 .
the latter is also true for @xmath243 and @xmath232 .
we also note that @xmath244 are undefined at @xmath234 .
( shaded or gray ) and @xmath245 ( blank or white ) are shown in the @xmath237 @xmath246 plane . from the left panel ( a ) , it is seen that @xmath247 for @xmath248 , i.e. , in the regimes @xmath238 ( superextensive ) and @xmath11 ( subextensive ) .
however , in the other superextensive regime @xmath236 , it can be positive or negative depending on the values of @xmath191 in @xmath249 .
middle panel ( b ) shows that in the superextensive regime , @xmath15 is negative for @xmath238 , however , it can be positive or negative in the sub - regime @xmath6 depending on the values of @xmath191 in @xmath232 . from right panel ( c )
we find that in subextensive plasmas , @xmath239 for @xmath240 and for @xmath232 except in some small regions of @xmath241 and @xmath191 in which @xmath242 .
the latter is also true for @xmath243 and @xmath232 .
note that @xmath244 are undefined at @xmath234 .
, width=672,height=240 ]
here , we follow the same analysis as in ref . .
though , the relevant analysis is standard , however , we repeat here for the sake of readers . it is well known that when the group velocity dispersive coefficient @xmath250 and the local nonlinear ( cubic ) term @xmath251 of an ordinary nls equation have the same sign , i.e. , @xmath242 , its plane wave solution exhibits instability against a plane wave perturbation of its amplitude and phase @xcite .
however , the present modified nls equation contains , in addition to the usual dispersive and nonlinear terms , the nonlocal nonlinear term which is associated with the resonant particles having the group velocity of the wave .
thus , it is of interest to examine how the nonlocal term contributes to the modulational instability and describes the nonlinear landau damping process . here
, we assume that , before modulation , the nls equation has a plane wave solution of the form @xcite @xmath252 where @xmath253 and @xmath51 are real functions of @xmath4 and @xmath5 . substituting the solution into eq .
we get @xmath254 @xmath255 .
\label{sigma}\ ] ] as before , we assume that the linear landau damping term is higher order than @xmath133 and linearize eqs . and by splitting up @xmath253 and @xmath51 into their equilibrium ( with suffix @xmath256 ) and perturbation ( with suffix @xmath257 ) parts as : @xmath258 @xmath259 where @xmath260 and @xmath261 are , respectively , the wave frequency and the wave number of modulation .
now , under the perturbations and , the solution can be expressed as @xmath262 } \notag\\ & & + a_2(k)\sin{[(k_0+\epsilon k)x-(\omega_0+\epsilon \lambda k+ \epsilon^2 \omega)t]}\notag \\ & & + \text{similar terms with}~ k\rightarrow -k~ \text{and}~ \omega\rightarrow -\omega , \label{phi - expansion}\end{aligned}\ ] ] where @xmath263 it follows that due to the linearization of eqs . and
, the electric potential @xmath3 [ eq . ] describes the three - wave interaction of the unperturbed carrier wave @xmath264 and two side bands with wave numbers and frequencies @xmath265 and @xmath266 .
now , substituting the perturbation expansions and into eqs . and , and
assuming the smallness of the landau damping coefficient , i.e. , @xmath267 , we obtain after eliminating @xmath268 and @xmath269 the following equation @xmath270 where the coupling between @xmath271 and @xmath272 appears due to the nonlocal nonlinear term ( the second term on the right - hand side ) in eq . .
thus , for nonzero values of @xmath271 and @xmath272 we obtain the following dispersion relation for electrostatic wave envelopes in collisionless nonextensive ep plasmas . @xmath273 the negative sign on the right - hand side of eq .
shows that whatever be the signs of @xmath16 and @xmath17 , the langmuir wave packet is always unstable due to the nonzero coefficient @xmath104 associated with the resonant particles having the wave group velocity . since eq .
is , in general , complex in @xmath260 , we seek a general solution of it by considering @xmath274 with @xmath275 being reals , and obtain @xmath276^{1/2}|k| , \label{omega - real}\ ] ] @xmath277^{1/2}|k| , \label{omega - img}\ ] ] where we consider the upper ( lower ) sign for @xmath278 . in
what follows , we consider two different limits of the wave amplitude . in the small amplitude limit with @xmath279 , eqs . and reduce to @xmath280 since a linear dispersion relation can be expanded as @xmath281 , we take the upper sign for @xmath282 and lower sign for @xmath283 in eq . .
thus , for @xmath284 , we have @xmath285 in which the imaginary part @xmath286 is solely due to the resonant particles having the group velocity of the wave envelope .
it follows that the langmuir wave packets under the modulation is unstable . since @xmath233 for both superextensive @xmath182 and subextensive @xmath183 plasmas , this instability is a kind of decay , and of course independent of @xmath16 and @xmath17 .
thus , in the small amplitude limit as above , i.e. , when the wave intensity @xmath287 is well below a critical value , the real part of @xmath260 relates the group velocity dispersion , while the imaginary part of @xmath260 describes the nonlinear landau damping process . in the latter ,
the wave energy is transferred from the higher frequency side bands to lower frequency ones . from eq .
we also find that the frequency shift can be positive or negative depending on the values of @xmath0 and @xmath191 as in fig .
for example , @xmath288 in the regimes @xmath238 ( superextensive ) and @xmath11 ( subextensive ) . however , in the other superextensive regime @xmath236 , it can be positive or negative depending on the values of @xmath191 in @xmath249 .
furthermore , for a fixed wave number of modulation @xmath261 , the frequency shift gets significantly reduced in the region @xmath192 .
this is a consequence of the results to the fact that in this region , the carrier wave frequency turn over with the group velocity going to zero values and then to negative values ( see fig . [ fig1 ] ) .
in the other region of @xmath0 , i.e. , for @xmath193 , the frequency @xmath289 increases with @xmath191 . in the large amplitude limit with @xmath290 ,
the frequency shift @xmath291 and the transfer rate @xmath292 can be obtained as @xmath293^{1/2 } , \label{omega - real - final}\ ] ] @xmath294^{1/2 } , \label{omega - img - final}\ ] ] which require @xmath239 .
thus , it turns out that when the wave intensity @xmath287 greatly exceeds a critical value , the frequency shift and the energy transfer rate can be obtained only in the regions of @xmath239 as in the middle and right panels of fig .
[ fig3 ] both in superextensive and subextensive plasmas .
furthermore , we note that @xmath291 and @xmath292 are proportional to @xmath295 instead of @xmath287 as in the small amplitude case .
in particular , for @xmath296 or when the cubic nonlinearity ( local ) greatly dominates over the local nonlinear term , the modulated wave becomes unstable for @xmath297 as in the ordinary nls equation .
next , from eq . we also find that for a given value of @xmath287 , the maximum value of the growth rate @xmath292 can be achieved at the wave number @xmath298 and for @xmath242 where @xmath299 the corresponding maximum values of @xmath291 and @xmath292 are thus obtained from eqs . and
as @xmath300 @xmath301 it follows that the maximum values of the frequency shift and the energy transfer rate for modulated waves can be achieved only in the regions of @xmath0 and @xmath191 where @xmath242 .
figure [ fig3 ] ( middle and right panels ) confirms that a wide range of values of @xmath191 as well as @xmath0 , both for superthermal and subextensive plasmas , exist for which @xmath242 . from the above results we also conclude that in contrast to the ordinary nls equation ( as in fluid theory ) in which the modulational instability occurs only for @xmath242 , the modulated langmuir wave packets in q - nonextensive plasmas always becomes unstable by the effects of resonant particles having the group velocity of the wave irrespective of the sign of @xmath242 or @xmath239 . in the former , the maximum values of the frequency shift and the growth rate
are achieved for arbitrary amplitude of the pump ( unperturbed ) wave which may not be obtained in maxwellian plasmas @xcite , whereas the latter gives asymptotic values of the same for larger values of the wave intensity . in a general manner
, we numerically examine the properties of @xmath291 and @xmath292 given by eqs . and for different values of the nonextensive parameter @xmath0 .
the results are displayed in fig .
[ fig4 ] for both superextensive and subextensive plasmas . from the upper panel of this figure
, we find that corresponding to the superextensive regime @xmath192 where the carrier wave frequency turn over with the group velocity ( fig .
[ fig1 ] ) , two subregions of @xmath191 exist , in one of which @xmath291 decreases having cutoffs at lower @xmath191 , while it increases with increasing values of @xmath0 and @xmath191 . in the other region of @xmath0 , namely , @xmath193
, the frequency shift is seen to increase with @xmath191 without any cutoff , while it decreases with increasing values of @xmath0 . for the damping rate @xmath292 ,
some different features are observed ( see the lower panel ) . here , as @xmath0 increases ,
the values of @xmath302 decrease , i.e. , more the number of superthermal particles the higher is the rate of energy transfer from high - frequency side bands to low - frequency ones .
[ upper panel ( a ) ] and the energy transfer rate @xmath303 [ lower panel ( b ) ] , given by eqs . and , are shown with respect to the nondimensional carrier wave number @xmath202 in superextensive @xmath304 and subextensive @xmath183 plasmas . in the superextensive subregion @xmath305 where the group velocity of the wave envelope vanishes ( fig .
[ fig1 ] ) , the frequency shift and the energy transfer rates are also seen to have cutoffs .
, width=480,height=240 ]
in absence of both the linear and nonlinear landau damping effects , i.e. , for @xmath115 , the modified nls equation reduces to the following ordinary nls equation @xmath306 in the case where the modulational instability occurs for @xmath242 , a stationary solution ( bright soliton ) of eq . can be obtained as @xcite @xmath307 where the amplitude @xmath308 and the phase @xmath309 are given by @xmath310 @xmath311 , \label{theta}\ ] ] with @xmath312 being a constant and @xmath313 denoting the transformation in the moving ( with velocity @xmath314 ) frame of reference . in this section
our aim is to determine the effect of a small amount of the nonlinear landau damping @xmath315 associated with the resonant particles having the group velocity of the wave envelope on the soliton solution . here
, we disregard the small effect of the linear landau damping rate @xmath122 which has been assumed to be higher order than @xmath133 . from eq .
it is evident that an initial perturbation of the form will decay to zero .
thus , one might expect that the amplitude @xmath316 is no longer a constant but can decrease slowly with time , i.e. , @xmath317 .
we consider @xmath318 and do the perturbation analysis of the nls equation with @xmath104 as the small parameter .
it can be easily verified that this assumption is valid for both superextensive and subextensive regions of @xmath0 as mentioned before .
we follow the similar approach as has been applied in different studies , however , to korteweg - de vries ( kdv ) equations ( see , e.g. , refs .
and references therein ) .
now , under the transformation @xmath313 , eq . reduces to @xmath319 where we rewrite @xmath320 to denote @xmath104 as small ( we will replace again @xmath321 by @xmath104 in the final solution ) and @xmath322 at @xmath323 . in what follows , to investigate the solution of eq .
we generalize the multiple time scale analysis with respect to @xmath321 , i.e. , we consider the solution as @xmath324 where @xmath325 , are functions of @xmath326 substituting eq . into eq . and equating the coefficients of the zeroth and first orders of @xmath321 , we successively obtain @xmath327 @xmath328 where @xmath329 and @xmath330 are given by @xmath331 it can easily be shown that under the boundary conditions , namely @xmath332 as @xmath333 , eq .
possesses a solution of the form @xmath334
@xmath335 @xmath336 for some real values of @xmath337 .
now , for the existence of solution of eq . , it is necessary that @xmath338 be orthogonal to all solution @xmath339 of @xmath340=0 $ ] which satisfy @xmath341 , where @xmath342 is the operator adjoint to @xmath343 defined by @xmath344dz=\int_{-\infty}^{\infty } \psi_2(z)l^+[\psi_1(z)]dz , \label{177 eq}\ ] ] with @xmath345 , and the only solution of @xmath340=0 $ ] is @xmath346 .
thus , we have @xmath347 which gives @xmath348-\frac{1}{\pi}\widetilde{\phi}_0^{3/2}{\cal p}\int_{-\infty}^{\infty}\int_{-\infty}^{\infty}\left(\frac{\text{sech}^2 z'}{z - z'}\right ) \text{sech}^2z \exp(i\theta_1z)dz dz'=0 . \label{eq1-phi0-tilde}\ ] ] equation is a first - order differential equation for the wave amplitude @xmath349 .
so , its solution can be obtained as @xmath350 where @xmath351 at @xmath352 and @xmath353 is given by ( rewriting now @xmath321 as @xmath104 ) @xmath354 { \cal{p } } \int_{-\infty}^{\infty}\int_{-\infty}^{\infty}\left(\frac{\text{sech}^2 z'}{z - z'}\right ) \text{sech}^2z \exp(i\theta_1z)dz dz'.\ ] ] thus , when @xmath242 , an approximate solitary wave solution of the nls equation with a small effect of the nonlinear landau damping is given by @xmath355 where @xmath356/2p$ ] .
on the other hand , when @xmath239 , a stationary solution ( dark soliton ) of eq .
can be taken as @xcite @xmath357 where @xmath308 and @xmath309 are different from those given by eqs . and , i.e. , @xmath358 @xmath359 , \label{theta - pq<0}\ ] ] with @xmath360 a constant .
proceeding in the same way as above up to eq . , it can be easily verified that @xmath361 , for some @xmath362 , is a solution of eq . if and only if @xmath336 . also , for the existence of a solution of eq .
we have the same necessary condition but with different @xmath363 .
the condition gives @xmath364 from which we obtain @xmath365 + \frac{1}{\pi}\widetilde{\phi}_0 ^ 3{\cal{p}}\int_{-\infty}^{\infty}\int_{-\infty}^{\infty}\left(\frac{\tanh^2 z'}{z - z'}\right ) \tanh^2 z \exp(i\theta_2z)dz dz'=0 , \label{eq2-phi0-tilde}\ ] ] where @xmath351 at @xmath352 .
as before , the solution of eq . in the case of @xmath239 is given by @xmath366 where ( rewriting @xmath321 as @xmath104 ) @xmath367}{\delta(\tau)(1-\cosh\pi\theta_2)+\theta_2 \sinh\left(\frac{\pi\theta_2}{2}\right ) } { \cal{p } } \int_{-\infty}^{\infty}\int_{-\infty}^{\infty}\left(\frac{\tanh^2 z'}{z - z'}\right ) \tanh^2z \exp(i\theta_2z)dz dz',\ ] ] with @xmath368 denoting the dirac delta function .
thus , when @xmath239 an approximate solitary wave solution of the nls equation with a small effect of the nonlinear landau damping is given by @xmath369 where @xmath309 is given by eq . . from eqs . and
, it is evident that the absolute value @xmath370 of the wave amplitude decays slowly with time @xmath5 with a small effect of the nonlinear landau damping .
figure [ fig5 ] exhibits a qualitative plot of the absolute value of @xmath3 given by eq . in plasmas with superextensive and subextensive velocity distributions .
it is seen that faster the decay of the wave amplitude , the larger is the percentage of superthermal particles in plasmas . is shown for both superextensive ( solid and dashed lines ) and subextensive ( dotted line ) plasmas .
it is seen that faster the decay of the wave amplitude , the larger is the percentage of superthermal particles in plasmas .
, width=480,height=240 ]
we have investigated the amplitude modulation and the nonlinear evolution of electrostatic wave envelopes in a collisionless electron - positron pair plasma in the context of tsallis @xmath0 nonextensive statistics .
starting from a set of vlasov - poisson equations and applying the reductive perturbation technique , the dynamics of the wave envelopes is shown to be governed by a nls equation with a nonlocal nonlinear term arising from resonant particles having the group velocity of the wave envelope .
such wave - particle resonance also modifies the local nonlinear ( cubic ) coupling coefficient of the nls equation .
furthermore , the nonextensive parameter @xmath0 , which measures the excess of superthermal particles in plasmas , is shown to modify the dispersive ( group velocity ) , local nonlinear ( cubic ) as well as the nonlocal nonlinear terms significantly . an interesting effect of @xmath0 is that a subregion @xmath371 of the superextensivity @xmath182 exists where the carrier wave dispersion curve @xmath372 can turn over with the group velocity @xmath373 going to zero and then to negative values ( fig .
[ fig1 ] ) .
such features of the dispersion curve have not been reported in the previous work @xcite where the same plasma model has been considered to investigate linear landau damping of langmuir oscillations in @xmath0-nonextensive ep plasmas . in ref .
these effects were absent due to truncation of the wave frequency up to @xmath374 .
furthermore , our results in the linear theory show that the electrostatic wave is always damped due to resonant particles having phase velocity of the wave in both superextensive and subextensive regimes and also for long - wavelength perturbations .
these are also in disagreement with the results of ref . where the possibility of growing instability has been predicted .
we , however , stress that such a growing instability should not appear for high - frequency oscillations in @xmath0-nonextensive ep pair plasmas .
this is due to the fact that since the wave phase velocity is assumed to be larger than the particle s velocity , the wave modes can be damped by losing their energy to the particles .
it is found that for a fixed @xmath0 , two subregions of @xmath191 exist , in one of which the linear damping rate @xmath196 becomes higher , while in the other it gets reduced with increasing values of @xmath191 .
on the other hand , for a fixed @xmath191 , the absolute value of @xmath196 increases with increasing values of @xmath0 , however , its value gets reduced in larger wavelength of perturbations . in the nonlinear regime , we have verified the conservation laws as applicable for a nls equation . it is found that unlike the ordinary nls equation , the nonlocal nonlinear term , associated with the nonlinear wave - particle resonance , violates the conservation laws , leading to a decay of the wave amplitude with time and thereby forbidding the existence of a steady state solution of the modified nls equation .
we show that the modulated wave packet is always unstable ( regardless of the sign of @xmath16 and @xmath17 ) due to the nonlocal nonlinear term which can describe the nonlinear landau damping process in which the wave energy is transferred from higher to lower frequency sidebands .
the frequency shift @xmath375 and the energy transfer rate @xmath376 for the modulated waves are also examined by the parameter @xmath0 in a general way as well as in the limits of small @xmath377 and large @xmath378 amplitudes , where @xmath379 is some critical value of the pump wave intensity @xmath287 .
it is found that both @xmath291 and @xmath292 attain their maximum values only when @xmath242 .
such maxima of @xmath291 and @xmath292 may not exist in pair plasmas or electron - ion plasmas with maxwellian distributions @xcite .
however , they assume some asymptotic values in the limit @xmath380 and when @xmath239 , which are @xmath381 instead of @xmath287 as in the opposite limit @xmath382 . the regions of @xmath0 and @xmath191 for which @xmath383 and @xmath384 are also obtained in both superextensive @xmath7 and subextensive @xmath183 regimes .
the general expressions of both @xmath291 and @xmath292 are also studied by the effects of @xmath0 .
it is found that corresponding to the superextensive regime @xmath192 where the carrier wave frequency vanishes with the group velocity ( fig .
[ fig1 ] ) , two subregions of @xmath191 exist , in one of which a significant reduction of @xmath291 occurs having cutoffs at lower @xmath191 , while it gets enhanced with increasing values of @xmath0 and @xmath191 .
the existence of such cutoffs are the consequences of the turn over effects of the carrier wave frequency as well as the group velocity of the wave envelope . in the other region of @xmath0 , namely , @xmath193
, the frequency shift is seen to increase with @xmath191 ( having no cutoff ) , while it decreases with increasing values of @xmath0 .
however , quite distinct features are observed for the energy transfer rate @xmath292 . here , as @xmath385 increases , the values of @xmath302 decrease , i.e. , more the percentage of superthermal particles the higher is the rate of energy transfer from high - frequency side bands to low - frequency ones .
we have also studied the effects of a small amount of the nonlinear landau damping @xmath286 on the solitary wave solution of the ordinary nls equation by assuming that the linear landau damping @xmath386 rate is higher order than @xmath133 .
it is found that the wave amplitude decays with time and the decay rate can be faster the larger is the number of superthermal particles in ep plasmas . to conclude , the results should be useful for the evolution of nonlinear wave envelopes and associated wave damping in collisionless pure ep plasmas or pure pair - ion plasmas such those in laboratory @xcite , space @xcite and astrophysical @xcite environments .
equating the coefficients of @xmath35 from eqs . and
for @xmath55 we successively obtain @xmath387 for nonzero perturbations , eq .
yields the following linear dispersion law : @xmath388 the small parameter @xmath35 , which measures the amplitude of the wave , can be related to the linear landau damping rate @xmath196 of the plasma wave as @xmath389 , where @xmath390 is a nonnegative integer and @xmath194 is the real part of @xmath391 .
next , from eq .
we have the plasma dielectric function @xmath392}{v-\omega / k}=0,\label{dk - omega}\ ] ] where for pair plasmas or ep plasmas @xmath148 .
if @xmath393 , i.e. , @xmath389 , then the dielectric function @xmath394 can be taylor expanded in the smallness of @xmath196 to obtain its real ( with suffix r ) and imaginary ( with suffix i ) parts as @xcite @xmath395}{v-\omega_r / k } , \label{dielectric real fnc}\ ] ] @xmath396_{v=\omega_r / k}. \label{dielectric i m fnc}\ ] ] next , neglecting the terms of order @xmath397 and higher in eqs . and , the expressions for @xmath194 and @xmath196 can be obtained from the following relations @xmath398 @xmath399
we equate the components for @xmath62 and @xmath63 .
thus , from eq . we have @xmath401 this gives @xmath402 where @xmath403 for @xmath404 has been introduced to anticipate that the solution in the linear approximation decays with landau damping rate . similarly , for @xmath62 and @xmath63 equating the coefficients of @xmath35 from eq .
we obtain @xmath405 next , substituting eq . into eq . and noting that @xmath406 we obtain @xmath407 where the contour @xmath408 enables us to remove @xmath409 in the denominator of eq . .
thus , by means of the dispersion relation , one must have @xmath410
we consider the second order , first harmonic modes with @xmath66 and @xmath79 .
thus , from eq .
we have @xmath411 which gives @xmath412 .
\label{f - vlasov - n2-l1}\ ] ] also , from eq .
we have for @xmath66 and @xmath79 the following equation @xmath413 this equation can be reduced , by substituting from eq .
the expression of @xmath414 into it and using the dispersion relation and the expression , to @xmath415 dv=0 , \label{phi - poisson - n2-l1-reduced}\ ] ] equation can be written in the form @xmath416
for @xmath88 we have from eqs . @xmath417
which by means of eqs . and yields @xmath89 .
\label{f_alpha2-n2-l2}\ ] ] next , the expression for @xmath418 is obtained from eq . for @xmath88 and using the relation as @xmath419 i.e. , @xmath420
for @xmath94 and @xmath67 we have from eq .
@xmath421 where we have used the relations .
this equation can further be reduced after few steps using eqs . , and to the following equation : @xmath422 where @xmath423 next , taking the fourier - laplace transform of eq .
with respect to @xmath95 and @xmath51 and using the initial condition we obtain @xmath424 where @xmath425 is defined as @xmath426 , \label{phi-1 - 1-mod}\ ] ] with @xmath427 now , for @xmath428 eq . reduces to @xmath429 thus , we obtain from eq .
the following relation @xmath97 where @xmath430 @xmath431 next , a substitution of eq .
back into eq . results a slow beat wave solution given by @xmath96 h(k,\omega ) .
\label{ff - alpha - n2-l0-final}\ ] ]
we can now use the expressions for the lower - order quantities so obtained to determine the components for @xmath94 and @xmath79 .
thus , we have from eq .
@xmath432 \right\rbrace \notag\\ & & + ik\frac{g_\alpha}{(\omega - kv)^2}v \frac{\partial}{\partial \zeta}\phi^{(1)}_1 + 2 \frac{e_\alpha}{m_\alpha } \frac{k^2}{\omega - kv}\frac{\partial}{\partial v}\left\lbrace \frac{g_\alpha}{\omega - kv}\right\rbrace \phi^{(1)}_{-1}\phi^{(2)}_2 \notag\\ & & -\frac{e_\alpha}{m_\alpha } \frac{k^2}{\omega - kv}\frac{\partial}{\partial v}\left\lbrace \frac{g_\alpha}{\omega - kv}\phi^{(2)}_2 -\frac{e_\alpha}{2m_\alpha } \frac{k}{\omega - kv}\frac{\partial}{\partial v}\left ( \frac{g_\alpha}{\omega - kv}\right ) \phi^{(1)}_{1}\phi^{(1)}_1\right\rbrace \phi^{(1)}_{-1 } -\frac{e_\alpha}{m_\alpha } \frac{k}{\omega - kv}\frac{\partial}{\partial v}f^{(2)}_{\alpha,0}\phi^{(1)}_{1}. \label{f - alpha - n3-l1}\end{aligned}\ ] ] this expression for @xmath433 is then substituted in eq . for @xmath100 to obtain the following equation @xmath434 where @xmath435 is unity for @xmath436 and vanishes otherwise , and @xmath292 denotes the path of integration around @xmath59 in the anticlockwise sense .
note that this integral results from the term @xmath437 representing the landau damping term .
now , substituting the expression into the coefficient of @xmath85 of the sixth term of eq .
we obtain the term as @xmath438\notag\\ & & \times\left[{{{\cal{w}}(k,\omega)}^2/\delta^{(c)}(k,\omega ) + { \cal{c } } ( k,\omega ) } \right ] , \label{sixth - term - phi - n3-l1}\end{aligned}\ ] ] where @xmath439 by deforming the contour @xmath408 to the landau contour with real @xmath260 we have @xmath440 thus , using the relation for @xmath425 , the functions defined by eqs . , and can , respectively , be expressed as follows : @xmath441 with @xmath442 @xmath443 similarly , @xmath444 with @xmath445 @xmath446 also , we have @xmath447 where @xmath448 @xmath449
we note that the group velocity @xmath450 is independent of @xmath261 and @xmath260 and we have the relation @xmath451 in terms of the constrained coordinate .
thus , we finally obtain the term given by eq . as @xmath452 where the symbols used are @xmath453 , \label{theta - k - omega}\ ] ] @xmath454 .
\label{phi - k - omega}\ ] ] here , the functions @xmath455 and @xmath292 represent contributions of the resonant particles having the group velocity of the wave envelope in ep plasmas .
now , the third term in eq . can be reduced by eliminating @xmath456 with the help of eq .
as @xmath457 in which the first term of it cancels the second term of eq . by means of the expression of the group velocity
furthermore , inspecting on the first term of eq .
we find that the derivative @xmath458 can be transformed into the derivative @xmath459 by redefining the variable @xmath460 .
finally , substitutions of eqs . and into eq .
give @xmath461 50 w. misner , k. s. thorne , and j. a. wheeler , _ gravitation _ ( freeman , san francisco , 1973 ) , p. 763 ; g. w. gibbons , s. w. hawking , and s. siklos , _ the very early universe _ ( cambridge university press , cambridge , uk , 1983 ) .
a. p. lightman , astrophys .
j. * 253 * , 842 ( 1982 ) ; m. l. burns and r. v. e. lovelace , _ ibid . _ * 262 * , 87 ( 1982 ) ; a. p. lightman and a. a. zdziarski , _ ibid . _ * 319 * , 643 ( 1987 ) ; m. y. yu , p. k. shukla , and l. stenflo , _ ibid .
_ * 309 * , l63 ( 1986 ) .
r. d. blandford and r. l. znajek , _ electromagnetic extraction of energy from kerr black holes _ mon . not .
. soc . * 179 * , 433 ( 1977 ) .
p. goldreich and w. h. julian , _ pulsar electrodynamics _ astrophys .
j. * 157 * , 869 ( 1969 ) .
j. f. c. wardle _ et al .
_ , _ electron - positron jets associated with the quasar 3c279 _ nature * 395 * , 457 ( 1998 ) . m. c. begelman , r. d. blandford , and m. d. rees , rev .
mod . phys . * 56 * , 255 ( 1984 ) ; h. r. miller and p. j. witta , in _ active galactic nuclei _
( springer , berlin , 1987 ) , p. 202 .
j. r. orsoz , r. a. remillard , c. d. bailyn , and j. e. mcclintock , astrophys .
j. * 478 * , l83 ( 1997 )
. g. sarri _
_ , _ generation of neutral and high - density electron
positron pair plasmas in the laboratory _ nat . commun .
* 6 * , 6747 ( 2015 ) . c. m. surko , m. leventhal , and a. passner , phys .
lett . * 62 * , 901 ( 1989 ) ; r. g. greaves and c. m. surko , _ ibid .
_ * 75 * , 3846 ( 1995 ) .
h. boehmer , m. adams , and n. rynn , phys .
plasmas * 2 * , 4369 ( 1995 ) . w. oohara and r. hatakeyama , phys . rev
. lett . * 91 * , 205005 ( 2003 ) .
w. oohara , d. date , and r. hatakeyama , phys .
lett . * 95 * , 175003 ( 2005 ) .
jao and l .- n .
hau , phys .
e * 89 * , 053104 ( 2014 ) .
e. saberian and a. esfandyari - kalejahi , phys .
e * 87 * , 053112 ( 2013 ) .
a. p. misra and a. roy chowdhury , phys .
e * 70 * , 058401 ( 2004 ) .
plasmas * 21 * , 022108 ( 2014 ) .
u. a. mofiz , phys .
a * 42 * , 960 ( 1990 ) .
f. a. asenjo et al . , phys .
e * 85 * , 046406 ( 2012 ) .
n. s. javan , phys .
plasmas * 19 * , 122107 ( 2012 ) .
e. ott and r. n. sudan , phys .
fluids * 12 * , 2388 ( 1969 ) ; _ ibid _ * 13 * , 1432 ( 1970 ) .
a. barman and a. p. misra , phys .
plasmas * 21 * , 073708 ( 2014 ) .
a. p. misra and a. barman , phys .
plasmas * 22 * , 073708 ( 2015 ) .
g. brodin , j. zamanian , and j. t. mendonca , phys .
scr . * 90 * , 068020 ( 2015 ) .
m. d. montgomery , s. j. bame , and a. j. hundhausen , j. geophys . res . *
73 * , 4999 ( 1968 ) ; m. maksimovic , v. pierrard , and p. riley , geophys .
* 24 * , 1151 ( 1997 ) ; i. zouganelis , j. geophys
a08111 ( 2008 ) . j. m. liu _
et al . _ ,
* 72 * ( 1994 ) 2717 .
v. pierrard and m. lazar solar phys .
* 267 * 153 ( 2010 ) and references therein .
a. renyi , acta math .
hungar.*6 * , 285 ( 1955 ) .
c. tsallis , j. stat .
phys . * 52 * , 479 ( 1988 ) .
f. caruso and c. tsallis , phys .
e * 78 * , 021102 ( 2008 ) .
f. d. nobre , m. a. rego - monteiro , and c. tsallis , phys .
* 106 * , 140601 ( 2011 ) .
s. guo , l. mei , and a. sun , annals .
* 332 * , 38 ( 2013 ) .
m. p. leubner , astrophys .
space sci . * 282 * , 573 ( 2002 ) .
l. landau , j. phys .
( ussr ) * 10 * , 25 ( 1946 )
. j. h. malmberg and c. b. wharton , phys .
* 13 * , 184 ( 1964 ) .
h. ikezi and y. kiwamoto , phys .
* 27 * , 718 ( 1971 ) .
y. h. ichikawa , suppl .
progress th .
* 55 * , 212 ( 1974 ) .
r. silva , jr .
, a. r. plastino , and j. a. s. lima , phys .
a * 249 * , 401 ( 1998 ) .
e. m. f. curado , braz .
* 29 * , 36 ( 1999 ) ; s. abe , physica ( amsterdam ) * 269a * , 403 ( 1999 ) . j. a. s. lima , r. silva , and a. r. plastino , phys . rev .
lett . * 86 * , 2938 ( 2001 ) .
t. taniuti and n. yajima , j. math . phys .
* 10 * , 1369 ( 1969 ) .
r. fedele , h. schamel , and p. k. shukla , phys .
* t98 * , 18 ( 2002 ) ; r. fedele , phys .
65 * , 502 ( 2002 ) ; r. fedele and h. schamel , eur .
j. b * 27 * , 313 ( 2002 ) . | the nonlinear theory of amplitude modulation of electrostatic wave envelopes in a collisionless electron - positron ( ep ) pair plasma is studied by using a set of vlasov - poisson equations in the context of tsallis @xmath0-nonextensive statistics . in particular , the previous linear theory of langmuir oscillations in ep plasmas [ phys .
rev .
e * 87 * , 053112 ( 2013 ) ] is rectified and modified . applying the multiple scale technique ( mst )
, it is shown that the evolution of electrostatic wave envelopes is governed by a nonlinear schrdinger ( nls ) equation with a nonlocal nonlinear term @xmath1 [ where @xmath2 denotes the cauchy principal value , @xmath3 is the small - amplitude electrostatic ( complex ) potential , and @xmath4 and @xmath5 are the stretched coordinates in mst ] which appears due to the wave - particle resonance .
it is found that a subregion @xmath6 of superextensivity @xmath7 exists where the carrier wave frequency can turn over with the group velocity going to zero and then to negative values .
the effects of the nonlocal nonlinear term and the nonextensive parameter @xmath0 are examined on the modulational instability ( mi ) of wave envelopes as well as on the solitary wave solution of the nls equation .
it is found that the modulated wave packet is always unstable ( nonlinear landau damping ) due to the nonlocal nonlinearity in the nls equation .
furthermore , the effect of the nonlinear landau damping is to slow down the amplitude of the wave envelope , and the corresponding decay rate can be faster the larger is the number of superthermal particles in pair plasmas . |
with an orbital period of only 11.4 minutes , 4u 1820 - 30 is the most compact low mass x - ray binary ( lmxb ) known ( stella et al .
thermonuclear x - ray bursts were discovered from this source by grindlay & gursky ( 1976 ) and attest that the primary is a neutron star . using vela
5b data priedhorsky & terrell ( 1984 ) found a 176 day periodic modulation of the x - ray flux from 4u 1820 - 30 .
bursts have only been observed when the accretion driven flux is near the low end of its observed range , @xmath12 ergs @xmath8 s@xmath2
. mass transfer in this system can be driven by gravitational radiation at the rate of @xmath13 g s@xmath2 .
the binary resides in the globular cluster ngc 6624 .
optical observations give a distance estimate of 7.6 kpc ( hesser & shawl 1985 ; rich , minniti , & liebert 1993 ) .
based on the peak flux of photospheric radius expansion x - ray bursts , vacca , lewin , & van paradijs ( 1986 ) estimated a likely distance of @xmath14 kpc .
the compact nature of the system requires that the secondary be a low mass helium dwarf ( see rappaport et al .
1987 ) , so that the accreted material likely has a very high helium abundance .
the observation of photospheric radius expansion bursts is consistent with the idea that helium flashes are the primary fuel for the 1020 s duration bursts most commonly observed .
modulations of the uv flux from 4u 1820 - 30 at the 11.4 minute orbital period were predicted by arons & king ( 1993 ) and subsequently discovered by anderson et al .
( 1997 ) . the uv modulations result from x - ray heating of the secondary and constrain the system inclination @xmath15 to the range @xmath16 .
smale et al . ( 1997 ) reported the discovery of kilohertz quasiperiodic oscillations ( qpo ) from 4u 1820 - 30 . since then further observations have revealed that the highest observed khz qpo appear to reach a saturation frequency near 1,050 hz , suggesting the presence of a last stable circular orbit as predicted by general relativity ( gr ) ( zhang , strohmayer & swank 1998 ; kaaret , ford & chen 1998 ) .
if this hypothesis is correct it also suggests that the neutron star in 4u 1820 - 30 may be as massive as @xmath17 ( see also arons & king 1993 ) .
constraints derived from the study of x - ray bursts can provide information on the neutron star mass and radius and thus might also provide a key test of these conclusions .
recently , cornelisse et al .
( 2000 ) reported the detection of a very long ( @xmath18 hr ) burst from 4u 1735 - 44 with the wide field camera ( wfc ) on _ bepposax_. spectral softening in this burst led the authors to conclude that it was thermonuclear in origin .
there have been other recent reports of very long thermonuclear bursts from lmxbs , for example , heise , in t zand & kuulkers ( 2000 ) report long bursts from ks 1731 - 260 and serpens x-1 , and kuulkers ( 2001 ) recently identified a likely super burst from gx 3 + 1 in rxte / asm data . as part of a rossi x - ray timing explorer ( rxte )
ao3 observing program to try and detect x - ray bursts as well as study khz qpo from 4u 1820 - 30 , rxte was observing the source on september 9 , 1999 ut when a long , powerful x - ray burst was observed .
this burst was @xmath19 times longer than the more frequent thermonuclear bursts seen from the source ( see for example haberl et al .
1987 ) . in this paper , we show that this super burst was thermonuclear in origin and that it was most likely fueled by burning of carbon ashes produced by the stable burning of helium accreted onto the neutron star in 4u 1820 - 30 . in this respect ,
4u 1820 - 30 differs from the other five super burst sources , which accrete a hydrogen / helium mixture and therefore produce much less carbon ( see cumming & bildsten 2001 ) .
a consequence of this is that the recurrence time for this source will be of the order of a decade , however , unless the carbon ignition is somehow `` triggered . ''
a very brief summary of the analysis in this paper was previously presented by strohmayer ( 2000 ) .
we also note that during the preparation of this paper a burst lasting several hours was observed during rxte observations of 4u 1636 - 53 in support of rxte proposal 50030 ( pi : strohmayer ) .
an analysis of this burst will be presented elsewhere .
the 4u 1636 - 53 event was independently found with the asm by wijnands ( 2001 ) , who also found a similar event in the asm archive which occurred @xmath20 years earlier .
the plan of this paper is as follows . in
2 we present the observed temporal and spectral characteristics as well as the energetics of the super burst . in 3 we discuss the implications of the discrete spectral components ( line and edge ) observed during the burst .
section 4 begins with a discussion of the production of carbon and the thermal structure of the `` ocean . ''
we assess the amount of carbon needed to produce a thermonuclear runaway and the evolution of the ensuing burst .
we conclude in 5 with a summary of our principal findings .
as mentioned above , 4u 1820 - 30 has a @xmath21 day x - ray intensity modulation ( priedhorsky & terrell 1984 ) . the all - sky monitor ( asm ) on board rxte has been monitoring the source for more than 5 years .
figure 1 shows a portion of the 212 kev asm lightcurve of 4u 1820 - 30 .
the dips down into the lowest flux state occur approximately at the 178 day period .
our rxte observations on mjd 51430 ( sep . 9 ,
1999 utc , indicated by the vertical arrow in the figure ) caught the source near its low flux state when bursts can be observed . in the weeks prior to our observation rxte had a malfunction in one of its antenna transponders which disabled one of the high - gain antennas with which rxte telemeters data to the ground . because of this antenna malfunction some data from our observation were unfortunately lost . for the proportional counter array ( pca )
we have only standard1 and standard2 data for the entire event .
these data provide 290 kev time histories with 1/8 s temporal resolution ( standard1 ) and 16 s spectral accumulations across the 290 kev bandpass ( standard2 ) , respectively .
although the pca bandpass is nominally 2 - 90 kev , for these data almost all source counts are in the 2 - 40 kev band .
we did obtain some high time resolution event mode data ( sampling rate of 1/8192 hz ) but only for two time intervals in the decaying portion of the burst .
unfortunately , we lost the high time resolution data during the onset of the burst
. we will investigate the timing properties of 4u 1820 - 30 at the time of the super burst in a subsequent publication .
the time history of the burst is shown in figure 2 .
the 290 kev pca standard1 lightcurve with 1/8 s time resolution is the higher time resolution histogram ( left axis ) , while the ( 830)/(28 ) kev hardness ratio from standard2 data is the lower resolution histogram ( right axis ) .
the burst was observed with 3 of the 5 proportional counter units ( pcus 0 , 2 , and 3 ) which comprise the pca .
note that the time axis is logarithmic .
the event began with a typical helium flash from 4u 1820 - 30 which extends from about 315 seconds in figure 2 .
however , before this first burst dies away the super burst erupts and is still decaying away 10,000 s later when our observation ended .
the event shows a clear ` precursor ' after which the x - ray flux dropped completely to background for a few seconds . a similar , but much weaker
precursor can be seen in the helium flash burst just prior to the onset of the super burst .
this behavior is characteristic of very powerful photospheric radius expansion bursts ( see hoffman et al . 1978 ; tawara et al .
1984 ; lewin , vacca & basinska 1984 ; van paradijs et al . 1990 ) .
note also the hardening of the spectrum following the precursor .
this is characteristic of the contraction phase of a radius expansion burst .
we found in the rxte archive one additional example of a helium flash burst from 4u 1820 - 30 . in figure 3
we compare this burst , observed on may 2 , 1997 at 17:32:39 utc , with the helium flash which preceded the super burst .
this he burst was observed with all 5 pcus , so in order to make a closer comparison we have scaled the observed count rate by 3/5 .
figure 3 shows the lightcurve of each burst at 1/8 s resolution from the standard1 data mode .
there are several things of note : both the preburst count rates as well as the peak count rate during the burst are similar , and each burst has a precursor or ` double peak ' which indicates photospheric radius expansion .
the above features of the time profile , combined with the observed behavior of the hardness ratio , are consistent with a thermonuclear origin for both bursts .
the much longer timescale of the super burst compared to the shorter , more frequent helium flashes suggests a larger reservoir of nuclear fuel released at much greater column depth below the surface .
these considerations suggest that carbon and/or oxygen are the likely fuel sources for the super burst .
later , in 4 , we will explore this scenario in greater detail .
we used the standard2 data to investigate the x - ray spectrum both before and throughout the burst . for most of the burst
we did not have spectral data in a high time resolution mode .
this was unfortunate , but since the evolution timescale of this burst is much longer than a typical 10 s burst it was only a significant problem near the onset of the burst where the data suggest the black body temperature was changing on a timescale shorter than our 16 s accumulations . before investigating the burst spectrum we analyzed the spectrum of the persistent , accretion driven flux .
we used the standard pca background estimation software , pcabackest , to determine the pca detector background .
for all the spectral results we used only the top layer of detectors 0 , 2 , and 3 of the pca .
we found that the persistent flux before the burst is well fit by a thermal bremsstrahlung model ( bremss in xspec ) with a temperature @xmath22 kev and flux of @xmath23 ergs @xmath8 s@xmath2 .
figure 4 shows the time history of the thermal bremsstrahlung temperature , @xmath24 ( top panel ) , and the derived 220 kev flux ( bottom panel ) . at a distance of 6.6 kpc
the observed flux corresponds to a luminosity of @xmath25 , consistent with previous measurements of the persistent luminosity when bursts are often observed ( clark et al .
1977 ; haberl et al .
1987 ) .
in most published analyses of burst spectral evolution the spectrum is obtained by subtracting off an estimate of the spectrum of the preburst , accretion driven flux .
this should be reasonable when the burst flux does not appreciably disturb the accretion flow , however , during powerful radius expansion bursts it may be that the burst driven wind and near - eddington flux can substantially effect the accretion process ( see , for example , walker & meszaros 1989 ) .
we investigated both subtracting off the preburst emission as a background and assuming that the accretion - driven flux is completely shut off by the burst .
for the latter case we simply subtracted off the appropriate detector background estimated with pcabackest . in general
we found that for the majority of the burst we were able to get better fits by subtracting off the preburst spectrum , so we will discuss our results using this method .
the best fit parameter values showed modest quantitative differences ( @xmath26 ) when estimated using the two different background assumptions , but all of our qualitative conclusions are robust in the sense that they were independent of which assumption we used for the background . because this burst is @xmath19 times longer than typical thermonuclear bursts the timescale on which the thermal continuum changes is also much longer .
this makes it possible to obtain very high signal to noise spectra during intervals in which the black body temperature remains fairly constant .
for example we were able to accumulate 64 s spectra throughout most of the burst and still obtain good fits to the continuum with single temperature black body models .
the spectrum during the burst is indeed thermal , and evolves in a way which is entirely consistent with thermonuclear bursts of shorter duration .
we found that a black body model with photoelectric absorption ( model wabs*bbodyrad in xspec ) alone usually did not provide an acceptable fit in the statistical sense , however , non - thermal continua gave much worse fits .
the residuals when fitting a black body function and photoelectric absorption strongly suggest the presence of an emission line near 6.4 kev as well as an absorption edge between 79 kev . to model these components we included a gaussian emission line as well as an edge in our spectral model ( model wabs*(bbodyrad+gaus)*edge in xspec ) . with these components included we were able to obtain acceptable fits with @xmath27 per degree of freedom @xmath28 .
figure 5 shows the characteristic residuals near the peak of the burst when the gaussian line and edge are removed from the model .
shown are the residuals , data minus model , in units of standard deviations .
these residuals appear to be qualitatively similar to those reported by van paradijs et al .
( 1990 ) for a powerful burst from 4u 2127 + 11 in m15 .
both the line and edge are strongly required to adequately model the data .
figure 6 shows the count rate spectrum and residuals from a typical fit near the peak of the burst including the line and edge components , and demonstrates the quality of fit we can achieve with this model . in general we find that the gaussian line at @xmath29 kev has finite width and often an energy centroid significantly less than 6.4 kev , suggestive of a line produced by reflection from a relativistic accretion disk in 4u 1820 - 30 .
day & done ( 1991 ) had suggested that such a signature should be present in burst spectra , but to our knowledge this is the first compelling observational evidence for such an effect , and it opens up the prospect of probing directly the properties of accretion disks during long thermonuclear bursts . with the disk reflection mechanism in mind
we also modeled the line as an fe fluorescence feature from an accretion disk ( model diskline in xspec ) .
since the absorption edge will also be broadened by motions in the disk , we also included a smeared edge in these fits ( model smedge in xspec ) .
we obtained excellent fits with this model for most of the burst duration except for about 100 s after the precursor . during this interval
the photosphere is extended and the black body temperature is evolving too quickly for the 16 s standard2 accumulations to resolve .
for this interval we would have greatly benefited from higher time resolution spectra .
we summarize our investigation of the time evolution of the burst spectrum in figures 7 , 8 and 9 .
figure 7 shows the time history of the burst from the standard1 data on a linear time axis .
the vertical dashed lines denote the region in which we derived black body fits .
the time evolution of the spectral parameters is shown in figures 8 and 9 . in each case
the timespan is that of the interval between the dashed vertical lines in figure 7 .
the derived black body flux , temperature , and inferred radius at 6.6 kpc are shown in figure 8 .
the evolution of the line and edge parameters is shown in figure 9 .
the evolution of the black body temperature and flux is very similar to that of other thermonuclear bursts , the only difference being that the timescale of the evolution is much longer . for comparison we also investigated the spectral evolution during the shorter ,
helium flash observed from 4u 1820 - 30 on may 2 , 1997 .
the peak black body temperature of both bursts approaches @xmath30 kev , and the peak fluxes are also consistent .
this leaves no doubt about the thermonuclear origin of the super burst .
the super burst from 4u 1820 - 30 provides a unique opportunity to try and use burst spectroscopy to constrain the radius of a neutron star .
the fact that the source distance is reasonably well constrained , the accreted matter is likely pure helium , and the burst lasted so long giving high signal to noise spectra
make such an effort well worthwhile .
our aim in this paper is to lay out the basic properties of the super burst and a likely mechanism for its production .
it is beyond the scope of this paper to present a detailed spectroscopic study in an attempt to derive constraints on the neutron star radius , however , we can make some simple estimates using our black body spectral fits and previous calculations of pure helium , neutron star atmospheres .
the spectrum of an x - ray burst is not expected to be strictly that of a black body .
many authors have investigated the effects of electron scattering on the spectrum of x - ray bursting atmospheres , which is the dominant opacity source under the conditions typical during bursts ( see for example , london , taam & howard 1986 ; ebisuzaki 1987 ; titarchuck 1994 ; madej 1991 ; madej 1997 ) . what is found is that the observed color temperature ( ie .
the temperature derived by fitting a black body function ) is higher ( harder ) than the actual effective temperature of the atmosphere by a factor which ranges from about 1.4 to 1.7 and is largely dependent on the local x - ray flux .
this correction factor is usually referred to as the hardening factor .
for example , ebisuzaki & nakamura ( 1988 ) provide analytic approximations for the hardening factors in x - ray bursting , pure helium atmospheres .
these factors depend only on the ratio of the local flux to the eddington flux . using these hardening factors
we can convert our measured color temperatures to effective temperatures and then infer the size of the neutron star using the formula ; @xmath31 where @xmath32 , @xmath33 , @xmath34 , and @xmath35 are the flux , effective temperature ( both measured far from the neutron star ) , stefan - boltzmann constant , and the source distance , respectively . here
@xmath36 and @xmath37 are the gravitational mass and radius of the star , both measured at the stellar surface . because of the general relativistic effects this expression does not directly give an estimate of the radius independent of the mass .
we used the flux and temperature measurements summarized in figure 8 along with the hardening factors from ebisuzaki & nakamura ( 1988 ) to evaluate the above expression ( assuming a distance of 6.6 kpc ) . for each pair of flux - temperature measurements we get an estimate of the quantity , @xmath38 the mean value of @xmath39 in these data is 18.3 km , which implies a mean radius range of 15.7 to 13.85 km for a mass range of 1.4 to 2 @xmath40 .
this range of radii is about a factor of two larger than that determined by haberl & titarchuk ( 1995 ) for 4u 1820 - 30 .
we will explore the implications of the super burst for the mass - radius relation in 4u 1820 - 30 in greater detail in a sequel .
the peak flux of @xmath41 ergs @xmath8 s@xmath2 during the super burst implies a peak isotropic luminosity of @xmath42 ergs s@xmath2 at a distance of 6.6 kpc .
a neutron star with a pure helium photosphere has an eddington luminosity of @xmath43 ergs s@xmath2 , which for reasonable neutron star masses and radii is consistent with our inferred peak luminosity . to determine the total energy in the observable burst we integrated the flux vs time profile , linearly interpolating across the gaps .
we find a total fluence of @xmath44 ergs @xmath8 . at a distance of 6.6 kpc
this implies a total energy release of @xmath45 ergs .
we note that these numbers should be considered lower limits since when our observations ended the burst flux was still decaying away .
moreover , it is likely that a substantial fraction of the energy released in the carbon burning layer is emitted as neutrinos or conducted inward and only released on a timescale much longer than the observed burst ( see our discussion in [ subsec : thermal - decay ] ) . nevertheless , the observed x - ray fluence is @xmath19 times greater than is released in a typical helium flash from 4u 1820 - 30 ( haberl et al .
the basic physics of nuclear burning instabilities on neutron stars indicates that carbon and oxygen burning can in principle provide this amount of energy , but not helium ( see for example lamb & lamb 1978 ) .
franco & strohmayer ( 1999 ) reported evidence for a @xmath29 kev emission line during the expansion phase of the may 2 , 1997 burst from 4u 1820 - 30 observed with rxte .
there have been several previous claims of absorption lines between 4 and 5 kev from several burst sources , but their reality and interpretation have been controversial ( see for example waki et al . 1984 ; nakamura , inoue , & tanaka 1988 ; magnier et al . 1989 ; foster , ross , & fabian 1987 ; madej 1989 ) , and we note that these previous line claims have not been confirmed by subsequent observations , for example , with the advanced satellite for cosmology and astrophysics ( asca ) .
our results leave no doubt that emission lines consistent with fe k@xmath46 fluorescence can be produced during x - ray bursts . what is the origin of these discrete spectral features ?
a number of authors have investigated line production mechanisms during x - ray bursts .
foster , ross , & fabian ( 1987 ) investigated the formation of fe features in the spectrum of cooling x - ray bursts .
their models predict broad k@xmath46 emission features should be present along with a ` blend ' of photoelectric edges near 9.1 kev .
we see both a broad line and an edge during the super burst , features similar to those predicted by foster , ross , & fabian ( 1987 ) , however , the observed energies of the features are somewhat problematic for this interpretation because the inferred redshifts are uncomfortably small for what would be expected from a canonical neutron star .
for example , when our observed black body temperature reaches a maximum , the inferred line centroid is about 6.2 kev ( see figures 8 and 9 ) .
if this resulted from a redshift of a 6.9 kev line of hydrogen - like fe , one would require a redshift @xmath47 which would imply a neutron star of radius 21.5 km for a mass of @xmath48 .
this is prohibitively large even for the stiffest estimated neutron star equations of state .
one might argue that the neutron star photosphere is still extended , but the temperature is thought to peak when touchdown of the photosphere has occurred .
if the rest energy of the line were 6.7 kev then the problem is only exacerbated .
it is possible , however , that because of the modest spectral resolution of these observations , the measured line centroid could be influenced by the choice of the continuum model , so these conclusions should still be considered cautiously . as mentioned previously , day & done ( 1991 )
suggested that a disk reflection component could be detected in the spectrum of x - ray bursts , and that detection of the absorption edge could be used to probe the ionization state of the disk .
we think this is the simplest interpretation of the discrete features we see in the burst spectra .
the inferred line centroids , which range from @xmath49 kev are comfortably consistent with an origin in the inner accretion disk
. relativistic motions in the disk also broaden the line profile .
furthermore , the inferred edge energies of @xmath50 kev are also consistent with this picture .
changes in the inferred energy of the absorption edge also suggest corresponding changes in the ionization state of the disk .
for example , the highest inferred edge energies correspond to the time when the black body flux and temperature were near their peaks , suggesting that the burst flux has a significant effect on the ionization state of the disk . the equivalent width of the @xmath29 kev line ranges from @xmath51 ev , but is typically about 130 ev .
this is reasonably consistent with recent estimates derived from spectral modelling of reflection from x - ray illuminated disks ( see for example , li , gu & kahn 2001 ; nayakshin & kallman 2001 ; ross , fabian & young 1999 ) . in principle , if the energies of the emission line and edge are known then one can constrain the ionization state of the gas .
turner et al .
( 1992 ) tabulate the run of fe k line and edge energies with ionization parameter in the context of fe k@xmath46 observations of seyfert galaxies ( see their figure 2 ) . based on this compilation ,
our line and edge energies appear consistent with an ionization parameter @xmath52 near the peak of the super burst .
the subsequent evolution of the edge and line energies through the burst appears consistent with a decreasing ionization parameter as the burst evolves .
this seems sensible , since the ionizing flux ( from the burst ) is decreasing with time .
however , more precise spectral modelling will be required to investigate the implications for disk structure in detail .
this will be left for a future study . during the decay of the burst
there is a highly significant hardening of the spectrum ( see figure 8 , near 3000 s ) .
this is evidenced by an increase in the black body temperature along with a decrease in the inferred black body radius .
even more interesting , the line flux and edge depth both decrease dramatically at the time of this spectral hardening ( see figure 9 ) .
this behavior is somewhat reminiscent of state changes in other accreting x - ray binaries , but happens on a much shorter timescale .
perhaps the inner disk is transitioning to a hot , optically thin condition , similar to an adaf ( see for example narayan , yi , & esin 1998 ) .
alternatively , we could be seeing a geometric effect of something in the system shadowing the disk so that the reflected component is greatly reduced , perhaps a warp in the accretion disk .
unfortunately , an earth occultation interval blocks our view of what happens next . also of note
is the fact that shortly after this hardening episode chaotic dipping can be seen in the lightcurve for an extended period .
although for part of this time the source is earth occulted , this dipping may be another indication of the accretion flow having been strongly affected by the burst .
another possibility for the discrete components is that we are seeing fluorescence in the wind material blown off the star by the super - eddington burst . in this case
one might expect to see a correlation of the line strength with the amount of absorbing material along the line of sight , and this does not seem to be the case .
it seems likely that at some level all of these processes are at work , but it seems most plausible that disk reflection is the dominant mechanism .
the uv modulations observed by anderson et al .
( 1997 ) suggest the system has an inclination @xmath53 ( see arons & king 1993 ) , which would provide a favorable geometry for reflection .
the energy resolution of the pca is not sufficient to resolve the line and edge in detail .
for example , a redshifted line from the neutron star photosphere could be obscured by the broad line produced by disk reflection .
however , additional observations of such a burst with much higher spectral resolution could help distinguish amongst these different scenarios , and could provide us with a powerful new probe of the accretion process in lmxbs .
having described the observation of the super burst and the evidence for it being thermonuclear in origin , we now examine its source of fuel and its ignition . in order for he burning to power the super burst ,
the accumulated he must not ignite until the base of the helium layer reaches densities of order @xmath54 ( e.g. , * ? ? ?
* ; * ? ? ?
this can only occur if the atmosphere is extremely cold and in particular requires a local accretion rate @xmath55 @xcite in the absence of any heat flux from the core .
such low accretion rates are not consistent with the observed flux and the presence of concurrent he flashes .
the fact that a helium flash went off just prior to the super burst also indicates that helium can not be the primary energy source for the giant burst .
we therefore discount the possibility that @xmath56 could be the fuel for the super burst .
as shown in the previous section , the spectral evolution of the burst is roughly consistent with the release of energy at large densities in the star .
the burning of carbon to iron - peak elements releases @xmath57 so that at least @xmath58 of carbon is needed to account for the energy released in the burst ( in [ subsec : thermal - decay ] we argue that this is likely an underestimate of the total burst energetics ) .
an @xmath59 mass of @xmath60 uniformly deposited over the surface of a neutron star of radius @xmath61 has a column density @xmath62 .
hydrostatic balance gives the pressure at the base of the layer , @xmath63 ; from the equation of state and the thermal conductivity ( see [ subsec : thermal - decay ] for details ) we may estimate the cooling timescale for a mass of newly formed @xmath64 to be @xmath65 , roughly consistent with the observed thermal decay timescale of the burst . here
@xmath66 is the specific heat , per unit mass , @xmath67 is the thermal conductivity , @xmath68 is the density , and @xmath69 is the column depth .
if mass transfer is driven by gravitational radiation we expect an accretion rate of @xmath70 onto the neutron star ; this level is also roughly consistent with the accretion rate required to account for the persistent luminosity of @xmath71 .
the precise rate is uncertain because of our lack of knowledge concerning the efficiency with which gravitational energy is converted to x - ray flux .
if this conversion were 100% efficient , then one would require @xmath72 ( assuming a 1.4 @xmath40 and @xmath73 neutron star ) ; a more realistic number , however , is probably 520 % , so that the actual mass accretion rate could be higher than @xmath74 .
isotropic ( over the stellar surface ) accretion at a rate of @xmath75 g s@xmath2 can deposit a sufficient amount of carbon to power the burst only after 0.20.6@xmath76@xmath77 , so such events should indeed be rare , consistent with only one such burst being observed from this source .
unstable carbon burning on accreting neutron stars was conjectured by @xcite and studied by @xcite , who found that for accretion rates of @xmath78@xmath79 , the timescale for recurrence would be of order 10100@xmath76@xmath77 , with the bursts having an energy of order @xmath80 , about 100 times greater than that measured for this burst .
while the original calculation of @xcite assumed accretion of hydrogen - rich material , recent calculations with large reaction networks for both unstable @xcite and steady - state burning @xcite find that only a small amount of @xmath60 is produced during hydrogen burning .
for this reason , the scenario presented here ( i.e. , the ignition of a carbon - rich layer ) is probably not relevant for the super bursts seen from 4u 173544 , ks 1731 - 260 , gx 3 + 1 , ser x-1 and 4u 1636 - 53 , which are likely accreting a mix of hydrogen and helium ( see @xcite ; heise , in t zand , & kuulkers 2000 ; @xcite ; wijnands 2001 ) .
the small amount of @xmath60 produced during rp - process burning might be a fuel for the other super - bursts ; for a discussion of this possibility , we refer the interested reader to @xcite .
while the burning of @xmath60 is a plausible cause of this super burst , the ignition of a pure carbon layer requires either high temperatures @xcite or a large mass of carbon @xcite . in this section ,
we address these constraints .
we begin ( [ subsec : c12-production ] ) by estimating the amount of carbon produced .
we then ( [ subsec : ignition - carbon - layer ] ) investigate the conditions for ignition of a carbon - rich layer .
we demonstrate , in the absence of a `` trigger , '' that a much larger mass of @xmath60 is required than that estimated from the observed burst energetics .
section [ subsec : thermal - decay ] provides a resolution to this dichotomy : namely , for very large bursts most of the heat is carried off by neutrinos , and of the heat conducted from the burning layer , a substantial amount flows _ inward _ , to be released on much longer timescales ( days to weeks ) .
the long timescale for thermal diffusion at great depths means that only a fraction of the total burst energy is radiated away in the few hours following the burst rise .
a consequence of requiring a massive carbon layer for ignition is that the recurrence times of such super bursts will be many years . whether or not @xmath60 is produced by the burning of @xmath56 depends largely on whether the burning is stable .
the ashes of stable he burning are composed mostly of @xmath60 at the lowest accretion rates compatible with stability @xcite . in contrast , unstable burning leads to much higher temperatures , which allow the reactions to run up to iron - peak nuclei .
numerical calculations find that for unstable burning only a trace amount of residual he is expected ( @xmath81 by mass ; @xcite ) and no @xmath60 . because this source accretes nearly pure he , the endpoint of the unstable nucleosynthesis will be iron - peak nuclei , rather than the heavier species produced during mixed h / he burning .
the luminosity of this source is variable by a factor of @xmath82 ( figure [ fig1 ] ) , and type i x - ray bursts are only observed when the luminosity is in its low state .
the lack of unstable burning when the luminosity is high is somewhat puzzling , since the inferred accretion rate is still less than what is necessary for stability , according to linear analysis and time - dependent calculations ( see * ? ? ?
* and references therein ) .
one possibility is that the burning is unstable , but that the flame front propagates at the slow conductive velocity @xcite .
whether such fronts can slowly propagate on a rapidly rotating neutron star is an open question ( for a discussion , see * ? ? ?
another possibility is that the accretion flow does not spread over the entire surface prior to he ignition , as suggested by @xcite as an explanation for patterns of unstable hydrogen / helium burning on z - sources .
if the efficiency of conversion of gravitational energy to x - ray luminosity is modest , then the inferred accretion rate can be substantially higher , reducing the discrepancy between theory and observation .
one constraint on the amount of carbon produced would be the recurrence time of a super burst .
we did search the rxte / asm archival lightcurve of 4u 1820 - 30 for additional instances of super bursts , but did not find any .
although the asm monitors the whole sky it does not do so all the time , so that the total time during which another super burst from 4u 1820 - 30 could have been seen is only about 25 days out of a time span of about 5 years .
since only one such event was seen and the total on source time was modest we can not place a strong constraint on the recurrence rate .
it is unlikely to be much larger than 0.05 day@xmath2 , but could easily be ten times lower than this .
we note that the source is only at or below the asm flux level at which thermonuclear bursts were observed for about @xmath83 of the time . for the purposes of this paper , we take the absence of observed type i x - ray bursts during the episodes of high - luminosity as indication that @xmath60 is produced . over each 178 day cycle
, we estimate that roughly @xmath84 of @xmath60 is formed . as we describe in the next section ,
it is quite plausible that the iron made in the bursts mixes with the carbon made during stable burning , so that the deep ocean is a mixture of the two .
the physics involved in the ignition of the carbon layer is detailed in @xcite ; here we review that discussion .
the scale height is of order @xmath85 , so that the carbon layer is susceptible to a thin - shell instability @xcite , and we may neglect the variation of gravity with depth .
we compute the thermal structure of the carbon layer following the method outlined in @xcite .
the relevant equations are those of continuity ( for each species ) , @xmath86 entropy , @xmath87 and flux , @xmath88 here @xmath89 and @xmath90 are the mass fraction and mass number of species @xmath91 , @xmath92 is avogadro s number , and @xmath93 and @xmath94 are the volumetric destruction and production rates of species @xmath91 via nuclear reactions . for the electron equation of state
, we use a table interpolation scheme of the helmholtz free energy @xcite .
the ionic equation of state includes coulomb interactions , which are parameterized by @xmath95 , where @xmath96 is the mean volume per ion . in this problem
, @xmath97 spans a range of values @xmath98 ( ions are weakly coupled ) to @xmath99 ( ions are crystalline ; see @xcite and references therein ) .
for @xmath100 the ions are in a liquid state , and we shall refer to this layer as the `` ocean '' .
the thermal conductivity @xmath67 contains contributions from both radiative transport ( in the diffusive approximation ) , and electronic thermal conduction .
radiative transport dominates at lower densities ; in this regime we use the rosseland mean of both thompson scattering ( with corrections for the relativistic , partially degenerate electrons ; @xcite ) and free - free absorption with an analytical approximation to the elwert factor @xcite , which reproduces the tabulated gaunt factor of @xcite . at higher densities
electron conduction dominates ; we include contributions from electron - electron scattering , for which we use the fit @xcite to the calculations of @xcite , and electron - ion scattering , for which we use the fits of @xcite .
the treatment of @xcite is valid in both the liquid and crystalline phases .
we are interested in the unstable ignition of @xmath60 .
prior to ignition , the heating from @xmath101 is negligible , and we do not include it in the thermal structure calculation . in the carbon ocean and iron crust , then , @xmath102 , the neutrino emissivity , for which we use the fits of @xcite to the pair , plasma , photoneutrino , bremsstrahlung , and recombination rates .
we change the composition discontinuously from a mixture of @xmath60 and @xmath64 ( we allow for the possibility of a mixed layer ) to @xmath64 at an interface column @xmath103 , which is chosen to be roughly where we expect the @xmath60 to have ignited . the thermal timescale , at a given column , is much faster than the timescale for accretion to supply that column of material . as a result , we may simplify our equations by presuming steady - state ( @xmath104 ) and converting the structure equations ( eq .
[ [ e : cont]][[e : fluxa ] ] ) from pde s into ode s . to further simplify calculations , we rewrite the equations with the column @xmath105 as the independent variable , @xmath106 , and neglect differential sedimentation of the ions . with these approximations ,
the velocity of the fluid is @xmath107 , where @xmath108 is the accretion rate per unit area , and the structure equations reduce to @xmath109 here @xmath110 is the adiabatic gradient and @xmath111 , where @xmath112 is the energy released per formed carbon .
equations ( [ eq : he])([eq : f ] ) depend on both the surface gravity @xmath113 ( through @xmath114 ) and the local accretion rate @xmath108 .
we fix @xmath115 to the value @xmath116 appropriate for a neutron star of mass @xmath117 and radius @xmath73 .
because the accretion rate varies , we first fix the temperature of the deep crust by solving equations ( [ eq : he])([eq : f ] ) at the mean accretion rate @xmath118 , which roughly corresponds to the persistent mean accretion rate of the source if uniformly deposited over the surface . to fix the composition of the atmosphere and the bottom of the he - rich layer , we compute the burning of he to c via the 3@xmath46 reaction . at low accretion rates ,
the @xmath119 reaction rate is negligible compared to the 3@xmath46 rate @xcite , and we therefore just track the abundances of @xmath56 and @xmath60 . for @xmath120 , the heating rate from the @xmath121 reaction , we use the analytical fit of @xcite .
boundary conditions are specified at both ends of the computational domain .
the thermal structure of the atmosphere is insensitive to the photosphere temperature ( it is , roughly speaking , a radiative - zero solution ) for @xmath122 , @xmath123 being the opacity ; for definiteness , we set @xmath124 at @xmath125 , where @xmath126 is the stefan - boltzmann constant .
the choice of boundary condition at the other end of the computational domain ( set to @xmath127 ) is more problematic .
the high thermal conductivity of the degenerate , relativistic electrons ensures that the ocean is well - coupled to the thermal state of the inner crust and core .
previous studies ( e.g. , * ? ? ?
* ; * ? ? ?
* ) simply took the flux to be zero at the inner boundary . for an accreting neutron star ,
continual compression of the deep crust forces a series of electron captures , neutron emissions , and pycnonuclear reactions @xcite ; in steady - state , most of the heat released from these reactions flows into the core but a small amount ( @xmath128 per accreted nucleon for @xmath129 ) flows outward @xcite .
the thermal timescale in the inner crust is of order years @xcite and so the temperature there depends mostly on the time - averaged accretion rate .
we therefore adjust the temperature at @xmath130 so that the flux from the deep crust is @xmath131 . with these approximations ,
we numerically integrate the equations ( [ e : cont])([e : fluxa ] ) using a stiff ode solver .
we integrate from each boundary and iteratively adjust the flux at the upper and lower boundaries so that the temperature and flux are continuous at a fitting point .
the result of this integration , over the range of @xmath105 relevant for ignition , is shown in figure [ fig : structure ] .
we show the temperature ( _ solid line _ ; left - side vertical axis ) and flux in units of mev per accreted nucleon ( _ dotted line _ ; right - side vertical axis ) . the arrow is to guide the eye and mark the value of @xmath132
the top axis shows the time required to accrete a given column at @xmath133 .
as we shall explain momentarily , it is plausible that the carbon and iron form a mixed layer ; we therefore change the mixture from @xmath134 to @xmath135 ( @xmath136 ) at @xmath137 .
the kink in the temperature at that point is because the conductivity decreases when we reset the composition .
we chose @xmath136 as a minimal case for ignition .
a larger @xmath138 would increase the ocean thermal conductivity and , for a fixed flux of @xmath139 , lower the ocean temperature .
this increases both the recurrence time and the total burst energetics ( although , as we shall explain in the next section , a larger @xmath138 will not increase the x - ray fluence substantially because the additional energy is lost to neutrinos ) .
a smaller @xmath138 may not ignite unstably @xcite . with a judicious use of hindsight ,
we set the base of the c / fe layer at @xmath140 ( roughly where the @xmath60 will ignite ) ; at greater column the composition is pure fe .
the point where the pure fe layer begins is marked in figure [ fig : structure ] by the large black dot . as explained in @xcite ,
thermal conduction is more efficient at cooling the ocean than neutrino emission at these temperatures @xcite .
the condition for ignition is crudely expressed as @xmath141 , where @xmath142 is an approximation to @xmath143 .
for the heating @xmath144 , we use the rate of @xcite and incorporate strong screening using the formalism of @xcite . in figure [ fig : structure ] , we show the points where ignition occurs ( _ heavy dashed curve _ ) . note that the strong temperature sensitivity of the @xmath101 reaction means that the location of this curve is relatively insensitive to the precise method of defining ignition .
as is evident from the figure , the temperature in the carbon - rich layer is too cold , by about a factor of 23 , for ignition to occur if the base of the carbon - rich layer is at @xmath145 .
the presence of heavy elements ( e.g. , @xmath146 ) in the ocean of the other super burst sources makes the carbon - containing layer hotter , for a given flux from the crust and core , and allows for a smaller ignition column @xcite . in this calculation
, the mass of carbon at the depth where ignition occurs is @xmath147 .
this is about a factor of 20 greater than that indicated by our estimate of the burst energetics . for a lesser @xmath138 ,
the conductivity is lower and the temperature at which ignition occurs increases ; as pointed out by @xcite , however , the carbon burning will become stable at low accretion rates and small carbon abundances . their estimate implies that for the conditions we assume here , @xmath148 is required for unstable ignition . consistent with this estimate , we find that for this calculation the lifetime , @xmath149 , is slightly less than the accretion timescale @xmath150 at the ignition point .
for this reason we do not consider smaller mass fractions of carbon .
the temperature at the base of the carbon layer is mostly determined by the flux flowing from deeper in the star .
the reason is that the heat flowing from the crust ( @xmath151 per accreted nucleon ) is larger than that generated by compressional heating ( @xmath152 per accreted nucleon ) , and the high conductivity of the degenerate plasma enforces a very shallow temperature gradient . at the time of the super burst , the accretion rate was lower than the mean , and the deep ocean was slightly colder than in this calculation , making ignition even more difficult .
as mentioned earlier , locally high accretion rates onto a small portion of the stellar surface may be an explanation for the lack of bursts during the high - luminosity state of this source . at higher local accretion rates ,
the ignition column decreases . to heat a confined carbon column to ignition at @xmath153 requires local accretion rates in excess of 5 times the local eddington rate @xcite .
it is implausible , however , that the accreted matter would fail to spread over the surface prior to reaching this column .
while strong magnetic fields ( @xmath154 ) could possibly confine such a large column @xcite , there is no evidence , such as pulsations in the persistent emission , for such strong organized fields in this source . what about heating of the carbon layer during a type i x - ray burst ?
the timescale to conduct the heat from the base of the he - rich layer to the surface is @xmath155 ( the decay timescale of the burst ) . unlike in the degenerate ocean
, the dominant opacity is thompson scattering , for which the equivalent radiative conductivity is independent of depth . as a result ,
the inward propagation of heat is inefficient ( as can be checked by a calculation similar to that described in equation ( [ eq : diffusion ] ) and the heating of the carbon layer is too small to force an ignition .
the fact that this source cyclically produces @xmath60 ( in the high - luminosity ) state and iron - peak elements ( through unsteady he burning in the low - luminosity state ) has some interesting consequences . during a he burst
, the burning produces mostly @xmath64 .
as the iron cools , it becomes denser than the underlying carbon , because the iron has fewer electrons per nucleon than helium .
the base of the unstable he burning layer is therefore susceptible to first a secular salt - finger and then a rayleigh - taylor instability .
for example , at an interface pressure of @xmath156 and a @xmath60 temperature of @xmath157 , the iron layer is less dense than the carbon layer only when its temperature is @xmath158 .
terrestrial rayleigh - taylor experiments and numerical simulations ( * ? ? ?
* and references therein ) suggest that the time @xmath159 for the interface ( between fluids of density @xmath68 and @xmath160 ) to traverse a distance @xmath161 is given by @xmath162^{1/2}$ ] , where @xmath163 is the atwood number and @xmath164 is uncertain by a factor of 2 . taking @xmath161 to be a scale height , we find that for this case @xmath165 .
thermal diffusion is ineffective on this timescale ; the characteristic distance over which heat can diffuse on this timescale is less than @xmath166 , which is much smaller than the pressure scale height ( @xmath167 ) .
the scaling of distance with @xmath168 comes about because of mergers between falling spikes into larger and larger drops ; in the case of a stratified medium , this effect is counteracted to some extent by compression of the falling spikes . as the iron `` plume '' descends ,
the ambient medium ( carbon ) is entrained , reducing the buoyancy contrast and eventually halting its descent @xcite .
because the mass of fe available after a he burst is much less than the mass of the carbon - rich layer , it is unlikely that the sinking fe will fall to the base of the carbon - rich layer . whether a descending plume would be able to spark an ignition of the carbon layer at @xmath169 , and whether a flame could propagate without quenching at those lower densities , is unknown . in any case , it is likely that at least some of the iron is mixed into the carbon layer and advected deeper into the star by continual accretion , so that ignition will occur in a mixed c / fe layer , as we have assumed .
detailed calculations are outside the scope of this paper ; for now we leave this as an interesting possibility .
as noted above , the ignition of @xmath60 , for conditions relevant to this source , requires either a large accumulated column of carbon - rich material , or a trigger , perhaps associated with an earlier he flash . in this subsection
, we explore the evolution of an unstable ignition at @xmath140 of a c / fe layer ( @xmath136 ) . unlike for normal ( 1020 s duration ) type i x - ray bursts , the energetics of this super burst are much greater than that inferred from the x - ray lightcurve .
there are two reasons for this .
the first is that neutrino cooling is more efficient than thermal conduction at the peak temperature ( @xmath170 ) reached in the burst .
the second reason is that thermal conduction into deeper layers is efficient and competes with the outward thermal transport . because of the great depth of ignition , the thermal time is of order @xmath171 and is much longer than the duration of the observation .
we now consider each of these cooling mechanisms in turn .
once the instability begins , the heating timescale @xmath172 is , by definition , shorter than both the thermal conduction and neutrino cooling timescales . at first the nuclear burning timescale is comparable to the thermal conduction time ( @xmath173 ; @xcite ) , but it accelerates with increasing temperature .
we calculate the peak temperature from the equation @xmath174 , where @xmath175 is the energy per gram released in burning @xmath60 to fe - peak elements . for a pure carbon layer
, the temperature would reach @xmath176 if all the carbon burned to iron .
note that even at this extreme temperature , the electrons are still ( mildly ) degenerate ; the fermi energy is @xmath177 .
in addition , the radiation pressure , at this temperature , is only @xmath178 , about a tenth of the total . unlike in he burning ,
the peak temperature is therefore not limited by the expansion of the burning layer if the base of that layer is sufficiently deep .
because of the high density , photodisintegration reactions are also unimportant until temperatures in excess of @xmath179 are reached , as can be checked by a saha - type equation . for the abundance ( @xmath136 ) used in figure [ fig : structure ] , @xmath180 . although neutrino cooling is unimportant for stabilizing the ignition of carbon on a neutron star @xcite , it is more efficient than thermal conduction when the burning layer is at its peak temperature .
in addition , thermal conduction to the surface is more efficient at lower column and competes more effectively with neutrino cooling . for a very deep layer
, however , the situation is different .
the thermal conduction timescale is longer and the peak temperature reached is much higher . to investigate the cooling
, we constructed a one - zone approximation of the layer following the burning , @xmath181 we set the initial temperature at @xmath182 but switched the composition to fe to simulate the ashes . at a temperature @xmath183 ,
the dominant neutrino - producing process is from electron - positron pairs . as a result ,
the neutrino emissivity _ decreases _ as density increases because of the rising electron fermi energy ( see fig .
[ fig : enu ] ) , and it is likely that a strong temperature gradient develops , so this one - zone calculation underestimates somewhat the neutrino flux .
our calculation is also limited in that it completely neglects convection . in figure
[ fig : compare - fluence ] , we show the neutrino ( _ solid line _ ) and conductive ( _ dotted line _ ) fluence , as a function of time , computed with equation ( [ eq : one - zone - neutrino ] ) for a fraction of carbon @xmath136 . for @xmath184 , the total conductive fluence of the burst only increases slightly ; any further heating is carried off by neutrinos . for this calculation ,
the total radiated energy is @xmath185 for a neutron star of @xmath73 radius about a factor of 10 greater than that observed , although for this calculation the total energy conducted to the surface after @xmath186 is only @xmath187 , consistent with that measured .
the evolution of the luminosity is shown in figure [ fig : luminosity ] .
for such a deep layer of carbon , the thermal evolution time is quite long ; the burst flux still exceeds that from accretion for several _ days _ , in agreement with calculations by @xcite for super bursts from hydrogen accreting sources .
while a one - zone approximation is tolerable at lower columns , it becomes suspect where the electrons are degenerate , since the conductivity increases with density .
the electron - thermal conductivity is @xmath188 where @xmath189 is the electron - ion relaxation time . here
@xmath190 and @xmath191 are the electron momentum and velocity evaluated at the fermi energy @xmath192 , @xmath193 and @xmath194 are the densities of electrons and nuclei , respectively , @xmath195 is the relativistic electron mass and @xmath196 is the coulomb logarithm . where relativistic , degenerate electrons dominate the pressure , the conductivity scales as @xmath197 ( neglecting variation in @xmath196 ) . because the thermal conductivity increases with depth ( pressure ) , heat deposited from unstable carbon burning can readily flow inwards . to illustrate the inward flow of heat
, we construct the simplest thermal transport problem by neglecting the variation of @xmath196 , taking the pressure to be solely that from the degenerate electrons , and writing the specific heat as @xmath198 ( since the electrons are more numerous than the ions , this is an adequate approximation for most of the cooling ) . with these simplifications ,
the thermal diffusion equation in non - dimensional form is @xmath199 over the domain @xmath200 .
here @xmath201 , @xmath202 , and the time is in units of @xmath203 .
the flux in these units is @xmath204 .
the background steady - state solution is one of constant flux , @xmath205^{1/2 } , \label{eq : uinit}\ ] ] where @xmath206 is the rise in temperature over the domain . to solve equation ( [ eq : diffusion ] ) numerically ,
we use the method of lines (;
@xcite ) , with a fifth - order polynomial spatial decomposition over logarithmically spaced collocation points and implicit integration in @xmath159 . for simplicity ,
the temperatures at both ends of the computational domain are fixed . at time @xmath207 the temperature profile is @xmath208 plus a smoothed top - hat profile representing the heating injected during unstable burning . in order to investigate the transport of heat following a perturbation
, we placed the boundaries at @xmath209 and @xmath210 and chose the top - hat perturbation to span @xmath211 , with an initial temperature of @xmath212 .
figure [ fig : fluence ] shows the fluence , defined here as @xmath213 , @xmath214 being the steady - state flux , as a function of time .
the solid curve denotes the fluence at the top of the layer ; the dotted curve denotes that at the bottom .
after @xmath215 , only about 0.16 of the total fluence has been radiated from the surface .
the inward - directed flux will raise the temperature in the deep ocean and crust slightly and will therefore increase the flux from the crust above the @xmath216 long - term value .
while all of the heat deposited in the ocean will eventually be radiated from the surface , only the fraction that is radiated immediately following the burst is discernible in the x - ray lightcurve . as in figure
[ fig : compare - fluence ] , the evolution timescale is roughly the thermal timescale at the base of the heated layer .
this simple calculation is only meant to be illustrative : it ignores neutrino cooling and in particular it underestimates the thermal transport to the surface .
a convective layer develops during the burst @xcite ; this convection may be enhanced by the stronger neutrino emissivity at lower densities ( fig .
[ fig : enu ] ) . to reproduce correctly the burst evolution and
the lightcurve requires an implicit evolution code , which is beyond the scope of this paper ; for now we note that the total fluence of the super burst is systematically underestimated , by perhaps as much as an order of magnitude . for both cooling calculations ( eqs .
[ [ eq : one - zone - neutrino ] ] and [ [ eq : diffusion ] ] ) , only a small fraction of the burst energy is radiated away after @xmath217 .
_ as a result , the fluence observed during the three hours following the burst is consistent with the burning of a large mass of carbon at great depth . _ the strongest constraint on this scenario is the recurrence time : to accrete a column of @xmath210 takes @xmath218 .
thermonuclear bursts provide a unique probe of the physics of neutron stars .
their properties depend on the accretion rate and composition of the accreted matter , the bulk properties of neutron stars , the thermal state of the neutron star and the detailed nuclear burning physics .
4u 1820 - 30 is an especially revealing system since its extreme properties place strong constraints on the composition of the accreted matter .
the observation of a super burst from 4u 1820 - 30 with @xmath219 times the energy release of a typical helium burst provides us with a compelling look at processes occurring at much greater depth in the neutron star `` ocean '' than we are normally privileged to witness .
we have investigated the ignition of a @xmath60-rich layer , presumably produced during the high - luminosity state when no type i x - ray bursts are observed . when the mass of this layer is @xmath220 , a thermal runaway ensues .
unlike the case discussed by @xcite , the lack of heavy rp - process ashes makes the carbon layer cooler for this source and a larger mass of carbon is required for ignition .
once the runaway begins , the temperature rises to @xmath221 .
the burning ends and the layer rapidly cools by neutrino emission until the temperature is @xmath222 , at which point thermal conduction becomes more efficient at cooling the layer . because of the great depth of the ignition , heat readily flows inward , and the temperature evolves on a timescale @xmath223 . as a result ,
the measured x - ray fluence is only a small fraction of the total , so that this scenario is consistent with the observation .
the recurrence time would provide a strong constraint on the ignition column . to accrete
an ignition column of @xmath210 takes @xmath224 . if the accretion rate were much higher than that inferred by the x - ray luminosity , than the ocean would be hotter , and the ignition column reduced .
recurrence times of less than a year would require that @xmath225 , however , or that the ignition be prematurely triggered .
as noted by @xcite , the heating from this burst can quench type i x - ray bursts until the cooling luminosity is @xmath226 @xcite .
if the scenario we outline is correct , then no type i x - ray bursts should have occurred in the week following the super burst .
distinguishing the elevated flux following the burst will be more difficult , given the variability of the accretion flux
. the strongest constraint on this scenario would be if another super burst were detected after a timescale much less than a decade .
this would suggest that the carbon ignition is triggered .
the fact that a he burst occurred immediately prior to the super burst is intriguing ; if the carbon ignition did occur at large densities , then the thermal instability began several hours prior to the observed rise of the burst .
it is possible that the flux from the carbon runaway could ignite any accumulated he .
while the calculations presented here are suggestive , they are very crude and could be substantially improved .
evolutionary calculations , with reaction networks , of the ignition and subsequent evolution of this burst , coupled with similar calculations for the super bursts seen from hydrogen accreting sources , can inform us about the relevant physics at work in the deep ocean . continued x - ray monitoring of the source should eventually give a better constraint on the recurrence time .
an important quantity is the amount of carbon produced ; for the scenario we outline here to work , roughly half of the accreted matter must burn to carbon so that the ignition is indeed unstable .
the x - ray spectrum during the burst reveals a broad emission line indicative of fe k@xmath46 fluorescence as well as an absorption edge in the 79 kev range .
the very high signal - to - noise spectra obtained during the burst make the detection of these discrete features extremely secure .
previous authors have predicted that such features might be produced by disk reflection ( see for example day & done 1991 ) , but to our knowledge this is the first confident detection
. detailed modeling of these features could provide a new probe of the accretion disk in bursting systems .
here we have only scratched the surface in terms of investigating the discrete components . an in - depth study and interpretation of the line and edge
will be presented in a future publication .
it is a pleasure to thank lars bildsten , alan calder , andrew cumming , erik kuulkers , yuri levin , craig markwardt , richard mushotzky , bob rosner , jean swank , frank timmes and jim truran for many helpful discussions .
we also thank frank timmes for providing the interpolation routines and tables for the equation of state used in the computations in this paper .
efb acknowledges support from an enrico fermi fellowship .
this work is partially supported by the department of energy under grant b341495 to the center for astrophysical thermonuclear flashes at the university of chicago .
80 natexlab#1#1 , s. f. , margon , b. , deutsch , e. w. , downes , r. a. , & allen , r. g. 1997 , , 482 , l69 arons , j. & king , i. r. 1993 , , 413 , l121 , l. 1995 , , 438 , 852 , l. 1998 , in the many faces of neutron stars , ed .
a. alpar , r. buccheri , & j. van paradijs , vol .
515 , nato asi ser .
c ( dordrecht : kluwer ) , 419 bildsten , l. 2000 , in cosmic explosions , ed . s. s. holt & w. w. zhang , g. s. & chechetkin , v. m. 1979 , soviet phys .- uspekhi , 127 , 263 , o. , blandford , r. , madau , p. , & koonin , s. 1990 , , 363 , 612 , e. f. 2000 , , 531 , 988 , e. f. & bildsten , l. 1998 , , 496 , 915 , e. f. , bildsten , l. , & rutledge , r. e. 1998 , , 504 , l95 , j. r. & yueh , w. r. 1976 , , 210 , 440 caughlan , g. r. & fowler , w. a. 1988 , at .
data nucl .
data tables , 40 , 283 , g. w. , li , f. k. , canizares , c. , hayakawa , s. , jernigan , g. , & lewin , w. h. g. 1977 , , 179 , 651 cornelisse , r. , heise , j. , kuulkers , e. , verbunt , f. , & int zand , j. j. m. 2000 , , 357 , l21 cumming , a. & bildsten , l. 2001 , , submitted , ( astro - ph/0107213 ) , c. s. r. & done , c. 1991 , , 253 , 35p ebisuzaki , t. 1987 , pasj , 39 , 287 ebisuzaki , t. & nakamura , n. 1988 , apj , 328 , 251 farouki , r. & hamaguchi , s. 1993 , , 47 , 4330 , a. j. , fabian , a. c. , & ross , r. r. 1987 , , 228 , 259 , l. m. & strohmayer , t. e. 1999 , in american astronomical society meeting , vol .
195 , 12609 , b. a. & woosley , s. e. 1982 , , 258 , 733 fushiki , i. & lamb , d. q. 1987 , , 317 , 368 , j. & gursky , h. 1976 , , 205 , l131 , f. , stella , l. , white , n. e. , gottwald , m. , & priedhorsky , w. c. 1987 , , 314 , 266 haberl , f. & titarchuk , l. 1995 , a&a , 299 , 414 , p. & zdunik , j. l. 1990 , , 227 , 431 hansen , c. j. & van horn , h. m. 1975 , , 195 , 735 , j. , int zand , j. j. m. , & kuulkers , e. 2000 , aas / high energy astrophysics division , 32 , 2803 + , j. e. & shawl , s. j. 1985 , , 97 , 465 , j. a. , lewin , w. h. g. , doty , j. , jernigan , j. g. , haney , m. , & richardson , j. a. 1978 , , 221 , l57 , n. a. & sunyaev , r. a. 1999 , astronomy letters , 25 , 269 itoh , n. , hayashi , h. , nishikawa , a. , & kohyama , y. 1996 , , 102 , 411 , n. , kuwashima , f. , ichihashi , k. , & mutoh , h. 1991 , , 382 , 636 , p. c. & li , f. k. 1980 , , 238 , 287 , p. , ford , e. c. , & chen , k. 1997 , , 480 , l27 , e. 2001 , ast .
, 68 , 1 , w. h. g. , vacca , w. d. , & basinska , e. m. 1984 , , 277 , l57 li , y. , gu , m. f. & kahn , s. m. 2001 , apj , submitted , ( astro - ph/0106163 ) litwin , c. , brown , e. f. , & rosner , r. 2001 , , 533 , 788 london , r. a. , taam , r. e. , & howard , w. m. 1986 , apj , 306 , 170 madej , j. 1997 , a&a , 320 , 177 madej , j. 1991 , apj , 376 , 161 , j. 1989 , , 339 , 386 madsen , n. k. , , & sincovec , r. f. 1979 , acm trans . math . soft . , 5 , 326 , e. , lewin , w. h. g. , van paradijs , j. , tan , j. , penninx , w. , & damen , e. 1989 , , 237 , 729 , n. , inoue , h. , & tanaka , y. 1988 , , 40 , 209 nayakshin , s. , kallman , t. r. 2001 , apj , 546 , 406 ogata , s. , ichimaru , s. , & van horn , h. m. 1993 , , 417 , 265 popham , r. & sunyaev , r. 2001 , , 547 , 355 , a. y. , baiko , d. a. , haensel , p. , & yakovlev , d. g. 1999 , , 346 , 345 , a. y. , chabrier , g. , & yakovlev , d. g. 1997 , , 323 , 415 , w. & terrell , j. 1984 , , 284 , l17 , s. , ma , c. p. , joss , p. c. , & nelson , l. a. 1987 , , 322 , 842 , r. m. , minniti , d. , & liebert , j. 1993 , , 406 , 489 ross , r. r. , fabian , a. c. & young , a. j. 1999 , mnras , 306 , 461 , d. h. 1959 , , 129 , 734 , h. , aprahamian , a. , barnard , v. , bildsten , l. , cumming , a. , ouellette , m. , rauscher , t. , thielemann , f .- k . , &
wiescher , m. 2001 , , in press , h. , bildsten , l. , cumming , a. , & wiescher , m. 1999 , , 524 , 1014 , a. p. , zhang , w. , & white , n. e. 1997 , , 483 , l119 , a. , levin , y. , & ushomirsky , g. 2001 , , submitted .
preprint available : astro - ph/0108074 , l. , white , n. e. , & priedhorsky , w. 1987 , , 312 , l17 , t. e. 2000 , aas / high energy astrophysics division , 32 , 2410 , r. e. & picklum , r. e. 1978 , , 224 , 210 , y. , kii , t. , hayakawa , s. , kunieda , h. , masai , k. , nagase , f. , inoue , h. , koyama , k. , makino , f. , makishima , k. , matsuoka , m. , murakami , t. , oda , m. , ogawara , y. , ohashi , t. , shibazaki , n. , tanaka , y. , miyamoto , s. , tsunemi , h. , yamashita , k. , & kondo , i. 1984 , , 276 , l41 , f. x. & swesty , f. d. 2000 , , 126 , 501 , f. x. & woosley , s. e. 1992 , , 396 , 649 titarchuk , l. 1994 , apj , 429 , 340 , a. a. 1966 , j. fluid mech . , 1 , 29 turner , t. j. , done , c. , mushotzky , r. , madejski , g. & kunieda , h. 1992 , apj , 391 , 102 urpin , v. a. & yakovlev , d. g. 1980 , , 24 , 126 , w. d. , lewin , w. h. g. , & van paradijs , j. 1986 , , 220 , 339 , j. , dotani , t. , tanaka , y. , & tsuru , t. 1990 , , 42 , 633 , i. , inoue , h. , koyama , k. , matsuoka , m. , murakami , t. , ogawara , y. , ohashi , t. , tanaka , y. , hayakawa , s. , tawara , y. , miyamoto , s. , tsunemi , h. , & kondo , i. 1984 , , 36 , 819 , m. a. & meszaros , p. 1989
, , 346 , 844 , r. 2001 , , 554 , l59 , s. e. & taam , r. e. 1976 , , 263 , 101 youngs , d. l. 1994 , lasers and particle beams , 12 , 725 , w. , strohmayer , t. e. , & swank , j. h. 1997 , , 482 , l167 , m. , timmes , f. x. , fryxell , b. , lamb , d. q. , olson , k. , calder , a. c. , dursi , l. j. , ricker , p. , rosner , r. , macneice , p. , & tufo , h. m. 2001 , , in press
10 pt figure 2 : time history of the x - ray flux from 4u 1820 - 30 during the super burst .
the pca lightcurve ( 2 - 60 kev ) at 1/8 s resolution is the higher time resolution trace ( left axis ) .
the lower time resolution curve is the ( 8 - 30)/(2 - 8) kev hardness ratio from standard2 data with 16 s resolution ( right axis ) .
figure 3 : comparison of the helium flash which preceded the super burst ( top panel ) and a helium flash which was observed on may 2 , 1997 ( bottom panel ) . shown for both bursts
are the 2 - 60 kev lightcurves from standard1 data with 1/8 s time resolution .
the rates are scaled to 3 pcu detectors .
figure 5 : residuals to a spectral fit during the peak of the burst using a black body function with photoelectric absorption . shown
are the residuals , data - model , in units of standard deviations .
the emission line feature just longward of 6 kev as well as the edge between 8 - 10 kev are cleary evident .
figure 6 : count rate spectrum and model ( top panel ) and residuals , data - model in units of standard deviations ( bottom panel ) for a spectral fit to an interval during the peak of the burst .
the model includes a black body function , a gaussian emission line near 6 kev and an absorption edge near 9 kev .
note that the error bars near the peak of the spectrum are almost smaller than the width of the lines .
figure 8 : time evolution of the bolometric flux ( top panel ) , black body temperature ( middle ) and the inferred black body radius ( bottom ) assuming a distance of 6.6 kpc .
the time interval shown corresponds to the interval between the vertical dashed lines in figure 7 .
the bolometric flux was derived by integrating the best - fit black body spectrum .
figure 9 : time evolution of the emission line and absorption edge components throughout the burst . shown
are the line centroid energy , and line flux from the diskline model ( top and 2nd from top ) along with the edge energy and depth ( bottom and next to bottom ) from the smedge model .
figure 10 : the thermal structure of the other layers of a neutron star accreting pure he at a local accretion rate @xmath227 .
the top axis indicates the time required for a column @xmath105 to be accreted at this rate .
plotted are the temperature ( _ solid line _ ) and flux ( _ dotted line _ ) in units of mev per accreted nucleon .
the arrow is to guide the eye to the value of @xmath228 at @xmath229 .
the dots indicates the base of the c / fe layer .
the condition for instability of the reaction @xmath101 is indicated by the heavy dashed curve .
figure 12 : the fluence , in units of @xmath231 , as a function of time , in units of @xmath186 , for the one - zone calculation ( eq . [ [ eq : one - zone - neutrino ] ] ) . shown
are both the neutrino ( _ solid _ line ) and conductive ( _ dotted _ line ) fluences .
the conductive flux evolves on the thermal diffusion timescale appropriate for a depth of @xmath140 .
figure 14 : the conductive fluence , in units of @xmath231 , for the model calculation described in eq .
[ [ eq : diffusion ] ] , as a function of time in units of @xmath186 .
we show both the outward - directed ( _ solid _ line ) and inward - directed ( _ dotted _ line ) fluence . as in fig .
[ fig : compare - fluence ] , the evolution of the fluence is on the thermal decay timescale at the base of the burning layer . | we present a detailed observational and theoretical study of a @xmath0 hr long x - ray burst ( the `` super burst '' ) observed by the rossi x - ray timing explorer ( rxte ) from the low mass x - ray binary ( lmxb ) 4u 1820 - 30 .
this is the longest x - ray burst ever observed from this source , and perhaps one of the longest ever observed in great detail from any source .
we show that the super burst is thermonuclear in origin .
its peak luminosity of @xmath1 ergs s@xmath2 is consistent with the helium eddington limit for a neutron star at @xmath3 kpc , as well as the peak luminosity of other , shorter , thermonuclear bursts from the same source .
the super burst begins in the decaying tail of a more typical ( @xmath4 s duration ) thermonuclear burst .
these shorter , more frequent bursts are well known helium flashes from this source .
the level of the accretion driven flux as well as the observed energy release of upwards of @xmath5 ergs indicate that helium could not be the energy source for the super burst .
we outline the physics relevant to carbon production and burning on helium accreting neutron stars and present calculations of the thermal evolution and stability of a carbon layer and show that this process is the most likely explanation for the super burst .
ignition at the temperatures in the deep carbon `` ocean '' requires @xmath6 times the mass of carbon inferred from the observed burst energetics unless the he flash is able to trigger a deflagration from a much smaller mass of carbon .
we show , however , that for large columns of accreted carbon fuel , a substantial fraction of the energy released in the carbon burning layer is radiated away as neutrinos , and the heat that is conducted from the burning layer in large part flows inward , only to be released on timescales longer than the observed burst .
thus the energy released during the event possibly exceeds that observed in x - rays by more than a factor of ten , making the scenario of burning a large mass of carbon at great depths consistent with the observed fluence without invoking any additional trigger .
a strong constraint on this scenario is the recurrence time : to accrete an ignition column of @xmath7 g @xmath8 takes @xmath9 g s@xmath2 ) yr .
spectral analysis during the super burst reveals the presence of a broad emission line between @xmath10 kev and an edge at @xmath11 kev likely due to reflection of the burst flux from the inner accretion disk in 4u 1820 - 30 .
we believe this is the first time such a signature has been unambiguously detected in the spectrum of an x - ray burst . |
emission from interstellar dust comprises a wide - ranging portion of the overall energy budget in galaxies , ranging from typically a few percent for ellipticals ( xilouris et al .
2004 ) , to tens of percent for normal star - forming galaxies ( dale et al . 2000 ) , to greater than 95 percent for many ultraluminous infrared galaxies ( soifer et al .
1984 ) . to take advantage of this energetically - important view of galaxies , many current and planned instruments are pushing infrared , submillimeter , and millimeter astronomy into new sensitivity regimes .
increasingly deep cosmological surveys carried out at these long wavelengths are already revealing a rich diversity of spectral energy distributions ( seds ) at every redshift ( e.g. yan et al .
2004 ; rowan - robinson et al .
2005 ) . in order to exploit these surveys to their full potential
, it is paramount to empirically characterize the details of individual galaxy infrared seds and their broad characteristics for a diverse sample of galaxies .
the 75 galaxies studied in detail in the sings legacy science program represent a fair sampling of the range of galaxies found in the local universe ( kennicutt et al .
the suite of _ spitzer space telescope _ cosmology surveys , on the other hand , will produce vast catalogs of infrared - bright sources at much larger distances .
tremendous advances in the understanding of galaxy infrared seds could potentially be made with such huge catalogs .
however , due to sensitivity and confusion limitations , these surveys will not constrain the far - infrared fluxes for most sources ( the initial swire catalog has detections at 70 and 160 for only 0.7% of its sources with photometric redshifts ; rowan - robinson et al .
2005 ) , whereas most sings targets will have robust detections at near- , mid- , and far - infrared wavelengths in addition to submillimeter data for a third of the sample .
in other words , though it only comprises 75 targets , the broad multi - wavelength coverage for the sings sample should enable unique advances in sed understanding .
in addition , inferring a galaxy s total infrared luminosity from just 24 or 24 + 70 measurements alone would be useful to cosmology surveys , and the collection of sings seds make it possible to calibrate these extrapolations and quantify the uncertainties involved .
finally , another important advantage to the sings sample lies in its proximity .
the unprecedented combination of sensitivity and angular resolution , coupled with the nearness of the sings targets , will enable us to learn about differences in the infrared seds as a function of the local ( sub - kpc to kpc ) environment .
the various cosmological efforts being pursued via the _
spitzer space telescope _ will profit from this compilation of sings infrared spectra .
the focus of this study is to make a preliminary exploration of the broad variations of sings 1 - 850 seds using both globally - integrated data and data stemming from smaller spatial scales within galaxies .
a complementary study of sings infrared spectra will concentrate on a narrower wavelength range to quantify the variations observed in the mid - infrared spectral shapes .
the full sings sample is described in kennicutt et al .
( 2003 ) . mid- and far - infrared images are now becoming available for a good portion of the sings sample .
the galaxies chosen for study in this work are simply those first sings galaxies with mid- and far - infrared broadband data in hand .
the sample spans a wide range in the infrared / optical ratio and morphological type . in this work we also present results on the infrared seds for various locations within
three sings galaxies : ngc 3031 ( m 81 ) , ngc 5194 ( m 51 ) , and ngc 7331 .
these three galaxies span large enough solid angles to allow many local environments to be explored .
figures [ fig : apertures1 ] and [ fig : apertures2 ] indicate the particular regions selected for analysis in each galaxy overlaid on ( unsmoothed ) 8.0 images .
apertures were selected using 8.0 and 24 images .
the diameter of each aperture is at least as large as 38 , the full width at half maximum ( fwhm ) of the poorest resolution point spread function in the dataset ( i.e. the 160 beam ) .
each image was smoothed to the 160 resolution before fluxes were extracted ( [ sec : local ] ) .
`` nuclear '' apertures are centered on the brightest central location ; `` arm '' regions follow the bright regions of the spiral patterns ; `` inter - arm '' regions sample the diffuse emission between spiral arms ; `` inner disk '' and `` outer disk '' apertures are centered on areas that appear to be a combination of arm and inter - arm regions .
the broad scope of the sings imaging program is described in kennicutt et al .
( 2003 ) , though a few modifications have since been incorporated , based on our initial validation data for ngc 7331 ( e.g. regan et al 2004 ) . in brief
, mid - infrared mosaics centered on each source are executed for all four irac channels ( fazio et al .
the mosaics are large enough and deep enough to detect , on average , emission out to the optical radius @xmath1 , in addition to providing generous sky coverage for excellent measurements of the background / foreground sky .
the removal of asteroids and detector artifacts is enabled via a second mapping that typically occurs a few days after the first visit .
the basic calibrated data images have been processed with the standard sings irac pipeline .
the final pixel scale is @xmath2075 .
all of the 24 , 70 , and 160 images are being obtained in the mips scan mode ( rieke et al .
2004 ) . as with the irac imaging strategy , the mips maps are large enough to detect emission out to the optical radius , there is sufficient sky coverage , and asteroid removal is enabled via two separate passes at each source .
the final pixel scales are 075 , 30 , and 60 at 24 , 70 , and 160 , respectively .
these pixel scales allow us to finely sample the point spread function at each wavelength and were chosen to match ( or be an integer multiple of ) the irac pixel scale .
the mips data were processed using the mips instrument team data analysis tool ( gordon et al .
note that the mips data processing has evolved since our validation work on ngc 7331 ; removal of stimflash latent images in the 70 data were only performed for ngc 7331 , for example .
further details of the mips data reduction procedures can be found in bendo et al .
( 2005 ) .
foreground stars ( and obvious background galaxies ) were edited from each irac and mips image before flux extraction was performed .
foreground stars were identified for this process by examining and color images .
global flux densities are extracted from apertures that cover the entire optical disk ( see table [ tab : fluxes ] ) .
the same apertures are used to measure flux densities in all mips and irac wavebands .
sky subtraction was carried out through the use of multiple sky apertures placed near the source , yet do not overlap with the faintest isophotes visible from the galaxy .
statistical uncertainties related to sky subtraction are usually less than 1% , but can be appreciable ( tens of percent ) for faint sources .
systematics in the irac calibration result in fluxes uncertain at the 10% level ( see table [ tab : calibration ] ) .
the uncertainties in the mips fluxes are also dominated by the uncertainties in the calibration : 10% at 24 and 20% at 70 and 160 .
the uncertainties listed in table [ tab : fluxes ] include both systematic and statistical uncertainties .
no aperture corrections have been applied to the irac fluxes listed in table [ tab : fluxes ] , nor in the plots throughout this work .
aperture corrections for extended sources are still unknown and none of our conclusions would rely on such small changes ( estimated to be between @xmath24% and 25% in pahre et al .
2004 ) .
submillimeter maps have been obtained with scuba for 27 sings galaxies , including 25 of the _ mipsirac in the present sample ( holland et al .
thirteen of the maps were taken by us in jiggle map mode , whereas 14 of the maps are archival jiggle or scan maps .
depending on the galaxy size , the maps run from @xmath23 to several arcminutes in extent .
the maps are large enough to capture essentially all of the submillimeter emission for the galaxies in this work except for those listed in table [ tab : submm ] with aperture correction factors .
based on the fractional mips flux contained within apertures similar to the submillimeter map sizes and assuming no far - infrared - submillimeter color gradients , for these galaxies it is likely that we have only detected 45 - 90% of the total submillimeter emission .
the submillimeter fluxes listed in table [ tab : submm ] have been corrected by the multiplicative factors in the same table .
the data were processed with the scuba user reduction facility ( jenness & lightfoot 1998 ) .
the data were first flatfielded and corrected for atmospheric extinction .
contributions from noisy bolometers were then removed , followed by spike removal . for the jiggle maps
the background signal was subtracted using the signal from several bolometers along the edges of the maps ( the scuba maps are large enough to include sky for all galaxies except for ngc 5033 ; the sky estimation for this galaxy is suspect ) . for the scan maps the signal baseline and sky background
were removed and then the data were calibrated , grouped according to chop throw and angle , and regridded .
the data were then smoothed with a 9 guassian beam and combined together using the remdbm procedure in surf .
a residual background was subtracted by smoothing the map with a boxcar function 150 - 300 in size and then subtracting the smoothed map from the original .
note that this step may additionally subtract some diffuse , extended emission .
further details of the submillimeter data processing can be found in bendo et al .
the data were calibrated using observations of submillimeter standards and regridded onto the sky plane .
the calibration of the 450 and 850 fluxes are uncertain at the 25% and 15% levels , respectively .
archival near - infrared , mid - infrared , and far - infrared data are used to supplement the and data , where available .
the 2mass near - infrared fluxes used in this study are the total fluxes listed in the nasa / ipac extragalactic database and are obtained by extrapolating the near - infrared surface brightness profiles out to about five disk scale lengths ( see jarrett et al .
fluxes from the 2mass all - sky extended source catalog exist for all but the six faintest of the 75 sings galaxies .
for these six sources we have extracted 2mass fluxes using the same apertures used to determine irac and mips fluxes .
flux uncertainty is at the 5% level .
archival 6.8 and 15 fluxes ( isocam ) are taken from dale et al .
( 2000 ) and roussel et al .
( 2001 ) for 13 sings galaxies .
in addition , we have extracted 6.8 and 15 fluxes from archival isocam data for another eight sources .
isocam fluxes are typically @xmath220% uncertain , though this percentage can substantially rise for very faint sources like holmberg ii
. finally , fluxes , accurate to @xmath220% and obtained from scanpi and hires data , are available in at least one wavelength for all but three sings targets .
imaging was obtained for ngc 3031 , ngc 5194 , and ngc 7331 at the kitt peak national observatory 2.1 m telescope in march 2001 ( ngc 3031 and ngc 5194 ) and november 2001 ( ngc 7331 ) .
integrations were 1800 s for the narrowband imaging and 420 s for the r band imaging that was used for continuum subtraction .
single pointings of the 104 field of view ( 03 pixel@xmath3 ) were sufficient to capture the emission for ngc 7331 , but mosaics were necessary to encompass the large angular diameters of ngc 3031 and ngc 5194 .
flux calibration was obtained through observations of spectrophotometric sources from massey et al .
the data were processed using standard iraf routines . a correction for in these relatively metal - rich galaxies
was made assuming a uniform value of /@xmath4 ( hoopes & walterbos 2003 ) .
fluxes were further corrected for internal extinction ( assuming @xmath51 mag ; e.g. kennicutt 1998 ) and foreground milky way extinction ( schlegel , finkbeiner , & davis 1998 ) . the overall uncertainty in the fluxes , including systematics , is estimated to be 25% .
figures [ fig : seds1]-[fig : seds8 ] show the infrared data for each galaxy ; mass , , , and data are shown where available . included in each plot
are two model curves and their sum .
the dashed curve is a dust - only model from the semi - empirical work of dale & helou ( 2002 ) on normal star - forming galaxies .
the dust - only templates have the advantage of being parametrized by a single parameter family , essentially defined by the far - infrared color .
the @xmath6 listed within each panel parametrizes the distribution of dust mass as a function of local heating intensity @xmath7 @xmath8 the dust curve ( @xmath6 ) is computed for each galaxy via 10,000 monte carlo simulations of a least - squares fit to the mips data . for each simulation
we add a random component to the mips fluxes , gaussian - scaled by the fluxes uncertainties .
the average @xmath6 and its dispersion is given in each panel .
we have likewise adopted a conceptually simple approach for the stellar template .
the dotted curve is the 900 myr continuous star formation , solar metallicity , salpeter imf ( @xmath9 ) curve from vazquez & leitherer ( 2005 ) fitted to the 2mass fluxes . though our current approach is mostly illustrative and avoids a more sophisticated suite of multi - parameter dust and stellar models , it is remarkable how well the data are generally fit by such a simple combination .
note that for a few galaxies the far - infrared emission appears to peak at a wavelength that is slightly longer than that in the most quiescent dust model ( i.e. @xmath10 ) .
the dust - only models will be updated , and a more refined approach will be pursued for the stellar profiles , once all the sings data are in hand .
finally , we note that the mid - infrared data for ngc 1377 are quite discrepant from the model predicted by the far - infrared fluxes .
roussel et al .
( 2003 ) suggest that ngc 1377 is in the initial throes of an intense starburst , and that this heavily extincted system is being observed at such an early stage of the star formation that it is yet to show any optical signatures or synchrotron radiation .
the unusual mid - infrared properties of ngc 1377 are studied in detail by roussel et al .
( in preparation ) . information on our collection of infrared spectra can be more compactly displayed in a single sed plot and in an infrared color - color diagram .
the examples presented in figure [ fig : seds_sample ] portray the variety of spectral shapes in the sings sample . arbitrarily normalizing the spectra at 8.0 emphasizes the significant dispersion in the sample s near - infrared / mid - infrared and far - infrared / mid - infrared colors .
the top panel of figure [ fig : global_colors ] shows the ( dust - only ) mid - infrared color as a function of the far - infrared color .
the stellar contribution is subtracted from the 8.0 and 24 measures via extrapolation of the 3.6 emission , assuming the 3.6 emission is completely stellar : @xmath11 where @xmath12 and @xmath13 are chosen such that the dust - only flux densities vanish for a standard stellar spectrum ( e.g. helou et al .
2004 ; engelbracht et al .
2005 ) . likewise , the ratio of 8.0 dust emission to _ total _ dust emission is plotted in the bottom panel of figure [ fig : global_colors ] using the relation @xmath14 where @xmath15 and the denominator is obtained from mips fluxes using equation 4 in dale & helou ( 2002 ) . the solid and dotted lines in figure [ fig : global_colors ]
respectively indicate the flavors of sed models presented in dale & helou ( 2002 ) and dale et al .
( 2001 ) , derived from the average global trends for a sample of normal star - forming galaxies observed by and .
the solid curve incorporates an improved approach to modeling the far - infrared / submillimeter emission ( dale & helou 2002 ) .
some of the low - metallicity objects lie dramatically below the canonical star - forming galaxy curves , consistent with the engelbracht et al .
( 2005 ) empirical finding that low metallicity galaxies show relatively low 8.0 emission .
these low - metallicity galaxies drive the scatter in the ratio between the inferred dust - only flux at 8 and the 8 flux as predicted from the dust model fitted to the mips fluxes .
the dispersion in this ratio is 2.90 and 1.82 dex including and not including , respectively , the low - metallicity systems .
a likely reason for the relatively low 8 flux for low - metallicity galaxies is a different balance between pah destruction and formation within these systems ( pahs dominate the emission at 8 ; see figures 8 and 12 in li & draine 2001 ) . for example , if pahs are formed by grain - surface chemistry in the interstellar medium ( so the grain formation rate would vary as the metallicity squared ) , and destroyed by gas - phase processes ( sputtering in shock waves , or photodestruction ) one might expect lower pah / dust ratios in low metallicity systems .
finally , we note that the sings sample fills in more of the quiescent portion of the color - color space ( e.g. small values of ) , and in fact some of the points lie beyond the most quiescent values predicted by the previous generation of models .
the sings sample spans a larger range of galaxy types , infrared / optical ratios , metallicities , and infrared colors than the sample from the key project on the interstellar medium of normal galaxies , so it is unsurprising that some of the data points in figure [ fig : global_colors ] do not exactly match the models predictions . to properly execute a multi - wavelength investigation within galaxies , all the images were first convolved to match the point - spread function of the 160 beam which has fwhm of 38 ( see engelbracht et al .
2004 ) . at the distances
to ngc 3031 , ngc 5194 , and ngc 7331 , 38 corresponds to spatial scales of @xmath20.6 , 1.5 , and 2.8 kpc , respectively .
after the data were all smoothed to the same resolution , sky - subtracted flux densities were extracted from the 170@xmath16 various regions within each galaxy ( figures [ fig : apertures1 ] and [ fig : apertures2 ] ) .
figure [ fig : local_colors ] shows the mid - infrared color as a function of the far - infrared color for the various regions in these three galaxies . in general ,
the observed seds are indicative of more intense radiation fields for regions closer to the nucleus and arms , with larger ratios due to hotter large grains and smaller ratios due to elevated small grain emission at 24 .
the relatively low values for near the nuclei and arm regions are unlikely explained by a diminution of the 8.0 flux as a result of the destruction of polycyclic aromatic hydrocarbons , since the radiation fields suggested by within these galaxies are fairly mild .
this issue is addressed in figure [ fig : local_dust ] .
the overall constancy of the 8.0-to - total infrared ratio in figure [ fig : local_dust ] supports the conclusion that the relatively low values of for high are due to elevated 24 emission and not a reduction in the pah emission .
another interesting facet of the data plotted in figure [ fig : local_dust ] is the comparatively small variation in 8.0-to - total dust emission : the typical scatter within each galaxy is @xmath215% even as varies by a factor of four .
this result reinforces the notion that the observed variations in do not arise from variations in 8.0-to - total dust emission , but rather from variations in 24-to - total dust emission .
another obvious feature to figure [ fig : local_colors ] is the substantial variations seen within a given galaxy s disk . the and colors vary by a factor of two or more within each disk , and the dispersion is larger the closer the galaxy .
ngc 3031 shows the largest scatter because in this galaxy we are sampling the smallest spatial scales and thus the widest variety of environments .
moreover , some of the local values are smaller than that predicted from the globally - integrated properties of the most quiescent galaxies studied by _ iso _ and _ iras _
( i.e. the leftmost values of the model curves displayed in figure [ fig : local_colors ] ) .
in other words , many of the local regions within these galaxies are more quiescent than what is seen for the global emission in normal star - forming galaxies .
this is not unexpected , as globally - integrated emission will inevitably encompass some hot spots , raising the overall average ratio above the typical cirrus value .
as alluded to above , the star formation rate is conceptually linked to the shape of the far - infrared spectrum
regions near enhanced star formation should have more intense interstellar radiation fields , and thus the large grains in such regions should be hotter than their counterparts in the diffuse interstellar medium .
however , before the advent of the _ spitzer space telescope _ and its superior angular resolving power , it was difficult to explore correlations between the far - infrared color and the local star formation rate in all but the closest of galaxies ( e.g. xu & helou 1994 ; xu & helou 1996 ) . figure [ fig : sfr ] shows the far - infrared color as measured by versus the local star formation rate for ngc 3031 , ngc 5194 , and ngc 7331 .
star formation rates were computed using the relation presented in kennicutt ( 1998 ) : sfr ( @xmath17 year@xmath18 ( ) ( ergs s@xmath3 ) .
the two quantities are clearly correlated ; a non - parametric ranking analysis indicates a correlation at the @xmath19 level .
the ratio is likewise correlated with the local star formation rate , and much less so .
these results suggest that both 24 and 70 trace warmer dust phases , consistent with morphological studies carried out by hinz et al .
( 2004 ) for m 33 , helou et al .
( 2004 ) for ngc 300 , gordon et al .
( 2004 ) , and rieke et al . for m 31
( private communication ) .
a large fraction of the 70 emission may come from the same grains that are thought to produce the 24 emission ( e.g. dsert , boulanger , & puget 1990 ) , grains that are particularly sensitive to the effects of local star formation .
one of the long - range goals of the sings project is to provide a reference set of empirical sed templates of nearby galaxies and regions within these galaxies that cover the full range of properties and interstellar environments found locally .
these templates can then be applied in turn to model the observed redshifted seds of galaxies observed at cosmological distances .
such templates are already available from and observations , such as the key project sample ( e.g. dale & helou 2002 ) . the sings sample is more diverse than the key project sample ( metallicity , colors , morphology , etc . ) , and will eventually provide a definitive set of reference seds for such applications .
the preliminary data presented here already provide some useful guidance ( and warnings ) for interpreting infrared observations of high - redshift galaxies from 24 or 8.0 data alone ( or together ) .
as a guide for -based cosmological studies , we show in figure [ fig : fir_z ] how the and the ( dust - only ) bolometric - to-24 ratios would change as a function of redshift for a variety of sings - like objects .
the tracks are pinned at redshift zero according to each source s observed infrared properties , and traced out to higher redshifts using the model of dale & helou ( 2002 ) .
figure [ fig : fir_z ] shows that sed variations remain a considerable source of systematic uncertainty even when the redshift is known .
given the range of infrared spectral energy distributions observed for this portion of the sings galaxy sample , extrapolating the total infrared luminosity from the 24 emission alone would yield uncertainties at the factor - of - five level .
this level of uncertainty can be reduced with some _ a priori _ constraints on the types and the redshifts of sources being studied ( e.g. papovich & bell 2002 ) , as well as information on the infrared colors ( e.g. ratio or their redshifted equivalents ) . however , for many current applications
even the redshifted 24 emission is inaccessible , and one must rely on the _ observed _ 24 flux , which at @xmath0 corresponds to 8.0 in the rest frame .
an in - depth analysis of the application of this index at high redshift will be presented in a separate paper ( smith et al . 2005 ) , but in the meantime we make some preliminary assessments based on the sed data shown in figures [ fig : global_colors ] and [ fig : local_dust ] . within individual large , metal - rich galaxies such as ngc 3031 , ngc 5194 , and ngc 7331
the ratio of 8.0-to - total infrared luminosity has a surprisingly low dispersion , less than a factor of two ( maximum to minimum ) within individual galaxies and hardly more among the three galaxies plotted in figure [ fig : local_dust ] .
however if one considers a much larger galaxy sample ( figure [ fig : global_colors ] ) , the total range in 8.0-to - total infrared luminosity ratios swells to more than a factor of twenty .
much of this large scatter can be attributed to a few metal - poor dwarf galaxies , in which the 8.0 pah emission is quenched . if one excludes the metal - poor dwarfs the total range of 8.0-to - total ratios shrinks to a factor of 10 , and this probably should be taken as a minimum systematic uncertainty in total infrared luminosities and star formation rates inferred from rest - frame 8.0 fluxes alone ( also see calzetti et al . 2005 ) .
if the metal abundance of the population can not be inferred _ a priori _ then the uncertainties may be considerably higher .
new imaging data are coupled with data from mass , , , and to investigate the properties of the 1 - 850 seds for _ mipsirac nearby galaxies from the sings sample .
the various shapes of the globally - integrated dust continuum generally follow the trends expected from the previous generation of infrared dust models that are based on and observations of normal star - forming galaxies from the key project sample .
however , the sings sample is more diverse than the key project sample ( metallicity , colors , morphology , etc ) , so it is perhaps not surprising that the sings sample shows larger variations in infrared colors .
we also take advantage of s unprecedented angular resolving power to explore over 170 regions within ngc 3031 , ngc 5194 , and ngc 7331 .
all three galaxies exhibit interesting trends in their infrared emission over the various environments sampled in this study . within each of these galaxies
we find a wide variety of spectral shapes , especially for ngc 3031 , the closest of the three targets and thus the galaxy for which we are able to sample the smallest spatial scales .
in particular we notice fairly constant ratios of dust - only 8.0 emission with respect to both bolometric dust emission , over large dynamic ranges of dust temperature as measured by .
the relative contribution from pahs does not appear to significantly vary on kiloparsec scales , suggesting that the variations observed in are largely driven by variations in the local 24 emission .
the emission from galaxies at 24 , along with that at 70 , traces grains that are particularly sensitive to changes in the local interstellar radiation field and thus can be used as proxies for gauging the strength of the local star formation ( e.g. calzetti et al . 2005 ) .
this dataset on a wide - ranging sample of local galaxies is useful for estimating how accurately one can predict the total infrared luminosity ( and star formation rate ) from just the 8.0 or 24 data alone . though the 8.0-to - total infrared ratio varies by less than a factor of two within ngc 3031 , ngc 5194 , and ngc 7331 , the global ratio varies by more than a factor of 20 among our diverse sample of galaxies .
the 24-to - total infrared ratio likewise spans a large range in the sings sample ( factor of 5 ) . in conjunction with models of infrared seds for normal star - forming galaxies
, we predict that the _ observed _ 24-to - total infrared ratio ( e.g. rest - frame 8.0-to - total at @xmath20 ) maintains at least this dispersion at most epochs back to a redshift of four .
as mentioned above , the characteristics of the infrared seds as seen by are broadly consistent with normal galaxy models developed based on and data , but with perhaps more significant dispersion .
once the complete sings dataset is in hand , the suite of empirical sed models referenced here will be updated according to the diversity of spectral shapes exhibited by the sings project .
a more generally comprehensive and quantitative analysis will be pursued as well .
for example , how much more dispersion in infrared colors is seen within galaxies versus between galaxies ?
how do the global spectral shapes depend on metallicity in this sample ? are the infrared surface brightnesses correlated with the far - infrared color or galaxy type ?
data from all 75 galaxies in the diverse sings sample will help to answer these questions .
we acknowledge e. xilouris for helpful input .
support for this work , part of the _ spitzer space telescope _
legacy science program , was provided by nasa through contract number 1224769 issued by the jet propulsion laboratory , california institute of technology under nasa contract 1407 .
this research has made use of the nasa / ipac extragalactic database which is operated by jpl / caltech , under contract with nasa .
this publication makes use of data products from the two micron all sky survey , which is a joint project of the university of massachusetts and the infrared processing and analysis center / california institute of technology , funded by the national aeronautics and space administration and the national science foundation .
dum bendo , g. et al .
2005 , , submitted calzetti , d. et al .
2005 , , in press dale , d.a .
2000 , , 120 , 583 dale , d.a .
2001 , , 549 , 215 dale , d.a . &
helou , g. 2002 , , 576 , 159 dsert , f.x , boulanger , f. & puget , j.l . 1990 , , 237 , 215 engelbracht , c.w .
2004 , , 154 , 248 engelbracht , c.w . ,
gordon , k.d . ,
rieke , g.h . ,
werner , m.w .
, dale , d.a . , & latter , w.b .
2005 , , 628 , l29 fazio , g. et al .
2004 , , 154 , 10 gordon , k. et al .
2004 , , 154 , 215 gordon , k. d. et al .
2005 , , 117 , 503 helou , g. et al .
2004 , , 154 , 253 hinz , et al .
2004 , , 154 , 259 holland , w.s .
1999 , , 303 , 659 hoopes , c.g . &
walterbos , r.a.m .
2003 , , 586 , 902 jarret , t.h . ,
chester , t. , cutri , r. , schneider , s.e . , & huchra , j.p . 2003 , , 125 , 525 jenness , t. & lightfoot , j. f. 1998 , `` _ reducing scuba data at the james clerk maxwell telescope _ '' , in _ astronomical data analysis software and systems vii _ , asp conf .
ser . , 145 , 216 kennicutt , r.c . 1998 , , 36 , 189 kennicutt , r.c .
2003 , , 115 , 928 li , a. & draine , b.t .
2001 , , 554 , 778 massey , p. , strobel , k. , barnes , j.v . , &
anderson , e. 1988 , , 328 , 315 pahre , m.a .
, ashby , m.l.n . , fazio , g.g .
, & willner , s.p .
2004 , , 154 , 229 papovich , c. & bell , e.f .
2002 , , 579 , l1 regan , m.w .
2004 , , 154 , 204 rieke , g. et al .
2004 , , 154 , 25 rowan - robinson , r. et al .
2005 , , 129 , 1183 roussel , h. et al .
2001 , , 369 , 473 roussel , h. , helou , g. , beck , r. , condon , j.j . , bosma , a. , matthews , k. , & jarrett , t.h .
2003 , , 593 , 733 schlegel , d.j . ,
finkbeiner , d.p .
, & davis , m. 1998 , , 500 , 525 soifer , b.,t . ,
neugebauer , g. , helou , g. , lonsdale , c.j . ,
hacking , p. , rice , w. , houck , j.r . , low , f.j . , & rowan - robinson , m. 1984 , , 283 , l1 vazquez , g.a .
& leitherer , c. 2005 , , 621 , 695 xilouris , e.m . , madden , s.c . , galliano , f. , vigroux , l. , & sauvage , m. 2004 , , 416 , 41 xu , c. & helou , g. 1994 , , 426 , 109 xu , c. & helou , g. 1996 , , 456 , 152 yan , l. et al .
2004 , , 154 , 60 lrrrrrrr c [ tab : fluxes ] ngc 0024 & & & & & 0.13 0.01 & 1.89 0.39 & 6.7231.36 ngc 0337 & 0.110.01 & 0.0700.007 & 0.17 0.02 & 0.50 0.05 & 0.65 0.07 & 8.83 1.77 & 18.30 3.69 ngc 0584 & 0.400.04 & 0.23 0.02 & 0.22 0.02 & 0.15 0.02 & 0.05 0.006 & 0.15 0.07 & 1.02 0.40 ngc 0628 & 0.950.10 & 0.58 0.06 & 1.50 0.15 & 3.62 0.36 & 3.08 0.31 & 29.73 5.95 & 116.64 23.34 ngc 0855 & 0.0460.005 & 0.0290.003 & 0.0220.003 & 0.0600.006 & 0.0820.008 & 1.38 0.28 & 2.09 0.44 ngc 0925 & & & & & 0.90 0.09 & 12.20 2.45 & 39.52 7.95 ngc 1097 & 1.360.14 & 0.85 0.09 & 1.89 0.19 & 4.29 0.43 & 6.40 0.64 & 43.40 8.68 & 144.63 28.93 ngc 1266 & 0.0590.006 & 0.0440.004 & 0.0660.007 & 0.12 0.01 & 0.84 0.08 & 9.64 1.93 & 9.14 1.84 ngc 1291 & 2.320.23 & 1.35 0.14 & 1.25 0.13 & 0.86 0.09 & 0.44 0.05 & 5.41 1.09 & 28.47 5.75 ngc 1316 & 2.720.27 & 1.63 0.16 & 1.46 0.15 & 0.75 0.08 & 0.36 0.04 & 4.22 0.85 & 9.66 1.94 ngc 1377 & 0.0620.006 & 0.0900.009 & 0.32 0.03 & 0.55 0.06 & 1.73 0.17 & 4.76 0.95 & 2.91 0.60 ngc 1404 & 0.800.08 & 0.46 0.05 & 0.42 0.04 & 0.21 0.02 & 0.0830.009 & 0.15 0.09 & 0.31 0.18 ngc 1512 & 0.430.04 & 0.26 0.03 & 0.34 0.03 & 0.59 0.06 & 0.42 0.04 & 5.40 1.08 & 21.85 4.38 ngc 1566 & 0.820.08 & 0.51 0.05 & 1.16 0.12 & 2.84 0.28 & 2.65 0.27 & 27.82 5.57 & 95.26 19.05 ngc 1705 & 0.0280.003 & 0.0190.002 & 0.0120.002 & 0.0220.002 & 0.0520.005 & 1.09 0.22 & 1.20 0.25 ngc 2403 & 2.060.21 & 1.38 0.14 & 2.79 0.28 & 5.53 0.55 & 5.64 0.56 & 75.58 15.12 & 231.56 46.32 holmberg ii & 0.0780.008 & 0.0600.006 & 0.0390.005 & 0.0320.005 & 0.17 0.02 & 3.18 0.64 & 4.05 0.87 m81 dwarf a & 0.0020.001 & 0.0010.001 & @xmath210.001 & @xmath210.001 & & & ddo 053 & 0.0060.001 & 0.0050.001 & 0.0030.001 & 0.0100.001 & 0.0280.003 & 0.31 0.07 & 0.32 0.11 ngc 2798 & 0.130.01 & 0.0860.009 & 0.32 0.03 & 0.84 0.08 & 2.51 0.25 & 14.70 2.94 & 18.45 3.69 ngc 2841 & 1.390.14 & 0.80 0.08 & 0.85 0.09 & 1.56 0.16 & 0.88 0.09 & 8.66 1.74 & 54.87 10.98 ngc 2915 & & & & & 0.0580.006 & 1.09 0.22 & 1.09 0.30 holmberg i & 0.0130.001 & 0.0080.001 & 0.0090.002 & 0.0100.002 & 0.0130.004 & 0.33 0.12 & 0.76 0.23 ngc 2976 & 0.470.05 & 0.30 0.03 & 0.64 0.06 & 1.36 0.14 & 1.33 0.13 & 16.99 3.40 & 46.81 9.40 ngc 3049 & 0.0440.004 & 0.0290.003 & 0.0780.008 & 0.18 0.02 & 0.41 0.04 & 2.27 0.46 & 4.05 0.82 ngc 3031 & 11.871.19 & 6.90 0.69 & 7.90 0.79 & 10.78 1.08 & 4.94 0.49 & 74.37 14.88 & 347.10 69.43 holmberg ix & 0.0080.001 & 0.0040.001 & @xmath210.006 & @xmath210.006 & & & m81 dwarf b & 0.0050.001 & 0.0040.001 & 0.0030.001 & 0.0030.001 & 0.0080.001 & 0.12 0.03 & 0.21 0.14 ngc 3190 & 0.410.04 & 0.25 0.03 & 0.30 0.03 & 0.44 0.04 & 0.26 0.03 & 4.34 0.87 & 13.19 2.65 ngc 3198 & 0.300.03 & 0.18 0.02 & 0.42 0.04 & 0.92 0.09 & 1.03 0.10 & 8.68 1.74 & 34.96 7.00 ic 2574 & 0.170.02 & 0.0960.01 & 0.0840.008 & 0.0890.009 & 0.27 0.03 & 4.61 0.92 & 10.31 2.12 ngc 3265 & & & & & 0.28 0.03 & 2.05 0.42 & 2.35 0.49 markarian 33 & 0.0290.003 & 0.0200.002 & 0.0630.006 & 0.17 0.02 & 0.82 0.08 & 3.34 0.67 & 3.46 0.71 ngc 3351 & 0.890.09 & 0.55 0.06 & 0.93 0.09 & 1.80 0.18 & 2.40 0.24 & 16.42 3.29 & 59.72 11.95 ngc 3521 & 2.230.22 & 1.44 0.14 & 3.29 0.33 & 8.36 0.84 & 5.36 0.54 & 49.85 9.97 & 206.65 41.35 lrrrrrrr c ngc 3621 & & & & & 3.30 0.33 & 40.21 8.04 & 126.15 25.24 ngc 3627 & 2.050.21 & 1.32 0.13 & 3.06 0.31 & 7.50 0.75 & 7.25 0.73 & 68.92 13.79 & 208.13 41.63 ngc 3773 & 0.0240.002 & 0.0150.002 & 0.0290.003 & 0.0610.006 & 0.13 0.01 & 1.22 0.25 & 2.12 0.48 ngc 3938 & 0.350.04 & 0.23 0.02 & 0.52 0.05 & 1.32 0.13 & 1.05 0.11 & 12.14 2.43 & 46.78 9.36 ngc 4125 & 0.700.07 & 0.39 0.04 & 0.30 0.03 & 0.19 0.02 & 0.0690.007 & 0.86 0.18 & 1.33 0.30 ngc 4236 & & & & & 0.53 0.05 & 7.08 1.42 & 18.87 3.85 ngc 4254 & 0.770.08 & 0.50 0.05 & 1.89 0.19 & 5.28 0.53 & 4.09 0.41 & 39.02 7.80 & 131.79 26.36 ngc 4321 & 1.040.10 & 0.68 0.07 & 1.55 0.16 & 3.88 0.39 & 3.33 0.33 & 32.28 6.46 & 128.41 25.68 ngc 4450 & 0.580.06 & 0.35 0.03 & 0.32 0.03 & 0.36 0.04 & 0.19 0.02 & 2.46 0.50 & 13.73 2.76 ngc 4536 & & & & & 3.37 0.34 & 22.49 4.50 & 54.39 10.89 ngc 4552 & 0.910.09 & 0.51 0.05 & 0.37 0.04 & 0.23 0.02 & 0.0620.006 & 0.0970.04 & 0.41 0.41 ngc 4559 & 0.390.04 & 0.25 0.03 & 0.53 0.05 & 1.13 0.11 & 1.08 0.11 & 14.32 2.87 & 46.81 9.37 ngc 4569 & 0.830.08 & 0.50 0.05 & 0.75 0.08 & 1.36 0.14 & 1.41 0.14 & 9.65 1.93 & 38.21 7.66 ngc 4579 & 0.950.10 & 0.55 0.06 & 0.66 0.07 & 0.97 0.10 & 0.74 0.07 & 8.21 1.65 & 39.07 7.82 ngc 4594 & 4.290.43 & 2.44 0.24 & 2.18 0.22 & 1.74 0.17 & 0.65 0.07 & 6.71 1.36 & 36.84 7.39 ngc 4631 & 1.380.14 & 0.89 0.09 & 3.22 0.32 & 7.89 0.79 & 7.97 0.80 & 98.78 19.76 & 269.01 53.80 ngc 4725 & 1.250.13 & 0.75 0.08 & 0.97 0.10 & 1.63 0.16 & 0.81 0.08 & 7.48 1.50 & 53.42 10.70 ngc 4736 & 3.950.40 & 2.46 0.25 & 3.61 0.36 & 6.97 0.70 & 5.50 0.55 & 69.89 13.98 & 170.28 34.06 ddo 154 & 0.00120.0004 & 0.00080.0003 & @xmath210.0012 & @xmath210.0009 & 0.00580.002 & 0.0430.03 & 0.26 0.14 ngc 4826 & 2.760.28 & 1.67 0.17 & 2.13 0.21 & 3.16 0.32 & 2.47 0.25 & 35.68 7.14 & 85.39 17.09 ddo 165 & & & & & 0.0110.003 & 0.14 0.05 & 0.27 0.15 ngc 5033 & 0.700.07 & 0.50 0.05 & 1.05 0.11 & 2.59 0.26 & 1.92 0.19 & 21.50 4.30 & 88.15 17.63 ngc 5055 & 2.610.26 & 1.64 0.16 & 3.47 0.35 & 7.60 0.76 & 5.59 0.56 & 59.76 11.95 & 286.34
57.27 ngc 5194 & 2.910.29 & 1.90 0.19 & 5.70 0.57 & 14.33 1.43 & 12.25 1.23 & 131.36 26.30 & 494.34 98.99 ngc 5195 & 0.910.09 & 0.54 0.05 & 0.56 0.06 & 0.85 0.09 & 1.31 0.05 & 10.85 2.17 & 12.34 2.49 tololo 89 & 0.0410.004 & 0.0260.003 & 0.0170.002 & 0.0770.008 & 0.25 0.03 & 1.52 0.31 & 2.69 0.59 ngc 5408 & 0.0560.006 & 0.0390.004 & 0.0500.005 & 0.0500.005 & 0.42 0.04 & 2.95 0.59 & 2.21 0.49 ngc 5474 & 0.110.01 & 0.0850.009 & 0.11 0.01 & 0.15 0.02 & 0.18 0.02 & 3.17 0.64 & 9.49 1.92 ngc 5713 & 0.220.02 & 0.15 0.02 & 0.35 0.04 & 1.52 0.15 & 2.28 0.23 & 17.23 3.45 & 34.77 6.96 ngc 5866 & & & & & 0.20 0.02 & 6.66 1.33 & 16.53 3.31 ic 4710 & 0.0760.008 & 0.0490.005 & 0.0550.006 & 0.0860.009 & 0.11 0.01 & 1.97 0.40 & 3.15 0.67 ngc 6822 & 2.240.22 & 1.41 0.14 & 1.91 0.19 & 0.75 0.08 & 2.51 0.25 & 53.21 10.65 & 136.22 27.27 ngc 6946 & 3.470.35 & 2.26 0.23 & 7.51 0.75 & 18.49 1.85 & 20.87 2.09 & 177.89 35.58 & 498.35 99.71 ngc 7331 & 1.75 0.17 & 1.08 0.11 & 2.36 0.24 & 5.40 0.54 & 3.92 0.39 & 56.49 11.30 & 164.12 32.89 ngc 7552 & 0.50 0.05 & 0.38 0.04 & 1.33 0.13 & 3.63 0.36 & 10.30 1.03 & 45.40 9.09 & 86.65 17.34 ngc 7793 & 0.85 0.08 & 0.50 0.05 & 1.33 0.13 & 2.49 0.25 & 1.97 0.20 & 29.86 5.97 & 119.53 23.91 lcccc [ tab : submm ] ngc 0337 & & 0.350.05 & & ngc 1097 & & 1.440.78 & & 2.09 ngc 2798 & & 0.190.03 & & 1.08 ngc 2976 & & 0.610.24 & & 1.56 ngc 3190 & & 0.190.04 & & 1.12 markarian 33 & & 0.040.01 & & ngc 3521 & & 2.110.82 & & 1.56 ngc 3627 & & 1.860.70 & & 1.53 ngc 4254 & & 1.010.54 & & 2.06 ngc 4321 & & 0.880.49 & & 2.19 ngc 4536 & & 0.420.11 & & 1.30 ngc 4569 & & 0.470.08 & & 1.11 ngc 4579 & & 0.440.07 & & ngc 4594 & & 0.370.11 & & 1.33 ngc 4631 & 30.7010.02 & 5.731.21 & 1.27 & 1.17 ngc 4736 & & 1.540.66 & & 1.67 ngc 4826 & & 1.230.31 & & 1.24 ngc 5033 & & 1.100.55 & & 1.93
ngc 5194 & & 2.610.39 & & ngc 5195 & & 0.260.04 & & ngc 5713 & & 0.570.12 & & 1.17 ngc 5866 & 0.790.20 & 0.140.02 & & ngc 6946 & 18.534.63 & 2.980.45 & & ngc 7331 & 20.568.10 & 2.110.38 & 1.44 & 1.11 ngc 7552 & & 0.800.17 & & 1.17 | the infrared nearby galaxies survey ( sings ) is carrying out a comprehensive multi - wavelength survey on a sample of 75 nearby galaxies .
the 1 - 850 spectral energy distributions are presented using broadband imaging data from , mass , , , and .
the infrared colors derived from the globally - integrated data are generally consistent with the previous generation of models that were developed based on global data for normal star - forming galaxies , though significant deviations are observed .
s excellent sensitivity and resolution also allow a detailed investigation of the infrared spectral energy distributions for various locations within the three large , nearby galaxies ngc 3031 ( m 81 ) , ngc 5194 ( m 51 ) , and ngc 7331 .
a wide variety of spectral shapes is found within each galaxy , especially for ngc 3031 , the closest of the three targets and thus the galaxy for which the smallest spatial scales can be explored .
strong correlations exist between the local star formation rate and the infrared colors and , suggesting that the 24 and 70 emission are useful tracers of the local star formation activity level .
preliminary evidence indicates that variations in the 24 emission , and not variations in the emission from polycyclic aromatic hydrocarbons at 8 , drive the variations in the colors within ngc 3031 , ngc 5194 , and ngc 7331 .
if the galaxy - to - galaxy variations in spectral energy distributions seen in our sample are representative of the range present at high redshift then extrapolations of total infrared luminosities and star formation rates from the observed 24 flux will be uncertain at the factor - of - five level ( total range ) .
the corresponding uncertainties using the redshifted 8.0 flux ( e.g. observed 24 flux for a @xmath0 source ) are factors of 1020 .
considerable caution should be used when interpreting such extrapolated infrared luminosities . |
after the completion of the human genome project has provided a blueprint of the dna present in each human cell @xcite , genomics research is now focusing on the study of dna variations that occur between individuals , seeking to understand how these variations confer susceptibility to common diseases such as diabetes or cancer .
the most common form of genomic variation are the so called _ single nucleotide polymorphisms _ ( snps ) , i.e. , the presence of different dna nucleotides , or _ alleles _ , at certain chromosomal locations .
the vast majority of snps are _ bi - allelic _ ,
i.e. , only two of the four possible dna bases are observed at the snp locus .
since human cells contain two copies of each chromosome ( with the exception of sex chromosomes in males ) , both snp alleles may be present in the dna of an individual .
determining the identity of alleles present in a dna sample at a given set of snp loci is called _
snp genotyping_. the continuous progress in high - throughput genomic technologies has resulted in numerous snp genotyping platforms combining a variety of allele discrimination techniques ( sequencing , direct hybridization , primer extension , allele - specific pcr , ligation , and cleavage , etc . ) , detection mechanisms ( fluorescence , mass spectrometry , etc . ) and reaction formats ( solution phase , solid support , bead arrays ) , see , e.g. , @xcite for comprehensive reviews . however , current technologies still offer an insufficient degree of multiplexing ( below 10,000 snps per assay ) for fully - powered genome wide disease association studies that require genotyping of large sets of user - selected snps @xcite .
the highest throughput is currently achieved by high - density mapping arrays produced by affymetrix , which can simultaneously genotype a fixed set of about 250,000 _ manufacturer selected _ snps per array . genotyping a comparable number of user - specified set of snps
would require an expensive and time - consuming re - design of array probes as well as a difficult re - engineering of the primer - ligation amplification protocol . among technologies that allow
genotyping of custom sets of snps one of the most successful ones is the use of dna tag arrays @xcite .
dna tag arrays consist of a set of dna strings called _ tags _ , designed such that each tag hybridizes strongly to its own _ antitag _ ( watson - crick complement ) , but to no other antitag .
the flexibility of tag arrays comes from combining solid - phase hybridization with the high sensitivity of single - base extension reactions , which has also been used for snp genotyping in combination with maldi - tof mass spectrometry @xcite .
a typical assay based on tag arrays performs snp genotyping using the following steps @xcite : ( 1 ) a set of _ reporter probes _ is synthesized by ligating antitags to the @xmath1 end of primers complementing the genomic sequence immediately preceding the snps of interest .
( 2 ) reporter probes are hybridized in solution with the genomic sample .
( 3 ) the hybridized @xmath2 ( primer ) end of reporter probes is extended by a single base in a reaction using the polymerase enzyme and dideoxynucleotides fluorescently labeled with 4 different dyes .
( 4 ) reporter probes are separated from the template dna and hybridized to a tag array .
( 5 ) finally , fluorescence levels are used to determine the identity of the extending dideoxynucleotides .
commercially available tag arrays have between 2,000 and 10,000 tags @xcite .
the number of snps that can be genotyped per array is typically smaller than the number of tags since some of the tags must remain unassigned due to cross - hybridization with the primers @xcite .
another factor limiting the wider use of tag arrays is the relatively high cost of synthesizing the reporter probes , which have a typical length of 40 nucleotides . in the @xmath0-mer array format @xcite ,
all @xmath3 dna probes of length @xmath0 are spotted or synthesized on the solid array substrate ( values of @xmath0 of up to @xmath4 are feasible with current high - density in - situ synthesis technologies ) .
this format was originally proposed for performing _
sequencing by hybridization ( sbh ) _ , which seeks to reconstruct an unknown dna sequence based on its @xmath0-mer spectrum @xcite .
however , the sequence length for which unambiguous reconstruction is possible with high probability is surprisingly small @xcite , and , despite several suggestions for improvement , such as the use of gapped probes @xcite and pooling of target sequences @xcite , the sbh scheme has not become practical so far . in this paper
we propose a new genotyping assay architecture combining multiplexed solution - phase single - base extension ( sbe ) reactions with sequencing by hybridization ( sbh ) using universal dna arrays such as all @xmath0-mer arrays .
snp genotyping using sbe / sbh assays requires the following steps ( see figure [ k - mer ] ) : ( 1 ) synthesizing primers complementing the genomic sequence immediately preceding snps of interest ; ( 2 ) hybridizing primers with the genomic dna ; ( 3 ) extending each primer by a single base using polymerase enzyme and dideoxynucleotides labeled with 4 different fluorescent dyes ; and finally ( 4 ) hybridizing extended primers to a universal dna array and determining the identity of the bases that extend each primer by hybridization pattern analysis . to the best of our knowledge the combination of the two technologies in the context of snp genotyping has not been explored thus far .
the most closely related genotyping assay is the generic polymerase extension assay ( pea ) recently proposed in @xcite . in pea ,
short amplicons containing the snps of interest are hybridized to an all @xmath0-mers array of _ primers _ that are subsequently extended via single - base extension reactions . hence ,
in pea the sbe reactions take place on solid support , similar to _ arrayed primer extension _ ( apex ) assays which use snp specific primers spotted on the array @xcite . as in @xcite
, the sbe / sbh assay leads to high array probe utilization since we hybridize to the array a large number of short extended primers .
however , the main power of the method lies in the fact that the sequences of the labeled oligonucleotides hybridized to the array are a priori known ( up to the identity of extending nucleotides ) .
while genotyping with sbe / sbh assays uses similar general principles as the pea assays proposed in @xcite , there are also significant differences .
a major advantage of sbe / sbh is the much shorter length of extended primers compared to that of pcr amplicons used in pea .
a second advantage is that _ all _ probes hybridizing to an extended primer are informative in sbe / sbh assays , regardless of array probe length ( in contrast , only probes hybridizing with a substring containing the snp site are informative in pea assays ) . as shown by the experimental results in section [ sec.results ] these advantages translate into an increase by orders of magnitude in multiplexing rate compared to the results reported in @xcite .
we further note that pea s effectiveness crucially depends on the ability to amplify very short ( preferably 40bp or less ) genomic fragments spanning the snp loci of interest .
this limits the achievable degree of multiplexing in pcr amplification @xcite , making pcr amplification the main bottleneck for pea assays .
full flexibility in picking pcr primers is preserved in sbe / sbh assays .
( a ) ( b ) + ( c ) ( d ) the rest of the paper is organized as follows . in section [ sec.formulations ]
we formalize two problems that arise in genotyping large sets of snps using sbe / sbh assays : the problem of partitioning a set of snps into the minimum number of `` decodable '' subsets , i.e. , subsets of snps that can be unambiguously genotyped using a single sbe / sbh assay , and that of finding a maximum decodable subset of a given set of snps .
we also establish hardness results for the latter problem .
in section [ sec.algos ] we propose several efficient heuristics . finally , in section [ sec.results ] we present experimental results on both randomly generated datasets and instances extracted from the ncbi dbsnp database , exploring achievable tradeoffs between the type / number of array probes and primer length on one hand and number of snps that can be assayed per array on the other .
our results suggest that the sbe / sbh architecture provides a flexible and cost - effective alternative to genotyping assays currently used in the industry , enabling genotyping of up to hundreds of thousands of user - selected snps per assay .
a set of snp loci can be unambiguously genotyped by sbe / sbh if every combination of snp genotypes yields a different hybridization pattern ( defined as the vector of dye colors observed at each array probe ) . to formalize the requirements of unambiguous genotyping ,
let us first consider a simplified sbe / sbh assay consisting of four parallel _ single - color _ sbe / sbh reactions , one for each possible snp allele . under this scenario ,
only one type of dideoxynucleotide is added to each sbe reaction , corresponding to the complement of the tested snp allele .
therefore , a primer is extended in such a reaction if the tested allele is present at the snp locus probed by the primer , and is left un - extended otherwise .
let @xmath5 be the set of primers used in a single - color sbe / sbh reaction involving dideoxynucleotide @xmath6a , c , g , t@xmath7 . from the resulting hybridization pattern we must be able to infer for every @xmath8 whether or not @xmath9 was extended by @xmath10 .
the extension of @xmath9 by @xmath10 will result in a fluorescent signal at all array probes that hybridize with @xmath11 .
however , some of these probes can give a fluorescent signal even when @xmath9 is not extended by @xmath10 , due to hybridization to other extended primers .
since in the worst case _ all _ other primers are extended , it must be the case that at least one of the probes that hybridize to @xmath11 does not hybridize to any other extended primer .
formally , let @xmath12 be the set of array probes .
for every string @xmath13 , let the _ spectrum of @xmath14 in @xmath15 _ , denoted @xmath16 , be the set of probes of @xmath15 that hybridize with @xmath14 .
under the assumption of perfect hybridization , @xmath16 consists of those probes of @xmath15 that are watson - crick complements of substrings of @xmath14 .
then , a set of primers @xmath5 is said to be _
decodable _ with respect to extension @xmath10 if and only if , for every @xmath8 , @xmath17 decoding constraints ( [ 1-color - weak ] ) can be directly extended to 4-color sbe / sbh experiments , in which each type of extending base is labeled by a different fluorescent dye . as before ,
let @xmath5 be the set of primers , and , for each primer @xmath8 , let @xmath18 be the set of possible extensions of @xmath9 , i.e. , watson - crick complements of corresponding snp alleles .
if we assume that any combination of dyes can be detected at an array probe location , unambiguous decoding is guaranteed if , for every @xmath8 and every extending nucleotide @xmath19 , @xmath20 in the following , we refine ( [ 4-color - weak ] ) to improve practical reliability of sbe / sbh assays .
more precisely , we impose additional constraints on the set of probes considered to be _ informative _ for each snp allele .
first , to enable reliable genotyping of genomic samples that contain snp alleles at very different concentrations ( as a result of uneven efficiency in the pcr amplification step or of pooling dna from different individuals ) , we require that a probe that is informative for a certain snp locus must not hybridize to primers corresponding to different snp loci , _ regardless of their extension_. second , since recent studies by naef et al . @xcite suggest that fluorescent dyes can significantly interfere with oligonucleotide hybridization on solid support , possibly destabilizing hybridization to a complementary probe on the array , in this paper we use a conservative approach and require that each probe that is informative for a certain snp allele must hybridize to a strict substring of the corresponding primer . on the other hand ,
informative probes are still required not to hybridize with any other extended primer , even if such hybridizations involve fluorescently labeled nucleotides .
finally , we introduce a _ decoding redundancy _ parameter @xmath21 , and require that each snp have at least @xmath22 informative probes , i.e. , probes that hybridize to the corresponding primer but do not hybridize to any other extended primer .
such a redundancy constraint facilitates reliable genotype calling in the presence of hybridization errors .
clearly , the larger the value of @xmath22 , the more hybridization errors that can be tolerated .
if a simple majority voting scheme is used for making allele calls , the assay can tolerate up to @xmath23 hybridization errors involving the @xmath22 informative probes of each snp .
furthermore , since the informative probes of a snp are required to hybridize _ exclusively _ with the primer corresponding to the snp , the redundancy requirement provides a powerful mechanism for detecting and gauging the extent of hybridization errors .
indeed , each unintended hybridization at an informative probe for a bi - allelic snp has a dye complementary to one of the snp alleles with probability of only 1/2 , and the probability that @xmath0 such errors pass undetected decreases exponentially in @xmath0 .
the refined set of constraints is captured by the following definition , where , for every primer @xmath24 and set of extensions @xmath25 , we let @xmath26 [ def.strong-primer ] a set of primers @xmath5 is said to be _ strongly @xmath22-decodable _ with respect to extension sets @xmath27 , @xmath8 , if and only if , for every @xmath8 , @xmath28 note that testing whether or not a given set of primers is strongly @xmath22-decodable can be easily accomplished in time linear in the total length of the primers . genotyping a large set of snps
will , in general , require more than one sbe / sbh assay .
this rises the problem of partitioning a given set of snps into the smallest number of subsets that can each be genotyped using a single sbe / sbh assay . for each snp locus
there are typically two different primers that can be used for genotyping . as shown in @xcite for the case of snp genotyping using tag arrays , exploiting this degree of freedom significantly increases achievable multiplexing rates .
therefore , we next extend our definitions to capture this degree of freedom .
let @xmath29 be the _ pool of primers _ that can be used to genotype the snp at locus @xmath30 .
similarly to definition [ def.strong-primer ] , we have : [ def.strong-pool ] a set of primer pools @xmath31 is said to be _ strongly @xmath22-decodable _ if and only if there is a primer @xmath32 in each pool @xmath29 such that @xmath33 is strongly @xmath22-decodable with respect to the respective extension sets @xmath34 , @xmath35
. primers @xmath36 above are called the _ representative primers _ of pools @xmath37 , respectively .
the snp partitioning problem can then be formulated as follows : * minimum pool partitioning problem ( mppp ) : * _ given primer pools @xmath31 , associated extension sets @xmath27 , @xmath38 , probe set @xmath15 , and redundancy @xmath22 , find a partitioning of @xmath5 into the minimum number of strongly @xmath22-decodable subsets .
_ a natural strategy for solving mppp , similar to the well - known greedy algorithm for the set cover problem , is to find a maximum strongly @xmath22-decodable subset of pools , remove it from @xmath5 , and then repeat the procedure until no more pools are left in @xmath5 .
this greedy strategy for solving mppp has been shown to empirically outperform other algorithms for solving the similar partitioning problem for pea assays @xcite . in the case of sbe / sbh ,
the optimization involved in the main step of the greedy strategy is formalized as follows : * maximum @xmath22-decodable pool subset problem ( mdpsp ) : * _ given primer pools @xmath31 , associated extension sets @xmath27 , @xmath38 , probe set @xmath15 , and redundancy @xmath22 , find a strongly @xmath22-decodable subset @xmath39 of maximum size .
in addition , for each pool @xmath40 , find its representative primer . _ unfortunately , as shown in next theorem , mdpsp is np - hard even for the case when the redundancy parameter is 1 and each pool has exactly one primer .
[ theorem.mdpsphard ] mdpsp is np - hard , even when restricted to instances with @xmath41 and @xmath42 for every @xmath43 .
we will use a reduction from the _ maximum induced matching _ problem in bipartite graphs , which is defined as follows : * maximum induced matching ( mim ) problem in bipartite graphs : * _ given a bipartite graph @xmath44 , find maximum size subsets @xmath45 , @xmath46 , with @xmath47 such that the subgraph of @xmath48 induced by @xmath49 is a matching . _
the mim problem in bipartite graphs is known to be np - hard even for graphs with maximum degree 3 @xcite .
let @xmath44 be such a bipartite graph with maximum degree 3 .
without loss of generality we may assume that every vertex in @xmath48 has degree at least 1 .
we will denote by @xmath50 the _ neighborhood _ of vertex @xmath51 , i.e. , the set of vertices adjacent with @xmath52 in @xmath48 .
we construct an instance of mdpsp as follows : let @xmath41 and @xmath53 . for every @xmath54
we add to @xmath15 a distinct probe @xmath55a , t@xmath56 ; note that this can be done since @xmath57a , t@xmath58 by our choice of @xmath59 . for every @xmath60 , with neighborhood @xmath61 , we construct a primer @xmath62c@xmath63c@xmath64 and set @xmath65 .
we use a similar construction for vertices @xmath66 with only 1 or 2 neighbors . note that in each case the pool @xmath67 consists of a single primer @xmath68 of length at most @xmath69 . for each constructed primer @xmath9 ,
the set of possible extensions is defined as @xmath70g , c@xmath7 . since the probes of @xmath15 contain only a s and t s , for every primer @xmath68 , @xmath66 , @xmath71 let @xmath45 , @xmath46 , @xmath47 , be subsets of vertices such that @xmath49 induces a matching in @xmath48 .
let @xmath72 .
for every @xmath73 , exactly one of @xmath52 s neighbors , denoted @xmath74 , appears in @xmath75 , because @xmath76 induces a matching .
furthermore , for each @xmath77 , @xmath78 , and therefore @xmath79 .
thus , for every @xmath73 , @xmath80 which means that @xmath81 is a strongly 1-decodable subset of pools of the same size as the induced matching of @xmath48 .
conversely , let @xmath81 be a strongly 1-decodable subset of @xmath5 , and let @xmath82 .
since @xmath81 is 1-decodable , for every primer @xmath68 with @xmath83 , there must exist a probe @xmath84 such that @xmath85 and @xmath86 for every @xmath87 . because @xmath88 , it follows that every vertex @xmath89 has a neighbor @xmath54 that is not a neighbor of any other @xmath77 .
let @xmath90 be such a neighbor ( pick @xmath90 arbitrarily if more than one vertex in @xmath91 satisfies above property ) , and let @xmath92 .
it is clear that @xmath76 induce a matching of size @xmath93 in @xmath48 .
thus , for every integer @xmath0 , there is a one - to - one correspondence between induced matchings of size @xmath0 in @xmath48 and strongly 1-decodable subsets of @xmath0 pools in the constructed instance of mdpsp , and np - hardness of mdpsp follows .
the reduction in the proof of theorem [ theorem.mdpsphard ] preserves the size of the optimal solution , and therefore any hardness of approximation result for the mim in bipartite graphs will also hold for mdpsp , even when restricted to instances with @xmath41 and @xmath42 for every @xmath43 . since duckworth et al .
@xcite proved that it is np - hard to approximate mim in bipartite graphs with maximum degree 3 within a factor of 6600/6659 , we get : [ corol.apx-hard ] it is np - hard to approximate mdpsp within a factor of 6600/6659 , even when restricted to instances with @xmath41 and @xmath42 for every @xmath43 .
in this section we describe three heuristic approaches to mdpsp .
the first one is a naive greedy algorithm that sequentially evaluates the primers in the given pools in an arbitrary order .
the algorithm picks a primer @xmath9 to be the representative of pool @xmath43 if @xmath9 together with the representatives already picked satisfy condition ( [ 4-color - strong ] ) .
the pseudocode of this algorithm , which we refer to as sequential greedy , is given in figure [ fig.seq ] .
the next two algorithms are inspired by the min - greedy algorithm in @xcite , which approximates mim in @xmath94-regular graphs within a factor of @xmath95 .
for the mim problem , the min - greedy algorithm picks at each step a vertex @xmath52 of minimum degree and a vertex @xmath96 , which is a minimum degree neighbor of @xmath52 .
all the neighbors of @xmath52 and @xmath96 are deleted and the edge @xmath97 is added to the induced matching .
the algorithm stops when the graph becomes empty .
each instance of mdpsp can be represented as a bipartite _ hybridization graph _ @xmath98 , with the left side containing all primers in the given pools and the right side containing the array probes , i.e. , @xmath15 .
there is an edge between primer @xmath9 and probe @xmath99 iff @xmath100 . as discussed in section [ sec.formulations ] , we need to distinguish between the hybridizations that involve fluorescently labeled nucleotides and those that do not .
thus , for every primer @xmath9 , we let @xmath101 and @xmath102 . similarly , for each probe @xmath99 , we let @xmath103 and @xmath104 . we considered two versions of the min - greedy algorithm when
run on the bipartite hybridization graph , depending on the side from which the minimum degree vertex is picked . in the first version , referred to as minprimergreedy , we pick first a minimum degree node from the primers side , while in the second version , referred to as minprobegreedy , we pick first a minimum degree node from the probes side .
thus , minprimergreedy greedy picks at each step a minimum degree primer @xmath9 and pairs it with a minimum degree probe @xmath105 .
minprobegreedy greedy , selects at each step a minimum degree probe @xmath106 and pairs it with a minimum degree primer @xmath9 in @xmath107 . in both algorithms , all neighbors of @xmath9 and @xmath106 and
their incident edges are removed from @xmath48 . also , at each step , the algorithms remove all vertices @xmath52 , for which @xmath108 .
these deletions ensure that the primers @xmath9 selected at each step satisfy condition ( [ 4-color - strong ] ) .
both algorithms stop when the graph becomes empty . as described so far , the minprimergreedy and minprobegreedy algorithms work when each pool contains only one primer and when the redundancy is 1 .
we extended the two variants to handle pools of size greater than 1 by simply removing from the graph all primers @xmath109 when picking primer @xmath9 from pool @xmath110 .
if the redundancy @xmath22 is greater than 1 , then whenever we pick a primer @xmath9 , we also pick it s @xmath22 probe neighbors from @xmath111 with the smallest degrees ( breaking ties arbitrarily ) .
the primer neighbors of all these @xmath22 probes will then be deleted from the graph . moreover
, the algorithm maintains the invariant that @xmath112 for every primer @xmath9 and @xmath113 for every probe @xmath106 by removing primers / probes for which the degree decreases below these bounds .
figures [ fig.min-primer ] and [ fig.min-probe ] give the pseudocode for the minprimergreedy , respectively the minprobegreedy greedy algorithms . for the sake of clarity , they use two subroutines for removing a primer vertex , respectively a probe vertex , which are described in figures [ fig.remove-primer ] and [ fig.remove-probe ] . algorithms minprimergreedy and minprobegreedy can be implemented efficiently using a fibonacci heap for maintaining the degrees of primers , respectively of probes .
let @xmath114 be the total number of primers in the @xmath115 pools , @xmath116 be the number of probes in @xmath15 , and @xmath0 be the size of the @xmath22-decodable set returned by the algorithm .
since each primer has bounded degree , the sorting of probe degrees requires @xmath117 total time .
the total number of edges in the hybridization graph is @xmath118 . by using a fibonacci heap , finding a minimum degree primer ( probe )
can be done in @xmath119 ( respectively @xmath120 ) and each primer degree update can be done in amortized @xmath121 time .
thus , the total runtime for minprimergreedy algorithm is @xmath122 , and the total runtime for minprobegreedy algorithm is @xmath123 .
we considered two types of data sets : * randomly generated datasets containing between 1,000 to 200,000 pools with 1 or 2 primers of length between 10 and 30 . * two - primer pools representing over 9 million reference snps in human chromosomes 1 - 22 ,
x , and y extracted from the ncbi dbsnp database build 125 .
we disregarded reference snps for which available flanking sequence was insufficient for determining two non - degenerate primers of desired length ( due , e.g. , to the presence of degenerate bases near the snp locus ) .
we used two types of array probe sets .
first , we used probe sets containing all @xmath0-mers , for @xmath0 between 8 and 10 .
all @xmath0-mer arrays are well studied in the context of sequencing by hybridization . however , a major drawback of all @xmath0-mer arrays is that the @xmath0-mers have a wide range of melting temperatures , making it difficult to ensure reliable hybridization results . for short oligonucleotides , a good approximation of the melting temperature
is obtained using the simple 2 - 4 rule of wallace @xcite , according to which the melting temperature of a probe is approximately twice the number of a and t bases , plus four times the number of c and g bases . as in @xcite , we define the _ weight _ of a dna string to be the number of a and t bases plus twice the number of c and g bases . for a given integer @xmath124 , a dna string is called a @xmath124-token if it has a weight @xmath124 or more and all its proper suffixes have weight strictly less than @xmath124 . since the weight of a @xmath124-token is either @xmath124 or @xmath125 , it follows that the 2 - 4 rule computed melting temperature of all @xmath124-tokens varies in a range of about @xmath126c . in our experiments we used probe sets consisting of all @xmath124-tokens , with @xmath124 varying between 11 and 13 .
the considered values of @xmath0 and @xmath124 were picked such that the resulting number of probes is representative of current array manufacturing technologies : there are roughly 65,000 8-mers , 262,000 9-mers , 1 million 10-mers , 86,000 11-tokens , 236,000 12-tokens , and 645,000 13-tokens the smaller probe sets can be spotted using current oligonucleotide printing robots , while the larger probe sets can be synthesized in situ using photolithographic techniques . in a first set of experiments on the randomly generated datasets we compared the three mdpsp algorithms on instances with primer length set to 20 , which is the typical length used , e.g. , in genotyping using tag arrays . in these experiments
the set of possible extensions was considered to be @xmath127a , c , t , g@xmath7 for all primers .
such a conservative choice gives an estimate of multiplexing rates achievable by sbe / sbh assays in more demanding genomic analyses such as microorganism identification by dna barcoding @xcite , in which a primer ( typically referred to as a _ distinguisher _ in this context ) may be extended by any of the dna bases in different microorganisms .
the results of these experiments for all @xmath0-mer and all @xmath124-token probe sets are presented in tables [ table.set1.k ] and [ table.set1.c ] , respectively .
the results show that using the flexibility of picking primers from either strand of the genomic sequence yields an improvement of up to 10% in the number of @xmath22-decodable pools .
the minprobegreedy algorithm typically produces better results compared to the minprimergreedy variant . on the other hand , neither sequential greedy nor minprobegreedy dominates the other algorithms for all range of instance parameters sequential greedy generally gives the better results for @xmath0-mer experiments with high redundancy values , while minprobegreedy generally gives better results for @xmath0-mer experiments with large number of pools and low redundancy and for @xmath124-token experiments . in the second set of experiments we ran the three mdpsp algorithms on datasets with the same primer length of 20 , pool size of 2 , and with the number of possible extensions of each primer set to 4 as in dna - barcoding applications , and to 2 as in snp genotyping .
the results for all @xmath0-mer and all @xmath124-token probe sets are given in tables [ table.set2.k ] and [ table.set2.c ] .
the relative performance of the algorithms is similar to that observed in the first set of experiments .
as expected , taking into account the reduced number of possible extensions increases the size of computed decodable pool subsets , often by more than 5% . in the third set of experiments we explored the degree of freedom given by the primer length .
for any fixed array probe set and redundancy requirement , we need a minimum primer length to be able to satisfy constraints ( [ 4-color - strong ] ) .
increasing the primer length beyond this minimum primer length is often beneficial , as it increases the number of array probes that hybridize with the primer .
however , if primer length increases too much , an increasing number of these probes become non - specific , and the multiplexing rate starts to decline .
figure [ fig.length ] gives the tradeoff between primer length and the size of the strongly @xmath22-decodable pool subsets computed by the three mdpsp algorithms for pools with 2 primers , 2 possible extensions per primer and all 10-mers , respectively all 13-tokens , as array probes .
we notice that the optimal primer length increases with the redundancy parameter .
.[table.set1.k ] size of the strongly @xmath22-decodable pool subset computed by the three mdpsp algorithms for primer length 20 and set of possible extensions @xmath127a , c , t , g@xmath7 , with redundancy @xmath128 and all @xmath0-mer probe sets for @xmath129 ( averages over 10 test cases ) .
[ cols="^,^,^,^,^,^,^,^,^ " , ] a. bendor , t. hartman , b. schwikowski , r. sharan , and z. yakhini . towards
optimally multiplexed applications of universal dna tag systems . in _ proc .
7th annual international conference on research in computational molecular biology _ , pages 4856 , 2003 .
heath and f.p . preparata . enhanced sequence reconstruction with dna microarray application . in _ proc .
7th annual international conference on computing and combinatorics ( cocoon ) _ , pages 6474 , 2001 .
hirschhorn , p. sklar , k. lindblad - toh , y .- m .
lim , m. ruiz - gutierrez , s. bolk , b. langhorst , s. schaffner , e. winchester , and e. lander .
: an array - based method for efficient single - nucleotide polymorphism genotyping . , 97(22):1216412169 , 2000 .
konwar , i.i .
mndoiu , a.c .
russell , and a.a .
improved algorithms for multiplex pcr primer set selection with amplification length constraints . in y .-
phoebe chen and l. wong , editors , _ proc .
3rd asia - pacific bioinformatics conference ( apbc ) _ , pages 4150 , london , 2005 .
imperial college press .
n. tonisson , a. kurg , e. lohmussaar , and a. metspalu .
arrayed primer extension on the dna chip - method and application . in mark
schena , editor , _ microarray biochip technology _ , pages 247263 .
eaton publishing , 2000 .
wallace , j. shaffer , r.f .
murphy , j. bonner , t. hirose , and k. itakura .
hybridization of synthetic oligodeoxyribonucleotides to phi chi 174 dna : the effect of single base pair mismatch .
, 6(11):63536357 , 1979 . | despite much progress over the past decade , current single nucleotide polymorphism ( snp ) genotyping technologies still offer an insufficient degree of multiplexing when required to handle user - selected sets of snps . in this paper
we propose a new genotyping assay architecture combining multiplexed solution - phase single - base extension ( sbe ) reactions with sequencing by hybridization ( sbh ) using universal dna arrays such as all @xmath0-mer arrays .
in addition to pcr amplification of genomic dna , snp genotyping using sbe / sbh assays involves the following steps : ( 1 ) synthesizing primers complementing the genomic sequence immediately preceding snps of interest ; ( 2 ) hybridizing these primers with the genomic dna ; ( 3 ) extending each primer by a single base using polymerase enzyme and dideoxynucleotides labeled with 4 different fluorescent dyes ; and finally ( 4 ) hybridizing extended primers to a universal dna array and determining the identity of the bases that extend each primer by hybridization pattern analysis . under the assumption of perfect hybridization , unambiguous genotyping of a set of snps requires selecting primers upstream of the snps such that each primer hybridizes to at least one array probe that hybridizes to no other primer that can be extended by a common base .
our contributions include a study of multiplexing algorithms for sbe / sbh genotyping assays and preliminary experimental results showing the achievable tradeoffs between the number of array probes and primer length on one hand and the number of snps that can be assayed simultaneously on the other .
we prove that the problem of selecting a maximum size subset of snps that can be unambiguously genotyped in a single sbe / sbh assay is np - hard , and propose efficient heuristics with good practical performance .
our heuristics take into account the freedom of selecting primers from both strands of the genomic dna as well as the presence of disjoint allele sets among genotyped snps .
in addition , our heuristics can enforce user - specified redundancy constraints facilitating reliable genotyping in the presence of hybridization errors .
simulation results on datasets both randomly generated and extracted from the ncbi dbsnp database suggest that the sbe / sbh architecture provides a flexible and cost - effective alternative to genotyping assays currently used in the industry , enabling genotyping of up to hundreds of thousands of user - specified snps per assay . |
based on the fact that the interaction between the @xmath3 meson and nucleon is attractive , haider and liu postulated the existence of a new kind of matter in form of @xmath3-nucleus bound states @xcite .
because the @xmath3 meson is neutral such a system can be formed only via the strong interaction which distinguishes it qualitatively from exotic atoms where the meson is bound via the electromagnetic force . therefore the search for a signature of an @xmath3-mesic nucleus is interesting on its own account , additionally the experimental determination of the width and binding energy of @xmath3-mesic nuclei would provide a valuable input for studies of the @xmath4 interaction .
it would provide information about properties of the @xmath5 embedded in nuclei , the behavior of the @xmath3 meson in the nuclear medium @xcite , and for the determination of the flavor singlet component of the @xmath3 meson @xcite .
the formation of a bound state can only take place in nuclei , for which the real part of the @xmath3nucleus scattering length is negative ( attractive nature of the @xmath3nucleus interaction ) , and the modulus of the real part of @xmath3nucleus scattering length is greater than the modulus of its imaginary part @xcite : @xmath6 the relatively small value of the s - wave @xmath3n scattering length known in 1980 s ( @xmath7 fm @xcite ) implied the possibility of forming an @xmath3mesic nuclei only for a @xmath8 12 @xcite .
this estimation was strengthened by calculations of li @xcite .
however , recent theoretical considerations of hadronic- and photoproduction of the @xmath3 meson result in a wide range of possible values of the @xmath3nucleon s - wave scattering lengths from @xmath9 fm up to @xmath10 fm , with the suggested average value of @xmath11 fm
. such a high value of the @xmath3nucleon scattering length may enable the formation of a bound @xmath3nucleus states in light nuclei region such as @xmath12 @xcite and even in the deuteron @xcite . according to the calculations including multiple scattering theory @xcite and skyrme model
@xcite an especially good candidate for binding is the @xmath13 system .
it should be stated that at present there is no univocal picture of this issue and results presented in article @xcite indicate that the ratio of the predicted width to the binding energy becomes bigger for lighter nuclei which can make an observation of the bound state in light nuclei more difficult .
however , there are promissing indirect experimental observations which may be interpreted as an indications of @xmath3-helium bound states .
for example analysis of the data from the close to threshold measurements of the total cross section for the @xmath14 reaction by spes-4 @xcite and spes-2 @xcite collaborations suggested a possible existence of the bound @xmath15 system since within the experimental errors the condition given in eq . [
11 ] could be fulfilled .
a search for a @xmath3nucleus bound state has also been performed in the hadronic channel at the cooler synchrotron cosy , where the cosy-11 and anke collaborations independently , using different detection setups , performed measurements of the excitation function and differential cross sections for the @xmath14 reaction in the vicinity of the kinematical threshold @xcite .
both groups used the momentum ramping technique of the beam deuterons in order to reduce the systematic errors .
measurements have been performed with beam momenta varying from below the reaction threshold , up to an excess energy of about 8.5 mev in the case of the cosy-11 experiment and about 11.5 mev in the case of the anke experiment .
data taken by the cosy-11 group were used to search for a signal of a @xmath16 bound state below the @xmath3 production threshold , via the @xmath17 and @xmath18 reactions @xcite , while the measurements above the threshold enabled the study of the forward - backward asymmetries of the differential cross sections and the extraction of the @xmath19 scattering length .
the data of both groups @xcite shows a variation in the phase of the s - wave amplitude in the near - threshold region , consistent with possible existence of a bound state @xcite .
the first direct experimental indications of a light @xmath3-nucleus bound system was reported by the taps collaboration in the @xmath3 photoproduction reaction @xmath20 @xcite . there
the difference between excitation functions for two ranges of the @xmath21 relative angle in the center - of - mass frame revealed a structure which was interpreted as a possible signature of a bound state .
however , it was claimed that due to the limited statistics the result could also be interpreted as an evidence for a virtual state @xcite .
therefore a new , statistically more significant experimental confirmation is needed .
the experimental method is based on a measurement of the excitation function for the chosen decay channels of the @xmath22 system and the search for a resonance - like structure below the @xmath22 threshold .
the measurement was performed by applying a continuous variation of the beam momentum around the threshold and @xmath23 meson production in @xmath24 and @xmath25 reactions @xcite ] . the relative angle between the outgoing @xmath26 pair which originates from the decay of the n*(1535 ) resonance created via absorption of the @xmath3 meson on a nucleon in the @xmath27 nucleus , is equal to 180@xmath28 in the @xmath29 reference frame .
it is smeared by about 30@xmath28 in the center - of - mass frame ( see fig . [ angle ] ) due to the fermi motion of the nucleons inside the @xmath27 nucleus .
the center - of - mass kinetic energies of the nucleon and pion originate from the mass difference @xmath30 and are around 50 mev and 350 mev , respectively .
[ angle ] shows that the distribution of the relative @xmath31 angle expected for the background due to the prompt @xmath32 reaction is much broader than the one expected from the decay of the bound state .
this will allow to control the background by comparing `` signal - rich '' and `` signal - poor '' regions .
the experiment was carried out with the wasa - at - cosy detector , and with the internal deutron beam of cosy scattered on a deuteron pellet target .
a detailed detector description can be found in @xcite .
we used a slowly ramped beam momentum scanning the range of momenta corresponding to a variation of the excess energy for the @xmath2he@xmath3 system from -51.4 mev to 22 mev .
this range is about three times wider than the width of the structure observed by the taps collaboration at mami in the @xmath20 process ( @xmath33 ) mev .
the charged pions and protons are registered in the forward detector as well as in the central detector .
@xmath34-ions are detected in the forward detector only .
[ ptheta ] presents an example of the angular distributions of ejectiles expected for the @xmath35 reaction .
the simulations were conducted under the assumption that the decay of the bound state is caused by the decay of the @xmath29 into a proton @xmath36 pair leaving the @xmath37 nucleus as a spectator with the fermi momentum which it possessed inside the @xmath38 nucleus . the fig .
[ ptheta ] shows that a significant fraction of the angular range of the outgoing particles is covered by the wasa - at - cosy detector setup .
in particular , about 80% of @xmath37 ions can be registered by means of the forward detector ( left panel of fig .
[ ptheta ] ) and a coincident measurement of all ejectiles from @xmath35 reaction can be performed with an efficency of 70 % which is flat in the whole range of the excess energy from -51.4 mev to 22 mev .
in june 2008 a test measurement of the production of @xmath39 , @xmath40 as well as of @xmath41 and @xmath42 in the dd collisions with the wasa - at - cosy facility was performed .
the analysis of the energy signals from the forward detector revealed that it is possible to identify clearly the @xmath37 ions . in fig.[de ] the experimental distribution of the energy loss in the window counter versus the energy deposited in the range hodoscope is compared to the expectation based on monte - carlo simulations .
unfortunately , at the time of the experiment the cooling system of the superconducting solenoid was not working and we had to perform the measurement without magnetic field . that fact excluded the possibility of a direct momentum determination in the central detector as well as the use of the standard wasa - at - cosy identification method for the charged particles registered in the central detector .
however , for the momentum determination of the charged particles in the central detector the information about directions of the particles in the central detector , as well as , the momentum of the @xmath27 nuclei can be used .
the absolute @xmath43 momentum was determined on the basis of energy losses in the forward detector and its direction was reconstructed from signals in the set of drift chambers built out of straw detectors .
the directions of charged particles in central detector were extracted from signals in the 17 layers of the central drift chamber and from the electromagnetic calorimeter . knowing the beam momentum , the @xmath43 momentum , and the directions of the two other particles one can derive the momenta of pion and proton assuming energy and momentum conservation ( see fig .
[ momrecovstrue ] ) .
energy loss in the plastic scintillator combined with the energy deposited in the electromagnetic calorimeter was used to identify protons and pions . at present
we have established the excitation function for the @xmath44 reaction only .
the result may be considered as a very conservative estimation of the upper limit for the prompt @xmath45 reaction .
the excitation function obtained is presented in fig .
[ excitations ] .
please note that the shape of the angular distribution ( right panel ) is consistent with the expectations shown in fig .
[ angle](right ) . in the left panel of fig .
[ excitations ] the excitation function for the angular range between 160 and 200 degrees is presented .
we conduct a search for the @xmath46 bound state with the wasa - at - cosy facility . because of the very high acceptance of the proton - pion pairs and good identification of helium nuclei the detector system is very well suited for this kind of experiment . in june 2008
we performed a first measurement of the excitation functions for the @xmath47 reaction .
the experiment will be continued in the year 2010 .
two weeks of cosy beamtime were already recommended by the cosy program advisory committee . the use of the magnetic field will permit better proton - pion identification , as well as a better momentum determination in the central detector . in two weeks of planned measurement , with the assumed luminosity of , we expect to reach a sensitivity in the order of a few nb with a statistical significance of 1 @xmath48 . in the case of no observation of a signal this result will significantly lower the upper limit for the existence of a bound state .
the work was supported by the european community - research infrastructure activity under the fp6 program ( hadron physics , rii3-ct-2004 - 506078 ) , by the german research foundation ( dfg ) , by the polish ministry of science and higher education through grants no . 3240/h03/2006/31 and 1202/dfg/2007/03 , and by the ffe grants from the research center jlich .
abc q. haider , l.c .
liu , _ phys .
lett . _ * b172 * , 257 ( 1986 ) .
t. inoue , e. oset , _ nucl .
phys . _ * a710 * , 354 ( 2002 ) . s. d. bass , a. w. thomas , _ phys .
* b634 * , 368 ( 2006 ) .
q. haider , l.c .
liu , _ phys .
lett . _ * c66 * , 045208 ( 2002 ) .
bhalerao , l.c .
liu , _ phys .
lett . _ * 54 * , 865 ( 1985 ) .
li _ et al .
_ , _ phys .
* b195 * , 515 ( 1987 ) . c. wilkin , _ phys
* c47 * , r938 ( 1993 ) .
s. wycech , a. m. green and j. a. niskanen , _ phys .
* c52 * , 544 ( 1995 ) .
a.m. green _ et al .
_ , _ phys .
rev . _ * c54 * , 1970 ( 1996 ) .
n. n. scoccola and d. o. riska , _ phys .
_ * b444 * , 21 ( 1998 ) . c. garcia - recio , j. nieves , t. inoue , e. oset , _ phys .
* b550 * , 47 ( 2002 ) .
j. berger , _ et al .
_ , _ phys .
lett . _ * 61 * , 919 ( 1988 ) .
b. mayer , _ et al .
_ , _ phys .
_ * c53 * , 2068 ( 1996 ) .
j. smyrski _ et al .
_ , _ phys .
* b 649 * , 258 ( 2007 ) .
adam _ et al .
_ , _ phys .
rev . _ * c75 * , 014004 ( 2007 ) . t. mersmann _ et al .
_ , _ phys .
* 98 * , 242301 ( 2007 ) .
j. smyrski _ et al .
_ , _ acta phys
. slovaca _ * 56 * , 213 ( 2006 ) .
w. krzemie _ et al .
_ , arxiv:0810.2330 , _ int .
phys . _ * a * ( 2009 ) in print .
j. smyrski _ et al .
* a790 * , 438 ( 2007 ) . c. wilkin , _ phys .
_ * b654 * , 92 ( 2007 ) . c. wilkin , _ aip conf .
proc . _ * 950 * , 23 ( 2007 ) .
m. pfeiffer , _ et al .
_ , _ phys .
lett . _ * 92 * , 252001 ( 2004 ) . c. hanhart , _ phys .
lett . _ * 94 * , 049101 ( 2005 ) .
j. smyrski _ et al .
_ , _ phys .
lett . _ * b474 * , 182 ( 2000 ) .
p. moskal , _ et al .
lett . _ * 80 * , 3202 ( 1998 ) .
wasa - at - cosy collaboration : h.h .
adam _ et al .
_ , arxiv : nucl - ex/0411038 .
redmer , m. janusz , these proceedings .
tryasuchev , a.v .
isaev , arxiv : nucl - th/0901.3242 . | we conduct a search for the @xmath0 bound state with the wasa - at - cosy facility via the measurement of the excitation function for the reaction @xmath1 . in first experiment
performed in june 2008 , we used cosy deuteron beam with a slowly ramped beam momentum corresponding to a variation of the excess energy for the @xmath2he@xmath3 system from -51.4 mev to 22 mev . here
we report on the status of the measurement and the data evaluation . |
further precise measurement of the anomalous magnetic dipole moment of the muon , conventionally denoted as @xmath2 , is now underway at brookhaven national laboratory ( bnl ) .
the perspective for the goal of this experiment is @xcite @xmath3 the recent report for its test - running course at bnl @xcite combined with the previous one at cern @xcite gives to muon @xmath0 @xmath4 where the numerals in the parenthesis represent the uncertainty in the final few digits .
thus the precision ( [ eq : future - precision ] ) amounts to the determination of its value by one further digit .
+ our primary interest is what we can learn when invoking such a improvement in muon @xmath0 experiment . at present
the standard model predicts @xmath5 which includes the up - dated estimate on the leading hadronic vacuum polarization contribution @xcite , ( see @xcite due to analysis without recourse to @xmath6 decay data ) , and @xmath7 contribution @xcite . the designed precision ( [ eq : future - precision ] ) is put forward to find out the existence of the @xmath8 , @xmath1 boson effect to muon @xmath0 @xcite @xmath9 the electron @xmath0 does not receive observable effects from @xmath1 and @xmath8 bosons and are entirely saturated by qed effect
this fact enables us to find out the validity of calculation scheme of quantum field theory , including the perturbative renormalization procedure .
rather electron @xmath0 provides the most accurate determination of the fine structure constant @xmath10 @xcite at present . from ( [ eq : weak_effect ] ) and ( [ eq : future - precision ] ) , the muon @xmath0 is expected not only to obtain a conceivable evidence for such a structure about the electroweak interactions as involved in standard model , but also has the potential to indicate the existence of much richer ingredient associated with some theoretical problem .
+ when we incline to use muon @xmath0 as a probe of new physics , the theme of the talk assigned to me as well , it would be important to recall the motivation or the merits of each model .
thus , here , i will focus on two concrete models considered to approach to hierarchy problem , although it is somewhat a conceptual viewpoint . here hierarchy problem
stands for the following question in some narrow sense ; what is assuring the stability of the electroweak scale , represented by @xmath8 boson mass , @xmath11 , against quantum fluctuation associated with high momentum modes below some cutoff scale .
the cutoff scale here is the scale at which the gauge interactions appearing in standard model would become subject to some kind of modification
. it may be the gut scale @xmath12 , at which the standard model gauge symmetry group is merged into a larger symmetry group .
or it may be the scale at which the gauge boson is resolved into more fundamental structure , for instance , into string .
+ here we take up two models considered with such a motivation .
one is tev scale gravity discussed in the next section , and the other supersymmetry in sec .
[ sec : susy ] .
[ sec : discussion ] concludes with remark on several facets for muon @xmath0 to probe new physics practically .
the laboratory experiments check the structure of gravity has been met up to the order of millimeters . with this in mind let us turn our attention to the scheme introduced a couple of years ago @xcite
. + let us imagine that our world is confined in a three - brane , the extended object with three - spatial directions , flowing in a higher dimensional space ( @xmath13-dimension in total ) .
these @xmath14-directions are compactified to an @xmath14-dimensional torus with the same length @xmath15 , for simplicity .
then there are an infinite tower of kaluza - klein states from the four - dimensional space - time point of view .
they are the states with non - zero momentum in the extra directions . in four - dimensional world
such a momentum appears as the mass , the scale of which is characterized by inverse of @xmath16 .
+ the behavior of static potential between two point - like sources prepared with separation @xmath17 illustrates an important aspect of the present setting @xcite . when @xmath17 is large compared to @xmath16 , the potential behaves as @xmath18 , nothing but the form of the usual newtonian one .
the planck scale @xmath19 characterizes its strength .
however , once the separation @xmath17 reaches below @xmath16 , the probability that kaluza - klein states get excited can not become neglected any more . thus counting those states which propagate essentially like massless states between two sources leads @xmath20 as @xmath17-dependence of the potential at short distance , characteristic of @xmath21-spatial dimension .
this only reflects the fact that the local structure less than the compactification scale is that of @xmath13-dimensional space - time .
a noteworthy point is that the strength of force is then characterized by another planck scale , @xmath22 ; @xmath23 this is the strength of the gravitational interaction in the bulk theory , the fundamental planck scale .
+ now we take @xmath16 equal to 1 mm , through @xmath24 ] which corresponds to the length scale one - order less than the current reach of the experiment on gravity . with two extra compactified directions
@xmath22 = 1 tev .
= 1 tev , which he heard at the seminar by arkani - hamed @xcite . ]
+ if the three - brane were further of dirichlet - brane type , on which the open string can attach , our standard model gauge bosons would become the tangential component of the ground state of such an open string .
then the gauge bosons would get resolve into strings higher than 1 tev as long as the string coupling constant is of order unity .
thus there is no hierarchy problem ab initio .
+ at present we are lacking the precise formulation of theories and the detail knowledge on its dynamical aspects ( especially on the compactification mechanism ) .
however the dimension counting argument with symmetry consideration has been an enough tool when we estimate the order of magnitude about some effect in low energy phenomena unless the effective theory description breaks down , although we rather expect to superstring theory that many miracles beyond this assumption occur .
+ the argument begins with the chiral symmetry for muon .
when muon was massless , the magnetic dipole coupling would be absent .
thus the magnitude of magnetic moment will be proportional to muon mass @xmath25 and the corresponding operator appears in the effective lagrangian : @xmath26 with the the extra effect @xmath27 due to the new structure characterized by mass scale @xmath22 . since the mass dimension of @xmath27 turns out to be minus one in four space - time dimension , the dimension counting now gives @xmath28 with the numerical constant @xmath29 of order unity .
insertion of ( [ eq : a_const ] ) into the effective interaction ( [ eq : muon_dipole ] ) with a slight rearrangement yields @xmath30\ , \bar{\mu } i\sigma^{\lambda\rho } f_{\lambda\rho}\,\mu.\ ] ] the quantity found in the square bracket of the above expression corresponds to the additional contribution to @xmath31 due to the presence of tev scale gravity .
thus the additional effect to @xmath31 can be read off as @xmath32 which becomes for @xmath22 = 1 tev @xmath33 thus a crude estimate shows that the effect from tev scale gravity is in the marginal situation to be detected even with the current accuracy ( [ eq : present_value ] ) .
the future accuracy ( [ eq : future - precision ] ) is quite adequate to detect the existence of new aspect of gravity characterized by tev scale .
now we will take our attention to the search of supersymmetry with the use of muon @xmath0 , which has been discussed as a machinary to check the various adovocated models or from some generic standpoint @xcite .
the detail formula and so on in this section will also be found in a separate literature @xcite .
+ in the narrow sense of the hierarchy problem defined in the introductory remark , supersymmetry makes it possible to extend the standard model gauge group into the larger gauge group in grand unified theory ( gut ) .
this is realized in the form of the cancellation of the dangerous quantum corrections within each supermultiplet .
the bosonic partner of muon , for instance , has not been observed yet so that it must be much heavier .
smuon can be let heavier by giving it a lifting - up mass , @xmath34 .
this procedure does not spoil quantum stability as far as @xmath34 is not so apart from the electroweak scale .
the scale @xmath34 works as the ultraviolet cutoff scale for the phenomena accessed by low momentum probe while it works as the infrared cutoff from the supersymmetric high - energy side .
it has been recognized that unification of the coupling constants of three gauge interactions , @xmath35 , @xmath36 and color @xmath37 , is achieved by assuming the particle content of the minimal supersymmetric extension of standard model above such @xmath34 . for the future purposes it deserves to recall here that the knowledge from precise measurement around @xmath1 pole plays an indispensable role to establish this fact .
+ now we examine the effect on muon @xmath0 induced from supersymmetric theories .
they come essentially from two diagrams .
one is the chargino - sneutrino loop , where the charginos are the admixture of @xmath38 gaugino @xmath39 and the charged higgsino .
the other is the neutralino - smuon loop , where the neutralinos are the admixture of their neutral counterparts .
the other contributions , such as the charged higgs loop one , are so small that they are irrelevant even for future study .
[ fig : g-2_50_200 ] is intended to demonstrate what magnitude of those effects to muon @xmath0 is expected .
the figure shows the supersymmetric contribution to muon @xmath0 as a function of @xmath40 parameter ( the supersymmetric mass common to two higgs supermultiplets ) , for relatively large @xmath41 ( the ratio of the vacuum expectation values of two higgs doublets ) , the supersymmetry breaking slepton mass set equal to 200 gev , and for three choices of supersymmetry breaking @xmath42 gaugino mass @xmath43 with @xmath44 gaugino mass given here through the gut relation @xmath45 as the weak effect is @xmath46 , the supersymmetric effect can become substantial .
actually the muon @xmath0 even with the current accuracy excludes the region of negative sign of @xmath40 for this set of the other parameters .
[ fig : g-2_50_200 ] is drawn without paying any attention to the other constraints on supersymmetric models already present .
the direct search of superpartners of the known species puts the lowest bound ( 93 gev ) to the lightest chargino mass , and the bound ( 78 gev ) to the mass of each lighter slepton @xcite .
in fact the chargino mass bound requires that the absolute magnitude of @xmath40 parameter be greater than about 100 gev for the gaugino mass in our interest , while the slepton mass bound demands @xmath47 less than about 400 gev . with those regions excluded fig .
[ fig : cnt_50_200 ] shows the contours each of which has equal magnitude of muon @xmath0 on the @xmath40-@xmath43 plane . since the future accuracy is much smaller than the interval between the neighboring contours in fig .
[ fig : cnt_50_200 ] , we have a great chance to observe a signal coming from the existence of supersymmetry through muon @xmath0 .
+ in @xmath40(horizontal direction)-@xmath43 plane for @xmath48 and @xmath49 gev .
the two islands are the regions escaping from any other constraints .
the contours are drawn with the interval of @xmath50 from @xmath51 to @xmath52 for @xmath53 .
darker face corresponds to smaller @xmath53 .
] now we are tempted to grasp the specific feature of muon @xmath0 in search of supersymmetric theory .
it will turn out that muon @xmath0 seems to have a peculiar property which is not shared by any other observables .
+ in the most honest region of parameter space @xmath54 , chargino - sneutrino loop contribution dominates over neutralino - smuon loop contribution .
the current basis analysis helps one to catch up with the qualitative dependence on the various parameters .
as long as @xmath55 , the chirality flips due to the vacuum expectation value @xmath56 of the higgs field giving mass to the up - type quarks , turning @xmath39 to the charged component of @xmath57 , which transformed to the charged @xmath58 due to @xmath40-term .
picking @xmath59 from @xmath56 and @xmath60 from a yukawa - type coupling involving muon , the dominant contribution in the present situation becomes @xmath61 although the overall sign needs a detail computation . from this expression
we can read off such properties that 1 .
the effect to muon @xmath0 is greatly enhanced for large @xmath41 @xcite .
in fact , when @xmath41 is small , the overall magnitude of susy effect is drastically reduced as shown in fig . [
fig : g-2_3_100 ] .
thus in this case the current experiment could not put any restriction on its existence .
but the future accuracy in muon @xmath0 is quite sufficient to explore it .
2 . the sign of this contribution is govern by the sign of @xmath40 .
it is interesting to remind that a large @xmath41 is a natural consequence of the gauge mediated susy breaking scenario , and elaborate analysis on muon @xmath0 has been performed in this line @xcite . or it is a necessary ingredient for unification of all yukawa coupling constants of the third generation . + as observed at both ends of @xmath40-direction in fig .
[ fig : g-2_3_100 ] , supersymmetric effect to muon @xmath0 does not decouple even if we can let the absolute magnitude of @xmath40 large while those other parameters remain fixed .
note that small @xmath41 case allows relatively large absolute magnitude of @xmath40 parameter without conflicting with slepton mass bound , as the mixing between left- and right - sleptons are proportional to a combination @xmath62 .
+ such a phenomenon can be understood from the following observation .
when @xmath47 is large the chargino - sneutrino effect decouples , but the neutralino - smuon effect increases .
let us consider a diagram in which the chiral flip occurs due to the mixing between the left and right - handed smuons in the current eigen - basis .
as the higgino does not propagate , suppression factor due to the inverse power of @xmath40 is now absent .
thus @xmath53 becomes proportional to @xmath63 .
( the sign is also opposite to the chargino - sneutrino effect ( [ eq : a_susy_1 ] ) . )
this is the reason for such a behavior in this large @xmath47 region .
+ from those observations , muon @xmath0 seems to play the major role to find the sign and the magnitude of @xmath40 term .
as far as i know , such a property sensible to @xmath40 is not shared by any other observables .
recall the following two facts , that is , 1 .
@xmath40 is a supersymmetric parameter which is associated with common mass of higgs supermultiplet .
thus this is a parameter independent of the supersymmetry breaking parameters by nature .
although supersymmetry assures the quantum stability of electroweak scale , supersymmetry does not set @xmath40 to this region at tree level automatically .
this is the most annoying matter called as
@xmath40 problem .
this problem stands out especially in the context of gut .
thus , once supersymmetry is established also by the other experiments , the determination of @xmath40 parameter through muon @xmath0 may develop further theoretical access to the origin of @xmath40 , the origin of electroweak scale .
+ before addressing to the future testing possibility in the small @xmath41 regime , we remind the additional constraints implied from precision measurement at @xmath1 pole .
as was mentioned , the result of this precision measurement has given an indispensable information to argue grand unification .
it also has killed the naive technicolor models .
thus we should discuss the effect on muon @xmath0 on the region of the parameter space consistent with those measurements .
+ they are summarized by four parameters .
three of these parametrizes the `` oblique '' corrections from new physics , with respect to `` reference '' standard model ; here we take the one specified by @xmath64 the last one is associated with the modification of coupling of bottom quark to @xmath1 boson .
this is neglected here by assuming that the squarks are so heavy enough that their effects decouple . since it has been recognized that the susy effect to @xmath8 boson mass and coupling of @xmath6 to @xmath1 is not relevant within the current accuracy@xcite , we concentrate on @xmath65 and @xmath66 parameters .
-@xmath66 parameters for @xmath67 .
slepton brings @xmath65 parameter to decrease ( each line follows the response to the change of @xmath68 . )
while chargino and neutralino tend to increase it ( each line shows the response to the change of @xmath40 parameter . ) .
therefore they partly cancels when added together ( the line with square dots for @xmath68 = 100 gev , with triangle dots for @xmath68 = 200 gev for @xmath41 = 3 ) . ]
[ fig : st_100 ] + fig .
[ fig : st_100 ] shows a constraint implied from @xmath65 and @xmath66 parameters .
, @xmath66 themselves here .
] the reference standard model is at the origin on this plane located in the contour of 90 % confidence level .
the slepton contribution brings @xmath65 parameter to negative , while the chargino and neutralino ones to positive .
a set of two lines in the left - hand side pursues the response of the slepton effects to the change of susy breaking slepton mass for two values of @xmath41 ( solid line for @xmath69 , dashed line for @xmath48 . )
the one on the right - hand side follows the response of the chargino effects against the change of @xmath40 parameter .
the solid line with the square ( triangular ) marks represents the locus followed by the sum of the these two contributions for the slepton mass 100 gev ( 200 gev ) and @xmath41 equal to 3 when @xmath40 is changed to about 500 gev .
thus such a parameter set with @xmath70 equal to 3 is allowed at 95 % confidence level .
but in the case of @xmath41 = 50 it is rather difficult to take slepton mass equal to 100 gev .
once the slepton mass is taken larger , for instance , at 200 gev , there is no restriction from this analysis .
+ now for @xmath69 the contour with equal @xmath53 in the @xmath40-@xmath43 plane is drawn in fig .
[ fig : cnt_3_100 ] . in @xmath40-@xmath43 plane for @xmath69 and @xmath71 gev .
the contours are drawn with the interval of @xmath72 between @xmath73 and @xmath74 for @xmath53 . ] with the future accuracy , which amounts to the interval between the neighboring contours in that figure , we can extract susy effect and may obtain precise information on the model .
now let us turn back to the theoretical uncertainty .
as was mentioned by several talks in this workshop , besides qed contribution , @xmath75 is also dominated by the leading order qcd contribution which arises through the hadronic vacuum polarization .
its improvement is now awaiting for the precise knowledge about the low energy hadron production cross section planned to be accumulated at novosivirsk , frascatti and beijing .
+ the hadronic light - by - light scattering effect @xcite , which requires purely theoretical evaluation , may become an obstacle .
thus the reduction of its error also needs further challenge .
+ to summarize we discussed the effect to muon @xmath0 from two candidates of models each of which accesses to `` hierarchy problem '' .
we found that the potential signatures are expected from the existence of both two candidates by future measurement of muon @xmath0 even on account of the precision measurement at @xmath1 pole .
but this program can not be accomplished without improvement in measurement of the hadron production cross section in low energy domain . +
* acknowledgement * + + the author thanks s. eidelman for hospitality at novosivirsk . he also thanks to n. sakai for occasion visiting at tit after this march , and to g. c. cho and k. hagiwara for the various comments on the preparatory content of talk .
10 r. m. carey , `` new measurement of the anomalous magnetic moment of the positive muon '' , phys .
82 * , 1632 ( 1999 ) .
talk by s. redin in this workshop .
j. bailey _
et al . _ , `` the anomalous magnetic moment of positive and negative muons '' , phys
68b * , 191 ( 1977 ) ; `` final report on the cern muon storage ring including the anomalous magnetic moment and the electric dipole moment of the muon , and a direct test of relativistic time dilation '' , nucl
b150 * , 1 ( 1979 ) ; f. j. m. farley and e. picasso , `` the muon @xmath0 experiments '' , in _ quantum electrodynamics _ , edited by t. kinoshita ( world scientific , singapore , 1990 ) , p. 479 .
m. davier and a. hocker , `` new results on the hadronic contributions to @xmath76 and to @xmath77 '' , phys .
b435 * , 427 ( 1998 ) , hep - ph/9805470 .
s. eidelman and f. jegerlehner , `` hadronic contributions to @xmath0 of the leptons and to the effective fine structure constant @xmath76 '' , z. phys .
* c67 * , 585 ( 1995 ) , hep - ph/9502298 ; b. krause , `` higher order hadronic contributions to the anomalous magnetic moment of leptons '' , phys . lett . *
b390 * , 392 ( 1997 ) , hep - ph/9607259 .
a. czarnecki , b. krause and w. j. marciano , `` electroweak corrections to the muon anomalous magnetic moment '' , phys .
lett . * 76 * , 3267 ( 1996 ) , hep - ph/9512369 .
t. kinoshita,``the fine structure constant '' rept .
59 * , 1459 ( 1996 ) .
n. arkani - hamed , s. dimopoulos and g. dvali , `` the hierarchy problem and new dimensions at a millimeter '' , phys . lett . *
b429 * , 263 ( 1998 ) , hep - ph/9803315 . n. arkani - hamed , s. dimopoulos and g. dvali , `` phenomenology , astrophysics and cosmology of theories with submillimeter dimensions and tev scale quantum gravity '' , phys . rev . *
d59 * , 086004 ( 1999 ) , hep - ph/9807344 .
m. l. graesser , `` extra dimensions and the muon anomalous magnetic moment '' , hep - ph/9902310 .
j. ellis , j. hagelin and d. v. nanopoulos , `` spin 0 leptons and the anomalous magnetic moment of the muon '' , phys .
116b * , 283 ( 1982 ) ; j. a. grifols and a. mendez , `` constraints on supersymmetric particle masses from @xmath77 '' , phys . rev . *
d26 * , 1809 ( 1982 ) ; d. a. kosower , l. m. krauss and n. sakai , `` low - energy supergravity and the anomalous magnetic moment of the muon '' , phys . lett . *
133b * , 305 ( 1983 ) .
u. chattopadhyay and p. nath , `` probing supergravity grand unification in the brookhaven @xmath0 experiment '' , phys . rev . * d53 * , 1648 ( 1996 ) hep - ph/9507386
. j. l. lopez , d. v. nanopoulos and x. wang , `` large @xmath77 in su(5 ) x u(1 ) supergravity models '' , phys .
* d49 * , 366 ( 1994 ) , hep - ph/9308336 ; t. moroi , `` the muon anomalous magnetic dipole moment in the minimal supersymmetric standard model '' , phys . rev . * d53 * , 6565 ( 1996 ) , hep - ph/9512396 ; _ ibid_. * d56 * , 4424 ( 1997 ) ( e ) ; f. borzumati , g. r. farrar , n. polonsky and s. thomas , `` soft yukawa couplings in supersymmetric theories '' , hep - ph/990244 m. carena , g. f. giudice and c. e. wagner , `` constraints on supersymmetric models from the muon anomalous magnetic moment '' , phys . lett . *
b390 * , 234 ( 1997 ) , hep - ph/9610233 ; e. gabrielli and u. sarid , `` low - energy signals for a minimal gauge mediated model '' , phys
* 79 * , 4752 ( 1997 ) , hep - ph/9707546 ; `` imminent phenomenology of a minimal gauge mediated model '' , phys . rev .
* d58 * , 115003 ( 1998 ) , hep - ph/9803451 .
g. c. cho , k. hagiwara and m. hayakawa , in preparation .
s. mihara , `` recent results from lep - ii '' , talk given in kek theory meeing on _ collider physics _ , jan 11 - 12 , 1999 ( tsukuba , january 1999 ) . g. degrassi and g. f. giudice , `` qed logarithms in the electroweak corrections to the muon anomalous magnetic moment '' , phys . rev . *
d58 * , 053007 ( 1998 ) , hep - ph/9803384 . private communication with
g. c. cho and k. hagiwara .
m. hayakawa , t. kinoshita and a. i. sanda , `` hadronic light - by - light scattering effect on muon @xmath0 '' , phys .
lett . * 75 * , 790 ( 1995 ) , hep - ph/9503463 ; `` hadronic light - by - light scattering contribution to muon @xmath0 '' phys . rev . *
d54 * , 3137 ( 1996 ) , hep - ph/9601310 ; j. bijnens , e. pallante and j. prades , `` hadronic light - by - light contributions to the muon @xmath0 in the large @xmath78 limit '' , phys .
lett . * 75 * , 1447 ( 1995 ) , hep - ph/9505251 ; erratum , _
ibid_. * 75 * , 3781 ( 1995 ) ; `` analysis of the hadronic light by light contributions to the muon @xmath0 '' , nucl .
* b474 * , 379 ( 1996 ) , hep - ph/9511388 ; m. hayakawa and t. kinoshita , `` pseudoscalar pole terms in the hadronic light - by - light scattering contribution to muon @xmath0 '' , phys .
* d57 * , 465 ( 1998 ) , hep - ph/9708227 . | here we invoke the current and future perspective on muon @xmath0 measurement when asking what the muon @xmath0 could tell about the underlying structure concerning with the hierarchy problem . here
we take up two such models , the presence of which turns out to alter the standard model prediction for muon @xmath0 significantly : one is tev scale gravity scenario , the other supersymmetric model , in the latter case of which the precision measurement up to @xmath1 boson mass is taken into account as an explicit constraint . |
binary neutron star ( bns ) mergers are the catastrophic fate of relativistic binary systems formed by a pair of neutron stars ( nss ) . the intense emission of gravitational waves ( gw ) that characterizes the final phases of the inspiral and its expected frequencies of @xmath3-@xmath4 hz make merging bnss one of the primary targets of ground based gw detectors , such as advanced ligo , advanced virgo , and , in the nearby future , kagra .
in addition , these events are expected to inject matter and energy into the interstellar medium , and to produce a rich variety of paramount astrophysical phenomena , including the nucleosynthesis of the heaviest elements , and the emission of characteristic electromagnetic transients from the freshly synthesized radioactive nuclei , and radio flares from the expanding ejecta .
a bns merger is expected to leave behind a compact object surrounded by a hot and dense accretion disk . inside this
remnant , gravitational energy is efficiently converted to internal energy and emitted in the form of neutrinos .
the nature of the central object depends primarily on intrinsic attributes of the merging system ( such as the nss masses and their ratio ) and on the still uncertain properties of hot nuclear matter above nuclear saturation density .
the direct collapse of the merging nss into a black hole ( bh ) is foreseen only for very massive colliding objects , while in most cases the production of a ( possibly metastable ) differentially rotating massive neutron star ( mns ) is expected .
if the mns mass exceeds the maximum allowed ns mass , the mns will collapse to a bh on a characteristic time scale .
the duration of this phase depends sensitively on several parameters and poorly understood processes happening inside the mns . among them
, we recall the amount of angular momentum and its distribution , the effects of thermal support and the consequences of neutrino cooling , the role and the properties of magnetic fields . the large amount of energy available in these events , together with the short time scale and the reduced volume that characterize the merger and its aftermath , led to the hypothesis that bns mergers represent ( together with the merger of bh - ns binaries ) a plausible progenitor system for short gamma - ray bursts ( grbs ) at cosmological distances .
observational evidences collected over the last years seem to support this hypothesis . however , a satisfactory and detailed explanation of how the merger of a compact binary system involving at least one ns powers a relativistic jet is still missing and the central engine of short grbs still remains veiled .
several different mechanisms have been suggested . among them , the conversion of energy by magnetohydrodynamical effects in bh - torus systems @xcite , the extraction of magnetic energy from highly magnetized , long - lived mns @xcite , and the annihilation of neutrino - antineutrino pairs in low density regions above the remnant @xcite .
the conditions needed for the formation and collimation of a relativistic jet above a merger remnant , as well as the interactions with the accretion disk , non - relativistic winds and dynamic ejecta , have been recently investigated with increasing details and accuracy @xcite .
neutrino emission from merger remnants has been the topic of several dedicated studies @xcite .
the annihilation of neutrino - antineutrino pairs above bh - torus systems has been extensively studied , also considering the effects of general relativity on the neutrino propagation @xcite .
annihilation rates above mns - disk system have been computed by and . in the former case , several snapshots from the first 100 ms of an axisymmetric merger aftermath simulation were post - processed using both a @xmath5 transport scheme and a simpler gray leakage scheme to model neutrino radiation . in the latter , four snapshots taken from long - term , axisymmetric simulations of merger aftermaths , and covering the 0 - 3 sec interval were post - processed with a monte carlo radiative transfer code and , again , with a simpler gray leakage scheme . in this work ,
we compute the energy and momentum deposition rates above bns merger remnant over a time scale of 400 ms , comparable with the life time of the disk . we post - process several merger configurations taken from a long - term , three dimensional simulation of the aftermath of a bns merger .
we assume a long - lived mns and specifically investigate the impact of the neutrino emission from the mns .
we also explore the impact of relativistic effects in the neutrino propagation on the annihilation rates , in the presence of a mns .
finally , we discuss our results in connection with the energy inferred from short grb observations , to address the question whether neutrino pair annihilation can be responsible for the production of relativistic jets above bns mergers .
the paper is structured as follows : in section [ sec : simulation ] , we present the merger aftermath simulation from which we take the remnant configuration and the neutrino emission properties .
our method to compute the annihilation rates , both in a newtonian and in a relativistic framework , is outlined in section [ sec : method ] .
section [ sec : results ] is devoted to the presentation of the annihilation calculations while in section [ sec : rolemns ] we analyze the impact of a long lived - mns on the annihilation rates .
our results are compared with short grb observations in section [ sec : discussion ] .
we summarize and conclude in section [ sec : conclusions ] .
our annihilation rate calculations are based on the numerical results of a three dimensional , newtonian simulation of the aftermath of a binary neutron star merger under the influence of neutrino cooling and heating @xcite .
the initial conditions for our model were taken from a high - resolution sph simulation of the merger of two equal mass ( 1.4 @xmath6 ) neutron stars @xcite .
we mapped the remnant configuration at @xmath7 ms after the merger inside our equally spaced cartesian grid . at this time
, the remnant was composed by an almost axisymmetric , stationary rotating mns with a mass of @xmath8 , surrounded by a thick accretion disk with a mass of @xmath9 .
we employed a rather stiff nuclear equation of state ( eos ) , hs(tm1 ) @xcite , for consistency with the initial conditions .
the radius of the mns , defined as the radial coordinate where the matter density drops below @xmath10 , was located in a range between 17 km and 25 km along the vertical and the equatorial directions , respectively .
neutrinos of all flavors were emitted from the remnant with a total luminosity larger than @xmath11 .
the re - absorption of a fraction of the emitted neutrinos inside the disk caused the formation of a neutrino - driven wind on a time scale of a few tens of milliseconds .
at the same time , matter expanded from the disk along the equatorial direction .
this was the beginning of the so - called disk evaporation @xcite .
the wind developed mainly from the disk and the funnel above the central neutron star has a lower density than the material extending in the equatorial direction .
we evolved the system for @xmath12 ms .
since our uniform spatial resolution ( @xmath13 ) was not sufficient to properly resolve the mns , we assumed the region inside @xmath14 as stationary , where @xmath14 is an oblate spheroid of @xmath15 as symmetry axis , center in the domain origin , semi - axis @xmath16 along @xmath17 and @xmath18 , and @xmath19 along @xmath15 , and matter density above @xmath20 .
profiles of the matter density and electron fraction inside the mns and in the innermost part of the disk at 40 ms inside our simulation are shown in figure [ fig : density and ye profiles ] , together with the incipient neutrino - driven wind .
( left ) , @xmath21 ( center ) and @xmath22 ( right ) , while lines with different colors and increasing thickness represent increasing neutrino energies . ]
neutrino physics is modeled using an energy dependent ( spectral ) leakage scheme @xcite .
all neutrino quantities are binned logarithmically in energy , with twelve energy bins between @xmath23 and @xmath24 .
the neutrino reactions included in the calculations are the emission , absorption , and scattering off free nucleons .
neutrino pair emission from electron - positron annihilation and nucleon - nucleon bremsstrahlung are also included in optically thick conditions , as well as absorption from their inverse reactions .
we model three independent neutrino species , @xmath25 , @xmath21 , and @xmath22 , the latter being a collective species for @xmath26 and @xmath27 ( anti)neutrinos .
spectral optical depths , @xmath28 , are computed by minimizing the line integral of inverse mean free paths along several possible straight propagation directions .
this allows the distinction between optically thin ( @xmath29 ) and optically thick ( @xmath30 ) regions .
we further distinguish between scattering , @xmath31 , and energy optical depths , @xmath32 @xcite .
we denote the neutrino surfaces as the surfaces where @xmath33 . in the case of @xmath31
they correspond to the last scattering surfaces . in the case of @xmath32
they locate the transition from a thermally coupled to a non - thermally coupled regime .
the position of the scattering and energy surfaces inside the remnant at 40 ms in our simulation are represented in figure [ fig : nu_surfaces ] , for all neutrino species and for five representative energies .
neutrinos are considered emitted isotropically in the half space denoted by @xmath34 , both from the neutrino scattering surfaces ( for radiation diffusing from optically thick conditions ) and from their production site ( for radiation emitted in optically thin conditions ) ( * ? ? ?
* appendix ) .
we review the most relevant properties of the neutrino emission obtained from our hydrodynamical simulation , since they represent a key ingredient for the calculations of the annihilation rates . in figure
[ fig : neutrino luminosities ] , we present the evolution of the neutrino luminosities ( @xmath35 left panel ) and root mean squared ( rms ) energies ( @xmath36 , right panel ) for all the three independent neutrino species , over the entire simulation time .
both quantities are measured at infinity , taking into account that a fraction of the emitted neutrinos ( especially @xmath25 s and , less significantly , @xmath21 s ) get absorbed on their way out . to show the effect of neutrino absorption outside the neutrino surfaces , we plot also the luminosities and rms energies obtained including the neutrinos emitted at the last scattering surfaces and outside , but assuming the neutrino absorptivity to be zero for @xmath37 .
we call them `` cooling '' luminosities and rms energies ( dashed lines ) .
neutrino luminosities are powered by the cooling of the remnant and by the process of accretion inside the disk .
after an initial phase ( @xmath38 ms ) when the disk dissipates internal perturbations and transients , and the total luminosity slightly increases , all the luminosities decrease with time .
the rms neutrino energies reflect the different depths and thermodynamical conditions at which neutrinos thermally decouple from matter .
while the values obtained for @xmath25 and @xmath21 are consistent with the ones reported in the literature , the values obtained for @xmath22 are smaller than expected ( by @xmath3920% ) .
as reported in , this discrepancy is due to the lack of spatial resolution in our simulation deep inside the mns , where the thermal decoupling of @xmath22 s occurs . due to the stationary treatment of the mns ,
the luminosities coming from the spheroid @xmath14 , @xmath40 , are practically constant during the simulation . to mimic the decrease produced by the cooling process
, we assume the evolution of the luminosities coming from the innermost part of the computational domain as an exponential decline : @xmath41 in analogy to the luminosity produced by a cooling proto - ns @xcite .
the cooling time scale @xmath42 is estimated as @xmath43 , where @xmath44 is the thermal component of the internal energy inside @xmath14 , calculated from the initial remnant configuration . from our simulation we obtained @xmath45 .
we consider the annihilation of neutrino pairs into electron positron pairs , i.e. @xmath46 with @xmath47 .
we calculate the energy and momentum deposition rates due to this process outside the most relevant neutrino surfaces , post - processing the input data obtained in our simulation of the aftermath of a binary ns merger .
since the power associated with neutrino - antineutrino annihilation is expected to be sub - dominant compared with the one of charged - current neutrino absorption reactions , we can safely compute the former in a post - processing step @xcite .
for consistency with the underlying simulation , we start computing the rate in a newtonian framework .
moreover , we assume the neutrino to be massless and we presently do not include the effects of neutrino oscillations in our analysis .
we calculate the local energy deposition rate @xmath48 for the neutrino flavor @xmath49 according to the following expression ( * ? ? ?
* ; * ? ? ?
* and references therein ) : @xmath50 we extend the previous formula to compute the deposited momentum @xmath51 : @xmath52 in eqs .
( [ eq : dessart ] ) and ( [ eq : dessart momentum ] ) , @xmath53 is the speed of light , @xmath54 the mass of the electron , @xmath55 the typical neutrino cross section , with @xmath56 the fermi constant , and @xmath57 the reduced planck constant .
furthermore , @xmath58 denotes the radiation intensity at position @xmath59 and time @xmath60 , for neutrinos with energy @xmath61 along the direction @xmath62 .
the infinitesimal solid angle and energy in the momentum space of the propagating radiation is represented by @xmath63 and @xmath64 , respectively .
the cosine of the angle between the momenta of the colliding neutrinos is given by @xmath65 .
equations ( [ eq : dessart ] ) and ( [ eq : dessart momentum ] ) apply to all neutrino species , provided the appropriate values of the weak coupling constants @xmath66 and @xmath67 , where @xmath68 . in the case of @xmath25 and @xmath21 , @xmath69 , while for @xmath22 and @xmath70 , @xmath71 .
note that in the expressions for @xmath72 and @xmath51 we have neglected phase space blocking for @xmath73 in the final state , as well as a term proportional to @xmath74 in the integrand .
in fact , the latter is @xmath75 times smaller for typical neutrino energies . since the neutrino intensities required in eqs .
( [ eq : dessart ] ) and ( [ eq : dessart momentum ] ) are not directly provided by our simulation , we compute them based on the local emissivities .
we fix one point @xmath76 outside the most relevant neutrino surfaces .
we denote the amount of energy @xmath77 transported across a surface element @xmath78 in a time @xmath79 by @xmath80 neutrinos in an energy interval @xmath81 around @xmath82 propagating inside a solid angle @xmath83 around @xmath84 , in terms of radiation intensity @xmath85 as @xmath86 .
we further express the amount of energy emitted at any point @xmath87 by a volume @xmath88 in a time @xmath79 in @xmath89 neutrinos with an energy interval @xmath90 around @xmath91 and into a solid angle @xmath92 around @xmath93 , as @xmath94 , where @xmath95 is the local emissivity .
conservation of particles and energy implies : @xmath96 where we took into account that a fraction of the emitted neutrinos get absorbed on their way to @xmath76 .
this amount is estimated using an exponential damping based on the difference of the energy optical depth @xmath97 between @xmath87 and @xmath76 , @xmath98 . inserting the definitions of intensity and emissivity into eq .
( [ eq : energy conservation between a and b ] ) , and rewriting @xmath99 , where @xmath100 denoted the distance between the two points , we obtain : @xmath101 we model the emission of neutrinos from the neutrino surfaces and from the optically transparent regions consistently with the treatment used in the post - merger simulation .
specifically , @xmath102 i.e. , we assume that radiation is emitted isotropically in the half space identified by the unitary vector @xmath103 and we take @xmath104 ( see section [ sec : simulation ] ) .
we discretize the three dimensional space with cylindrical coordinates @xmath105 where @xmath49 , @xmath106 and @xmath107 can vary between 1 and @xmath108 , @xmath109 and @xmath110 , respectively . for the neutrino energy , we use the same energy bins that we employed in our post - merger simulation , @xmath111 , with @xmath112 . from the results of our simulation ,
we extract cylindrically averaged values for the spectral optical depths @xmath113 and for the ( solid angle integrated ) neutrino emission rates in each energy bin @xmath114 , @xmath115 .
integrating eq .
( [ eq : isotropic emissivity ] ) over the whole solid angle , we can express @xmath116 in terms of @xmath117 : @xmath118 if we discretize eq .
( [ eq : dessart ] ) according to @xmath119 inserting eqs .
( [ eq : intensity_omega_e ] ) and ( [ eq : eta and r_nu relationship ] ) we obtain @xmath120 where @xmath121 is the volume element in cylindrical coordinates , @xmath122 , and @xmath123 is the heaviside step function . for the discrete version of eq .
( [ eq : dessart momentum ] ) , it is enough to replace @xmath124 with @xmath125 in eq .
( [ eq : dessart_discretized 2 ] ) . starting from the local expression of the energy deposition rate , eq . ( [ eq : dessart ] ) , we compute the volume - integrated energy deposition rate , @xmath126 , as @xmath127 following , we define @xmath128 as the volume outside the most relevant neutrino surfaces , and in which the radial component of the total momentum deposited by neutrinos is pointing outwards .
in addition , we designate the annihilation efficiency as : @xmath129 since the integrand in eq .
( [ eq : q non rel ] ) is proportional to @xmath130 , it is natural to assume that : @xmath131 \ , , \label{eq : simple parametrization}\ ] ] where @xmath132 and @xmath133 are geometrical factors that contain the integral of all the residual spatial and geometrical dependencies . in the next sections , we will test the validity of this parametrization , based on detailed numerical results , using both the luminosities at infinity and the cooling luminosities .
finally , we define @xmath134 as the deposited cumulative energy : @xmath135 alongside with a global annihilation efficiency , @xmath136 the hydrodynamics simulation and the annihilation rate calculations presented in the previous sections are performed in a newtonian framework .
however , matter in the disk moves close to keplerian , and typical orbital speeds are given by : @xmath137 where @xmath138 and @xmath139 are the msn mass and radius , respectively .
moreover , the ratio between the the schwarzschild radius @xmath140 and the actual mns radius is @xmath141 thus , relativistic effects due to the fast motion of matter and to the intense gravitational field are potentially relevant .
a consistent relativistic calculation requires to start from the results of relativistic hydrodynamical simulations of a neutron star merger , and to compute the annihilation rates in a relativistic framework . due to the newtonian character of our input simulations ,
this is not entirely possible .
however , we can compute the impact of special and general relativistic effects on the neutrino propagation and on the annihilation process using our newtonian input data .
we consider the following effects on the propagation from the emission to the interaction points : the doppler and beaming effects due to the fast motion of the emitting fluid elements , and the redshift and light bending effects due to motion of the radiation in the gravitational potential .
a detailed explanation of the adopted spacetime model , as well as of the implementation of the relativistic effects is presented in [ apx : implementation rel effects ] .
following and , we compute the volume - integrated energy deposition rates in the local frames @xmath142 as well as the energy deposition rates measured by an infinitely distant observer : @xmath143 where the redshift due to the gravitational field is also included ) and ( [ eqn : relativisitic at infinity volume integrated energy rate ] ) differ by the indexes of the metric tensor : in the former expression only the spatial part of @xmath144 is considered , while the full metric is used in the latter . ] . each of these two rates can be integrated over time to obtain the cumulative energies , @xmath145 and @xmath146 .
we perform the calculation of the annihilation rates for twenty equally spaced time steps , ranging from 0 to 380 ms .
since the input data are cylindrically symmetric , we expect also @xmath72 and @xmath51 to be axisymmetric .
thus , we compute the rates only on the @xmath147 plane and we assume them to be valid for any polar angle @xmath148 . for the discretization of the @xmath148 angle required by our calculations , we choose @xmath149 ( @xmath150 ) and we test the convergence of our results with respect to this parameter ( @xmath151 for @xmath152 ) . to save computing time
, we consider the emission of neutrinos only from cells with @xmath153 .
also in this case , this simplification introduces an error lower , at most , than a few percents .
in figure [ fig : anni rate different timesteps ] , we show the local energy deposition rate @xmath154 at four different times , both for electron flavor ( left panel ) and heavy flavor ( right panel ) neutrino pairs .
these rates have been obtained using our standard ( i.e. , non - relativistic ) approach .
we notice that in both cases and at each time the energy deposition rate is more intense close to the remnant and decreases for increasing distances , as a consequence of the decrease of the neutrino intensity .
moreover , even if the rates decrease with time due to the decline of the neutrino luminosities , their spatial distribution remains qualitatively similar during the evolution of the system .
however , the results obtained for the different flavors differ significantly . in the @xmath25-@xmath21 case ,
the energy deposition is more intense and distributed over a wider volume , including the regions above the mns and the innermost part of the disk . in the @xmath22-@xmath70 case ,
the energy deposition is concentrated above the mns and the decrease of @xmath155 with time is less pronounced , due to the milder reduction of the @xmath22 luminosities .
the deposition rate for the linear momentum is presented in figure [ fig : anni momentum rate different timesteps ] , again for four different times .
the arrows indicate the direction of @xmath156 projected on the vertical plane . due to the axisymmetry of the neutrino emissivities , we expect the azimuthal component of the deposited momentum to be negligible .
the geometry of the mns and of the disk influences the emission direction of the neutrinos and , in turn , the direction of @xmath156 , according to eq .
( [ eq : dessart momentum ] ) .
in particular , the mns produces an almost spherical emission inside the funnel above its surface .
this combines with the axisymmetric emission from the disk .
the latter happens predominantly at the transition between the mns and the disk , where neutrinos are emitted towards the funnel and , on average , perpendicularly to the neutrino surfaces ( see figure [ fig : nu_surfaces ] ) . at larger radii ,
the average emission direction from the disk becomes parallel to the @xmath15 axis or even pointing outwards .
since in the case of @xmath22 the neutrino emission comes mainly from the mns , @xmath157 presents a more radial structure .
we notice that at larger times , when the disk has shrunk its radius and consumed its mass , @xmath158 also develops into a more radial configuration .
a comparison between the energy and momentum deposition reveals that the intensity of the latter tracks the one of the former .
however , the dependence on the colliding angle @xmath159 contained in eqs .
( [ eq : dessart ] ) and ( [ eq : dessart momentum ] ) , and the vector sum that characterizes the momentum deposition imply that @xmath160 .
in particular , the more isotropic and symmetric the local neutrino distribution , the less efficient is the momentum deposition compared to the energy one . in figure [ fig : anni rate momentum ratio different timesteps ] , we present @xmath161 and confirm that close to the msn and to the disk , where the propagation directions are expected to be more isotropic , the ratio is significantly smaller than 1 , reaching even 0.2 , while it increases up to 0.8 at larger distances .
we repeat the pair annihilation calculations including special and general relativistic effects , and we present the results for the local energy deposition rate in figure [ fig : anni rate with gr different timesteps ] .
the inclusion of the relativistic effects does not change the qualitative features of the energy deposition rate .
a careful comparison between the energy deposition rates provided by @xmath162 pairs shows that the light bending and the gravitational blueshift increase the annihilation rates immediately above the mns ( by @xmath163 ) , while the gravitational redshift , together with the beaming of the radiation emitted by matter rotating inside the disk , reduces the energy deposition rate in the funnel and above the innermost part of the disk .
where the rates are larger , the reduction is usually @xmath164 , but it increases when the distance from the mns and the disk increases , as well as at later times . in the case of @xmath165 pairs ,
the dominant effect is provided by the gravitational redshift and the result is a reduction of the energy deposition rate everywhere in the computational domain . the gain in annihilation efficiency due to the light bending
is attenuated by the dominant contribution of the mns to the @xmath22 luminosities .
the qualitative features of the momentum deposition rate computed by taking into account the relativistic effects in the neutrino propagation are very similar to the ones obtained in our reference newtonian calculations . in figure
[ fig : anni rate momentum ratio with gr different timesteps ] , we present the ratio between the modulus of the momentum and the energy deposition rates .
a comparison with the analogue figure obtained in the newtonian framework , figure [ fig : anni rate momentum ratio different timesteps ] , shows that the light bending reduces the efficiency at which a net momentum is deposited more significantly , with respect to the energy deposition rate , especially close to the mns surface and in the funnel above it . in figure
[ fig : anni rate integrated ] , we present the volume integrated rates , @xmath126 ( left panel ) and cumulative deposited energy , @xmath134 ( right panel ) , for both the reference newtonian calculations , eqs .
( [ eq : q non rel])-([eq : e non rel ] ) , and including relativistic corrections , ( [ eqn : relativisitic local volume integrated energy rate ] ) .
we first notice that , for the volume element in our cylindrical discretization , one has @xmath166 thus , intense deposition regions located at larger cylindrical radii contribute more significantly than close to the rotational axis .
the integration over the volume allows to quantify the difference between the different neutrino flavors : @xmath167 is @xmath39 60 times larger than @xmath168 .
due to this large difference , the contribution of @xmath22 s is significantly sub - dominant for the overall energy deposition process . the approximated expression ( [ eq : simple parametrization ] )
accounts for this considerable disparity . in particular , comparing the electron and the heavy flavors , @xmath169 and @xmath170 .
the inclusion of relativistic effects in the neutrino propagation changes only marginally the intensity of the energy deposition rate .
in particular , the energy blueshift and light bending happening close to the mns are globally compensated by the redshift and the light beaming in the funnel and above the disk .
overall , the integrated energy deposition measured by local observers differs only by a few percent in the newtonian and in the relativistic case .
the difference becomes more relevant ( @xmath171 ) only when @xmath172 and @xmath173 are considered : this is due to the inclusion of the gravitational redshift required for the deposited energy to travel from the deposition site to infinity .
the volume - integrated energy deposition rate for @xmath25 and @xmath21 ranges between @xmath174 and @xmath175 , with a clear decreasing trend with time . the total amount of energy deposited by neutrino pair annihilations in the funnel above the merger remnant reaches @xmath176 erg at 380 ms in our newtonian calculation , while it reduces to @xmath177 erg in the relativistic calculations when the energy at infinity is considered .
we notice that , even if the deposition rate has significantly decreased at the end of our simulation , this energy has not saturated yet .
this is due to the longer time scales of neutrino diffusion from the mns , @xmath178 , and of disk consumption , @xmath179 @xcite .
in order to investigate the role of the mns in the process of neutrino pair annihilation , we decompose the local emissivity @xmath116 into a contribution from the disk ( @xmath180 ) and one from the neutron star ( @xmath181 ) . all neutrinos that originate inside the spheroid @xmath14 are tagged as `` neutron star ( ns ) neutrinos '' , while all the others are identified as `` disk ( ds ) neutrinos '' .
we notice that this distinction refers to the origin place , but not necessarily to the emission place .
while all ds neutrinos are emitted from the disk region , a fraction of the ns neutrinos can diffuse out from the disk before being emitted .
this happens especially for those neutrinos emitted by the mns along the equator . in the case of an accretion disk around a bh
, the disk luminosity can be related to the accretion rate at the bh horizon , @xmath182 , where @xmath183 is a coefficient that quantifies the efficiency at which the rest mass of the accreting matter is converted into emitted radiation ( typically , @xmath184 , ) . here
we present the behavior of the luminosity and accretion rate based on our simulation and discuss the significant differences due to the presence of a mns .
the distinction between neutrinos coming from the mns and from the disk allows to decompose each neutrino luminosity into a mns and a disk contribution . in figure
[ fig : decomposed neutrino luminosities ] , we show for each independent species the disk and the mns components .
the mns contributions originate mainly from the diffusion of neutrinos from the hot and dense central remnant .
their decrease reflects the progressive cooling of the mns and of the matter accreting on its surface .
the larger disk components are more related to the accretion process and present a more pronounced reduction .
the evolution of the accretion rate inside the disk , @xmath185 is shown in the left panel of figure [ fig : accretion rate and luminosity ] . since there is no clear inner boundary where to measure the flux of the infalling matter
, we compute the accretion rate as the time derivative of the mass contained inside the densest part of the disk ( @xmath186 ) , outside a cylindrical radius @xmath187 : @xmath188 where @xmath123 is the heaviside step function .
the choice of @xmath189 ensures the exclusion of the mns . since the expansion of matter in the @xmath190-driven wind and in the viscous component happen around a few times @xmath191 , our definition of @xmath185 takes into account only the disk consumption due to the accretion process . to verify our assumption
, we also compute @xmath192 where @xmath193 is the cylindrical surface of radius @xmath194 and @xmath195 is an infinitesimal surface element pointing in the positive radial direction .
we choose @xmath196 since for r @xmath197 the disk is characterized by a global infalling bulk motion along the radial direction , while at smaller radii , radially outgoing flows are also observed . in the initial phase ( @xmath198 ) , accretion onto the mns
occurs mainly from high density regions located within @xmath194 , thus @xmath199 . once the disk has reached an almost stationary configuration ( @xmath200 ) , the two accretion rates behave similarly and @xmath201 .
the integral of @xmath185 reveals that the mns has accreted @xmath39 0.1 @xmath6 within 400 ms . in the right panel of figure
[ fig : accretion rate and luminosity ] , we compare the total disk luminosity @xmath202 ( using both the total and the cooling luminosities ) with the luminosity obtained from the accretion rate , @xmath203 assuming @xmath204 .
we notice that in the presence of a mns @xmath205 is no more proportional to @xmath206
. in particular , the decline of the disk luminosity is slower than the decrease of the accretion rate .
the difference with the bh case is due to the presence of the long - lived mns , which has several consequences : \1 ) the mns surface does not act as an inner , open boundary , and the transition between the mns and the disk happens in a high density , high temperature region where the accreted matter settle on the mns surface .
the steep density gradient that characterizes this innermost part of the disk and the mns surface translates into a large pressure gradient .
while for @xmath207 the radial flow inside the disk is distinguished by an inward bulk motion , at smaller radii it is characterized by the presence of both inward and outward radial flows , resulting from the combination of large scale inflow , bounces on the mns surface and fast orbital motion .
\2 ) even if the innermost part of the disk is moderately optically thick @xmath208 , matter inside it can still emit neutrinos .
they diffuse out on a time scale of a few ms and are emitted at the neutrino surfaces .
\3 ) due to the disk finite size and consumption , the density profile inside the disk decreases and the disk becomes progressively more transparent to neutrinos .
in particular , the neutrino diffusion time scale , @xmath209 where @xmath210 , decreases with time . at the same time , the accretion process partially compensates the internal energy emitted by the neutrino radiation . if @xmath211 is the specific internal energy of matter the ratio @xmath212 increases with time in the innermost part of the disk . in summary , in the presence of a mns
, the larger efficiency at which the internal energy can be radiated by neutrinos , together with the longer time that matter has to cool , partially compensates the disk consumption and the accretion rate attenuation .
the splitting of the local emissivity into a neutron star and a disk component allows also the decomposition of the local radiation intensity , @xmath213 . since we deal with a pair process , the integrand in eq .
( [ eq : dessart ] ) for the local energy deposition rate @xmath214 can be expanded in four different contributions according to : @xmath215 .
\label{eq : q contributions } \end{aligned}\ ] ] in the left panel of figure [ fig : decomposition of the annihilation rate ] , we show the different contributions for electron flavor neutrinos at 40 ms , alongside with their sum . in the following
, we will refer to the combined neutron star - disk contribution as @xmath216 $ ] , if not stated differently .
the different contributions deposit energy at different locations above the remnant .
the ns - ns contribution is more intense immediately above the surface of the mns and it fades rapidly for increasing radial distances .
this reflects the quasi - spherical nature of the neutrino surfaces .
the annihilation is efficient only close to the emission surfaces , where the intensities are larger and the cosine of the average propagation angle is @xmath217 . for larger distances , the intensities decrease as @xmath218 and
enter the forward - peaked regimes , @xmath219 @xcite .
the largest energy deposition rates for the ds - ds contribution are located above the region marking the transition between the mns and the disk .
the energy rate in the funnel above the mns is still high , but it decreases while approaching the rotational axis .
this behavior points to the fact that most of the ds - ds pair annihilations happen close to the neutrino emission surfaces ( for radiation emitted from contiguous zones of the disk and annihilating at all angles ) rather than in the funnel , confirming results from .
the larger intensities of the former configuration compensate for the better collision angle of the latter . finally , for the ns&ds contribution , neutrinos moving away from the mns annihilate with those emitted by the disk towards the funnel .
the properties of this contribution are a combination of the properties of the previous two .
the results is an intense energy deposition both close to the mns surface and above the transition region between the mns and the disk .
we integrate the energy deposition rates over the volume for the three different contributions and the results are shown in the right panel of figure [ fig : decomposition of the annihilation rate ] .
the ds - ds and the ns&ds contributions are comparable throughout the whole simulation time .
this result points out the potential relevance of a long - lived mns , in comparison with a bh - torus system .
in fact , in the latter case , only the ds - ds contribution is expected to be present and the total energy deposited by neutrino pair annihilation can be reduced by a factor of ( at least ) two . the steeper decrease of the ds - ds contribution is due to its quadratic dependence on the more rapidly decreasing disk luminosity , @xmath220 , see eq .
( [ eq : simple parametrization ] ) , while for the ns&ds contribution we expect @xmath221 . despite the intense energy deposition associated with the ns - ns contribution in the funnel above the mns ,
the volume integrated rate is significantly sub - dominant , especially at earlier times .
this is a consequence of the smaller volume where the intense energy deposition occurs , see eq .
( [ eq : volume element dependences ] ) . to summarize the spatial dependence of the energy deposition , in figure [ fig : decomposition of the annihilation rate , angle and density cuts ] we present the volume integral of the energy deposited by @xmath162 pair annihilation at three different times and in selected regions of the domain . in particular , the integration is performed only for regions where polar angles are smaller than @xmath222 or greater than @xmath223 ( left panel ) , and densities are smaller than @xmath224 ( right panel ) .
the energy deposited closer to the rotational axis and at lower densities is foreseen to contribute more to the formation of a jet .
a comparison between the different terms reveals that the contributions involving the mns deposit a more significant fraction of their energy at high densities ( @xmath225 ) compared with the ds - ds contribution .
however , the latter deposits energies also at larger angular distances from the rotational axis ( @xmath226 ) . finally , we show the impact of relativistic effects on the annihilation process . from the left panel of figure [ fig : relativity effect on components ] , we can infer that the ds - ds contribution is affected the most by relativistic effects . on the one hand , the beaming effect due to the matter motion in the disk , together with the doppler effect and the gravitational redshift ,
reduces the annihilation rate in the funnel above the mns . on the other hand , the gravitational blueshift and the light bending occurring close to the mns and to the innermost neutrino surfaces increase the rates immediately above the mns surface .
the same effects are visible , even if less pronounced , for the ns&ds contributions . on the contrary , for the ns - ns contribution
, the dominant gravitational redshift reduces the annihilation rates almost everywhere . to quantify the global impact of the relativistic effects
, we integrate the different contributions over the deposition volume , according to eqs .
( [ eq : q non rel ] ) , ( [ eqn : relativisitic local volume integrated energy rate ] ) and ( [ eqn : relativisitic at infinity volume integrated energy rate ] ) .
the results are shown in the right panel of figure [ fig : relativity effect on components ] . in the case of the ds - ds contribution
, relativistic effects can even marginally increase the integrated energy deposition rate , as measured by local observers , leading to @xmath227 . in the case of the ns&ds and ns - ns contributions ,
the larger volume and the more pronounced decrease in the energy deposition efficiency at large radii compensate the increase close to the remnant , such that @xmath228 .
a more detailed analysis of the impact of special and general relativistic effects on the energy deposition rates is presented in [ apx : analysis rel effects ] . in eq .
( [ eq : simple parametrization ] ) , we have introduced a possible parametrization of the volume - integrated energy deposition rates , @xmath229 .
the advantage of this expression is to provide global information about the energy deposition rate based on basic properties of the neutrino emission ( @xmath35 , @xmath230 , neutrino mean energies ) and on the geometry of the system ( @xmath231 , @xmath232 ) . in the following ,
we verify the validity of our expression and compute the unknown geometrical factors .
we take again advantage of the luminosity decomposition and we evaluate the geometrical factors separately for each of the contributions appearing in eq .
( [ eq : q contributions ] ) , i.e. @xmath233 and @xmath234 with @xmath235 .
since our decomposition distinguishes among contributions presenting different emission properties , we further assume that @xmath236 in eq .
( [ eq : simple parametrization ] ) and obtain @xmath237 \ , , \label{eq : simple parametrization decomposed}\ ] ] where @xmath238 can be both electron or heavy flavor neutrino pairs . differently from before , here we distinguish between neutrinos coming from the mns and antineutrinos coming from the disk , and vice versa , for the @xmath25,@xmath21 contributions . in the left panel of figure [ fig : g factor lcool ] , we present the temporal evolution of @xmath239 for all the four different contributions to @xmath240 .
we tested both the usage of the luminosities at infinity ( solid lines ) , @xmath241 , and the cooling luminosities ( dashed line ) , @xmath242 ( see section [ sec : simulation ] ) in eq .
( [ eq : simple parametrization decomposed ] ) .
we first notice that , despite the different geometrical properties of the mns and disk emission , the geometrical factors show similar values and similar trends with time , within 10% of their average values .
this result corroborates the validity of parametrization ( [ eq : simple parametrization decomposed ] ) .
the geometrical factors obtained by the cooling luminosities exhibit a smaller spread , both among them and in time .
this reflects the different impact of neutrino absorption outside the neutrino surfaces for the different contributions .
in particular , the factor @xmath1 between the results computed using @xmath35 and @xmath243 can be explained as @xmath244 , where @xmath245 and @xmath246 are the average values of the energy optical depth at the last scattering surface . since for @xmath25 , the scattering and the energy optical depth are close ( see figure ( [ fig : nu_surfaces ] ) ) , @xmath247 , while for @xmath21 the difference is larger , @xmath248 .
the situation is qualitatively different for heavy flavor neutrinos ( not shown in the figure ) .
their values are such that @xmath249 , @xmath250 , and @xmath251 . since there is no significant @xmath22 absorption outside the neutrino surfaces , @xmath252 .
thus , if we compare with the results obtained with @xmath242 and @xmath253 , we conclude that the geometrical factors for @xmath22 are larger .
we speculate that the origin of these larger values depends on the smaller extension of the neutrino surfaces . on the one hand , this provides a more isotropic emission , but on the other hand leads to less spatially distributed emission and larger neutrino densities . after having explored the dependency of the energy deposition rates on the neutrino luminosities
, we can quantify the annihilation efficiency , as defined in eq .
( [ eq : annihilation efficiency ] ) . in the right panel of figure
[ fig : g factor lcool ] , we plot @xmath254 for the total rates and the four different contributions . in the latter cases , at the denominator , we select only the relevant luminosities , i.e. @xmath255 . also in this case , we compute @xmath254 both using the total ( solid lines ) and the cooling luminosities ( dashed lines ) , and obviously the former are systematically larger than the latter .
the annihilation efficiency of the single components ranges between 0.03% up to 0.15% .
the behavior of the different components and their relative strength results from the non - trivial combinations of neutrino luminosities , geometrical factors , and mean energies .
the most efficient energy conversion happens for the ds - ds contribution , while the less efficient is the ns - ns contribution .
we also compute the efficiency for the total rates ( orange line ) and find that it varies between 0.2% ( at the beginning , when the annihilation is dominated by the more efficient ds - ds and ns&ds contributions ) and 0.07% ( at the end , when the ns - ns contribution becomes more relevant ) .
we also notice that @xmath256 is larger than any other contribution .
this is due to the fact that each component of the luminosity powers two different energy deposition contributions .
thus , we can write @xmath257 , where @xmath258 . .table with the values obtained from the exponential interpolation of the different components of @xmath229 ( left ) , @xmath259 ( center ) , and @xmath260 ( right ) , for @xmath162 ( top ) and @xmath165 ( bottom ) , according to eq .
( [ eq : interpolation ] ) .
the integers in bracket correspond to the power of ten of the number . [ cols="^,^,^,^,^,^,^,^ " , ] the results obtained so far rely on calculations of the neutrino luminosities based on one single hydrodynamical model .
however , the large variety of initial conditions ( ns masses , mass ratios and spins ) are expected to translate into a potentially large variety of neutrino luminosities .
moreover , the uncertainties on the nuclear eos , as well as the impact of relativistic dynamics and different neutrino treatments , can introduce noticeably differences in the intensity of the neutrino luminosities between different numerical models .
specifically , gr merger simulations tends to produce hotter remnant and larger neutrino luminosities @xcite .
a similar effect is observed using softer nuclear eos .
moreover , gray moment transport schemes seem to suggest lower @xmath25 luminosities , compared with gray leakage schemes @xcite . to address the impact of diverse luminosities on the total deposited energy , we employ the proportionality between @xmath261 and @xmath229 in eq .
( [ eq : simple parametrization decomposed ] ) , separately for each of the four distinct contributions appearing in eq .
( [ eq : q contributions ] ) .
we first assume to rescale separately @xmath35 and @xmath230 , i.e. @xmath262 and @xmath263 , by a constant factor @xmath264 . in figure [
fig : effect of luminosity rescaling , nu - antinu ] , we present the total cumulative deposited energy as a function of @xmath265 and @xmath266 .
the four different panels refer to four different bh formation times and the @xmath267 case corresponds to our standard simulation .
luminosities that are twice as large as the computed ones lead to an energy deposition rate four times larger , due to the linear dependence of @xmath229 on the product of @xmath35 and @xmath230 .
assuming that neutrino pair annihilation is the only source of energy powering short grbs and that the deposited energy is converted very efficiently in kinetic energy and photons inside the relativistic jet , we can compare the grbs inferred energy with the energy computed in our models .
thus , in the same figures , we show also the inferred total energy from the observation of eleven short grbs , for which also an estimate of the jet opening angle is available , see table [ tab : grb observations ] , derived from . in four cases ( a - d )
, the temporal steepening of the afterglow decline rate is attributed to jet breaks , and this interpretation provides a measure of the jet opening angle . for the others ,
the absence of evidences of jet breaks translates in a lower limit for @xmath268 , depending on the time of the last observation . in these cases , an upper limit of @xmath269
is assumed @xcite .
following , the true energy is computed from the isotropized total energy according to @xmath270 in case the mns collapses quickly to a bh ( @xmath271 ) , the results obtained from our simulation are incompatible with all available observations .
also in the case of a long - lived mns ( @xmath272 ) , our results are compatible only with the lower limits of the less energetics grbs
. significantly larger luminosities ( @xmath273 ) are needed to explain the energetics of a large fraction of the observed short grbs .
a very delayed collapse to bh ( @xmath274 ) decreases the required increase in luminosity by a factor @xmath275 , compared with an early collapse ( @xmath276 ) .
nevertheless , the most energetic short grbs seem to require @xmath277 .
we repeat the calculations rescaling the ns and the ds contributions , without distinguishing @xmath25 from @xmath21 : @xmath278 and @xmath279 , and again @xmath280 . in figure
[ fig : effect of luminosity rescaling , ns - disk ] , we present the total cumulative deposited energy , extrapolated at one second , as a function of @xmath281 and @xmath282 , for four different bh formation times . since the ns - ns contribution is smaller than any other contribution involving radiation coming from the disk , there is no more symmetry between @xmath281 and @xmath282 , and variations along @xmath282 have a significantly larger impact .
thus , hotter disks and long - lived mns provide larger energy deposition , and they are necessary ingredients to explain short grbs energetics with neutrino pair annihilation ( at least for bursts of low and medium energy ) .
in this work , we have investigated the energy and momentum deposition operated by the annihilation of @xmath190-@xmath283 pairs above the remnant of a bns merger over a time scale comparable with the expected disk lifetime ( @xmath284 ) . in particular , we have analyzed the implications of a long - lived mns and studied the impact of relativistic effects on the neutrino propagation and annihilation . for our study , we have used results of the first long - term , three dimensional simulations of the aftermath of a bns merger after the influence of neutrino cooling and heating @xcite . the neutrino emission was modeled via a spectral leakage scheme and neutrino pair annihilation rates have been computed outside the neutrino surfaces using a detailed , spectral ray - tracing algorithm .
our major findings are : * the presence of a mns ( instead of a bh ) in the center increases the annihilation rate by a factor of @xmath1 , due to the interaction between neutrinos coming from the mns with antineutrinos coming from the disk , and vice versa .
moreover , it increases the efficiency at which neutrinos can be emitted from the disk , due to the larger emission time and slower accretion process on the mns surface ; * energy and momentum depositions operated by neutrino pair annihilation are more intense closer to the neutrino surfaces , i.e. immediately above the mns surface and above the region that marks the transition between the mns and the disk ( @xmath285 , @xmath286 ) .
neutrino annihilation above the mns occurs at lower polar angles , but larger matter densities , than neutrino annihilation above the innermost part of the disk ( @xmath287 ) .
the location of the annihilation rate and the level of baryonic pollution are expected to influence the dynamical effect of the energy and momentum deposition ; * the net momentum provided by neutrino annihilation is mostly pointing outwards : upwards above the mns and the densest part of the disk , more radially at large distances from the center . the efficiency at which net momentum is deposited , with respect to the energy deposition , is low close to the remnant ( @xmath288 ) , while efficient ( @xmath289 ) far from it ; * the energy and momentum depositions operated by heavy flavor neutrinos are more than one order of magnitude ( @xmath290 ) smaller than the ones by electron flavor neutrinos ; * volume - integrated energy deposition rates change from @xmath291 at the beginning of our simulation to @xmath292 at the end ( 400 ms ) .
they are , in good approximation , proportional to the product of the neutrino luminosities , @xmath293 .
interestingly , the proportionality term that contains all the geometrical dependences of the process ( @xmath231 ) is fairly similar for all the different contributions of @xmath229 and rather constant in time ; * the efficiency at which the emitted neutrinos annihilate above the remnant is rather low ( @xmath294 ) , especially in comparison with the efficiency at which neutrinos are emitted from the disk ( @xmath295 ) ; * the inclusion of relativistic effects in the neutrino propagation does not change the results we have obtained with the newtonian calculations qualitatively . from a more detailed quantitative analysis , we observe that the beaming of the neutrino radiation emitted from the rotating disk reduces the annihilation rate above the disk and in the funnel ; the gravitational redshift further reduces the amount of energy deposited at large distances ( a few tens of kilometers ) from the remnant , while close to the compact object neutrinos emitted from the disk experience gravitational blueshift and light bending , which increase their energy deposition efficiency , but reduce the momentum deposition one ; * we computed a cumulative energy deposition of @xmath296 at 400 ms and estimated @xmath2 at 1 s. an early collapse ( @xmath297 ) of the mns to a bh decreases the total energy deposition by @xmath298 , due to the progressive decrease of the annihilation rate .
a late collapse ( @xmath299 ) is significantly less relevant . in the following ,
we compare our results with some recent calculations of neutrino pair annihilation above compact binary mergers in the literature .
the initial conditions used in our bns merger aftermath simulation ( section [ sec : simulation ] ) are very close to the ones used by .
thus , this work allows a direct comparison , even if our calculations span a much longer time . in their work , computed the annihilation rates during the first 100 ms after the bns merger using two different approaches : first , a neutrino gray leakage scheme and an annihilation rate formalism based on and ; second , a @xmath300 neutrino transport scheme @xcite and a moment formalism for the annihilation rate calculations . comparing with our results , on the one hand our spatial distribution of the energy deposition rate , figure [ fig : anni rate different timesteps ] , is very similar to the one obtained by using their second , more accurate approach .
on the other hand , significant qualitative differences are visible when comparing with results obtained with their gray leakage scheme .
in particular , despite the usage of a leakage scheme , we do not observe a difference of five orders of magnitude between electron and heavy flavor neutrino annihilation rates , rather of a factor 10 - 100 , more similar to the @xmath300 results .
this comparison reveals that more accuracy can be reached using spectral approaches .
moreover , it confirms that taking into account that neutrinos diffusing from optically thick regions are ultimately emitted from the last scattering neutrino surfaces is key to describe the neutrino intensities and their angular distribution in optically thin conditions ( rather than being emitted isotropically from their production site ) .
this effect is more evident for heavy flavor neutrinos , since they decouple much deeper inside the remnant .
a more gentle decrease of the total deposited energy with time is observed in our calculations , compared with the steeper decrease found by and .
consequently , if we integrate the @xmath300 results obtained by during the first 100 ms , we obtain @xmath301 , smaller than our result @xmath302 .
this is due to the different temporal evolution of the neutrino luminosities between and .
this difference depends on the diverse neutrino treatments and dimensionality of the two models , as already discussed in .
the larger accretion rate obtained in our simulation powers more intense neutrino luminosities , which decrease more gradually than in
. computed neutrino annihilation rates in models of compact binary merger remnants using a monte carlo radiative transfer code for neutrinos , and comparing with a gray leakage scheme @xcite . if we restrict the volume integrals of the energy deposition rate only to the two cones of 45@xmath303 around the rotational axis , the energy deposited at 400 ms is @xmath304 , more than six times larger than their results at 3 s assuming a long - lived mns ( @xmath305 ) .
since we do not expect the energy deposition to be significantly large for @xmath306 , we explain this discrepancy with differences in the two models .
in particular , the differences that discussed in relation with apply also to our case . their mns is larger , while their disk is significantly smaller ( by a factor of 6 ) and less dense than ours . as a consequence ,
their optically thin disk produces lower luminosities , and a more spherical and dilute radiation distribution .
the different role of the disk in the two models is confirmed by the bh - torus results . in their calculations ,
the cumulative energy in the 45@xmath303 cones at 300 ms is @xmath307 , much lower than the corresponding disk - disk contribution we have computed in our model @xmath308 .
our disk - disk contribution can be also compared with the results of , where the annihilation of neutrinos above bh - torus systems was computed .
our results at 400 ms , @xmath309 , differ only by @xmath310 with the results of their equal mass ( 1.45 @xmath6 - 1.45 @xmath6 ) binary merger remnant calculation , despite the usage of different ns masses , eoss , hydrodynamics and neutrino treatments .
in particular , the smaller and faster decreasing neutrino luminosities are compensated by a larger global annihilation efficiency , @xmath311 , compared with our results ( @xmath312 at peak ) .
finally , we notice that the effects of the inclusion of relativistic effects in the neutrino propagation are qualitatively compatible with results presented in and , for bh - torus calculations . in the latter case
, the lack of an extended compact object in the center increases the positive effects close to the bh horizon and the annihilation rate , compared with newtonian calculations .
since our hot mns is rather extended ( @xmath313 ) , these effects are less relevant . for a softer nuclear eos , we expect a large impact and an increase of the energy deposition rates . the comparison of the deposited energy we have computed with the inferred energy of observed short grbs does not support the annihilation of neutrino pairs as a sufficient mechanism to power short grb jets .
neutrino pair annihilation is still a possible central engine only if the luminosities are significantly larger than the ones we have computed .
a factor of 2 - 3 is usually necessary to explain low energy short grbs , while a much larger increase ( @xmath314 ) is necessary for the most energetic ones .
smaller jet opening angles ( @xmath315 ) reduce the intrinsic grb energetics and require smaller luminosity magnification factors .
hotter merger remnants , emerging from general relativistic simulations , possibly employing softer nuclear eoss , are expected to power significantly larger neutrino luminosities ( up to a few times @xmath316 in the first ms after the merger ) .
however , it is unclear if such higher luminosities can be sustained for a long enough time scale .
moreover , general relativistic models predict less massive disks , with presumably smaller lifetimes .
the deposition of an amount of energy compatible with short grb energetics is not enough to explain the formation of a relativistic jet .
the presence of a relatively baryon - free region , where the deposition occurs , is also required to accelerate matter to relativistic speeds .
recently , simulated the formation and expansion of jets powered by neutrino pair annihilation in bh - torus systems .
they pointed out the possible role of the dynamic ejecta as an obstacle for the formation of a relativistic jet , especially in the case of bns mergers . on the one hand
, a long - lived mns can potentially pollute more heavily the regions above the remnant via baryon - rich ejecta , in the form of @xmath190-driven @xcite or magnetically - driven @xcite winds .
on the other hand , the presence of a large scale @xmath190-driven wind , mainly generated from the disk surface , can collimate the jet within a cone of small opening angle .
moreover , our results suggest that a long - lived mns increases the amount of deposited energy and net momentum significantly .
nevertheless , the required high luminosities and the difficulties in keeping the funnel above the remnant baryon - free suggest that it is unlikely that neutrino pair annihilation can power the formation of relativistic jets alone .
if they still provide the bulk of the jet energy , another mechanism is necessary to keep the funnel free from intense baryon contaminations .
otherwise , an alternative engine is needed and it can work together with neutrino annihilation . if this requires the gravitational collapse of the mns , the propagation of the jet inside non - relativistic baryon - rich ejecta imposes constraints to the duration of the mns phase , @xmath317 , @xcite .
we thank raffaella margutti , giancarlo ghirlanda and gabriele ghisellini for useful discussions .
we acknowledge support from the helmholtz - university investigator grant no .
vh - ng-825 , and from the european research council grant no .
677912 europium .
this work was also supported by a grant from the swiss national supercomputing centre ( cscs ) under project i d 667 .
in this appendix , we illustrate in detail the implementation of relativistic effects in our annihilation rate calculations .
we model the spacetime outside the mns as a stationary spherically symmetric spacetime generated by the mass @xmath138 . in
the global coordinates @xmath318 the metric @xmath144 is @xmath319 where @xmath320 . since @xmath321 , we neglect the effects of the accretion disk mass .
we also ignore the effect of the mns rotation on the spacetime .
the spin parameter of the mns is @xmath322 , where @xmath323 is the mns angular momentum extracted from our simulation .
although the mns spin is not negligible , @xmath324 is still significantly lower than 1 and @xmath325 .
thus , we assume that rotational effects on the neutrino propagation do not change the results obtained with the schwarzschild metric qualitatively .
we compute the energy ( @xmath326 ) and momentum ( @xmath327 ) deposition rates according to eqs .
( [ eq : dessart ] ) and ( [ eq : dessart momentum ] ) , valid in the local stationary frame of each annihilation point . in this locally inertial frame , we can introduce a fixed tetrad @xmath328 , i.e. an orthornormal tetrad fixed with respect to the spatial global coordinates .
this basis is defined such that @xmath329 , where @xmath330 is the minkoswki metric . for the fiducial observer associated with this frame ,
all the relevant physical quantities are defined as in the flat spacetime .
let s consider neutrinos of energy @xmath331 emitted around a direction @xmath332 from a point @xmath87 with emissivity @xmath333 , as measured by the local stationary observer .
these neutrinos travel to @xmath76 ( the annihilation point ) where they arrive with an energy @xmath334 and a direction @xmath335 , as measured by the local stationary observer .
the energy at the two points are related by the gravitational redshift formula : @xmath336 the direction @xmath335 is obtained by solving the geodetic motion of a null particle in the schwarzschild metric traveling from @xmath87 to @xmath76 . in section ( [ subsec : geodetic motion ] ) , we detail our method to compute the geodetic motion .
now we want to compute the neutrino intensity at b based on the emissivity at @xmath87 : the observer at @xmath76 measures a number of neutrinos equal to the number of particles emitted at @xmath87 towards @xmath76 , diminished by the particles absorbed along the propagation path .
we estimate the latter quantity similarly to what we have done in the non - relativistic case , eq .
( [ eq : energy conservation between a and b ] ) : @xmath337 expressing eq .
( [ eq : relation between a and b neutrinos in gr ] ) in terms of the local intensity at @xmath76 and emissivity at @xmath87 , we obtain @xmath338 and from that @xmath339 the calculation of @xmath340 is also reported in section ( [ subsec : geodetic motion ] ) . at the point @xmath87 , the radiation is emitted by matter moving with a velocity @xmath341 , as seen by the stationary observer .
the frame comoving with the fluid at @xmath87 is related to the local stationary frame by a lorentz boost @xmath342 .
if we define @xmath343 , @xmath344 , @xmath345 , and @xmath346 , then the energy measured by the local static observer and the emission energy in the comoving frame , @xmath347 , are related by the doppler shift formula : @xmath348 the emissivity transforms between the local and the comoving frames according to @xmath349 we notice here that , instead of the usual transformation law for the emissivity , stating that @xmath350 is a relativistic invariant , we adopted the receiver point of view , because we express the received intensity at @xmath76 based on the emissivity at @xmath87 , eq .
( [ eq : energy comoving frame ] ) . in this case
, the time interval does not transform as @xmath351 ( emitter s point of view ) , but as @xmath352 , due to the doppler effect . in the comoving frame
, the emission is assumed to originate from a volume @xmath353 and to be isotropic in the half plane defined by a direction @xmath354 , according to eq .
( [ eq : isotropic emissivity ] ) . to relate our relativistic calculations with the input data coming from our newtonian simulation
, we make the following assumptions .
first , we assume that eq .
( [ eq : eta and r_nu relationship ] ) is valid in the comoving reference frame .
second , we assume that the favored emission direction @xmath355 of our newtonian calculations corresponds to @xmath356 , the average emission direction as seen by the local stationary observer .
the sign of the scalar product @xmath357 , required in eq .
( [ eq : isotropic emissivity ] ) for eq .
( [ eqn : emissivity transformation law ] ) , is computed using @xmath358 and @xmath359 , two vectors parallel to @xmath360 and @xmath361 , respectively , and obtained from the lorentz transformation of the corresponding transformed propagation directions @xmath362 and @xmath363 : @xmath364\ ] ] and @xmath365 \ , .\ ] ] thus , @xmath366 \right ) .
\label{eq : sign of inner product}\end{aligned}\ ] ] in a first step , the spectral neutrino fluxes at the annihilation point are computed from the simulation data according to the expressions eqs .
( [ eqn : intensity in arrival point ] ) , ( [ eqn : emissivity transformation law ] ) and ( [ eq : sign of inner product ] ) , on an energy grid shifted with respect to the grid used in our hydrodynamics simulation , according to eqs .
( [ eqn : energy shift ] ) and ( [ eq : energy comoving frame ] ) . in a second step
, the spectral intensities are interpolated back on the original energy grid .
the interpolation is performed in such a way that the energy - integrated neutrino flux is preserved . here
, we clarify the method we have adopted to compute the geodetic motion of a null particle between @xmath87 and @xmath76 in the metric ( [ eqn : metric ] ) . we first find the rotation @xmath367 that transforms the global coordinate system to a new coordinate system @xmath368 such that the origin , and the points @xmath87 and @xmath76 lie on the plane @xmath369 . due to angular momentum conservation
, the geodetic motion of neutrinos is also confined on this plane .
we define @xmath370 and @xmath371 as the polar angles of @xmath87 and @xmath76 on the @xmath369 plane , and @xmath372 , where @xmath373 is the angle between the position vectors of @xmath87 and @xmath76 in the original coordinate system .
due to the axisymmetry of the problem around the @xmath374 axis , the choice of @xmath375 is arbitrary . since @xmath367 is a spatial rotation , @xmath376 and @xmath377 .
the solution of the elliptic integrals that describe exactly the geodetic motion would be computationally too expensive for our annihilation rate calculations . thus , we use a method derived from the approximated analytic expression provided in .
the trajectory of a null particle , expressed in polar coordinates @xmath378 on the @xmath369 plane , can be approximated by @xmath379^{1/2 } \nonumber \\ & - \frac{r_g ( 1-\cos{(\phi'-\delta)})}{2(1+\cos{(\phi'-\delta ) } ) } \ , .
\label{eq : second trajectory equation } \end{aligned}\ ] ] in the previous expression , @xmath380 is the impact parameter , one of the trajectory integrals , and at any point of the trajectory , it can be related to the propagation angle @xmath381 by @xmath382 the propagation angle @xmath381 is defined as the angle between the local propagation direction of the neutrino and the local radial direction , measured in clockwise direction . if we denote @xmath383 as the initial propagation angle , the deflection angle @xmath384 is equal to @xmath385 concerning @xmath386 , if @xmath387 , then @xmath388 . in this case
, @xmath389 takes values in the interval @xmath390 $ ] , such that @xmath391 and @xmath392 . on the other hand , if @xmath393 , then @xmath394 . in this other case
, @xmath389 takes values in the interval @xmath395 $ ] , and @xmath391 and @xmath396 .
the approximated formula replaces the exact elliptic integrals with high accuracy for @xmath397 .
this approximation is justified because in our case @xmath398 and @xmath399 .
let s assume first that @xmath400 .
our method consists on the numerical solution of the equation @xmath401 , where @xmath402^{1/2 } \nonumber \\ & + \frac{r_g ( 1-\cos{(\phi_b-\delta)})}{2(1+\cos{(\phi_b-\delta ) } ) } \ , , \label{eqn : newrap equation}\end{aligned}\ ] ] obtained by imposing the passage of the trajectory from @xmath87 _ and _ @xmath76 . if a solution of eq .
( [ eqn : newrap equation ] ) is found , it allows the calculation of the propagation directions @xmath403 and @xmath404 in the rotated local reference frames . in the cases where @xmath405 , we use the time invariance of the geodetic motion .
we solve for the trajectory of a neutrino traveling from @xmath76 to @xmath87 and then invert the propagation directions @xmath406 .
finally , we apply the inverse rotation @xmath407 to these vectors to find the components of @xmath332 and @xmath335 for the original local stationary observers .
the quantity @xmath408 , required in eq .
( [ eqn : intensity in arrival point ] ) , is computed based on our approximated expressions of the trajectory of the null geodesics . in particular
, we consider radiation emitted at @xmath87 with an initial small spread @xmath409 in the propagation angle around the direction @xmath410 , such that @xmath411 .
this radiation reaches @xmath76 with a central direction @xmath412 , within a surface @xmath413 .
the quantity @xmath414 is the length of the perpendicular displacement @xmath415 , defined such as @xmath416 assuming to have expressed the final direction as @xmath417 .
the displacement can also be expressed in terms of the variation of the polar coordinates @xmath418 if we consider the expression of the trajectory , eq .
( [ eqn : newrap equation ] ) , as a function of both @xmath148 and @xmath383 , we compute the total differential as : @xmath419 comparing eqs .
( [ eqn : expression for displacement 1 ] ) and ( [ eqn : expression for displacement 2 ] ) and using ( [ eqn : trajectory differentials ] ) , we obtain @xmath420 ^ 2 \ , .
\label{eq : ratio domega da}\ ] ]
special and general relativistic corrections have potentially different effects on the four different contributions to the energy deposition rate presented in eq . ( [ eq : q contributions ] ) , due to the different origin and emission geometry . in the color
coded panels of figure [ fig : detailed relativity effect on components ] , we show the ratios between @xmath421 computed with different levels of approximations for the propagation of the neutrinos outside the neutrino surfaces .
the rates refer to 40 ms in our simulation .
we consider the standard newtonian rates , @xmath214 , ( section [ sec : method newtonian ] ) , as well as the relativistic ones , @xmath326 ( section [ sec : method rel ] and [ apx : implementation rel effects ] ) . as intermediate step , we consider also @xmath422 , the energy deposition rate obtained by taking into account only special relativistic effects due to the fast motion of matter inside the remnant ( relativistic doppler and beaming ) , but neglecting all the relativistic effects due to the curved spacetime ( i.e. , we assume @xmath423 in the spacetime metric ) .
the different quadrants correspond to the different contributions to @xmath421 .
the ratio @xmath424 ( left panel ) shows that the beaming effect and the doppler effect reduce the efficiency of the annihilation process significantly in the funnel and , more in general , far from the disk plane , for all the different contributions . at the same time
, they increase the energy deposition efficiency close to the remnant .
since neutrinos coming from the disk are more subject to these effects , the largest relative increase is verified for the ds - ds contribution .
general relativistic effects modify special relativistic deposition rates mainly in the proximity of the mns , where the curvature is more pronounced . in particular , the ns - ns contribution benefits only of the radiation bending immediately above the mns , while in the rest of the volume the redshift decreases the energy deposition rate . for the ds - ds contribution , the better collision angle due to radiation bending and the gravitational blueshift of neutrinos moving towards smaller radii increase significantly the deposition energy efficiency . a similar ,
although weaker , effect is observed also in the case of the ns&ds contribution . in this case , neutrinos coming from the mns are more subject to the gravitational redshift and this explains the faster decrease of the @xmath214 ratio at larger radii , compared with the ds - ds case .
xx g , salafia o s , pescalli a , ghisellini g , salvaterra r , chassande - mottin e , colpi m , nappo f , davanzo p , melandri a , bernardini m g , branchesi m , campana s , ciolfi r , covino s , gtz d , vergani s d , zennaro m tagliaferri g 2016 _ a&a _ * 594 * , a84 . | binary neutron star mergers are plausible progenitor candidates for short gamma - ray bursts ( grbs ) ; however , a detailed explanation of their central engine is still lacking . the annihilation of neutrino pairs has been proposed as one of the possible powering mechanisms .
we present calculations of the energy and momentum deposition operated by neutrino pair annihilation above merger remnants .
starting from the results of a detailed , three - dimensional simulation of the aftermath of a binary neutron star merger , we compute the deposition rates over a time scale comparable to the life time of the disk ( @xmath0 s ) , assuming a long - lived massive neutron star ( mns ) .
we model neutrino emission using a spectral leakage scheme and compute the neutrino annihilation rates using a ray - tracing algorithm .
we find that the presence of the mns increases the energy deposition rate by a factor @xmath1 , mainly due to the annihilation of radiation coming from the mns with radiation coming from the disk .
we compute the impact of relativistic effects and discover that , despite they can significantly change the local rate intensity , the volume - integrated results are only marginally decreased .
the cumulative deposited energy , extrapolated to 1 sec , is @xmath2 .
a comparison with the inferred short grb energetics reveals that in most cases this energy is not large enough , even assuming small jet opening angles and a long - lived mns .
significantly more intense neutrino luminosities ( a factor 5 - 10 larger ) are required to explain most of the observed short grb .
we conclude that it is unlikely that neutrino pair annihilation can explain the central engine of short grbs alone .
december 2016 _ keywords _ : accretion disks , gamma - ray burst , neutrinos , relativistic processes , stars : neutron . |
exclusive production of the higgs boson can be an alternative to the present studies of higgs in inclusive processes .
there is recently a growing theoretical interest in studying exclusive processes .
only a few processes have been measured so far , mostly at the tevatron ( see @xcite and references therein ) .
khoze , martin and ryskin developed an approach in the language of off - diagonal unintegrated gluon distributions .
this approach was applied to exclusive production of higgs boson @xcite . in our recent papers
we applied the same formalism to exclusive production of @xmath0 and @xmath2 quarks .
quite large cross sections have been found @xcite .
the cross section for the standard model higgs production is of the order of 1 fb for light higgs @xcite .
the dominant @xmath2 decay channel is therefore preferential from the point of view of statistics .
it was argued that the leading - order contribution is rather small using a so - called @xmath6 = 0 rule .
here we show a quantitative calculation which goes beyond this simple rule . in our calculation
we include exact matrix element for massive quarks and the 2 @xmath7 4 phase space .
this fully four - body calculation allows to impose cuts on kinematical variables .
different types of backgrounds to higgs production were studied before e.g. in ref.@xcite .
let us concentrate on the simplest case of the production of @xmath8 pair in the color singlet state .
color octet state would demand an emission of an extra gluon which considerably complicates the calculations .
we do not consider the @xmath9 contribution as it is higher order compared to the one considered here .
we write the amplitude of the exclusive diffractive @xmath8 pair production @xmath10 in the color singlet state as + @xmath11 where @xmath12 are helicities of heavy @xmath13 and @xmath14 , respectively . above @xmath15 and @xmath16
are the off - diagonal unintegrated gluon distributions in nucleon 1 and 2 , respectively .
the longitudinal momentum fractions of active gluons are calculated based on kinematical variables of outgoing quark and antiquark : @xmath17 and @xmath18 , where @xmath19 and @xmath20 are transverse masses of the quark and antiquark , respectively , and @xmath21 and @xmath22 are corresponding rapidities . the bare amplitude above is subjected to absorption corrections . the absorption corrections are taken here in a simple multiplicative form .
let us consider the subprocess amplitude for the @xmath8 pair production via off - shell gluon - gluon fusion .
the vertex factor @xmath23 in expression ( [ amplitude ] ) is the production amplitude of a pair of massive quark @xmath13 and antiquark @xmath14 with helicities @xmath24 , @xmath25 and momenta @xmath26 , @xmath27 , respectively .
the color singlet @xmath8 pair production amplitude can be written as @xcite @xmath28 the tensorial part of the amplitude reads : @xmath29 the coupling constants @xmath30 . in the present calculation
we take the renormalization scale to be @xmath31 or @xmath32 .
the exact matrix element is calculated numerically .
analytical formulae are shown explicitly in @xcite .
the off - diagonal parton distributions ( i=1,2 ) are calculated as @xmath33}{d \log k_t^2 } |_{k_t^2 = q_{it}^2 } \ ; f(t_i ) \end{array } \label{kmr - off - diagonal - ugdfs}\ ] ] where @xmath34 is a sudakov - like form factor relevant for the case under consideration .
it is reasonable to take a running ( factorization ) scale as : @xmath35 or @xmath32 .
the factor @xmath36 here can not be calculated from first principles in the most general case of off - diagonal ugdfs .
it can be estimated in the case of off - diagonal collinear pdfs when @xmath37 and @xmath38 . then @xmath39 .
typically @xmath40 1.3 1.4 at the tevatron energy .
the off - diagonal form factors are parametrized here as @xmath41 . in practical calculations
we take @xmath42 = 2 gev@xmath43 . in the original kmr approach the following prescription for the effective transverse momentum
is taken : @xmath44 and @xmath45 . in evaluating @xmath46 and
@xmath47 needed for calculating the amplitude ( [ amplitude ] ) we use different collinear distributions .
it was proposed @xcite to express the @xmath48 form factors in eq .
( [ kmr - off - diagonal - ugdfs ] ) through the standard sudakov form factors as : @xmath49 the cross section for the four - body reaction is calculated as @xmath50 the details how to conveniently reduce the number of kinematical integration variables are given elsewhere .
let us proceed now with the presentation of differential distributions of charm quarks produced in the edd mechanism .
in this case we have fixed the scale of the sudakov form factor to be @xmath52 .
such a choice of the scale leads to a strong damping of the cases with large rapidity gaps between @xmath13 and @xmath53 . in the left panel of fig .
[ figx ] we show distribution in rapidity .
the results obtained with the kmr method are shown together with inclusive gluon - gluon contribution .
the effect of absorption leads to a damping of the cross section by an energy - dependent factor . for the tevatron this factor is about 0.1 .
if the extra factor is taken into account the edd contribution is of the order of 1% of the dominant gluon - gluon fusion contribution .
the corresponding rapidity - integrated cross section at @xmath1 = 1960 gev is : 6.6 @xmath54b for the exact formula , 2.4 @xmath54b for the simplified formula ( see eq .
( [ kmr - off - diagonal - ugdfs ] ) ) . for comparison
the inclusive cross section ( gluon - gluon component only ) is 807 @xmath54b . or @xmath55 ( left ) and transverse momentum distribution of @xmath56 or @xmath55 ( right ) . the top curve is for inclusive production @xcite while the two lower lines are for the edd mechanism with leading - order collinear gluon distribution @xcite .
the solid line is calculated from the exact formula and the dashed line for the simplified formula @xcite .
an extra cut on the momenta in the loop @xmath57 = 0.26 gev@xmath58 was imposed .
absorption effects were included by multiplying the cross section by the gap survival factor @xmath59 = 0.1 .
, title="fig:",width=188 ] or @xmath55 ( left ) and transverse momentum distribution of @xmath56 or @xmath55 ( right ) .
the top curve is for inclusive production @xcite while the two lower lines are for the edd mechanism with leading - order collinear gluon distribution @xcite .
the solid line is calculated from the exact formula and the dashed line for the simplified formula @xcite .
an extra cut on the momenta in the loop @xmath57 = 0.26 gev@xmath58 was imposed .
absorption effects were included by multiplying the cross section by the gap survival factor @xmath59 = 0.1 .
, title="fig:",width=188 ] in the right panel of fig .
[ figx ] we show the differential cross section in transverse momentum of the charm quark .
compared to the inclusive case , the exclusive contribution falls significantly faster with transverse momentum than in the inclusive case . in parallel to the exclusive @xmath2 production , we calculate the differential cross sections for exclusive higgs boson production .
compared to the standard kmr approach here we calculate the amplitude with the hard subprocess @xmath61 taking into account off - shellness of the active gluons .
the details of the off - shell matrix element can be found in ref .
in contrast to the exclusive @xmath62 production @xcite , due to a large factorization scale @xmath63 the off - shell effects for @xmath61 give only a few percents .
the same unintegrated gluon distributions based on the collinear distributions are used for the higgs and continuum @xmath64 production . in the case of exclusive higgs production
we calculate the four - dimensional distribution in the standard kinematical variables : @xmath65 and @xmath66 . assuming the full coverage for outgoing protons we construct the two - dimensional distributions @xmath67 in higgs rapidity and transverse momentum .
the distribution is used then in a simple monte carlo code which includes the higgs boson decay into the @xmath68 channel .
it is checked subsequently whether @xmath4 and @xmath5 enter into the pseudorapidity region spanned by the central detector .
invariant mass distribution for @xmath1 = 14 tev and for @xmath69 corresponding to the atlas / cms detectors .
the absorption effects were taken into account by multiplying by the gap survival factor @xmath59 = 0.03.,width=226 ] invariant mass distribution for @xmath1 = 14 tev and for @xmath4 and @xmath5 jets in the rapidity interval @xmath70 corresponding to the atlas detector .
the absorption effects for the higgs boson and the background were taken into account by multiplying by the gap survival factor @xmath59 = 0.03 .
the left panel shows purely theoretical predictions , while the right panel includes experimental effects due to experimental uncertainty in invariant mass measurement.,title="fig:",width=226 ] invariant mass distribution for @xmath1 = 14 tev and for @xmath4 and @xmath5 jets in the rapidity interval @xmath70 corresponding to the atlas detector .
the absorption effects for the higgs boson and the background were taken into account by multiplying by the gap survival factor @xmath59 = 0.03 .
the left panel shows purely theoretical predictions , while the right panel includes experimental effects due to experimental uncertainty in invariant mass measurement.,title="fig:",width=226 ] in fig .
[ fig : dsigma_dmbb_pdfs ] we show the most essential distribution in the invariant mass of the centrally produced @xmath2 pair , which is also being the missing mass of the two outgoing protons . in this calculation
we have taken into account typical detector limitations in rapidity @xmath71 .
we show results with different collinear gluon distributions from the literature : grv @xcite , cteq @xcite , gjr @xcite and mstw @xcite .
the results obtained with radiatively generated gluon distributions ( grv , gjr ) allow to use low values of @xmath72 whereas for other gluon distributions an upper cut on @xmath73 is necessary .
the lowest curve in fig.[fig : dsigma_dmbb_pdfs ] represents the @xmath74 contribution @xcite .
while the integrated over phase space @xmath74 contribution is rather small , it is significant compared to the double - diffractive component at large @xmath75 100 gev .
this can be understood by a damping of the double diffractive component at large @xmath76 by the sudakov form factor @xcite . in addition , in contrast to the double - diffractive component the absorption for the @xmath74 component is very small and in practice can be neglected . in the left panel of fig.[fig : dsigma_dmbb_fully ] we show the double diffractive contribution for a selected ( cteq6 @xcite ) collinear gluon distribution and the contribution from the decay of the higgs boson including natural decay width calculated as in ref .
@xcite , see the sharp peak at @xmath76 = 120 gev .
the phase space integrated cross section for the higgs production , including absorption effects with @xmath77 is somewhat less than 1 fb . the result shown in fig.[fig : dsigma_dmbb_fully ] includes also the branching fraction for br(@xmath78 0.8 and the rapidity restrictions .
the second much broader breit - wigner type peak corresponds to the exclusive production of the @xmath79 boson with the cross section calculated as in ref .
the exclusive cross section for @xmath1 = 14 tev is 16.61 fb including absorption .
the branching fraction br(@xmath80 0.15 has been included in addition .
in contrast to the higgs case the absorption effects for the @xmath79 production are much smaller @xcite .
the sharp peak corresponding to the higgs boson clearly sticks above the background . in the above calculations we have assumed an ideal measurement . in reality
the situation is , however , much worse as both protons and in particular @xmath4 and @xmath5 jets are measured with a certain precision which automatically leads to a smearing in @xmath76 .
experimentally instead of @xmath76 one will measure rather two - proton missing mass ( @xmath81 ) .
the experimental effects are included in the simplest way by a convolution of the theoretical distributions with the gaussian smearing function @xmath82 with @xmath83 = 2 gev which is determined mainly by the precision of measuring forward protons . in the right panel
we show the two - proton missing mass distribution when the smearing is included .
now the bump corresponding to the higgs boson is below the @xmath2 background . with the experimental resolution assumed above the identification of
the standard model higgs seems rather difficult .
the situation for some scenarios beyond the standard model may be better .
can the situation be improved by imposing further cuts ? in fig .
[ fig : y3y4_edd ] we show the distribution for the edd background in @xmath86 and @xmath84 .
in contrast to the higgs @xcite the cross section for the @xmath2 continuum has maxima far from the diagonal . this can be used to impose cuts on quark / antiquark rapidities . in fig .
[ fig : dsigma_dmbb_cuts ] ( left panel ) we show the result for a more limited range of @xmath4 and @xmath5 rapidity , i.e. not making use of the whole coverage of the main lhc detectors . here we omit the @xmath79 contribution and concentrate solely on the higgs signal .
now the signal - to - background ratio is somewhat improved .
this would be obviously at the expense of a deteriorated statistics .
similar improvements of the signal - to - background ratio can be obtained by imposing cuts on jet transverse momenta .
detailed studies of the role of cuts is discussed in @xcite .
v. a. khoze , a. d. martin and m. g. ryskin , phys .
b * 401 * , 330 ( 1997 ) ; + a. b. kaidalov , v. a. khoze , a. d. martin and m. g. ryskin , eur .
j. c * 33 * , 261 ( 2004 ) .
r. maciua , r. pasechnik and a. szczurek , phys .
b * 685 * , 165 ( 2010 ) .
r. maciua , r. pasechnik and a. szczurek , arxiv:1006.3007 [ hep - ph ] , in print in phys .
* d*. r. s. pasechnik , o. v. teryaev and a. szczurek , eur .
j. c * 47 * , 429 ( 2006 ) .
r. s. pasechnik , a. szczurek and o. v. teryaev , phys .
d * 78 * , 014007 ( 2008 ) ; + r. s. pasechnik , a. szczurek and o. v. teryaev , t.d .
coughlin and j.r .
forshaw , jhep 1001 , 121 ( 2010 ) . | we discuss exclusive double diffractive ( edd ) production of higgs boson and heavy quark - heavy antiquark pairs at high energies .
differential distributions for @xmath0 at @xmath1 = 1.96 gev and for @xmath2 at @xmath1 = 14 tev are shown and discussed .
irreducible leading - order @xmath3 background to higgs production is calculated in several kinematical variables .
the signal - to - background ratio is shown and several improvements are suggested by imposing cuts on @xmath4 ( @xmath5 ) transverse momenta and rapidities . |
it is known that the spatial derivative of the solution to the ( backward ) kolmogorov equation can be represented as an expectation of a functional of the solution of an sde with some weight , namely the so - called bismut - elworthy - li ( bel ) formula as shown in @xcite and extended in @xcite . in @xcite
the authors use techniques from malliavin calculus to prove bel formula and employ it for the computation of sensitivities of financial options , also known as _
greeks_. in many applications , it is very natural to expect that the coefficients of a stochastic differential equation ( sde ) may depend on properties of the law of the solution , such as dependence on its moments .
here , we want to extend the formula to mean - field type sdes following the essence of @xcite and show that such generalisation is actually non - trivial , requiring more regularity of the solution in the sense of malliavin .
first , we give a relationship between the malliavin derivative and the spatial derivative of the solution with respect to the initial condition .
already here we see that such generalisation involves an extra factor which is no longer adapted , thus requiring more ( malliavin ) regularity on the solution which is not immediate .
fortunately , if @xmath0 and @xmath1 are lipschitz continuous in space , then the solution is twice malliavin differentiable , as it is shown in @xcite , and hence a formula using the skorokhod integral may be expected .
using such relation one can find the bel formula in this context .
some merely illustrative examples are provided in order to give a better insight on the effect of mean - field sdes in the bel formula . in the last examples we carry out some simulations to show that the malliavin method is more efficient compared to a finite difference method , especially when the function involved is discontinuous .
the paper is organised as follows : in section [ frame ] we collect some summarised basic facts on malliavin calculus needed for the derivation of the main results of the paper . in section [ sectionbel ] we include all intermediate steps towards the main result which is the bismut - elworthy - li formula in the context of mean - field sdes .
finally , section [ sectionappl ] is devoted to provide some illustrative examples of this generalised bismut - elworthy - li formula with simulations .
the findings are similar to those in @xcite , the use of bismut - elworthy - li s formula serves as a much more efficient method to compute sensitivities with respect to initial data , especially when the function involved in the expectation is highly irregular .
* notations : * let @xmath2 denote the non - negative real numbers . denote by @xmath3 the euclidean norm in @xmath4 , @xmath5 .
given a banach space @xmath6 , denote by @xmath7 its associated norm .
let @xmath8 integers and @xmath9 be the space of @xmath10 times malliavin differentiable random variables with all @xmath11-moments .
denote by @xmath12 , @xmath13 the malliavin derivative as introduced in ( * ? ? ?
* chapter 1 , section 1.2.1 ) and @xmath14 its dual operator ( skorokhod integral ) . denote by @xmath15 the domain of @xmath14 ( skorokhod integrable processes ) .
denote the trace of a matrix @xmath16 by @xmath17 and by @xmath18 its transpose .
for a ( weakly ) differentiable function @xmath19 , @xmath20 , denote by @xmath21 , respectively by @xmath22 , ( weak ) differentiation with respect to the first ( space ) variable @xmath23 , respectively the second ( space ) variable @xmath24 .
our main results centrally rely on tools from malliavin calculus .
we here provide a concise introduction to the main concepts in this area . for deeper information on malliavin calculus
the reader is referred to i.e. @xcite .
let @xmath25\}$ ] be a standard wiener process on some complete filtered probability space @xmath26 where @xmath27}$ ] is the @xmath28-augmented natural filtration .
denote by @xmath29 the set of simple random variables @xmath30 in the form @xmath31 ) , \ \
f\in c_0^{\infty } ( { \mathbb r}^n).\ ] ] the malliavin derivative operator @xmath32 acting on such simple random variables is the process @xmath33\}$ ] in @xmath34)$ ] defined by @xmath35define the following norm on @xmath29 : @xmath36))}=e[|f|^2]^{1/2}+ e\left [ \int_0^t |d_t f|^2 dt\right]^{1/2}.\end{aligned}\ ] ] we denote by @xmath37 the closure of the family of simple random variables @xmath29 with respect to the norm given in and we will refer to this space as the space of malliavin differentiable random variables in @xmath38 with malliavin derivative belonging to @xmath38 . in the derivation of the probabilistic representation for the delta , the following chain rule for the malliavin derivative will be essential : [ chainrule ] let @xmath39 be continuously differentiable with bounded partial derivatives .
further , suppose that @xmath40 is a random vector whose components are in @xmath37 .
then @xmath41 and @xmath42.\ ] ] the malliavin derivative operator @xmath43)$ ] admits an adjoint operator @xmath44 where the domain @xmath45 is characterised by all @xmath46)$ ] such that for all @xmath47 we have @xmath48 \leq c \|f\|_{1,2},\ ] ] where @xmath49 is some constant depending on @xmath50 . for a stochastic process
@xmath51 ( not necessarily adapted to @xmath52 ) we denote by @xmath53 the action of @xmath14 on @xmath50 .
the above expression ( [ skorokhod ] ) is known as the skorokhod integral of @xmath50 and it is an anticipative stochastic integral .
it turns out that all @xmath52-adapted processes in @xmath54)$ ] are in the domain of @xmath14 and for such processes @xmath50 we have @xmath55 i.e. the skorokhod and it integrals coincide . in this sense
, the skorokhod integral can be considered to be an extension of the it integral to non - adapted integrands . as for the it integral , there is also a corresponding isometry property for the skorokhod integral .
the proof of this can e.g. be found in ( * ? ? ?
* theorem 6.17 . ) .
let @xmath50 be a process such that @xmath56 for a.e .
@xmath57 $ ] and @xmath58 < \infty.\ ] ] then @xmath50 is skorokhod integrable and @xmath59 = e\left [ \int_0^t u(t)^2 dt+ \int_0^t \int_0^t d_t u(s ) d_s u(t ) dsdr\right].\ ] ] the dual relation between the malliavin derivative and the skorokhod integral implies the following important formula : [ duality ] let @xmath47 be @xmath60-measurable and @xmath61 . then @xmath62 = e\left [ \int_0^t u(t ) d_t f dt\right].\end{aligned}\ ] ] the following is the corresponding integration by parts formula for the skorokhod integral .
see e.g. ( * ? ? ?
* theorem 3.15 . ) . [ ibp ]
let @xmath61 and @xmath47 such that @xmath63
. then @xmath64
the object of study is a _
mean - field _ type _ stochastic differential equation _ ( sde ) of the form @xmath65\\ \rho_t : = & \ , e[\varphi(x_t ) ] , \quad \pi_t : = e[\psi(x_t ) ] \end{split}\end{aligned}\ ] ] where @xmath66 , @xmath67 , @xmath68\times { \mathbb r}^d \times { \mathbb r}^d \rightarrow { \mathbb r}^d$ ] , @xmath69\times { \mathbb r}^d \times { \mathbb r}^d \rightarrow { \mathbb r}^d$ ] , @xmath70 , @xmath71 , @xmath72 are measurable functions and @xmath25\}$ ] is an @xmath73-dimensional brownian motion on some probability space @xmath26 equipped with the natural filtration augmented by all @xmath28-null sets , denoted by @xmath27}$ ] .
we will usually consider the solution as a function of @xmath74 and hence write @xmath75 to stress this fact .
otherwise , we will just write @xmath76 . moreover , we will assume the following conditions as in @xcite * the functions @xmath77 and @xmath78 , @xmath79 are continuously differentiable with bounded lipschitz derivatives uniformly with respect to @xmath80 $ ] .
* assume @xmath81 and the matrix @xmath82 is uniformly elliptic and admits a right pseudo - inverse .
* the functions @xmath83 and @xmath84 are continuously differentiable with bounded lipschitz derivatives .
the following proposition is extracted from the work by buckdahn , li , peng and rainer in @xcite where they show that under the hypothesis above , equation admits a stochastic flow of diffeomorphisms , in particular , the function @xmath85 is @xmath28-a.s .
classically differentiable .
this property is crucial to define the first variation process and relate it to the malliavin derivative of the solution .
[ diff ] let @xmath86\}$ ] be the unique global strong solution of ( [ sde ] ) .
then the function @xmath85 is continuously differentiable , @xmath28-a.s .
see @xcite .
next proposition shows that the first variation process is invertible for every @xmath87 .
[ detyt ] let @xmath88\}$ ] be the solution to the following matrix - valued linear sde @xmath89\ ] ] where @xmath90 and @xmath91 , @xmath79 .
then @xmath92 as a consequence , @xmath93 is @xmath28-a.s .
invertible for every @xmath80 $ ] .
we want to show that @xmath94 < \infty\ ] ] for every integer @xmath95 .
indeed , in virtue of ( stochastic ) liouville s formula which can be found in @xcite , one has @xmath96 observe that since the processes @xmath97 are in @xmath98)$ ] and @xmath52-adapted , the stochastic integrals appearing in the exponent are martingales .
hence , @xmath99 is an @xmath52-martingale with @xmath100 = e[m_0]=1 $ ] .
thus , using cauchy - schwarz inequality we have for every @xmath95 @xmath101 \leq e\left[\exp \left\{(2p^2-p)\sum_{k=1}^m \int_0^t { \mathrm{tr}}b_u^k dw_u^k-2p \int_0^t { \mathrm{tr}}a_u du\right\ } \right].\end{aligned}\ ] ] in particular , the claim is reduced to showing that @xmath102 } \left|e\left[\exp\left\{\lambda \int_0^t { \mathrm{tr}}a_u du\right\}\right]\right| + \sup_{t\in [ 0,t ] } \left|e\left[\exp\left\{\lambda \int_0^t \sum_{k=1}^m ( { \mathrm{tr}}b_u^k)^2 du\right\}\right]\right|<\infty\ ] ] for every @xmath103 which clearly holds since @xmath104 and @xmath105 , @xmath79 are uniformly bounded .
the following statement is one of the main observations for the derivation of the bismut - elworthy - li formula in the mean - field context .
it can be seen as a generalisation of the well - known relation between the first variation process @xmath106 in the non mean - field context and the malliavin derivative , see e.g. ( * ? ? ?
* ch.2 , sec.2.3.1 ) .
[ mallsob ] let @xmath86\}$ ] be the solution of
. then for every @xmath107 $ ] , @xmath108 one has the following relationship between the spatial derivative and the malliavin derivative of @xmath75 @xmath109 where @xmath110 @xmath111 denotes the right pseudo - inverse of @xmath1 , @xmath88\}$ ] is the @xmath112 fundamental matrix satisfying @xmath89\ ] ] and where @xmath113\}$ ] , @xmath114\}$ ] , @xmath115\}$ ] , @xmath116\}$ ] , @xmath70 are matrix valued processes defined as : @xmath117 for @xmath79 .
differentiating with respect to @xmath118 we have that @xmath119 satisfies the following matrix - valued linear equation @xmath120 using the notations in the statement of the theorem , we can solve ( [ matrixjacobian ] ) and express @xmath121 as @xmath122 where @xmath88\}$ ] is the @xmath123 fundamental matrix satisfying @xmath124 and @xmath125.\ ] ] by the well - known classical relation , see e.g. ( * ? ? ?
* chapter 2 , section 2.3.1 ) , it is true that , @xmath126 where @xmath111 denotes the right pseudo - inverse of @xmath1 and hence the relation follows . for the relation @xmath127 , @xmath108 to hold in the mean - field setting one also needs the property that @xmath85 defines a stochastic semiflow . in the mean - field case
we point out that the fact that @xmath0 , @xmath1 , @xmath83 and @xmath84 are continuously differentiable with bounded lipschitz derivative implies this fact in virtue of @xcite .
it is shown in @xcite that sde ( [ sde ] ) is twice malliavin differentiable when the vector field @xmath0 does not depend on the law of @xmath128 and one has additive noise . nevertheless , using the same method one can prove the same result since the dependence on @xmath129 $ ] and @xmath130 $ ] does not bring stochasticity to the equation . in the sense that the malliavin derivative of @xmath76 for every fixed @xmath80 $ ] takes the same form as in the usual linear setting .
henceforth , we will assume the following technical condition for simplicity .
* there exists a bijection @xmath131 , for every @xmath80 $ ] , such that the function @xmath132 \times { \mathbb r}^d\times { \mathbb r}^d \times { \mathbb r}^d \rightarrow { \mathbb r}^d$ ] defined as @xmath133\\ & + \frac{1}{2 } \partial_x^2 \lambda_t(\lambda_t^{-1}(x ) ) \left [ \sum_{i=1}^d \sigma(t,\lambda_t^{-1}(x),\pi_t)[e_i ] , \sum_{i=1}^d \sigma(t,\lambda_t^{-1}(x),\pi_t)[e_i]\right]\end{aligned}\ ] ] where @xmath134 is a basis of @xmath4 , is lipschitz continuous uniformly with respect to @xmath80 $ ] .
the reason of the above condition is to use it s formula on the process @xmath135 so that @xmath136 satisfies an sde with additive noise for which the results from @xcite can be applied .
although , it might seem that the class of such processes is small , it covers a wide variety of models which are relevant in applications , such as for instance geometric - type models .
[ malldiff ] let @xmath86\}$ ] be the unique global strong solution of ( [ sde ] ) .
then we have @xmath137 for every @xmath80 $ ] .
see @xcite .
[ skorokhodi ] let @xmath80 $ ] and define @xmath138 then @xmath139 . since @xmath140
, see ( * ? ? ?
* proposition 1.3.1 . )
, one needs to show that @xmath141 . in both cases it suffices to show that the integrands are malliavin differentiable . for the lebesgue integral it is immediate as for the it integral one may justify this by ( * ? ? ?
* lemma 1.3.4 . ) .
now , since @xmath142 , @xmath108 and @xmath143 for every @xmath80 $ ] we have together with proposition [ detyt ] that @xmath144 for every @xmath80 $ ] .
the result follows since the functions @xmath145 , @xmath105 and @xmath146 are continuously differentiable in the first variable with bounded derivative .
[ skorokhodii ] the process @xmath147 @xmath148 , is skorokhod integrable .
indeed , it is the product of an adapted process , hence skorokhod integrable and by proposition [ skorokhodi ] a skorokhod integrable random variable .
[ belthm ] let @xmath86\}$ ] be the unique global strong solution of ( [ sde ] ) .
let @xmath149 be a measurable function such that @xmath150 .
define the function @xmath151.\ ] ] then @xmath152^\ast\delta w_s\right]^\ast\end{aligned}\ ] ] where @xmath153 denotes transposition , @xmath106 is the fundamental matrix obtained in theorem [ mallsob ] and @xmath154 here @xmath155 \rightarrow { \mathbb r}$ ] is an integrable function such that @xmath156 and @xmath157 otherwise .
the functions @xmath145 , @xmath105 , @xmath146 , @xmath79 are defined as in theorem [ mallsob ] .
in fact , the skorokhod integral in can be expressed as an it integral plus a finite variation term .
indeed , since @xmath158 , integration by parts for the skorokhod integral , see theorem [ ibp ] or ( * ? ? ? * theorem 3.15 ) , yields @xmath159^\ast dw_s \ , u(t)\\ & \hspace{4cm}- \int_0^t a(s ) [ \sigma^{-1}(s , x_s^x,\pi_s^x)y_s d_s u(t)]^{\ast } ds\bigg)\bigg]^\ast . \end{split}\end{aligned}\ ] ] we will carry out the proof in four steps .
first , we will show the formula for smooth functions @xmath160 with compact support . then extend it to any continuous and bounded function @xmath160 by using a limit argument .
we then get rid of the continuity by employing a monotone class argument .
finally , we consider any general function with the property that @xmath161 . * step 1 : * assume first that @xmath160 is infinitely differentiable with compact support .
by theorem [ mallsob ] we have @xmath162 then multiplying both sides by the function @xmath163 and integrating over @xmath164 $ ] we have @xmath165 as a consequence , @xmath166 \\ = & \ , e\left [ \phi ' ( x_t^{x } ) \int_0^t a(s ) d_s x_t \sigma^{-1}(s , x_s^x,\pi_s^x)y_s u(t ) ds\right ] \\
= & \ , e\left [ \int_0^t a(s ) d_s \phi(x_t ) \sigma^{-1}(s , x_s^x,\pi_s^x)y_s u(t ) ds\right]\\ = & \ , e\left [ \phi(x_t ) \int_0^t a(s ) \left[\sigma^{-1}(s , x_s^x,\pi_s^x)y_s u(t)\right]^\ast \delta b_s\right]^\ast\end{aligned}\ ] ] where we have used relation ( [ mallintrel ] ) , the chain rule for the malliavin derivative ( backwards ) and the duality formula for the malliavin derivative which is justified by corollary [ skorokhodii ] . *
step 2 : * assume @xmath160 is bounded and continuous , in particular , @xmath150 .
we can approximate @xmath160 by a sequence of smooth functions @xmath167 with compact support such that @xmath168 a.e .
as @xmath169 .
define @xmath170^\ast\delta w_s\right]^\ast.\ ] ] to make reading clearer introduce the notation @xmath171^{1/2}$ ] and the matrix -valued process @xmath172 , @xmath173 .
then the objects @xmath174 $ ] and @xmath175 are well - defined since @xmath150 and using cauchy - schwarz inequality we have @xmath176 ds + \int_0^t \int_0^t d_s \xi_r d_r \xi_s dr ds\right]^{1/2}\ ] ] where we used it s isometry property for skorokhod integrals , see theorem [ duality ] or e.g. ( * ? ? ?
* theorem 6.17 . ) .
observe that the first term is bounded since @xmath163 and @xmath111 are uniformly bounded and @xmath50 has integrable trajectories .
the second term is bounded since @xmath177 is malliavin differentiable for every @xmath164 $ ] because @xmath178 is malliavin differentiable for every @xmath57 $ ] in virtue of proposition [ detyt ] in connection with proposition [ malldiff ] as for @xmath179 , due to corollary [ skorokhodii ] .
now , we approximate @xmath180 by @xmath181 $ ] .
it is clear that @xmath182 a.e .
and now we can use the bismut - elworthy - li formula on @xmath183 in order to estimate @xmath184 .
indeed , again by cauchy - schwarz inequality and it s isometry for the skorokhod integral we have @xmath185^{1/2 } e[|w(t)|^2]^{1/2}\end{aligned}\ ] ] where @xmath186 denotes the malliavin weight .
now since @xmath187 and @xmath160 are continuous and bounded we have for every compact subset @xmath188 @xmath189 hence , @xmath180 is continuously differentiable with @xmath190 .
* step 3 : * let us denote @xmath191 it is clear that @xmath192 is a multiplicative class , i.e. @xmath193 then @xmath194 .
further , let @xmath195 be the class of functions @xmath196 for which holds . from step 2
we have @xmath197 .
then @xmath195 is a monotone vector space on @xmath4 , see e.g. @xcite for definitions .
indeed , from dominated convergence we have monotonicity .
in fact , if @xmath198 such that @xmath199 with @xmath200 and @xmath160 is bounded then @xmath201 .
furthermore , denote by @xmath202 where @xmath203 denotes the borel @xmath1-algebra in @xmath2
. then we are able to apply the monotone class theorem , see e.g. ( * ? ? ?
* theorem 8) and conclude that @xmath195 contains all bounded and @xmath204-measurable functions @xmath205
. nevertheless , @xmath204 coincides with the borel @xmath1-algebra of @xmath4 since @xmath192 contains all continuous bounded functions .
so we conclude that @xmath195 contains all bounded borel measurable functions on @xmath4 .
* step 4 : * the last step is then to approximate any @xmath206-measurable function @xmath205 such that @xmath161 by a sequence @xmath167 of bounded @xmath206-measurable functions .
for example , @xmath207 then @xmath208 for each @xmath209 .
define @xmath210 $ ] .
then by cauchy - schwarz inequality and it s isometry we know @xmath211^{1/2},\ ] ] for any compactum @xmath188 and some finite constant @xmath212 . finally , observe that clearly one has @xmath213^{1/2 } \xrightarrow{n\to \infty } 0\ ] ] thus proving the result .
in this section we wish to give a rather simple but illustrative example of how the dependence on the expectation of the solution may give rise to more complicated terms when deriving the bismut - elworthy - li formula . in one of the examples we adopt the context of finance where the formula has a broad use for the computation of the so - called delta sensitivities which
, in short , is the sensitivity of prices of contracts with respect to the initial value of the price of the stock taken into consideration
. we will consider the price of an option written on a stock whose dynamics depend on the expectation of the price process .
then we provide two numerical examples in order to demonstrate that the bismut - elworthy - li formula , or the so - called malliavin method for computing the delta is numerically more efficient than the usual finite difference method even when the function @xmath160 is discontinuous .
let @xmath214\}$ ] represent the price dynamics of some asset with initial price @xmath215 governed by the following sde @xmath216 , \quad t\in [ 0,t ] , \quad s_0^x = x>0\end{aligned}\ ] ] where @xmath217 and @xmath218 .
let @xmath219 , @xmath80 $ ] , @xmath220 with @xmath221 be the risk - less asset and @xmath222 a pay - off function .
then the price of a european option at current time with maturity @xmath67 ( under the risk - neutral valuation approach ) is given by @xmath223\ ] ] where @xmath224 is the risk - neutral measure , i.e. @xmath225\ ] ] where @xmath226\ ] ] is the market price of risk process .
it follows that @xmath227 obtained as the solution of a riccati equation . also , we have @xmath228 and @xmath229
. then the @xmath230-sensitivity of an option @xmath160 written on @xmath231 is given by @xmath232.\ ] ] let us find a simpler expression for the stochastic integral . using the integration by parts formula for the skorokhod integral , see theorem [ ibp ] , we find that @xmath233 and hence , taking @xmath234 we find that under the risk - neutral measure @xmath224 , the @xmath230-sensitivity is given by @xmath235\ ] ] with malliavin weight @xmath236 finally , observe that if we ignore the dependence on @xmath237 $ ] , e.g. taking @xmath238 then we obtain @xmath239= e^{-rt } e_{p } \left[\phi(s_t^x ) \frac{\hat{w}(t)}{tx\sigma}\right],\ ] ] where @xmath240 is a standard brownian motion under @xmath28 and hence the @xmath230 coincides with the classical one .
consider now the following sde , @xmath241 , \quad t\in [ 0,t ] , \quad x_0^x = x>0,\end{aligned}\ ] ] for some suitable continuously differentiable function @xmath242 .
let us compute the derivative of @xmath243 $ ] for an irregular function @xmath160 .
this special `` geometric - type '' case shows that whenever @xmath50 is deterministic then the delta @xmath247 is an rescaled version of the classical delta and they coincide when @xmath248 is constant , indeed .
let , for instance , @xmath258 for some fixed @xmath259 , also known as a _
european call option _ in the context of finance .
then we use a monte carlo method to compute the above expression and compare it to the following finite difference method scheme hereunder @xmath260-e[\phi(x_t^{x})]}{h } , \quad h\approx 0.\ ] ] , @xmath261 , @xmath262 , @xmath263 and @xmath264 with @xmath265 to the right and @xmath266 to the left . on bottom parameters set to @xmath267 , @xmath261 , @xmath268 , @xmath269 and @xmath264 with @xmath265 to the right and @xmath266 to the left ] in the upper left figure , @xmath265 for the finite difference method and the two methods are seemingly giving similar accurate results , although the malliavin method is more efficient in number of iterations .
if one wishes to decrease @xmath270 in order to gain precision we can see how the finite difference method becomes unstable ( upper right and lower right figures ) . in conclusion
, the integration by parts formula seems to be a much more efficient tool for the computation of sensitivities for mean - field sdes , at least , in this setting . ,
@xmath261 , @xmath262 , @xmath263 and @xmath264 with @xmath265 to the right and @xmath266 to the left . on bottom parameters
set to @xmath267 , @xmath261 , @xmath268 , @xmath269 and @xmath264 with @xmath265 to the right and @xmath266 to the left ] the conclusions here are clear .
the regularity of the function @xmath160 plays an important role .
we see that the bias in the finite difference method seems high and it becomes unstable when decreasing the values of @xmath270 . on the contrary
, the bismut - elworthy - li formula gives a better approximation of the sensitivity , even when the function @xmath160 is discontinuous .
d. baos , t. nilssen , _ malliavin and flow regularity of sdes .
application to the study of densities and the stochastic transport equation _ , ( to appear in stochastics an international journal of probability and stochastic processes ) . | we generalise the so - called bismut - elworthy - li formula to a class of stochastic differential equations whose coefficients might depend on the law of the solution .
we give some examples of where this formula can be applied to in the context of finance and the computation of greeks and provide with a simple but rather illustrative simulation experiment showing that the use of the bismut - elworthy - li formula , also known as malliavin method , is more efficient compared to the finite difference method . |
there is little variety in temperature and density among the quantum - degenerate neutral gases produced currently in over one hundred laboratories in seventeen countries .
this commonality may be surprising because , unlike solids , gases have no lower bound in density .
however to within a factor of ten , the density of an ultracold gas is @xmath0@xmath1 , about six orders of magnitude lower than the density of an ideal gas at standard temperature and pressure .
why this particular density ? for @xmath2@xmath1 , loss processes , such as a three - particle collision leading to the formation of a deeply bound molecule , become faster than rethermalization from elastic collisions . for @xmath3@xmath1 , the characteristic energy and temperature scales of the quantum gas
become impractically small , particularly if one is interested in anything beyond the non - interacting ideal gas , and the thermalization rate of the gas becomes slow compared to parasitic heating rates and the vacuum - limited lifetimes of the frigid gas samples .
this common density sets a common characteristic energy and length scale for experiments on cold gases . from the inter - particle spacing @xmath4 , typically 300 nm , one would guess the energy scale for the physics of such gases to be @xmath5 , where @xmath6 is planck s constant and @xmath7 is the atomic mass ; more formally , one often takes the fermi energy @xmath8 , where @xmath9 is the fermi wave vector . for @xmath10rb ,
this energy scale is around @xmath11ev , corresponding to a temperature of about 500 nk , and a frequency of about 10 khz.rb as a typical example .
rubidium was also the first gaseous element to be bose condensed , and is still the most commonly used species for ultracold boson experiments .
] this energy scale also ( roughly ) defines the temperature at which a gas becomes quantum degenerate .
the onset of quantum degeneracy can also be considered from the comparison of length scales . from the system temperature
, we can define the thermal debroglie wavelength , @xmath12 where @xmath13 is the boltzmann constant .
when the de broglie wavelength is comparable to the interparticle spacing , the coherent matter waves associated with the various particles in the gas are forced to overlap , meaning ( pictorially ) that the number of independent quantum states in the gas becomes comparable to the number of gas particles . at this point ,
the quantum statistics of particles come into play in describing the nature of the gas .
the ultra - low temperature of quantum gases is therefore simply a consequence of their necessarily low density .
since the gases in question are roughly a billion times less dense than liquid helium , they are degenerate at a temperature a million times lower than the lambda point of helium .
the low temperature scale required for the study of quantum degenerate gases has been , and continues to be , the prime technical challenge in ultracold - atom research . as discussed in sec .
[ sec : cycle ] , cooling gases from room temperature to quantum degeneracy relies on a hybrid of cooling methods .
these methods took decades to develop . even today , with cutting - edge techniques , gases can not match the extreme quantum degeneracy of electrons within a solid in a dilution refrigerator , where the ratio @xmath14 is around @xmath15 ; in contrast , for cold atom experiments to date , this ratio goes no lower than @xmath16 .
thus , advancing the frontier cold atom experiments still requires the continued development of cooling techniques .
pursuing this frontier is tremendously appealing , as one expects that neutral gases would be able to explore the rich physics of spin liquids , topological quantum matter , pure bcs superfluidity , and perhaps d - wave pairing in lattices @xcite .
however , even within the current technical limits , the combination of low density and ultracold temperature creates an extraordinary opportunity to study many - body systems with cold atoms .
atomic gases are subject almost exclusively to pair - wise interactions that can be characterized completely by considering the physics of an isolated two - atom molecule .
now , a molecule of , say , two @xmath10rb atoms is not the simplest quantum object , and its properties can not be determined _
ab initio_. however , after taking into account a range of data from various forms of molecular spectroscopy , it is possible to set up a reliable model that describes the scattering properties of two isolated atoms extremely well , and , thereby , provides the only inputs needed to specify an accurate description of a many - atom quantum gas comprising @xmath10rb atoms
. moreover , at the very low kinetic energies characteristic of atomic gases , the partial - wave treatment of binary collisions is of great utility in reducing the complete characterization of atomic interactions simply to the specifications of the s - wave scattering length , @xmath17 , and , on rare occasions , to the range of the potential , @xmath18 . as explained in sec .
[ sec : interactions ] , the scattering length can be controlled by tuning the structure of the two - atom molecular potential , e.g. using magnetic - field - tunable feshbach resonances . for such reasons ,
ultracold - atom `` materials '' can be regarded as faithful experimental renditions of a known many - body hamiltonian .
experiments on quantum gases have therefore been dubbed as _ quantum simulations _ of various models ( i.e. hamiltonians ) of many - body quantum physics .
although cold - atom systems have non - universal eccentricities just like any other materials , it is comforting to know that even these eccentricities can , in principle , be characterized _
in this section we hope to provide the reader with `` ( almost ) everything you wanted to know about a neutral atom but were afraid to ask . ''
mostly , we will concern ourselves with the magnetic and electric susceptibility of a single ground state atom , the latter due to coupling of the ground state to excited states by electric fields . in sec .
[ sec : interactions ] we will extend this discussion to include interactions between two ground - state atoms . at the time of writing this chapter , eleven elements have been bose - condensed : hydrogen ( h ) , five alkali metals ( li , na , two isotopes of k , rb , cs ) , three earth - alkaline and similar metals ( ca , three isotopes of sr , four isotopes of yb ) , two transition metals ( cr , dy ) , and one metastable noble gas ( he * ) . five of these elements also have naturally occuring fermionic isotopes that have been cooled to quantum degeneracy : @xmath19he * , @xmath20li , @xmath21k , @xmath10sr , and @xmath22yb .
alkalis were the first to be bose - condensed , and remain popular due to the simplicity of their electronic structure : a single unpaired electron is optically active , just as in hydrogen , while the remaining electronics form a relatively inert core . in the absence of external fields ,
there are three terms in the hamiltonian of hydrogen - like atoms , @xmath23 where @xmath24 contains the non - relativistic kinetic energy of the electrons and the coulomb interaction between them ; @xmath25 is the `` fine structure '' term that includes relativistic corrections to @xmath24 , electron spin , and spin - orbit terms ; and @xmath26 is the `` hyperfine structure '' term induced by the spin and electric quadrupole moment of the nucleus . in the russell - saunders ( @xmath27 ) coupling scheme , which is relevant when the electronic valence shells are sparsely populated as is the case for hydrogen - like atoms , the eigenstates of @xmath28 share a common set of quantum numbers : @xmath29 ( principal quantum number ) , @xmath30 ( total orbital angular momentum ) , and @xmath31 ( total overall angular momentum ) .
for instance , the ground state of rb is @xmath32 , meaning the optically active electron is in the @xmath33 orbital , with @xmath34 , and an angular momentum of @xmath35 .
the `` 2 '' superscript gives the degeneracy for the total electronic spin of @xmath36 . for alkali atoms ,
the two lowest - lying excited states are @xmath37 and @xmath38 .
different from the situation in hydrogen , the energies of these levels differ from that of the @xmath29s state because the higher angular momentum of the p states pulls the electron orbit to a higher radius , sampling less of the charge density of the core electrons .
such a many - electron mechanism ( known as the `` quantum defect '' ) is weak for excited states with higher principal quantum numbers . for the alkalis ,
this @xmath29s-@xmath29p transition is the lowest - energy and strongest optical transition .
the resonant frequency occurs at several hundred thz ( roughly 1.5 ev of energy ) , and is therefore accessed by light in the near - infrared to visible range , generated easily from solid - state or dye laser systems . for atoms with nuclear spin ,
@xmath26 further splits the fine - structure eigenstates . the total angular momentum , which is a sum of the electron orbital and spin angular momentum and of the nuclear spin , is called @xmath39 . for example , for @xmath10rb
the nuclear spin is @xmath40 , which , added to the ground - state electronic angular momentum of @xmath36 ( all of it from spin ) , gives either @xmath41 or @xmath42 .
these states are split by 6.8 ghz in @xmath10rb , and similar amounts in other hydrogen - like atoms , small enough that the upper state does not decay spontaneously . for illustration ,
the atomic level structure of @xmath10rb is shown in fig .
[ fig : rbspectrum ] . for this alkali atom ,
strong transitions are found at optical wavelengths of 780 and 795 nm ( per fraunhofer , these are the d2 and d1 lines , respectively ) .
each of these lines is split into well - resolved transitions between the various ground- and excited - state hyperfine levels .
. a naturally occurring gas of @xmath10rb would be in a mixture of its 8 electronic ground states , in the 5s manifold . at low magnetic field ,
these states are well characterized by the total atomic angular momentum , with quantum number @xmath39 , and the magnetic quantum number @xmath43 .
ground - state hyperfine interactions split the @xmath41 and @xmath42 levels by a frequency of @xmath446.8 ghz .
laser cooling , trapping , and imaging typically involve optical transitions to the 5@xmath45 states .
these are split into two disparate lines , labeled d1 and d2 , by the fine structure interaction , and split further into states of distinct total angular momentum @xmath46 by the hyperfine interaction . in @xmath10rb ,
these excited state hyperfine levels are split by about 100 mhz between neighboring lines , greater than the natural linewidth ( 6 mhz ) of the optical transitions . for some other alkali elements ,
the excited state hyperfine structure is not resolved .
especially important are the two `` cycling '' transitions , shown with dashed arrows , on the d2 line , for which spontaneous emission from the excited state can not return the atom to an electronic ground state other than the one being driven .
such cycling transitions are favored for laser cooling and optical probing since thousands of photons can be scattered by the atom without its departing from the set of states being interrogated.,width=377 ] rb . * left column : * at low field , states are well characterized by their total angular momentum @xmath39 , and their energies deviate linearly from their zero - field hyperfine splitting by @xmath47 .
states that can be trapped at low field are indicated with a thick black trace . *
right column : * at high field , the energies are again linear in field , varying as @xmath48 . * central column : * between these two linear regimes is a quadratic regime in which neither @xmath39 nor @xmath49 are good quantum numbers.,width=510 ] applying a magnetic field to the atom adds another term to the hamiltonian , @xmath50 , where @xmath51 is the magnetic dipole operator and @xmath52 is the external field .
this term breaks the rotational symmetry of the hamiltonian of an isolated atom . solving for the eigenvalues of the electronic ground and excited states in an applied magnetic field
is a standard exercise in an atomic physics course .
for the case of a @xmath35 electronic state , such as the ground - state of alkali atoms , the energy eigenvalues are expressed conveniently by the breit - rabi formula , @xmath53 where @xmath54 , @xmath55 , @xmath56 , @xmath57 is the hyperfine constant , and @xmath58 . fig . [
fig : rbbr ] shows the the breit - rabi level diagram for the ground state of @xmath10rb .
we point out several of its features . for weak magnetic fields ,
the magnetic dipole term @xmath59 can be treated as a perturbation for which the total atomic angular momentum @xmath39 is still conserved .
the zeeman energy shifts are then linear : @xmath60 , where @xmath43 is the magnetic quantum number @xmath43 ( projection of @xmath39 along the field axis ) , @xmath61 is the bohr magneton , and the factor @xmath62 is given by land s formula . as the field
is increased , the field - induced coupling between hyperfine spin states first becomes evident in a quadratic zeeman shift , proportional to @xmath63 .
for example , for the @xmath64 levels @xmath10rb , this shift is 70 hz per square gauss .
the quadratic zeeman shift plays an important role in the physics of spinor bose gases , as discussed in the chapter by fetter and foot . at very high fields , the magnetic dipole term dominates the energy eigenstructure . with the hyperfine interaction
now treated as a perturbation , the atomic levels become increasingly well described by the quantum numbers @xmath49 and @xmath65 , these being the individual spin projections of the electron and nuclear angular momenta . since the magnetic moment of the electron is much larger than that of the nucleus , the energy states break up into a higher - energy set of @xmath66 eigenstates and a lower - energy set of @xmath67 eigenstates , as shown in fig .
[ fig : rbbr ] . in this regime , @xmath68 . in
each @xmath49 set , level spacing is nearly constant at @xmath69 , with a weak dependence on magnetic field .
this slow variation becomes useful in performing rf spectroscopy on resonantly interacting fermi gases of @xmath20li , for instance , to observe interaction shifts from the single - particle transition energy @xmath70 .
the variation of atomic energies in an applied magnetic field finds important application in the magnetic trapping of ultracold gases . to achieve such trapping ,
atoms are placed in an inhomogeneous magnetic field .
assuming an itinerant atom follows the magnetic field orientation and magnitude adiabatically , the spatial variation of the magnetic field magnitude leads directly to a spatially varying magnetic potential energy .
traps are formed typically with fields in the low - field regime discussed above , and , since magneto - static field maxima in conductor - free regimes are not allowed by maxwell s equations , only `` weak - field - seeking '' atomic states , i.e. those with @xmath71 , can be trapped magnetically . for example , for @xmath10rb , only the @xmath72 , @xmath73 and @xmath74 states are trapped at the limit of zero magnetic field state of @xmath10rb @xcite . ]
( see fig .
[ fig : rbbr ] ) .
the simplest trap is a magnetic quadrupole field , whose minimum field is zero .
however , spin - flip losses occur when atom pass too close to the zero - field point .
long - lived magnetically trapped samples use a ioffe - pritchard configuration @xcite , in which a two - dimensional magnetic quadrupole is combined with a magnetic dipole along the third axis .
to lowest order , the combined field is @xmath75 where @xmath76 is the holding field , @xmath77 is the quadrupole gradient , and @xmath78 is the dipole field curvature . for @xmath79 ,
the potential is harmonic in all directions , with frequencies @xmath80 and @xmath81 .
the depth of the trap is limited by off - axis saddle points .
the maximum magnetic dipole moment of an atom is about one bohr magneton per unpaired electron .
thus , in a trap where the magnetic field strength varies from zero at the trap bottom to , say , 0.1 t at the rim of the trap potential , a one-@xmath61 atom experiences a trap depth of about 100mk , a temperature that is attainable by a precursor stage of laser or perhaps buffer - gas cooling .
transition - metal atoms that have been recently laser - cooled can have rather large magnetic moments , e.g. 6 @xmath61 for cr or 10 @xmath61 for dy , coming from the large number of unpaired electrons in the partially filled shells of these atoms .
however , the large dipole moment of these atoms also makes the weak - field - seeking states more unstable to dipolar relaxation collisions wherein the magnetic potential energy is converted to kinetic energy due to transitions among zeeman states . thus ,
magnetic trapping of these high - spin atoms has been less useful than for hydrogen - like atoms
. the ground states of alkaline earth atoms , with two paired electrons , lack an electronic magnetic moment , and thus can not be magnetically trapped . even though neutral atoms with unpaired electrons have a _
dipole moment , eigenstates of @xmath82 do not have an _ electric _ dipole moment .
however , an applied electric field can mix eigenstates of opposite parity to induce a dipole moment . barring degeneracies between states of opposite parity ( such as occurs for excited states of hydrogen ) , the resultant stark shift is quadratic in field strength . for heteronuclear ( `` polar '' ) molecules ,
the existence of nearly degenerate states of opposite parity means that the stark shift becomes linear beyond some relatively weak polarizing field . restricting our discussion to the quadratic stark shift typical of ground - state atoms , static electric fields create an attractive potential ( the dc stark shift ) with a strength @xmath83 , where @xmath84 is the dc polarizability and @xmath85 is the electric field strength .
alkali atoms have @xmath86cm@xmath87/v .
since fields greater than @xmath88v / m typically cause electrode discharge , static potentials can not be larger than a few @xmath89k , in temperature units .
this potential depth is roughly @xmath88 times shallower than the typical magnetic trap depths discussed in sec .
[ sec : btraps ] .
another problem with electrostatic traps is that field maxima are ( again ) not allowed by maxwell s equations , so an electrostatic potential must be combined with another type of potential to form a stable trap for atoms .
fortunately , the stark potential can be greatly enhanced near optical resonances .
considering just a `` two - level atom , '' i.e. where we can consider only the coupling between one ground state @xmath90 and one excited state @xmath91 of the atom , the expectation value of the dipole operator @xmath92 is @xmath93 where @xmath94 is the rabi frequency . on resonance ,
population oscillates between ground and excited states at the frequency @xmath94 .
far from resonance , when the detuning @xmath95 between the driving frequency @xmath96 and the atomic transition frequency @xmath97 is much larger than the transition line width @xmath98 , the ground state feels a second - order energy shift ( sometimes called the `` ac stark shift '' ) , with the value @xmath99 as in the static case , this energy scales like @xmath100 , but now we identify this as the intensity of an electromagnetic wave .
when @xmath101 , i.e. as we approach the limit of the dc stark shift , the above formula must be corrected by the addition of counter - rotating terms @xcite . comparing the dc and ac stark shifts , we see that the resonant enhancement of the stark shift is approximately @xmath102 . in principle , since optical resonances can have a quality factor @xmath103 of @xmath104 , a tremendous enhancement is possible . in practice ,
optical traps are tuned many line widths from resonance to avoid heating from light scattering , so @xmath102 is typically tens to hundreds .
we see that the sign of the ac stark shift varies with that of the detuning of the drive field .
referring to the case of @xmath10rb and considering only the levels shown in fig .
[ fig : rbspectrum ] , light at a wavelength below 780 nm is blue - detuned with respect to both the principal transitions .
thus , a ground - state @xmath10rb atom exposed to such light experiences a repulsive ac stark potential , and will be attracted to low - intensity spots of the light field . in contrast , light at a wavelength above 795 nm is red - detuned with respect to all transitions from the ground - state , so that @xmath10rb atoms experience an attractive ac stark potential and are pulled toward the high - intensity regions .
light with a wavelength between those of the d1 and d2 transitions can yield either attractive or repulsive potentials depending on the relative strengths of coupling to the two transitions .
these relative strengths depend on the atomic hyperfine and zeeman state and the polarization of the light field .
such a dependence allows the ac stark shift to be used to create state - dependent potentials or to exert fictitious magnetic fields .
the ac stark potential allows one to trap quantum gases optically @xcite .
the simplest optical trap is formed by a focused gaussian beam of red - detuned light .
a 10w beam focused to a minimum beam waist of 20@xmath89 m creates an electric field strength of @xmath105v / m .
such a beam at a wavelength of 1064 nm induces a trap depth of 2.4mk for rubidium . in general
, we see that optical potentials can be deeper than electrostatic potentials , but not as deep as typical magnetic traps .
optical traps have the tremendous benefit of allowing one to trap _ any _ magnetic sub - level of the ground state manifold . for @xmath10rb , for instance , there are eight states among the @xmath41 and @xmath42 hyperfine levels .
as discussed above , only three of these can be magnetically trapped with a linear zeeman effect , whereas an optical trap can hold all eight .
a further implication is that the magnetic field can be varied arbitrarily if the atoms are trapped optically .
this will have important implications in sec .
[ sec : feshbach ] . in a solid
, electron band structure arises from the periodic potential created by ions in a crystalline arrangement .
a similar potential landscape can be created for cold atoms by interfering multiple laser beams .
two intersecting plane - wave beams of the same optical frequency create an intensity pattern that varies sinusoidally in space .
following eq .
( [ eq : shift ] ) , this intensity pattern produces a periodic potential of the form @xmath106 where @xmath107 is the magnitude of the potential , and where @xmath108 , the difference between the wavevectors of the intersecting beams , defines the @xmath109 axis .
a single pair of beams creates a stack of pancake - like potentials , with tight confinement along one axis due to the spatial interference term and loose confinement along the other axes defined by the radial profile of the intersecting laser beams .
when the intersecting laser beams are weak , the atoms may tunnel between neighboring pancakes over experimentally reasonable time scales .
when the laser beams are sufficiently intense , such tunneling is negligible , and , at sufficiently low temperature , the atomic motion may be reduced to just two dimensions . starting from this basic configuration , a wide range of trapping potentials can be generated .
adding more laser beams , one can create periodic confinement also in the @xmath110 and @xmath111 directions .
the simplest three - dimensional lattice potential is one in which three pairs of beams are used along orthogonal axes , with each pair at a different optical frequency so that cross - interference between pairs can be ignored .
the resulting potential is @xmath112 the intensities @xmath113 in this cubic lattice can be chosen arbitrarily , and even dynamically , to produce atomic gases that live effectively in three- , two- , one- , or zero - dimensional spaces .
even the wave vectors can be varied dynamically , in an `` accordion '' fashion @xcite or to create a rotating lattice @xcite .
although square nets and cubic lattices have dominated cold atom experiments to date , all of the bravais lattices can be created @xcite .
so far we have tacitly assumed that lattice potentials were created with parallel linear polarizations of light .
however the polarization of each laser beam can be chosen arbitrarily .
the intensity at @xmath114 of two interfering beams with wave vectors @xmath115 and complex polarizations @xmath116 is @xmath117 where @xmath118 is the intensity of each individual beam .
however it is not the intensity alone , but also the optical polarization that determines the potential , as follows from the definition of the rabi frequency in eq .
( [ eq : omegar ] ) . treating the ac stark effects more completely ,
accounting for both diagonal and off - diagonal second - order coupling between the different spin states of the electronic ground state , leads to a diverse set of optical lattice potentials , such as spin - dependent lattices , where the external potential depends on the internal state of the atom @xcite , or lattices with significant spin - orbit coupling , in which bands can have non - zero chern number @xcite .
dynamic control of the relative polarization can be used for spin - dependent transport and , thereby , to generate widespread entanglement between the spin and motion states of lattice - trapped atoms @xcite .
considering also the possibility of optical superlattices ( formed by overlaying lattices of different periodicity ) , lattices for atoms with several stable electronic states ( notably the alkali - earth atoms ) , lattices that combine both optical and radio - frequency manipulation of the trapped atoms , and so forth , one might expect that atomic physicists will be as busy studying lattice trapped atoms as condensed - matter physicists have been studying crystal bound electrons !
however , as one considers designing a new family of optical lattice potentials , one should keep in mind the limitation that optical potentials are not entirely conservative . the electric dipole moment
induced dynamically in an atom by a near - resonant laser field does not respond exactly in phase with the laser drive .
the out of phase ( imaginary ) electric susceptibility of the atom induces the atom to emit light spontaneously .
the recoil from these randomly emitted photons and also the random transitions between electronic states of differing polarizability ( dipole - force fluctuations ) heat the atom mechanically @xcite .
many of the machinations that yield the most intriguing optical lattice potentials often require the use of laser light that is fairly close to atomic resonances , so that the laser detuning from various transitions be significantly different . perhaps realizing the more exotic optical lattice configurations will have to wait for more exotic atoms and molecules to be introduced to the ultracold regime .
as discussed above , the broad relevance of research on quantum gases rests upon the idea that the many - body quantum systems behave similarly regardless of the details of their construction . yet , in considering analogies between electrons in solid - state materials and atoms in artificial potentials , one can not escape the fact that electrons , being charged , interact via the long - range coulomb interaction , while neutral atoms do not .
rather , in most neutral atomic gases , interactions are short - ranged as defined by the angstrom - scale effective range of the two - atom molecular potential .
the absence of long - range interactions among atoms does not necessarily invalidate the analogy between atomic and electronic materials , since in many materials , the coulomb interactions among mobile electrons are screened at long range , leaving only an effective short - range interaction between electron - like ( or hole - like ) quasiparticles .
however in other contexts , the absence of electric charge and long - range interactions lead to a major departure of ultracold atom systems from typical solids , for instance _ vis a vis _ the connection between plasma oscillations and bogoliubov modes in charged superconductors or the realization of high - order and long - range spin - spin interactions in magnetic materials .
the presence of dipolar interactions in some neutral gases is interesting exception to the rule , and is discussed at the end of this section .
k , shows little temperature dependence .
however the p - wave cross - section ( closed symbols ) , measured using spin - polarized atoms , exhibits a threshold behavior , falling two orders of magnitude between 100@xmath89k and 5@xmath89k .
reproduced from ref .
@xcite with permission , courtesy of the aps .
, width=264 ] at the low density of atomic gases , interactions can be treated as a two - atom quantum scattering problem governed by the relevant molecular potential .
a complete treatment of such scattering is quite complicated , owing to the many potential - energy surfaces that arise as two compound quantum objects atoms with internal states describing electronic motion , electron spin , and nuclear spin approach one another .
however , much of this complexity is swept under the rug due to the near - zero incident energy of a colliding atom pair .
one appeals to the partial wave treatment of scattering , in which the incident matter waves engaged in a collision are decomposed into eigenstates of angular momentum . in this basis , a centrifugal barrier exists for any non - zero angular momentum , for example p - wave or d - wave collisions .
however as shown in fig .
[ fig : freezep ] , even for p - wave collisions , the magnitude of the barrier is @xmath1190.1mk , much higher than the sub-@xmath89k temperatures of quantum degenerate gases .
therefore only isotropic s - wave scattering is significant for ultracold atoms .
far from the innards of the molecular potential , such s - wave scattering results only in a phase difference , @xmath120 , between the incoming and outgoing debroglie waves , which is given as @xcite @xmath121 where the wavevector @xmath122 relates to the relative momentum of the colliding pair . in the ultracold regime , all the gory details of the interaction potential may be neglected ( though sometimes it pays to keep them in mind ) by retaining just two quantities that characterize the scattering phase : the s - wave scattering length @xmath17 , and the effective range of the potential @xmath123 .
the effective range term itself is negligible for collision energies much less than @xmath124 , which is roughly 100@xmath89k for @xmath10rb . for @xmath10rb in the polarized @xmath125 state ,
the triplet scattering length @xmath126 nm @xcite .
the scattering length also determines the elastic collision cross section , given for distinguishable particles as @xmath127 for identical bosons , there is an additional factor of two : @xmath128 , whereas for identical fermions , s - wave collisions are forbidden by the pauli principle , and @xmath129 .
figure [ fig : freezep ] shows direct measurement of the latter effect in a spin - polarized gas of @xmath21k .
since the details of the interaction potential are unimportant , it is convenient to replace it formally with a simpler potential the zero - range contact potential @xmath130 with @xmath131 being the relative spatial coordinate which yields the same s - wave scattering length by identifying @xmath132 .
subtle divergence problems that arise sometimes from this pseudo - potential approach are avoided by using regularized potential , @xmath133 @xcite . from dimensional arguments , and in mean - field theory
, it follows that the product @xmath134 , with @xmath29 being the number density , quantifies the per - atom interaction energy .
this energy is seen to be proportional to the scattering length , i.e. we associate @xmath135 with repulsive interactions and @xmath136 with attractive interactions .
the possibility of realizing both attractive and repulsive interactions may seem counter - intuitive , given that the interaction energy between polarizable ground - state atoms is always _ negative _ at long - range ( at the typical distance between atoms ) due to van der waals interactions .
more properly , the per - atom interaction energy should be thought of as the influence of scattering on the _ kinetic energy _ of the interacting gas . to visualize this , consider two atoms trapped in a box with linear dimensions @xmath137 , which we assume to be large .
their kinetic energy goes as @xmath138 . due to scattering
, the two - body wavefunction behaves at long range as if the linear dimension of this box now has a length @xmath139 . for @xmath135 , the volume inhabited by a collection of atoms
is effectively reduced due to the asymptotic effects of scattering , causing their kinetic energy to rise . for @xmath136 , adding more atoms seemingly _ increases _ the volume available to the gas , causing its kinetic energy to diminish .
li across seven decades , from extremely small attractive interactions to extremely strong repulsive interactions . in - situ
cloud size is used to measure the scattering length ( points ) .
the mean and standard error of approximately ten shots taken at each field is shown .
results of a coupled - channels calculation are shown by the solid line .
the feshbach resonance fit is indicated by the dashed line .
the inset shows the extracted values of @xmath17 near the zero crossing .
values of the scattering length above @xmath140 bohr radii are not accurate due to beyond mean - field effects . reproduced from ref .
@xcite with permission , courtesy of the aps .
, width=302 ] the treatment of three - dimensional scattering discussed above is modified in fascinating ways by the confinement of atoms to lower - dimensional spaces , through which interaction strengths are rescaled and various confinement - induced collisional resonances and bound states can occur .
these effects are discussed in the chapter by hadzibabic and khl . in the presence of a magnetic field , free atoms and bound molecules
experience a differential zeeman shift since their magnetic dipole moments are not equal . at certain serendipitous values of the field ,
this differential energy shift can bring a bound dimer state into resonance with the energy of two asymptotically free atoms .
this condition is called a feshbach resonance @xcite .
even though the atoms do not actually form a dimer , the phase shift of their s - wave collision is modified by the proximity of the resonance . as shown in fig .
[ fig : tuning ] , by using a magnetic field , the ultracold gas can be tuned from weakly to strongly interacting , and chosen to be either repulsive or attractive .
such an exquisite control of interactions is possible with cold atoms because the interactions are binary .
liquid helium is roughly a billion times more dense , meaning that particles are a thousand times closer to one another . at the microscopic level , it s an `` atomic traffic jam '' the mean separation is comparable to the range of the interatomic potential , and no clean separation of energy scales is possible . for neutral gases , the feshbach resonances can be milli - gauss
wide , showing that the control over interactions through delicate tuning of the two - body molecular potential is not compromised by many - body effects .
experiments with feshbach resonances use optical traps , for two reasons
. first , feshbach resonances typically occur at fields of hundreds of gauss ( @xmath141mt ) , whereas the magnetic - field minima of magnetic traps are often at the single - gauss level ( @xmath142mt ) .
second , the feshbach resonances are accessed at the lowest magnetic fields in anti - trapped zeeman states . as discussed in sec .
[ sec : btraps ] , @xmath143 atoms can not be trapped magnetically .
an alternative to the magnetic feshbach resonance is the coupling to molecular states with electric fields , either static @xcite or optical @xcite .
these techniques are especially important for atoms with paired electrons , such as rare earth alkalis , since the hyperfine interaction couples free states to bound states in magnetic feshbach resonances .
optically induced feshbach resonances are being explored as a way to produce novel systems with rapid temporal and spatial variations in the interaction strength . to date , however , these optical feshbach resonances appear to be accompanied by the significant disturbance of the gas by spontaneous light scattering .
sweeping the magnetic field across a feshbach resonance can cause the adiabatic binding of free atoms into the loosely bound molecular states which induce the resonance .
these `` feshbach molecules '' are quite unusual in that they are bound by mere khz - level binding energies ( @xmath144 wavenumbers ! ) and have sizes in the nanometer range @xcite .
the generic few - body physics of feshbach - resonating atoms has been explored in a number of landmark experiments with cold atoms .
feshbach molecules are also centrally relevant to studies of the bcs - bec crossover in degenerate fermi gases , across which bose - condensed feshbach molecules transform smoothly into cooper pairs in a bcs superfluid .
the formation of feshbach molecules is also an important stepping stone for the creation of ground - state molecules by photo - association .
the magnetic dipole moment we discussed above for magnetic trapping can also lead to inter - atomic interactions .
atoms in excited states ( especially in highly excited rydberg states ) can have strong electric dipole moments , but they are typically not stable when colliding .
molecules offer the possibility of strong electric dipole moments for both trapping and interactions .
uniformly oriented dipoles interact according to the potential @xmath145 where @xmath146 is the angle between the dipole orientation and @xmath147 the separation between particles .
a potential scaling as @xmath148 at large distance is defined as short - ranged and can be treated using the pseudo - potential approach ( sec . [
sec : as ] ) so long as @xmath149 , where @xmath150 is the number of dimensions .
for instance , van der waals interactions arising from fluctuation - induced dipole moments fall off like @xmath151 ( or faster ) .
thus , such interactions can be summarized by the scattering length @xmath17 and otherwise ignored .
however the @xmath152 interaction between permanent dipole moments is , by this definition , long - ranged in three dimensions , and must be treated by a different approach @xcite . for neutral atoms , `` dipolar interactions ''
is nearly synonymous with `` long range interactions . ''
one can compare the strength of the magnetic dipole interaction to that of the s - wave contact interaction by calculating the magnetic self - energy of a trapped gas to the contact interaction strength @xmath153 ( sec .
[ sec : as ] ) , yielding the ratio @xmath154 .
for the alkali gases , with a typical s - wave scattering length of tens of bohr radii and a magnetic moment of one @xmath61 , @xmath155 is typically below @xmath16 , so that magnetic dipole interactions are small compared to the overall contact interaction energy of a single - component bose or fermi gas . however , dipolar interactions may still be significant in determining spin ordering in multi - component alkali gases .
most notably , for @xmath10rb , the contact interaction energy varies only slightly for different magnetic states , as determined by a small difference in s - wave scattering lengths for collisions among different pairs of atomic spin states . in face of this small variation , the magnetic dipolar energy may indeed be significant @xcite .
more dramatic effects of the magnetic dipole interaction are seen in atomic gases with larger magnetic moments , and in which the s - wave scattering length can be tuned near zero .
seminal experiments on such `` quantum ferrofluids '' were performed using cr gases held in anisotropic optical traps .
fixing the orientation of the atomic magnetic moment ( of magnitude @xmath156 ) to that of the guiding magnetic field , and varying the orientation of that field with respect to the axes of the optical trap , the interactions were converted from being effectively repulsive ( for atoms spread out preferentially in directions transverse to the dipolar axis ) , allowing large condensates to remain stable , to attractive , whereupon the condensates collapsed @xcite .
even stronger dipolar effects are expected in other atomic and molecular gases .
neutral atoms in rydberg states have strong electric polarizabilities due to large - radius electron orbitals , in addition to long - range van der waals interactions .
however the collisional instability of rydberg atoms prevents a simple comparison to ground - state dipoles .
heteronuclear molecules polarize in a relatively weak electric field to have an electric dipole moment on the order of 1 debye .
when normalized with respect to a typical @xmath17 , a dimensionless strength @xmath157 is expected .
quantum degenerate gases of polar molecules have yet to be produced , but several teams are close @xcite and expected to succeed in the near future .
the constellation of measurement tools available for studies of materials in the solid state reflects the robustness of such materials .
the copious scattering of high - energy photons , electrons and neutrons , injections of decaying muons , and the scratching of solid probes across the surface can all be tolerated by solid materials without necessarily destroying them .
in contrast , cold - atomic materials are delicate .
the momentum transferred to an atom upon scattering just a single near - infrared photon will typically exceed the typical thermal , fermi , and interaction energy in a quantum vapor .
material probes poking into a nano kelvin temperature gas will quickly boil the atoms away , ruling out most scanning probe techniques .
the delicacy of atomic many - body systems implies that the act of extracting data from such systems is extremely influential on their evolution .
efficiency is essential , and accounting for the back action of measurements is necessary for their interpretation .
the lion s share of measurements on cold gases are performed optically .
the lightly interacting atoms in a quantum gas are strong , narrow - band optical emitters , as discussed in sec .
[ sec : structure ] .
however ion - based detection can be used for fast detection of single atoms ( see sec . [
sec : singleatom ] ) , especially for metastable gases that ionize upon impact with an electrode surface . in conventional condensed - matter physics research ,
one makes or identifies a material sample , prepares it for measurement , inserts it into a measurement apparatus , and takes measurements for as long one needs to gather reliable data on the phenomenon being investigated .
the same sample might be used for months of data taking or for several sequential types of measurement , and might be stored in a drawer for further investigations down the line .
ultracold atomic materials are quite different
. the material must be constructed prior to each measurement , starting from a new , hot atomic vapor .
the typical experimental sequence is ( 1 ) using laser cooling and trapping to gather atoms from the vapor into a magneto - optical trap , with temperatures on the order of 100 @xmath89k , ( 2 ) trapping the atoms in a conservative potential , e.g. a magnetic ( sec .
[ sec : btraps ] ) or optical - dipole ( sec .
[ sec : etraps ] ) trap , ( 3 ) evaporatively cooling the atoms by gradually lowering the depth of the conservative trap and letting atypically high energy atoms escape the trap , reaching temperatures in the quantum - degenerate regime ( usually sub-@xmath89k ) and ( 4 ) putting the final touches on the material by turning on the system hamiltonian ( interaction strength , lattice type , spin admixture , etc . )
that one wants to examine .
once finally produced , the lifetime of the ultracold material is short , and so the entire experimental sequence for probing the material is quick , typically much shorter than the time it took to prepare the material in the first place . at the end of each measurement ,
the sample is discarded .
this cycle is repeated at the cycle time of a few seconds to a few minutes , depending on the speed of the accumulation and cooling stages . in examining a graph of data from a cold atom experiment ( for example , figs .
[ fig : freezep ] or [ fig : tuning ] ) , one should value the fact that each point on the graph represents one or several repetitions of a make - probe - discard experimental run . in light of this protocol , it is somewhat stunning to hear of cold - atom experiments that require hundreds or thousands of `` shots , '' each reproducing a gas under almost identical conditions , to obtain the high precision required to reveal new phenomena or test recent theories
. cold atoms are conventionally probed by optical imaging .
probe light at a well defined optical frequency is sent through the atomic gas and imaged onto a camera . assuming the gas is sufficiently thin that we can neglect double refraction , and neglecting scattering at large numerical aperture , such a probe measures the column - integrated complex dielectric susceptibility tensor , truncated to the two polarizations allowed for light propagating along the imaging axis . in other words ,
information about the atomic gas is encoded onto the absorption , phase shift , and polarization state of the optical probe field , properties that can be extracted by appropriate imaging methods .
for example , in _ absorption imaging _ , one images the probe light directly onto a camera . comparing images taken with and without the atomic gas within the field of view , one records the fractional transmission of probe light , @xmath158 , for each camera pixel ( labeled by position @xmath159 ) . in the simple case of linear scattering from atoms in a single initial internal state ,
the areal density of the gas is given by @xmath160 , with @xmath161 being the light absorption cross section given the atomic initial state and the optical polarization and detuning . yet ,
this simple interpretation of the absorption signal may be complicated by various factors such as saturation , optical pumping , polarization rotation , the displacement and acceleration of atoms due to light forces from the probe , collective scattering , inhomogeneous broadening , and others .
some of these complications are obviated by probing the gas with probe light well above the saturation intensity , so that atoms scatter light at a known maximum rate ( one - half the excited - state decay rate ) , and counting the number of absorbed photons @xcite ( or detecting the metered fluorescence @xcite ) .
an advantage of absorption imaging is that an atom can scatter very many photons in a single imaging pulse , improving the signal - to - noise ratio of the optical measurement . indeed , several groups have quieted absorption imaging to the point of measuring atomic distributions with sub - poissonian noise where the atom number @xmath162 in a given region is measured with uncertainty @xmath163 to reveal correlations generated by interatomic interactions or quantum statistics @xcite .
absorption imaging has two primary drawbacks .
first , absorption imaging is typically only a single - shot probe .
the photons `` absorbed '' by the atoms in such imaging are , in actuality , re - scattered out of the imaging system .
such photon scattering imparts significant momentum to atoms within the gas , typically adding sufficient energy and entropy that the atomic quantum system is irrevocably disturbed and must be discarded and created anew for further measurements .
the destructiveness of the measurement makes it challenging to use absorption imaging to characterize interesting temporal dynamics , such as equilibration dynamics , responses to sudden variations in the system hamiltonian , collective excitations , or temporal noise correlations . if the response of the system to a temporal perturbation is reproducible
, one can construct a time - series of measurements from many repeated experiments with variable delay before the absorption probe .
we note it is possible to get around the single - shot limitation by applying absorption imaging to just a small fraction of atoms extracted from the quantum gas .
for example , frelich _ et al . _ used a weak microwave pulses to transfer @xmath10rb atoms to a different hyperfine ground state before imaging them with light that was sufficiently far detuned from the optical transitions of the remaining gas that the trapped sample continued to evolve with little disturbance @xcite .
second , absorption imaging is also of limited use in probing gases at high optical densities , where imaging noise leads to large uncertainty in the measured column density . for this reason absorption imaging is rarely used to measure trapped atoms directly , _ in situ_. rather , absorption imaging is applied typically to gases _ ex situ _ , after they ve been released from their trap and allowed to expand significantly before probing a method known as _ time - of - flight imaging_. alternately , one may use _ dispersive imaging _ to probe high - density , trapped atomic gases _ in situ_. to attain the highest imaging resolution , one uses probe light sufficiently detuned so that diffraction , rather than refraction , is dominant , a condition achieved when the phase shift imparted on probe light on the order of 1 radian or less .
for high optical density gases , i.e. those for which absorption imaging is problematic , this condition also implies that optical dispersion dominates over optical absorption .
information on the atomic gas is now encoded in the phase of the transmitted light .
this phase is detected by one of several common imaging techniques , such dark - field @xcite , phase - contrast @xcite , or polarization - contrast @xcite imaging .
dispersive imaging is still somewhat destructive , owing to the residual off - resonant absorption by the gas .
the destructiveness is mitigated only by imaging at lower signal - to - noise ratios ( e.g. by using fewer photons or probe light detuned further from resonance ) ; only so much information can be garnered from the gas before it is destroyed . nevertheless , by parsing this information it is possible to obtain several _ in situ _ images of a gas , e.g. to measure the time evolution ( fig .
[ fig : soundprop ] ) or to measure several properties ( fig .
[ fig : larmor ] ) of a single sample .
, a repulsive potential is introduced suddenly within the middle of the elongated condensate .
images are taken at 1.3 ms intervals , beginning at @xmath164 ms .
the density excesses ( red ) travel along the condensate , revealing the speed of sound .
figure adapted from ref .
@xcite.,width=288 ] rb gas .
( a ) several brief polarization - rotation images are taken , measuring one component of the magnetization transverse to an applied magnetic field , while such magnetization undergoes larmor precession between image frames .
such larmor precession is evident by selecting data from a common location in each image frame ( yellow band in a ) and observing the temporal oscillation ( shown in b ) .
( c ) analyzing data from each image pixel , one extracts the spatially varying larmor precession amplitude and phase , or , equivalently , the two components of the transverse magnetization ( here denoted as @xmath165 and @xmath166 , represented either in gray scale , or in a color scale with brightness indicating amplitude and hue indicating orientation ) .
this image analyzes a degenerate gas equilibrating to a ferromagnetic state , as reported in ref .
@xcite.,width=5 ] the imaging probes discussed above are analogous to transmission electron or x - ray microscopy in which real - space images are obtained of solid - state materials . alternately , solid - state materials can be characterized in reciprocal or momentum space , e.g. by using laue or bragg x - ray diffraction to identify lattice planes and crystal orientations or neutron diffraction to identify magnetic order .
such scattering experiments on solids are most powerful in the form of angle resolved spectroscopy , by which one determines both the momentum and the energy absorbed by a material sample by scattering a probe particle .
a similar family of methods has been developed to characterize the structure of atomic gases by angle - resolved light scattering .
as for the methods used for solid - state materials , bragg scattering of light can be used to determine the dynamic and static structure factors of a quantum gas , revealing , for example , the role of interactions in suppressing density fluctuations @xcite or shifting excitation energies @xcite .
a notable difference in the quantum - gas applications is that coherent laser light interacts strongly with the gas and also possesses a wavelength that is on par with the physically relevant length scales of the gaseous sample .
for this reason , experimental applications of _ bragg spectroscopy _ typically make use of _ stimulated _ bragg scattering to characterize the sample . rather than quantifying the momentum and energy transfer by spectroscopic analysis of ( few ) scattered particles ,
one specifies such quantities _ a priori _ and with excellent resolution by the wavevector and energy difference between several plane - wave light beams impinging on the sample .
the excitation strength is then quantified , e.g. by detecting the amplification ( or suppression ) of one of the driving light fields @xcite , or the energy or momentum deposited in the gas @xcite .
the similar method of _ modulation spectroscopy _ is well suited to the study of lattice - trapped quantum gases .
instead of the optical plane waves used for bragg scattering , here , the optical excitation is effected by standing - wave modes of light , these being the optical lattice beams themselves .
the carrier and modulation sidebands of the amplitude - modulated lattice beams define the energy transferred to the medium in transitions at constant quasi - momentum .
such modulation spectroscopy has been used to characterize the mott insulator formed by lattice - trapped bose gases @xcite and the particle - hole excitations in fermionic samples @xcite .
a distinct feature of research on atomic gases is that measurements need not be confined to the perturbative regime .
for example , modulation spectroscopy is performed often by modulating the potential energy ( the periodic optical lattice ) by tens of percent . on the one hand ,
the use of such a non - perturbative measurement complicates its interpretation . on the other hand ,
such strong perturbations produce interesting non - equilibrium systems worthy of study in their own right , for example , bose - einstein condensates with macroscopic occupancy of a phonon state @xcite and dynamic systems in which the initial excitation of the system is fed back via matter - wave mixing or optical cavities @xcite .
the limits of measurement sensitivity are reached when a high - fidelity record of each and every atom of a many - body system is obtained .
remarkably , recent experimental developments are converging on this limit .
several approaches are being pursued .
for instance , it is possible to detect each atom in a two- @xcite or three - dimensional @xcite array .
to achieve such a measurement , atoms in an optical lattice are exposed to near - resonant light that is scattered copiously by the atoms as they are being actively laser cooled .
such fluoresced light is then imaged with sufficient spatial resolution to isolate the emission of each atom from that of its neighbor .
high measurement sensitivity has also been attained for metastable atomic species , such as the lowest state of orthohelium @xcite . upon striking a detector surface ,
a single metastable atom may undergo auger decay , releasing electrons that can be fed into a multi - channel electron multiplier .
other single - atom - sensitive techniques include cavity - enhanced optical detection @xcite and multiphoton ionization @xcite .
ultimately , such methods may allow for a detailed measurement of the state of an interacting many - body quantum system , providing a wealth of information beyond the standard correlation functions that are measured typically for solid - state systems .
as discussed above , the atomic physics toolbox provides ready means to extract data from gaseous systems .
however , to turn such data into a meaningful characterization of system properties , particularly in forms that are comparable to those that characterize solid - state materials , requires another round of ingenuity .
many of the major recent advances in cold atomic physics represent such ingenuity . here
we provide a few examples of how the raw data of measurements are interpreted as revealing the underlying physics of atomic quantum gases .
we are often asked , `` how do you measure such low temperatures ? ''
one measures the temperature , or other thermodynamic property , of a conventional material by connecting it with a reliable and calibrated sensor , which is a second material the properties of which are already fully understood . while different in practice , the concept behind measuring bulk properties of quantum gases is the same : temperature and pressure are measured by reference to the properties of an ideal gas .
for example , at thermal equilibrium , the velocity distribution of a non - degenerate ideal gas is given by the boltzmann distribution , @xmath167 .
one may then determine the temperature by measuring how fast the particles in such a gas ( of mass @xmath7 ) are moving . in principle , this can be done by seeing how quickly the gas escapes from its container once the trapping potential is turned off .
this simple concept can also be extended to some of the degenerate and interacting systems that are under study today .
one notes that the most highly excited atoms in the gas are less affected by interactions , and so a reliable temperature measurement can still be made by examining the high - velocity tails . however ,
with increasing interactions , one can not rely on the time - of - flight distribution being a true reflection of the _ in - situ _ velocity distribution : the gas particles scatter off each other during the expansion from the trap , resulting in thermal cooling and a modification of the velocity distribution .
other approaches to measuring velocity distributions , such as atom interferometry @xcite , also fail in strongly interacting gases as the doppler and interaction - induced effects are difficult to disentangle .
an alternate approach is to base one s measurements on the _ in situ _ spatial distribution of the gas
. analyses of such distributions rely typically on the local density approximation , i.e. that even in the inhomogeneity introduced by the non - constant trapping potential across the volume of the trapped gas , properties of the gas measured locally namely the density , compressibility , pressure , magnetization , etc .
are interrelated by the equation of state of a homogeneous quantum gas .
in fact , the inhomogeneous trapping potential of the gas becomes the vehicle by which such properties are measured in the first place .
the constant chemical potential of an equilibrated gas may be written as @xmath168 where , by the local density approximation , we interpret @xmath169 as the `` local '' chemical potential that determines local properties of the gas according to its equation of state .
consider that one measures the _ in situ _ density profile @xmath170 over a region where the trapping potential has a non - zero gradient @xmath171 .
one thereby obtains two thermodynamic quantities simultaneously : @xmath172 and @xmath173 . together with a means of determining @xmath174 and the temperature @xmath175 , one obtains an experimental characterization of the equation of state .
the fact that quantum gases decay away even as one wants to study them is certainly inconvenient . yet , this annoyance can also be turned to one s advantage . in a paper entitled `` coherence ,
correlations and collisions ... '' burt _ et al .
_ observed that bose - einstein condensates in their apparatus were decaying remarkably more slowly than non - degenerate gases @xcite .
examining closely the rate of loss of atoms from their gases , they found that three - body decay , losses that occur when a pair of atoms relax into a bound molecular state with the aid of a third collision partner , were diminished because the coherent matter - wave of a bose - einstein condensate contains fewer trios of atoms at short distances from each other than does an incoherent thermal gas .
the six - fold reduction of the zero - range third order correlation function , @xmath176 , is familiar also from the quantum optics of thermal and coherence sources of light .
detailed studies of loss rates ascribed to two- , three- , and four - body processes have been used to reveal the short - range correlations that arise in a number of strongly interacting systems . suppressed losses revealed the `` fermionization '' of one - dimensional bose gases as they enter the tonks - girardeau regime @xcite .
sharp increases and decreases in the three - body loss rate indicate the formation of three - body bound states due to strong two - body interactions @xcite , as predicted by efimov . a reduced loss rate in two - state fermi gases with strong interactions revealed the onset of local spin correlations and also the rapid formation of bound - state molecules in systems swept across feshbach resonances @xcite .
the lowest temperatures attained in cryogenic physics , in the @xmath89k range for bulk materials , are infernally high compared to the temperatures of quantum degenerate gases .
thus , by necessity , atomic gases are kept extremely well isolated from their thermal environment by placing the gas within an ultrahigh vacuum environment , with pressures below @xmath177 torr .
the residual background gas within the chamber , at thermal equilibrium with the chamber walls ( say at room temperature ) , is so dilute that its effect is just to deplete the trapped gas slowly , since background - gas collisions typically impart more energy to an atom than the depth of trap in which it is contained .
radiative coupling to the atomic gas is minimal since the atomic absorption spectrum is so sparse ( atoms are a very low emissivity object ) .
this latter observation may have to be revisited for atoms and , particularly , for molecules with more dense and complex spectra .
this extreme isolation preserves the cold gas long enough for it to arrive at low temperature by evaporative cooling and to be manipulated and probed by the experimentalist .
however , the isolation also presents us with a challenge : without an external thermal bath with which to exchange energy , or an external reservoir with which to exchange particles , can the atomic materials be truly regarded as having reached equilibrium ?
thermodynamically , quantum gases should perhaps be described as a microcanonical ensemble ; however , for a gas constantly subject to evaporative cooling , perhaps that picture is also inadequate . in any case
, one should exercise caution in importing findings from a grand - canonical quantum - statistical model , where systems are characterized by fixed chemical potential and temperature , rather than fixed energy and particle number .
the same concerns apply to quantum gases with a magnetic degree of freedom . in condensed - matter studies of magnetism ,
the magnetic susceptibility is typically measured by applying a magnetic field to the sample , and measuring its resultant magnetization . however , in quantum gases , the magnetization of the system can also be a conserved quantity given the absence of a `` spin reservoir . ''
one way to connect the results of such models to the properties of gases trapped in inhomogeneous potentials is via the local density approximation , a method described above ( sec .
[ sec : lda ] ) . aside from being limited by the presence of several conserved quantities , equilibration of quantum gases is also limited by the competition between the slowness of equilibration dynamics at low temperature and the finite lifetime or parasitic heating rates of the trapped gas .
as we demand increasingly precise information about the quantum gases in our laboratories , regarding , for example , the locations of transitions in a phase diagram or of subtle spatial features in a trapped gas , we require the atomic systems to be increasingly close to equilibrium , as highlighted by recent investigations of the effects of interactions on the bose - einstein condensation phase transition @xcite or of fluctuations in two - dimensional bose - hubbard systems @xcite
. the question of equilibration will continue to be important as cold - atom experiments make ever - deeper inroads into many - body quantum physics . on the flip side ,
the persistence of non - equilibrium states in quantum gases may be turned into one of their most appealing features . as discussed in the chapter by lamacraft and moore , quantum gases offer us access to non - equilibrium dynamics in complex quantum systems that may be impossible to study otherwise .
ultracold atomic physics provides an alternate approach to studying an increasing family of open problems in condensed - matter physics .
the experimental procedures are very different in the two fields , each being informed by a different intellectual legacy and by the different properties of the materials being studied .
but our understanding of the universal physics that is common to these fields is enriched by the complementary information and perspectives provided by the two research approaches . it is tempting to distinguish atomic physics research as more of a `` materials by design '' rather than a `` materials by happenstance '' approach to materials science .
an idealistic view of this field posits that atomic physics experiments are a form of quantum simulation where one first identifies a many - body quantum physics model that one wants to understand better , and then one rigs up this model into an actual physical realization using cold atoms .
one can muse whether such simulations are to be regarded as experimental research per se , or whether they represent a form of `` experimental many - body quantum theory . ''
however , the reality is that the field of cold atom physics is mostly driven by new experimental capabilities , rather than by the prior identification of a singular simulation goal .
the systems selected for study are those that are just now becoming possible to study because of the development of a new technique for preparing the system or probing its properties .
understanding these techniques is essential for knowing where the field is going next .
we hope the present chapter , and the remaining works in this book , help convey this knowledge .
we thank d.s . jin and r. hulet for permission to use figures from their works , j. guzman for graphical assistance , and other members of our research teams for their insights .
d.m.s - k acknowledges generous support from the nsf , afosr and dtra
. j.h.t . acknowledges nserc , afosr , and cifar .
both authors are supported by a grant from the aro with funding from the darpa ole program .
10 url # 1`#1`urlprefixhref # 1#2#2 # 1#1 s. peil , j. v. porto , b. l. tolra , j. m. obrecht , b. e. king , m. subbotin , s. l. rolston , w. d. phillips , patterned loading of a bose - einstein condensate into an optical lattice , phys .
a 67 ( 5 ) ( 2003 ) 051603 .
r. a. williams , j. d. pillet , s. al - assam , b. fletcher , m. shotter , c. j. foot , dynamic optical lattices : two - dimensional rotating and accordion lattices for ultracold atoms , opt .
express 16 ( 21 ) ( 2008 ) 1697716983 .
k. k. ni , s. ospelkaus , m. h. g. de miranda , a. peer , b. neyenhuis , j. j. zirbel , s. kotochigova , p. s. julienne , d. s. jin , j. ye , a high phase - space - density gas of polar molecules , science 322 ( 5899 ) ( 2008 ) 231235 .
j. deiglmayr , a. grochola , m. repp , k. mrtlbauer , c. glck , j. lange , o. dulieu , r. wester , m. weidemller , formation of ultracold polar molecules in the rovibrational ground state , phys .
lett . 101 ( 13 ) ( 2008 ) 133004 .
s. b. papp , j. m. pino , r. j. wild , s. ronen , c. e. wieman , d. s. jin , e. a. cornell , bragg spectroscopy of a strongly interacting @xmath179rb bose - einstein condensate , phys .
( 2008 ) 135301 .
j. heinze , s. gtze , j. s. krauser , b. hundt , n. flschner , d. s. lhmann , c. becker , k. sengstock , multiband spectroscopy of ultracold fermions : observation of reduced tunneling in attractive bose - fermi mixtures , phys .
107 ( 13 ) ( 2011 ) 135303 .
f. henkel , m. krug , j. hofmann , w. rosenfeld , m. weber , h. weinfurter , highly efficient state - selective submicrosecond photoionization detection of single atoms , phys .
105 ( 25 ) ( 2010 ) 253001 .
e. burt , r. ghrist , c. myatt , m. holland , e. cornell , c. wieman , coherence , correlations , and collisions : what one learns about bose - einstein condensates from their decay , phys .
79 ( 1997 ) 337 .
b. l. tolra , k. m. ohara , j. h. huckans , w. d. phillips , s. l. rolston , j. v. porto , observation of reduced three - body recombination in a correlated 1d degenerate bose gas , phys .
92 ( 19 ) ( 2004 ) 1904014 .
t. kraemer , m. mark , p. waldburger , j. g. danzl , c. chin , b. engeser , a. d. lange , k. pilch , a. jaakkola , h. c. ngerl , r. grimm , evidence for efimov quantum states in an ultracold gas of caesium atoms , nature 440 ( 7082 ) ( 2006 ) 315318 .
g. b. jo , y. r. lee , j. h. choi , c. a. christensen , t. h. kim , j. h. thywissen , d. e. pritchard , w. ketterle , itinerant ferromagnetism in a fermi gas of ultracold atoms , science 325 ( 5947 ) ( 2009 ) 15211524 . | experiments on solid - state materials and atomic quantum gases are increasingly investigating similar concepts in many - body quantum physics . yet
, the flavor of experiments on the gaseous atomic materials is different from that of conventional materials research . here , we summarize some aspects of atomic physics and some of the common technical elements of cold - atom experiments which underlie the investigations described in the remaining chapters of this volume .
the broad appeal of research on quantum gases relies on the universality of many - body quantum physics .
for example , regardless of whether it is constructed using electrons ( as in superconductors ) , neutrons ( as in neutron stars ) , or neutral atoms of different hyperfine spin ( as in ultracold lithium gases ) , a system of strongly interacting , mobile fermions will show the same phenomenology .
similarly , nonlinear and quantum optics can be treated in a common theoretical scheme , regardless of whether the bosonic fields under investigation correspond to massless photons or massive rubidium atoms .
such universality allows ultracold atomic physics to contribute significantly to fields as diverse as condensed matter physics , high energy astrophysics , and quantum optics . as such
, the collection of works within this book is meant to bridge the gap between practitioners of these diverse fields so as to make the exchange among them more productive . yet , despite these appealing similarities , there do remain system - specific considerations that must be kept in mind in comparing physical systems built from different basic ingredients .
this chapter discusses some of the atom - specific aspects of ultracold atomic physics experiments .
we focus on two main topics : the common experimental techniques of quantum gas experiments and the nature of atom - atom interactions
. our treatment will be at a colloquium level ; for more detailed explanations , references are given . |
in order to keep the presentation simple , i prefer to deal only with binary variables ( ising spins ) @xmath0 and hamiltonian containing up to two - body interactions , i.e. external fields and pairwise couplings .
thus , the most general model i want to study is defined by the following joint probability distribution over @xmath1 ising variables @xmath2\ ; , \label{measure}\ ] ] where the partition function @xmath3 is a normalizing constant , that depends on all the couplings @xmath4 and the external fields @xmath5
. please notice that the temperature parameter has been absorbed in the definition of external fields and couplings .
all the required information about the model is encoded in the free - energy @xmath6 in the rest of this section i summarize the most common mfa to the free - energy : i am particularly interested in deriving the self - consistency equations for the magnetizations that are used in section [ sec : corr ] for obtaining 2-point correlations .
the simplest mfa , also known as naive mf ( nmf ) , approximates the model in terms of local magnetizations @xmath7 , where the angular brackets represent the average w.r.t .
the measure in eq.([measure ] ) .
the corresponding approximation to the free - energy is @xmath8 + \sum_i h_i m_i + \sum_{i \neq j } j_{ij } m_i m_j\;,\ ] ] where @xmath9 and the @xmath10 must be fixed according to the self - consistency equations @xmath11\;.\label{mimf}\ ] ] a better mfa can be obtained by considering also the onsager reaction term @xcite , leading to the following tap approximated free - energy and self - consistency equations @xmath12 + \nonumber \\ & & + \sum_i h_i m_i + \sum_{i \neq j } \left ( j_{ij } m_i m_j + \frac12 j_{ij}^2 ( 1-m_i^2)(1-m_j^2 ) \right)\ ; , \label{ftap}\\ m_i & = & \tanh\left[h_i + \sum_j j_{ij } \big(m_j - j_{ij } ( 1-m_j^2 ) m_i\big ) \right]\;.\label{mitap}\end{aligned}\ ] ] in the tap approximation , when computing the marginal probability of spin @xmath13 ( i.e. its magnetization @xmath10 ) , the reaction term modifies the marginal probabilities of the neighboring spins , @xmath14 , in order to try to remove the effect of the spin @xmath13 under study .
it has been recognized @xcite that @xmath15 and @xmath16 are only the first two terms of the expansion of @xmath17 in small couplings @xmath18 at fixed magnetizations @xmath19 .
this expansion contains @xcite both loop terms , like @xmath20 , and terms with higher powers of a single coupling , i.e. @xmath21 : the latter terms , that correspond to considering recursively the reaction to the reaction between spins @xmath13 and @xmath22 , can be resummed and lead to the ba .
the ba gives a description of the model in terms of magnetizations @xmath10 and connected correlations @xmath23 between neighboring spins ( i.e. spins connected by a non - zero coupling @xmath24 ) .
the ba can be derived in two equivalent ways . the first way consists in finding values of @xmath25 and @xmath26 minimizing the following free - energy @xmath27 + \nonumber\\ + \sum_i ( 1-d_i ) \left [ h\left(\frac{1+m_i}{2}\right ) + h\left(\frac{1-m_i}{2}\right ) \right ] + \sum_i h_i m_i + \sum_{i \neq j } j_{ij } ( c_{ij } + m_i m_j)\;,\label{eq : fba}\end{aligned}\ ] ] where @xmath28 is the degree of spin @xmath13 , i.e. the number of its neighboring spins .
in eq.([eq : fba ] ) the last two terms correspond to the average value of the energy at given magnetizations and neighbouring correlations , while the first two terms correspond to the entropy of the bethe approximation to the joint probability distribution of the @xmath1 spin variables , @xmath29 where the first product runs over all pair of neighboring spins and the two - spins and single - spin marginal probabilities are given respectively by @xmath30/4 $ ] and @xmath31 .
the conditions @xmath32 can be solved analytically and lead to @xmath33 where @xmath34
. please note that eq.([eq : ip ] ) is identical to eq.(26 ) in ref . and
this is a further confirmation that resumming all 2-spin terms in the plefka expansion leads to the ba .
moreover eq.([eq : ip ] ) has been used in the literature @xcite as the independent - pair ( ip ) approximation for inferring couplings from magnetizations and correlations : such an approximation infers the coupling @xmath24 by assuming spins @xmath13 and @xmath22 form an isolated pair with magnetizations @xmath10 and @xmath35 and correlation @xmath36 . unfortunately under this ip approximation computing the external fields in not immediate and moreover even the estimates of the couplings are rather poor ( see section [ sec : numeric ] ) . by making the substitution @xmath37 in @xmath38 one
can obtain the bethe free - energy only in terms of magnetizations , from which the self - consistency equations for the magnetizations can be derived .
however this derivation requires a rather complicated algebra and i prefer to obtain the same equations in a much simpler alternative way . in the so - called cavity method @xcite local magnetizations @xmath10 and neighbouring
correlations @xmath36 are expressed in terms of some auxiliary variables , the cavity magnetizations @xmath39 ( i.e. the mean value of @xmath13 in the absence of a neighboring spin @xmath22 ) : @xmath40 cavity magnetizations must satisfy the self - consistency equations @xmath41 \;.\label{spemij}\ ] ] these equations are often solved by an iterative algorithm known as belief propagation ( bp ) @xcite : in case of convergence , the fixed point of bp gives directly the bethe free - energy that admits an expression in terms of cavity magnetizations only @xcite . in order to obtain a closed set of self - consistency equations in the magnetizations @xmath25 ,
let me solve eqs.([mi]-[mj ] ) for the cavity magnetizations and find @xmath42 where @xmath43 the sign in front of the square root has been chosen such that @xmath44 as it should .
a consistency check can be made by substituting expressions ( [ mijmji ] ) in eq.([cij ] ) to obtain again the result in eq.([cijprimo ] ) .
finally , combining eq.([mi ] ) and eq.([spemij ] ) , it is possible to obtain the self consistency equation for the magnetizations under the ba : @xmath45\;.\label{miba}\ ] ] it is fair to comment that the use of this formula for finding bethe magnetizations is not a good idea : indeed an iterative solution of eq.([miba ] ) is typically more unstable than bp solving eq.([spemij ] ) .
my interest in this formula is that it involves only physical magnetizations ( not cavity ones ) and can be used to obtain correlations ( see section [ sec : corr ] ) and to solve in a fast way the inverse ising problem ( see section [ sec : numeric ] ) .
a series expansion of the exponent in eq.([miba ] ) for small couplings gives @xmath46 and one recognizes that the first two terms of the expansion are the naive mf approximation and the onsager reaction term .
this expansion should make clearer that the ba is a way of considering recursively all the reactions between a pair of neighboring variables .
a preliminary step to solve the inverse ising problem by any mfa is to derive an analytical expression for the pairwise correlations as a function of the coupling constants .
actually , the mfa discussed in section [ sec : mfa ] do not provide information about the correlation between distant variables : indeed , naive mf and tap approximations give @xmath47 for any pair of variables , and the ba only provides an expression for correlation between neighboring spins , see eq.([cijprimo ] ) , which is trivially @xmath48 in case of null magnetizations . nonetheless , a closed set of equations for the connected correlations with the parameter @xmath36 appearing in the ba : the two coincide only when the ba is exact .
] , @xmath49 for any pair @xmath50 , can be derived from the magnetizations self - consistency equations , eqs.([mimf ] ) , ( [ mitap ] ) , ( [ miba ] ) , through the linear response @xcite @xmath51 the inverse correlation matrices @xmath52 for the three mfa discussed above are given by the following expressions : @xmath53\delta_{ij } - \left(j_{ij } + 2 j_{ij}^2 m_i m_j \right)\;,\label{cinvtap}\\ \text{bethe } \quad ( c_\text{ba}^{-1})_{ij } & = & \left[\frac{1}{1-m_i^2}-\sum_k \frac{t_{ik}f_2(m_k , m_i , t_{ik})}{1-t_{ik}^2 f(m_k , m_i , t_{ik})^2}\right]\delta_{ij } - \frac{t_{ij}f_1(m_j , m_i , t_{ij})}{1-t_{ij}^2 f(m_j , m_i , t_{ij})^2}\;,\qquad\label{cinvba}\end{aligned}\ ] ] where @xmath54 and @xmath55 . from these expressions
one can obtain directly any correlation by simply computing the inverse of a matrix .
please note that eq.([cinvba ] ) gives exactly the same solution found by the suscprop iterative algorithm @xcite , which is presently considered one among the best inference algorithms .
the main advantage of eq.([cinvba ] ) is that it always provides the correlation matrix , even in those cases where suscprop does not converge to the fixed point .
moreover inverting a matrix takes roughly the same time of a single iteration of suscprop , and so using eq.([cinvba ] ) is much faster than running suscprop , even when the latter converges . nevertheless , it is fair to notice that the use of eq.([cinvba ] ) does not solve all the problems related to the lack of convergence of suscprop .
indeed , during the many tests i have run , i noticed that often the lack of convergence of suscprop does correspond to the ba fixed point becoming unphysical : in these cases , by inverting the correlation matrix provided by eq.([cinvba ] ) , one gets an unphysical correlation matrix ( e.g. a correlation matrix with negative diagonal elements ! ) . in this sense
the lack of convergence of suscprop gives a warning that the `` blind '' use of eq.([cinvba ] ) does not provide .
so , a general suggestion when using the above formulas , providing an analytical expression for the correlation matrices under a mfa , is to check explicitly the physical consistency of the outcome .
one may comment that eq.([cinvba ] ) contains the magnetizations and the iterative computation of these ( i.e. the bp algorithm ) suffers the same convergence problems of suscprop : this is easy to prove , given that the homogeneous suscprop equations are nothing but the iterative equations for evolving under bp a small perturbation in the magnetization , and so bp is unstable if suscprop does not converge .
however there are provably convergent algorithms for the computation of magnetizations under the ba @xcite : the use of these algorithms in conjunction with eq.([cinvba ] ) allows a direct computation of correlations under the ba .
moreover there are situations where magnetizations are known a priori and eq.([cinvba ] ) can be applied directly : e.g. when symmetries in the probability measure force magnetizations to be zero , or in the inverse ising problem , where magnetizations are given as an input to the problem .
in the rest of the paper i deal mainly with these two cases .
a preliminary ranking of mfa can be done on the basis of how good are their estimates of correlations , given the couplings .
indeed i expect that the better is this estimate , the better will be the solution to the inverse problem . for simplicity
i concentrate on models with no external fields and the couplings are multiplied by a parameter @xmath56 ( the inverse temperature ) such that the difficulty of the inference problem increases with @xmath56 . in case of null magnetizations , the expressions for the inverse correlation matrices simplify a lot @xmath57\delta_{ij } - j_{ij}\;,\\ \text{bethe } \qquad ( c_\text{ba}^{-1})_{ij } & = & \left[1+\sum_k \frac{t_{ik}^2}{1-t_{ik}^2}\right]\delta_{ij } - \frac{t_{ij}}{1-t_{ij}^2}\;,\label{cinvbam0}\end{aligned}\ ] ] since @xmath58 and @xmath59 . given that for @xmath60 the expressions for the correlation matrices are much simpler ,
i report also those that can be obtained from the plefka expansion at the third and fourth order @xmath61\delta_{ij } - \left ( j_{ij } + \frac23 j_{ij}^3 \right)\;,\\ 4^\text{th } \text { order } \qquad ( c_4^{-1})_{ij } & = & \left[1 + \sum_k j_{ik}^2 + 2 \sum_{k,\ell } j_{ik } j_{k\ell } j_{\ell i } + \frac13 \sum_k j_{ik}^4 + 2 \sum_{k,\ell , m } j_{ik } j_{k\ell } j_{\ell m } j_{mi } \right]\delta_{ij } + \nonumber\\ & & - \left ( j_{ij } + \frac23 j_{ij}^3 + 2 j_{ij}^2 \sum_k j_{ik } j_{kj } \right)\;.\label{last}\end{aligned}\ ] ] the purpose is to understand if and how much does the estimate of the correlation matrix improve by adding terms in the plefka expansion .
i have tested the accuracy of formulas in eqs.([first]-[last ] ) for ferromagnetic ( @xmath62 ) and spin glass ( @xmath63 ) models defined on fully connected ( fc ) topologies , on a 2d square lattice and on a 3d cubic lattice . in diluted versions of these models
a fraction @xmath64 of couplings has been set to zero . in models defined on fc graphs
the couplings have been normalized such as to have a critical inverse temperature @xmath65 in the thermodynamic limit .
the discrepancy between true correlations @xmath66 and those inferred @xmath67 is defined as @xmath68 in figure [ plot_2d_all ] and [ plot_other ] i report the typical behavior of the error @xmath69 between exact and estimated correlation matrices for 5 different mfa .
figure [ plot_2d_all ] shows results for models defined on a 2d square lattice , while figure [ plot_other ] refers to fc and 3d topologies . in order to compare the mfa estimates with the exact correlation matrices i am studying here small systems , but the qualitative behavior does not change for larger sizes .
( the qualitative behavior does not change for larger sizes ) . ] for typical samples of the fully - connected spin glass ( sk model ) of size @xmath70 and of the 3d diluted ferromagnet of size @xmath71 . ]
although the quantitative behavior of @xmath69 depends on the specific sample , some general statements can be made : naive mf is typically the worst mfa and shows many spurious singularities ( roughly one for each peak in @xmath69 ) ; tap and @xmath72 order approximations typically show no ( or very rare ) singularities ; the best estimate is typically provided by ba and tap , with ba being the best unless it has a singularity ( in this case tap becomes the best at lower temperatures , higher @xmath56 ) .
these results suggest that increasing the number of terms in the plefka expansion does not always improve the estimate of the correlation matrix ( as one could have naively expected ) . on the basis of these preliminary results
i believe it is relevant to consider only tap and ba for the inverse ising problem , together with other inference methods ( see section [ sec : inverse ] ) . in the left panel of figure [ plot_other ] the results obtained by tap and ba
are almost perfectly superimposed ( indeed the former in not well visible ) .
this is expected since the tap approximation is exact for the sk model at high temperatures ( @xmath73 ) in the large @xmath1 limit : so the ba can not improve it , but in @xmath74 corrections .
indeed a careful analysis shows a tiny improvement of ba over tap around the critical temperature , where @xmath74 corrections are stronger .
please note that in the right panel of figure [ plot_other ] the high temperature ( small @xmath56 ) behavior of @xmath69 is very different than in previous plots : indeed for @xmath75 , @xmath69 goes to a constant , instead of decreasing with a power law in @xmath56 ( as in figure [ plot_2d_all ] and in the left panel of figure [ plot_other ] ) .
this is due to the fact that the comparison has not been made with the exact correlation matrix , but with correlations measured from a monte carlo ( mc ) simulation .
actually , in this case , i have used the wolff algorithm and the correlation matrix has been computed from @xmath76 independent measures .
the difference between the error due to the mfa and the error due to mc noisy data can be better appreciated in figure [ plot_2d_mcvsexact ] : in the high temperature region the error does not decrease below a limiting value given roughly by the inverse of the square root of the number of measures .
expressions in eqs.([first]-[last ] ) are intrinsically approximated , and turn out to be correct only in some particular cases .
naive mf and tap approximations ( as well as @xmath77 and @xmath72 orders approximations ) , being the first orders in a small couplings expansion , are exact only in the limit of very weak couplings ( either high temperature or fully - connected models in the large @xmath1 limit ) .
the ba , on the contrary , is exact also for coupling intensities @xmath78 , but only if the interacting network is a tree ; on random graph models ( which are locally tree - like ) the ba turns out to be correct as long as the model has only one state ( modulo the known symmetries ) . on any other model
those expressions are approximated and it is worth trying to improve it .
let me first notice that any of the above mfa returns in general a value for the self - correlation differing from the exact one , i.e. @xmath79 ( for simplicity i consider the case of null magnetizations , but the argument is general ) .
this fact can be easily explained , noticing that all the above mfa assume that correlations along loops are vanishingly small ( at least in the large @xmath1 limit ) . on the contrary , on any loopy graph , like e.g. regular lattices , correlations along loops are important and may alter significantly the mean - field estimates .
a general solution to this problem is still not available , although many work is in progress to include loop contributions to mfa @xcite .
what i am proposing here is a simple heuristic improvement .
once the correlation matrix @xmath80 is computed by one of the approximations described in section [ sec : corr ] , a properly _ normalized _ correlation matrix can be defined @xmath81 by definition @xmath82 and also off - diagonal element may approximate better the true correlations .
the reason for this is that the loops neglected in mfa actually modify in a similar way both self - correlations @xmath83 and off - diagonal correlations @xmath80 , and the heuristic normalization in eq.([normc ] ) is assuming that the modifying factor only depends on the loop structure around sites @xmath84 and @xmath85 ( which is certainly wrong for distant sites , but may be a reasonable approximation for closed - by sites ) . for a typical sample of the 2d diluted ferromagnet of size @xmath86 .
the error @xmath69 has been computed with respect to the exact correlation matrix and with respect to the one measured in mc simulations .
full points show the error obtained with the normalization trick . ] in figure [ plot_2d_mcvsexact ] full points show that the error @xmath69 in the ba decreases by roughly one order of magnitude if normalized correlations are used . on the contrary ,
the tap result is not very sensitive to this normalization : the reason is that the estimates of the self - correlations in tap remain quite close to the right value , specially if compared to ba estimates that diverge at the singularity ( mark by a peak in figure [ plot_2d_mcvsexact ] ) . on the right of such a peak
the error obtained by the normalized ba is not reported because eq.([normc ] ) can not be used , since several ba estimates for self - correlations are negative .
this is the problem of the ba fixed point becoming ( strongly ) unphysical , already discussed in section [ sec : corr ] : indeed by running suscprop on this sample one would observe convergence only for @xmath56 smaller than the peak location .
i would like to stress again that checking the physical consistency of a solution based on a mfa is very important : for the sample shown in figure [ plot_2d_mcvsexact ] , even without knowing the exact correlations , one should switch from the ba to tap , when the former reaches the singularity ( that manifests e.g. in suscprop not converging or in self - correlations diverging ) , the correlation matrix can be computed by mean of eq.([cinvba ] ) .
however in the general case , bp does not converge in presence of long range correlations , i.e. after the singularity , and one must resort to other mfa . ] .
moreover there are cases ( e.g. homogeneous fc models ) where the spurious singularity induced by the mfa in a system of finite size is such that @xmath83 and @xmath80 diverge with the same law at the spurious critical point , while the normalized correlation @xmath87 stays finite ( and much closer to the true one ) .
for example for the fc ferromagnetic model the normalized correlation @xmath88 estimates the true correlation with an error roughly half than the one of @xmath89 for any of the 5 mfa considered here .
i consider 4 different approximations for solving the inverse ising problem .
the simplest one is the independent - pair ( ip ) approximation , already discussed in section [ sec : mfa ] and recalled here for convenience @xmath90 among the mfa which can be derived from the plefka expansion , i consider only tap and ba , because are those performing better in the direct problem of estimating correlations ( see section [ sec : corr ] ) .
the corresponding expressions for the inferred couplings can be obtained by solving the equation @xmath91 for tap and the equation @xmath92 for the ba , thus leading to @xmath93\;.\label{ba}\end{aligned}\ ] ] the fourth approximation i am considering has been obtained from a small correlation expansion by sessak and monasson @xcite and has been further simplified in ref . to the following expression @xmath94 for each approximation ,
i measure the error in inferred couplings @xmath95 with respect to the true ones @xmath24 by the following expression @xmath96 i study both diluted ferromagnetic model with a fraction @xmath97 of non - zero couplings ( @xmath98 ) and undiluted spin glass models ( @xmath99 with probability @xmath100 ) .
i also consider several topologies : 2d square lattices , 3d cubic lattices , random regular graphs with fixed degree @xmath101 and fully connected ( fc ) graphs . in the latter case
the couplings are normalized in order to have a phase transition at @xmath65 in the thermodynamic limit .
i restrict the study to models of small sizes , with @xmath1 ranging between 20 and 100 , because these are the sizes for problems of biological interest .
moreover , as discusses below , the number @xmath102 of independent measurements of the correlation matrix that make inferred coupling reasonably good grows linearly with @xmath1 and so for larger systems the number of measurements needed become too large .
the data shown in section [ sec : numeric ] have been obtained with @xmath103 independent measures of the correlation matrix ( unless differently stated ) and going to much larger values seems to me rather unrealistic if compared with practical applications .
the trick of normalizing the correlation matrix to improve inference ( see section [ sec : trick ] ) can be extended to the inverse ising problem . in practice
, it corresponds to solve _ all _ the equations relating the inverse correlation matrix @xmath52 to the couplings @xmath24 , including also those for the diagonal elements which are usually ignored .
let me illustrate the new method for the simple case of the tap approximation with null magnetizations . in this case ,
solving the inverse ising problem only on the off - diagonal elements is equivalent to solve the equations @xmath104 but the diagonal equations are in general unsatisfied @xmath105 where @xmath106 is the inverse correlation matrix estimated by tap once coupling @xmath24 are given .
please notice that diagonal elements @xmath107 are fully determined once the off - diagonal elements are known .
following eq.([normc ] ) i would like to normalize the estimated matrix @xmath106 , and produce a normalized inverse correlation matrix @xmath108 , matching better the true inverse correlation matrix @xmath52 .
in other words i would like to solve the equations @xmath109 where the @xmath1 variables @xmath110 are exactly those necessary to solve the new @xmath1 diagonal equations .
physically speaking , @xmath111 should be the self - correlation @xmath83 produced by the mfa , when using the right couplings @xmath24 .
when the @xmath110 becomes very different from 1 , then the mfa is working very badly ; however i expect situations where some of the errors produced by the mfa can be compensated by this normalization trick .
unfortunately the solution to the new equations , those involving both @xmath24 and @xmath110 , does not have an analytical expression and need to be solved numerically .
i have adopted an iterative solution , which is very fast ( when it converges ) . in practice ,
i start with all @xmath112 and then , iteratively , first i solve the off - diagonal equations , thus getting an estimate for the couplings , and then i solve the diagonal equations to obtain a new estimate for the @xmath113 s .
i repeat this iterative procedure , updating @xmath113 s with a damping factor , until the @xmath113 s variations is below a threshold ( typically @xmath114 ) . from the many tests i have run
, i noticed that this normalization trick is more relevant for unfrustrated models ( as the diluted ferromagnets studied below ) or models containing regions very weakly frustrated ( those usually leading to griffith singularities ) .
most probably in these weakly frustrated regions correlations get self - reinforced by the loops ( ignored in the mfa ) and thus the normalization trick may improve coupling estimates . on the contrary for strongly disordered models , like spin glasses , the effect of the normalization trick depends a lot on the disorderd sample and it does not seem to give a clear improvement on average . the use of the normalization trick for improving inference in the inverse ising problem may resemble the diagonal - weight trick introduced in ref . , but it is actually very different . in the diagonal - weight trick , the self - couplings @xmath115 are allow to take non - zero values in order to solve all the equations @xmath116 , while in the normalization trick the self - couplings @xmath115 remain null .
it has been shown @xcite that the the first order approximation ( nmf ) with the diagonal - weight trick improves over the second order approximation ( tap ) in estimating magnetizations .
however the estimates for the couplings do not improve at all , because the off - diagonal equations are left unchanged by the diagonal - weight trick . on the contrary ,
the normalization trick used here does change the estimates for the couplings . moreover , being based on the very general requirement that self - correlations must take the right value , it can be applied to any approximation .
let me start by making some general statements that summarize the numerical results shown in this section .
among the four approximations studied ( ip , tap , sm and ba ) it seems in general true that : * ip always provides the worst estimate , especially at high temperatures ( low @xmath56 ) ; * ba always outperforms tap ; * between sm and ba , the former is better in the high temperature ( low @xmath56 ) phase , while the latter is better at lower temperatures ( higher @xmath56 values ) ; * at low @xmath56 , the error @xmath117 in couplings inference is completely dominated by the uncertainty on the correlation matrix and it does not depend on the inference method : for this reason the range where sm is the best method becomes tiny , especially for noisy data ; * at high @xmath56 , the error @xmath117 produced by tap and sm diverge , the one by ip stays limited , but very high , and only ba may have a reasonable error ; * for diluted ferromagnetic models the normalization trick works fine and thus ba with the normalization trick is the best method overall ; * in presence of an external field , i.e. when magnetizations are different from zero , tap and ba stop working at high enough @xmath56 values ( i.e. these methods do not admit a solution ) ; given that at high @xmath56 neither ip nor sm provide acceptable inferred couplings , i conclude that in such a situation the inverse ising problem needs to be solved by other methods not explored in the present work . in inferred couplings on the number @xmath102 of independent measures for the correlation matrix .
the model is a ferromagnet of @xmath70 variables on a regular random graph of fixed degree @xmath101 , whose critical temperature is marked by the vertical dotted line . ]
let me start discussing the high temperature ( low @xmath56 ) regime . in figure
[ fig : numsam ] are reported the errors @xmath117 in inferring the couplings of a ferromagnetic model of @xmath70 variables on a regular random graph of fixed degree @xmath101 .
for simplicity i am plotting only the results obtained with sm and ba .
data in the upper curves have been computed using a correlation matrix averaged over @xmath102 independent measurements , while data in lower curves have been computed from the exact correlation matrix .
it is clear that the low @xmath56 regime is completely dominated by the uncertainty in the correlation matrix ( as already noticed in ref .
@xcite ) , and the error in this regime is independent on the inference method used .
for this reason the good performances of sm in this regime are actually washed out and , even for @xmath103 measurements , the improvement of sm over ba is very limited ( see figure [ fig : numsam ] ) .
moreover such an improvement tends to become smaller by increasing the system size because in the low @xmath56 regime the error goes like @xmath118 i think that comparing inference methods by using the exact correlation matrix is rather unrealistic , given that in any practical application the correlations are always known with some uncertainty . so in presenting below the numerical results i always consider the case with @xmath103 independent measurements for the magnetizations and the correlations .
( left ) and @xmath119 ( right ) .
the vertical dotted lines mark the locus of the ferromagnetic phase transition in the thermodynamic limit.,title="fig : " ] ( left ) and @xmath119 ( right ) .
the vertical dotted lines mark the locus of the ferromagnetic phase transition in the thermodynamic limit.,title="fig : " ] . shown
are typical samples for 4 different dilutions : 0.6 , 0.7 , 0.8 and 0.9 .
the vertical dotted lines mark the loci of the ferromagnetic phase transitions in the thermodynamic limit.,title="fig : " ] . shown
are typical samples for 4 different dilutions : 0.6 , 0.7 , 0.8 and 0.9 .
the vertical dotted lines mark the loci of the ferromagnetic phase transitions in the thermodynamic limit.,title="fig : " ] . shown
are typical samples for 4 different dilutions : 0.6 , 0.7 , 0.8 and 0.9 .
the vertical dotted lines mark the loci of the ferromagnetic phase transitions in the thermodynamic limit.,title="fig : " ] . shown
are typical samples for 4 different dilutions : 0.6 , 0.7 , 0.8 and 0.9 .
the vertical dotted lines mark the loci of the ferromagnetic phase transitions in the thermodynamic limit.,title="fig : " ] in figure [ ferrobethe ] i am showing the error in the couplings inferred by several approximation for a ferromagnet on a random regular graph with fixed degree @xmath101 .
the two panels correspond to sizes @xmath70 ( left ) and @xmath119 ( right ) and show that the qualitative behavior is mostly size independent .
also the dependence on the specific sample ( i.e. , on the random graph ) is rather weak .
the data in figure [ ferrobethe ] support many of the statements written above : ( i ) ip is a very bad approximation even in the low @xmath56 regime ; ( ii ) ba always outperforms tap ; ( iii ) sm is better than ba only in the low @xmath56 regime , but here the error is dominated by the uncertainty in the correlation and increases with the system size , so the improvement of sm over ba is tiny ; ( iv ) errors in tap and sm diverge for large @xmath56 , while those in ip and ba remains finite , although very large ; ( v ) the normalization trick works nicely and gives actually the best result in a wide range of temperatures .
the data for tap with the normalization trick ( label `` tap norm '' ) are interrupted because at large @xmath56 the iterative procedure i am using for finding the parameters @xmath120 stops converging .
the same qualitative conclusions can be reached by studying a diluted ferromagnet on a 2d square lattice for several different dilutions ( see figure [ ferro2d ] ) . in particular the relative quality of the approximations
seems to be independent on the dilution , and the ba with the normalization trick outperforms the other inference methods .
however i notice that , while the error of ba ( with and without the normalization trick ) at the critical temperature is roughly independent on the dilution , the errors made by tap and sm tend to increase when the dilution is stronger and the system becomes more heterogenous . .
shown are two typical samples with different dilutions .
the vertical dotted lines mark the loci of the ferromagnetic phase transitions in the thermodynamic limit.,title="fig : " ] . shown
are two typical samples with different dilutions .
the vertical dotted lines mark the loci of the ferromagnetic phase transitions in the thermodynamic limit.,title="fig : " ] all the conclusions reached for the 2d case , perfectly apply also to the case of a diluted ferromagnetic model on a 3d cubic lattice ( see figure [ ferro3d ] ) .
so it is very reasonable to conclude that the statements made at the beginning of this section apply to any diluted ferromagnet independently on the specific topology . ) on different topologies : random regular graphs with fixed degree 4 ( upper panels ) , 3d cubic and 2d square lattices ( lower panels ) .
the vertical dotted lines mark the loci of the spin glass phase transitions in the thermodynamic limit.,title="fig : " ] ) on different topologies : random regular graphs with fixed degree 4 ( upper panels ) , 3d cubic and 2d square lattices ( lower panels ) .
the vertical dotted lines mark the loci of the spin glass phase transitions in the thermodynamic limit.,title="fig : " ] ) on different topologies : random regular graphs with fixed degree 4 ( upper panels ) , 3d cubic and 2d square lattices ( lower panels ) .
the vertical dotted lines mark the loci of the spin glass phase transitions in the thermodynamic limit.,title="fig : " ] ) on different topologies : random regular graphs with fixed degree 4 ( upper panels ) , 3d cubic and 2d square lattices ( lower panels ) .
the vertical dotted lines mark the loci of the spin glass phase transitions in the thermodynamic limit.,title="fig : " ] also for spin glass models the conclusions reached above apply very well : the relative goodness of the various approximations is roughly unchanged .
the only major difference is that the normalization trick does not provide any more a clear improvement : its performances strongly depend on the disorder sample . in figure [ sgmodels ]
i am reporting the error on the inferred coupling for spin glass models on a random regular graph with fixed degree @xmath101 , on a 3d cubic lattice and a 2d square lattice .
again , willing to suggest a general purpose inference method , the choice is clearly in favor of the ba .
please notice that by running suscprop it would be impossible to obtain the results shown in figure [ sgmodels ] , because of the limited range of convergence of such an algorithm .
indeed for spin glass models on a random graph suscprop stops converging around the critical temperature and for spin glass models on a regular lattice it stops converging even before , well into the high temperature phase : for example for the a spin glass model on a 2d square lattice it converges up to @xmath121 @xcite and on a 3d cubic lattice up to @xmath122 @xcite . in this sense , the use of the new formula in eq.([ba ] ) is really innovative .
( upper left , just for comparison ) , sk model @xmath123 ( upper right ) , sk model @xmath124 ( lower left ) and spin glass on a random regular graph with fixed degree @xmath101 and @xmath125 ( lower right ) .
the vertical dotted lines mark the loci of the spin glass phase transitions in the thermodynamic limit . in presence of an external field
tap and ba cease to have a solution for high enough @xmath56 values.,title="fig : " ] ( upper left , just for comparison ) , sk model @xmath123 ( upper right ) , sk model @xmath124 ( lower left ) and spin glass on a random regular graph with fixed degree @xmath101 and @xmath125 ( lower right ) .
the vertical dotted lines mark the loci of the spin glass phase transitions in the thermodynamic limit . in presence of an external field
tap and ba cease to have a solution for high enough @xmath56 values.,title="fig : " ] ( upper left , just for comparison ) , sk model @xmath123 ( upper right ) , sk model @xmath124 ( lower left ) and spin glass on a random regular graph with fixed degree @xmath101 and @xmath125 ( lower right ) .
the vertical dotted lines mark the loci of the spin glass phase transitions in the thermodynamic limit . in presence of an external field
tap and ba cease to have a solution for high enough @xmath56 values.,title="fig : " ] ( upper left , just for comparison ) , sk model @xmath123 ( upper right ) , sk model @xmath124 ( lower left ) and spin glass on a random regular graph with fixed degree @xmath101 and @xmath125 ( lower right ) .
the vertical dotted lines mark the loci of the spin glass phase transitions in the thermodynamic limit . in presence of an external field
tap and ba cease to have a solution for high enough @xmath56 values.,title="fig : " ] let me finally come to the most surprising case : frustrated models in presence of an external field . as shown in figure [ sgfield ] , once more the relative level of accuracy of the 4 approximations tested is the same , but the is a major difference with respect to case of zero field ( which has been reported in the upper left panel of figure [ sgfield ] for reader convenience ) . at high enough @xmath56 values , both tap and ba cease to have a solution to eqs.([eq : tap ] ) and ( [ eq : ba ] ) or equivalently the expressions under the square roots in eqs.([tap ] ) and ( [ ba ] ) become negative . to my knowledge
, this fact has been never noticed in the past , although the tap approximation for inferring couplings is largely used .
in the appendix i sketch the analytical solution to the simplest model showing this phenomenon , namely a system of 3 spins connected by antiferromagnetic couplings in presence of an external field : such an analytical solution should convince the reader that the phenomenon is not due to any numerical inaccuracy related to the complexity of the models studied here , but it can be mathematically proved in a very simple model . the absence of solutions in tap and ba is evident in figure [ sgfield ] where the corresponding curves are interrupted at a @xmath56 value that becomes smaller for larger fields . beyond the point where ba stops providing inferred couplings , one should resort to other inference methods .
unfortunately at that point both sm and ip already give quite large errors , that keep growing fast .
so in practice none of the methods studied in the present work is valid for inferring couplings in a frustrated model in presence of an external field at low enough temperatures .
the purpose of the present work is to make a detailed comparison among several approximation for solving the inverse ising problem , i.e. , estimating coupling and fields form magnetizations and correlations .
after having explained how to derive the mean - field approximations based on the plefka expansion ( naive mean - field , tap , bethe approximations , etc ... ) , i have ranked these approximations on the basis of how good they are in solving the direct problem ( i.e. , computing the correlations given the couplings ) .
tap and bethe turned out to be in general the best approximations available .
secondly i have derived the new analytical formula ( [ ba ] ) for inferring couplings from magnetizations and correlations under the bethe approximation .
this formula allow to infer couplings without running the susceptibility propagation algorithm , thus avoiding all the serious problems related to the lack of convergence of such an algorithm .
after having summarized the formulas giving the inferred couplings for the 4 approximations tested ( independent - pair , tap , bethe and the small correlation expansion of ref . ) , i have introduced a trick that , normalizing the correlation matrix , improves the tap and the bethe approximations in case of models being unfrustrated or weakly frustrated .
finally i have presented the results of the comparison among the 4 approximations for inferring couplings in diluted ferromagnetic models and spin glass models .
i have used several different topologies : fully connected graphs , regular random graphs , 3d cubic lattices and 2d square lattices . at the beginning of section [ sec : numeric ] a list of general statements about the performances of these approximations in solving the inverse ising problem
the bottom - line suggestion is to use the bethe approximation , eq.([ba ] ) , eventually with the normalization trick if the model is weakly frustrated or unfrustrated . in case of frustrated models with an external field (
that is with non - zero magnetizations ) i have found an important limitation for the tap and bethe approximations : at low enough temperatures these approximations stop having a solution and can not be used any more for solving the inverse ising problem .
this is a fundamental limitation , that take place also in very simple systems ( see the appendix ) and that was not noticed before .
moreover , when the bethe approximation stops inferring couplings , the other methods already have a rather large error .
so , in my opinion , it is still an open problem to find an approximation that , using only the correlation matrix , is able to solve the inverse ising problem in a frustrated model with a field at low enough temperatures .
i acknowledge interesting conversations with a. decelle , s. franz , a. pelizzola , i. perez - castillo and j. raymond , and financial support by the italian research minister through the firb project no .
rbfr086nn1 on `` inference and optimization in complex systems : from the thermodynamics of spin glasses to message passing algorithms '' . after the completion of the present manuscript i learnt of a manuscript by h. c. nguyen and j. berg @xcite showing a formula similar to eq.([ba ] ) .
in this appendix i show explicitly that the formulas derived with tap and ba for solving the inverse ising problem do not always admit a solution for the case of frustrated models . in order to simplify the computation i focus on the simplest model showing this problem , namely a system of 3 spins interacting with antiferromagnetic couplings of intensity @xmath126 , in presence of an external field of intensity @xmath127 ,
whose probability distribution is @xmath128\;.\ ] ] thanks to the symmetries in the above measure , each spin has the same local magnetization @xmath129 and each pair of spins has the same correlation @xmath130 .
when using the tap approximation for the inverse problem one has to solve the following equation for each coupling @xmath24 , @xmath131 and the above equation admit a solution only if its discriminant is non - negative : @xmath132 in the present case , the discriminant is the same for each coupling and it is a function of the two parameters @xmath134 and @xmath127 , that i report schematically in figure [ fig : discr ] .
the full curve shown in figure [ fig : discr ] corresponds to @xmath135 and has two asymptotes at @xmath136 and @xmath137 .
it is clear that for any non - zero field @xmath127 and any antiferromagnetic coupling @xmath134 the inference method based on the tap approximation will fail at sufficiently small temperatures ( i.e. , large absolute values of @xmath127 and @xmath134 ) .
the same phenomenon happens also for the inference method based on the ba . in this case
the discriminant that may become negative is @xmath138 in figure [ fig : discr ] the dashed line corresponds to @xmath139 and has two asymptotes at @xmath140 and along the line @xmath141 ( meaning that for this simple system of 3 spins the ba can work even a very low temperatures if the external field is large enough ) .
t. heskes , c. albers , and h. kappen , in proceedings of the conference on uncertainty in artificial intelligence , 2003 , edited by christopher meek and uffe kjaerulff ( morgan kaufmann , san francisco , 2003 ) , p. 313 . | the inverse ising problem consists in inferring the coupling constants of an ising model given the correlation matrix .
the fastest methods for solving this problem are based on mean - field approximations , but which one performs better in the general case is still not completely clear . in the first part of this work , i summarize the formulas for several mean - field approximations and i derive new analytical expressions for the bethe approximation , which allow to solve the inverse ising problem without running the susceptibility propagation algorithm ( thus avoiding the lack of convergence ) . in the second part ,
i compare the accuracy of different mean field approximations on several models ( diluted ferromagnets and spin glasses ) defined on random graphs and regular lattices , showing which one is in general more effective .
a simple improvement over these approximations is proposed .
also a fundamental limitation is found in using methods based on tap and bethe approximations in presence of an external field
. mean - field approximations ( mfa ) are very important tools in statistical mechanics , since they provide an approximated description of a physical system in terms of few parameters ( e.g. local magnetizations ) . among mfa
the one based on the bethe approximation ( ba ) is very effective . in recent years
the ba originally derived for the ferromagnetic model on regular lattices @xcite has been extended , under the name of cavity method , to models having arbitrary couplings and topologies @xcite .
although the ba is exact only for tree - like topologies , its application to models defined on random graphs has proved very successful : see e.g. the cases of low - density parity check codes , spin glasses and constraint satisfaction problems , all nicely reviewed in ref . .
the inverse ising problem , originally known as boltzmann machine learning , consists in inferring coupling constants of an ising model ( both pairwise interactions and external fields ) given the vector of magnetizations and the matrix of pairwise correlations . in recent years
the inverse ising problem has received a lot of attention , specially in connection to inference in biological problems @xcite .
the inverse ising problems can be viewed as the dual problem with respect to the ` direct ' problem of estimating magnetizations and correlation given the hamiltonian .
so , under any mfa , the inverse problem can be solved by inverting ( if possible ) the analytic expressions that give the magnetizations and the correlations as a function of interactions and fields .
although mfa usually do not provide directly the correlations between distant variables , these correlations can be computed by using the linear response theorem @xcite . within the ba , a very fast and efficient way to estimate correlations between any pair of variables is given by the susceptibility propagation ( suscprop ) algorithm recently introduced in ref . .
suscprop is an iterative algorithm for solving a set of self - consistency equations : when it converges is very fast , but sometimes it may not converge . indeed
the application of the ba to the inverse ising problem has been limited up to now by the range of convergence of suscprop @xcite .
in this work i present an analytical expression for the fixed point of suscprop , thus avoiding any problem related to its lack of convergence .
actually such an expression already appeared in ref . , but was unknown to the statistical mechanics community ( including the author ) : otherwise there would be no need for the suscprop algorithm introduced in ref . .
in the first part of this work i derive new analytical expressions for solving the inverse ising problem under the ba .
these analytical expressions , allow for a fair comparison among different mfa , in a wide range of temperatures , both for the problem of estimating 2-point correlations given the couplings ( direct problem ) and for the problem of estimating couplings given correlations and magnetizations ( inverse problem ) . in the present work
i consider mfa obtained from the so - called plefka expansion @xcite and from a small correlations expansion @xcite .
for the inverse ising problem i compare methods that take in input only the correlation matrix .
more complex , but usually slower , inference methods exist that require many samples of equilibrium configurations @xcite and not only the correlation matrix . improving over these mfa would be very welcome . it is well known
that mfa ignore loops , so correcting these mfa by adding the loops contributions would be the right direction to follow .
however at present all the methods which has been developed to consider explicitly the loops @xcite do not provide analytical expressions for the correlations , which are simple enough to be inverted .
for this reason , i have not considered these improved algorithms in the comparison of mfa for solving the inverse ising problem .
nonetheless , i am proposing a simple improvement of inference methods ( both for the direct and the inverse problems ) based on the idea that the loops may modify similarly self - correlations and correlations between close - by variables . |
the seniority quantum number was introduced by racah for the classification of electrons in an atomic @xmath1 configuration @xcite .
he assumed a spin - independent interaction @xmath2 between the electrons with the property @xmath3 that is , there is no interaction unless the two electrons orbital angular momenta @xmath4 are coupled to a combined angular momentum of @xmath5 .
racah was able to derive a closed formula for the interaction energy among @xmath6 electrons and to prove that any eigenstate of the interaction ( [ e_pairl1 ] ) is characterized by a ` seniority number ' @xmath7 , a quantum number additional to the total orbital angular momentum @xmath8 , the total spin @xmath9 and the number of electrons @xmath6 .
he also showed that @xmath7 corresponds to the number of electrons that are not in pairs coupled to @xmath5 @xcite .
racah s original definition of seniority made use of coefficients of fractional parentage .
he later noted that simplifications arose through the use of group theory @xcite .
seniority turned out to be a label associated with the orthogonal algebra @xmath10 in the classification @xmath11 where the dots indicate intermediate algebras , if any exist .
the number of states available to a single electron in an @xmath4 orbit is @xmath12 .
all states of the @xmath1 configuration therefore belong to the totally antisymmetric irreducible representation ( ir ) @xmath13 $ ] of @xmath14 .
furthermore , the natural scheme for electrons in an atom is @xmath15 coupling which corresponds to the reduction @xmath16 , where the orbital degrees of freedom are contained in @xmath17 and the spin degrees of freedom in @xmath18 . for any value of @xmath4
the unitary algebra @xmath17 contains the orthogonal subalgebra @xmath10 which in turn contains so(3 ) , associated with the total orbital angular momentum @xmath8 . the group - theoretical classification ( [ e_clasl ] ) allowed racah to derive a number of important results in the theory of complex atomic spectra .
the pairing force ( [ e_pairl1 ] ) , however , is a poor approximation to the coulomb interaction between electrons and for a more physically relevant application of seniority we have to turn to nuclei .
the discussion of seniority in atoms and in nuclei differs in two aspects : ( i ) @xmath15 coupling is a good first - order approximation in atoms while in nuclei it is rather @xmath19 coupling and ( ii ) electrons are identical particles while nucleons come in two kinds , neutrons and protons .
let us postpone the discussion of the second complication until sect .
[ s_isospin ] and concentrate in this section on the case of identical nucleons ( either all neutrons or all protons ) .
we impose the additional restriction that the identical nucleons are confined to a single-@xmath0 shell , deferring the discussion of the many-@xmath0 case to sect .
[ s_many - j ] .
it turns out that a pairing force of the type @xmath20 is a reasonable first - order approximation to the strong interaction between identical nucleons . in eq .
( [ e_pairj1me ] ) @xmath0 is the total ( orbital+spin ) angular momentum of a single nucleon and @xmath21 results from the coupling of two of them . since the pairing property now refers to the total @xmath0 of the nucleons , there is no need for a separate treatment of orbital and spin degrees of freedom as in eq .
( [ e_clasl ] ) , and the classification becomes in fact simpler : @xmath22 [ 1^n]&&[1^\upsilon]&&&&j \end{array}. \label{e_clasj1}\ ] ] seniority is associated with the ( unitary ) symplectic algebra @xmath23 which replaces the orthogonal algebra @xmath10 of the atomic case .
since the nucleons are identical , all states of the @xmath24 configuration belong to the totally antisymmetric ir @xmath13 $ ] of @xmath25 .
the irs of @xmath23 therefore must be totally antisymmetric of the type @xmath26 $ ] .
the allowed values of seniority are @xmath27 or 0 .
the angular momentum content for a given seniority @xmath7 can also be worked out @xcite but no simple general rule is available .
an alternative , simpler definition of seniority can be given which relies on the existence of an su(2 ) symmetry of the pairing hamiltonian @xcite . in second quantization
the pairing interaction ( [ e_pairj1me ] ) is written as @xmath28 with @xmath29 where @xmath30 creates a nucleon in the shell @xmath0 with projection @xmath31 .
the commutator of @xmath32 and @xmath33 leads to the operator @xmath34=(2\hat n_j-2j-1)/2\equiv2\hat s^j_z$ ] , which thus equals , up to a constant , the number operator @xmath35 .
since the three operators @xmath36 close under commutation , @xmath37=\pm \hat s^j_\pm$ ] and @xmath34=2\hat s^j_z$ ] , they form an su(2 ) algebra , referred to as the quasi - spin algebra .
this algebraic structure allows an analytical solution of the pairing hamiltonian . from the commutation relations it follows that @xmath38 , which shows that the pairing hamiltonian can be written as a combination of casimir operators belonging to su(2 ) and @xmath39 .
the associated eigenvalue problem can be solved instantly , yielding the energy expression @xmath40 $ ] .
the quantum numbers @xmath9 and @xmath41 can be put in relation to the seniority @xmath7 and the nucleon number @xmath6 , @xmath42 and @xmath43 , leading to the energy expression @xmath44 .
this coincides with the original expression given by racah , eq .
( 50 ) of ref .
@xcite , after the replacement of the degeneracy in @xmath15 coupling , @xmath12 , by the degeneracy in @xmath19 coupling , @xmath45 .
while this analysis shows that the eigenstates of a pairing interaction carry good seniority , it does not answer the question what are the necessary and sufficient conditions for a general interaction to conserve seniority .
let us specify a rotationally invariant two - body interaction @xmath2 by the matrix elements @xmath46 with @xmath47 .
the necessary and sufficient conditions for the conservation of seniority can then be written as @xmath48 with @xmath49 , and where the symbol between curly brackets is a racah coefficient .
these conditions have been derived previously in a variety of ways @xcite .
although ( [ e_condi ] ) determines all constraints on the matrix elements @xmath50 by varying @xmath49 , it does not tell us how many of those are independent .
this number turns out to be @xmath51 , the number of independent seniority @xmath52 states @xcite .
no condition on the matrix elements @xmath50 is obtained for @xmath53 , 5/2 and 7/2 , one condition for @xmath54 , 11/2 and 13/2 , and so on . as a result ,
identical nucleons in a single shell with @xmath55 conserve seniority for _ any _ interaction @xcite .
clearly , the conditions ( [ e_condi ] ) are much weaker than the requirement that the interaction be of pairing character but still many of the results of the quasi - spin formalism remain valid . for instance , the ground state of an even even nucleus still can be written in the form ( [ e_cond1 ] ) . the main restriction of the concept of seniority as defined so far , concerns the fact that the nucleons are confined to a single-@xmath0 shell . to lift this restriction
, we turn to the generalization presented in the next section .
the quasi - spin algebra can be generalized to the case of several degenerate shells ( which we assume to be @xmath56 in number ) by making the substitutions @xmath57 and @xmath58 .
therefore , if a semi - magic nucleus can be approximated as a system of identical nucleons interacting through a pairing force and distributed over several degenerate shells , the formulas of the quasi - spin formalism should apply .
in particular , the ground states of even even semi - magic nuclei will have a ` superfluid ' structure of the form @xmath59 where @xmath60 represents the vacuum ( _ i.e. _ , the doubly - magic core nucleus ) .
the su(2 ) quasi - spin solution of the pairing hamiltonian ( [ e_pairj1 ] ) leads to several characteristic predictions : a constant excitation energy ( independent of @xmath6 ) of the first - excited @xmath61 state in even even isotopes , the linear variation of two - nucleon separation energies as a function of @xmath6 , the odd even staggering in nuclear binding energies , the enhancement of two - nucleon transfer .
a more generally valid model is obtained if one imposes the following condition on the hamiltonian : @xmath62,\hat s_+]= \delta\left(\hat s_+\right)^2 , \label{e_gsen}\ ] ] where @xmath63 creates the lowest two - nucleon eigenstate of @xmath64 and @xmath65 is a constant .
this condition of generalized seniority , which was proposed by talmi @xcite , is much weaker than the assumption of a pairing interaction and , in particular , it does not require the commutator @xmath66 $ ] to yield ( up to a constant ) the number operator a property which is central to the quasi - spin formalism . in spite of the absence of a closed algebraic structure , it is still possible to compute the exact ground - state eigenvalue but hamiltonians satisfying ( [ e_gsen ] )
are no longer necessarily completely solvable .
an exact method to solve the problem of identical nucleons distributed over non - degenerate levels interacting through a pairing force was proposed a long time ago by richardson @xcite based on the bethe _ ansatz _ @xcite . as an illustration of richardson s
approach , we supplement the pairing interaction with a one - body term , to obtain the following hamiltonian : @xmath67 where @xmath68 are single - particle energies .
the solvability of the hamiltonian ( [ e_pairjm1 ] ) arises as a result of the symmetry @xmath69 where each su(2 ) algebra pertains to a specific @xmath0 .
whether the solution of ( [ e_pairjm1 ] ) can be called superfluid depends on the differences @xmath70 in relation to the strength @xmath71 . in all cases
the solution is known in closed form for all possible choices of @xmath68 .
it is instructive to analyze first the case of @xmath72 nucleons because it gives insight into the structure of the general problem .
the two - nucleon , @xmath73 eigenstates can be written as @xmath74 with @xmath75 coefficients that are determined from the eigenequation @xmath76 where @xmath77 is the unknown eigenenergy . with some elementary manipulations
this can be converted into the secular equation @xmath78 , with @xmath79 , from where @xmath75 can be obtained up to a normalization constant , @xmath80 .
the eigenenergy @xmath77 can be found by substituting the solution for @xmath75 into the secular equation , leading to @xmath81 this equation can be solved graphically which is done in fig .
[ f_rich2 ] for a particular choice of single - particle energies @xmath68 and degeneracies @xmath82 , appropriate for the tin isotopes with @xmath83 protons and neutrons distributed over the 5082 shell .
fermions distributed over @xmath84 single - particle orbits .
the sum @xmath85 is plotted as a function of @xmath77 ; the intersections of this curve with the line @xmath86 ( dots ) then correspond to the solutions of the richardson equation.,width=529 ] in the limit @xmath87 of weak pairing , the solutions @xmath88 are obtained , as should be .
of more interest is the limit of strong pairing , @xmath89 . from the graphical solution
we see that in this limit there is one eigenstate of the pairing hamiltonian which lies well below the other eigenstates with approximately constant amplitudes @xmath75 since for that eigenstate @xmath90 .
hence , in the limit of strong pairing one finds a @xmath73 ground state which can be approximated as @xmath91 where @xmath92 . because of this property this state is often referred to as the collective @xmath9 state , in the sense that all single - particle orbits contribute to its structure .
this result can be generalized to @xmath6 particles , albeit that the general solution is more complex .
on the basis of the two - particle problem one may propose , for an even number of particles @xmath6 , a ground state of the hamiltonian ( [ e_pairjm1 ] ) of the form ( up to a normalization constant ) @xmath93 which is known as the bethe _ ansatz _ @xcite .
each pair in the product is defined through coefficients @xmath94 in terms of an energy @xmath95 depending on @xmath96 which labels the @xmath97 pairs .
this product indeed turns out to be the ground state provided the @xmath95 are solutions of @xmath97 coupled , non - linear equations @xmath98 known as the richardson equations @xcite .
note the presence of a second term on the left - hand side with differences of the unknowns @xmath99 in the denominator , which is absent in the two - particle case .
in addition , the energy of the state ( [ e_gsrich ] ) is given by @xmath100 .
a characteristic feature of the bethe _ ansatz _ is that it no longer consists of a superposition of _ identical _ pairs since the coefficients @xmath101 vary as @xmath96 runs from 1 to @xmath97 .
richardson s model thus provides a solution that covers all possible hamiltonians ( [ e_pairjm1 ] ) , ranging from those with superfluid character to those with little or no pairing correlations @xcite .
an important remaining restriction on the form of the pairing hamiltonian ( [ e_pairjm1 ] ) is that it contains a single strength parameter @xmath71 whereas , in general , the interaction might depend on @xmath0 and @xmath102 , leading to @xmath103 strengths @xmath104 . in nuclei , often the assumption of a separable interaction is made which , in the case of pairing , leads to strengths @xmath105 in terms of @xmath56 parameters @xmath106 .
this restriction leads to the following pairing hamiltonian : @xmath107 as yet , no closed solution of the general hamiltonian ( [ e_pairjm2 ] ) is known but three solvable cases have been worked out : 1 .
the strengths @xmath106 are constant ( independent of @xmath0 ) .
this case was discussed above .
the single - particle energies @xmath68 are constant ( independent of @xmath0 ) .
the solution was given by pan _
et al . _
@xcite 3 .
there are two levels .
the solution was given by balantekin and pehlivan @xcite .
about ten years after its introduction by racah , seniority was adopted in nuclear physics for the @xmath19-coupling classification of nucleons in a single-@xmath0 shell @xcite .
the main additional difficulty in nuclei is that one deals with a system of neutrons and protons , and hence the isospin @xmath108 of the nucleons should be taken into account .
the generalization of the classification ( [ e_clasj1 ] ) for identical nucleons toward neutrons and protons reads as follows : @xmath109 [ 1^n]&&[h]&&[\sigma]&&&&j&&t \end{array } , \label{e_clasj2}\ ] ] where @xmath110 $ ] and @xmath111 $ ] are young tableaux associated with @xmath25 and @xmath23 . in general , @xmath45 labels are needed to characterize an ir of @xmath25 , @xmath110=[h_1,h_2,\dots , h_{2j+1}]$ ] , and @xmath112 labels are needed for an ir of @xmath23 , @xmath111=[\sigma_1,\sigma_2,\dots,\sigma_{j+1/2}]$ ] . to ensure overall antisymmetry under @xmath113 , the young tableaux of @xmath25 and @xmath114
must be conjugate , that is , one is obtained from the other by interchanging rows and columns . since the young tableau associated with @xmath114
is determined by the nucleon number @xmath6 and the total isospin @xmath108 as @xmath115 $ ] , the young tableau of @xmath25 must therefore be @xmath116=[\overbrace{2,2,\dots,2}^{n/2-t},\overbrace{1,1,\dots,1}^{2t}].\ ] ] since an ir of @xmath25 has at most @xmath45 labels , it follows that @xmath117 .
furthermore , all non - zero labels in @xmath111 $ ] must be either 2 or 1 and the young tableau of @xmath23 must therefore be of the form @xmath118=[\overbrace{2,2,\dots,2}^{\upsilon/2-t},\overbrace{1,1,\dots,1}^{2t}].\ ] ] the ir of @xmath23 is thus characterized by _ two _ labels : the seniority @xmath7 and the ` reduced isospin ' @xmath119 .
the former has the same interpretation as in the like - nucleon case while the latter corresponds to the isospin of the nucleons which are not in pairs coupled to @xmath73 .
the group - theoretical analysis is considerably more complex here than in the case of identical nucleons and , in addition , for each value of @xmath0 one is faced with a different reduction problem associated with @xmath120 .
it is therefore advantageous to go over to a quasi - spin formulation of the problem and , as was shown by helmers @xcite , this is possible for whatever value of the intrinsic quantum number of the particles ( which is @xmath121 for nucleons ) . if the pairing interaction is assumed to be isospin invariant , it is the same in the three @xmath122 channels , neutron neutron , neutron proton and proton proton , and eq .
( [ e_pairj1 ] ) can be generalized to @xmath123 where the dot indicates a scalar product in isospin . in terms of the nucleon creation operators @xmath124 , which now carry also isospin indices ( with @xmath121 ) ,
the pair operators are @xmath125 where the coupling refers to angular momentum and to isospin .
the index @xmath126 ( isospin projection ) distinguishes neutron neutron ( @xmath127 ) , neutron
proton ( @xmath128 ) and proton proton ( @xmath129 ) pairs .
there are thus three different pairs with @xmath73 and @xmath122 and they are related through the action of the isospin raising and lowering operators @xmath130 . by considering the commutation relations between the different operators , a closed algebraic structure is obtained , generated by the pair operators @xmath131 , the number operator @xmath132 and the isospin operators @xmath130 and @xmath133 .
the quasi - spin algebra of neutrons and protons in degenerate @xmath0 shells turns out to be so(5 ) , by virtue of which the hamiltonian ( [ e_pairj2 ] ) is analytically solvable @xcite .
a further generalization is possible in @xmath15 coupling . for a neutron and a proton there
exists a different paired state with _
parallel _ spins .
the most general pairing interaction for a system of neutrons and protons is therefore of the form @xmath134 where the pair operators are defined as @xmath135 where @xmath136 creates a nucleon in the shell @xmath4 with projection @xmath137 , spin projection @xmath138 and isospin projection @xmath139 .
the hamiltonian ( [ e_pairl2 ] ) contains two parameters @xmath71 and @xmath140 , the strengths of the isovector and isoscalar components of the pairing interaction .
while in the previous case the single strength parameter @xmath71 just defines an overall scale , this is no longer true for a generalized pairing interaction and different solutions are obtained for different ratios @xmath141 .
in general , the eigenproblem associated with the interaction ( [ e_pairl2 ] ) can only be solved numerically ; for specific choices of @xmath71 and @xmath140 the solution of @xmath142 can be obtained analytically @xcite .
a closed algebraic structure is obtained , formed by the pair operators ( [ e_pair3 ] ) , their commutators , the commutators of these among themselves , and so on until closure is attained .
the quasi - spin algebra in this case turns out to be so(8 ) , with 28 generators , consisting of the pair operators @xmath131 and @xmath143 , the number operator @xmath132 , the spin and isospin operators @xmath144 and @xmath145 , and the gamow teller - like operator @xmath146 , which is a vector in spin and isospin .
the symmetry character of the hamiltonian ( [ e_pairl2 ] ) is obtained by studying the subalgebras of so(8 ) .
of relevance are the subalgebras @xmath147 , @xmath148 , @xmath149 , @xmath150 and @xmath151 , which can be placed in the following lattice of algebras : @xmath152 by use of the explicit form of the generators of so(8 ) and its subalgebras , and their commutation relations @xcite , the following relations can be shown to hold : @xmath153- { \frac 1 2}\hat c_2[{\rm so}_t(3)]- { \frac 1 8}(2\omega-\hat n)(2\omega-\hat n+6 ) , \nonumber\\ \hat s_+\cdot\hat s_-+ \hat p_+\cdot\hat p_-&= & { \frac 1 2}\hat c_2[{\rm so}(8)]- { \frac 1 2}\hat c_2[{\rm so}(6)]- { \frac 1 8}(2\omega-\hat n)(2\omega-\hat n+12 ) , \nonumber\\ \hat p_+\cdot\hat p_-&= & { \frac 1 2}\hat c_2[{\rm so}_s(5)]- { \frac 1 2}\hat c_2[{\rm so}_s(3)]- { \frac 1 8}(2\omega-\hat n)(2\omega-\hat n+6),\end{aligned}\ ] ] with @xmath154 and where @xmath155 $ ] is the @xmath156-order casimir operator of the algebra @xmath157 .
this shows that the interaction ( [ e_pairl2 ] ) in the three cases ( i ) @xmath158 , ( ii ) @xmath159 and ( iii ) @xmath160 , can be written as a combination of casimir operators of algebras belonging to a chain of _ nested _ algebras of the lattice ( [ e_lattice ] ) .
they are thus the dynamical symmetries of the so(8 ) model .
the nature of ` so(8 ) superfluidity ' can be illustrated in the specific example of the ground state of even
even @xmath161 nuclei . in the so(6 )
limit of the so(8 ) model the exact ground - state solution can be written as @xcite @xmath162 this shows that the superfluid solution acquires a _
structure in the sense that it reduces to a condensate of bosons each of which corresponds to four nucleons .
since the boson in ( [ e_quartet ] ) is a scalar in spin and isospin , it can be thought of as an @xmath96 particle ; its orbital character , however , might be different from that of an actual @xmath96 particle . a quartet structure is also present in the two so(5 ) limits of the so(8 ) model , which yields a ground - state wave function of the type ( [ e_quartet ] ) with either the first or the second term suppressed .
a reasonable _ ansatz _ for the @xmath161 ground - state wave function of the so(8 ) pairing interaction ( [ e_pairl2 ] ) with arbitrary strengths @xmath71 and @xmath140 is therefore @xmath163 where @xmath164 is a parameter that depends on the ratio @xmath141 .
the condensate ( [ e_trial ] ) of @xmath96-like particles provides an excellent approximation to the @xmath161 ground state of the pairing hamiltonian ( [ e_pairl2 ] ) for any combination of @xmath71 and @xmath140 @xcite .
it should nevertheless be stressed that , in the presence of both neutrons and protons in the valence shell , the pairing hamiltonian ( [ e_pairl2 ] ) is _ not _ a good approximation to a realistic shell - model hamiltonian which contains an important quadrupole component .
these results can be generalized to the case of several non - degenerate shells .
in fact , the richardson equations ( [ e_richeqs ] ) are valid for the quasi - spin symmetry su(2 ) but they are known for any lie algebra @xcite .
closed solutions have been obtained for a system of neutron and protons with a pairing interaction of pure isovector character and of equal isovector and isoscalar strength , based on the so(5 ) and the so(6 ) quasi - spin algebras , respectively @xcite .
in this section the concept of seniority is illustrated with an application to the physics of cold atoms .
if atoms in a bose einstein condensate ( bec ) are trapped by optical means @xcite , their hyperfine spins ( or spins ) are not frozen in one particular direction but are essentially free but for their mutual interactions . as a result
, the atoms do not behave as scalar particles but each of the components of the spin is involved in the formation of the bec .
this raises interesting questions concerning the structure of the condensate and how it depends on the spin - exchange interactions between the atoms .
such questions were addressed in a series of theoretical papers by ho and co - workers @xcite who obtained solutions based on a generating function method . in the case of spin-1 atoms
the problem of quantum spin mixing was analyzed by law _
et al . _
@xcite who proposed an elegant solution based on algebraic methods .
it is shown here that an exact solution is also available for the spin value @xmath165 ( for any number of atoms @xmath6 ) which allows the analytic determination of the structure of the ground state of the condensate .
this was simultaneously and independently pointed out in refs .
@xcite .
we consider a one - component dilute gas of trapped bosonic atoms with arbitrary ( integer ) hyperfine spin @xmath166 .
in second quantization the hamiltonian of this system has a one - body and a two - body piece that can be written as ( in the notation of ref .
@xcite ) @xmath167 where @xmath168 , @xmath169 is the mass of the atom , and @xmath170 and @xmath171 are the atomic field annihilation and creation operators associated with atoms in the hyperfine state @xmath172 with @xmath173 , the possible values of all summation indices in ( [ e_hambec ] ) .
the trapping potential @xmath174 is assumed to be the same for all @xmath175 components . according to the assumptions outlined in ref .
@xcite , the atomic field creation and annihilation operators at zero temperature can be approximated by @xmath176 , @xmath177 , @xmath173 , where @xmath178 is a single wave function ( independent of @xmath179 ) and @xmath180 and @xmath181 are annihilation and creation operators , satisfying the usual boson commutation rules . in this approximation
the entire hamiltonian ( [ e_hambec ] ) can be rewritten as @xmath182 where the coefficients @xmath183 and @xmath184 are related to those in the original hamiltonian ( [ e_hambec ] ) and with @xmath185 .
exactly solvable hamiltonians with rotational or so(3 ) invariance are now found by the determination of all lie algebras @xmath157 satisfying @xmath186 .
the canonical reduction of u(@xmath175 ) is of the form as encountered by racah ( see sect .
[ s_birth ] ) , @xmath187 defining a class of solvable hamiltonians of the type @xmath188 + a_2\hat c_2[{\rm u}(2f+1)]+ b\,\hat c_2[{\rm so}(2f+1)]+ c\,\hat c_2[{\rm so}(3 ) ] , \label{e_hamsol}\ ] ] where @xmath189 , @xmath190 , @xmath191 , and @xmath192 are numerical coefficients .
the solvability properties of the original hamiltonian ( [ e_hamgen ] ) now follow from a simple counting argument . for atoms with spin
@xmath193 the solvable hamiltonian ( [ e_hamsol ] ) has three coefficients @xmath189 , @xmath190 , and @xmath192 [ since so(@xmath175)=so(3 ) ] while the general hamiltonian ( [ e_hamgen ] ) also contains three coefficients @xmath183 , @xmath194 , and @xmath195 .
( note that the coupling of two spins to odd @xmath196 is not allowed in the approximation of a common spatial wave function , so no @xmath197 term occurs . )
they can be put into one - to - one correspondence . for atoms with spin @xmath165
both the solvable and the general hamiltonian contain four coefficients ( @xmath189 , @xmath190 , @xmath191 , and @xmath192 _ versus _
@xmath183 , @xmath194 , @xmath195 , and @xmath198 ) which also can be put into one - to - one correspondence .
hence the general hamiltonian ( [ e_hamgen ] ) is solvable for @xmath199 .
the same counting argument shows that it is no longer solvable for @xmath200 .
the case of interacting @xmath193 atoms was discussed by law _
et al . _
@xcite who identified the existence of two possible condensate ground states : one with all atoms aligned to maximum spin @xmath201 and a second with pairs of atoms coupled to @xmath202 . whether the condensate is aligned or paired
depends on a single interaction parameter . with the technique explained above , the phase diagram for atoms with spin @xmath165 can also be derived .
the results are exact and valid for arbitrary @xmath6 .
the entire spectrum is determined by the eigenvalue expression together with the necessary branching rules .
in particular , the allowed values of total spin @xmath196 for a given seniority @xmath7 are derived from the @xmath203 branching rule given by @xmath204 with @xmath205 and @xmath206 .
it is now possible to determine all possible ground - state configurations of the condensate and their quantum numbers @xmath207 and @xmath208 @xcite .
the character of the ground state does not depend on the coefficients @xmath209 since the first two terms in the expression ( [ e_hamsol ] ) give a constant contribution to the energy of all states .
although this contribution is dominant , the spectrum - generating perturbation of the hamiltonian is confined to the last two terms and depends solely on the coefficients @xmath191 and @xmath192 which are related to the original interactions @xmath184 according to @xmath210 and @xmath211 .
the phase diagram displays a richer structure than in the @xmath193 case .
there is an aligned phase where the seniority is maximal , @xmath212 , and all spins are aligned , @xmath213 .
secondly , there is a low - seniority ( paired ) and consequently low - spin phase . for even @xmath6 ,
this corresponds to @xmath214 .
the aligned and paired phases are also encountered for interacting @xmath193 atoms . for @xmath165 a third phase occurs characterized by high seniority ( _ i.e. _ , unpaired ) and low total spin , @xmath215 with @xmath216 or 1 .
since the hamiltonian ( [ e_hamsol ] ) is solvable for @xmath165 , all eigenstates , and in particular the three different ground states , can be determined analytically .
the general expressions given by chacn _
et al . _
@xcite reduce to @xmath217 where the @xmath165 atoms are denoted as @xmath218 bosons . in the second of these expressions
it is assumed that @xmath6 is even and in the third that @xmath219 ; other cases are obtained by adding a single boson @xmath220 or a @xmath221 pair .
the @xmath222 are the so - called traceless boson operators @xcite which are defined as @xmath223 ( see also chapt . 8 of ref .
the wave functions ( [ e_wave ] ) are the _ exact _ finite-@xmath6 expressions for the eigenstates of the hamiltonian ( [ e_hamsol ] ) .
since in the large-@xmath6 limit the traceless boson operators @xmath224 become identical to @xmath225 , one arrives at a simple interpretation of the three types of configurations : ( i ) spin - aligned , ( ii ) condensed into _
pairs of atoms _ coupled to @xmath202 , and ( iii ) condensed into _ triplets of atoms _ coupled to @xmath202 . in conclusion ,
the consideration of seniority is crucial in obtaining results concerning bose
einstein condensates consisting of atoms with spin . since all eigenstates of interacting atoms with spin @xmath199
are known analytically , this opens up the possibility to study the relaxation properties of such condensates using their exact , macroscopic wave functions .
in addition , preliminary studies indicate that seniority can be exploited even when @xmath200 .
these problems are currently under investigation @xcite .
this paper is dedicated to the memory of marcos moshinsky .
the two years i have spent in mexico as a visitor and the many hours with marcos as a teacher , were crucial to my formation as a physicist . without him i never could have written this paper . | the use of the seniority quantum number in many - body systems is reviewed .
a brief summary is given of its introduction by racah in the context of atomic spectroscopy .
several extensions of racah s original idea are discussed : seniority for identical nucleons in a single-@xmath0 shell , its extension to the case of many , non - degenerate @xmath0 shells and to systems with neutrons and protons . to illustrate its usefulness to this day , a recent application of seniority is presented in bose
einstein condensates of atoms with spin . |
let @xmath0 be a simple graph on @xmath4 vertices .
the eigenvalues of @xmath0 are the eigenvalues of its adjacency matrix , which are denoted by @xmath5 .
the estrada index of @xmath0 , put forward by estrada @xcite , is defined as @xmath6 the estrada index has multiple applications in a large variety of problems , for example , it has been successfully employed to quantify the degree of folding of long - chain molecules , especially proteins @xcite , and it is a useful tool to measure the centrality of complex ( reaction , metabolic , communication , social , etc . ) networks @xcite .
there is also a connection between the estrada index and the extended atomic branching of molecules @xcite . besides these applications
, the estrada index has also been extensively studied in mathematics , see @xcite .
ili and stevanovi @xcite obtained the unique tree with minimum estrada index among the set of trees with a given maximum degree .
zhang , zhou and li @xcite determined the unique tree with maximum estrada indices among the set of trees with a given matching number . in @xcite ,
du and zhou characterized the unique unicyclic graph with maximum estrada index .
wang et al .
@xcite determined the unique graph with maximum estrada index among bicyclic graphs with fixed order , and zhu et al .
@xcite determined the unique graph with maximum estrada index among tricyclic graphs with fixed order .
more mathematical properties on the estrada index can be founded in @xcite .
a graph is bipartite if its vertex set can be partitioned into two subsets @xmath7 and @xmath8 so that every edge has one end in @xmath7 and the other end in @xmath8 .
we denote a bipartite graph g with bipartition @xmath9 by @xmath10 $ ] .
if @xmath10 $ ] is simple and every vertex in @xmath7 is joined to every vertex in @xmath8 , then @xmath0 is called a complete bipartite graph . up to isomorphism
, there is a unique complete bipartite graph with parts of sizes @xmath11 and @xmath4 , denoted @xmath12 . for an edge subset @xmath13 of the complement of @xmath0
, we use @xmath14 to denote the graph obtained from @xmath0 by adding the edges in @xmath13 .
a matching in a graph is a set of pairwise nonadjacent edges .
if @xmath15 is a matching , the two ends of each edge of @xmath15 are said to be matched under @xmath15 , and each vertex incident with an edge of @xmath15 is said to be covered by @xmath15 .
a maximum matching is one which covers as many vertices as possible .
the number of edges in a maximum matching of a graph @xmath0 is called the matching number of @xmath0 and denoted by @xmath16 .
let @xmath17 be the set of bipartite graphs on @xmath4 vertices with @xmath18 .
a cut vertex(edge ) of a graph is a vertex(edge ) whose removal increases the number of components of the graph .
a(an ) vertex(edge ) cut of a graph is a set of vertices(edges ) whose removal disconnects the graph .
the connectivity(edge - connectivity ) of a graph @xmath0 is defined as @xmath19 let @xmath20(@xmath21 ) denote the set of bipartite graphs on @xmath4 vertices with @xmath22(@xmath23 ) . for other undefined terminology and notation we refer to bondy and murty @xcite . in @xcite , du , zhou and xing
determined the graphs with maximum estrada indices among graphs with given number of cut vertices , connectivity , and edge connectivity , respectively . in this paper
, we consider bipartite graphs , and characterize the unique bipartite graph with maximum estrada indices among @xmath17 , @xmath20 and @xmath21 , respectively .
denote by @xmath24 the @xmath25-th spectral moment of a graph @xmath0 , i.e. , @xmath26 .
it is well - known @xcite that @xmath24 is equal to the number of closed walks of length @xmath25 in @xmath0 .
then @xmath27 for @xmath4-vertex graphs @xmath28 and @xmath29 , if @xmath30 for all positive integers @xmath25 , then by eq.(1 ) we have that @xmath31 with equality if and only if @xmath32 for all positive integers @xmath25 .
let @xmath25 be a positive integer .
for @xmath33 , let @xmath34 denote the set of @xmath35-walks of length @xmath25 in @xmath0 , and let @xmath36 .
for convenience , let @xmath37 and @xmath38 . for graphs @xmath28 and @xmath29 with @xmath39 and @xmath40 , if @xmath41 for all positive integers @xmath25 , then we write@xmath42 , and if @xmath43 and there is a positive integer @xmath44 such that @xmath45 , then we write @xmath46 . for convenience ,
we write @xmath47 for @xmath48 , and @xmath49 for @xmath50 .
[ lem:2.1 ] @xcite let @xmath0 be a graph .
then for any edge @xmath51 , one has @xmath52
@xcite if a graph @xmath0 is bipartite , and if @xmath53 is the nullity ( = the multiplicity of its eigenvalue zero ) of @xmath0 , then @xmath54 where cosh stands for the hyperbolic cosine @xmath55 $ ] , whereas @xmath56 denotes summation over all positive eigenvalues of the corresponding graph .
as is well known @xcite that the spectrum of a complete bipartite graph @xmath57 is @xmath58 @xmath59 @xmath60(@xmath61 times ) . by the definition , we have [ lem:2.2 ] @xcite @xmath62 by the monotonicity of @xmath63 , it is obvious that [ coro:2.3 ] @xmath64 [ lem:2.4 ] let @xmath0 be a non - trivial graph with @xmath33 such that @xmath65 . then for any @xmath66 , one has @xmath67 for any walk @xmath68 ,
let @xmath69 be the walk obtained from @xmath70 by replacing its first and last vertex @xmath71 by @xmath72 .
this is practical since @xmath65 .
obviously , @xmath73 and @xmath74 is a bijection from @xmath75 to @xmath76 , and so @xmath77 .
we can similarly construct a bijection from @xmath75 to @xmath34 or @xmath78 .
so we have @xmath79 as desired .
[ lem:2.5 ] let @xmath57 be the complete bipartite graph with @xmath80 and @xmath81 .
for any @xmath82 , one has that for any @xmath83 and @xmath84 , @xmath85 + furthermore , @xmath86 .
let @xmath87 be an @xmath88-walk of length @xmath89 . since @xmath0 is a complete bipartite graph
, it is straightforward that @xmath90 and @xmath91 for @xmath92 .
moreover , we know that each @xmath93 can be arbitrarily chosen from @xmath7 and each @xmath94 can be arbitrarily chosen from @xmath8 .
hence , for fixed @xmath95 and @xmath96 there are @xmath97 walks of length @xmath89 between them , that is , @xmath98 for any @xmath83 .
similarly , we can obtain @xmath99 for any @xmath100 . by the definition of @xmath101
, we have @xmath102 the proof is complete .
let @xmath103 be an independent set of @xmath28 and @xmath104 an independent set of @xmath29 .
we denote @xmath105 as the graph obtained from @xmath28 and @xmath29 by identifying @xmath106 with @xmath107 for each @xmath108 ( @xmath109 ) .
we denote the identified vertex set in @xmath105 by @xmath110 .
likewise , we can also get @xmath111 from @xmath112 and @xmath113 , where the two independent sets that should be identified are @xmath114 and @xmath115 , respectively .
[ lem:2.6 ] let @xmath116 and @xmath117 be the graphs of order @xmath4 defined as above satisfying the following conditions : 1 . for any @xmath82 ,
@xmath118 2 .
for any @xmath119 @xmath120 then for any @xmath82 , @xmath121 .
furthermore , @xmath122 , with equality holds if and only if all the equalities in ( [ 0.1 ] ) and ( [ 0.2 ] ) hold . for any @xmath82 ,
let @xmath123 denote the set of closed walks of length @xmath25 in @xmath0 , we can see that @xmath124 where @xmath125 is the set of closed walks of length @xmath25 in @xmath0 containing both vertices in @xmath126 and vertices in @xmath127 . similarly , one has @xmath128 where @xmath129 is the set of closed walks of length @xmath25 in @xmath130 containing both vertices in @xmath131 and vertices in @xmath132 . by ( [ 0.1 ] )
, we know that @xmath133 and @xmath134 . we only need to show that @xmath135 .
in fact , there exists an injection from @xmath125 to @xmath129 . in the following
, we will construct such an injection . for any @xmath136 , by ( [ 0.2 ] ) we know that for any @xmath137
, @xmath138 so there exist an injection @xmath139 from @xmath140 to @xmath141 , and an injection @xmath142 from @xmath143 to @xmath144 for any @xmath145 and any @xmath146 , we will omit the subscript of @xmath139 and @xmath142 if there is no confusion about the first and last vertices of the walks we considered
. for any @xmath147 , we call a maximal @xmath28 walk of @xmath70 a 1-block , and a maximal @xmath29 walk of @xmath70 a 2-block . from the definition
, we have that the ends of a 1-block and a 2-block are both contained in @xmath110 .
since @xmath125 is the set of closed walks of length @xmath25 in @xmath28 and contains both vertices in @xmath148 and vertices in @xmath127 , there exist at least one 1-block and one 2-block , and the 1-blocks and 2-blocks appear one by one alternately with equal number .
hence we can decompose @xmath70 as follows : @xmath149 where @xmath150 is a 1-block of length @xmath151 , and @xmath152 is a 2-block of length @xmath153 .
we define a map @xmath154 from @xmath155 to @xmath156 as follows : @xmath157 then @xmath158 is a closed walk in @xmath156 . since both @xmath139 and @xmath142 are injection , we can easily deduce that @xmath154 is an injection .
thus , @xmath159 and any @xmath146 , @xmath139 and @xmath142 are bijections , that is , all the qualities in ( [ 0.2 ] ) hold .
hence , we have @xmath160 therefore , the result follows .
a covering of a graph @xmath0 is a vertex subset @xmath161 such that each edge of @xmath0 has at least one end in the set @xmath162 .
the number of vertices in a minimum covering of a graph @xmath0 is called the covering number of @xmath0 and denoted by @xmath163 . _ ( the knig - egervry theorem , @xcite ) . _ in any bipartite graph , the number of edges in a maximum matching is equal to the number of vertices in a minimum covering .
let @xmath164 $ ] be a bipartite graph such that @xmath165 . from lemma 3.1 , we know that @xmath166 .
let @xmath110 be a minimum covering of @xmath0 and @xmath167 , @xmath168 .
without loss of generality , suppose that @xmath169 in the following analysis .
set @xmath170 , @xmath171 .
we have that @xmath172 since @xmath110 is a covering of @xmath0 .
let @xmath173 $ ] be a bipartite graph with the same vertex set as @xmath0 such that @xmath174 .
obviously , @xmath0 is a subgraph of @xmath175 . from lemma 2.1
, we know that @xmath176 with equality holds if and only if @xmath177 .
let @xmath178 then we have the following conclusion : [ 1 ] + figure 1 .
@xmath175 and @xmath179 let @xmath175 and @xmath179 be the graph defined above ( see figure 1 )
. then one has @xmath180 with equality holds if and only if @xmath181 .
let @xmath182 $ ] , @xmath183 $ ] , and @xmath184 $ ] , @xmath185 $ ] .
we can see that @xmath186 , @xmath187 and @xmath188 .
furthermore , @xmath189 , and @xmath190 with @xmath191 . by lemma [ lem:2.5 ] , we have @xmath192 since @xmath169 , we have @xmath193 .
furthermore , as both @xmath29 and @xmath194 are bipartite graphs , one has @xmath195 for any @xmath82 .
now condition 1 of lemma [ lem:2.6 ] is satisfied . for any @xmath196 ,
by lemma [ lem:2.5 ] we know that for any @xmath137 , @xmath197 as @xmath169 , we have @xmath198 with equality holds if and only if @xmath199 .
together with @xmath200 @xmath201 @xmath202 , condition 2 of lemma [ lem:2.6 ] is satisfied .
so we have @xmath203 , with equality holds if and only if @xmath199 , i.e. , @xmath181 . by ( [ 1.1 ] ) and ( [ 1.2 ] ) ,
together with corollary [ coro:2.3 ] , it is straightforward to see that among the graphs in @xmath17 , @xmath204 is the unique graph with maximum estrada index .
for two complete bipartite graphs @xmath57 and @xmath205 , we define a graph @xmath206 , where @xmath207 is the union of two graphs , @xmath208 is an empty graph of order @xmath209 and @xmath210 is a graph operation that joins all the vertices in @xmath211 to the vertices belonging to the partitions of cardinality @xmath212 in @xmath57 and @xmath213 in @xmath205 ( see figure .
2 ) , respectively . let @xmath238 be an eigenvector of @xmath239 corresponding to the eigenvalue @xmath240 . by the eigenvalue - equations , for any @xmath241 , @xmath242
thus , for any eigenvalue of @xmath215 with @xmath243 , one has @xmath244 if @xmath245 .
so , we know that the eigenvalue of @xmath215 which is not equal to 0 satisfies : @xmath246 as the root of ( [ a ] ) is also the root of @xmath247 then we have that @xmath248 where @xmath249 are the different positive roots of ( [ 5 ] ) .
we may assume that @xmath250 and @xmath251 .
then @xmath252 , and we can get @xmath253 then we have @xmath254 and @xmath255 let @xmath256 . as @xmath257
we have @xmath258 . on the other hand , let @xmath259 and @xmath260 be the maximum root of @xmath261
, we will show @xmath262 .
in fact , as @xmath263 we have @xmath264 .
together with @xmath265 , we can get @xmath262 .
thus , by ( [ 7 ] ) and ( [ 8 ] ) we have @xmath266 , i.e. , @xmath267 . by lemma [ lem:2.2 ] , we have @xmath270 . as in the proof of lemma [ lem:4.2 ] , one has @xmath271 where @xmath272 and @xmath273 are the different positive roots of @xmath274 , where @xmath275 without loss of generality , we assume that @xmath276
. then we have @xmath277 where the @xmath278 holds since @xmath268 , i.e. , @xmath279 .
together with @xmath280 , we have @xmath281 . [ lem:4.3 ]
let @xmath0 be a graph with maximum estrada index in @xmath20 and @xmath287 be a minimum vertex cut .
if @xmath288 has a nontrivial component @xmath28 , then @xmath288 has exactly two components , and the other component which is distinct from @xmath28 can not be nontrivial .
let @xmath289 be the components of @xmath288 .
suppose @xmath290 .
then , we can add some appropriate edges in @xmath0 between @xmath291 so that the resulting graph @xmath130 is still bipartite .
it is obvious that @xmath292 . by lemma [ lem:2.1 ]
, we have @xmath293 .
this contradicts the fact that @xmath0 has the maximum estrada index among graphs in @xmath20 , and so we have @xmath294 . if both @xmath28 and @xmath29 are nontrivial with bipartition @xmath295 and @xmath296 , respectively .
let @xmath297 be the bipartition of @xmath287 induced by the bipartition of @xmath0 . now joining all possible edges between the vertices of @xmath13 and @xmath298 , @xmath299 and @xmath300 , and @xmath301 and @xmath302
, we get a graph @xmath303 in @xmath20 such that @xmath304 . therefore , we assume @xmath305 ; see figure 3 .
suppose that @xmath28 and @xmath29 are the two nontrivial components of @xmath288 .
the bipartition of @xmath28 is @xmath306 and the bipartition of @xmath29 is @xmath296 .
let @xmath297 be the bipartition of @xmath287 induced by the bipartition of @xmath0 . now joining all possible edges between the vertices of @xmath13 and @xmath298 , @xmath299 and @xmath300 , and @xmath301 and @xmath302
, we get a graph @xmath303 in @xmath20 such that @xmath304 .
therefore , we assume @xmath305 ; see figure 3 . if there exists some vertex @xmath307 in @xmath288 such that @xmath308 , then forming a complete bipartite graph within the vertices of @xmath309 we would get a graph in @xmath20 with larger estrada index .
thus , we may assume that each vertex in @xmath288 has a degree greater than @xmath209 .
let @xmath310 of edges joining @xmath311 and the vertices of @xmath301 .
note that @xmath313 is the vertex cut of order @xmath209 , hence @xmath314 . without loss of generality
, we may assume that @xmath315 and note that @xmath316 , hence @xmath317 .
we now choose a subset @xmath318 of @xmath300 such that @xmath319 .
let @xmath320 it is routine to check that @xmath321 with bipartition @xmath322 .
we claim that @xmath323 we color the edges @xmath324 blue if @xmath325 and red if @xmath326 . for any @xmath82 and @xmath327 , let @xmath328 be the closed walk of length @xmath89 that is obtained by changing all the blue edges @xmath324 to @xmath329 and all the red edges @xmath330 which is incident with a red edge to @xmath331 , where @xmath332 is a vertex of @xmath13 .
it is obvious that @xmath333 and @xmath334 is an injection .
hence we have for any @xmath335 , @xmath336 . together with @xmath337 , we get @xmath338 . furthermore , as @xmath339 , we know that @xmath340 .
so , we get our conclusion .
all the components of @xmath288 are singletons . in this case
, we have @xmath343 . for @xmath344 ,
it is nothing to say since @xmath345 . for @xmath346 , by lemma [ lem:4.2 ] , @xmath347 , which contradicts the maximality of @xmath0 .
one component of @xmath288 , say @xmath28 , contains at least two vertices . by lemma [ lem:4.3 ] , we know that @xmath288 has exactly two components @xmath28 and @xmath29 , with @xmath348 . therefore , there exist @xmath349 with @xmath350 , such that @xmath351
. then by lemma [ lem:4.1 ] and lemma [ lem:4.2 ] , we have @xmath352 .
hence we have @xmath353 , i.e. , @xmath354 .
this completes the proof . as is well known @xcite that @xmath356 , hence for any @xmath357 there exists @xmath358 such that @xmath359 .
then by theorem [ thm:4.3 ] , we have @xmath360 . on the other hand , @xmath361 and @xmath362 as @xmath358 , and hence we get our conclusion . | the estrada index of a graph @xmath0 is defined as @xmath1 , where @xmath2 @xmath3 are the eigenvalues of the adjacency matrix of @xmath0 . in this paper
, we characterize the unique bipartite graph with maximum estrada index among bipartite graphs with given matching number and given vertex - connectivity , edge - connectivity , respectively .
+ * keywords : * estrada index ; walk ; maximum matching ; vertex cut ; connectivity + * ams subject classification : * 05c50 , 15a18 , 05c35 , 05c90 , 92e10 [ section ] [ theorem]lemma [ theorem]corollary [ theorem]theorem |
hck was supported in part by the korea science and engineering foundation ( kosef ) grant funded by the korea government ( mest ) ( no.2010 - 0011308 ) .
from eq . , the time derivative of the scalar field is given by @xmath112 where we have used eq . to get @xmath113 now
, we obtain the relation between the two times @xmath1 and @xmath114 by @xmath115 note that the arrow of time of the two time is dependent on the sign of @xmath116 . because the exponent of @xmath117 is always larger than @xmath118
, we see that the cosmological time go to infinity at @xmath6 . at @xmath119 ,
the exponent of @xmath120 is always smaller than one for positive @xmath83 .
therefore , the time will take a finite value , which we choose to zero . at @xmath121 ,
the exponent of @xmath122 is equal to or larger than one if @xmath8 .
therefore , @xmath123 for @xmath51 and @xmath114 takes a finite value for @xmath45 , which we choose to zero .
explicitly , eq . can be integrated in a closed form , @xmath124 , \label{t : l}\end{aligned}\ ] ] where @xmath125 will be used to set the initial time to be zero .
however , it is too complex to use directly .
we may simply note that the arrow of the time is the same as the increase of @xmath1 for @xmath126 and is reversed for @xmath127 and use the asymptotic forms below . around @xmath128 ,
the relation becomes , @xmath129 around @xmath130 , eq
. becomes @xmath131 now we write down the evolutions in @xmath132 case . with @xmath38 limit
, the generating function becomes @xmath133 the time derivative of the scalar field and @xmath1 with respect to the cosmological time is given by @xmath134 for @xmath128 , the second equation of eq .
becomes @xmath135 for @xmath130 , the relation with the cosmological time becomes @xmath136 r. r. caldwell , r. dave and p. j. steinhardt , astrophys .
space sci . *
261 * , 303 ( 1998 ) .
r. r. caldwell , r. dave and p. j. steinhardt , phys .
lett . * 80 * , 1582 ( 1998 ) [ astro - ph/9708069 ] ; n. a. bahcall , j. p. ostriker , s. perlmutter and p. j. steinhardt , science * 284 * , 1481 ( 1999 ) [ astro - ph/9906463 ] ; e. j. copeland , m. sami and s. tsujikawa , int .
j. mod .
d * 15 * , 1753 ( 2006 ) [ hep - th/0603057 ] .
p. k. townsend , `` cosmic acceleration and m theory , '' in : proceedings of icmp2003 , lisbon , hep - th/0308149 .
a. kehagias and g. kofinas , class .
* 21 * , 3871 ( 2004 ) [ gr - qc/0402059 ] .
l. p. chimento , class .
* 15 * , 965 ( 1998 ) .
j. g. russo , phys .
b * 600 * , 185 ( 2004 ) [ hep - th/0403010 ] .
p. k. townsend and m. n. r. wohlfarth , phys .
* 91 * , 061302 ( 2003 ) [ hep - th/0303097 ] .
r. emparan and j. garriga , jhep * 0305 * , 028 ( 2003 ) [ hep - th/0304124 ] .
q. shafi and c. wetterich , phys .
b * 129 * , 387 ( 1983 ) ; f. lucchin and s. matarrese , phys .
d * 32 * , 1316 ( 1985 ) ; j. d. barrow , a. b. burd , d. lancaster , class .
quantum grav .
* 3 * , 551 ( 1986 ) ; a. b. burd and j. d. barrow , nucl .
b * 308 * , 929 ( 1988 ) ; j. j. halliwell , phys .
b * 185 * , 341 ( 1987 ) .
b. ratra and p. j. e. peebles , phys . rev .
d * 37 * , 3406 ( 1988 ) ; l. p.
chimento , class .
grav . * 15 * , 965 ( 1998 ) .
l. p. chimento , a. e. cossarini and n. a. zuccala , class .
* 15 * , 57 ( 1998 ) .
e. j. copeland , a. r liddle and d. wands , phys .
d * 57 * , 4686 ( 1998 ) [ gr - qc/9711068 ] .
p. k. townsend and m. n. r. wohlfarth , phys .
* 91 * , 061302 ( 2003 ) [ hep - th/0303097 ] ; r. emparan and j. garriga , jhep * 0305 * , 028 ( 2003 ) [ hep - th/0304124 ] .
e. elizalde , s. nojiri , and s. d. odintsov , phys .
d * 70 * , 043539 ( 2004 ) .
[ arxiv : hep - th/0405034 ] ; e. elizalde , s. nojiri , s. d. odintsov , d. saez - gomez , and v. faraoni , phys .
d * 77 * , 106005 ( 2008 ) .
[ arxiv:0803.1311 [ hep - th ] ] m. a. reyes , `` on exact solutions to the scalar field equations in standard cosmology , '' arxiv:0806.2292 [ gr - qc ] .
d. bazeia , c. b. gomes , l. losano and r. menezes , phys .
b * 633 * , 415 ( 2006 ) [ astro - ph/0512197 ] .
i. y. .arefeva , a. s. koshelev and s. y. .vernov , phys .
d * 72 * , 064017 ( 2005 ) [ astro - ph/0507067 ] .
s. y. .vernov , teor .
fiz . * 155 * , 47 ( 2008 ) [ theor .
phys . * 155 * , 544 ( 2008 ) ] [ astro - ph/0612487 ] ; i. y. .arefeva , n. v. bulatov and s. y. .vernov , theor . math .
* 163 * , 788 ( 2010 ) [ arxiv:0911.5105 [ hep - th ] ] .
h. -c .
kim , arxiv:1211.0604 [ gr - qc ] .
r. de ritis , g. marmo , g. platania , c. rubano , p. scudellaro and c. stornaiolo , phys .
d * 42 * , 1091 ( 1990 ) .
d. s. salopek and j. r. bond , phys .
d * 42 * ( 1990 ) 3936 .
j. sonner , p. k. townsend and , phys .
d * 74 * , 103508 ( 2006 ) [ hep - th/0608068 ] .
k. -i .
maeda , k. yamamoto and , phys .
d * 87 * , 023528 ( 2013 ) [ arxiv:1210.4054 [ astro-ph.co ] ] .
e. teo , phys .
b * 609 * , 181 ( 2005 ) [ hep - th/0412164 ] . | we present a new way describing the solution of the einstein - scalar field theory with exponential potential @xmath0 in spatially flat friedmann - robertson - walker space - time .
we introduced a new time variable , @xmath1 , which may vary in @xmath2 $ ] .
the new time represents the state of the universe clearly because the equation of state at a given time takes the simple form , @xmath3 .
the universe will inflate when @xmath4 . for @xmath5 ,
the universe ends with its evolution at @xmath6 .
this implies that the equation of state at the end of the universe is nothing but @xmath7 . for @xmath8
, the universe ends at @xmath9 , where the equation of state of the universe is one . on the other hand
, the universe always begins with @xmath10 at @xmath11 .
the discovery of cosmic acceleration in 1997 , which is regarded to be originated from a dark energy , changed the modern cosmology from the basics . in the context of fundamental theories
, it is an important problem to understand the origin of the dark energy
. the dimensional reduction of higher dimensional m / string theory typically give rise to a scalar fields with exponential potentials coupled to four - dimensional gravity .
it is usually argued that these the homogeneous scalar field slowly rolling on the potential may give rise to a dark energy , which is typically called as quintessence @xcite .
it is interesting to understand whether exponential potentials could describe observational data for the late - time cosmic acceleration . in cosmology
, exponential potentials were much investigated in the past and general exact solutions have appeared @xcite . in particular ,
a general solution in four dimensions was obtained in ref .
@xcite . especially , in ref .
@xcite , it was also shown that in the simplest case of a homogeneous scalar field coupled to an exponential potential can be solved in a direct way in @xmath12-dimensions by introducing new variables which decouple the system . in ref .
@xcite , it was shown an example of cosmology which starts with a decelerating expansion , at some point it experiences a transitory period of acceleration , and it ends with decelerating expansion again . the origin of the acceleration was further clarified in ref . @xcite . as noted in ref .
@xcite , the explicit general solution will present a clear view in parametric space on the physical origin of the acceleration .
we are interested in the universe which is spatially flat , homogeneous , and isotropic , with metric : @xmath13 where @xmath14 is the scale factor .
we find exact cosmological solutions of einstein equation coupled to a scalar field @xmath15 with action in standard form , @xmath16,\ ] ] where we set @xmath17 .
the dynamics of the scalar field and gravity can be dealt with a pair of equations @xmath18 where overdot and prime denote derivatives with respect to time and the scalar field , respectively , and @xmath19 is the hubble parameter .
we consider the exponential potential , @xmath20 without loss of generality we set @xmath21 . in the case of an exponential potentials , the scalar cosmology in four dimensions
were investigated @xcite and general exact solutions were found @xcite .
interesting properties of the cosmological solutions with exponential solutions were also discussed @xcite . in refs .
@xcite , it was shown that the equations of motion and is equivalent to the ` _ generating equation _ ' : @xmath22 ^ 2,\end{aligned}\ ] ] which is typically called as the hamilton - jacobi equation , supplemented by @xmath23 summarizing , the two coupled differential equations and with respect to time is reduced to one non - linear first order differential equation with respect to the scalar field supplemented by the equation giving the dynamics .
we solve eq . directly to obtain the generating function in the cases of the exponential potentials .
one may easily guess a specific solution of the generating function since the derivative of the exponential is nothing but an exponential : @xmath24 where a real generating function of this form exists only when @xmath25 .
this example was dealt in ref .
@xcite and its fixed point properties including contributions from perfect fluid were studied in ref . @xcite .
the scalar field and scale factor corresponding to this is given by @xmath26 the whole cosmological solutions with the potential were studied in ref .
@xcite by changing the equations into two riccati equations using a couple of coordinates transformation .
the model was also studied in terms of nether charge method in ref .
@xcite and hamilton - jacobi method @xcite .
the model is extended to include a perfect fluid numerically in ref .
@xcite .
the solution corresponds to a solution approaching to the fixed point @xmath27 in ref .
@xcite . in this work corresponds to @xmath28 in ref .
@xcite .
] another two fixed points @xmath29 in the reference corresponds to @xmath30 , which is not relevant at the present situation . in this work
, we present the whole generating functions by solving eq . directly .
the general solution of eq . representing expanding universe is @xmath31 where we restrict the range of @xmath1 to be @xmath32 . for the range of @xmath33 ,
the generating function describes the potential @xmath34 , which we are not interested in at the present cosmological situation .
@xmath35 is chosen to be a parameter characterizing the maximum possible value of the scalar field during evolution , which will be shown below , and the scalar field evolves as @xmath36 .\end{aligned}\ ] ] for @xmath37 , this equation is ill - defined .
but , by using @xmath38 limit , we get @xmath39 the solution corresponds to the limit @xmath40 with @xmath6 . the schematic plot for the time evolution of the scalar field is given in fig .
[ fig : scalar ] .
[ cols= " < , < " , ] the equation of state parameter of the universe can be expressed in a quite simple form : @xmath41 because of the simplicity , it becomes easier to understand the behavior of the universe as a function of @xmath1 .
the equation of state becomes that of the cosmological constant at @xmath42 .
the universe inflates if @xmath43 , which gives @xmath44 .
the relation between the cosmological time and the time @xmath1 is explicitly given in appendix [ sec : t ] .
rather than explicitly describing the whole detail , let us remind the following key results : for @xmath45 , there are two corresponding universes .
one begins at @xmath46 ( @xmath47 ) and ends at @xmath6 ( @xmath48 ) and the other begins at @xmath49 ( @xmath47 ) and ends at @xmath6 ( @xmath50 ) .
remember that the time arrow is reversed in the second case . for @xmath51
, the universe begins at @xmath46 ( @xmath47 ) and ends at @xmath9 ( @xmath48 ) . noting @xmath52 , the scale factor , using eq . in appendix
[ sec : t ] , is given by @xmath53 integrating , we get the scale factor as a function of @xmath1 , @xmath54 the metric , now , can be written as , @xmath55 these solutions are not new but found in ref .
@xcite in a different form . for various specific parameter values ,
the solutions were also found in ref .
@xcite . in the rest of this work
, we analyze the behavior of the universe for each parameter space .
let us consider the case with @xmath56 first .
because the scale factor at @xmath6 diverges , there are two independent universes divided by the range of @xmath1 with @xmath57 and @xmath58 , in which the time runs in @xmath59 .
first consider the universe resides in @xmath57 .
the universe begins at @xmath46 with @xmath60 .
the universe is in a singular state because the scalar energy density diverges , which can be seen from eq . .
as the scalar field climbs up the potential , its velocity monotonically decreases due to the friction of the hubble parameter and the inclination of the potential .
the hubble parameter also decreases until @xmath1 reaches @xmath61 . at this time , the universe starts to inflate because the speed of the scalar field is slowed down and the scalar potential is big enough .
the scalar field arrive at @xmath62 at time @xmath42 and its velocity vanishes there .
thereafter , it turns its direction and starts to decrease .
however , note that the scalar velocity vanishes at @xmath6 .
therefore , even if the scalar field decreases continually , its velocity approaches zero in the future .
this results in a interesting result : if @xmath63 , the universe inflates eternally even though the scalar field continually decreases so that the scalar potential goes to zero . on the other hand ,
if @xmath64 , the inflation will end at @xmath61 and the universe enters into the decelerating expansion .
the universe will ends at @xmath6 .
therefore , the equation of state at the end of the universe is simply given by @xmath65 let us obtain the asymptotic form of the scalar field and the scale factor for @xmath66 .
the asymptotic form , in fact , is given by the fixed point solution .
then , from eq .
, one gets @xmath67 then , one can integrate eqs . and to obtain the same evolutions of the scale factor and the scalar field as in eq . .
for @xmath68 , the attractor denotes the inflationary attractor . on the other hand , for @xmath69 ,
it is not an inflationary attractor but is still an attractor approaching a state with its equation of state @xmath70 .
we next consider the case with @xmath71 .
the universe begins at @xmath9 with @xmath72 and @xmath73 . at the beginning ,
the universe starts from a singular state .
even though the scalar field slides down the potential , its speed can not always increase because the friction due to the hubble parameter competes with the slope of the potential .
the scalar velocity starts to decrease at some @xmath15 . at @xmath61 ,
its speed becomes small enough to support the inflation of the universe .
if @xmath74 , the universe will go into eternal inflating stage at @xmath61 . on the other hand , if @xmath75 , the universe fails to get into the inflating stage but continues the decelerating expansion .
the later evolutions can also be approximated by the fixed point solution .
the cosmological time runs @xmath59 during @xmath77 .
the universe begins at @xmath46 with @xmath60 and @xmath73 . at the beginning ,
the universe starts from a singular state . as the scalar field climbs up the potential , the velocity ( therefore kinetic energy ) of the scalar field decreases due to the friction of the hubble parameter and the slope of the potential . for @xmath78 ,
the kinetic energy is small enough to inflate the universe .
the velocity vanishes at @xmath79 , when @xmath42 .
afterward , the scalar field starts to slide down the potential .
however , as seen in eq . , the scalar velocity vanishes at @xmath9 .
therefore , it can not increase at all times but starts to decrease at some field value . in the future
, the scalar velocity goes to zero as @xmath60 .
the universe , after the temporal inflation , goes into decelerating expansion .
let us now consider the asymptotic behavior of the solutions in the cosmological time .
we first consider the early universe around @xmath80 .
using eq . , the scalar field and the scale factor behaves as @xmath81 where @xmath82 is a lengthy function of @xmath35 and @xmath83 and @xmath84 we next consider the later time solution with @xmath85 . as @xmath50 , the scalar field and the scale factor behaves as @xmath86 where @xmath87 is a lengthy function of @xmath35 and @xmath83 and @xmath88 comparing @xmath89 and @xmath90
, we find that the scale factor during the inflationary period is enhanced by @xmath91 unless @xmath92 , this value is not big enough to support the present observational data for inflation .
the asymptotic form of the scalar field and the scale factor are explicitly dependent on the choice of @xmath35 and deviates from the fixed point solution .
in fact , this corresponds to a kinetic attractor , with the equation of state @xmath93 corresponding to matter with dominance of kinetic energy with equation of state , @xmath94 the solution in this case can be obtained by taking the @xmath96 limit in eq . .
the evolution of the scalar field was given in eq . .
the derivative of the scale factor with respect to @xmath1 is the same as in eq . with @xmath37 . integrating this equation with eq .
, we get the scale factor , @xmath97 .\ ] ] the equation of state is also the same as eq . . as @xmath98 , the cosmological time runs @xmath59 .
the universe begins from @xmath46 with @xmath73 and @xmath99 . both of the initial scalar velocity and hubble parameter diverges . due to the friction from the hubble parameter , the scalar velocity decreases with time and
the scalar field starts to approach @xmath35 . at @xmath44 ,
the scalar velocity becomes small enough to support the inflation .
the scalar field takes its maximum value @xmath35 at @xmath42 and then bounces back to decrease . on the whole ,
the evolution of the scale factor and the scalar field is completely different from that of @xmath100 .
we now write down the asymptotic form of the solutions for @xmath80 using eq . , @xmath101 on the other hand , around @xmath85 using eq . , the scale factor becomes @xmath102^{1/3 } + \cdots\ ] ] where @xmath103 .
notice that the limiting behaviors are completely different from the cases of @xmath104 .
this limiting behavior can also be seen in the specific solutions in ref .
@xcite . in summary
, we directly attacked the hamilton - jacobi equation for the scalar cosmology with exponential potential @xmath0 .
we have reproduced the known solutions and introduced a new time variable @xmath1 , which varies in @xmath77 . in describing the evolution of the universe
, we have found that the new time variable is very convenient to identify the state of the universe at a given time .
the most beautiful property of it is that the equation of state of the scalar field is determined by the value very easily , @xmath105 . from this
, we notice that the universe will expands with accelerating rate if @xmath44 .
the evolution of the universe are characterized by the value of @xmath83 . for @xmath106 ,
the universe ends at @xmath6 .
therefore , the equation of state at the end of the universe is nothing but @xmath107 . for @xmath108
the universe ends with eternal later time inflation irrespective of the initial condition of the scalar field . on the other hand , for @xmath109
, the universe will ends at @xmath9 and the equation of state of the universe will approach to @xmath10 , implying the kinetic dominance .
the temporal inflation during @xmath78 makes the scale factor is increased by the factor @xmath110^{1/3 } $ ] .
however , this is not enough to explain the cosmic inflation or the present later time accelerating expansion . to see how useful the variable @xmath1 in describing the scalar cosmology is ,
more general systems need to be analyzed .
for example , in addition to the scalar field one may include a perfect fluid , an axion @xcite , or a gauge field @xcite .
if the scalar field plays the role of the dark energy , the value of @xmath83 should be very close to zero , @xmath111 for dark energy dominated future .
however , as pointed out by townsend @xcite , there is a conjecture saying that such spacetime with future event horizon can not arise from classical compactification of string / m - theory . for partial proof for this conjecture ,
consult ref .
@xcite .
if this is true , the value should be restricted to be @xmath69 discarding the previous possibility . |
13 g.g .
barnafldi , p. lvai , g. papp , g. fai and b.a .
cole , arxiv : hep - ph/0703059 ( 2007 ) .
s. albino , b.a .
kniehl and g. kramer , 181 ( 2005 ) ; _ ibid . _ * b734 * 50 ( 2006 ) .
star preliminary data , presented by ming yao on sqm07 conference | hadron production has been calculated in a pqcd improved parton model for @xmath0 , @xmath1 and heavy ion collisions .
we applied kkp and akk fragmentation functions .
our jet fragmentation study shows , that hadron ratios at high @xmath2 depend on quark contribution mostly and less on the gluonic one .
this finding can be seen in jet - energy loss calculations , also .
we display the suppression pattern on different hadron ratios in @xmath3 collisions at lhc energies . the precision of pqcd based parton model calculations was enhanced during the last decade .
the calculated spectra allow to make predictions not only for the hadron yields , but for sensitive particle ratios and nuclear modifications .
for the calculation of particle ratios new fragmentation functions are needed not only for the most produced light mesons , but for protons also . from the experimental point of view one
requires identified particle spectra by rhic and lhc .
especially the alice detector has a unique capability to measure identified particles at highest transverse momenta via herenkov detectors .
the @xmath4 and @xmath5 ratios can be measured up to @xmath6 gev / c and @xmath7 gev / c respectively . here
we calculate hadron ratios in our next - to - leading order pqcd improved parton model based on ref.@xcite with intrinsic transverse momenta , determined by the expected c.m .
energy evolution along the lines of ref.@xcite .
the presented ratios are based on @xmath8 , @xmath9 and @xmath10 spectra which were calculated by akk fragmentation functions @xcite .
first we compare calculated particle ratios to the data of the star collaboration measured in @xmath11 collisions at @xmath12 @xmath13gev rhic energy @xcite .
predictions for high-@xmath2 hadron ratios at rhic and at lhc energies in most central ( @xmath14 ) @xmath3 collisions are also shown in fig .
[ fig1 ] . on the _ left panel _ of fig .
[ fig1 ] , particle ratios are compared to @xmath11 collisions at @xmath15 @xmath13gev star @xmath16 ( _ dots _ ) and @xmath17 ( _ triangles _ ) data .
the agreement between the rhic data and the calculations at rhic energy can be considered acceptable at @xmath18 gev / c , with an opacity of @xmath19 . however , at lower momenta , where pqcd is no longer reliable , the ratios differ from the calculated curves .
the _ right panel _ shows calculations for @xmath3 collisions for @xmath20 @xmath13tev energy . using a simple @xmath21 estimation
, we expect a @xmath22 opacity in most central @xmath3 collisions . for comparison
, we plotted the @xmath23 and @xmath24 values also .
the lower- and intermediate-@xmath2 variation of the hadron ratios arise from the different strengths of the jet quenching for quark and gluon contributions@xcite . due to the quark dominated fragmentation
, the difference disappears at high-@xmath2 in the ratios . and @xmath17 ratios in @xmath25 collisions at rhic and lhc energies .
rhic curves are compared to star@xcite data at @xmath26 200 @xmath13gev.,title="fig : " ] and @xmath17 ratios in @xmath25 collisions at rhic and lhc energies .
rhic curves are compared to star@xcite data at @xmath26 200 @xmath13gev.,title="fig : " ] acknowledgments : special thanks to prof .
john j. portman for computer time at kent state university .
our work was supported in part by hungarian otka t047050 and nk62044 , by the u.s .
department of energy under grant u.s .
de - fg02 - 86er40251 , and jointly by the u.s . and
hungary under mta - nsf - otka oise-0435701 . |
in contrast to hadronic machines , a high energy @xmath5 collider in the tev range will be a rather unique place to discover and study charged higgses in a clean environment . these would be produced either in pairs @xcite , our main concern here , or in ( rare ) production in association with @xmath6 .
it was first found in @xcite that loop corrections from matter fermions and their susy partners ( mainly the @xmath7 sector ) , are likely to change the tree - level result at @xmath8 gev @xcite , by as much as @xmath9 dip in the cross - section .
the effect could even lie between @xmath10 and @xmath11 and perhaps grow out of perturbative control , though in a realistic range of the model - parameters .
such a sensitivity to loop effects appears to be related to the fact that at tree - level the @xmath12 and @xmath13 mediated process is exclusively controlled by @xmath14 gauge invariance and thus knows nothing about the structure of the non - standard extension whatsoever .
it is thus important in a first step to go beyond @xcite and study the full 1-loop structure , in order to understand the sensitivity to the various model - parameters and the origin of large effects , whether in the mssm or in a general thdm - ii . in a second step
one should perform a renormalization group analyses , eventually taking into account constraints from sugra - gut models , and ask whether the allowed regions of the model - parameters could still lead to large or even uncontrollable loop effects in charged higgs pair production . in the present study
we resume the first of these steps and improve upon the results of ref @xcite by including : * a ) * the complete higgs sector contributions ( self - energies , vertices and boxes ) ; * b ) * the infrared part , including initial and final soft photon radiation as well as @xmath15 and @xmath16 boxes ; * c ) * the complete set of charginos / neutralinos/@xmath17/@xmath18 box diagrams ; we performed the analysis in the on - shell renormalization scheme defined in @xcite . since at tree - level
only standard model couplings appear in @xmath0 , we need only to supplement the above scheme with the wave function and mass renormalization conditions of the charged higgs field .
it is also convenient to take the charged higgs mass @xmath19 , rather than that of the cp - odd scalar @xmath20 , as the input on - shell physical mass ( see @xcite for more details ) .
it will turn out that besides the sensitivity to the heavy quark - squark sector , there are on one hand model - independent large effects from the soft photon radiation , and on the other possibly important effects in the vertex corrections involving the purely higgs sector .
the latter case occurs when deviations from the tree - level mssm triple - self - couplings are allowed .
one of the issues will be the possibility of signing a clear difference between the mssm and a non - supersymmetric thdm - ii .
in this section we outline a general parametrization describing deviations from the tree - level higgs self - couplings as given in the mssm .
let us recall first that the most general gauge invariant scalar potential ( corresponding to cp - invariant thdm - ii ) is given by ( see for instance ref .
@xcite ) : @xmath21 & & + \lambda_{4}((\phi_{1}^+\phi_{1})(\phi_{2}^+ \phi_{2})-(\phi_{1}^+\phi_{2})(\phi_{2}^+\phi_{1}))+ \lambda_{5 } ( re(\phi^+_{1}\phi_{2 } ) -v_{1}v_{2})^2+\lambda_{6}im(\phi^+_{1}\phi_{2})^2 \label{pot}\end{aligned}\ ] ] where all @xmath22 are real and @xmath23 , @xmath24 denote the vacuum expectation values of the two higgs fields .
we have also omitted an additive constant in eq .
( [ pot ] ) .
thus one has in general 7 free parameters , i.e. the @xmath22 s and @xmath25 , @xmath26 being fixed by the electroweak scale .
let us now define @xmath27 where the @xmath28 s @xmath29 in general .
it is well known @xcite that ( softly broken ) supersymmetry forces the @xmath28 s to zero . from this
follow the usual tree - level relations among the higgs masses , @xmath30 and @xmath31 , as well as the negative sign of @xmath32 . in this case the triple scalar couplings read ( here we only show those entering the 1-loop corrections in @xmath0 ) : @xmath33 as one can easily see , none of the couplings in eq .
( [ gh ] ) can be enhanced at tree - level in the supersymmetric case . in the non supersymmetric case
however , @xmath34 , there is an important difference between @xmath35 and @xmath36 on one side , and the three remaining couplings of eq .
( [ gh ] ) on the other .
the couplings involving the charged goldstone bosons @xmath37 retain their initial form , so that a possible enhancement is due only to deviations from the supersymmetric higgs - boson - mass sum rules . in contrast
@xmath35 and @xmath36 will deviate from their forms in eq .
( [ gh ] ) through a non trivial dependence on @xmath31 , @xmath38 , @xmath39 , etc ... .
these aspects will be of importance in what follows .
starting from eq.(2 ) it is possible to express all couplings and mass relations in terms of the non - vanishing @xmath28 s , @xmath31 and one of the @xmath22 s .
( the general expressions are somewhat involved and will be given elsewhere ) .
here we would like to insist on a physically interesting special case : it is possible to define a deviation from the mssm in such a way that _ all _ the mssm tree - level relations ] among the higgs masses , @xmath30 and @xmath31 remain valid .
this occurs provided we require the @xmath28 s to satisfy @xmath40 under such conditions we have only three free parameters which we can choose as @xmath31 , @xmath41 and @xmath39 .
the mssm corresponds then to a particular value of @xmath39 , @xmath42 which is a function of @xmath31 and @xmath41 and leads to @xmath43 as a special case of eq .
( [ quasisusy ] ) . for this reason eq .
( [ quasisusy ] ) will be dubbed quasi - supersymmetric .
it can be thought of as corresponding to the situation where all the two - higgs doublet spectrum has been experimentally unraveled and found to be consistent with the mssm higgs mass sum rules but with no other direct evidence for supersymmetry .
although not generic , the quasi - supersymmetric parametrization would then be a very useful device in terms of which the tests of the higgs self - couplings can be expressed .
as far as the charged higgs sector is concerned only @xmath44 and @xmath45 can deviate from their supersymmetric values given in eq.([gh ] ) .
we give here for illustration their behavior in the large @xmath31 limit ( with @xmath46 and @xmath47 @xmath48 where @xmath49 , @xmath50 , @xmath51 , @xmath52 and @xmath53 are given in eq.([gh ] ) .
note that triple couplings of gauge bosons to higgses are never enhanced , and quartic couplings enter exclusively one - loop diagrams that vanish with @xmath54 .
it is thus interesting to note that the sensitivity to large @xmath31 resides exclusively in @xmath35 of eq .
( [ largetbeta ] ) , while @xmath36 is more sensitive to @xmath55 in this limit .
a more general investigation of the full higgs sector can be persued along the same lines .
we now present some numerical results , using the exact expressions for the higgs self - couplings in terms of the 3 free quasi - susy parameters , @xmath31 , @xmath38 and @xmath39 . in fig.1
we show the percentage contribution to the integrated cross section , for each sector separately , as a function of @xmath56 .
one finds that the higgs sector contributions can counterbalance those of the heavy quarks found in @xcite for large @xmath57 , but only near threshold and for @xmath46 .
far from threshold all of the effects , except for @xmath58 boxes , become negative . furthermore the `` neutral '' model - independent contributions , including soft bremsstrahlung , obtained by adding one photon ( or z ) line to the tree diagrams depend loosely on @xmath19 or @xmath56 and contribute at the level of @xmath59 for a soft photon cut @xmath60 . in fig.2
we show ( excluding those `` neutral '' contributions ) the integrated cross - section for two values of @xmath19 and @xmath57 . in thdm - ii the total loop effect increases ( negatively ) with increasing @xmath57 , the farther one goes from production threshold . in the mssm ( @xmath61 )
the leading effects come exclusively from the heavy quark - squark sector and the conclusions of @xcite remain unaltered in this case .
the @xmath62/@xmath18 contributions do not exceed a few percent despite the large number of diagrams .
for instance we evaluated the full set of supersymmetric boxes , ( taking into account fermion - number violation through the rules of ref .
@xcite ) , and found that they largely cancel among each other , leading at most to @xmath63 ( negative ) effect for a wide range of sparticle masses .
finally we should stress that the large effects in thdm - ii are not an artifact of the quasi - susy parametrization , they are also present even if the tree - level higgs mass sum rules are relaxed .
this raises the question of whether such effects would appear in the mssm as a result of the running of the @xmath22 s @xcite .
we conclude that the charged higgs sector seems to offer a non - trivial structure beyond the tree - level , which can complement the information gained from the neutral sector .
yet a more elaborate strategy involving the full - fledged higgs sector still needs to be worked out .
99 s. komamiya , phys.rev .
d38 ( 1988 ) 2158 ; a. arhrib , m. capdequi peyranere , g. moultaka phys.lett .
b 341 ( 1995 ) 313 ; m.bhm , w.hollik and h.spiesberger , fortschr.phys.34 ( 1986 ) 11 ; a.brignole et al . , in proceedings of the workshop on @xmath64 collisions at @xmath65 gev
: + the physics potential , ed .
p.m.zerwas , desy 92 - 123 , p.613 ; + a.djouadi , j.kalinowski , p.m.zerwas , _ ibid .
_ p.83 and z.phys.c57 ( 1993 ) 569 ; j.f.gunion and h.e.haber , nucl .
b272 ( 1986 ) 1 ; a.denner , h.eck , o.hahn and j.kbelbeck nucl .
b387 ( 1992 ) 467 ; h.e . haber and r. hempfling ,
d48 ( 1993 ) 4280 ; | we study the 1-loop corrections to the charged higgs production both in the minimal supersymmetric standard model ( mssm ) and in a more general type ii two - higgs - doublet model ( thdm - ii ) .
we consider the full set of corrections ( including soft photon contributions as well as box diagrams ) , and define a parametrization that allows a comparison between the two models .
besides the soft photon radiation there can be prominent model - dependent effects .
-0.25 in .3 cm _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ hep - ph / yymmxxx + pm/96 - 16 + june 1996 _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ * radiative corrections to @xmath0 linear colliders workshop _ , annecy
gran sasso
hamburg 1995 , ed .
p. zerwas ] * a. arhrib@xmath1 and g. moultaka@xmath2 + @xmath2 physique mathmatique et thorique , e.s.a .
du cnrs n@xmath3 5032 , + universit montpellier ii , f-34095 montpellier france + @xmath4 l.p.t.n .
, facult des sciences semlalia , b.p .
s15 , marrakesh , morocco * radiative corrections to @xmath0 * a. arhrib@xmath1 and g. moultaka@xmath2 + @xmath2 physique mathmatique et thorique , e.s.a .
du cnrs n@xmath3 5032 , + universit montpellier ii , f-34095 montpellier france + @xmath4 l.p.t.n .
, facult des sciences semlalia , b.p .
s15 , marrakesh , morocco |
modified @xmath0 gravity has become one of the most popular mechanisms to generate a late accelerated expansion in the universe without the need of introducing new fields @xcite .
it was also one of the first consistent models for early inflation @xcite . during the past fifteen years or so
, several specific @xmath0 models have been thoroughly analyzed in many scenarios , but only a few of them can survive the classical tests ( e.g. solar system , binary pulsar ) while predicting the correct accelerating expansion , and in general , a successful cosmological model , both at the background and at the perturbative level .
therefore , it is still unclear until what extent this kind of alternative theories of gravity can recover _ all _ the successes of general relativity ( gr ) while making new testable predictions . as concern
black hole solutions , the situation of @xmath0 gravity can , in some sense , differ from gr , and in other sense be almost the same .
the last statement is related with the content of section [ sec : theory ] about the existence of the same kind of vacuum black - hole solutions found in gr , while the former concerns the existence of hairy solutions , or lack thereof , that we analyze in all the rest of the paper .
perhaps the first and simplest theorem concerning bh solutions in gr was the birkhoff s theorem ( bt ) . roughly speaking
, this theorem establishes that in vacuum all spherically symmetric ( ss ) spacetimes are also static , and those that are asymptotically flat ( af ) are represented by a one - parameter family of solutions , namely , the ubiquitous schwarzschild solution , where the parameter is interpreted as the ( adm ) _ mass _ @xmath1 of the spacetime ( see @xcite for a discussion ) .
remarkably , when including an electric field , the ss solution can be extended as to include the charge @xmath2 of the bh , it is the well known two - parameter reissner
nordstrm ( rn ) solution . in the af case ,
both the schwarzschild and the rn solution are @xmath3 solutions of the einstein s field equations ( also termed _
ricci flat _ solutions ) .
when including a cosmological constant @xmath4 , the schwarzschild and rn black holes become a two and three - parameter family respectively , and the solutions are asymptotically de sitter ( ads ) or asymptotically anti de sitter ( aads ) , depending if @xmath5 or @xmath6 , respectively . in the decade of 1960 s , motivated by the discovery of the kerr solution , several theorems (
notably , the _ uniqueness theorems _ ) were established for stationary and axisymmetric spacetimes both in vacuum and with an electromagnetic field ( see @xcite for details and reviews ) .
one of the main consequences of those theorems is that in the af case the solutions of the einstein field equations under such symmetries are characterized _ only _ by three parameters , the mass @xmath1 , the charge @xmath2 , and the angular momentum @xmath7 of the bh .
these solutions are known as the kerr
newman family , which extends the schwarzschild and rn black holes to more general spacetimes : stationary and axisymmetric . due to the apparent simplicity of such solutions ,
wheeler established the so called _ no hair conjecture _ , a statement that `` doomed '' all possible stationary afbh solutions from having any other parameters than those three ( @xmath8 and @xmath7 ) .
this conjecture has been reinforced thereafter by the elaboration of several _ no - hair _ theorems ( nht s ) that forbid the existence of bh solutions with more parameters associated with other kinds of matter fields ( see @xcite for a review ) . among such theorems
one can mention those that include several kinds of scalar fields . eventually , this conjecture proved to be `` false '' , for instance , within the einstein yang mills system @xcite , and einstein - scalar - field system with `` exotic '' potentials that can be negative @xcite . or when including rotation like in the einstein - boson - field system @xcite .
nonetheless , since most , if not all of such _ hairy _ solutions are unstable , the community ( or at least part of it ) consider those solutions as _ weak _ counterexamples to the wheeler s no - hair conjecture .
therefore , it has been tantalizing to extend the conjecture in the following more precise , although still informal , statement : _ the only stable stationary afbh s are within the kerr
newman family _ @xcite .
the proposal of alternative theories of gravity as a possible solution to the dark - matter and dark - energy problems and to other theoretical problems ( e.g. inflation , gravity renormalization ) has motivated people to generalize several of the theorems and conjectures mentioned above , and which pertain to gr , to the realm of other modified - gravity proposals .
while it is out of the scope of the present paper to review all such attempts , we shall simply focus on @xmath0 metric gravity . unless otherwise stated , by @xmath0 gravity we mean a theory that departs from the gr function @xmath9 . as concerns
this kind of theory , a large amount of analysis has been devoted to establish an analogue of the bt for the ss situation @xcite . however , the reality is that no rigorous bt exists today in @xmath0 gravity , as far as we are aware . in fact
, if one such theorem were proved , certainly should be restricted to some specific @xmath0 models .
moreover , the theorem should establish at least four things , upon fixing the boundary conditions ( i.e. regularity and asymptotic conditions ) : 1 ) _ staticity : _ the only spherically symmetric solutions in vacuum [ i.e. without any matter field associated with the standard model of particle physics or any other field that is not associated with the lagrangian @xmath0 ] are necessarily static ( i.e. the existence of a static killing field should be proved from the spherically symmetric assumptions ) ; 2 ) _ existence : _ the existence of an exact static spherically symmetric ( sss ) solution in vacuum ; 3 ) _ uniqueness : _ the sss solution found in point ` 2 ) ' is the only solution in vacuum ( or prove otherwise ) ; 4 ) the conditions under which the solution in point ` 2 ) ' matches or not the exterior solution of an sss extended body .
so far , in vacuum only a few bh exact solutions exist in @xmath0 gravity , and those that are genuine af , ads or aads correspond simply to the same kind of solutions found in gr , where @xmath11 everywhere in the spacetime ( with @xmath12 , @xmath13 or @xmath14 , respectively ) .
it is unclear if other solutions exist with the same kind of asymptotics but with a varying @xmath15 in the domain of outer communication of the bh .
we shall elaborate more about this point below to be more precise .
furthermore , in the presence of matter ( i.e. a star - like object ) , it is possible to find sss solutions where @xmath15 can vary in space @xcite , however , those solutions are not exact , but given only numerically , and it is unclear if the exterior part ( i.e. the vacuum part ) of those solutions is the same solution found when matter is totally absent in the spacetime , if it exists at all , as it happens in gr where the exterior solution of extended objects under such symmetries is always given by the vacuum schwarzschild solution . now , despite the absence of such bt , some nht s have been proved in this kind of theories . in order to do so
, people have resorted to the equivalence between a certain class of @xmath0 models ( notably , those where @xmath16 and @xmath17 , where the subindex indicates differentiation ) with scalar - tensor theories ( stt ) .
the point is that one performs a conformal transformation from the original _ jordan frame _ to the so - called _ einstein frame _ where the conformal metric appears to be coupled minimally to gravity and a new scalar - field @xmath18 emerges , where @xmath19 , which is also coupled minimally to the conformal metric but endowed with an `` exotic '' potential @xmath20 .
thus , the available nht s constructed for the einstein - scalar - field system in gr can be applied for these theories as well ( see section [ sec : stt ] ) , notably in vacuum , and when the spacetime is af and the potential satisfies the condition @xmath21
@xcite for the potential .
this proof is similar to bekenstein s @xcite which assumes only _ stationarity _ as opposed to _
staticity_. ] .
we stress that the applicability of such nht s is possible because the non - minimal coupling between the scalar field @xmath18 and the matter fields that usually appears under the einstein frame obviously vanishes in the absence of the matter .
the only caveat of this method is that the potential @xmath20 is not given a priori but is the result of the specific @xmath0 model considered ab initio , and thus , @xmath20 can be negative or even not well defined ( i.e. it can be multivalued ) , which in turn can jeopardize the use of the nht s .
consequently , the existing nht s in @xmath0 gravity can reduce the kind of afsssbh solutions that are available in some specific models , but do not rule out completely the absence of _ geometric hair_. in this context , by ( geometric ) _ hairy _ solutions within @xmath0 gravity we mean afsssbh solutions where the ricci scalar is not _ trivial _
( i.e. constant ) , but rather a function that interpolates non - trivially between the horizon and spatial infinity .
thus , when the condition @xmath21 fails and the nht s are not applicable one can resort to a numerical analysis for evidence about the existence of such hair or its absence thereof . at this respect it is important to stress that regularity conditions have to be imposed at the inner boundary , namely , at the bh horizon @xmath22 in order to prevent the presence of singularities there . in section [ sec : regcond ] and
appendix [ sec : regcond2 ] we obtain such regularity conditions and then in sections [ sec : expmodels ] and [ sec : numerical ] we present analytical and numerical evidence , respectively , showing that hairy solutions are absent in several specific @xmath0 models proposed as dark - energy alternatives in cosmology .
in particular , the models considered in section [ sec : numerical ] are precisely those for which the nht s can not be applied as the corresponding potential @xmath20 can be negative or is not even well defined . on the other hand ,
when such hairy solutions are absent , one may still find the trivial solution @xmath23 for which the field equations reduce to the einstein field equations with an effective cosmological constant @xmath24 , and an effective gravitational constant @xmath25 where @xmath26 is a solution of an algebraic equation involving @xmath0 and @xmath27 .
this includes the case where @xmath28 .
therefore , in such circumstances , all the best known bh solutions found in gr exist also in @xmath0 gravity simply by replacing the usual cosmological constant @xmath4 by @xmath29 , and the newton s gravitational constant @xmath30 by @xmath31 .
in view of this we shall argue in section [ sec : theory ] that such solutions are so trivial ( i.e. trivial in the context of @xmath0 gravity ) that almost nothing new arise from them . finally , we mention that some `` non - trivial '' exact sssbh solutions have been reported in the literature as a result of very ad hoc @xmath0 models @xcite .
notwithstanding such solutions _ can not _ be considered as hairy solutions because they have unusual asymptotics , and therefore , the corresponding `` hairless '' solution @xmath32 ( including the ricci - flat solution ) does not even exist with the same kind of asymptotics .
we shall discuss one such solution in section [ sec : exactsols ] .
the article is organized as follows : in section [ sec : theory ] we discuss in a general setting the conditions for the existence of several trivial bh solutions . in section [ sec : sss ] we focus on sss spacetimes and provide the corresponding differential equations to find bh solutions .
we also discuss some exact solutions that will be used later to test a numerical code constructed to solve the equations .
the boundary conditions appropriate to solve these equations with the presence of a bh are given in section [ sec : sss ] in form of _ regularity conditions _ at the horizon .
no - hair theorems and the properties of @xmath0 gravity formulated in the einstein frame are analyzed in section [ sec : stt ] . in that section
we also provide strong numerical evidence about the absence of hair for several @xmath0 models when the nht s do not apply .
our conclusions and final remarks are presented in section [ sec : concl ] .
several appendices at the end of the article complement the ideas of the main sections .
the general action for a @xmath0 theory of gravity is given by @xmath33 = \!\ !
\frac{f(r)}{2\kappa } \sqrt{-g } \ : d^4 x + i_{\rm matt}[g_{ab } , { \mbox{\boldmath{$\psi$ } } } ] \ ; , \ ] ] where @xmath34 ( we use units where @xmath35 ) , we extend our units so that @xmath36 as well . ] , and @xmath0 is a sufficiently smooth ( i.e. @xmath37 ) but otherwise an a priori arbitrary function of the ricci scalar @xmath15 .
the first term corresponds to the modified gravity action , while the second is the usual action for the matter , where @xmath38 represents schematically the matter fields .
the field equation arising from the action ( [ f(r ) ] ) under the metric approach is @xmath39 where @xmath27 stands for @xmath40 ( we shall use similar notation for higher derivatives ) , @xmath41 is the covariant dalambertian and @xmath42 is the energy - momentum tensor of matter resulting from the variation of the matter action in ( [ f(r ) ] ) .
it is straightforward , although a non - trivial result , to show that the conservation equation @xmath43 holds also in this case ( see appendix [ sec : consemt ] for a proof ) . in turn , this latter leads to the geodesic equation for free - fall particles @xmath44 .
therefore , the weak - equivalence principle ( for point test particles ) is also incorporated in this theory as well .
actually @xmath0 metric gravity preserves all the axioms of gr but the one that assumes that the field equations for the metric @xmath45 must be of second order .
clearly the only case where this happens is for @xmath9 , which leads to gr plus a cosmological constant ( hereafter gr@xmath4 ) .
now , taking the trace of eq .
( [ fieldeq1 ] ) yields @xmath46\,\,\,,\ ] ] where @xmath47 . when using ( [ tracer ] ) in ( [ fieldeq1 ] ) and
after some elementary manipulations we obtain @xcite @xmath48 } \ ; .\ ] ] equations ( [ fieldeq3 ] ) and ( [ tracer ] ) are the basic equations that we have used systematically in the past to tackle several problems in cosmology and astrophysics @xcite , and that we plan to use in this article as well . now , apart from the gr@xmath4 theory for which @xmath49 , @xmath50 , and @xmath51 , for more general models , one imposes the conditions @xmath52 , for a positive @xmath31 , and @xmath17 , for stability @xcite
however , in this paper we shall sometimes relax these two assumptions in order to explore its consequences for the sake of finding bh solutions . in vacuum , that is when @xmath53 , or more generally , in the presence of matter fields where @xmath54 , like in electromagnetism or yang mills theory , eq .
( [ tracer ] ) admits in principle the trivial exact solution @xmath55 where @xmath56 is a solution of the algebraic equation @xmath57/f_{rr}=0 $ ] .
in particular , if @xmath58 and @xmath59 , as often happens in potentially viable @xmath0 models , then @xmath56 is an algebraic solution of @xmath60 that we call @xmath26 . for such kind of solutions the field
( [ fieldeq3 ] ) reduces to @xmath61 where @xmath62 we assume @xmath63 .
moreover , the condition @xmath64 is an _ stability _ condition that is usually imposed in order to avoid exponentially growing modes when perturbing around the value @xmath65 . on the other hand ,
if there exists other value @xmath11 , that we call @xmath66 ( in order to avoid confusion with @xmath26 ) such that @xmath67 blows up at @xmath68 , but @xmath69 then @xmath66 can be also a possible trivial solution of eq .
( [ tracer ] ) .
however , in such a case one must be extremely cautious as in eq .
( [ fieldeq3 ] ) may appear products of the sort @xmath70 ( or @xmath71 ) .
even if such a product is finite , namely zero , still the interpretation of such solution would be problematic as @xmath72 would be singular everywhere . at
what extent such `` singularity '' is physical , is something that one should clarify . in sec .
[ sec : exactsols ] , we shall be dealing with an sss solution where the ricci scalar is not constant but @xmath73 as @xmath74 .
hence , in that example @xmath75 , @xmath76 and @xmath77 as @xmath74 .
however , these pathologies , as peculiar as they may be , does not concern us too much in this article , since we will be mainly interested in situations where they are absent .
now , in the particular and simplest scenario where @xmath65 is one of the trivial solutions of eq .
( [ tracer ] ) , we see that the field equation ( [ einsred ] ) corresponds to gr with the usual `` bare '' cosmological constant @xmath4 and the bare newton s
constant @xmath30 replaced by @xmath29 and @xmath31 , respectively .
therefore , in that occurrence all the solutions that exist in gr exist in @xmath0 gravity as well when taking into account the above replacements .
in particular , the afbh solutions with @xmath3 that exist in gr , like the kerr
newman family and its sss limit , exist also in @xmath0 gravity if @xmath78 , i.e. , if @xmath79 . on the other hand , the de sitter or anti - de sitter bh solutions associated with the kerr
newman family with a cosmological constant @xcite exist also in @xmath0 gravity if @xmath80 .
it is note that bh solutions @xmath81 , which from the point of view of @xmath0 gravity are _ trivial _ , have been systematically reported in the literature as something new or special ( e.g. see refs .
however , as we just showed , the existence of such solutions stems from the fact that @xmath26 exists in various @xmath0 models , like the ones we consider in section [ sec : expmodels ] . in turn
, the existence of such a trivial solution is just a standard demand for @xmath0 theories to produce a late acceleration expansion : the cosmological constant @xmath29 emerge while the universe evolves towards the solution @xmath82 as the universe expands and matter dilutes .
so nothing exceptional , astonishing or radically different from the already known bh solutions in gr are to be expected in this kind of trivial solutions in @xmath0 gravity . furthermore , and as a consequence of these remarks , the bekenstein hawking entropy defined for such bh s , has @xmath31 instead of @xmath30 in the formula .
that is , the modified entropy is @xmath83 @xcite , where @xmath84 is the area of the bh event horizon , instead of just @xmath85 .
let us consider now the `` odd '' scenario where @xmath26 is such that @xmath86 , and @xmath87 . then eq .
( [ tracer ] ) admits @xmath88 as one possible trivial solution . nonetheless , the possible solutions for the metric that satisfy eq .
( [ fieldeq3 ] ) _ degenerate _ and a whole spectrum of solutions can emerge , besides the usual af vacuum solutions mentioned above .
such solutions are _ all _ the possible solutions of the einstein equation @xmath89 compatible with a _ null
_ ricci scalar @xmath88 .
more specifically , if one considers an @xmath0 model in vacuum such that @xmath90 at @xmath3 ( implying in turn @xmath91 ) but assuming @xmath92 , then eq . ( [ tracer ] ) is solved trivially .
moreover the field eq .
( [ fieldeq3 ] ) in vacuum reduce to @xmath93 which is trivially satisfied for @xmath94 even if @xmath95 .
thus , in that instance the model admits unexpectedly all the possible solutions associated with the non - vacuum einstein equation @xmath89 compatible with @xmath96 .
now , given such scenario one has to deal with a problematic interpretation of the new global quantities that appear in the bh solutions as integration constants since in reality we are dealing with an @xmath0 model in vacuum . in non - vacuum
gr the global quantities ( other than @xmath1 and @xmath7 ) are ascribed to properties associated with the matter described by @xmath42 ( like the electric charge @xmath2 or the nodes of the yang
mills field ) .
therefore the same quantities would appear in this degenerate scenario , but clearly they can not have the same interpretation in vacuum . in section [ sec : expmodels ] within the framework of model 1 , we shall encounter one such example which is associated with the model @xmath97 . in this model
@xmath3 is a trivial solution and clearly @xmath94 , thus all possible solutions of einstein equations with a traceless energy - momentum tensor are in principle allowed , for instance , afsss solutions .
in fact , one such solution is exactly the analogue of the rn , where the roll of the charge is played by a new quantity that appears as an integration constant .
this interesting situation was also analyzed recently in ref .
@xcite and remarked before in ref .
we shall elaborate more about this in sections [ sec : exactsols ] and [ sec : expmodels ] , notably , concerning the issue of _ uniqueness _ and the bt .
having clarified the fact that @xmath0 gravity naturally admits trivial bh solutions where @xmath98 everywhere , notably the ones with @xmath3 , and which correspond to the same solutions found in gr modulo a trivial redefinition of constants , the main goal of this paper is an effort to find afsss solutions where @xmath99 is a non - trivial solution of the field equations , where @xmath100 is some radial coordinate .
this means that if such solution exists , @xmath99 should interpolate in a non - trivial manner between the event ( killing ) horizon of the bh and the asymptotic region .
if exist , this is what we might call a _ hairy _ solution . in order to find such a solution we enforce suitable regularity conditions at the horizon .
these conditions provide a very specific form of derivatives of several variables at the horizon , notably , the first derivative of the ricci scalar .
these regularity conditions are extremely important as they prevent the presence of bh solutions that are pathological at the horizon .
in particular , they prevent a singularity in the scalar - degree of freedom at the horizon , a singularity that would be otherwise considered as _
physical _ as opposed to a coordinate singularity . in section
[ sec : regcond ] and in appendix [ sec : regcond2 ] we provide such conditions in its full form and compare them with similar regularity conditions reported first in @xcite , but amended in @xcite .
examples of such pathologies ( singularities ) are common in stt when using conformal methods to generate sss exact solutions without enforcing the regularity condition of the scalar field at the bh horizon @xcite .
thus , the fact that the field equations may not be satisfied at the horizon in those examples cast serious doubts about its relevance as genuine counterexamples to the no - hair conjecture .
the problem of describing a sss space - time in vacuum ( @xmath101 ) within the framework of @xmath0 gravity reduces to solving the field equations ( [ fieldeq3 ] ) and ( [ tracer ] ) for the following sss metric : @xmath102 where the _ mass function _ @xmath103 provides the adm mass in the asymptotic region provided the spacetime is af .
component ) . in those spacetimes
the equivalent of the adm mass corresponds typically to a suitable renormalization of @xmath104 ( see footnote [ f : ads ] ) . ]
the function @xmath105 indicates the extent to which the equality @xmath106 is satisfied or infringed by the components of the einstein tensor or equivalently , by the corresponding components of the effective energy - momentum tensor given by the r.h.s of eq .
( [ fieldeq3 ] ) taking @xmath101 there .
so , if @xmath107 , in particular zero , it means that the equality holds exactly everywhere in the spacetime .
this situation includes some of the the best known sss spacetimes .
the field equations for the metric eq .
( [ metric ] ) have been obtained in a rather convenient form in @xcite with a slightly different but equivalent parametrization .
we present their final form based on those equations but without the matter terms : @xmath108 + \bigg [ \frac { 2 ( rm^{\prime } - m ) } { ( r - 2 m ) r } - \delta ' - \frac{2}{r } \bigg]r ' \,\,\,,\\ \label{msss } m ' & = & \frac{m}{r } + \frac{1}{2(2f_{r } + rr'f_{rr})}\bigg\ { -\frac{4f_{r}m}{r } + \frac{r^2}{3}(rf_{r } + f ) \nonumber\\ & + & \frac{rr'f_{rr}}{f_{r } } \bigg [ \frac{r^2}{3}(2rf_{r } - f ) - \frac{4 m}{r } f_{r } + 2rr'f_{rr}\big(1 - \frac { 2m}{r}\big)\bigg ] \bigg\ } \,\,\,,\\ \label{deltasss } \big(1- \frac{2m}{r}\big)\delta ' & = & \frac{1}{2(2f_{r } + rr'f_{rr})r}\bigg\ { \frac{2r^2}{3}(2f - rf_{r } ) + \frac{rr'f_{rr}}{f_{r}}\big [ \frac{r^2}{3}(2rf_{r } - f ) - 2f_{r } \big ] \nonumber\\ & - & 4(r - 2m)r'f_{rr } + \frac{2(f_{r } + rr'f_{rr } ) ( r - 2 m ) r'f_{rr}}{f_{r } } \bigg\}. \end{aligned}\ ] ] in order to find bh solutions , eqs .
( [ tracersss])([deltasss ] ) have to be solved from the bh horizon at @xmath109 to the asymptotic region which in the af or anti - de sitter cases corresponds to spatial infinity . on the other hand , in a de sitter background a cosmological horizon can be reached at some @xmath110 . in this paper
we shall focus only in af solutions , therefore we deal only with the bh horizon .
suitable boundary conditions at @xmath22 are to be imposed .
typically one assumes _ regularity conditions _ that enforce the verification of the field equations there ( see section [ sec : regcond ] ) .
for instance , the value @xmath111 enforces the existence of the killing horizon , where @xmath22 can have , in principle , any arbitrary non - negative value .
the condition @xmath112 is also arbitrary as it simply determines the value of @xmath105 in the asymptotic region which can be redefined by a suitable change of the @xmath113 . in other words ,
the field equations are left invariant if one performs the transformation @xmath114 , a property that holds due to the existence of a static killing vector field .
so without loss of generality we can take @xmath115 . as concerns @xmath116 , hereafter @xmath117 , this value is usually fixed so as to obtain the desired asymptotic value for @xmath15 . typically , but not necessarily ( see below ) , a value @xmath118 ( where @xmath119 stands for @xmath74 ) gives rise to af spacetimes with @xmath120 . for de sitter backgrounds enforcing the boundary conditions
can be more involved since a cosmological horizon at @xmath121 ( @xmath122 ) is present , and in this case the value @xmath117 is to be fixed so as to recover the regularity conditions at @xmath121 as well . ] . due to the complexity of the field equations ( [ tracersss])([deltasss ] )
one usually resorts to a numerical solution , in which case , the value @xmath117 is fixed by a _ shooting method _ ( cf .
section [ sec : numerical ] ) .
this value , which provides the adequate asymptotic behavior for @xmath99 , is also related with the the convergence of @xmath103 to the _ adm mass_. equation ( [ msss ] ) can be written as in gr in the form @xmath123 where @xmath124 can be readoff from eq .
( [ msss ] ) . in this kind of coordinates the adm mass @xmath125 for af spacetimes is given by @xmath126 , also called the komar mass @xcite . in order for @xmath1 to converge to @xmath125
one requires @xmath127 asymptotically with @xmath128 .
that is , we require that the effective energy - density falls off faster than @xmath129 . otherwise @xmath103 can diverge asymptotically as @xmath130 if @xmath131 or @xmath132 if @xmath133 in the case of anti - de sitter backgrounds the corresponding `` adm '' mass is given by @xmath134 where @xmath135 ( @xmath136 ) .
a rigorous definition for the _ mass _ in this kind of backgrounds has been given by several authors @xcite .
this mass is like the komar mass @xcite but specialized to the sss scenario .
for instance , when the mass function is given by @xmath137 where @xmath138 is a constant ( like in kottler
schwarzschild de sitter metrics ) , @xmath138 coincides with the mass found from the formal definitions @xcite . in the context of @xmath0 gravity
@xmath139 must behave as @xmath140 with @xmath128 for @xmath141 to converge . ] . as a consequence ,
if the spacetime containing a bh possesses a non - trivial ricci scalar @xmath99 , this must behave as @xmath142 , namely @xmath143 with @xmath128 in the af scenario .
therefore , the shooting method is implemented such that @xmath99 behaves asymptotically in the previous manner using @xmath117 as a control parameter .
if no such behavior exists for all @xmath117 , this implies that af spacetimes with a _ non - trivial _
ricci scalar do _ not _ exist .
thus , scalar - hair would be absent and the only possible af solutions are @xmath3 . there exist spacetimes with asymptotic behaviors different from af or ads / aads with interesting properties .
one of them correspond to a spacetime that is af except for a deficit solid angle @xmath144 .
this kind of spacetimes are typically generated by topological defects , like strings and global monopoles @xcite .
the deficit angle is related with the `` symmetry breaking '' scale @xmath145 of the mexican hat potential by @xmath146 @xcite .
the deficit angle produces typically a repulsion instead of an attraction of test bodies in the gravitational field generated by these defects . in this kind of spacetimes the mass function @xmath147 asymptotically .
it is somehow remarkable that such spacetimes can have a well defined mass @xcite which , in the case of spherical symmetry , can be computed , in practice , by a suitable renormalization of @xmath126 .
this situation corresponds to @xmath148 .
it is then tantalizing to define the `` adm '' mass in this case simply by @xmath149 , where @xmath150 . in the next section
we shall see that in fact such a mass is not exactly @xmath151 but rather proportional to it where the constant of proportionality depends on the deficit angle @xmath152 .
furthermore , @xmath153 asymptotically .
hence , we have here an explicit example where the ricci scalar is not trivial and it also vanishes asymptotically , and yet the spacetime is not af .
below we present an explicit @xmath0 model for which an exact sss of this sort can be found , except that the deficit angle is not produced by a topological defect , but by the underlying modified - gravity model itself .
we finish this section by giving explicitly the kind of trivial solution @xmath154 that we alluded in section [ sec : theory ] . for sss spacetimes
we have @xmath155 with @xmath156 this is the kottler schwarzschild de sitter solution , where @xmath157 is a constant of integration of eq .
( [ msss ] ) which is identified with the `` adm '' mass ( cf . ref .
@xcite ) . in the cases where @xmath28
, the solution reduces to the schwarzschild solution .
it is a straightforward exercise to check that eqs .
( [ mrtriv])([derpotcond ] ) solve exactly eqs .
( [ tracersss])([deltasss ] ) , regardless of the @xmath0 model , provided the model satisfies eq .
( [ derpotcond ] ) and @xmath158 , @xmath159 , @xmath160 and @xmath161 .
as we show in the next section , when the @xmath0 model is given explicitly @xmath26 writes in terms of its fundamental parameters [ cf .
( [ lefffrsz ] ) ] .
let us consider the model @xmath162 where @xmath163 is a parameter with units @xmath164^{-1}$ ] , and @xmath165 is another parameter of the model which is related to an effective cosmological constant as we show below .
this @xmath0 model and a variant of it was considered in the past by several authors @xcite . in the sss scenario the metric @xmath166 with the mass function , ricci scalar , and @xmath105 given , respectively , by @xmath167
solve eqs .
( [ tracersss])([deltasss ] ) exactly , as one can verify by straightforward substitutions . here
, @xmath2 is an integration constant .
taking into account electromagnetic and yang mills fields , this solution was extended in ref .
when @xmath168 this solution was part of a more general class of solutions associated with the model @xmath169
however , for @xmath170 the asymptotic behavior of those solutions was not analyzed by those authors as we do here .
the coordinates are defined such that , @xmath171 , @xmath172 , @xmath173 with @xmath174 if @xmath175 and @xmath176 if @xmath177 , where @xmath22 and @xmath121 corresponds to the location of the event and cosmological horizons , respectively , that we analyze below .
the metric ( [ gfrsz ] ) possesses a deficit angle @xmath178 , @xmath179 , `` charge '' @xmath2 and a cosmological constant @xmath180 ( see appendix [ sec : afdefang ] for more details ) . in the current case @xmath181
@xcite where @xmath182 .
the divergent terms ( linear and cubic in @xmath100 ) appear in this renormalization of mass since , as we remarked before , the spacetime has a deficit angle ( associated with the linear term ) and also a cosmological constant ( associated with the cubic term ) . using ( [ mrexact ] )
we conclude @xmath183 , thus , @xmath184 .
examples of spacetimes with a deficit angle and with zero - mass bh are not new @xcite . the metric with a deficit angle given by eq .
( [ gfrsz ] ) is a solution with a cosmological constant , @xmath185 .
for instance , taking @xmath186 we can introduce @xmath187 for convenience .
the location of the black hole horizon depends on the value of @xmath2 .
there are three possibilities : a ) if @xmath188 , @xmath189 , the event horizon of the black hole is located at @xmath190 ; b ) if @xmath191 ( @xmath192 ) and @xmath193 , there are two horizons which are located at @xmath194 , and @xmath195 .
in particular , for an _ extreme _ black hole with @xmath196 the horizon is given by @xmath197 ; c ) if @xmath198 , then @xmath199 .
the aads spacetime with a deficit angle is a solution with a negative cosmological constant @xmath185 ( with @xmath200 ) and @xmath201 .
notice that for @xmath202 an horizon does not exist like in the usual anti - de sitter ( i.e. anti - de sitter spacetime without the deficit angle ) .
however , for @xmath203 an horizon exists and is located at @xmath204 .
when @xmath205 , the spacetime turns to be af except for a deficit angle . in this case only for @xmath206 there is an horizon at @xmath207 . finally when @xmath208 , the spacetime is simply the minkowski spacetime with a deficit angle .
all other cases correspond to naked singularities . in absence of naked singularities
a straightforward calculation of other scalars , like @xmath209 and @xmath210 , show that the only _ physical _ singularity appear at @xmath211 . even if @xmath208 , such scalars are @xmath212 and @xmath213 while @xmath214 , and thus , the singularity at @xmath211 is entirely due to the deficit angle .
since the coordinates used so far do not cover the entire manifold one can look for analytic extensions using kruskal - like coordinates .
these extensions and the construction of penrose diagrams are out of the scope of the current paper and will be reported elsewhere .
now , for @xmath200 , there is no cosmological horizon , and thus , one can analyze the solution as @xmath74 .
it is then interesting to note that the asymptotic value of the ricci scalar @xmath215 is not an algebraic solution of @xmath216 which is @xmath217 but rather a pole of @xmath218^{-3/2}/2 $ ] .
the trivial solution @xmath217 provides a cosmological constant @xmath219 which is different from the actual one @xmath220 . now , even if @xmath221^{-1/2}$ ] , @xmath218^{-3/2}/2 $ ] and @xmath222^{-5/2}/4 $ ] blow up at @xmath223 , the quantities that appear in eq . ( [ tracersss ] ) behave well asymptotically ( i.e. they are finite ) as @xmath224 : @xmath225 moreover the quantities that appear in the r.h.s of eq .
( [ fieldeq3 ] ) behave also well asymptotically and give rise to the cosmological constant @xmath226 as we show next . take for instance the @xmath227 component of eq .
( [ fieldeq3 ] ) in vacuum . for our purposes
it is more convenient to take the mixed components .
then @xmath228 now let us analyze the asymptotic behavior of each of the terms at the r.h.s of eq .
( [ grr ] ) : @xmath229 therefore , to leading order we obtain @xmath230 so the r.h.s of eq .
( [ fieldeq3 ] ) is well behaved asymptotically , and it is just a constant which we can precisely identify with effective cosmological constant @xmath226 , with @xmath200 . notice that this constant emerged not only from the last two terms at the r.h.s of eq .
( [ grr ] ) but also from the contribution of the first two , which one would naively think that they do not contribute as @xmath224 asymptotically .
however , as mentioned in sec .
[ sec : theory ] , a closer look shows that one actually has in those two terms something like @xmath70 asymptotically .
this is why it was necessary to perform the correct asymptotic analysis which leads then to contribution @xmath231 due to the first two terms of the r.h.s . of eq .
( [ grr ] ) . of course one can perform the same asymptotic analysis in the full set of equations ( [ tracersss])([deltasss ] ) to find that all of them behave well and consistently as @xmath224 , since from both the l.h.s and the r.h.s one obtains exactly the same behavior .
this is otherwise expected as we have explicitly the exact solution from which one can compute @xmath232 and @xmath233 and @xmath234 to confirm that nothing diverges as @xmath224 .
the definition of this cosmological constant @xmath226 is consistent with the canonical form that the metric coefficients @xmath235 and @xmath236 take in ( [ gfrsz ] ) in these coordinates .
for instance , in terms of @xmath237 they read @xmath238 where @xmath239 .
hence , we conclude that when @xmath72 has a pole precisely at the anti - de sitter point , the cosmological constant @xmath237 does not arise simply from the last term of eq .
( [ grr ] ) , like in the analysis performed in sec .
[ sec : theory ] where @xmath24 .
that analysis was valid provided that as @xmath82 the following two necessary conditions were satisfied : @xmath240 and @xmath160 , which as emphasized above , is not the actual case for this exact solution .
finally , we mention that for @xmath241 , which corresponds to a null cosmological constant , the quantity @xmath242 that appears in eq .
( [ tracersss ] ) vanishes as the solution approaches the asymptotic value @xmath243 , even if @xmath76 .
moreover , the term @xmath244 also vanishes asymptotically since @xmath245 and @xmath246 .
therefore , a posteriori one can understand why eq .
( [ tracersss ] ) is well behaved asymptotically . it is important to stress that in the previous works @xcite the above physical and geometric interpretation of the metric ( [ gfrsz ] ) was completely absent and therefore , its meaning was rather unclear . in the nonvacuum case , some but not all of the aspects discussed above for this exact solution were elucidated @xcite .
so far we have mainly discuss the case @xmath200 .
regarding @xmath186 , the ricci scalar will tend to @xmath165 asymptotically but will never reach this value since well before @xmath74 the cosmological horizon is reached by the solution .
therefore , @xmath247 and @xmath248 .
that is , the possible solution @xmath224 with @xmath249 is never reached asymptotically due to the presence of the cosmological horizon . now , as concerns the trivial solution @xmath23 , the model ( [ frsz ] ) admits the solution @xmath217 , which solves @xmath240 .
notice that @xmath250^{-3/2}/2 $ ] .
therefore the sss solution is given by the metric ( [ sds ] ) where @xmath251 in this case the cosmological constant is given by @xmath29 , instead of @xmath226 .
we emphasize that the solution ( [ sds ] ) is _ not _ approached asymptotically by the solution ( [ gfrsz ] ) , since as we mentioned before , the latter has a deficit angle @xmath239 , while ( [ sds ] ) has @xmath252 . in summary ,
the model ( [ frsz ] ) admits the two exact solutions ( [ gfrsz ] ) and ( [ sds ] ) , one where the ricci scalar is constant everywhere and one where it varies with the radial coordinate @xmath100 .
however , it is important to remark that both solutions are not two different solutions with the same boundary conditions , but two different solutions with different boundary conditions .
even if we put @xmath253 , @xmath202 , one solution still has a deficit angle while the other does not .
moreover , both solutions have different cosmological constants @xmath29 and @xmath254 , and so the spacetimes are not even the same asymptotically when @xmath255 . on the other hand , one could try to make coincide the inner boundaries ( i.e. the event horizons ) of both solutions artificially as well as the value @xmath117 by fixing @xmath2 and @xmath157 , but still both spacetimes would have different global quantities ( mass and charge ) besides the respective deficit angles ( @xmath252 and @xmath239 ) and the respective cosmological constants ( @xmath29 and @xmath254 ) .
namely , one solution would have zero `` adm '' mass and non zero charge , while the other solution would have non - zero `` adm '' mass and zero charge .
two final remarks are in order .
the first one concerns the deficit angle solution ( [ gfrsz ] ) .
it is possible to show that in fact one can obtain the @xmath0 model ( [ frsz ] ) by using a kind of `` reconstruction method '' .
this consists by imposing a solution in the form of eqs .
( [ gfrsz ] ) with ( [ mrexact])([deltafrsz ] ) . that is with @xmath256 and with @xmath1 and @xmath15
written in terms of finite powers of @xmath100 .
the condition @xmath256 leads then to @xmath257 , i.e. , @xmath258 , which in turn provide a differential equation for @xmath0 in terms of @xmath100 , @xmath99 and @xmath103 .
however , given such a power expansion , one can invert and write @xmath259 , to obtain a differential equation for @xmath0 and @xmath15 solely , which when solved provides ( [ frsz ] ) .
this kind of `` tricks '' , which are also common to find exact solutions in gr with exotic sources @xcite , have been applied in modified theories of gravity in the past @xcite .
physically , one usually proceeds in the opposite way we just described .
that is , one constructs or proposes an explicit @xmath0 model in order to fit some observations , for instance , in cosmology , and then one asks if such a model admits or not an exact solution in one or other scenario . at this regard , and as a second remark related with our comment at the end of the previous section , we mention that almost all physically viable @xmath0 models admit the trivial solutions @xmath154 given by the schwarzschild de sitter / anti - de sitter solution eqs .
( [ sds])([delta0 ] ) , where the parameter @xmath165 is replaced by a different constant according to the different @xmath0 models .
for instance , in section [ sec : expmodels ] we will see more specific examples .
these are the kind of trivial solutions that we alluded in section [ sec : theory ] that several authors have systematically reported as something `` special '' in @xmath0 gravity , while we see that they are nothing more than the usual solution found in gr with @xmath260 and @xmath261 .
in particular the mass parameter @xmath157 can be redefined as @xmath262 where @xmath263 can be taken as the fiducial mass associated with the spacetime . moreover , the rn or the kerr
newman bh solutions with or without @xmath29 can also be found in @xmath0 gravity when considering matter with a traceless energy - momentum tensor as these corresponds also to the same trivial solutions @xmath81 , including the af ones when @xmath78 .
we remark that there are more exact solutions for other choices of @xmath264 in the model @xmath265
@xcite , not only for @xmath170 , which corresponds to the model we have just analyzed taking @xmath241 . nevertheless , most of those solutions are still trivial or have exotic asymptotics . in section [ sec : expmodels ] ( cf . model 1 ) we discuss the case @xmath266 , which was not covered in ref .
@xcite , as their equations become singular precisely for @xmath266 .
finding physically interesting exact bh solutions different from the trivial ones proves to be difficult when the @xmath0 model is complicated , like the physically viable models that have passed many cosmological and solar - system tests ( e.g. models 45 of section [ sec : expmodels ] ) .
in particular , if one is interested in genuine af or ads / aads type of spacetimes .
in such an instance , one has then to appeal to a numerical analysis . at this regard
, we stress that we have used the two exact solutions presented in this section as a testbed for a fortran code developed to solve numerically eqs .
( [ tracersss])([deltasss ] ) for more complicated and `` realistic '' @xmath0 models , submitted to suitable boundary ( regularity ) conditions that represent the presence of a black hole ( i.e. when an horizon is present ) .
these _ regularity conditions _ are presented next .
in order to obtain the regularity conditions at the horizon @xmath267 , whether is the event ( inner ) horizon or the cosmological ( outer ) horizon , we expand the variables as follows : @xmath268 where @xmath269 stands for @xmath103 , @xmath99 , @xmath270 . when replacing these expansions in eq .
( [ tracersss ] ) and demanding that the derivatives of these variables are finite at the horizon one obtains after long but straightforward algebra the following regularity condition : @xmath271f_{rr } } } \left|\rule{0mm}{0.6cm}\right._{r = r_h } \,\,\,,\ ] ] where as stressed , all the quantities in this equation are to be evaluated at the horizon @xmath267 , which in principle is any positive value , whether @xmath272 is the event ( inner ) horizon ( denoted by @xmath22 ) or the ( outermost ) cosmological horizon ( @xmath121 ) . in this paper
we will only be interested in finding af solutions , and thus , the spacetime will contain only the event ( inner ) horizon . in turn , eq .
( [ msss ] ) provides : @xmath273 hereafter the quantities evaluated at the horizon will be written with a subindex `` @xmath274 '' , the regularity conditions correspond to a spacetime with a _ regular origin .
_ in this case , @xmath275 , @xmath276 , and @xmath277 .
the so called _ solitons _ are localized field configurations that are _ globally regular _ , including the origin . in gr there
exist examples of spacetimes with matter that allow for _ solitons _ and sssafbh with hair , like in the einstein
yang mills system @xcite and in the einstein - scalar - field system with an asymmetric scalar - field potential @xcite .
it is to be mentioned that regularity conditions of this kind were proposed first in @xcite , and then rectified in @xcite .
those authors did not use exactly the same differential equations ( [ msss ] ) and ( [ deltasss ] ) that we used here , which were a consequence of eqs .
( [ fieldeq3 ] ) and ( [ tracer ] ) . instead , they departed from the much more involved field equation ( [ fieldeq1 ] )
. nevertheless our regularity conditions ( [ rprimereg ] ) and ( [ mprimereg ] ) are equivalent to those of @xcite . in order to illustrate the consistency of these two regularity conditions , we take three @xmath0 models and their exact solutions .
first , @xmath278 , which corresponds to gr plus a cosmological constant . the sss solution is like in ( [ sds ] ) but with @xmath4 instead of @xmath279 in this case eq .
( [ rprimereg ] ) , when multiplied by @xmath72 , gives an identity @xmath280 , since the numerator at the r.h.s @xmath281 . on the other hand , since @xmath282 then @xmath283 .
therefore , @xmath284 . on the other hand , the r.h.s of eq .
( [ mprimereg ] ) yields @xmath285 , which corresponds precisely to the de sitter / anti - de sitter value obtained from eq .
( [ mrexact ] ) .
the second case corresponds to the trivial constant solution discussed in previous sections : @xmath286 , @xmath287 , where @xmath26 given by @xmath288 and such that @xmath159 , @xmath289 and @xmath290 .
then eq .
( [ rprimereg ] ) gives @xmath291 , which is clearly compatible with the trivial solution , whereas the r.h.s of eq .
( [ mprimereg ] ) yields @xmath292 , where @xmath293 .
a less trivial test to our regularity conditions is provided by the exact solution given by eqs .
( [ mrexact])([deltafrsz ] ) . for that solution
it is easy to verify that both sides of eq .
( [ rprimereg ] ) give @xmath294 while eq .
( [ mprimereg ] ) yields @xmath295 in both sides .
now , eqs .
( [ tracersss])([deltasss ] ) do not provide the regularity conditions for @xmath234 and @xmath296 at the horizon .
we require to differentiate eqs .
( [ tracersss ] ) and ( [ deltasss ] ) with respect to @xmath100 one more time . when doing so and when replacing the expansions in the form eq .
( [ fexp ] ) it is possible to obtain the values @xmath297 and @xmath298 .
this is a lengthy but otherwise straightforward calculation . in the appendix
[ sec : regcond2 ] we provide the explicit expressions .
as remarked in the introduction , there are very well known nht s for the einstein - scalar field system , hereafter einstein-@xmath299 system ( i.e. einstein - hilbert gravity minimally coupled to a _
real _ scalar field @xmath299 ) when the potential associated with a scalar field @xmath299 verifies the non - negativity condition @xmath21 . on the other hand
, it is also well known that provided @xmath17 one can map @xmath0 into a scalar - tensor theory ( stt ) by defining the scalar @xmath300 .
this is the so called jordan - frame representation of the @xmath0 theory . under this framework and in the presence of matter , the scalar field @xmath301 turns out to be coupled _
non minimally _ to the curvature but minimally to matter .
a further conformal transformation allows one to write the theory in the so called einstein frame , where a new scalar field @xmath299 with a new potential @xmath20 , is now coupled minimally to the curvature but non - minimally to the matter sector . nonetheless , in the absence of matter , and still under the einstein frame , the theory look exactly as an einstein-@xmath299 system , and thus
, one can turn to the nht s and see whether they apply or not for the @xmath0 model at hand .
since we are interested precisely in finding sss black holes in vacuum , we can thus exploit this equivalence to see if hair is absent or if it can exist .
thus , given a specific @xmath0 model , we carry out the following protocol : 1 ) we check if the model allows a `` trivial '' solution @xmath3 ( leading to schwarzschild solution ) .
if the model does , then it means that the model is a priori able of allowing af hairy solutions ( i.e. solutions where @xmath15 would be a non - trivial function of @xmath100 that interpolates from the event horizon , with value @xmath117 , to spatial infinity where @xmath3 ) .
many @xmath0 models used as geometric dark - energy admit also the trivial solution @xmath3 .
thus , in this paper we focus only in such kind of models and analyze if geometric hair can also exist . in a future work
, we shall analyze the existence of hair in ads or aads black - hole solutions ; 2 ) we write the model as a stt in the einstein frame . if the potential @xmath20 is not negative , then the nht s apply and we conclude that afsss hairy solutions _ can not _ exist in such model .
in fact , if @xmath20 turns to be strictly positive , the schwarzschild solution does not even exist , only the de sitter type of solutions exist at best ; 3 ) if the potential has negative branches then we proceed to analyze if hairy solutions exist by solving numerically the field equations of sec .
[ sec : sss ] under the regularity conditions provided in secs .
[ sec : regcond ] and the appendix [ sec : regcond2 ] . in order to fix the ideas ,
let us now briefly review the different scalar - field transformations required to analyze properly an @xmath0 model under the stt approach .
let us consider for simplicity the following gravitational action without matter @xmath302,\ ] ] where at this point @xmath2 is an auxiliary scalar - field that depends on @xmath15 and @xmath303 with the parameter appearing in the metric ( [ gfrsz ] ) . ] .
now , if @xmath304 ( in particular @xmath305 ) the variation of this action with respect to the metric give rise to field equations equivalent to the original theory ( [ f(r ) ] ) in vacuum provided @xmath306 .
thus the action ( [ qframe ] ) is equivalent to ( [ f(r ) ] ) if we enforce @xmath307 ( see appendix [ sec : bdidentification ] for a further discussion ) .
furthermore , one can introduce the scalar field @xmath308 and write a dynamically equivalent action as @xmath309 where @xmath310 is defined as follows : @xmath311.\ ] ] by _ dynamically equivalent _
, we mean that we obtain field equations that are completely equivalent to the original action ( see appendix [ sec : bdidentification ] ) . to do so , we can treat the scalar - field @xmath301 as metric - independent , and @xmath15 as independent of @xmath301 , and thus take the action as a functional of both @xmath45 and @xmath301 .
thus , variation of this action with respect to the metric leads to field equations equivalent to eq .
( [ fieldeq1 ] ) , while variation with respect to @xmath301 simply leads to @xmath312 .
we see then that the action ( [ jordanf ] ) is equivalent to the action of a brans dicke like theory , with a brans dicke parameter @xmath313 ( implying that the kinetic term associated with the gradients of @xmath301 is absent ) and with a potential @xmath314 ( see appendix [ sec : bdidentification ] for the details ) .
clearly a stt of this sort ( i.e. one with @xmath313 ) but without a potential would be incompatible with the bound @xmath315 @xcite , which is required for the theory to pass the solar system tests , and thus , would be automatically ruled out
. however , the presence of this potential makes possible for certain @xmath0 models to pass those tests even if @xmath313 .
this depends if the potential allows for the emergence of the _ chameleon _ mechanism ( cf .
@xcite ) , but not all the potentials have this property . while we will consider some @xmath0 models that seem to pass such tests thanks to the chameleon mechanism , our main purpose in this paper is to analyze the issue about the existence or absence of hairy bh solutions in such models rather to test their observational viability .
thus , there are models that we use only for that purpose and which may fail the observational tests .
it is said that , the actions ( [ qframe ] ) and ( [ jordanf ] ) are written in the so called jordan frame ( jf ) , where by _ frame _ is to be understood as a set of physical variables .
as we remarked , in reality we have only introduced a new scalar - field variable without changing the metric in any way , thus this formulation of the theory is completely equivalent to the original one ( at least in the sectors where @xmath17 ) , and thus , the field equations are also equivalent . as mentioned before , the energy - momentum tensor of matter is conserved in the jf ( see the appendix [ sec : consemt ] ) , and therefore it is with respect to this frame that point - test particles follow geodesics . furthermore one can define a new scalar field and a conformal metric as follows : @xmath316 in terms of these new variables the gravitational action ( [ jordanf ] ) takes the form @xcite ) we have omitted in the integral the term @xmath317 , which can be converted into a surface term , that we assume to vanish . ] @xmath318,\ ] ] where all the quantities with a _
tilde _ are defined with respect to the metric @xmath319 .
this corresponds precisely to the einstein - hilbert action coupled minimally to a scalar field @xmath299 .
notice that , unlike the jf action , a kinetic term appears due to the conformal transformation .
the action ( [ einsteinf ] ) is written in what is known as the _
einstein frame _ ( ef ) , where the potential @xmath20 is defined as @xmath320)$ ] , which will be given explicitly when the model @xmath0 be provided ( see section [ sec : expmodels ] ) .
if we include the matter action then the scalar field @xmath299 will be coupled non minimally to the matter fields . in this paper
we are only interested in the vacuum case , so the field equations obtained from the action ( [ einsteinf ] ) are simply those of the einstein-@xmath299 system : @xmath321\,\,\,,\\ \label{kgef } \box^{\!\!\!\tilde { } } \,\,\phi & = & \frac{d\mathscr{u}}{d\phi } \,\,\,.\end{aligned}\ ] ] as we emphasized previously , based on this equivalence and in view of our interest in finding sss and af non - trivial black holes in @xmath0 gravity , we have to take into account the nht s
@xcite which are valid when @xmath21 .
the theorems roughly establish that _ whenever the condition @xmath21 holds , given an afsss spacetime containing a black - hole ( with a regular horizon ) within the einstein-@xmath299 system , the only possible solution is the hairless schwarzschild solution .
_ here by _
hairless _ we mean that the scalar field @xmath322 , i.e. , the scalar field is constant everywhere in the domain of outer communication of the bh and it is such that @xmath323 , in order to prevent the presence of a cosmological constant which would spoil the af condition . the nht s can be avoided if the potential has negative branches , notably at the horizon @xcite .
so in our case , given an @xmath0 model , we have only to check if the corresponding potential satisfies or not the condition @xmath21 . in the affirmative case , we conclude that sss and af hairy black holes are absent .
nevertheless , when this condition fails , one usually need to resort to a numerical treatment in order to analyze if a black hole can support scalar hair or not . before concluding this section ,
a final remark is in order .
there is an important relationship between the critical points of the potential @xmath20 , notably the extrema , and those of the `` potential ''
@xmath324 defined such that @xcite(see also appendix [ sec : bdidentification ] ) @xmath325 in section [ sec : theory ] we denoted the extrema of @xmath324 by @xmath26 .
it is straightforward to verify @xmath326 therefore , provided @xmath327 and @xmath328 ( i.e. the conditions for a well defined map to the einstein frame ) , we see that the extrema of @xmath20 correspond precisely to @xmath26 .
however , care must be taken when @xmath329 or @xmath330 , as it may happen in some models that we will encounter in the next section .
when this happens , the conformal transformation ( [ conft ] ) becomes singular or ill defined as @xmath300 . finally , for @xmath0 models where @xmath72 is not strictly positive , notably , where @xmath72 can vanish at some @xmath331 , called _ weak singularity _
model 5 in section [ sec : expmodels ] ) it will be useful to introduce the `` potential ''
@xmath332 defined via @xmath333 the finite or divergent behavior of @xmath334 also provides insight about @xmath335 . for instance , if @xmath334 is finite , it means that @xmath335 vanishes like @xmath336 .
furthermore , @xmath337 can supply further information about the possible trivial solutions @xmath32 . as we discussed in section
[ sec : theory ] and also in section [ sec : exactsols ] , if at @xmath68 where @xmath338 and @xmath339 vanishes , @xmath68 can be one trivial solution of eq .
( [ tracer ] ) in vacuum or more generally , when the matter has a traceless energy - momentum tensor , a solution that can be different from @xmath65 if @xmath340 . in this section
we focus on some @xmath0 models that also admit the trivial solution @xmath3 , and check whether or not they satisfy the condition @xmath21 for which the nht s apply .
the results are summarized in table [ tab : models ] at the end of this section . in order to obtain the ef potential @xmath20 from @xmath310
let us recall the relationships @xmath341 ) \,\,\,,\end{aligned}\ ] ] the first one is obtained from eq .
( [ phichi ] ) , while the second one is a definition .
* model 1 : * @xmath342 , where @xmath264 , @xmath343 and @xmath344 are positive parameters of the theory .
@xmath344 fixes the scale for each @xmath264 and @xmath343 can be chosen to be proportional to @xmath344 .
this model has been thoroughly analyzed in the past in several scenarios ( see @xcite and references therein ) . for this model
, @xmath345 if we focus in the domain @xmath346 , then for @xmath347 , @xmath348 . in fact the value @xmath349 corresponds to a degenerate situation @xmath350 that we discuss below . for @xmath351 , @xmath352 as @xmath353 and
vice versa .
so , @xmath354 for @xmath351 . by inverting the relationship ( [ mod1chir ] ) and using eq .
( [ jfpot ] ) followed by the use of eqs .
( [ chiphi ] ) and ( [ efpot ] ) , the two potentials read @xmath355 defined for @xmath356 , and @xmath357 defined for @xmath358 .
the condition @xmath359 holds if @xmath347 , while for @xmath360 , the potential is @xmath361 , but in both cases the potential does not have minima , at least not for a finite @xmath299 .
moreover , we do not consider @xmath362 because @xmath363 which can give rise to an effective negative gravitational constant .
for @xmath364 , corresponding to gr , @xmath365 as one can see directly from eq .
( [ jfpot ] ) . for this model
the nht s a priori apply since @xmath359 .
the fact that the potential is strictly positive and have no minima imply that the solution @xmath3 can not even exist in the ef .
the point is that for @xmath351 the solution @xmath3 corresponds to @xmath352 ( @xmath366 ) , while for @xmath347 the same solution corresponds to @xmath367 ( @xmath368 ) .
we see then that in both cases the mapping to the stt in the ef is ill defined precisely at @xmath349 where @xmath369 , while @xmath370 as @xmath352 is not even well defined as it turns to be multivalued . ] .
this problem at @xmath349 exacerbates for @xmath266 that we discuss below .
we conclude that for this @xmath0 model afsss simply can not exist under the ef . in the original formulation
, the theory also degenerates at @xmath3 for @xmath347 since @xmath94 .
afsss solutions exist but they are no unique as we are about to see .
as we remarked briefly at the end of section [ sec : theory ] , for this class of @xmath0 models a quite degenerate situation may occur .
to fix the ideas , let us focus on the case @xmath266 in the original formulation , since in the stt approach the maps breakdown at @xmath3 as we just mentioned , given that @xmath371 .
so in this case the field eqs .
( [ fieldeq3 ] ) and ( [ tracer ] ) in vacuum reduce to @xmath372 } \,\,\,,\\
\label{field2r2 } \box r & = & 0 \,\,\,.\end{aligned}\ ] ] where we used @xmath373 , @xmath374 and @xmath375 in eqs .
( [ field1r2 ] ) and ( [ field2r2 ] ) model @xmath376 and @xmath377 .
thus we can expect degenerate solutions ( in the sense described in the main text ) for @xmath378 as well . ] .
therefore we see that @xmath3 is a trivial solution of eq .
( [ field2r2 ] ) . on the other hand , for such trivial solution eq .
( [ field1r2 ] ) is satisfied for any @xmath95 compatible with @xmath379 . for instance
, this can be satisfied for any solution of the metric satisfying the einstein equation @xmath380 , with @xmath381 .
this _ degeneracy _ is somehow remarkable as shows that solutions of the field equations in @xmath0 gravity may not be unique , as illustrated by this simple model . in the afsss scenario ,
one bh solution is clearly the schwarzschild solution , but other solutions are possible @xcite , which would be interesting to know to what kind of matter content they correspond in pure gr . in this example , the afsssbh solutions are unique as concerns the solution @xmath3 , since at the horizon @xmath382 , but they are highly non unique as concerns the metric .
that is , to the trivial solution @xmath3 of eq .
( [ tracersss ] ) , one can associate any solution for the metric satisfying @xmath380 , with @xmath381 , like the schwarzschild solution , the rn solution , a solution within the einstein yang mills system , etc . as emphasized in @xcite , this @xmath0 model provides a specific example showing that a generalization of the birkhoff s theorem similar to the one elucidated in the introduction , simply can not exist in general .
it is enlightening to stress that this degenerate situation in vacuum is in a way similar , but opposite , to the `` degeneracy '' that appears in gr with matter sources : given @xmath383 , eq .
( [ fieldeq1 ] ) or eq .
( [ fieldeq3 ] ) reduce to the einstein field equation , whereas ( [ tracer ] ) reduce to @xmath384 , with @xmath50 .
this means that this equation is satisfied identically regardless of the value @xmath385 , which in general , is not zero because @xmath386 . in other words , in gr the metric
is constrained to satisfy the einstein s field equation , but @xmath15 is not constrained to satisfy any differential equation like eq .
( [ fieldeq1 ] ) .
finally we mention that when @xmath387 , the model with @xmath266 also admits the trivial solution eqs .
( [ sds])([derpotcond ] ) the potential @xmath20 is constant in the ef , thus the trivial solution @xmath388 gives rise to the schwarzschild de sitter solution just like in the original variables . ] .
incidentally , for this model the algebraic condition ( [ derpotcond ] ) is satisfied for any @xmath26 .
that is , @xmath26 emerges as an integration constant independent of the parameters of the model .
therefore the value @xmath389 depends on the assigned value for @xmath26 .
in particular , taking @xmath78 we just recover the usual schwarzschild solution as mentioned above . for any other @xmath390
the model admits the trivial solution @xmath3 , but the solution for the metric is not unique either as the degeneracy emerge as well in a similar way to the case @xmath266 .
* model 2 : * @xmath391 , where @xmath392 is a positive dimensionless constant and @xmath393 is a positive parameter that fixes the _ scale_.
this model was proposed by starobinsky as an alternative to explain the early inflationary period of the universe @xcite . for this model , @xmath394 in principle
the model is defined for @xmath395 . however ,
if we impose @xmath396 , then we require @xmath397 .
in fact if we allow @xmath398 the transformation to stt is not well defined , and the model degenerates at @xmath349 in the original variables as @xmath350 . if we focus on solutions with @xmath399 then @xmath400 .
furthermore , @xmath401 and @xmath72 are both finite at @xmath3 for any @xmath402 . proceeding like in the previous model ,
this one has associated the following potentials @xmath403 the potential @xmath20 is defined for @xmath404 .
the region @xmath405 corresponds to @xmath400 , while @xmath406 corresponds to @xmath407 .
clearly @xmath21 , with a global minimum located at @xmath408 where the potential vanishes ( see figure [ fig : staroinf ] ) .
the model admits the solution @xmath3 corresponding to @xmath409 and @xmath408 .
the trivial solution @xmath3 is the only root of @xmath410 . for this model
the nht s apply , and therefore afsssbh hairy solutions are absent .
although the model is defined for @xmath411 in order to be compatible with inflation , in the context of bh s one can in principle consider @xmath412 as a way to evade the nht s because then @xmath413 ( in that instance the condition @xmath414 , implies @xmath415 and the global minimum becomes a global maximum ) .
it turns out , however , that potentials that are negative around a maximum but vanishes there ( in this case @xmath416 ) sometimes admit non - trivial solutions that seem hairy and af would be unstable and never settle into a stationary configuration . ] .
we shall discuss a numerical example of this kind later , but suffice is to say that the bh solutions that one finds may vanish asymptotically but can have an _ oscillatory _ behavior that make them not genuinely af . in order to illustrate this ,
suppose that asymptotically the metric component @xmath236 behaves like @xmath417 ( where @xmath418 and @xmath419 are some constants , and @xmath420 ) , then the mass function @xmath421 oscillates ( it may even diverge if @xmath422 ) , and thus , it does not really converge to a finite value in the limit @xmath74 , a value that one would identify with the adm mass .
yet @xmath423 as @xmath74 .
thus , for this kind of solutions the spacetime is not authentically af . these arguments can be justified using the following heuristic analysis .
let us consider eq .
( [ kgef ] ) and neglect the non - flat spacetime contributions from the metric .
moreover , expanding @xmath20 around its maximum ( which is equivalent to expand @xmath424 around its minimum ) gives @xmath425 . with these simplifying assumptions
it is easy to see that the sss solution of eq .
( [ kgef ] ) is @xmath426 where @xmath427 , and @xmath428 , @xmath429 are constants .
now , at leading order when @xmath74 , we can take the conformal factor @xmath409 in eq .
( [ conft ] ) and both metrics ( the jordan and the einstein frame metrics ) coincide asymptotically .
thus the energy - density contribution @xmath430 is given by @xmath431}{r^2 } \,\,\,,\ ] ] which is not positive definite .
as a consequence , the mass functions is not positive definite either .
in fact , the mass function behaves asymptotically as @xmath432/m$ ] , which oscillates with @xmath100 and does not converges to a definite value ( the komar mass ) . as we mentioned above
, this heuristic analysis confirms the behavior of the full numerical solution for potentials of this kind .
one such sort of solutions are shown for the model 4 in section [ sec : numerical ] below .
( in units of @xmath433 ; @xmath299 is given in units of @xmath434 ) associated with the model 2 with @xmath435 .
the potential verifies @xmath21 for any @xmath411 and therefore , the nht s apply .
the minimum at @xmath408 is associated with the schwarzschild solution where @xmath436.,width=321,height=188 ] * model 3 : * @xmath437 , where @xmath438 is a dimensionless constant , and @xmath439 is a positive parameter that fixes the scale .
this model was proposed by miranda _
et al . _
@xcite as a possible solution for the accelerated expansion of the universe while being free of singularities during the cosmic evolution .
this model is also viable for constructing relativistic extended objects .
notwithstanding , it has problems at the level of the solar system and grow of structure @xcite .
although it may not be a realistic model in all the scenarios , it is worth consider it in this context due to its simplicity .
the model is defined for @xmath440 and the condition @xmath16 restricts @xmath15 in the range @xmath441 . in particular for @xmath439 and @xmath438 positive , as we assumed for this model , whenever @xmath16 is satisfied the condition @xmath440 is also satisfied .
furthermore @xmath442 is finite at @xmath3 , as well as @xmath443 and @xmath27 . for this model @xmath444 the domain @xmath445 corresponds to @xmath446 , and to @xmath447 .
the potentials are @xmath448 , \\
\mathscr{u}(\phi ) & = & \frac{r_{\ast}}{2\kappa}e^ { -2\sqrt{\frac{2\kappa}{3}}\phi}\bigg [ \alpha_1\ln\bigg ( \frac{\alpha_1 } { 1 - e^{\sqrt{\frac{2\kappa}{3}}\phi}}\bigg ) - e^{\sqrt{\frac{2\kappa}{3}}\phi } + 1 - \alpha_1 \bigg].\end{aligned}\ ] ] the potential @xmath20 is thus defined for @xmath447 .
the solution @xmath3 corresponds to @xmath449 in this model .
however , for @xmath450 one is led to @xmath398 , which by construction is not allowed in the ef frame .
therefore this solution can not be present in that frame for those values of @xmath438 .
in fact , for such values of @xmath438 the potential @xmath20 has a minimum at some @xmath451 , but @xmath452 ( see figure [ fig : mirandau ] ) corresponds to one of the roots @xmath80 of @xmath453 , which leads to a schwarzschild de sitter type of solution . ] . according to our protocol ,
a strictly positive potential can not allow for the schwarzschild solution .
therefore , such solution can not be recovered from the ef approach . as a consequence ,
the nht s also rule out the existence of afsss hairy bh solutions . in the original variables
the trivial solution can be recovered since @xmath3 is a root of @xmath453 , but then the effective gravitational constant @xmath31 becomes negative , as the condition @xmath454 fails at @xmath3 .
now , for @xmath455 , @xmath456 if @xmath368 , which is the minimum , a situation similar to the model 1 . finally , for @xmath457 , the potential satisfies @xmath458 ( see the middle panel of figure [ fig : mirandau ] ) .
in this case the nht s also apply , and the schwarzschild solution in the ef is associated with a local minimum @xmath459 at which @xmath460 .
the trivial solution @xmath461 is associated with the trivial solution @xmath3 .
the overall conclusion for this model is that hair is in general forbidden , i.e. , non - trivial afsssbh solutions @xmath99 attempting to interpolate between @xmath117 and @xmath3 can not exist if @xmath16 . more specifically :
1 ) for @xmath450 the ef representation precludes the presence of hair in afsssbh according to the nht s since the solution @xmath3 does not even exist as the potential @xmath20 never vanishes ( the potential is only defined for @xmath16 ) .
if in the original jf variables one permits the possibility of having negative values for @xmath27 ( which in turn implies negative values for @xmath31 ) , then the trivial solution @xmath3 may exist .
but then when looking for a hairy solution that interpolates from @xmath117 to @xmath3 , the solution may cross the value @xmath462 if @xmath463 , a value which is associated with the _ singularity _ @xmath464
. such hairy solution would be rather pathological if it exists at all .
on the other hand , if we impose the condition @xmath363 , in order to avoid that singularity , then @xmath15 turns to be restricted in the range @xmath465 .
in particular one would require @xmath466 to allow for the solution @xmath3 to exist .
this range for @xmath15 restricts severely the region in which one can look for an optimal shooting @xmath117 ( see section [ sec : numerical ] ) , unless @xmath467 .
all in all , the numerical exploration shows that hairy solutions seem to be absent in all these scenarios ; 2 ) for @xmath457 the nht s apply straightforwardly , and therefore hairy solutions can not exist either .
unlike the previous subclass ( @xmath450 ) , the solution @xmath468 corresponding to @xmath461 [ @xmath460 ] can also be recovered in the ef since for this range of values of @xmath438 the condition @xmath470 is fulfilled .
( in units of @xmath471 ; @xmath299 is given in units of @xmath434 ) associated with the model 3 for different values of @xmath438 in the range @xmath472 .
for such values @xmath20 is strictly positive . in particular for @xmath455 the potential vanishes when @xmath368 . for @xmath466
the potential has a global minimum at @xmath451 which leads to a de sitter type of solutions @xmath461 in the ef and @xmath473 in the jf .
middle panel : potential @xmath20 for five values of @xmath438 in the interval @xmath474 . for @xmath475 ,
the potential satisfies @xmath458 and it vanishes at the global minimum @xmath476 leading to the schwarzschild solution where @xmath461 in the ef and @xmath436 in the jf ( see the right panel ) .
right panel : the function @xmath477 is depicted .
the zeros of this function at @xmath3 , which are mapped to @xmath476 in the ef , lead to the schwarzschild solution .
the zeros at @xmath478 ( in units of @xmath479 ) in principle lead to a schwarzschild anti - de sitter solution , but @xmath480 and @xmath481 , @xmath76 , @xmath482 there.,title="fig:",width=222,height=188 ] ( in units of @xmath471 ; @xmath299 is given in units of @xmath434 ) associated with the model 3 for different values of @xmath438 in the range @xmath472 .
for such values @xmath20 is strictly positive . in particular for @xmath455 the potential vanishes when @xmath368 . for @xmath466 the potential has a global minimum at @xmath451 which leads to a de sitter type of solutions @xmath461 in the ef and @xmath473 in the jf .
middle panel : potential @xmath20 for five values of @xmath438 in the interval @xmath474 . for @xmath475 ,
the potential satisfies @xmath458 and it vanishes at the global minimum @xmath476 leading to the schwarzschild solution where @xmath461 in the ef and @xmath436 in the jf ( see the right panel ) .
right panel : the function @xmath477 is depicted .
the zeros of this function at @xmath3 , which are mapped to @xmath476 in the ef , lead to the schwarzschild solution .
the zeros at @xmath478 ( in units of @xmath479 ) in principle lead to a schwarzschild anti - de sitter solution , but @xmath480 and @xmath481 , @xmath76 , @xmath482 there.,title="fig:",width=222,height=188 ] ( in units of @xmath471 ; @xmath299 is given in units of @xmath434 ) associated with the model 3 for different values of @xmath438 in the range @xmath472 . for such values
@xmath20 is strictly positive . in particular for @xmath455 the potential vanishes when @xmath368 . for @xmath466
the potential has a global minimum at @xmath451 which leads to a de sitter type of solutions @xmath461 in the ef and @xmath473 in the jf .
middle panel : potential @xmath20 for five values of @xmath438 in the interval @xmath474 . for @xmath475 ,
the potential satisfies @xmath458 and it vanishes at the global minimum @xmath476 leading to the schwarzschild solution where @xmath461 in the ef and @xmath436 in the jf ( see the right panel ) .
right panel : the function @xmath477 is depicted .
the zeros of this function at @xmath3 , which are mapped to @xmath476 in the ef , lead to the schwarzschild solution .
the zeros at @xmath478 ( in units of @xmath479 ) in principle lead to a schwarzschild anti - de sitter solution , but @xmath480 and @xmath481 , @xmath76 , @xmath482 there.,title="fig:",width=222,height=188 ] we turn now our attention to three @xmath0 models that have been analyzed recently in cosmology and which are some of the most successful ones as concerns the cosmological and the solar - system tests .
however , as we show below , for two of these models ( i.e. models 5 and 6 ) the potential @xmath20 is not even well defined as it turns to be multivalued .
therefore , the conclusion about the absence or existence of hair in these two models has to be obtained numerically using the original formulation ( sections [ sec : theory ] and [ sec : sss ] ) as opposed to the efstt approach .
* model 4 : * @xmath483 where @xmath484 is a dimensionless constant that usually is taken to be positive for a successful phenomenology and @xmath485 is a positive parameter that fixes the scale .
for instance , @xmath486 ensures @xmath17 and @xmath72 never vanishes at a finite @xmath15 regardless of the sign of @xmath484 .
this exponential model has been analyzed in the past by several authors @xcite .
for this model the scalar field @xmath301 is @xmath487 the condition @xmath16 holds provided @xmath488 , in which case @xmath407 , where @xmath489 as @xmath490 . in particular , for @xmath491 the condition @xmath16 is satisfied if @xmath492 .
the potential @xmath324 has only a global minimum at @xmath3 , i.e. , @xmath493 @xcite , and thus , it leads to the schwarzschild solution ( cf .
the right panel of figure [ fig : expu ] ) . for @xmath494
the potential @xmath324 has a local maximum at @xmath3 , and the potential develops in addition a local minimum at some @xmath495 and a global minimum at some @xmath496 @xcite .
these extrema correspond to trivial solutions @xmath65 associated with the schwarzschild anti - de sitter and schwarzschild de sitter solutions respectively .
notice , however , that @xmath497 since the condition @xmath488 fails at @xmath3
. therefore , in this case the schwarzschild solution has @xmath498 .
since the mapping @xmath499 is defined only for @xmath356 , we conclude that for @xmath494 the condition @xmath488 implies @xmath496 for the efstt approach to be well defined and thus , like in the previous model , one can not recover the solution @xmath3 . as a consequence
, we require the original formulation of the theory to analyze if hairy solution can exist for these values of @xmath484 .
but , again , a hairy solution , if exists , can encounter the singularity at @xmath462 before approaching the asymptotic value @xmath497 , notably if @xmath463 . the potentials are @xmath500\,\,\,,\\ \mathscr{u}(\phi ) & = & = \frac { r_{e}}{2\kappa}e^{-2\sqrt{\frac{2\kappa}{3}}\phi}\bigg [ \big ( e^ { \sqrt{\frac{2\kappa}{3}}\phi } -1 \big)\ln\big ( \frac{\lambda_e}{1-e^ { \sqrt{\frac{2\kappa}{3}}\phi } } \big ) + \lambda_e + e^{\sqrt{\frac{2\kappa}{3}}\phi } - 1 \bigg ] \,\,\,.\end{aligned}\ ] ] which are valid in the domain @xmath407 and @xmath501 , respectively . the potential @xmath458 is depicted in figure [ fig : expu ] ( left panel ) for various values of @xmath502 . for this model the nht s apply . as emphasized above , for @xmath503
the potential @xmath20 is strictly positive , in particular its minimum , and therefore the corresponding solution @xmath3 can not be recovered from the efstt approach , but only the de sitter type of solution @xmath154 which is associated with the minimum of @xmath20 . on the other hand , for @xmath504 the potential vanishes at its minimum which leads to the solution @xmath3 that allows one to recover the schwarzschild solution . if for a moment we dismiss the condition @xmath17 and consider @xmath505 we can evade the nht s because now @xmath20 , defined for @xmath506 , is never positive ( see the middle panel of figure [ fig : expu ] ) .
the new domain is a consequence of eq .
( [ chiexp ] ) which yields @xmath507 .
thus for @xmath508 the scalar field @xmath301 is defined in the domain @xmath509 , which in turns leads to @xmath506 .
the potentials depicted in figure [ fig : expu ] ( middle panel ) suggest that hairy solutions might exist .
for instance , in the jf variables , such solution @xmath99 would interpolate between the horizon @xmath117 and its value @xmath3 at spatial infinity .
it turns out that indeed such solution can be found numerically ( see the next section ) but the spacetime is not authentically asymptotically flat as discussed above within the framework of model 2 .
the the mass @xmath126 never converges to a well defined value as the mass function behaves asymptotically @xmath510 , where @xmath511 and @xmath512 is an oscillating but presumably a bounded function ( cf .
figure [ fig : frexpnohair ] ) .
( in units of @xmath513 ; @xmath299 is given in units of @xmath434 ) associated with the model 4 depicted for three values of @xmath486 .
the potential satisfies @xmath458 for @xmath514 and therefore the nht s apply .
in particular for @xmath515 the potential is strictly positive and has a global minimum which leads to a de sitter type of solution @xmath516 . for
@xmath517 the potential can vanish at the global minimum which is associated with a solution @xmath243 .
the panel displays one such example for @xmath518 ( cf .
right panel ) . for @xmath519
the potential vanishes when @xmath368 .
middle - panel : potential @xmath20 depicted for three values of @xmath505 .
we appreciate that @xmath520 and thus the nht s can be evaded .
the potential vanishes at the global maximum , and therefore the solution @xmath243 exists ( see the right panel ) . in principle
the potential suggests that a hairy bh solution @xmath521 might exist interpolating between some @xmath522 at the horizon and @xmath523 at infinity where @xmath524 .
however , the numerical analysis shows ( cf . figure [ fig : frexpnohair ] )
that such solutions are not genuinely asymptotically flat as the mass function at infinity is not well defined ( i.e. it never converges to a fixed value ) .
the local minimum @xmath525 that appears for each @xmath526 is associated with the anti - de sitter solutions ( see right panel ) .
right panel : the function @xmath527 for four values of @xmath484 shows the zeros at which the trivial solutions @xmath154 exist , including the solution @xmath243 ( corresponding to the places where @xmath20 vanishes as depicted in the left and middle panels ) and the anti - de sitter solutions ( one zero @xmath120 for each @xmath505 is appreciated ) corresponding to @xmath528 where the local minimum @xmath525 as shown in the middle panel.,title="fig:",width=222,height=188 ] ( in units of @xmath513 ; @xmath299 is given in units of @xmath434 ) associated with the model 4 depicted for three values of @xmath486 .
the potential satisfies @xmath458 for @xmath514 and therefore the nht s apply .
in particular for @xmath515 the potential is strictly positive and has a global minimum which leads to a de sitter type of solution @xmath516 . for
@xmath517 the potential can vanish at the global minimum which is associated with a solution @xmath243 .
the panel displays one such example for @xmath518 ( cf .
right panel ) . for @xmath519
the potential vanishes when @xmath368 .
middle - panel : potential @xmath20 depicted for three values of @xmath505 .
we appreciate that @xmath520 and thus the nht s can be evaded .
the potential vanishes at the global maximum , and therefore the solution @xmath243 exists ( see the right panel ) . in principle
the potential suggests that a hairy bh solution @xmath521 might exist interpolating between some @xmath522 at the horizon and @xmath523 at infinity where @xmath524 .
however , the numerical analysis shows ( cf . figure [ fig : frexpnohair ] )
that such solutions are not genuinely asymptotically flat as the mass function at infinity is not well defined ( i.e. it never converges to a fixed value ) .
the local minimum @xmath525 that appears for each @xmath526 is associated with the anti - de sitter solutions ( see right panel ) .
right panel : the function @xmath527 for four values of @xmath484 shows the zeros at which the trivial solutions @xmath154 exist , including the solution @xmath243 ( corresponding to the places where @xmath20 vanishes as depicted in the left and middle panels ) and the anti - de sitter solutions ( one zero @xmath120 for each @xmath505 is appreciated ) corresponding to @xmath528 where the local minimum @xmath525 as shown in the middle panel.,title="fig:",width=222,height=188 ] ( in units of @xmath513 ; @xmath299 is given in units of @xmath434 ) associated with the model 4 depicted for three values of @xmath486 .
the potential satisfies @xmath458 for @xmath514 and therefore the nht s apply . in particular for @xmath515 the potential is strictly positive and has a global minimum which leads to a de sitter type of solution @xmath516 . for
@xmath517 the potential can vanish at the global minimum which is associated with a solution @xmath243 .
the panel displays one such example for @xmath518 ( cf .
right panel ) . for @xmath519
the potential vanishes when @xmath368 .
middle - panel : potential @xmath20 depicted for three values of @xmath505 .
we appreciate that @xmath520 and thus the nht s can be evaded .
the potential vanishes at the global maximum , and therefore the solution @xmath243 exists ( see the right panel ) . in principle
the potential suggests that a hairy bh solution @xmath521 might exist interpolating between some @xmath522 at the horizon and @xmath523 at infinity where @xmath524 .
however , the numerical analysis shows ( cf . figure [ fig : frexpnohair ] )
that such solutions are not genuinely asymptotically flat as the mass function at infinity is not well defined ( i.e. it never converges to a fixed value ) .
the local minimum @xmath525 that appears for each @xmath526 is associated with the anti - de sitter solutions ( see right panel ) .
right panel : the function @xmath527 for four values of @xmath484 shows the zeros at which the trivial solutions @xmath154 exist , including the solution @xmath243 ( corresponding to the places where @xmath20 vanishes as depicted in the left and middle panels ) and the anti - de sitter solutions ( one zero @xmath120 for each @xmath505 is appreciated ) corresponding to @xmath528 where the local minimum @xmath525 as shown in the middle panel.,title="fig:",width=222,height=188 ] * model 5 : * @xmath529 $ ] where @xmath530 is a dimensionless constant , @xmath531 a dimensionless parameter and @xmath532 is a positive parameter that fixes the scale .
this model was proposed by starobinsky @xcite as a mechanism for generating the late accelerating expansion while satisfying several _ local _ observational tests .
we analyzed this and the model 3 in the past in the cosmological setting @xcite and for constructing star - like objects @xcite using the approach of section [ sec : theory ] . in this paper
we take @xmath533 and explore several values of @xmath531 ( see section [ sec : numerical ] ) . for this model
the conditions @xmath16 ( for a positive @xmath31 ) and @xmath17 do not hold in general .
in fact , @xmath72 vanishes at @xmath534 .
since @xmath72 appears in the denominator of eq .
( [ tracersss ] ) , the vanishing of @xmath72 was termed by starobinsky a _
weak singularity_. one can appreciate these features from figure [ fig : starou ] ( right panel ) where the `` potential '' @xmath332 is depicted .
we see that @xmath535 at @xmath536 where @xmath72 vanishes .
thus , the _ weak singularities _ at @xmath536 can not be `` cured '' by the term @xmath537 in @xmath453 because such term does not vanish there , and which otherwise could have lead to a finite @xmath538 . therefore any solution @xmath99 intending to interpolate between @xmath117 and @xmath3 such that @xmath539 or @xmath540 will irremediably encounter the _
weak singularities _ at @xmath536 where we expect a singular behavior in eq .
( [ tracersss ] ) . as a consequence ,
our search for a numerical bh solutions with non - trivial @xmath15 was limited mostly in the range @xmath541 ( see section [ sec : numerical ] ) ( as opposed to the minkowski `` point '' @xmath78 ) in order to recover an effective cosmological constant asymptotically ( in _ time _ ) , and thus , to mimic the dark energy . in that scenario
the actual numerical solution @xmath542 is always positive and larger than @xmath543 , thus , the solution never crosses the _ weak _
singularity @xcite .
something similar takes place for the model 6 . ] . for this model
the potential @xmath324 has several extrema ( see the middle panel of figure [ fig : starou ] ) , in particular , a global minimum at @xmath3 with @xmath544 , @xmath545 , which allows one to recover the schwarzschild solution .
notice that the global minimum at @xmath3 corresponds to the global maximum of @xmath332 .
this is because @xmath546 and @xmath547 is positive , whereas @xmath548 , and so @xmath549 is negative . on the other hand @xmath453 at @xmath536
is well behaved there .
the other extrema , a local maximum and minimum , lead to two schwarzschild de sitter solutions with positive @xmath550 now , the inversion @xmath551 required to recover the potential @xmath310 and then the potential @xmath20 demands @xmath17 or @xmath552 . that is , the inversion is possible when @xmath300 is a monotonic function of @xmath15 , which is not the case for this model . in principle
one could perform the inversion picewise in very specific domains of the model but not in all the domain where the model is defined . in view of this drawback
the potential @xmath20 is not well defined .
in fact it is multivalued as we are about to see .
its expression can not be given in closed form but only in parametric representation through the equations @xmath553^{1+q } } \,\,\,,\\
\phi ( r ) & = & \sqrt{\frac{3}{2\kappa } } { \rm ln}\chi ( r ) \,\,\,,\\
\mathscr{u}(\phi(r ) ) & : = & v(\chi[\phi(r ) ] ) \,\,\,.\end{aligned}\ ] ] the form of the potential @xmath20 is shown in figure [ fig : starou ] ( left panel ) . given that @xmath20 is not single valued it is a priori unclear how to establish a method to solve the differential equations in the efstt approach and decide unambiguously which value of @xmath20 to assign for a given @xmath299 .
hence we conclude that one can not obtain any rigorous result from this frame using this potential , let alone trying to implement the nht s . but
even if we tried to do so , the lower branch of the potential does not satisfy the condition @xmath458 required by the theorem to prevent the existence of hair . in view of this , any strong conclusion about the existence or absence of hair must be obtained from the original formulation of the theory that was presented in section [ sec : sss ] .
furthermore , due to the complexity of the model itself and of the differential equations , a numerical analysis is in order . in the next section we provide the numerical results that show evidence about the absence of hairy afsss black holes in this model .
( in units of @xmath554 ; @xmath299 is given in units of @xmath434 ) associated with the starobinsky model 5 with @xmath555 and @xmath556 .
the potential @xmath20 is multivalued and has negative branches , therefore , the nht s can not be applied .
the arrows indicate the trajectory of the parametric plot for increasing values of @xmath15 ( @xmath15 is in units of @xmath532 ) .
one of the marks ( in gray ) indicate the value @xmath3 corresponding to @xmath408 at which @xmath557 . the second mark ( in blue ) indicates the starting point of the parametric plot at @xmath558 . incidentally , for @xmath490 the field @xmath559 [ cf .
( [ f_rstaro ] ] and the potential returns to its starting point .
middle panel : potential @xmath324 showing explicitly the extrema ( maximum or minimum ) where the trivial solutions @xmath560 exist .
the global minimum at @xmath3 leads to the schwarzschild solution with @xmath436 .
the other extrema are associated with the schwarzschild de sitter solutions @xmath560 . for this particular model
any non - trivial solution @xmath99 interpolating between @xmath117 ( the value at the bh horizon ) and @xmath3 ( the asymptotic value ) is confined within the range @xmath561 which corresponds to @xmath562 , i.e. , values of @xmath15 close to the global minimum at @xmath3 . outside this range
, the _ weak singularities _ can be reached by @xmath15 ( see the right panel ) .
right panel : the function @xmath332 is depicted showing the places where @xmath563 diverges . these places called _ weak singularities _ are located at @xmath564 , where @xmath565 ( these values are denoted generically by @xmath66 in the main text ) . at such values
( [ tracersss ] ) blows up.,title="fig:",width=222,height=188 ] ( in units of @xmath554 ; @xmath299 is given in units of @xmath434 ) associated with the starobinsky model 5 with @xmath555 and @xmath556 .
the potential @xmath20 is multivalued and has negative branches , therefore , the nht s can not be applied .
the arrows indicate the trajectory of the parametric plot for increasing values of @xmath15 ( @xmath15 is in units of @xmath532 ) .
one of the marks ( in gray ) indicate the value @xmath3 corresponding to @xmath408 at which @xmath557 . the second mark ( in blue ) indicates the starting point of the parametric plot at @xmath558 . incidentally , for @xmath490 the field @xmath559 [ cf .
( [ f_rstaro ] ] and the potential returns to its starting point .
middle panel : potential @xmath324 showing explicitly the extrema ( maximum or minimum ) where the trivial solutions @xmath560 exist .
the global minimum at @xmath3 leads to the schwarzschild solution with @xmath436 .
the other extrema are associated with the schwarzschild de sitter solutions @xmath560 . for this particular model
any non - trivial solution @xmath99 interpolating between @xmath117 ( the value at the bh horizon ) and @xmath3 ( the asymptotic value ) is confined within the range @xmath561 which corresponds to @xmath562 , i.e. , values of @xmath15 close to the global minimum at @xmath3 . outside this range
, the _ weak singularities _ can be reached by @xmath15 ( see the right panel ) .
right panel : the function @xmath332 is depicted showing the places where @xmath563 diverges . these places called _ weak singularities _ are located at @xmath564 , where @xmath565 ( these values are denoted generically by @xmath66 in the main text ) . at such values
( [ tracersss ] ) blows up.,title="fig:",width=222,height=188 ] ( in units of @xmath554 ; @xmath299 is given in units of @xmath434 ) associated with the starobinsky model 5 with @xmath555 and @xmath556 .
the potential @xmath20 is multivalued and has negative branches , therefore , the nht s can not be applied .
the arrows indicate the trajectory of the parametric plot for increasing values of @xmath15 ( @xmath15 is in units of @xmath532 ) .
one of the marks ( in gray ) indicate the value @xmath3 corresponding to @xmath408 at which @xmath557 . the second mark ( in blue ) indicates the starting point of the parametric plot at @xmath558 . incidentally , for @xmath490 the field @xmath559 [ cf .
( [ f_rstaro ] ] and the potential returns to its starting point .
middle panel : potential @xmath324 showing explicitly the extrema ( maximum or minimum ) where the trivial solutions @xmath560 exist .
the global minimum at @xmath3 leads to the schwarzschild solution with @xmath436 .
the other extrema are associated with the schwarzschild de sitter solutions @xmath560 . for this particular model
any non - trivial solution @xmath99 interpolating between @xmath117 ( the value at the bh horizon ) and @xmath3 ( the asymptotic value ) is confined within the range @xmath561 which corresponds to @xmath562 , i.e. , values of @xmath15 close to the global minimum at @xmath3 . outside this range
, the _ weak singularities _ can be reached by @xmath15 ( see the right panel ) .
right panel : the function @xmath332 is depicted showing the places where @xmath563 diverges . these places called _ weak singularities _ are located at @xmath564 , where @xmath565 ( these values are denoted generically by @xmath66 in the main text ) . at such values
( [ tracersss ] ) blows up.,title="fig:",width=222,height=188 ] * model 6 : * @xmath566 , where @xmath567 and @xmath392 are two dimensionless constants , and like in previous models , @xmath568 fixes the scale .
this model was proposed by hu and sawicky @xcite , and it is perhaps one of the most thoroughly studied @xmath0 models ( see ref .
@xcite for a review ) . in the cosmological context , @xmath567 and
@xmath392 were fixed as to obtain adequate cosmological observables , like the actual dark and matter content in the universe .
for instance taking @xmath569 , their values are @xmath570 , and @xmath571 @xcite . notice that the model 5 with @xmath572 and this model with @xmath266 are essentially the same . like in the previous model
, the conditions @xmath16 and @xmath17 are not met in general , therefore , the potential @xmath20 is multivalued and has negative branches as well .
it can be plotted using a parametric representation as in the model 5 : @xmath573^{2 } } \,\,\,,\\
\phi ( r ) & = & \sqrt{\frac{3}{2\kappa } } { \rm ln}\chi ( r ) \,\,\,,\\
\mathscr{u}(\phi(r ) ) & : = & v(\chi[\phi(r ) ] ) \,\,\,.\end{aligned}\ ] ] figure [ fig : hsu ] depicts the potential @xmath20 ( left panel ) where one can appreciate the pathological features .
in fact , in this model a _ weak _ singularity @xmath565 is located precisely at @xmath3 , i.e. , the value that @xmath99 should reach asymptotically in the af scenario , and it is also the value corresponding to the ( hairless ) schwarzschild solution that we should be able to recover . nevertheless , and unlike model 5 , this singularity in eq .
( [ tracer ] ) or in eq .
( [ tracersss ] ) disappears for some values of @xmath264 because @xmath538 is finite or vanishes at @xmath3 in this model .
namely , @xmath538 vanishes at @xmath3 for @xmath574 .
thus , in this range of @xmath264 , the model admits the trivial solution @xmath3 .
we do not consider the case @xmath575 as this model reduces to @xmath576 , which amounts to gr plus a cosmological constant .
for @xmath577 we find @xmath578 , which does not even vanish .
therefore , this means that @xmath3 does not solve eq .
( [ tracersss ] ) , not trivially nor asymptotically . for @xmath378 and @xmath579 , there is indeed a _
weak _ singularity at @xmath3 where @xmath580 ( cf . the right panel of figure [ fig : hsu ] ) . for @xmath360
the quantity @xmath581 blows up , thus we consider only @xmath582 , notably , for the numerical analysis of section [ sec : numerical ] .
in this model the potential @xmath324 has an minimum at @xmath3 for any @xmath583 , which allows for the schwarzschild solution whenever @xmath584 at @xmath3 ( i.e. for @xmath585 ) .
however , when @xmath586 at @xmath3 and for which @xmath587 the schwarzschild solution may not even exist . in those situations , non - trivial solutions where @xmath99 vanishes asymptotically will encounter such singularity ( see section [ sec : numerical ] ) . like in the model 5 , any analysis using the ill - defined potential @xmath20 for the hu
sawicky model is not robust .
we then turn to a numerical analysis using the original formulation of the theory .
this is presented in the next section .
( in units of @xmath588 ; @xmath299 is given in units of @xmath434 ) associated with the hu
sawicky model 6 for @xmath569 with the values @xmath567 and @xmath392 as in the main text .
the potential @xmath20 is multivalued and has negative branches and the nht s can not be applied .
the arrows and marks have the same meaning as in figure [ fig : starou ] .
middle panel : potential @xmath324 showing explicitly the extrema where the trivial solutions @xmath560 may exist . at the local minimum @xmath3 occurs a _ weak _ singularity where @xmath565 ( see the right panel ) .
right panel : potential @xmath332 .
two weak singularities where @xmath589 are located at @xmath3 and @xmath590 . in particular , the singularity at @xmath3 precludes the search for numerical af hairy solutions with @xmath591 as @xmath74 since the `` singularity '' is encountered at finite @xmath100.,title="fig:",width=222,height=188 ] ( in units of @xmath588 ; @xmath299 is given in units of @xmath434 ) associated with the hu
sawicky model 6 for @xmath569 with the values @xmath567 and @xmath392 as in the main text .
the potential @xmath20 is multivalued and has negative branches and the nht s can not be applied .
the arrows and marks have the same meaning as in figure [ fig : starou ] .
middle panel : potential @xmath324 showing explicitly the extrema where the trivial solutions @xmath560 may exist . at the local minimum @xmath3 occurs a _ weak _ singularity where @xmath565 ( see the right panel ) .
right panel : potential @xmath332 .
two weak singularities where @xmath589 are located at @xmath3 and @xmath590 . in particular , the singularity at @xmath3 precludes the search for numerical af hairy solutions with @xmath591 as @xmath74 since the `` singularity '' is encountered at finite @xmath100.,title="fig:",width=222,height=188 ] ( in units of @xmath588 ; @xmath299 is given in units of @xmath434 ) associated with the hu
sawicky model 6 for @xmath569 with the values @xmath567 and @xmath392 as in the main text .
the potential @xmath20 is multivalued and has negative branches and the nht s can not be applied .
the arrows and marks have the same meaning as in figure [ fig : starou ] .
middle panel : potential @xmath324 showing explicitly the extrema where the trivial solutions @xmath560 may exist . at the local minimum @xmath3 occurs a _ weak _ singularity where @xmath565 ( see the right panel ) .
right panel : potential @xmath332 .
two weak singularities where @xmath589 are located at @xmath3 and @xmath590 . in particular , the singularity at @xmath3 precludes the search for numerical af hairy solutions with @xmath591 as @xmath74 since the `` singularity '' is encountered at finite @xmath100.,title="fig:",width=222,height=188 ] [ cols="^,^,^,^ " , ] as we discussed in the previous section , in some circumstances it is possible to formulate the original @xmath0 model as a stt in the ef where the scalar - field turns to be coupled minimally to the ef metric but it is subject to a potential @xmath20 .
if this potential verifies the condition @xmath592 , then the nht s apply and , at least in the region where @xmath16 and @xmath17 , we can assert that hair ( where @xmath521 or equivalently @xmath99 are not trivial solutions ) is absent , in which case , the only possible af solutions are at best @xmath593 and @xmath468 .
this conclusion follows for the models 1 - 4 in the sectors where their parameters allow for the @xmath3 solution and led to @xmath592 . on the other hand
, we mentioned that models 2 and 4 can have potentials @xmath20 with negative branches if we allow for the parameters @xmath392 and @xmath484 to be negative .
negative values of such parameters are not usually considered in cosmology , but for the sake of finding hairy solutions , we can contemplate them . because the nht s do not apply when @xmath20 is negative , notably at the horizon ,
the problem of hair reopens when this happens . at this regard ,
several strategies are available to solve it : 1 ) show an explicitly exact afsss black hole solution with hair ; 2 ) prove analytically the absence of it ( i.e. extend the nht s ) ; 3 ) show numerical evidence about one or the other . given that the differential equations presented in section [ sec : sss ] are very involved , strategies 1 or 2 might lead to a dead end , thus we opted for option three . in particular , this strategy seems even the most adequate as concerns the models 5 and 6 , where the potential @xmath20 is not even well defined .
we proceed to solve numerically eqs .
( [ tracersss])([deltasss ] ) subject to the regularity conditions at the horizon provided in sec .
[ sec : regcond ] and in the appendix [ sec : regcond2 ] .
the only free conditions are the value @xmath22 and @xmath117 .
the methodology is roughly as follows .
one starts by fixing the size of the black hole @xmath22 , and then looks for @xmath117 so that @xmath353 as @xmath594 .
this `` boundary - value '' problem is solved using a _ shooting method _
@xcite within a runge kutta algorithm .
we have implemented a similar methodology for constructing star - like objects in @xmath0 gravity in the past @xcite .
numerical solutions with non - trivial hair with _ asymmetric _
( non positive definite ) potentials have been found previously within the einstein-@xmath299 system using similar techniques @xcite . as we will see in the next section , for certain @xmath0 models
is not even necessary to perform a shooting as the dynamics of @xmath15 naturally drives @xmath353 asymptotically for a given @xmath117 .
now , as we mentioned previously , for the af solutions to exist , it is not sufficient that @xmath353 as @xmath594 . in section [ sec : exactsols ]
we analyzed one exact solution where this happens precisely , and yet , the solution is not af but has a deficit angle . in that case
the mass function @xmath103 diverges at least linearly with @xmath100 .
it is then crucial to ensure that the mass function converges to a constant value ( that we assume to be the komar or , equivalently , the adm mass ) in order to claim for a genuinely af solution . as a matter of fact
, we used that exact solution as a testbed for our code .
that is , we took the model eq .
( [ frsz ] ) as _ input _ and recovered numerically the exact solution provided by eqs .
( [ gfrsz])([deltafrsz ] ) , notably for @xmath595 .
notice that in this case @xmath99 is not trivial .
figure ( [ fig : frsz ] ) depicts the analytic and the numerical solutions superposed , showing an excellent agreement between the two .
typical numerical errors are depicted in figure [ fig : relerr ] .
we also checked that the trivial solutions @xmath154 that exist in several of the models 1 - 5 were recovered numerically when starting with @xmath596 and which lead to the _ hairless _ kottler schwarzschild de sitter solutions , including the plain af schwarzschild solution when @xmath28 .
additionally we devised other _
internal _ tests to verify the consistency of our code .
these tests are similar to those implemented in our analysis of star - like objects @xcite , and are independent of the fact that exact solutions are available or not .
let us turn our attention to the specific models that deserved a detailed numerical exploration .
( the exact and numerical solutions are superposed ) computed using the model eq .
( [ frsz ] ) for @xmath168 ( i.e. null cosmological constant ) .
the ricci scalar is not trivial and vanishes asymptotically , however , the spacetime is not exactly asymptotically flat but has a deficit angle [ see eqs .
( [ gfrsz])([deltafrsz ] ) ] . at the horizon
@xmath22 the ricci scalar satisfies the regularity conditions .
middle - panel : the mass function @xmath103 is not constant but grows linearly with the coordinate @xmath100 due to the deficit angle .
right - panel : metric components @xmath597 , @xmath236 and their product @xmath598 . in the middle and right panels
the exact and numerical solutions are superposed as well ( cf . figure [ fig : relerr]).,title="fig:",width=222,height=188 ] ( the exact and numerical solutions are superposed ) computed using the model eq .
( [ frsz ] ) for @xmath168 ( i.e. null cosmological constant ) .
the ricci scalar is not trivial and vanishes asymptotically , however , the spacetime is not exactly asymptotically flat but has a deficit angle [ see eqs .
( [ gfrsz])([deltafrsz ] ) ] . at the horizon
@xmath22 the ricci scalar satisfies the regularity conditions .
middle - panel : the mass function @xmath103 is not constant but grows linearly with the coordinate @xmath100 due to the deficit angle .
right - panel : metric components @xmath597 , @xmath236 and their product @xmath598 . in the middle and right panels
the exact and numerical solutions are superposed as well ( cf . figure [ fig : relerr]).,title="fig:",width=222,height=188 ] ( the exact and numerical solutions are superposed ) computed using the model eq .
( [ frsz ] ) for @xmath168 ( i.e. null cosmological constant ) .
the ricci scalar is not trivial and vanishes asymptotically , however , the spacetime is not exactly asymptotically flat but has a deficit angle [ see eqs .
( [ gfrsz])([deltafrsz ] ) ] . at the horizon
@xmath22 the ricci scalar satisfies the regularity conditions .
middle - panel : the mass function @xmath103 is not constant but grows linearly with the coordinate @xmath100 due to the deficit angle .
right - panel : metric components @xmath597 , @xmath236 and their product @xmath598 . in the middle and right panels
the exact and numerical solutions are superposed as well ( cf . figure [ fig : relerr]).,title="fig:",width=222,height=188 ] as depicted in figure [ fig : frsz ] ( left panel ) .
similar relative errors ( not depicted ) are found for the mass function @xmath103 and the metric components . , width=245,height=188 ] * model 4 : * we consider the model 4 with @xmath505 . in this sector of @xmath484
the potential @xmath20 is not positive definite and the model may admit hairy solutions because the nht s do not apply .
notwithstanding , the only solutions with a non - trivial ricci scalar @xmath99 that we find numerically are not exactly af .
the ricci scalar vanishes asymptotically in an oscillating fashion as @xmath594 , but the mass function @xmath103 does not converge but oscillates as well and grows unboundedly as @xmath599 with @xmath600 ( see figure [ fig : frexpnohair ] ) . this behavior is similar to the one provided by the heuristic analysis within the model 2 , except that here we take into account the full system of equations . despite such behavior
the metric components , which depend on @xmath103 , remain bounded in the asymptotic region .
this can be partially understood by looking to @xmath601 , and realize that for @xmath602 the non - oscillating part of this metric component behaves like @xmath603 .
so if @xmath604 , that part of @xmath236 may converge to @xmath605 very slowly , so slowly that one can not even notice it by looking to the numerical outcome .
this behavior seems to be generic for any @xmath505 and any @xmath117 .
the conclusion is that we do not find any genuinely afsss black hole solution in this model . .
left - panel : ricci scalar for three different values of @xmath117 .
the solutions vanish asymptotically .
middle panel : the mass function @xmath103 corresponds to the solution of @xmath15 with @xmath606 shown in the left panel . similar plots for @xmath103 are found for the other two solutions of @xmath15 . the non - converging behavior of @xmath103 as @xmath594 indicates that the spacetime is not af .
right panel : metric components for the solution depicted in the middle panel . the metric components @xmath597 and @xmath236 and their product @xmath598 are bounded but oscillate as @xmath594 corroborating that the resulting spacetime is not af.,title="fig:",width=222,height=188 ] .
left - panel : ricci scalar for three different values of @xmath117 .
the solutions vanish asymptotically .
middle panel : the mass function @xmath103 corresponds to the solution of @xmath15 with @xmath606 shown in the left panel . similar plots for @xmath103 are found for the other two solutions of @xmath15 . the non - converging behavior of @xmath103 as @xmath594 indicates that the spacetime is not af .
right panel : metric components for the solution depicted in the middle panel . the metric components @xmath597 and @xmath236 and their product @xmath598 are bounded but oscillate as @xmath594 corroborating that the resulting spacetime is not af.,title="fig:",width=222,height=188 ] .
left - panel : ricci scalar for three different values of @xmath117 .
the solutions vanish asymptotically .
middle panel : the mass function @xmath103 corresponds to the solution of @xmath15 with @xmath606 shown in the left panel . similar plots for @xmath103 are found for the other two solutions of @xmath15 . the non - converging behavior of @xmath103 as @xmath594 indicates that the spacetime is not af .
right panel : metric components for the solution depicted in the middle panel . the metric components @xmath597 and @xmath236 and their product @xmath598 are bounded but oscillate as @xmath594 corroborating that the resulting spacetime is not af.,title="fig:",width=222,height=188 ] finally , let us focus on the models 5 and 6 that led to pathological potentials in the efstt description . * model 5 : * for the starobinsky model we limit our search for a shooting value @xmath117 first in the region @xmath607 and then in @xmath608 in order to avoid crossing the _ weak _ singularities at @xmath609 when @xmath99 tries to reach the asymptotic value @xmath3 .
we never found a successful shooting parameter leading to an authentic asymptotically flat solution .
two examples of this kind of solutions are depicted in figures [ fig : starono - hairoscillating ] and [ fig : starono - hair ] .
figure [ fig : starono - hairoscillating ] shows that for @xmath556 the solutions are similar to the exponential model 4 with @xmath610 depicted in figure [ fig : frexpnohair ] .
thus , the asymptotic behavior does not correspond to an af spacetime . for @xmath611 , we find situations where the ricci scalar decreases monotonically to a constant value without oscillating as one can see in the left panel of figure [ fig : starono - hair ] . however , this constant is not related with the trivial solution @xmath154 which is the solution of the algebraic equation @xmath612 . in fact
, what happens is that @xmath613 as @xmath74 , as we can see from the middle panel of figure [ fig : starono - hair ] , and also @xmath614 , therefore , by looking at eq .
( [ tracersss ] ) , we appreciate that the combination @xmath615/(1 - 2m / r)$ ] goes to zero even if @xmath616 . in this way
, we see that @xmath617 and @xmath618 , while @xmath619 asymptotically , which solves eq .
( [ tracersss ] ) .
this behavior of @xmath103 explains why the metric components vanish asymptotically ( see the right panel of figure [ fig : starono - hair ] ) . before vanishing
we see that @xmath620 at @xmath621 precisely where @xmath622 . at this value of @xmath100 ,
the component @xmath623 .
thus , we conclude that the af behavior is not recovered either for this and other values @xmath624 and different @xmath117 . in summary , we find strong numerical evidence that for the starobinsky model 5 afsssbh with geometric hair do no exist .
this conclusion is obtained by changing the parameters in several combinations as well as the values @xmath117 . * model 6 : * the numerical analysis of the hu
sawicky model 6 requires more care because @xmath72 vanishes at @xmath3 for several values of @xmath264 ( @xmath625 ) .
the _ weak _ singularity where @xmath626 diverges can be reached where @xmath565 , except if @xmath537 vanishes at the same @xmath15 .
in fact for @xmath625 and @xmath579 the quantity @xmath627 always diverges . in figure
[ fig : hsu ] ( right panel ) , one can appreciate this divergence for @xmath569 . for @xmath577
the quantity @xmath627 remains finite , but @xmath628 , so @xmath3 can not be a possible asymptotic solution of eq .
( [ tracersss ] ) .
thus , the case @xmath577 is irrelevant for af solutions . in consequence , for @xmath625 and @xmath579 ,
the _ weak _ singularity is approached as @xmath15 tries to reach its asymptotic value and at this point eq .
( [ tracersss ] ) becomes singular .
therefore , in this range of @xmath264 we never find a well behaved solution regardless of the shooting parameter @xmath117 and @xmath15 systematically shows a divergent behavior in the numerical solutions
. in the interval @xmath629 the quantity @xmath627 vanishes and we do not find any pathologies as @xmath353 .
the numerical solutions are very similar to the oscillating solutions found in the starobinsky model 5 and the exponential model 4 .
figure [ fig : hs - nohairoscillating ] shows a prototype of such solutions . as we remarked before , for @xmath266 the hu
sawicky model is essentially the starobinsky model with @xmath572 , thus it not surprising to find , at least for these values of @xmath264 , such an oscillating behavior for all of our trial values @xmath117 .
our conclusions seem to be insensitive for several values of the constants @xmath567 and @xmath392 .
hence , the numerical evidence indicates that af solutions with non - trivial hair are absent as well in the hu
sawicky model . and @xmath533 .
left panel : ricci scalar for three values of @xmath117 ( including @xmath630 leading to the trivial solution @xmath436 ) .
@xmath591 asymptotically , however , for the non - trivial solutions ( dotted lines ) the mass function @xmath103 does not converge ( see the middle panel ) .
middle panel : mass function @xmath103 associated with the solution of @xmath15 with @xmath631 as shown in the left panel .
similar plots for @xmath103 are found , but not depicted , for the other non - trivial solution of @xmath15 , while @xmath632 when @xmath436 , corresponding to the schwarzschild solution .
right panel : metric components associated with the solution of middle panel . the metric components @xmath597 and @xmath236 and their product @xmath598 oscillate as @xmath594 corroborating that the resulting spacetime is not af . here
@xmath15 is given in units of @xmath633 , and @xmath1 and @xmath100 in units of @xmath634 ( @xmath635).,title="fig:",width=222,height=188 ] and @xmath533 .
left panel : ricci scalar for three values of @xmath117 ( including @xmath630 leading to the trivial solution @xmath436 ) .
@xmath591 asymptotically , however , for the non - trivial solutions ( dotted lines ) the mass function @xmath103 does not converge ( see the middle panel ) .
middle panel : mass function @xmath103 associated with the solution of @xmath15 with @xmath631 as shown in the left panel .
similar plots for @xmath103 are found , but not depicted , for the other non - trivial solution of @xmath15 , while @xmath632 when @xmath436 , corresponding to the schwarzschild solution .
right panel : metric components associated with the solution of middle panel . the metric components @xmath597 and @xmath236 and their product @xmath598 oscillate as @xmath594 corroborating that the resulting spacetime is not af . here
@xmath15 is given in units of @xmath633 , and @xmath1 and @xmath100 in units of @xmath634 ( @xmath635).,title="fig:",width=222,height=188 ] and @xmath533 .
left panel : ricci scalar for three values of @xmath117 ( including @xmath630 leading to the trivial solution @xmath436 ) .
@xmath591 asymptotically , however , for the non - trivial solutions ( dotted lines ) the mass function @xmath103 does not converge ( see the middle panel ) .
middle panel : mass function @xmath103 associated with the solution of @xmath15 with @xmath631 as shown in the left panel .
similar plots for @xmath103 are found , but not depicted , for the other non - trivial solution of @xmath15 , while @xmath632 when @xmath436 , corresponding to the schwarzschild solution .
right panel : metric components associated with the solution of middle panel . the metric components @xmath597 and @xmath236 and their product @xmath598 oscillate as @xmath594 corroborating that the resulting spacetime is not af . here
@xmath15 is given in units of @xmath633 , and @xmath1 and @xmath100 in units of @xmath634 ( @xmath635).,title="fig:",width=222,height=188 ] and @xmath533 .
left panel : the numerical solution shows that the spacetime is not af as @xmath99 does not vanish asymptotically .
it has rather a spurious de sitter behavior when @xmath636 asymptotically .
middle panel : the mass function @xmath103 does not converge asymptotically but decreases to a very large _ negative _ values .
this is even opposite to the growing behavior @xmath637 that should be expected if the spacetime were genuinely asymptotically de sitter .
right panel : the metric components @xmath597 , @xmath236 and their product @xmath598 vanish asymptotically .
this behavior confirms that the spacetime is not even genuinely asymptotically de sitter where a cosmological horizon is expected at @xmath638 where @xmath639 and @xmath640.,title="fig:",width=222,height=188 ] and @xmath533 .
left panel : the numerical solution shows that the spacetime is not af as @xmath99 does not vanish asymptotically .
it has rather a spurious de sitter behavior when @xmath636 asymptotically .
middle panel : the mass function @xmath103 does not converge asymptotically but decreases to a very large _ negative _ values .
this is even opposite to the growing behavior @xmath637 that should be expected if the spacetime were genuinely asymptotically de sitter .
right panel : the metric components @xmath597 , @xmath236 and their product @xmath598 vanish asymptotically .
this behavior confirms that the spacetime is not even genuinely asymptotically de sitter where a cosmological horizon is expected at @xmath638 where @xmath639 and @xmath640.,title="fig:",width=222,height=188 ] and @xmath533 .
left panel : the numerical solution shows that the spacetime is not af as @xmath99 does not vanish asymptotically .
it has rather a spurious de sitter behavior when @xmath636 asymptotically .
middle panel : the mass function @xmath103 does not converge asymptotically but decreases to a very large _ negative _ values .
this is even opposite to the growing behavior @xmath637 that should be expected if the spacetime were genuinely asymptotically de sitter .
right panel : the metric components @xmath597 , @xmath236 and their product @xmath598 vanish asymptotically .
this behavior confirms that the spacetime is not even genuinely asymptotically de sitter where a cosmological horizon is expected at @xmath638 where @xmath639 and @xmath640.,title="fig:",width=222,height=188 ] and and @xmath641 and @xmath642 .
the units are like in figure [ fig : starono - hair ] but with @xmath568 instead of @xmath633 .
the description of panels is similar to figure [ fig : starono - hairoscillating].,title="fig:",width=222,height=188 ] and and @xmath641 and @xmath642 .
the units are like in figure [ fig : starono - hair ] but with @xmath568 instead of @xmath633 .
the description of panels is similar to figure [ fig : starono - hairoscillating].,title="fig:",width=222,height=188 ] and and @xmath641 and @xmath642 .
the units are like in figure [ fig : starono - hair ] but with @xmath568 instead of @xmath633 .
the description of panels is similar to figure [ fig : starono - hairoscillating].,title="fig:",width=222,height=188 ]
we have argued that generically @xmath0 gravity contains trivial solutions @xmath643 .
this property allows to mimic an effective cosmological constant that can produce the observed accelerated expansion in the universe . in the context of black - holes (
stationary and axisymmetric , or static and spherically symmetric ) the same property allows to find the same solutions known in gr ( under the same symmetries ) with or without a cosmological constant .
the only difference between both type of solutions ( i.e. the solutions found in one or the other theory ) is that the two fundamental constants involved ( the newton s gravitational constant and the cosmological constant ) are replaced by the effective ones , @xmath31 and @xmath29 in @xmath0 gravity .
therefore , all solutions of this kind reported in the literature within the framework of @xmath0 gravity do not provide any deeper knowledge than the ones we already know in gr .
we then focused in the problem of finding non - trivial ( hairy ) afsss black hole solutions where the ricci scalar is _ not _ constant in the domain of outer communication of the black hole but varies with the radial coordinate @xmath100 .
within that aim we provided the equations to study this scenario for an arbitrary @xmath0 model and derived the conditions for a regular black hole .
we then proceeded to analyze some specific models . prior to a thorough numerical analysis
, we studied the models under the scalar - tensor approach within the einstein - frame , stressing that in vacuum @xmath0 gravity takes the same form as in the einstein - scalar - field system .
therefore , this method makes possible to check if the available no - hair theorems for afsss can be applied , and thus , sparing an unnecessary numerical effort .
thus , for the models where the nht s apply , we concluded that _ geometric _ hair is absent . in those situations
the numerically analysis simply confirms the nht s . for the cases where the resulting scalar - field potential @xmath20 does not satisfy the condition @xmath458 required by the nht s , we turned to a detailed numerical study .
our conclusion is that we did not find any such hair in any of the models where the nht s do not apply .
we also discussed some exact solutions that seem to represent af hairy black holes , however , we showed that in fact the @xmath100 dependence in the ricci scalar is due to the presence of a deficit angle .
therefore such solutions are not really af .
similarly we report numerical solutions that are _ not _ genuinely af as the the mass function never converges to the komar mass , a feature necessary to prove asymptotic flatness .
some of the solutions have an asymptotic oscillatory behavior but without encountering any singularity , while others produce singularities in the equations at a finite @xmath100 .
it remains thus an open question to determine if hair exist in @xmath0 gravity . in order to settle the question in the affirmative it is tantalizing to depart from the einstein-@xmath299 system with a potential that has negative branches and that allow for hairy black holes @xcite , and
then perform an `` inverse '' conformal transformation to obtain an @xmath0 model .
however , in practice this seems to be difficult in a closed explicit form .
moreover , even if this is possible , the resulting @xmath0 model would be rather artificial and would need to be submitted to the usual test ( i.e. cosmological , solar system , binary pulsar , etc . ) to be better motivated physically , irrespective of the issue of hair .
some of the exact black hole solutions that have been reported in the literature ( including in general relativity with exotic energy - momentum tensors ) have been obtained using similar `` tricks '' or ad hoc confections , but presented afterwards in the more _ logical _ direction within the aim of enhance their merit .
this is not the exception in @xmath0 gravity , where an @xmath0 model can be deduced by demanding the existence of some kind of exact solution , most of time , deprived of actual physical interest .
for instance , one can impose @xmath256 ( in area coordinates ) , which is valid provided @xmath644 , and then obtain a differential equation for @xmath0 which can be solved if the assumed form for @xmath645 is simple enough .
in fact , by using this method we were able to recover the model @xmath646 given by ( [ gfrsz ] ) .
notice however that such solution was not af , and thus , it can not be used as a counterexample to the no - hair conjecture in @xmath0 gravity . in this paper
we limited ourselves to the case of af spacetimes . in a future investigation
we will analyze the case of hair in asymptotically de sitter and anti - de sitter spacetimes .
for such spacetimes the no - hair theorems for the af scenario require amendments due the different asymptotic conditions , and the numerical treatment , although similar to the one presented in this work , is sufficiently different to require a detailed and separate analysis .
m.s . is supported partially by dgapa
papiit unam grant in107113 and sep conacyt grants cb-166656 and cb-239639 ; l. g. j. acknowledges conacyt postdoctoral fellowship 236937 and also l. amendola and the cosmology group at the university of heidelberg for their hospitality .
the simplest way to show the conservation equation @xmath647 in @xmath0 metric gravity is departing from eq .
( [ fieldeq1 ] ) written as : @xmath648 one can use the commutation relation @xmath649 since @xmath27 is a scalar function ( i.e. in @xmath0 metric gravity we assume the absence of torsion : the l.h.s of eq .
( [ fieldeq1ap ] ) must be a symmetric tensor ) .
so , applying the covariant derivative @xmath650 to eq .
( [ fieldeq1ap ] ) we obtain @xmath651 where @xmath652 and @xmath653 were used . according to the bianchi identities @xmath654 therefore the first and third terms of l.h.s of eq .
( [ divfieldeq1 ] ) cancel out , yielding @xmath655 furthermore @xmath656 this result implies that the l.h.s of eq .
( [ divfieldeq2 ] ) vanishes identically , leading then to @xmath657 we can summarize this result as a theorem : _ in @xmath0 metric gravity ( under the same basic axioms assumed in gr concerning the manifold and the metric ) the generalized tensor @xmath658 obeys a generalized bianchi identity _
in principle the regularity conditions at the horizon ( [ rprimereg ] ) , ( [ mprimereg ] ) together with @xmath660 , @xmath661 , @xmath662 , seem enough data to solve the system of ordinary differential eqs .
( [ tracersss ] ) , ( [ msss ] ) , and ( [ deltasss ] ) .
nevertheless , in view of the method that we devised to solve these equations numerically , care must be taken when evaluating at the horizon the right - hand - side of eqs .
( [ tracersss ] ) and ( [ deltasss ] ) .
therefore at this place we also need to know the values of @xmath297 and @xmath298 at the horizon when starting the numerical integration .
if we do not impose these supplementary regularity conditions a mild numerical error is made at @xmath272 , but we want to avoid this and impose also the right conditions on @xmath297 and @xmath298 . as we mentioned in sec . [
sec : regcond ] , in order to find these two conditions we require first to obtain an expression for @xmath663 and @xmath664 , then as before , develop the quantities around @xmath272 , and finally impose that @xmath663 and @xmath664 are finite at the horizon .
we simply provide the final outcome of this process , and the regularity conditions turn out to be as follows : @xmath665 + r^{2}f f^{2}_{r}\big [ 10f -r\big ( 13f_{r } -4rf_{rr}\big ) \big ] \nonumber\\ & + & 2f^{3}_{r}\big [ 6f + rf_{r } \big ( 2r^{2}r - 3\big ) \big]\bigg\ } \bigg\{f_{rr}\big [ 2f_{r}\big ( 3 - r^{2}r\big ) + r^{2}f\big]\big[\big(-41r^{2}f + 2f_{r } \big ( 13r^{2}r - 63\big ) \big)r^{2}rf_{r } \nonumber\\ & + & 2r^{2}\big(7r^{2}f^{2 } + 18f_{r } \big ( \frac{4 f_{r}}{r^{2 } } + 3 f\big ) \big ) \big]\bigg\}^{-1 } \left|\rule{0mm}{0.6cm}\right._{r = r_h } \,\,\,,\end{aligned}\ ] ] @xmath666 + f^{2}_{r}\big [ r^{2}rf_{r}f_{rrr } \big ( 2f_{r } \big ( 63r -\frac{72}{r^{2 } } - 13 r^{2}r^{2}\big ) -3f \big ( 120 - 31 r^{2}r\big )
\big ) \nonumber \\ & + & r^{2}f_{r}f_{rr}\big ( 3f_{r}\big ( 22r - 5r^{2}r^{2 } - \frac{24}{r^{2}}\big ) -3f\big ( 20 - 9 r^{2}r\big ) \big ) + 288f f_{r}f_{rrr } + 3r^{2}f f^{2}_{rr}\big ( 22r - \frac{24}{r^{2 } } - 5r^{2 } r^{2}\big ) \big]\bigg\}\nonumber \\
& \times & \bigg\ { f^{3}_{rr}\big[14r^{2}f + f_{r } \big ( 24 - 13 r^{2}r\big ) \big ] \big[r^{2}f + 2f_{r } \big(3- r^{2}r \big ) \big]^{3 } \bigg\}^{-1 } \left|\rule{0mm}{0.6cm}\right._{r = r_h } \,\,.\end{aligned}\ ] ] given that these two regularity conditions are quite involved , we performed minimal tests to prove its validity .
like in sec .
[ sec : regcond ] we used the exact solutions provided in section [ sec : sss ] .
first we used the solution eqs .
( [ mrtriv])([derpotcond ] ) , notably for the model @xmath667 , and then we considered the model ( [ frsz ] ) with the solution provided by eqs .
( [ mrexact])([deltafrsz ] ) . in both solutions @xmath668 .
this means that the r.h.s of eq .
( [ regdeltaprime ] ) must vanish in the two cases .
we verified that indeed this happens . on the other hand ,
the exact solution @xmath669 obtained from the model @xmath667 leads simply to @xmath670 , whereas @xmath671 for the model ( [ frsz ] ) . in both cases we checked that the r.h.s of eq .
( [ regrbiprime ] ) gives respectively these two values at the horizon .
we are therefore confident that our expressions are correct and so we enforced them in the numerical treatment presented in section [ sec : numerical ] .
when introducing the action ( [ qframe ] ) in this way , we can recognize two things : 1 ) let us consider the functions @xmath672 $ ] , and @xmath673 , where @xmath674 , and in this appendix a _
_ indicates differentiation with respect to the argument of the corresponding function .
so , assuming @xmath675 to be a convex function , i.e. , @xmath676 , then clearly @xmath677 if @xmath678 , and the inverse of this function @xmath679 allows to define @xmath680 . in this way we see that @xmath681 is no other than the legendre transformation of @xmath675 , and @xmath682 .
moreover , @xmath683 defines in turn the legendre transformation of @xmath681 ( at least in the region where @xmath684 ) which in this case the condition @xmath685 , simply leads to @xmath686 .
2 ) the condition for the second legendre transformation can be imposed in the action by considering @xmath678 as a lagrange multiplier , so that the variation with respect to @xmath301 leads to @xmath312 .
this is the formal construction when treating @xmath0 theories as stt , and in practice it is achieved by taking the action ( [ jordanf ] ) . in the following
we perform explicitly the transformation between the original @xmath0 theory and the special class of brans dicke model @xmath313 supplemented with a potential .
the brans dicke action with a potential @xmath687 is given by@xcite @xmath688= \frac{1}{2\kappa}\int \!\ !
d^4 x \ :
\sqrt{-g } \big [ \phi r - \frac{\omega_{\rm bd}(\phi)}{\phi}g^{ab}(\nabla_a\phi)(\nabla_b\phi ) -w(\phi)\big ] + i_{\rm matt}[g_{ab } , { \mbox{\boldmath{$\psi$ } } } ] \,\,\,.\ ] ] for comparison with the @xmath0 model we shall focus only on the case @xmath313 .
thus , the field equations read @xcite @xmath689 on the other hand when introducing @xmath300 eq . ( [ fieldeq1 ] ) simply reads @xmath690 where @xmath691 .
moreover , this equation can be written as @xmath692\,\,\,,\ ] ] where we have made explicit the functional dependence @xmath693 , which means that if @xmath694 , one can in principle invert the definition @xmath695 , and obtain @xmath693 , and thus @xmath691 . therefore ,
if we choose the potential @xmath696 and identify @xmath697 , then eq .
( [ ebdfield ] ) becomes exactly eq .
( [ fieldeq1jf2 ] ) . moreover , the trace of eq .
( [ fieldeq1jf2 ] ) reads @xmath698 with the above identification of the fields @xmath699 and the potential @xmath700 we appreciate that eq . ( [ phibdfield ] ) also becomes eq .
( [ chifield ] ) where one can easily verify that the expression @xmath701 coincides exactly with @xmath702 .
henceforth , we conclude that @xmath0 theory is equivalent to a brans dicke theory with @xmath313 and a potential @xmath703 .
sss spacetimes with zero charge that are asymptotically flat except for a deficit angle have the asymptotic form @xmath704 after a redefinition of coordinates @xmath705 , and @xmath706 the metric acquires the standard angle - deficit form @xmath707 under this parametrization the coefficient @xmath708 is then identified with the adm mass associated with this kind of spacetimes @xcite . in the same way
, an sss metric which is ads or aads with a deficit angle @xmath709 can be transformed into @xmath710 notice that the cosmological constant @xmath4 did not require to be redefined in order to obtain the standard metric ( [ ssslambda2 ] ) .
we then need @xmath179 , @xmath239 and @xmath711 in order to recover the metric ( [ gfrsz ] ) in the standard form when @xmath202 .
+ finally , the metric of an sss that is ads or aads with a deficit angle and endowed with a charge @xmath531 @xmath712 is transformed into @xmath713 taking @xmath714 .
the quantity @xmath715 is presumably the actual charge when @xmath716 .
again , taking @xmath179 , @xmath239 and @xmath711 we recover the metric ( [ gfrsz ] ) written in the standard form but now with @xmath717 given by @xmath718 .
s. nojiri and s. d. odintsov , int . j. geom .
phys . 4 , 115 ( 2007 )
; ; s. capozziello , m. de laurentis , and v. faraoni , arxiv : 0909.4672 ; s. capozziello , and m. de laurentis , arxiv : 1108.6266 ; t. clifton , p. g. ferreira , a. padilla , and c. skordis , ; s. capozziello , and m. francaviglia , ; t. p. sotiriou and v. faraoni , ; a. de felice , and s. tsujikawa ,
. p. t. chrusciel , contemp . math . * 170 * , 23 ( 1994 ) ; m. heusler , _ black holes uniqueness theorems _ , cambridge univ . press , cambridge ( uk ) , 1996 ;
d. c. robinson , in _
kerr spacetime : rotating black holes in general relativity _ ,
d. l. wiltshire , m. visser , and s. m. scott , cambridge univ . press ,
cambridge ( uk ) , 2009 , p.115 - 143 .
t. multamki , and i. vilja , ; ; k. kainulainen , j. piilonen , v. reijonen , and d. sunhede , ; k. kainulainen , and d. sunhede , ; k. henttunen , t. multamki , and i. vilja , ; t. kobayashi , and k. maeda , ; ; e. babichev , and d. langlois , ; ; a. upadhye , and w. hu , ; s. capozziello , m. de laurentis , s. d. odintsov , and a. stabile , ; s. s. yazadjiev , d. d. doneva , k. d. kokkotas , and kalin v. staykov , arxiv:1402.4469 s. capozziello , a. stabile , and a. troisi , ; ; a. de la cruz dombriz , a. dobado , and a. l. maroto , ; ; a. larraaga , pramana journal of physics , * 78 * , 697 ( 2012 ) [ arxiv : 1108.6325 ] ; j. a. r. cembranos , a. de la cruz dombriz , and p. jimeno romero , arxiv : 1109.4519 ; t. moon , y. s. myung , and e. j. son , ; a. sheykhi , ; s. habib mazharimousavi , m. halilsoy , and t. tahamtan , . l. f. abbott , and s. deser , ; g. w. gibbons , s. w. hawking , g. t. horowitz , and m. j. perry , ; a. ashtekhar , and a. magnon , ; w. boucher , g. w. gibbons , and g. t. horowitz , ; m. henneaux , and c. teitelboim , ; v. balasubramanian , and p. krauss , ; g. w. gibbons , ; a. ashtekhar , and s. das , ; p. chruciel , and g. nagy , ; adv
* 5 * , 697 ( 2002 ) .
g. cognola , e. elizalde , s. nojiri , s. d. odintsov , l. sebastiani , s. zerbini , ; e. linder , ; l. yang , c. c. lee , l. w. luo , c. q. geng , ; k. bamba , c. q. geng , c. c. lee , ; e. elizalde , s. nojiri , s. d. odintsov , l. sebastiani , s. zerbini , ; e. elizalde , s. d. odintsov , l. sebastiani , s. zerbini , . | we discuss with a rather critical eye the current situation of black hole ( bh ) solutions in @xmath0 gravity and shed light about its geometrical and physical significance .
we also argue about the meaning , existence or lack thereof of a birkhoff s theorem in this kind of modified gravity .
we focus then on the analysis and quest of _ non - trivial _
( i.e. hairy ) _ asymptotically flat _ ( af ) bh solutions in static and spherically symmetric ( sss ) spacetimes in vacuum having the property that the ricci scalar does _ not _ vanish identically in the domain of outer communication . to do so , we provide and enforce the _ regularity conditions _ at the horizon in order to prevent the presence of singular solutions there .
specifically , we consider several classes of @xmath0 models like those proposed recently for explaining the accelerated expansion in the universe and which have been thoroughly tested in several physical scenarios . finally , we report analytical and numerical evidence about the _ absence _ of _ geometric hair _ in afsssbh solutions in those @xmath0 models .
first , we submit the models to the available no - hair theorems , and in the cases where the theorems apply , the absence of hair is demonstrated analytically . in the cases where the theorems do not apply , we resort to a numerical analysis due to the complexity of the non - linear differential equations . within that aim , a code to solve the equations numerically was built and tested using well know exact solutions . in a future investigation
we plan to analyze the problem of hair in de sitter and anti - de sitter backgrounds .
= 1 |
consider the classical regression quantile model : given independent observations @xmath4 , with @xmath5 fixed ( for fixed @xmath6 ) , the conditional quantile of the response @xmath7 given @xmath8 is @xmath9 let @xmath10 be the koenker
bassett regression quantile estimator of @xmath11 .
@xcite provides definitions and basic properties , and describes the traditional approach to asymptotics for @xmath10 using a bahadur representation : @xmath12 where @xmath13 is a brownian bridge and @xmath14 is an error term .
unfortunately , @xmath14 is of order @xmath15 [ see , e.g. , @xcite and @xcite ] .
this might suggest that asymptotic results are accurate only to this order . however , both simulations in regression cases and one - dimensional results [ @xcite ( @xcite ) ] justify a belief that regression quantile methods should share ( nearly ) the @xmath16 accuracy of smooth statistical procedures ( uniformly in @xmath17 ) .
in fact , as shown in @xcite , @xmath18 has a limit with zero mean and that is independent of @xmath19 .
thus , in any smooth inferential procedure ( say , confidence interval lengths or coverages ) , this error term should enter only through @xmath20 .
nonetheless , this expansion would still leave an error of @xmath21 ( coming from the error beyond the @xmath14 term in the bahadur representation ) , and so would still fail to reflect root-@xmath2 behavior .
furthermore , previous results only provide such a second - order expansion for fixed @xmath17.=1 it must be noted that the slower @xmath22 error rate arises from the discreteness introduced by indicator functions appearing in the gradient conditions . in fact , expansions can be carried out when the design is assumed to be random ; see @xcite and @xcite , where the focus is on analysis of the @xmath23 bootstrap .
specifically , the assumption of a smooth distribution for the design vectors together with a separate treatment of the lattice contribution of the intercept does permit appropriate expansions .
unfortunately , the randomness in @xmath24 means that all inference must be in terms of the average asymptotic distribution ( averaged over @xmath24 ) , and so fails to apply to the generally more desirable conditional forms of inference .
specifically , unconditional methods may be quite poor in the heteroscedastic and nonsymmetric cases for which regression quantile analysis is especially appropriate .
the main goal of this paper is to reclaim increased accuracy for conditional inference beyond that provided by the traditional bahadur representation .
specifically , the aim is to provide a theoretical justification for an error bound of nearly root-@xmath2 order uniformly in @xmath17 .
define @xmath25 we first develop a normal approximation for the density of @xmath26 with the following form : @xmath27 for @xmath28 , where @xmath29 .
we then extend this result to the densities of a pair of regression quantiles in order to obtain a `` hungarian '' construction [ @xcite ( @xcite ) ] that approximates the process @xmath30 by a gaussian process to order @xmath31 , where @xmath32 ( uniformly for @xmath33 ) .
section [ sec2 ] provides some applications of the results here to conditional inference methods in regression quantile models .
specifically , an expansion is developed for coverage probabilities of confidence intervals based on the [ @xcite ] difference quotient estimator of the sparsity function .
the coverage error rate is shown to achieve the rate @xmath34 for conditional inference , which is nearly the known `` optimal '' rate obtained for a single sample and for unconditional inference .
section [ sec3 ] lists the conditions and main results , and offers some remarks .
section [ sec4 ] provides a description of the basic ingredients of the proof ( since this proof is rather long and complicated ) .
section [ sec5 ] proves the density approximation for a fixed @xmath17 ( with multiplicative error ) .
section [ sec6 ] extends the result to pairs of regression quantiles ( theorem [ den2d ] ) , and section [ sec7 ] provides the `` hungarian '' construction ( theorem [ hung ] ) with what appears to be a somewhat innovative induction along dyadic rationals .
as the impetus for this work was the need to provide some theoretical foundation for empirical results on the accuracy of regression quantile inference , some remarks on implications are in order . [ remark1 ] clearly , whenever published work assesses the accuracy of an inferential method using the error term from the bahadur representation , the present results will immediately provide an improvement from @xmath35 to the nearly root-@xmath2 rate here .
one area of such results is methods based directly on regression quantiles and not requiring estimation of the sparsity function [ @xmath36 .
there are several papers giving such results , although at present it appears that their methods have theoretical justification only under location - scale forms of quantile regression models .
specifically , @xcite introduced confidence intervals ( especially for fitted values ) based on using pairs of regression quantiles in a way analogous to confidence intervals for one - sample quantiles .
they showed that the method was consistent , but the accuracy depended on the bahadur error term .
thus , results here now provide accuracy to the nearly root-@xmath2 rate of theorem [ th2 ] . a second approach directly using the dual quantile process
is based on the regression ranks of @xcite .
again , the error terms in the theoretical results there can be improved using theorem [ th1 ] here , though the development is not so direct . for a third application
, @xcite showed that the regression quantile process interpolated along a grid of mesh strictly larger than @xmath1 is asymptotically equivalent to the full regression quantile process to first order , but ( because of additional smoothness ) will yield monotonic quantile functions with probability tending to 1 .
however , their development used the bahadur representation , which indicated that a mesh of order @xmath37 balanced the bias and accuracy and bounded the difference between @xmath38 and its linear interpolate by nearly @xmath39 . with some work , use of the results here would permit a mesh slightly larger than the nearly root-@xmath2 rate here to obtain an approximation of nearly root-@xmath2 order .
[ remark2 ] inference under completely general regression quantile models appears to require either estimation of the sparsity function or use of resampling methods .
the most general methods in the ` quantreg ` package [ @xcite ] use the `` difference quotient '' method with the [ @xcite ] bandwidth of order @xmath37 , which is known to be optimal for coverage probabilities in the one - sample problem .
as noted above , expansions using the randomness of the regressors can be developed to provide analogous results for unconditional inference .
the results here ( with some elaboration ) can be used to show that the hall sheather estimates provide ( nearly ) the same rates of accuracy for coverage probabilities under the conditional form of the regression quantile model . to be specific ,
consider the problem of confidence interval estimation for a fixed linear combination of regression parameters : @xmath40 .
the asymptotic variance is the well - known sandwich formula @xmath41 where @xmath42 is the sparsity , @xmath43 ( with @xmath44 being the gradient ) , and where @xmath24 is the design matrix .
following @xcite , the sparsity may be approximated by the difference quotient @xmath45 .
standard approximation theory ( using the taylor series ) shows that @xmath46 the sparsity may be estimated by @xmath47 and the sparsity ( [ sadef ] ) may be estimated by inserting @xmath48 in @xmath49 .
then , as shown in the , the confidence interval @xmath50 has coverage probability @xmath51 , which is within a factor of @xmath3 of the optimal hall sheather rate in a single sample .
furthermore , this rate is achieved at the ( optimal ) @xmath52-value @xmath53 , which is the optimal hall sheather bandwidth except for the @xmath54 term . since the optimal bandwidth depends on @xmath55 , the optimal constant for the @xmath56
can not be determined , as it can when @xmath24 is allowed to be random [ and for which the @xmath57 term is explicit ] .
this appears to be an inherent shortcoming for using inference conditional on the design .
note also that it is possible to obtain better error rates for the coverage probability by using higher order differences .
specifically , using the notation of ( [ deldef ] ) , @xmath58 as a consequence , the optimal bandwidth for this estimator is of order @xmath59 , and the coverage probability is accurate to order @xmath60 ( except for logarithmic factors ) .
[ remark3 ] a third approach to inference applies resampling methods .
as noted in the , while the @xmath23 bootstrap is available for unconditional inference , the practicing statistician will generally prefer to use inference conditional on the design .
there are some resampling approaches that can obtain such inference .
one method is that of @xcite , which simulates the binomial variables appearing in the gradient condition .
another is the `` markov chain marginal bootstrap '' of @xcite [ see also @xcite ] . however
, this method also involves sampling from the gradient condition .
the discreteness in the gradient condition would seem to require the error term from the bahadur representation , and thus leads to poorer inferential approximation : the error would be no better than order @xmath61 even if it were the square of the bahadur error term . while some evidence for decent performance of these methods comes from ( rather limited ) simulations , it is often noticed that these methods perform perhaps somewhat more poorly than the other methods in the ` quantreg ` package of @xcite . clearly , a more complete analysis of inference for regression quantiles based on the more accurate stochastic expansions here would be useful
under the regression quantile model of section [ sec1 ] , the following conditions will be imposed : let @xmath62 denote the coordinates of @xmath8 except for the intercept ( i.e. , the last @xmath63 coordinates , if there is an intercept ) .
let @xmath64 denote the conditional characteristic function of the random variable @xmath65 , given @xmath8 .
let @xmath66 and @xmath67 denote the conditional density and c.d.f
. of @xmath7 given @xmath8 .
[ co1 ] for any @xmath68 , there is @xmath69 such that @xmath70 uniformly in @xmath71 .
[ co2 ] @xmath72 are uniformly bounded , and there are positive definite @xmath73 matrices @xmath74 and @xmath75 such that for any @xmath68 ( as @xmath76 ) @xmath77 uniformly in @xmath71 .
[ cof ] the derivative of @xmath78 is uniformly bounded on the interval @xmath79 .
two fundamental results will be developed here .
the first result provides a density approximation with multiplicative error of nearly root-@xmath2 rate . a result for fixed @xmath17
is given in theorem [ th5 ] , but the result needed here is a bivariate approximation for the joint density of one regression quantile and the difference between this one and a second regression quantile ( properly normalized for the difference in @xmath17-values ) .
let @xmath80 for some @xmath68 , and let @xmath81 with @xmath82 for some @xmath83 .
here , one may want to take @xmath84 near 1 [ see remark ( 1 ) below ] , though the basic result will often be useful for @xmath85 , or even smaller .
define @xmath86.\end{aligned}\ ] ] [ th1 ] [ den2d ] under conditions [ co1 ] , [ co2 ] and [ cof ] , there is a constant @xmath49 such that for @xmath87 and @xmath88 at @xmath89 and @xmath90 , respectively , satisfies @xmath91 where @xmath92 is a normal density with covariance matrix @xmath93 having the form given in ( [ diffcov ] ) .
the second result provides the desired `` hungarian '' construction : [ th2 ] [ hung ] assume conditions [ co1 ] , [ co2 ] and [ cof ] .
fix @xmath94 with @xmath83 , and let @xmath95 be dyadic rationals with denominator less than @xmath96 . define @xmath97 to be the piecewise linear interpolant of @xmath98 [ as defined in ( [ bndef ] ) ] .
then for any @xmath68 , there is a ( zero - mean ) gaussian process , @xmath99 , defined along the dyadic rationals @xmath95 and with the same covariance structure as @xmath100 ( along @xmath101 ) such that its piecewise linear interpolant @xmath102 satisfies @xmath103 almost surely .
some remarks on the conditions and ramifications are in order : \(1 ) the usual construction approximates @xmath104 by a `` brownian bridge '' process .
theorem [ hung ] really only provides an approximation for the discrete processes at a sufficiently sparse grid of dyadic rationals . that the piecewise linear interpolants converge to the usual brownian bridge follows as in @xcite .
the critical impediment to getting a brownian bridge approximation to @xmath105 with the error in theorem [ hung ] is the square root behavior of the modulus of continuity .
this prevents approximating the piecewise linear interpolant within an interval of length greater than ( roughly ) order @xmath106 if a root-@xmath2 error is desired . in order to approximate the density of the difference in @xmath105 over an interval between dyadic rationals ,
the length of the interval must be at least of order @xmath107 ( for @xmath83 ) .
clearly , it will be possible to approximate the piecewise linear interpolant by a brownian bridge with error @xmath108 , and thus to get arbitrarily close to the value of @xmath109 for the exponent of @xmath2 . for most purposes
, it might be better to state the final result as @xmath110 for any @xmath111 ( where @xmath112 is the appropriate brownian bridge ) ; but the stronger error bound of theorem [ hung ] does provide a much closer analog of the result for the one - sample ( one - dimensional ) quantile process .
\(2 ) the one - sample result requires only the first power of @xmath113 , which is known to give the best rate for a general result .
the extra addition of @xmath114 in the exponent is clearly needed for the density approximation , but this may be only a technical assumption .
nonetheless , i conjecture that some extra amount is needed in the exponent .
\(3 ) conditions [ co1 ] and [ co2 ] can be shown to hold with probability tending to one under smoothness and boundedness assumptions of the distribution of @xmath115 .
nonetheless , the condition that @xmath116 be bounded seems rather strong in the case of random @xmath115 .
it seems clear that this can be weakened , though probably at the cost of getting a poorer approximation .
for example , @xmath116 having exponentially small tails might increase the bound in theorem [ hung ] by an additional factor of @xmath113 , and algebraic tails are likely worse .
however , details of such results remain to be developed .
\(4 ) similarly , it should be possible to let @xmath117 , which defines the compact subinterval of @xmath17-values , tend to zero . clearly ,
letting @xmath118 be of order @xmath106 would lead to extreme value theory and very different approximations . for slower rates of convergence of @xmath118 ,
bahadur expansions have been developed [ e.g. , see @xcite ] and extension to the approximation result in theorem [ hung ] should be possible .
again , however , this would most likely be at the cost of a larger error term .
\(5 ) the assumption that the conditional density of the response ( given @xmath115 ) be continuous is required even for the usual first order asymptotics .
however , one might hope to avoid condition [ cof ] , which requires a bounded derivative at all points .
for example , the double exponential distribution does not satisfy this condition .
it is likely that the proofs here can be extended to the case where the derivative does not exist on a finite set ( or even on a set of measure zero ) , but dropping differentiability entirely would require a rather different approach .
furthermore , the apparent need for bounded derivatives in providing uniformity over @xmath17 in bahadur expansions suggests the possibility that some differentiability is required .
\(6 ) theorem [ den2d ] provides a bivariate normal density approximation with error rate ( nearly ) @xmath1 when @xmath119 and @xmath120 are fixed . when @xmath121 , of course , the error rate is larger .
note , however , that the slower convergence rate when @xmath122 does not reduce the order of the error in the final construction since the difference @xmath123 is of order @xmath124 .
the development of the fundamental results ( theorems [ den2d ] and [ hung ] ) will be presented in three phases . the first phase provides the density approximation for a fixed @xmath17 , since some of the more complicated features are more transparent in this case .
the second phase extends this result to the bivariate approximation of theorem [ den2d ] .
the final phase provides the `` hungarian '' construction of theorem [ th2 ] . to clarify the development ,
the basic ingredients and some preliminary results will be presented first .
[ ingredient1 ] begin with the finite sample density for a regression quantile [ @xcite , @xcite ] : assume @xmath7 has a density , @xmath125 , and let @xmath17 be fixed . note that @xmath38 is defined by having @xmath6 zero residuals ( if the design is in general position ) .
specifically , there is a subset , @xmath52 , of @xmath6 integers such that @xmath126 , where @xmath127 has rows @xmath128 for @xmath129 and @xmath130 has coordinates @xmath7 for @xmath131 .
let @xmath132 denote the set of all such @xmath6-element subsets .
define @xmath133 as described in @xcite , the density of @xmath134 evaluated at the argument @xmath135 is given by @xmath136 here , the event in the probability above is the event that the gradient condition holds for a fixed subset , @xmath137 , where @xmath138 , with @xmath139 the rectangle that is the product of intervals @xmath140 [ see theorem 2.1 of @xcite ] , and where @xmath141 [ ingredient2 ] since @xmath142 is approximately normal , and @xmath143 is bounded , the probability in ( [ finite ] ) is approximately a normal density evaluated at @xmath89 . to get a multiplicative bound
, we may apply a `` cramr '' expansion ( or a saddlepoint approximation ) .
if @xmath144 had a smooth distribution ( i.e. , satisfied cramr s condition ) , then standard results would apply .
unfortunately , @xmath144 is discrete .
the first coordinate of @xmath144 is nearly binomial , and so a multiplicative bound can be obtained by applying a known saddlepoint formula for lattice variables [ see @xcite ] .
equivalently , approximate by an exact binomial and ( more directly , but with some rather tedious computation ) expand the logarithm of the gamma function in stirling s formula . using either approach
, one can show the following result : [ th3 ] [ bin ] let @xmath145 , @xmath146 be any interval of length @xmath147 containing @xmath148 , and let @xmath149 .
then @xmath150 where @xmath151 . a proof based on multinomial expansions is given for the bivariate generalization in theorem [ den2d ] .
note that this result includes an extra factor of @xmath152 .
this will allow the bounds to hold except with probability bounded by an arbitrarily large negative power of @xmath2 .
this is clear for the limiting normal case ( by standard asymptotic expansions of the normal c.d.f . ) . to obtain such bounds for
the distribution of @xmath144 will require some form of bernstein s inequality .
such inequalities date to bernstein s original publication in 1924 [ see @xcite ] , but a version due to @xcite may be easier to apply .
[ ingredient3 ] using theorem [ bin ] , it can be shown ( see section [ sec4 ] ) that the probability in ( [ finite ] ) may be approximated as @xmath153 where the first coordinate of @xmath154 is a sum of @xmath2 i.i.d . @xmath155
random variables , the last @xmath156 coordinates are those of @xmath144 , and @xmath157 . since we seek a normal approximation for this probability with multiplicative error , at this point one might hope that a known ( multidimensional ) `` cramr '' expansion or saddlepoint approximation would allow @xmath154 to be replaced by a normal vector ( thus providing the desired result ) .
however , this will require that the summands be smooth , or ( at least ) satisfy a form of cramr s condition .
let @xmath158 denote the last @xmath159 coordinates of @xmath160 .
one approach would be to assume @xmath158 has a smooth distribution satisfying the classical form of cramr s condition . however , to maintain a conditional form of the analysis , it suffices to impose a condition on @xmath161 , which is designed to mimic the effect of a smooth distribution and will hold with probability tending to one if @xmath158 has such a smooth distribution .
condition [ co1 ] specifies just such an assumption .
note that the characteristic functions of the summands of @xmath162 , say , @xmath163 , will also satisfy condition [ co1 ] [ equation ( [ x - cond1 ] ) ] and so should allow application of known results on normal approximations .
unfortunately , i have been unable to find a published result providing this and so section [ sec5 ] will present an independent proof . clearly , some additional conditions will be required .
specifically , we will need conditions that the empirical moments of @xmath164 converge appropriately , as specified in condition [ co2 ] .
finally , the approach using characteristic functions is greatly simplified when the sums , @xmath154 , have densities .
again , to avoid using smoothness of the distribution of @xmath165 ( and thus to maintain a conditional approach ) , introduce a random perturbation @xmath166 which is small and has a bounded smooth density ( the bound may depend on @xmath2 ) .
section [ sec4 ] will then prove the following : [ th4 ] [ sina ] assume conditions [ co1 ] and [ co2 ] and the regression quantile model of section [ sec1 ] .
let @xmath89 be the argument of the density of @xmath167 , and suppose @xmath168 for some constant @xmath169 .
then a constant @xmath170 can be chosen so that @xmath171 where @xmath172 has mean @xmath173 and covariance @xmath174 , @xmath170 can be arbitrarily large , and @xmath166 is a small perturbation [ see ( [ vbound ] ) ] . following the proof of this theorem , it will be shown that the effect of @xmath166 can be ignored , if @xmath166 is bounded by @xmath175 , where @xmath176 may depend on @xmath169 ( but not on @xmath170 ) .
[ ingredient4 ] expanding the densities in ( [ finite ] ) is trivial if the densities are sufficiently smooth .
the assumption of a bounded first derivative in condition [ cof ] appears to be required to analyze second order terms ( beyond the first order normal approximation ) .
[ ingredient5 ] finally , summing terms involving @xmath177 in ( [ finite ] ) over the @xmath178 summands will require vinograd s theorem and related results from matrix theory concerning adjoint matrices [ see @xcite ] .
the remaining ingredients provide the desired `` hungarian '' construction .
[ ingredient6 ] extend the density approximation to the joint density for @xmath179 and @xmath180 ( when standardized ) .
a major complication is that one needs @xmath181 , making the covariance matrix tend to singularity .
thus , we focus on the joint density for standardized versions of @xmath179 and @xmath182 .
clearly , this requires modification of the proof for the univariate case to treat the fact that @xmath183 converges at a rate depending on @xmath184 .
the result is given in theorem [ den2d ] .
[ ingredient7 ] extend the density result to obtain an approximation for the quantile transform for the conditional distribution of differences @xmath183 ( between successive dyadic rationals ) .
this will provide ( independent ) normal approximations to the differences whose sums will have the same covariance structure as the regression quantile process ( at least along a sufficiently sparse grid of dyadic rationals ) .
[ ingredient8 ] finally , the hungarian construction is applied inductively along the sparse grid of dyadic rationals .
this inductive step requires some innovative development , mainly because the regression quantile process is not directly expressible in terms of sums of random variables ( as are the empiric one - sample distribution function and quantile function ) .
let @xmath185 be the last @xmath186 coordinates of @xmath144 and @xmath187 be the interval @xmath188 . then , @xmath189 where @xmath190 is the set @xmath191 shifted as indicated above .
note that by hoeffding s inequality [ @xcite ] , for any fixed @xmath169 , the shift satisfies @xmath192 except with probability bounded by @xmath193 .
thus , we may apply theorem [ bin ] [ equation ( [ binbd ] ) ] with @xmath194 equal to the shift above to obtain the following bound ( to within an additional additive error of @xmath193 ) : @xmath195 where @xmath196 and @xmath184 is a bound on @xmath197 , which may be taken to be of the form @xmath198 ( by hoeffding s inequality ) .
finally , we obtain @xmath199 where the first coordinate of @xmath154 is a sum of @xmath2 i.i.d .
@xmath155 random variables and the last @xmath200 coordinates are those of @xmath144
. to treat the probability involving @xmath154 , standard approaches using characteristic functions can be employed . in theory , exponential tilting ( or saddlepoint methods ) should provide better approximations , but since we require only the order of the leading error term , we can proceed more directly . as in @xcite ,
the first step is to add an independent perturbation so that the sum has an integrable density : specifically , for fixed @xmath201 let @xmath166 be a random variable ( independent of all observations ) with a smooth bounded density and for which ( for each @xmath201 ) @xmath202 where @xmath176 will be chosen later .
define @xmath203 we now allow @xmath143 to be any ( arbitrary ) set , say , @xmath204 .
thus , @xmath205 has a density and we can write [ with @xmath206 @xmath207 where @xmath208 denotes the characteristic function of the random variable @xmath209 .
break domain of integration into 3 sets : @xmath210 , @xmath211 , and @xmath212 . on @xmath213 ,
expand @xmath214 .
for this , compute @xmath215 \\ & = & x_i x_i ' \tau ( 1 - \tau ) + { { \mathcal}{o } } \bigl ( \| x_i \|^3 \| \delta\|^2 / n \bigr).\end{aligned}\ ] ] hence , using the boundedness of @xmath216 , @xmath217 and @xmath218 ( on this first interval ) , @xmath219 where @xmath220 and @xmath221 are defined in condition [ co2 ] [ see ( [ gdef ] ) and ( [ hdef ] ) ] . for the other two intervals on the @xmath222-axis , the integrands will be bounded by an additive error times @xmath223 since @xmath224 . on @xmath225 ,
the summands are bounded and so their characteristic functions satisfy @xmath226 for some constant @xmath227 .
thus , on @xmath228 , @xmath229 for some constant @xmath230 . therefore , integrating times @xmath231 provides an additive bound of order @xmath232 , where @xmath233 and ( for any @xmath170 ) @xmath234 can be chosen sufficiently large so that @xmath235 .
finally , on @xmath212 , condition [ co1 ] [ see ( [ x - cond1 ] ) ] gives an additive bound of @xmath236 directly and , again ( as on the previous interval ) , an additive error bounded by @xmath237 can be obtained .
therefore , it now follows that we can choose @xmath170 ( depending on @xmath169 , @xmath176 , @xmath234 and @xmath238 ) so that @xmath239 from which theorem [ sina ] follows .
finally , we show that the contribution of @xmath166 can be ignored : @xmath240 where @xmath241 denotes the symmetric difference of the sets .
since @xmath166 is bounded and @xmath242 , this symmetric difference is contained in a set , @xmath49 , which is the union of @xmath243 ( boundary ) parallelepipeds each of the form @xmath244 , where @xmath245 is a rectangle one of whose coordinates has width @xmath246 and all other coordinates have length 1 .
thus , applying theorem [ sina ] ( as proved for the set @xmath247 ) , @xmath248 where @xmath227 and @xmath249 are constants , and @xmath176 may be chosen arbitrarily large .
[ th5 ] [ den1d ] assume conditions [ co1 ] , [ co2 ] , [ cof ] and the regression quantile model of section [ sec1 ] .
let @xmath89 be the argument of the density of @xmath250 and suppose @xmath251 for some constant @xmath169 .
then , uniformly in @xmath252 ( for @xmath253 ) , @xmath254 where @xmath255 denotes the normal density with covariance @xmath256 with @xmath220 and @xmath221 given by ( [ gdef ] ) and ( [ hdef ] ) .
recall the basic formula for the density ( [ finite ] ) : @xmath257 by theorem [ sina ] , ignoring the multiplicative and additive error terms given in this result and setting @xmath258 , @xmath259 since @xmath260 is bounded by a constant times @xmath61 on @xmath261 and the last integral equals @xmath262 . by ingredient [ ingredient4 ] ,
the product is @xmath263 this gives the main term of the approximation as @xmath264 the penultimate step is to apply results from matrix theory on adjoint matrices [ specifically , the cauchy binet theorem and the `` trace '' theorem ; see , e.g. , @xcite , pages 9 and 87 ] : the sum above is just the trace of the @xmath6th adjoint of @xmath265 , which equals @xmath266 .
the various determinants combine ( with the factor @xmath267 ) to give @xmath268 , which provides the asymptotic normal density we want .
finally , we need to combine the multiplicative and additive errors into a single multiplicative error .
so consider @xmath269 ( for some constant @xmath169 ) .
then , the asymptotic normal density is bounded below by @xmath270 for some constant @xmath227 .
thus , since the constant @xmath170 ( which depends on @xmath176 , @xmath234 , @xmath238 and @xmath271 ) can be chosen so that the additive errors are smaller than @xmath272 , the error is entirely subsumed in the multiplicative factor : @xmath273 .
we first prove theorem [ den2d ] , which provides the bivariate normal approximation .
proof of theorem [ den2d ] the proof follows the development in theorem [ den1d ] .
the first step treats the first ( intercept ) coordinate .
since the binomial expansions were omitted in the proof of theorem [ bin ] , details for the trinomial expansion needed for the bivariate case here will be presented . the binomial sum in the first coordinate of ( [ sndef ] )
will be split into the sum of observations in the intervals @xmath274 , @xmath275 and @xmath276 . the expected number of observations in each interval is within @xmath6 of @xmath2 times the length of the corresponding interval .
thus , ignoring an error of order @xmath277 , we expand a trinomial with @xmath2 observations and @xmath278 and .
let @xmath279 be the ( trinomially distributed ) number of observation in the respective intervals and consider @xmath280 .
we may take @xmath281\\[-8pt ] k_2 & = & { \mathcal{o } } \bigl ( { a_n ( \log n)^{1/2 } } \bigr),\nonumber\end{aligned}\ ] ] since these bounds are exceeded with probability bounded by @xmath282 for any ( sufficiently large ) @xmath169 .
so @xmath283 , where @xmath284 expanding ( using sterling s formula and some computation ) , @xmath285 & & \hspace*{37.8pt } { } - \biggl(n p_1 + k_1 + \frac{1}{2}\biggr ) \log\biggl ( np_1 + \frac{k_1 + 1}{n p_1 } \biggr ) \\[-1pt ] & & \hspace*{37.8pt } { } - \biggl(n p_2 + k_2 + \frac{1}{2}\biggr ) \log \biggl ( n p_2 + \frac { k_2 + 1}{n p_2 } \biggr)\\[-1pt ] & & \hspace*{37.8pt } { } - \biggl(n(1-p_1-p_2)-k_1-k_2 + \frac{1}{2}\biggr ) \\[-1pt ] & & \hspace*{48.8pt } { } \times\log\biggl ( n(1-p_1-p_2 ) - \frac{k_1+k_2 -1}{n(1-p_1-p_2 ) } \biggr ) + { \mathcal{o } } \biggl ( { \frac { 1}{np_2 } } \biggr ) \biggr\ } \\[-1pt ] & = & \frac{1}{2\pi } \exp\biggl\{\frac{1}{2}\log n - n p_1 \log p_1 - \biggl(k_1 + \frac{1}{2}\biggr ) \log(n p_1)\\[-1pt ] & & \hspace*{37.8pt } { } - n p_2 \log p_2 - \biggl(k_2 + \frac{1}{2}\biggr ) \log(n p_2)\\[-1pt ] & & \hspace*{37.8pt } { } - n ( 1 - p_1 - p_2 ) \log(1 - p_1 - p_2 ) -\biggl(k_1 + k_2 + \frac{1}{2 } \biggr ) \\[-1pt ] & & \hspace*{48.8pt } { } \times\log\bigl(n(1 - p_1 - p_2)\bigr ) - \frac{k_1 ^ 2}{np_1 } - \frac{k_2 ^ 2}{np_2 } \\[-1pt ] & & \hspace*{125.1pt}{}- \frac { ( k_1+k_2)^2}{n(1-p_1-p_2 ) } + { \mathcal{o } } \biggl ( { \frac{k_2 ^ 3}{(np_2)^2 } } \biggr ) \biggr\ } \\[-1pt ] & = & \frac{1}{2\pi } \exp\biggl\{- \log n - \biggl(np_1 + k_1 + \frac{1}{2}\biggr ) \log p_1 - \biggl(np_2 + k_2 + \frac{1}{2}\biggr ) \log p_2 \\[-1pt ] & & \hspace*{37pt } { } - \biggl(n(1-p_1-p_2 ) -k_1 - k_2 + \frac{1}{2}\biggr ) \log(1-p_1 - p_2 ) \\[-1pt ] & & \hspace*{73.5pt } { } - \frac{k_1 ^ 2}{np_1 } - \frac{k_2 ^ 2}{np_2 } - \frac { ( k_1+k_2)^2}{n(1-p_1-p_2 ) } + { \mathcal{o } } \biggl ( { \frac { ( log n)^{3/2 } } { n a_n^2 } } \biggr ) \biggr \ } , \\[-1pt ] b & = & \exp\bigl\{(np_1+k_1 ) \log p_1 + ( np_2 + k_2 ) \log
p_2 \\[-1pt ] & & \hspace*{18.5pt } { } + \bigl(n(1-p_1-p_2 ) -k_1 - k_2 \bigr ) \log(1-p_1-p_2 ) \bigr\}.\end{aligned}\ ] ] therefore , @xmath286 & & \hspace*{19.3pt } { } - \frac{k_1 ^ 2}{np_1 } - \frac{k_2 ^ 2}{np_2 } - \frac { ( k_1+k_2)^2}{n(1-p_1-p_2 ) } + { \mathcal{o } } \biggl ( { \frac { ( log n)^{3/2 } } { n a_n^2 } } \biggr ) \biggr \}.\end{aligned}\ ] ] some further simplification shows that @xmath287 gives the usual normal approximation to the trinomial with a multiplicative error of @xmath288 [ when @xmath289 and @xmath290 satisfy ( [ ki ] ) ] .
the next step of the proof follows that of theorem [ sina ] ( see ingredient [ ingredient3 ] ) .
since the proof is based on expanding characteristic functions ( which do not involve the inverse of the covariance matrices ) , all uniform error bounds continue to hold .
this extends the result of theorem [ sina ] to the bivariate case : @xmath291 & & \qquad= p \bigl\ { z_1 \in a_{h_1 } / \sqrt{n } , z_2 \in a_{h_2 } / \sqrt{n } \bigr\ } \\[-1pt ] & & \qquad= p \bigl\ { z_1 \in a_{h_1 } / \sqrt{n } \bigr\ } \times p \bigl\{(z_2 - z_1 ) / \sqrt{n } \in ( a_{h_2 } - z_2 ) / \sqrt{n } | z_1 \bigr\}\nonumber\end{aligned}\ ] ] for appropriate normally distributed @xmath292 ( depending on @xmath2 ) . this last equation is needed to extend the argument of theorem [ den1d ] , which involves integrating normal densities .
the joint covariance matrix for @xmath293 is nearly singular ( for @xmath294 small ) and complicates the bounds for the integral of the densities .
the first factor above can be treated exactly as in the proof of theorem [ den1d ] , while the conditional densities involved in the second factor can be handled by simple rescaling .
this provides the desired generalization of theorem [ den1d ] .
thus , the next step is to develop the parameters of the normal distribution for @xmath295 [ see ( [ bndef ] ) , ( [ rndef ] ) ] in a usable form .
the covariance matrix for @xmath296 has blocks of the form @xmath297 where @xmath298 with @xmath220 and @xmath221 given in condition [ co2 ] [ see ( [ gdef ] ) and ( [ hdef ] ) ] . expanding @xmath299 about @xmath300 ( using the differentiability of the densities from condition [ cof ] ) , @xmath301 where @xmath302 are derivatives of @xmath220 at @xmath119 (
note that @xmath303 ) .
straightforward matrix computation now yields the joint covariance for @xmath295 : @xmath304 where @xmath305 are uniformly bounded matrices .
thus , the conditional distribution of @xmath306 given @xmath307 has moments @xmath308 & = & ( \tau_2 - \tau_1 ) \lambda_{11}^{-1 } \delta_{12 } / \bigl(\tau_1 ( 1 - \tau_1)\bigr ) , \\[-2pt ] \label{covcond } \operatorname{cov } \bigl [ r_n | b_n(\tau_1 ) \bigr ] & = & ( \tau_2 - \tau_1 ) \biggl [ \delta_{22}^ * - \frac{\tau_2 - \tau_1}{\tau_1 ( 1 - \tau_1 ) } \delta_{21}^ * \lambda_{11}^{-1 } \delta_{12}^ * \biggr]\end{aligned}\ ] ] and analogous equations also hold for @xmath309 .
finally , recalling that @xmath310 , the second term in ( [ zcond ] ) can be written @xmath311 thus , since the conditional covariance matrix is uniformly bounded except for the @xmath312 factor , the argument of theorem [ den1d ] also applies directly to this conditional probability .
finally , the above results are used to apply the quantile transform for increments between dyadic rationals inductively in order to obtain the desired `` hungarian '' construction .
the proof of theorem [ hung ] is as follows : proof of theorem [ hung ] ( i ) following the approach in @xcite , the first step is to provide the result of theorem [ den2d ] for conditional densities one coordinate at a time . using the notation of theorem [ den2d ] ,
let @xmath313 and @xmath314 be successive dyadic rationals ( between @xmath117 and @xmath315 ) with denominator @xmath316 .
so @xmath317 .
let @xmath318 be the @xmath319th coordinate of @xmath320 [ see ( [ rndef ] ) ] , let @xmath321 be the vector of coordinates before the @xmath319th one , and let @xmath322 .
then the conditional density of @xmath323 satisfies @xmath324 for @xmath325 , @xmath326 , and @xmath327 , and where @xmath328 and @xmath329 are easily derived from ( [ econd ] ) and ( [ covcond ] ) .
note that @xmath328 has the form @xmath330 where @xmath331 can be bounded ( independent of @xmath2 ) and @xmath332 can be bounded away from zero and infinity ( independent of @xmath2 ) .
this follows since the conditional densities are ratios of marginal densities of the form @xmath333 ( with @xmath334 satisfying theorem [ den2d ] ) .
the integral over @xmath335 has the multiplicative error bound directly .
the remainder of the integral is bounded by @xmath336 , which is smaller than the normal integral over @xmath337 ( see the end of the proof of theorem [ den1d ] ) .
\(ii ) the second step is to develop a bound on the ( conditional ) quantile transform in order to approximate an asymptotic normal random variable by a normal one .
the basic idea appears in @xcite .
clearly , from ( [ condden ] ) , @xmath338 for @xmath339 , @xmath340 , and @xmath327 . by condition [ cof ] , the conditional densities ( of the response given @xmath115 )
are bounded above zero on @xmath341 .
hence , the inverse of the above versions of the c.d.f.s also satisfy this multiplicative error bound , at least for the variables bounded by @xmath342 .
thus , the quantile transform can be applied to show that there is a normal random variable , @xmath343 , such that @xmath344 so long as @xmath345 and the quantile transform of @xmath345 are bounded by @xmath342 . using the conditional mean and variance
[ see ( [ mean1 ] ) ] , and the fact that the random variables exceed @xmath342 with probability bounded by @xmath336 ( where @xmath169 can be made large by choosing @xmath49 large enough ) , there is a random variable @xmath346 that can be chosen independently so that @xmath347 except with probability bounded by @xmath336 .
\(iii ) finally , the `` hungarian '' construction will be developed inductively .
let @xmath348 and consider induction on @xmath349 .
first consider the case where @xmath350 ; the argument for @xmath351 is entirely analogous .
define @xmath352 , where @xmath227 bounds the big - o term in any equation of the form ( [ couple ] ) .
let @xmath204 be a bound [ uniform over @xmath353 on @xmath354 in ( [ couple ] ) .
the induction hypothesis is as follows : there are normal random vectors @xmath355 such that @xmath356 except with probability @xmath357 , where for each @xmath358 , @xmath359 has the same covariance structure as @xmath360 , and where @xmath361 note : since the earlier bounds apply only for intervals whose lengths exceed @xmath362 ( for some positive @xmath363 ) , @xmath358 must be taken to be smaller than @xmath364 .
thus , the bound in ( [ epsdef ] ) becomes @xmath365 , as stated in theorem [ den2d ] . to prove the induction result , note first that theorem [ den2d ] ( or theorem [ den1d ] ) provides the normal approximation for @xmath366 for @xmath367 .
the induction step is proved as follows : following @xcite , take two consecutive dyadic rationals @xmath368 and @xmath369 with @xmath370 odd .
so @xmath371/2^{\ell-1 } = \tau\bigl([k/2 ] , \ell-1\bigr).\ ] ] condition each coordinate of @xmath372 on previous coordinates and on @xmath373 , \ell-1))$ ] .
let @xmath374 be one such coordinate .
now , as above , define @xmath375 by @xmath376 , \ell-1\bigr)\bigr ) + r(k , \ell).\ ] ] from ( [ couple ] ) , there is a normal random variable @xmath377 such that @xmath378 , \ell-1\bigr)\bigr ) - z_n(k , \ell ) \bigr| \leq \varepsilon_n^*.\ ] ] by the induction hypothesis for @xmath379 , @xmath373 , \ell-1 ) $ ] is approximable by normal random variables to within @xmath380 ( except with probability @xmath336 ) .
thus , a coordinate @xmath381 , \ell-1 ) $ ] is also approximable with this error , and the error in approximating @xmath382 , \ell- 1 ) $ ] is bounded by @xmath380 times @xmath383 .
finally , since @xmath384 is independent of these normal variables , the errors can be added to obtain @xmath385 therefore , except with probability less than @xmath386 , the induction hypothesis ( [ induct1 ] ) holds with error @xmath387 and the induction is proven .
the theorem now follows since the piecewise linear interpolants satisfy the same error bound [ see @xcite ] .
[ re1 ] under the conditions for the theorems here , the coverage probability for the confidence interval ( [ confint ] ) is @xmath388 , which is achieved at @xmath389 ( where @xmath227 is a constant ) .
sketch of proof recall the notation of remark [ remark2 ] in section [ sec2 ] . using theorem [ th1 ] and the quantile transform as described in the first steps of theorem [ th2 ] ( and not needing the dyadic expansion argument ) ,
it can be shown that there is a bivariate normal pair @xmath390 such that @xmath391\\[-8pt ] { \sqrt{n } } \bigl ( \hat\delta(h_n ) - \delta(h_n)\bigr ) & = & z + r^*_n,\qquad r^*_n = { { \mathcal}{o}}_p \bigl ( n^{-1/2 } ( \log n)^{3/2 } \bigr).\nonumber\end{aligned}\ ] ] note that from the proofs of theorems [ th1 ] and [ th2 ] , the @xmath392 terms above are actually @xmath393 terms except with probability @xmath336 where @xmath169 is an arbitrary fixed constant .
the `` almost sure '' results above take @xmath394 , but @xmath395 will suffice for the bounds on the coverage probability here .
now consider expanding @xmath397 .
first , note that under the design conditions here , @xmath398 will be of exact order @xmath61 ; specifically , if @xmath24 is replaced by @xmath399 , all terms involving @xmath400 will remain bounded , and we may focus on @xmath401 .
note also that for @xmath402 , the terms in the expansion of @xmath403 tend to zero [ specifically , @xmath404 .
so the sparsity , @xmath397 , may be expanded in a taylor series as follows : @xmath405 where @xmath406 is a ( gradient ) vector that can be defined in terms of @xmath407 and @xmath11 ( and its derivatives ) , @xmath408 is a quadratic function ( of its vector argument ) and @xmath409 is a cubic function . note that under the design conditions , all the coefficients in @xmath406 , @xmath408 and @xmath409 are bounded , and so it is not hard to show that all the terms in @xmath410 tend to zero as long as @xmath411 .
specifically , if @xmath412 is of order @xmath413 , then all the terms in @xmath410 tend to zero . also , @xmath55 is within a @xmath3 factor of @xmath414 and @xmath415 is even smaller .
finally , @xmath112 is a difference of two quantiles separated by @xmath416 , and so @xmath417 has variance proportional to @xmath52 .
thus , @xmath418 .
thus , not only does @xmath419 , but powers of this term greater than 2 will also be @xmath420 .
it follows that the coverage probability may be computed using only two terms of the taylor series expansion for the normal c.d.f . :
@xmath421 note that the ( normal ) conditional distribution of @xmath422 given @xmath112 is straightforward to compute ( using the usual asymptotic covariance matrix for quantiles ) : the conditional mean is a small constant ( of the order of @xmath412 ) times @xmath112 , and the conditional variance is bounded .
expanding the lower probability in the same way and subtracting provides some cancelation
. the contribution of @xmath14 will cancel in the @xmath423 differences , and is negligible in subsequent terms since @xmath424 .
similarly , the @xmath425 term will appear only in the @xmath423 difference where it contributes a term that is @xmath426 times a term of order @xmath427 , and will also be negligible in subsequent terms .
also , the @xmath428 term will only appear in @xmath423 , as higher powers will be negligible .
the only remaining terms involve @xmath429 .
for the first power ( appearing in @xmath423 ) , @xmath430 .
for the squared @xmath112-terms in @xmath431 , since var(@xmath417 ) is proportional to @xmath412 , @xmath432 , and all other terms involving @xmath112 have smaller order . therefore , one can obtain the following error for the coverage probability : for some constants @xmath230 and @xmath433 , the error is @xmath434 ( plus terms of smaller order ) . since @xmath55 is of order nearly @xmath61
, the first terms have nearly the same order . using @xmath435
, it is straightforward to find the optimal @xmath412 to be a constant times @xmath436 , which bounds the error in the coverage probability by @xmath437 . | traditionally , assessing the accuracy of inference based on regression quantiles has relied on the bahadur representation .
this provides an error of order @xmath0 in normal approximations , and suggests that inference based on regression quantiles may not be as reliable as that based on other ( smoother ) approaches , whose errors are generally of order @xmath1 ( or better in special symmetric cases ) .
fortunately , extensive simulations and empirical applications show that inference for regression quantiles shares the smaller error rates of other procedures .
in fact , the `` hungarian '' construction of komls , major and tusndy [ _ z . wahrsch .
verw .
gebiete _ * 32 * ( 1975 ) 111131 , _ z. wahrsch . verw .
gebiete _ * 34 * ( 1976 ) 3358 ] provides an alternative expansion for the one - sample quantile process with nearly the root-@xmath2 error rate ( specifically , to within a factor of @xmath3 ) .
such an expansion is developed here to provide a theoretical foundation for more accurate approximations for inference in regression quantile models .
one specific application of independent interest is a result establishing that for conditional inference , the error rate for coverage probabilities using the hall and sheather [ _ j .
r. stat .
soc .
ser .
b stat .
methodol . _
* 50 * ( 1988 ) 381391 ] method of sparsity estimation matches their one - sample rate . |
the central @xmath2100 pc around the galactic center is an extremely complex region containing a variety of astrophysical activities : i.e. , cold and warm molecular clouds , star clusters / formation , supernova remnants ( snrs ) , and hii regions , to name a few . earlier galactic plane surveys with the _ tenma _ and _ ginga _ satellites have discovered the highly ionized fe emission line at @xmath3 @xmath2 6.7 kev , which is strongly enhanced toward the galactic center region @xcite .
_ asca _ and _ bepposax _ observations have confirmed the presence of the ionized fe emission line as well as detecting emission lines from other highly ionized elemental species of si , s , and ar @xcite .
these results suggested that the difuse x - ray emission in the plane and the galactic center regions arises from thermal plasmas with multiple temperatures and ionization states , although the temperatures of the individual plasma components are inconclusive ( e.g. , koyama et al .
1996 ; kaneda et al . 1997 ; valinia & marshall 1998 ; park & ebisawa 2001 ) . the origins of these hot plasmas have been suggested to be multiple supernova explosions and magnetic confinement of the interstellar medium ( ism ) ( e.g. , kaneda et al .
1997 and references therein ) . besides the apparent hot thermal components , the presence of giant molecular cloud complexes ( e.g. , sanders et al . 1984 ) , as well as a number of hard x - ray emitting point sources near the galactic center ( e.g. , skinner et al .
1987 ) has led to the prediction that fluorescent x - ray emission should arise where the cold , dense ism is illuminated by external x - ray sources @xcite .
this proposed new class of x - ray sources has been discovered with _
asca _ by detecting low - ionization - state fe line emission ( fei @xmath4 fexvii ) in the galactic center region emerging at a characteristic energy of @xmath3 @xmath2 6.4 kev @xcite .
a detailed analysis of this low ionization , or `` neutral '' 6.4 kev fe line emission toward the galactic giant molecular cloud sgr b2 has successfully described this fluorescent x - ray emission from the molecular cloud in terms of an x - ray reflection nebula ( xrn ) model @xcite .
strong past activity of the supermassive galactic black hole candidate sgr a * was proposed to be the primary external illuminating source to produce the observed neutral fe line emission from sgr b2 .
a similar xrn feature has also been detected in another galactic center molecular cloud , sgr c @xcite .
alternatively , such neutral fe line emission may also be explained by the bombardment of energetic electrons on metal - rich sn ejecta fragments @xcite .
massive stars appear to be forming at several sites in the central few hundred parsecs @xcite , so the proposed unresolved sn ejecta scenario for the origin of the observed neutral fe line emission is plausible as well .
the x - ray line emission from the hot / cold ism is therefore a useful tool to probe the origins of the complex structure of the ism in the galactic center .
the poor angular resolutions of previous detectors were , however , insufficient for such studies in complex regions of the galactic center . with the superb sub - arcsecond angular resolution and good sensitivity in the 0.3 @xmath4 8 kev band , the _ chandra x - ray observatory _ provides an excellent opportunity to investigate the nature of the diffuse x - ray emission in the galactic center regions .
preliminary results from earlier _ chandra _ observations showed that the diffuse x - ray emission features in the galactic center appear to have a variety of spectral and morphological characteristics @xcite .
the interpretations of these diffuse x - ray features have however been highly speculative primarily due to the limited photon statistics with previous data .
recently , we have performed deep observations of the galactic center with _
the exposure has reached @xmath2600 ks , which is an order of magnitude deeper than previous observations , so that these deep _ chandra _ observations represent an unprecedented wealth of x - ray data on the galactic center .
in fact , these _ chandra _ data have resolved over @xmath22300 point sources within @xmath220 pc of the galactic center @xcite .
besides these point sources , including the target object sgr a * , the _ chandra _ data show a complex structure of the diffuse x - ray emission within the 17@xmath0 @xmath1 17@xmath0 field of view ( fov ) .
we here present the latest deep _
observations of the galactic center . the high angular resolution _
chandra _ data with decent photon statistics from the deep exposure allow us extensive imaging and spectral analysis of detected bright filamentary features .
the observation is described in [ sec : obs ] , and the x - ray image analysis is presented in [ sec : image ] and [ sec : ewi ] .
we present the results from spectral analysis for some selected emission features in [ sec : spec ] , and compare these results with molecular line emission features in [ sec : mc ] .
some implications are then discussed in
[ sec : disc ] .
a summary is presented in [ sec : summary ] .
for the last three years , the galactic supermassive black hole candidate sgr a * has been monitored , as part of guaranteed time observation ( gto ) and guest observer ( go ) programs , with the advanced ccd imaging spectrometer ( acis ) @xcite on board the _ chandra x - ray observatory _ ( table [ tbl : tab1 ] ) . as of 2002 june , combining 11 _ chandra_/acis observations ( except for obsid 1561a , which was severely contaminated by a bright transient source within the fov ) ,
the total exposure has reached @xmath2590 ks , which is the deepest ever observation of the galactic center region in x - rays .
we first applied the algorithm developed by townsley et al .
( 2002a ) to correct the spatial and spectral degradation of the acis data caused by radiation damage , known as charge transfer inefficiency ( cti ; townsley et al .
[ 2000 ] ) .
the expected effects of the cti correction include an increase in the number of detected events and improved event energies and energy resolution @xcite .
we then screened the data by status , grade , and the flight timeline filter .
we have also removed observation time intervals of strong flaring in the background .
all individual event files were then reprojected to the tangent plane at the radio position of sgr a * ( ra[j2000 ] = 17@xmath5 45@xmath6 [email protected] , dec[j2000 ] = @xmath429@xmath8 00@xmath0 28@xmath9118 ) in order to generate the composite data .
the detailed descriptions of these data reduction processes and the resulting broadband raw image from the composite data may be found in muno et al .
( 2003a ) .
during the monitoring observations of sgr a * ( ra[j2000 ] = 17@xmath5 45@xmath6 [email protected] , dec[j2000 ] = @xmath429@xmath8 00@xmath0 28@xmath90 ) , the 17@xmath0 @xmath1 17@xmath0 fov of the acis - i array also imaged the interstellar surroundings of sgr a * , which reveals the complex environment within @xmath220 pc of the galactic center with unprecedented detail in x - rays ( hereafter , we assume a distance of @xmath10 = 8 kpc to the galactic center [ reid 1993 ] ) .
a `` true - color '' x - ray image of the galactic center region is presented in figure [ fig : fig1 ] .
each subband image has been exposure - corrected utilizing the exposure map produced by muno et al .
( 2003a ) , and adaptively smoothed to achieve signal - to - noise ( s / n ) ratio of 4 by using the _ ciao _ tool _
csmooth_. while over @xmath22000 point - like sources prevail over the fov , the point source contribution to the observed x - ray flux is @xmath210% and the bulk of the x - ray emission is _ truly _ diffuse @xcite .
figure [ fig : fig1 ] indeed demonstrates the presence of diffuse extended x - ray emission features on a variety of angular scales .
the target object sgr a * is located near the center of the image where the bright x - ray emission from a massive star cluster within the inner parsec of the galactic center dominates .
the bright diffuse x - ray emission to the immediate east of sgr a * is the snr sgr a east @xcite .
symmetric lobes of soft x - ray emission extended from sgr a * , perpendicular to the galactic plane , are indicated with the yellow dotted line in figure [ fig : fig1 ] .
these features may represent a bipolar outflow from sgr a * , which are extensively discussed by baganoff et al .
( 2003 ) and morris et al .
( 2003 , in preparation ) .
outside of these complex central regions are larger angular scale ( @xmath2a few arcmin ) x - ray features , consisting of a few x - ray `` loops '' around sgr a * ( e.g. , the white dotted curves in figure [ fig : fig1 ] ) .
the soft nature of these loop - like features is typical of thermal shocks interacting with the galactic ism . considering significant star - forming activities in the galactic center regions ,
these red , soft x - ray features are perhaps caused by supernova blast waves or stellar winds / bubbles .
we also detect a few hard x - ray knots and filaments .
the bright knot to the south of sgr a * ( region a in figure [ fig : fig1 ] ) is the x - ray counterpart of the bright , nonthermal radio knot in the boundary shell of the snr [email protected] @xcite .
we have identified at least one more continuum dominated arc - like feature in the northeast of sgr a * ( region b in figure [ fig : fig1 ] ) , which shows a hard photon index .
the detailed analysis of these continuum x - ray sources are presented elsewhere ( baganoff et al . 2003 ; sakano et al .
2003 ; morris et al .
2003 , in preparation ; yusef - zadeh et al .
2003 , in preparation ) .
in contrast , some other blue , hard filaments to the northeast of sgr a * ( regions 1 , 2 , and 3 in figure [ fig : fig1 ] ) exhibit strong neutral fe k@xmath11 line emission @xcite . the origin of these neutral fe line sources has been elusive @xcite , and we will discuss the detailed spectral characteristics of these neutral fe line features in [ sec : spec ] .
in order to examine the overall distributions of the diffuse emission line features around the galactic center region , we construct _ equivalent width _ ( ew ) images for the detected atomic emission lines , following the method described by park et al .
( 2002 ) . after removing
all detected point sources from the broadband image ( see muno et al .
[ 2003a ] for the details of the point source detection ) , subband images for the line and continuum bandpasses were extracted for each spectral line of interest ( table [ tbl : tab2 ] ) .
these subband images were adaptively smoothed to achieve a s / n ratio of 3 @xmath4 4 .
the underlying continuum was calculated by logarithmically interpolating between images made from the higher and lower energy `` shoulders '' of each broad line .
the estimated continuum flux was integrated over the selected line width and subtracted from the line emission .
the continuum - subtracted line intensity was then divided by the estimated continuum on a pixel - by - pixel basis to generate the ew images for each element . in order to avoid noise in the ew maps caused by poor photon statistics near the ccd chip boundaries
, we have set the ew values to zero where the estimated continuum flux is low .
we also set the ew to zero where the integrated continuum flux is greater than the line flux .
we present the ew images of the fe he@xmath12 ( @xmath3 @xmath2 6.7 kev ) , the neutral fe k@xmath11 ( @xmath3 @xmath2 6.4 kev ) , s he@xmath12 + ly@xmath12 ( @xmath3 @xmath2 2.5 kev ) and si he@xmath12 ( @xmath3
@xmath2 1.8 kev ) line emission ( figure [ fig : fig2 ] ) .
we note that cosmic x - ray background ( cxb ) and the particle background ( pb ) , both of which have not been subtracted in the ew generation , could have contaminated our ew images to some extent .
we found that the cxb contribution to the observed fluxes is negligible ( @xmath131% ; e.g. , muno et al .
the pb contamination appears to be more significant than the cxb and may contribute up to @xmath220%@xmath430% of the estimated continuum emission where the surface brightness is low .
the ew estimations are thus affected by the pb contamination by up to @xmath220% .
this embedded contamination is not however critical for our purposes of the ew images : i.e. , we intended to utilize these ew images in order to qualitatively investigate overall variation and distribution of the bright x - ray line emission over the galactic center region .
we do not use the estimated absolute ew values unless the actual regional spectal analysis has been performed .
the adaptive smoothing also introduces some systematic uncertainties for the small - scale , faint ew features which we do not rely on for the analysis .
the large - scale ( @xmath14arcmin ) , relatively bright ew features are , on the other hand , consistent with the actual spectral analysis and are reliable in the context with previous observations ( see the following discussions and sections ) . despite the embedded uncertainties , these ew images are therefore useful as first - order information in order to investigate the overall variation of the x - ray line emission within the fov and to qualitatively discuss some noticeable characteristics .
we also note that the line flux maps as well as the ew maps might also be useful to understand the line emission distributions . since the line flux maps show the _ same _ overall intensity variations as the ew maps , we do not explicitly present the line flux maps although our discussion on the line features is in parts based on the line flux distribution in addition to the ew variations .
the fe he@xmath12 line ew is generally enhanced along the plane .
the enhancements in the fe he@xmath12 line ew of this galactic plane component are marginal ( roughly around @xmath2400 ev ) , which is consistent with the _ asca _ measurements of the hard galactic ridge x - ray emission ( grxe ) component ( @xmath3 @xmath2 8 kev ) @xcite .
this marginal ew enhancements are also consistent with those from the detected point sources in the galactic center @xcite
. a fraction of this fe he@xmath12 ew should thus originate from unresolved point sources .
the substantial ew variations on arcmin scales , however , suggest that a bulk of this hot fe line emission has a diffuse origin arising from a hot ( @xmath3 @xmath2 5 @xmath4 10 kev ) plasma .
sgr a east is a strong fe he@xmath12 line source @xcite , and is immediately recognized as an extremely bright source with an @xmath21@xmath153 diameter in the center of the fe he@xmath12 ew image ( figure [ fig : fig2]a ) .
the estimated fe he@xmath12 line ew is @xmath22500 ev @xmath4 5000 ev within the central @xmath230@xmath16 diameter region of sgr a east , which is consistent with the results from the spectral analysis by maeda et al .
this strong fe ew appears to be from hot fe - rich ejecta heated by the reverse shock of the snr sgr a east as posited by maeda et al .
( 2002 ) .
the neutral fe ew is largely distributed along the plane , and , in contrast to the he - like fe ew , is strongly enhanced in the northeast of sgr a * ( figure [ fig : fig2]b ) .
the bright , hard fe knots in the northeast of sgr a * ( regions 1 , 2 , and 3 in figure [ fig : fig1 ] ) are featureless in the fe he@xmath12 ew image ( figure [ fig : fig2]a ) , but remarkably bright in the neutral fe ew image ( figure [ fig : fig2]b ) .
the regions to the northeast of sgr a * are also bright in the si and s ew images ( figure [ fig : fig2]c and figure [ fig : fig2]d ) .
the si and s maps represent high ionization states in hot thermal plasmas which are typically produced by snr and/or star - forming activity .
the extinction toward the hot s and si emission in the northeast of sgr a * is consistent with that of typical column density toward the galactic center ( see [ sec : spec ] , also muno et al .
2003b ) , which indicates the galactic center origin of the hot gas rather than a nearby foreground component .
it is interesting that these high - ionization - state hot s , si material and the low - ionization - state cold fe sources are in general coincident in the northeast of sgr a*. on the other hand , albeit that the angular resolution and the positional accuracy of the ew images are significantly degraded due to the adaptive smoothing of faint diffuse emission , the one - to - one correlations between the bright s / si ew and the neutral fe ew enhancements appear to be weak : i.e. , the locations of the strong enhancements of hot and cold gas are probably not `` exactly '' coincident .
these cold , neutral fe line emitting gas and the hot , ionized s / si line emitting plasma also appear to have galactic center molecular cloud counterparts ( see [ sec : mc ] ) .
these features suggest that the high- and low - ionization - state line emission might be associated with the star - forming / sn activities nearby to the dense molecular cloud complexes in the northeast of sgr a*. both of s / si and neutral fe ews are weak in the southwest of sgr a * , which appears to be caused by intrinscally less activities there ( for more discussion , see [ sec : mc ] and [ sec : disc ] ) .
we perform a spectral analysis of a few diffuse emission features , particularly those with strong neutral fe ews .
the individual acis observations comprising our images utilized different roll - angles : i.e. , the first three observations have @xmath2270@xmath8 roll - angles and the remainder @xmath280@xmath8 .
the fe knots have thus been detected on either acis - i0 or acis - i3 chips depending on the roll - angle . about @xmath285% of the total exposure
was taken with roll - angles of @xmath280@xmath8 , however , so the bulk of the photons from these fe knots have been detected on the acis - i0 chip .
we thus utilize the response matrices appropriate for acis - i0 .
the fe knots 1 , 2 , and 3 contain @xmath212000 , @xmath224000 , and @xmath211000 photons , respectively , and we bin the data to contain a minimum of 60 @xmath4 200 counts per channel for the spectral fitting .
these emission features are fairly bright and have well - confined structures , and so we estimate the background contamination by extracting spectra from nearby source - free regions after removing all detected point - like sources from the data . for the spectral analysis of our cti - corrected data , we utilize the response matrices appropriate for the spectral redistribution of the ccd , as generated by townsley et al .
( 2002b ) .
the low energy ( @xmath3 @xmath13 1 kev ) quantum efficiency ( qe ) of the acis has degraded because of the molecular contamination on the optical blocking filter . although effects from this qe degradation on our heavily absorbed spectra are most likely negligible , we correct this time - dependent qe degradation by modifying the ancillary response function ( arf ) for each extracted spectrum by utilizing the idl acisabs software .
the x - ray spectral characteristics of knots 1 and 2 are similar to each other : i.e. , a strong 6.4 kev fe emission line with underlying power - law continuum of photon index @xmath17 @xmath2 2 @xmath4 3 ( figure [ fig : fig3]a and figure [ fig : fig3]b ) .
both spectra are heavily absorbed ( @xmath18 @xmath2 3.5 @xmath1 10@xmath19 @xmath20 ) and the photon fluxes are insignificant below @xmath3 @xmath13 4 kev .
we thus present the best - fit parameters based on spectral fittings in the 4 @xmath4 8 kev band ( table [ tbl : tab3 ] ) .
the measured ew of the fe line is @xmath21 kev for both features , which is consistent with that of the sgr c cloud @xcite .
we note that the knot 1 spectrum may also suggest the presence of an fe k@xmath21 line at @xmath3 @xmath2 7 kev ( figure [ fig : fig3]a ) .
we have added another gaussian line component in order to fit this weak feature , assuming a narrow line width .
the best - fit k@xmath21 line center energy is 7.02@xmath22 kev with an ew of @xmath2200 ev .
the measured k@xmath21 to k@xmath11 line flux ratio is [email protected] .
this k@xmath21 line feature is not apparent in the knot 2 spectrum , at least with the current data .
we place an upper limit ( 2@xmath24 ) on the k@xmath21 line ew of 120 ev for knot 2 .
the fe knot 3 spectrum shows the same characteristics as the spectrum of the other two knots in the hard x - ray band ( @xmath3 @xmath25 4 kev ) : i.e. , a strong 6.4 kev fe k@xmath11 line with underlying continuum of @xmath17 @xmath2 1.8 ( figure [ fig : fig3]c ) .
the measured 6.4 kev fe line ew is @xmath21.3 kev .
this spectrum also appears to show a weak fe k@xmath21 line feature as seen in the knot 1 spectrum .
the best - fit line center energy of this weak feature , the measured ew , and the line flux ratio between k@xmath21 to k@xmath11 lines are consistent with those of knot 1 .
the knot 3 spectrum , on the other hand , is less absorbed ( @xmath18 @xmath2 1.5 @xmath1 10@xmath19 @xmath20 ) than knots 1 and 2 , revealing remarkably different features than knots 1 and 2 in the soft band ( @xmath3 @xmath26 4 kev ; figure [ fig : fig3]c ) .
the soft band x - ray spectrum of knot 3 is dominated by thermal emission with a strong s he@xmath12 line . for comparison
, a single power law model fit is relatively poor ( @xmath27 = 1.4 ) and the inclusion of an additional thermal plasma model improves the overall fit significantly based on an f - test .
the strong s line emission from the soft component of the knot 3 spectrum indicates x - ray emission from a thermal plasma having an electron temperature of @xmath28 @xmath2 0.3 kev .
the metal abundance for the soft component appears to be high and is best - fitted with a few times solar , although unconstrained with the current data ( e.g. , a 2@xmath24 lower limit is @xmath20.5 solar ) .
this soft thermal component appears to be associated with the galactic center molecular clouds ( [ sec : mc ] ) , suggesting its location in the galactic center instead of a nearby foreground origin .
the spectral fit with a two - component absorption for knot 3 indeed indicates a significant foreground column ( n@xmath29 @xmath2 1 @xmath1 10@xmath19 @xmath20 ) even only for the soft thermal component .
the nearby foreground origin for this soft thermal emission is therefore unlikely .
the best - fit parameters for the knot 3 spectrum are presented based on the spectral fitting in the 2 @xmath4 8 kev band ( table [ tbl : tab3 ] ) .
the spatial distribution and spectral x - ray characteristics revealed by the data suggest that some of the diffuse x - ray emission features in the galactic center be associated with the cold , dense component of the ism and/or active star - forming regions .
we thus examine galactic center molecular cloud maps in order to investigate correspondences between dense clouds and the detected x - ray emission features . considering the high molecular densities in the galactic center region , the cs molecule , with its relatively high critical density , @xmath30(h@xmath31 ) @xmath2 10@xmath32 @xmath33 , is an appropriate tracer of dense molecular clouds there .
the cs line maps are , unlike those of co , expected to be nearly free of strong contamination by the galactic disk and are more representative of the molecular column density rather than the surface temperatures of clouds ( tsuboi et al .
1999 and references therein ) .
the cs data also provide an angular resolution of @xmath230@xmath16 ( tsuboi et al .
1999 ) , allowing a meaningful cross - comparison with the high resolution _
chandra _ image .
we have thus searched for potential molecular cloud counterparts to x - ray features by investigating various velocity components of the 49 ghz cs ( j = 1@xmath40 ) line emission @xcite .
the overall distribution of dense molecular clouds in the velocity range @xmath34 = @xmath425 to @xmath3550 km s@xmath36 is along the galactic plane , and we compare their distributions with our ew images of neutral fe and highly ionized s line emission ( figure [ fig : fig4 ] ) .
the molecular cloud contours in the presented velocity ranges display a general correspondence with the x - ray emission - line ew distributions , particularly to the northeast of sgr a*. the detailed comparisons indeed suggest an intimate relationship , but higher spatial resolution in the molecular line maps will be required to unambiguously demonstrate the nature and geometry of the physical relationship .
the + 40 to + 50 km s@xmath36 clouds , which , according to vollmer et al .
( 2003 ) , are located close to , but slightly in front ( @xmath25 pc ) of sgr a * , are well correlated spatially with the s and the neutral fe ew images throughout the region to the northeast of sgr a * ( figure [ fig : fig4]e and figure [ fig : fig4]f ) .
the most prominent cloud at + 10 to + 30 km s@xmath36 the `` 20 km s@xmath36 molecular cloud '' located to the south of sgr a * is located substantially in front ( @xmath220 @xmath4 50 pc ) of sgr a * @xcite , and does not show an x - ray emission line counterpart . in this same velocity range , there is a smaller molecular cloud in the northeast corner of the fov ( figure [ fig : fig4]c and figure [ fig : fig4]d ) which is adjacent to the radio arc and may be interacting with the arc @xcite .
this cloud also appears to be spatially correlated with the most prominent emission in the ew images , and with knot 3 in particular .
this correlation may be best demonstrated by the high angular resolution 98 ghz cs ( j=2@xmath41 ) data @xcite overlaid with the neutral fe line ew image ( see the lower - left corner inset in figure [ fig : fig4]c ) .
the line - of - sight locations of the negative velocity clouds remains undetermined . except for knot 1 ( see below ) , the molecular emission in the @xmath425 to @xmath45 km s@xmath36 range appears to be anti - correlated with the s and neutral fe ew maps in the northeast corner of the fov ( figure [ fig : fig4]a and [ fig : fig4]b ) .
some of the bright , hard x - ray filaments have interesting spatial correspondences with cs clouds .
knot 1 falls upon a strong cs line peak at negative velocities ( figure [ fig : fig4]a ) , although it also coincides with a prominent ridge in the + 50 km s@xmath36 cloud ( figure [ fig : fig4]e ) .
knot 2 is superimposed on the edge of those same molecular clouds , so its relationship with those clouds , if any , is unclear . and , as mentioned above , knot 3 has plausible molecular counterparts in the velocity ranges + 10 to + 30 km s@xmath36 and + 40 to + 50 km s@xmath36 ( figure [ fig : fig4]c and figure [ fig : fig4]e ) . toward the south and the southwest of sgr a *
, the overall broadband surface brightness , as well as the si and s line fluxes , are relatively low , and little correspondence with molecular gas is evident .
the foreground `` 20 km s@xmath36 '' cloud , as represented by the emission integrated from + 10 to + 30 km @xmath36 in figures [ fig : fig4]c and [ fig : fig4]d , is prominent in this region .
this dense molecular cloud complex ( @xmath37 @xmath2 2 @xmath1 10@xmath32 @xmath33 ) implies a significant column density ( @xmath18 @xmath2 4 @xmath1 10@xmath19 @xmath20 ) and a correspondingly large visual extinction , @xmath38 @xmath14 400 mag @xcite , so this cloud should be optically thick in the soft x - ray band .
this possibly accounts for the observed low intensities in the broadband surface brightness and for the soft x - ray line fluxes to the south of sgr a*. the relative lack of detected x - ray point sources toward this 20 km s@xmath36 cloud @xcite is consistent with a strong absorption effect .
we also note that the si , s , and neutral fe line fluxes and the ews to the southwest of sgr a * where the negative velocity cloud is present , but where the 20 km s@xmath36 cloud is absent , are weak .
this stands in contrast to the strong line emission in si , s , and neutral fe toward dense clouds in the northeast portion of the fov .
the low line fluxes in this southwest region can not be caused by absorption in the 20 km s@xmath36 cloud simply because of the absence of this cloud in this direction .
it is also unlikely to be caused by absorption in the negative velocity cloud because there is no evidence for substantial obscuration of the x - ray point sources in this southwest region toward the negative velocity cloud @xcite .
thus , the weak x - ray line features to the southwest of sgr a * , where the 20 km s@xmath36 cloud is absent , appear to be intrinsic : that is , this region is lack of supernova activity , or perhaps of energetic star - forming activity compared to the northeast regions .
based on the spectral analysis presented in
[ sec : spec ] , we realize that the characteristics of the knot 3 spectrum are consistent with those of knot 1 , except for the low absorbing column density which reveals the soft thermal component .
the column density toward knot 3 ( @xmath18 @xmath2 1.5 @xmath1 10@xmath19 @xmath20 ) is consistent with typical columns to the galactic center @xcite .
knot 1 ( and 2 ) , on the other hand , has significantly higher column ( @xmath18 @xmath2 3.5 @xmath1 10@xmath19 @xmath20 ) .
we speculate that the larger column for knot 1 ( and 2 ) might be attributable to the negative velocity molecular cloud in this direction .
consistent with this hypothesis is the fact that sgra east , which has the same column as knot 3 @xcite , is not coincident with emission in the @xmath425 to @xmath45 km s@xmath36 cs map ( figure [ fig : fig4 ] ) .
the apparent angular size of the @xmath425 to @xmath45 km s@xmath36 cloud to the northeast of sgr a * is @xmath26@xmath0 , so the physical extent is @xmath215 pc . if the bulk of the excess column toward knot 1 ( @xmath18 @xmath2 2 @xmath1 10@xmath19 @xmath20 ) is attributed to this cloud , a cloud density of @xmath30 @xmath2 4 @xmath1 10@xmath39 @xmath33 is implied . assuming a simple spherical geometry for the cloud , the total mass of the cloud is @xmath40 @xmath2 2 @xmath1 10@xmath41 @xmath42 .
albeit crude estimates , the derived density and mass of the cloud are not unreasonable compared with other giant molecular cloud complexes in the galactic center . based on these x - ray absorption features , we therefore propose that this cloud is in the `` foreground '' of fe knot 1 ( and 2 ) , although some portions appear to be associated with knot 1 itself .
the soft x - ray emission to the northeast of sgr a * , as presented by the s ew images ( figure [ fig : fig4]b ) , is apparently anti - correlated with the negative velocity cs emission , which can be accounted for by the presence of this absorbing foreground cloud .
the bright , hard x - ray filaments to the northeast of sgr a * stand out for having strong neutral fe emission .
the spectral characteristics of knot 1 , for example , are consistent with those of the sgr b2 and sgr c clouds , both of which have been plausibly described by an xrn model illuminated by past activity of sgr a * @xcite .
knot 1 has a cometary morphology , with the `` head '' in the north and extended south by @xmath21@xmath0 , along a minor ridge of the negative velocity cs cloud .
the spectral features of the strong 6.4 kev fe k@xmath11 line , a weak 7.02 kev fe k@xmath21 line , and the underlying continuum with @xmath17 @xmath2 2 are typical for the proposed xrn model @xcite .
while the fe k@xmath11 ew of @xmath141 kev is weaker than that from the sgr b2 xrn @xcite , it nonetheless suggests that the fe line emission is dominated by irradiation from external sources ( sunyaev & churazov 1998 and references therein ) . since the thomson cross - section is dependent on the scattering angle of the incident photon , while the fluorescent line emission is isotropic , the observed ew the ratio of fluorescent line intensity to the scattered continuum is inversely proportional to 1 + cos@xmath43@xmath44 , where @xmath44 is the angle subtended at the scatterer between the observer and the illuminating source .
if we assume that sgr a * was the primary external illuminating source for knot 1 , the lower fe ew for the knot 1 spectrum , relative to that of sgr b2 , therefore suggests that @xmath44 is near 0 or @xmath45 and thus that the primary reflecting cloud toward knot 1 is located in front of or behind the plane between sgr a * and sgr b2 .
( the ew may as well depend on other cloud parameters , such as the optical depth , abundances , and the ionization state .
extensive studies of the determinants of the ew characteristics may thus require detailed xrn simulations , which are beyond the scope of the current work . )
the fe k@xmath21 to k@xmath11 line flux ratio , @xmath20.15 , is consistent with the laboratory prediction for the xrn model @xcite .
the sharp absorption edge at @xmath3 @xmath2 7.1 kev is also in good agreement with previously reported xrn interpretations @xcite .
the strong neutral fe line emission from knot 1 is thus well described by the xrn model .
knot 2 shares the overall spectral characteristics of knot 1 .
however , the molecular counterpart is weak or absent for knot 2 , and the apparent absence of the fe k@xmath21 line in the knot 2 spectrum is problematical for the xrn model .
the identification of knot 2 as an xrn might thus be inconclusive .
the knot 3 spectrum shows similar features to the knot 1 spectrum in the hard band ( @xmath3 @xmath25 4 kev ) , but reveals a substantial contribution from a thermal component in the soft band ( @xmath3 @xmath26 4 kev ) . the implied electron temperature ( @xmath28 @xmath2 0.3 kev ) and
likely high abundances might be suggestive of emission from sn ejecta heated by a blast wave reverse shock .
this suggested contribution from snr(s ) to the observed knot 3 spectrum is not surprising because there are dense molecular cloud complexes in these regions of the galactic center , which may contain active star - forming regions .
in fact , the cs maps indicate associations of dense clouds with the x - ray emitting material at the position of knot 3 ( figures [ fig : fig4]c,[fig : fig4]d,[fig : fig4]e , and [ fig : fig4]f ) .
based on the xrn model , a past state of dramatically enhanced activity of sgr a * , with an x - ray luminosity of @xmath46 @xmath14 10@xmath47 ergs s@xmath36 , has been proposed as the primary external illuminating source for the sgr b2 and sgr c clouds @xcite .
the overall large - scale geometry of the neutral fe ew enhancements to the northeast of sgr a * is consistent with the fe emission lying predominantly within a conical region with an apex toward sgr a * ( figure [ fig : fig2]b ) .
we thus consider that sgr a * might have also been the external irradiating source for the detected neutral fe knots . in the reflection models ,
the luminosity of the irradiating external source can be derived in terms of the observed 6.4 kev photon flux , the distance to the reflecting cloud from the source , and the apparent angular size of the cloud as viewed from the source position @xcite .
assuming a solar abundance for fe and a thomson scattering depth @xmath48@xmath49 @xmath2 0.1 ( @xmath18 @xmath2 10@xmath19 @xmath20 ) , we derive an irradiating luminosity from sgr a * , @xmath46 @xmath2 10@xmath50 ergs s@xmath36 for knot 1 and @xmath46 @xmath2 10@xmath51 ergs s@xmath36 for knot 3 .
these luminosities are @xmath22 @xmath4 3 orders of magnitude higher than that of the quiescent state flux of sgr a * @xcite , but only @xmath260.1% of those which might have produced the xrne in sgr b2 and sgr c clouds .
the x - ray luminosity of sgr a * from the recently detected @xmath23 hr flaring event ( @xmath46 @xmath2 10@xmath50 ergs s@xmath36 ; baganoff et al .
[ 2001 ] ) is comparable to the derived irradiating luminosity . considering the physical sizes of these fe knots ( @xmath22 pc for the angular sizes of @xmath21@xmath0 at @xmath10 = 8 kpc ) ,
however , the duration of the irradiating event must have been a few years or longer instead of on the order of @xmath2hours .
the inferred sgr a * activity , if it was the irradiating source , should thus have been persistent much longer than the observed short flares .
the inferred irradiating luminosities might have also been generated by galactic transient sources such as low mass x - ray binaries .
there are likely a number of transient sources in active regions such as the galactic center .
in fact , several transient x - ray sources have been detected near the galactic center with x - ray luminosities of @xmath46 @xmath2 10@xmath50 @xmath4 10@xmath52 ergs s@xmath36 ( e.g. , branduardi et al .
1976 ; maeda et al . 1996
; pavlinsky et al .
although durations of x - ray transient outbursts can be as long as a decade , typical burst periods are relatively short on the order of months .
( e.g. , wijnands 2001 ) .
the nearby x - ray transient sources may thus not be completely ruled out as contributors to the observed neutral fe line emission , but unlikely . in terms of continuous irradiation sources , it is noteworthy that the derived irradiating fluxes are consistent with the x - ray luminosity of sgr a east @xcite .
the sky position of sgr a east is between sgr a * and the fe knots , and so the steady emission from sgr a east might have also contributed the reflected neutral fe line emission at the positions of the observed fe knots . the measured temperatures for sgr a east , @xmath28 @xmath2 2 kev with _ chandra _ @xcite and @xmath28 @xmath2 4 kev with _ xmm - newton _
@xcite , however , appear to be too soft to produce a strong fe line at @xmath3 = 6.4 kev . moreover , if sgr a east is a steady illuminating source capable of producing the observed xrn features , a strong 6.4 kev fe line feature would also have been expected from the nearby neutral ism , for example , the circumnuclear disk of sgr a * ( e.g. , latkoski et al .
1999 and references therein ) , which is apparently not the case .
steady emission from sgr a east is thus ruled out as a primary irradiating source . alternatively , the characteristic 6.4 kev fe line emission may originate from unresolved sn ejecta . fast nonthermal particles accelerated by the fermi mechanism in collisionless magnetohydrodynamic shocks diffuse through cold , metal - rich ejecta knots , and produce low - ionization - state k@xmath11 line x - ray emission @xcite .
this is most effective when the ejecta knot is propagating through dense clouds ( @xmath37 @xmath2 1000 @xmath33 ) @xcite .
the model predicts an underlying , non - thermal bremsstrahlung continuum which is harder ( @xmath17 @xmath2 1.5 ) than that expected from the xrn model ( @xmath17 @xmath2 2 ) .
x - ray line broadenings caused by the fast - moving unresolved sn ejecta are also expected .
the presence of the soft thermal component , particularly in knot 3 , could be apparent evidence for such a contribution from unresolved sn ejecta .
although the observed continuum of knot 3 appears to be softer than the typical prediction from the sn ejecta model , the large uncertainties on the best - fit photon index of the observed underlying continuum still make the ejecta model a plausible candidate to describe the spectum .
although a soft thermal component is not detected from knot 1 and knot 2 , the neutral fe line characteristics are nearly identical for all of these knots .
the absence of soft thermal component from knot 1 and knot 2 may thus be merely due to absorption , and then the supernova ejecta model can not be ruled out as a source of knot 1 and knot 2 as well .
the estimated fe line luminosity ( @xmath46 @xmath2 10@xmath53 ergs s@xmath36 ) can be produced by fast - moving ejecta from a single sn explosion interacting with a dense molecular cloud @xcite .
we note that large - scale correlations between the neutral fe line and the sio ( j=1@xmath40 ) emission in the galactic center regions have been observed @xcite .
this gas phase sio emission due to grain processing by strong shocks , being spatially coincident with the 6.4 kev fe line features , may support the sn ejecta origin of the neutral fe line emission in the galactic center .
one way to distinguish between the xrn and sn ejecta models is to make conclusive measurements of the fe line broadenings .
the detector position of knot 3 is close to the ccd readout ( rows @xmath26 256 ) , so that the energy resolution is achieved down to the instrumental limit ( @xmath2150 ev fwhm ) @xcite .
the measured line width of the 6.4 kev line is then @xmath24 = 10@xmath54 ev for the fitted gaussian line profile .
this is only a weak detection of line width , yet suggests that the fe line width might have been resolved , thus indicating a line broadening in this region . if confirmed ( by follow - up observations ) , this line broadening would provide support for the contribution from unresolved sn ejecta to the observed neutral fe line emission .
observations with instruments of good spectral resolution such as _ astro - e2 _ should be helpful for this purpose .
better constraints on the photon index of the underlying continuum may also be helpful in order to understand the detailed nature of these features .
follow - up deep observations with _ xmm - newton _ , by utilizing the high sensitivity compared to the _ chandra_/acis , would be useful for this purpose .
the sn ejecta model also predicts strong infrared emission from the radiative shocks that accompany ejecta fragments @xcite , and so infrared line observations would help to discriminate models as well .
although the sn ejecta contribution to the observed neutral fe line emission appears to be plausible , we note a caveat that non - thermal radio emission features have not been reported in the northeast regions of sgr a * within the acis fov .
this absence of the non - thermal radio features is thus admittedly a difficult problem in terms of the snr interpretation .
the presence of highly ionized elemental species , si , s , and fe indicates that thermal diffuse x - ray emission , as well as the nonthermal component , also prevails in the galactic center .
these emission lines from the he - like species over the wide range of photon energies ( @xmath3 @xmath2 2 kev for si / s and @xmath3 @xmath2 7 kev for fe ) confirm the multiple - temperature origin for the diffuse thermal emission in the galactic center / plane as indicated by the _ asca
_ data ( e.g. , kaneda et al . 1997 ) .
the thermal components are , with some exceptions of bright features near sgr a * , relatively faint and large - scale features .
the direct spectral analysis of this diffuse thermal emission thus requires substantially more careful treatment than the bright , well - confined , nonthermal neutral fe line features .
we will present the extensive spectral analysis of the galactic center thermal diffuse x - ray emission in subsequent papers ( muno et al . 2003b ; morris et al .
2003 , in preparation ) .
we here qualitatively discuss the overall characteristics of the thermal emission based primarily on the image analysis .
the si and s line ews trace enhancements of the soft component of the thermal emission , which is enhanced to the northeast of sgr a * ( figure [ fig : fig2]c and figure [ fig : fig2]d ) .
these regions are close to the radio arc and the arches cluster , where the formation of massive stars has taken place in at least two episodes over the past 5 million years ( figer et al .
1999 ; yusef - zadeh et al .
massive galactic center molecular clouds are present in these regions ( c.f . , [ sec : mc ] ) , so the observed si and s line emission most likely arises from the hot ism heated by powerful stellar winds and/or supernova blast waves .
as we have discussed in previous sections , the overall broadband x - ray surface brightness and the s / si ews are low in the southwest regions , which is uncorrelated with the absorption effect .
this suggests that the active star forming processes are present _ only _ in the northeast of sgr a * and that such activity is not manifested in the southwest .
knot 3 provides a sample spectrum of this soft thermal component .
as presented in the previous section , this thermal spectrum shows characteristics typical of hot plasma produced by snrs .
the _ asca _ observations of the grxe have successfully demonstrated that the soft component emission in the plane can be furnished by snrs with a sn rate of @xmath2one per century @xcite .
for the case of knot 3 , the angular size of the emission region ( @xmath21@xmath0 ) corresponds to @xmath22 pc at a distance of 8 kpc , which is typical for young galactic snrs , but might also be of a relatively older age in a dense environment such as the galactic center .
the hard thermal component is representd by the he - like fe line ew ( figure [ fig : fig2]a ) . except for the brightest central feature of sgr a east , the highly ionized fe line emission is primarily distributed along the plane .
in contrast to the other ews in figure [ fig : fig2 ] , this hot fe emission shows no particular enhancements in the northeast of sgr a*. the _ asca _ data suggested that this hot component might not be adequately described by the snrs , because such a hot plasma may not be confined in the plane by the galactic gravity and/or by typical interstellar pressure @xcite . in order to maintain this hot plasma in the plane , a high sn rate and a low ionization state of the plasma
are required @xcite .
the electron temperature of the plasma required to produce the observed strong 6.7 kev fe line emission ( @xmath28 = 5 @xmath4 10 kev ) is also unusually high for typical snrs , although we do detect the 6.7 kev fe line from metal - rich ejecta in young snrs ( e.g. , hughes et al .
2000 ; lewis et al .
2003 ) . this hot gas with @xmath28 @xmath2 5@xmath410 kev might rather have been produced when cooler plasma is confined and compressed by strong galactic magnetic fields @xcite . the required magnetic field strength to achieve this is @xmath55 = 20@xmath430 @xmath56 g .
the interstellar magnetic field in the galactic center ( @xmath55 @xmath21 mg , e.g. , yusef - zadeh & morris 1987 ; lang et al .
1999 ) is sufficiently high for this model .
this magnetic confinement model might not require recent massive star - forming activity or sn explosions in order to produce the observed hot plasma .
the relatively uniform distribution of the he - like fe ew along the plane may also indicate this hot plasma s association with the strong magnetic field in the galactic center along the plane rather than with star - forming regions .
deep _ chandra_/acis observations of the galactic center reveal a complex spatial and spectral structure of diffuse x - ray emission with significant photon statistics . the thermal ( e.g. , hot si and s ) and nonthermal ( e.g. , neutral fe ) x - ray line emission is generally enhanced in a region lying to the northeast of sgr a * , a region in which massive star - formation has been active over the past 5 million years ( e.g. , the arches cluster ) , and in which dense galactic center molecular clouds are currently present .
the highly ionized line emission from si and s in these regions is therefore most likely associated with the hot ism heated by the shocks from sne and/or massive stellar winds , while shocks interacting with dense , cold ism may produce emission in the neutral fe line .
curiously , we detect little or no enhancement of the emission lines to the south and southwest of sgr a*. this weak line emission to the south is apparently the result of an absence of activity there , in contrast to the northeast regions , and it is owed in part to the substantial absorption by the foreground 20 km s@xmath36 cloud .
in contrast , the highly ionized he - like fe line emission shows a more uniform distribution along the plane , with no apparent enhancements to the northeast of sgr a*. this feature suggests that the young star - forming and/or sne activities have less effect on the he - like fe distribution than they do on the neutral fe line emission .
the hot fe line emission might rather be caused by the hot ism trapped and then heated by the strong galactic center magnetic fields .
the spectral characteristics of the strong neutral fe line emitting features to the northeast of sgr a * are plausibly described by the xrn model .
we identify dense molecular counterparts for these fe knots , which is strong support for the hypothesized xrn origin . if past activity of sgr a * is the primary external illuminating source , the required luminosities are @xmath46 @xmath2 10@xmath50@xmath410@xmath51 ergs s@xmath36 with durations of at least a few years .
alternatively , nonthermally - accelerated electrons by sn shocks penetrating back into dense , cold supernova ejecta may also explain the observed neutral fe line features .
the presence of active star - forming regions and the dense molecular clouds near the neutral fe features are supportive of this scenario .
unambiguous discriminations between these models are however difficult with the current data .
follow - up observations with the current and upcoming observatories may be helpful for a detailed understanding of the origins for these features .
baganoff , f. k. , bautz , m. w. , brandt , w. n. , chartas , g. , feigelson , e. d. , garmire , g. p. , maeda , y. , morris , m. , ricker , g. r. , townsley , l. k. , & walter , f. 2001 , nature , 413 , 45 baganoff , f. k. , maeda , y. , morris , m. , bautz , m. w. , brandt , w. n. , cui , w. , doty , j. p. , feigelson , e. d. , garmire , g. p. , pravdo , s. h. , ricker , g. r. , & townsley , l. k. 2003 , apj , 591 , 891 bamba , a. , murakami , h. , senda , a. , takaki , s. , yokogawa , j. , & koyama , k. 2002 , proceedings of the symposium `` new visions of the x - ray universe in the xmm - newton and chandra era '' , noordwijk - nl , nov . 2001 ( astro - ph/0202010 ) branduardi , g. , ives , j. c. , sanford , p. w. , brinkman , a. c. , & maraschi , l. 1976 , mnras , 175 , 47 bykov , a. m. 2002 , a&a , 390 , 327 figer , d. f. , kim , s. s. , morris , m. , serabyn , e. , rich , r. m. & mclean , i. s. 1999 , apj , 525 , 750 garmire , g. p. , bautz , m. w. , ford , p. g. , nousek , j. a. , & ricker , jr . , g. r. 2003 , `` x - ray and gamma - ray telescopes and instruments for astronomy '' , proceedings of spie , ed . , j. e. trmper & h. d. tananbaum , vol .
4851 , 28 ho , p. t. , jackson , j. m. , barrett , a. h. , & armstrong , t. 1985 , apj , 288 , 575 hughes , j. p. , rakowski , c. e. , burrows , d. n. , & slane , p. o. 2000 , apj , 528 , l109 kaneda , h. , makishima , k. , yamauchi , s. , koyama , k. , matsuzaki , k. , & yamasaki , n. y. 1997 , apj , 491 , 638 koyama , k. , makishima , k. , tanaka , y. , & tsunemi , h. 1986 , pasj , 38 , 121 koyama , k. , ikeuchi , s. , & tomisaka , k. 1986b , pasj , 38 , 503 koyama , k. , awaki , h. , kunieda , h. , takano , s. , tawara , y. 1989 , nature , 339 , 603 koyama , k. , maeda , y. , sonobe , t. , takeshima , t. , tanaka , y. , & yamauchi , s. 1996 , pasj , 48 , 249 lang , c. , morris , m. , & echevarria , l. 1999 , apj , 526 , 727 latvakoski , h. m. , stacey , g. j. , gull , g. e. , & hayward , t. l. 1999 , apj , 511 , 761 lewis , k. t. , burrows , d. n. , hughes , j. p. , slane , p. o. , garmire , g. p. , & nousek , j. a. 2003 , apj , 582 , 770 maeda , y. , koyama , k. , sakano , m. , takeshima , t. , & yamauchi , s. 1996 , pasj , 48 , 417 maeda , y. , baganoff , f. k. , feigelson , e. d. , morris , m. , bautz , m. w. , brandt , w. m. , burrows , d. n. , doty , j. p. , garmire , g. p. , pravdo , s. h. , ricker , g. r. , & townsley , l. k. 2002 , apj , 570 , 671 makishima , k. 1997 , in x - ray imaging and spectroscopy of cosmic hot plasmas , ed . , f. makino ( tokyo : universal academy press ) , 137 martin - pintado , j. , de vicente , p. , rodriguez - fernndez , n. j. , fuente , a. , & planesas , p. 2000 , a&a , 356 , l5 morris , m. 1993 , apj , 408 , 496 muno , m. p. , baganoff , f. k. , bautz , m. w. , brandt , w. n. , broos , p. s. , feigelson , e. d. , garmire , g. p. , morris , m. r. , ricker , g. r. , & townsley , l. k. 2003a , apj , 589 , 225 muno , m. p. , baganoff , f. k. , bautz , m. , broos , p. s. , feigelson , e. d. , garmire , g. p. , morris , m. r. , park , s. , ricker , g. r. , & townsley , l. k. 2003b , apj , submitted murakami , h. , koyama , k. , sakano , m. , & tsujimoto , m. 2000 , apj , 534 , 283 murakami , h. , koyama , k. , tsujimoto , m. , maeda , y. , & sakano , m. 2001a , apj , 550 , 297 murakami , h. , koyama , k. , & maeda , y. 2001b , apj , 558 , 687 park , s. , & ebisawa , k. , 2001 , apj , 547 , 777 park , s. , roming , p. w. a. , hughes , j. p. , slane , p. o. , burrows , d. n. , garmire , g. p. , & nousek , j. a. , 2002 , apj , 564 , l39 pavlinsky , m. n. , grebenev , s. a. , & sunyaev , r. a. 1994 , apj , 425 , 110 reid , m. j. 1993 , ara&a , 31 , 345 sakano , m. , mishiuchi , m. , maeda , y. , koyama , k. , & yokogawa , j. 1998 , iaus , `` the central regions of the galaxy and galaxies '' , ed . , y. sofue ( london : kluwer ) , 184 , 443 sakano , m. , warwick , r. s. , & decourchelle , a. 2002 , the proceedings of `` japan - germany workshop on galaxies and clusters of galaxies '' , shuzenji , japan , ( astro - ph/0212464 ) sakano , m. , warwick , r. s. , decourchelle , a. , & predehl , p. 2003 , mnras , 340 , 747 sanders , d. v. , solomon , p. m. , & scoville , n. z. 1984 , apj , 276 , 182 sidoli , l. , & mereghetti , s. 1999 , , 349 , l49 skinner , g. k. , willmore , a. p. , eyles , c. j. , bertram , d. , & church , m. j. 1987 , nature , 330 , 544 sunyaev , r. , markevitch , m. , & pavlinsky , m. 1993 , apj , 407 , 606 sunyaev , r. , & churazov , e. 1998 , mnras , 297 , 1279 townsley , l. k. , broos , p. s. , garmire , g. p. , & nousek , j. a. 2000 , apj , 534 , l139 townsley , l. k. , broos , p. s. , nousek , j. a. , & garmire , g. p. 2002a
, nuclear instruments & methods in physics research section a , 486 , 751 townsley , l. k. , broos , p. s. , chartas , g. , moskalenko , e. , nousek , j. a. , & pavlov , g. g. 2002b , nuclear instruments & methods in physics research section a , 486 , 716 tsuboi , m. , ukita , n. & handa , t. 1997 , apj 481 , 263 tsuboi , m. , handa , t. & ukita , n. 1999 , apjs , 120 , 1 valinia , a. , & marshall , f. e. , 1998 , apj , 505 , 134 vollmer , b. , zylka , r. , & duschl , w. j. 2003 , a&a , in press ( astro - ph/0306200 ) wijnands , r. 2001 , in proceedings of the 113th meeting of the astronomical society of the pacific , `` the high energy universe at sharp focus : chandra science '' ( astro - ph/0107600 ) yamauchi , s. , & koyama , k. 1995 , pasj , 47 , 439 yusef - zadeh , f. , & morris , m. 1987 , apj , 320 , 545 yusef - zadeh , f. , law , c. , wardle , m. , wang , q. d. , fruscione , a. , lang , c. c. , & cotera , a. 2002 , apj , 570 , 665 zylka , r. , mezger , p. g. , & wink , j. e. 1990 , a&a , 234 , 133 ccc 0242 & 1999 - 09 - 21 & 41 + 1561a & 2000 - 10 - 26 & 36 + 1561b & 2001 - 07 - 14 & 14 + 2943 & 2002 - 05 - 22 & 35 + 2951 & 2002 - 02 - 19 & 12 + 2952 & 2002 - 03 - 23 & 12 + 2953 & 2002 - 04 - 19 & 12 + 2954 & 2002 - 05 - 07 & 12 + 3392 & 2002 - 05 - 25 & 167 + 3393 & 2002 - 05 - 28 & 158 + 3663 & 2002 - 05 - 24 & 38 + 3665 & 2002 - 06 - 03 & 90 + cccc si ( he@xmath12 ) & 1700 @xmath4 1900 & 1550 @xmath4 1640 & 2100 @xmath4 2150 + s ( he@xmath12 + ly@xmath12 ) & 2350 @xmath4 2700 & 2100 @xmath4 2150 & 2730 @xmath4 2810 + fe ( neutral ) & 6250 @xmath4 6500 & 5000 @xmath4 6100 & 7150 @xmath4 7300 + fe ( he@xmath12 ) & 6550 @xmath4 6800 & 5000 @xmath4 6100 & 7150 @xmath4 7300 + cccccccc 1 & 1.99@xmath57/ - & 32.9@xmath58 & 6.40@xmath59 & 1.19@xmath60 & [email protected] & 1.3 & 52.9/51 + 2 & 3.12@xmath61/ - & 36.8@xmath62 & 6.40@xmath59 & 1.03@xmath63 & [email protected] & 2.4 & 33.7/44 + 3 & 1.77@xmath64/0.31@xmath65 & 15.8@xmath66 & 6.39@xmath67 & 1.29@xmath60 & [email protected] & 1.5 & 134.7/121 + + | over the last three years , the galactic center region has been monitored with the _ chandra x - ray observatory_. besides the x - ray emission from the target object , sgr a * , diffuse x - ray emission was detected throughout most of the 17@xmath0 @xmath1 17@xmath0 field of view . with 11
_ chandra _ observations through 2002 june , the total effective exposure reaches @xmath2590 ks , providing significant photon statistics on much of the detailed structure within the faint , often filamentary , diffuse x - ray emission .
the _ true - color _ x - ray image and the _ equivalent width _ images for
the detected elemental species of the galactic center region demonstrate that the diffuse x - ray features have a broad range of spatio - spectral properties .
enhancements of the low - ionization - state fe line emission ( @xmath3 @xmath2 6.4 kev ) to the northeast of sgr a * can be interpreted as fluorescence within the dense interstellar medium resulting from irradiation by hard , external x - ray sources .
they may also be explained by emission induced by the bombardments by high energy particles on the ambient medium , such as may accompany unresolved supernova ejecta intruding into dense interstellar medium .
the detection of molecular cloud counterparts to the 6.4 kev fe line emitting features indicates that these fe line features are associated with dense galactic center clouds and/or active star - forming regions , which strongly supports the proposed origins of the x - ray reflection and/or supernova ejecta for the fe line emission .
we detect highly ionized s and si lines which are generally coincident with the low - ionization - state fe line emission and the dense molecular clouds in the northeast of sgr a*. these hot plasmas are then most likely produced by massive star - forming activities and/or supernova remnants in the galactic center . in contrast , we find that highly ionized he - like fe line emission ( @xmath3 @xmath2 6.7 kev ) is primarily distributed along the plane instead of being concentrated in the northeast of sgr a*. the implied high temperature and the relatively uniform , but , strong alignment along the plane are consistent with the magnetic confinement model , suggesting this hot gas component has been re - heated by the strong interstellar magnetic fields in the galactic center to produce the observed he - like fe line emission . |
dusty plasmas ( dp ) have recently attracted a great deal of attention due to a variety of new phenomena observed in them and the novel physical mechanisms involved in their description @xcite . in addition to known plasma electrostatic waves @xcite , e.g. ion - acoustic waves ( iaw ) , new oscillatory modes arise in dp @xcite , among which the _ dust - ion acoustic wave _ ( diaw ) , essentially corresponding to ion - acoustic ( ia ) oscillations in a strong presence of dust .
the linear properties of the iaws and diaws have been quite extensively studied and now appear well understood @xcite . as far as nonlinear effects are concerned ,
various studies have pointed out the possibility of the formation of diaw - related localized structures , due to a mutual compensation between nonlinearity and dispersion , including small - amplitude pulse solitons , shocks and vortices @xcite .
furthermore , the propagation of nonlinearly modulated dia wave packets was studied in ref .
@xcite , in addition to the formation of localized envelope soliton
modulated waves due to the modulational instability of the carrier waves .
a very interesting known approach is the lagrangean description of a nonlinear wave profile . in the context of electrostatic plasma waves ,
this formalism has been employed in studies of electron plasma waves @xcite and , more recently , ion - acoustic @xcite and dust - acoustic @xcite waves .
our aim here is to extend previous results by applying the lagrangean formalism to the description of nonlinear diaws propagating in dusty plasmas .
we shall consider the nonlinear propagation of dust - ion - acoustic waves in a collisionless plasma consisting of three distinct particle species ` @xmath0 ' : an inertial species of ions ( denoted by ` @xmath1 ' ; mass @xmath2 , charge @xmath3 ; @xmath4 denotes the absolute of the electron charge ) , surrounded by an environment of thermalized electrons ( mass @xmath5 , charge @xmath6 ) , and massive dust grains ( mass @xmath7 , charge @xmath8 , both assumed constant for simplicity ) ; @xmath9 denotes the charge state of dust grains ; we leave the choice of dust grain charge sign @xmath10 ( @xmath11 for negative / positive dust charge ) open in the algebra .
charge neutrality is assumed at equilibrium .
the hydrodynamic poisson system of equations which describe the evolution of the ion ` fluid ' in the plasma are considered .
the ion number density @xmath12 is governed by the continuity equation @xmath13 where the mean velocity @xmath14 obeys @xmath15 the electric field @xmath16 is related to the gradient of the wave potential @xmath17 , which is obtained from poisson s equation @xmath18 , viz .
@xmath19 alternatively , one may consider @xmath20 we assume a near - boltzmann distribution for the electrons , i.e. @xmath21 ( @xmath22 is the electron temperature and @xmath23 is boltzmann s constant ) .
the dust distribution is assumed stationary , i.e. @xmath24 . the overall quasi - neutrality condition at equilibrium then reads @xmath25 by choosing appropriate physical scales , eqs .
( 1)-(3 ) can be cast into a reduced ( dimensionless ) form .
let us define the ion - acoustic speed @xmath26 .
an appropriate choice for the space and timescales , @xmath27 and @xmath28 , are the effective debye length @xmath29 and the ion plasma period @xmath30 , respectively .
alternatively , one might leave the choice of @xmath27 ( and thus @xmath28 ) arbitrary following an idea suggested in refs .
@xcite ) which leads to the appearance of a dimensionless dispersion parameter @xmath31 in the formulae .
the specific choice of scale made above corresponds to @xmath32 ( implied everywhere in the following , unless otherwise stated ) ; however , we may keep the parameter @xmath33 to ` label ' the dispersion term in the forthcoming formulae . for one - dimensional wave propagation along the @xmath34 axis , eqs .
( [ densityequation ] ) - ( [ poisson ] ) can now be written as @xmath35 where all quantities are dimensionless : @xmath36 , @xmath37 and @xmath38 ; the scaling quantities are , respectively : the equilibrium ion density @xmath39 , the effective sound speed @xmath40 ( defined above ) and @xmath41 . the ( reduced ) electron and dust background density @xmath42 is defined as @xmath43 where we have defined the dp parameter @xmath44 , and made use of eq .
( [ neutrality ] ) .
note that both @xmath45 and @xmath42 reduce to unity at equilibrium .
we shall define , for later reference , the function @xmath46 given by eq .
( [ defhatn1 ] ) and its inverse function @xmath47 viz .
@xmath48 implies @xmath49 .
we note that the dependence on the charge sign @xmath50 is now incorporated in @xmath51 ; retain that @xmath52 ( @xmath53 ) corresponds to negative ( positive ) dust . remarkably , since the dust - free limit is recovered for @xmath54 , the results to be obtained in the following are also straightforward valid for ion - acoustic waves propagating in ( dust - free ) e - i plasma , upon setting @xmath54 in the formulae
. the well known diaw dispersion relation @xmath55 @xcite is obtained from eqs .
( [ densityequation ] ) to ( [ neutrality ] ) . on the other hand , the system ( [ reducedeqs ] )
yields the reduced relation @xmath56 , which of course immediately recovers the former dispersion relation upon restoring dimensions ( regardless , in fact , of one s choice of space scale @xmath27 ; cf .
definition of @xmath33 ) .
however , some extra qualitative information is admittedly hidden in the latter ( dimensionless ) relation .
should one consider a very long space scale @xmath57 ( i.e. @xmath58 ) , one readily obtains @xmath59 ( unveiling the role of @xmath33 as a characteristic dispersion control parameter ) .
finally , the opposite limit of short @xmath27 ( or infinite @xmath33 ) corresponds to ion plasma oscillations ( viz .
@xmath60 = constant ) .
let us introduce the lagrangean variables @xmath61 , which are related to the eulerian ones @xmath62 via @xmath63 see that they coincide at @xmath64 .
accordingly , the space and time gradients are transformed as @xmath65 where we have defined the quantity @xmath66 note that the convective derivative @xmath67 is now plainly identified to @xmath68 .
also notice that @xmath0 satisfies @xmath69 and @xmath70 as a matter of fact , the lagrangean transformation defined here reduces to a galilean transformation if one suppresses the evolution of @xmath71 , i.e. for @xmath72 ( or @xmath73 , hence @xmath74 ) . furthermore
, if one also suppresses the dependence in time @xmath75 , this transformation is reminiscent of the travelling wave ansatz @xmath76 , which is widely used in the sagdeev potential formalism @xcite .
the lagrangean variable transformation defined above leads to a new set of reduced equations @xmath77 \label{l3 } \\
\biggl ( \frac{\partial } { \partial \tau } - \alpha^{-1 } u \ , \frac{\partial } { \partial \xi } \biggr ) e(\xi , \tau ) & = & - 4 \pi z_i e n(\xi , \tau ) u(\xi , \tau ) \ , , \label{l4}\end{aligned}\ ] ] where we have temporarily restored dimensions for physical transparency ; recall that the ( dimensionless ) quantity @xmath78 , which is in fact a function of @xmath79 , is given by ( [ defhatn1 ] ) .
one immediately recognizes the role of the ( inverse of the ) function @xmath80 as a density time evolution operator ; cf .
( [ l1 ] ) @xcite .
poisson s equation is now obtained by eliminating @xmath79 from eqs .
( [ l2 ] , [ l3 ] ) @xmath81 note that a factor @xmath82 should appear in the left - hand side if one rescaled eq .
( [ l5 ] ) as described above ; cf .
the last of eqs .
( [ reducedeqs ] ) . this will be retained for later reference , with respect to the treatment suggested in ref .
@xcite ( see discussion below ) . in principle , our aim is to solve the system of eqs .
( [ l1 ] ) to ( [ l4 ] ) or , by eliminating @xmath79 , eqs .
( [ l1 ] ) , ( [ l2 ] ) and ( [ l5 ] ) for a given initial condition @xmath83 , and then make use of the definition ( [ lagrange - def ] ) in order to invert back to the eulerian arguments of the state moment variables ( i.e. density , velocity etc . ) .
however , this abstract scheme is definitely not a trivial task to accomplish .
multiplying eq . ( [ l3 ] ) by @xmath84 and then adding to eq .
( [ l4 ] ) , one obtains @xmath85 combining with eq .
( [ l2 ] ) , one obtains @xmath86 where @xmath87 is the ion plasma frequency ( defined above ) . despite its apparent simplicity , eq .
( [ nloscil ] ) is _ neither _ an ordinary differential equation ( ode ) since all variables depend on _ both _ time @xmath75 and space @xmath88 _ nor _ a closed evolution equation for the mean velocity @xmath84 : note that the ( normalized ) background particle density @xmath42 depends on the potential @xmath79 and on the plasma parameters ; see its definition ( [ defhatn1 ] ) .
the evolution of the potential @xmath89 , in turn , involves @xmath84 ( via the quantity @xmath80 ) and the ion density @xmath90 .
( [ nloscil ] ) suggests that the system performs nonlinear oscillations at a frequency @xmath91 .
near equilibrium , the quantity @xmath92 is approximately equal to unity and one plainly recovers a linear oscillation at the ion plasma frequency @xmath87 .
quite unfortunately this apparent simplicity , which might in principle enable one to solve for @xmath84 and then obtain @xmath93 in terms of @xmath62 and vice versa ( cf .
davidson s treatment for electron plasma oscillations in ref .
@xcite ; also compare to ref .
@xcite , setting @xmath94 therein ) , is absent in the general ( off - equilibrium ) case where the plasma oscillations described by eq .
( [ nloscil ] ) are intrinsically _ nonlinear_. since eq .
( [ nloscil ] ) is in general not a closed equation for @xmath71 , unless the background density @xmath42 is constant ( i.e. independent of @xmath79 , as in refs .
@xcite ) , one can _ neither _ apply standard methods involved in the description of nonlinear oscillators on eq .
( [ nloscil ] ) ( cf . ref .
@xcite ) , _ nor _ reduce the description to a study of eqs .
( [ nloscil ] , [ l6 ] ) ( cf . ref .
@xcite ) , but rather has to retain all ( or rather five ) of the evolution equations derived above , since five inter - dependent dynamical state variables ( i.e. @xmath45 , @xmath71 , @xmath95 , @xmath79 and @xmath0 ) are involved .
this procedure will be exposed in the following section .
let us consider weakly nonlinear oscillations performed by our system close to ( but not at ) equilibrium
. the basis of our study will be the reduced system of equations @xmath96 which follow from the lagrangean eqs .
( [ l1 ] ) to ( [ l5 ] ) by scaling over appropriate quantities , as described in [ reducedeuler ] @xcite .
this system describes the evolution of the state vector , say @xmath97 ( @xmath98 ) , in the lagrangean coordinates defined above .
we will consider small deviations from the equilibrium state @xmath99 , by taking @xmath100 , where @xmath101 ( @xmath102 ) is a smallness parameter .
accordingly , we shall taylor develop the quantity @xmath103 near @xmath104 , viz .
@xmath105 , in order to express @xmath42 as @xmath106 where the coefficients @xmath107 ( @xmath108 ) , which are determined from the definition ( [ defhatn1 ] ) of @xmath42 , contain all the essential dependence on the plasma parameters , e.g. @xmath109 ; making use of @xmath110 , one readily obtains @xmath111 remember that for @xmath54 ( i.e. for vanishing dust ) one recovers the expressions for iaws in e - i plasma . following the standard reductive perturbation technique @xcite , we shall consider the stretched ( slow ) lagrangean coordinates @xmath112 ( where @xmath113 will be determined later ) .
the perturbed state of ( the @xmath114th @xmath115 component of ) the state vector @xmath116 is assumed to depend on the fast scales via the carrier phase @xmath117 , while the slow scales enter the argument of the ( @xmath114th element s ) @xmath118th harmonic amplitude @xmath119 , viz .
@xmath120 ( where @xmath121 ensures reality ) . treating the derivative operators as @xmath122 and substituting into the system of evolution equations
, one obtains an infinite series in both ( perturbation order ) @xmath123 and ( phase harmonic ) @xmath124 .
the standard perturbation procedure now consists in solving in successive orders @xmath125 and substituting in subsequent orders .
the method involves a tedious calculation which is however straightforward ; the details of the method are presented e.g. in ref .
@xcite , so only the essential stepstones need to be provided here .
the equations obtained for @xmath126 determine the first harmonics of the perturbation @xmath127 where @xmath128 denotes the potential correction @xmath129 .
the cyclic frequency @xmath130 obeys the dispersion relation @xmath131 , which exactly recovers , once dimensions are restored , the standard iaw dispersion relation @xcite mentioned above .
proceeding in the same manner , we obtain the second order quantities , namely the amplitudes of the second harmonics @xmath132 and constant ( ` direct current ' ) terms @xmath133 , as well as a finite contribution @xmath134 to the first harmonics ; as expected from similar studies , these three ( sets of 5 , at each @xmath135 ) quantities are found to be proportional to @xmath136 , @xmath137 and @xmath138 respectively ; the lengthy expressions are omitted here for brevity .
the ( @xmath139 , @xmath140 ) equations provide the compatibility condition : @xmath141 ; @xmath142 is therefore the group velocity @xmath143 at which the wave envelope propagates .
it turns out that @xmath144 decreases with increasing wave number @xmath145 ; nevertheless , it always remains positive . in order @xmath146 , the equations for @xmath147 yield an explicit compatibility condition in the form of a nonlinear schrdinger
type equation ( nlse ) @xmath148 recall that @xmath149 denotes the amplitude of the first - order electric potential perturbation .
the ` slow ' variables @xmath150 were defined above . the _ dispersion coefficient _ @xmath151
is related to the curvature of the dispersion curve as @xmath152 .
one may easily check that @xmath151 is negative ( for all values of @xmath145 ) . the _ nonlinearity coefficient _
@xmath153 is due to carrier wave self - interaction .
it is given by the expression @xmath154 \ , .
\label{qcoeff}\ ] ] where the coefficients @xmath155 were defined above . for low wavenumber @xmath145
, @xmath153 goes to @xmath156 as @xmath157 according to the standard analysis @xcite , we can linearize around the plane wave solution of the nlse ( [ nlse ] ) @xmath158 ( @xmath159 : complex conjugate ) notice the amplitude dependence of the frequency shift @xmath160 and then assuming the perturbation @xmath161 to be of the form : @xmath162 . substituting into ( [ nlse ] ) , one thus readily obtains @xmath163 .
the wave will thus be _ stable _ ( @xmath164 ) if the product @xmath165 is negative .
however , for positive @xmath166 , instability sets in for wavenumbers below a critical value @xmath167 , i.e. for wavelengths above a threshold @xmath168 ; defining the instability growth rate @xmath169 , we see that it reaches its maximum value for @xmath170 , viz .
@xmath171 we see that the instability condition depends only on the sign of the product @xmath165 , which may be studied numerically , relying on the expressions derived above .
the nlse ( [ nlse ] ) is long known to possess distinct types of localized constant profile ( solitary wave ) solutions , depending on the sign of the product @xmath165 @xcite .
remember that this equation here describes the evolution of the wave s envelope , so these solutions represent slowly varying localized envelope structures , confining the ( fast ) carrier wave .
the analytic form of these excitation can be found in the literature ( see e.g. in @xcite for a brief review ) and need not be derived here in detail .
let us however briefly summarize those results . following ref .
@xcite , we may seek a solution of eq .
( [ nlse ] ) in the form @xmath172 , where @xmath173 , @xmath174 are real variables which are determined by substituting into the nlse and separating real and imaginary parts .
the different types of solution thus obtained are summarized in the following . for @xmath175
we find the _
( bright ) envelope soliton _ @xmath176 \ , , \label{bright}\ ] ] which represents a localized pulse travelling at the envelope speed @xmath177 and oscillating at a frequency @xmath178 ( at rest ) .
the pulse width @xmath27 depends on the maximum amplitude square @xmath179 as @xmath180 .
since the product @xmath165 is always positive for long wavelengths , as we saw above , this type of excitation will be rather privileged in dusty plasmas .
the bright - type envelope soliton is depicted in fig .
[ figure1]a , b. + for @xmath181 , we obtain the _ dark _ envelope soliton ( _ hole _ ) @xcite @xmath182^{1/2 } \pm \rho_1 \ , \tanh \biggl(\frac{z - u_e\ , t}{l ' } \biggr ) \ , , \nonumber \\ \theta & = & \frac{1}{2 p } \
, \biggl [ u_e z \ , - \biggl(\frac{1}{2 } u_e^2 - 2 p q \rho_1 \biggr ) \,t \biggr ] \ , , \label{darksoliton}\end{aligned}\ ] ] which represents a localized region of negative wave density ( shock ) travelling at a speed @xmath177 ; see fig .
[ figure1]c .
again , the pulse width depends on the maximum amplitude square @xmath183 via @xmath184 .
finally , still for @xmath181 , one also obtains the _ gray _ envelope solitary wave @xcite @xmath185^{1/2 } \
, , \label{greysoliton}\ ] ] which also represents a localized region of negative wave density . comparing to the dark soliton ( [ darksoliton ] ) , we note that the maximum amplitude @xmath186 is now finite ( non - zero ) everywhere ; see fig .
[ figure1]d .
the the pulse width of this gray - type excitation @xmath187 now also depends on an independent parameter @xmath188 which represents the modulation depth ( @xmath189 ) .
the lengthy expressions which determine the phase shift @xmath190 and the parameter @xmath188 , which are omitted here for brevity , can be found in refs .
@xcite . for @xmath191 ,
one recovers the _ dark _ soliton presented above .
an important qualitative result to be retained is that the envelope soliton width @xmath27 and maximum amplitude @xmath173 satisfy @xmath192 ( see above ) , and thus depend on ( the ratio of ) the coefficients @xmath151 and @xmath153 ; for instance , regions with higher values of @xmath151 ( or lower values of @xmath153 ) will support wider ( spatially more extended ) localized excitations , for a given value of the maximum amplitude .
contrary to the kdv soliton picture , the width of these excitations does not depend on their velocity .
it does , however , depend on the plasma parameters , e.g. here @xmath109 .
the localized envelope excitations presented above represent the slowly varying envelope which confines the ( fast ) carrier space and time oscillations , viz .
@xmath193 for the electric potential @xmath79 ( and analogous expressions for the density @xmath12 etc .
( [ 1st - order - corrections ] ) ) .
the qualitative characteristics ( width , amplitude ) of these excitations , may be investigated by a numerical study of the ratio @xmath194 : recall that its sign determines the type ( bright or dark ) of the excitation , while its ( absolute ) value determines its width for a given amplitude ( and vice versa ) . in fig .
[ figure2 ] we have depicted the behaviour of @xmath195 as a function of the wavenumber @xmath145 and the parameter @xmath109 : higher values of @xmath109 correspond to lower curves .
remember that , for any given wavenumber @xmath145 , the dust concentration ( expressed via the value of @xmath109 ) determines the soliton width @xmath27 ( for a given amplitude @xmath173 ; see discussion above ) since @xmath196 .
therefore , we see that the addition of _ negative _ dust generally ( @xmath197 ) results to higher values of @xmath195 ( i.e. wider or higher solitons ) , while _ positive _ dust ( @xmath53 ) has the opposite effect : it reduces the value of @xmath195 ( leading to narrower or shorter solitons ) . in a rather general manner , bright type solitons ( pulses )
seem to be rather privileged , since the ratio @xmath195 ( or the product @xmath165 ) of the coefficients @xmath151 and @xmath153 is positive in most of the @xmath198 plane of values .
one exception seems to be very the region of _ very _ low values of @xmath109 ( typically below @xmath199 ) , which develops a negative tail of @xmath195 for small @xmath145 ( @xmath200 ) : thus , a very high ( @xmath201 per cent ) electron depletion results in pulse destabilization in favour of dark - type excitations ( fig .
[ figure1]c , d ) . strictly speaking
, @xmath195 also becomes negative for very high wave number values ( @xmath202 ) ; nevertheless , we neglect for rigor this region from the analysis , in this ( long wavelength @xmath142 ) fluid picture ( for a weak dust presence , short @xmath142 diaws may be quite strongly damped ; however , this result may still be interesting for a strong presence of dust , when landau damping is not a significant issue @xcite ) .
by combining the lagrangean system of eqs .
( [ l1 ] ) to ( [ l5 ] ) , one obtains the ( reduced ) evolution equation @xmath203 \ , , \label{lfinal}\ ] ] where the function @xmath204 [ defined in eq .
( [ defg ] ) ] is evaluated at @xmath205 \ , .\ ] ] note that the ion density @xmath45 has been scaled by its equilibrium value @xmath39 , to be distinguished from the initial condition @xmath206 . despite its complex form ,
the nonlinear evolution equation ( [ lfinal ] ) can be solved exactly by considering different special cases , as regards the order of magnitude of the dispersion related parameter @xmath33 .
this treatment , based on ref .
@xcite , will only be briefly summarized here , for the sake of reference .
first , one may consider very short scale variations , i.e. @xmath207 ( or @xmath208 ) .
this amounts to neglecting collective effects , so oscillatory motion within a debye sphere is essentially decoupled from neighboring ones . by considering @xmath209 and @xmath104 ( i.e. @xmath210 )
, one may combine eqs .
( [ l4 ] ) and ( [ lfinal ] ) into @xmath211 which , imposing the initial condition @xmath212 , yields the solution @xmath213 note that if the system is initially at equilibrium , viz .
@xmath214 , then it remains so at all times @xmath215 .
now , one may go back to eq .
( [ l1 ] ) and solve for @xmath216 , which in turn immediately provides the mean fluid velocity @xmath71 @xmath217 via ( [ l0 ] ) , and then @xmath218 and @xmath89 . finally , the variable transformation ( [ lagrange - def ] ) may now be inverted , immediately providing the eulerian position @xmath34 in terms of @xmath88 and @xmath75
. we shall not go into further details regarding this procedure , which is essentially analogue ( yet not identical ) to davidson s treatment of electron plasma oscillations . quite interestingly , upon neglecting the dispersive effects , i.e. setting @xmath219 , eq .
( [ lfinal ] ) may be solved by separation of variables , and thus shown to possess a nonlinear special solution in the form of a product , say @xmath220 @xcite .
this calculation was put forward in ref .
@xcite ( where the study of iaw in a single electron temperature plasma was argued to rely on an equation quasi - identical to eq .
( [ lfinal ] ) ) .
however , the solution thus obtained relies on doubtful physical grounds , since the assumption @xmath221 , which amounts to remaining close to equilibrium
the last of eqs .
( [ reducedeqs ] ) , implies an infinite space scale @xmath27 ( recall the definition of @xmath33 ) , contrary to the very nature of the ( localized ) nonlinear excitation itself .
rather not surprisingly , this solution was shown in ref .
@xcite to decay fast in time , in both eulerian and lagrangean coordinates .
therefore , we shall not pursue this analysis any further .
we have studied the nonlinear propagation of dust ion acoustic waves propagating in a dusty plasma . by employing a lagrangean formalism
, we have investigated the modulational stability of the amplitude of the propagating dust ion acoustic oscillations and have shown that these electrostatic waves may become unstable , due to self interaction of the carrier wave .
this instability may either lead to wave collapse or to wave energy localization , in the form of propagating localized envelope structures .
we have provided an exact set of analytical expressions for these localized excitations .
this study complements similar investigations which relied on an eulerian formulation of the dusty plasma fluid model @xcite .
in fact , the lagrangean picture provides a strongly modified nonlinear stability profile for the wave amplitude , with respect to the previous ( eulerian ) description ; this was intuitively expected , since the passing to lagrangean variables involves an inherently nonlinear transformation , which inevitably modifies the nonlinear evolution profile of the system described .
however , the general qualitative result remains in tact : the dust ion acoustic - type electrostatic plasma waves may propagate in the form of localized envelope excitations , which are formed as a result of the mutual balance between dispersion and nonlinearity in the plasma fluid .
more sophisticated descriptions , incorporating e.g. thermal or collisional effects , may be elaborated in order to refine the parameter range of the problem , and may be reported later .
this work was funded by the _ sfb591 ( sonderforschungsbereich )
universelles verhalten gleichgewichtsferner plasmen : heizung , transport und strukturbildung _
german government programme .
support by the european commission ( brussels ) through the human potential research and training network via the project entitled : `` complex plasmas : the science of laboratory colloidal plasmas and mesospheric charged aerosols '' ( contract no .
hprn - ct-2000 - 00140 ) is also acknowledged . eqs .
( [ system - reduced ] ) are derived from eqs .
( [ l1 ] , [ l2 ] , [ l3 ] ) , @xmath223 and ( [ property1 ] ) , respectively .
we have avoided the appearance of @xmath224 cf .
( [ l4 ] , [ l5 ] ) for analytical convenience . | ion - acoustic modes propagating in unmagnetized dusty plasmas are studied by applying a generic collisionless fluid model .
an eulerian - to - lagrangean variable transformation leads to a new system of evolution equations , which may be combined into a single ( rather complex ) equation for the mean ion density .
the linear and weakly nonlinear oscillations regimes are studied and their modulational stability is investigated .
this study extends known previous results on the lagrangian description of electron- or ion - acoustic electrostatic plasma modes .
the relevance with the description of dust - ion acoustic waves propagating in a dusty plasma is also discussed . |
in this paper we consider a single electron living in the one - dimensional configuration space with a junction .
thus , we consider it the schrdinger particle .
we mathematically simplify our physical set - up in the following : we regard the one - dimensional configuration space with the junction as the space @xmath0 $ ] .
the segment @xmath1 $ ] is the junction then .
we have been interested in the boundary condition of the wave functions of the non - relativistic electron so that the electron s energy operator ( i.e. , the schrdinger operator ) becomes an observable ( i.e. , self - adjoint ) .
we have considered some self - adjoint extensions of the minimal schrdinger operator @xmath2 for this configuration space in @xcite from the point of the view of a quantum device .
we then showed that there are some cases where the wave functions on which a self - adjoint extension acts have their own phase factor at the boundary , @xmath3 , when the wave functions pass through the junction . on the other hand
, we showed that the wave functions do not have such a phase factor when they respectively stay in the left island @xmath4 and in the right island @xmath5 , namely , when there is no exchange between the wave functions living individual islands .
actually , all the self - adjoint extensions of our minimal schrdinger operator @xmath2 can be parameterized by @xmath6 following the von neumann s theory . here
, @xmath7 is the unitary group of the degree @xmath8 . in @xcite
the appearance of the phase factor was shown only in the cases where diagonal entries of the unitary matrix are all zero ( i.e. , @xmath9 ) . in this paper
, we complete the result in @xcite .
as asserted in theorem [ theo : main2 ] below , we will characterize the boundary conditions of all the self - adjoint extensions of the minimal schrdinger operator @xmath2 by just two types of boundary conditions proposed in @xcite .
more precisely , we will actually construct the boundary condition from every matrix of @xmath7 .
we will then give a tunnel - junction formula concerning the phase factor for the non - relativistic electron as the schrdinger particle .
we will propose a mathematical idea to make a qubit from a schrdinger particle through the tunnel - junction formula by controlling the phase factor , even though the spin of the schrdinger particle can not be used . in the case
where the electron spin should be considered , we studied similar problem for a single relativistic electron as the dirac particle @xcite .
first up , we prepare some mathematical tools and recall some results to state our main theorem , theorem [ theo : main2 ] . we respectively define function spaces @xmath10 and @xmath11 for our configuration space @xmath12 by @xmath13 and @xmath14 here @xmath15 is the set of all the square lebesgue - integrable functions on @xmath12 , and @xmath16 denotes the boundary @xmath17 of @xmath12 .
thus , @xmath18 satisfies the boundary condition : @xmath19 for @xmath20 and @xmath21 .
we denote by @xmath22 the domain of a linear operator @xmath23 throughout this paper .
now we introduce our starting object : ( minimal schrdinger operator ) : the @xmath24-dimensional _ schrdinger operator _
@xmath2 is defined by @xmath25 we call the operator @xmath2 the _ minimal schrdinger operator_. following ( * ? ? ?
* theorems 8.25(b ) and 8.22 ) respectively , we realize that the minimal schrdinger operator @xmath2 is closed symmetric , and its adjoint operator @xmath26 is given as @xmath27 then , the operation of @xmath26 is the same as that of @xmath2 though their domains are different from each other .
so , we call the adjoint operator @xmath26 the _ maximal schrdinger operator _ , and moreover , we employ the following naming : for every subspace @xmath28 with the condition , @xmath29 , we call the restriction , @xmath30 , of @xmath26 on @xmath28 the _ schrdinger operator_. we will investigate every boundary condition of the wave functions on which individual self - adjoint extension acts .
it was proved in @xcite that both deficiency indices are @xmath8 : @xmath31 , where the deficiency index @xmath32 is defined as the dimension of the individual deficiency subspace @xmath33 : @xmath34 .
thus , the minimal schrdinger operator has uncountably many self - adjoint extensions .
more precisely , following the general theory of differential equation and solving simple differential equations : @xmath35 , we can obtain orthonormal bases , @xmath36 , of the deficiency subspaces @xmath37 , respectively : @xmath38 and @xmath39 with the normalization factor @xmath40 so that @xmath41 and @xmath42 .
the uniqueness of the differential equations tells us that the individual bases are complete and @xmath31 .
we here note the following relations to use later : @xmath43 and @xmath44 following the von neumann s theory @xcite , all the self - adjoint extensions @xmath45 of the minimal schrdinger operator @xmath2 are given as a restriction of the maximal schrdinger operator @xmath26 on a proper subspace @xmath46 with @xmath47 : @xmath48 .
then , von neumann s theory @xcite provides the following proposition : [ prop : von - neumann ] there is a one - to - one correspondence between self - adjoint extensions @xmath45 of the minimal schrdinger operator @xmath2 and unitary operators @xmath49 so that the correspondence is given in the following : for every unitary operator @xmath50 , the corresponding self - adjoint extension @xmath51 is defined by @xmath52 and then its operation is @xmath53 conversely , for every self - adjoint extension @xmath45 of the minimal schrdinger operator @xmath2 , there is the corresponding unitary operator @xmath54 so that @xmath55 .
we here introduce some mathematical notation and terminology .
we denote by @xmath56 the set of all extended real numbers : @xmath57 .
since @xmath58 , the deficiency subspaces @xmath37 are @xmath8-dimensional hilbert spaces .
we here remember that the sets @xmath59 and @xmath60 are respectively the complete orthonormal systems of the deficiency subspace @xmath61 and @xmath62 .
we identify unitary operators @xmath63 from @xmath61 to @xmath62 with @xmath64 unitary matrices @xmath65 , making the correspondence by @xmath66 , @xmath67 , where @xmath68 and @xmath69 .
so , we often identify the unitary operator @xmath63 with the unitary matrix @xmath70 , and write @xmath71 .
we denote by @xmath72 the unitary group of degree @xmath73 throughout this paper .
the representation of our @xmath74 in this paper is then : @xmath75 we say that @xmath63 is _ diagonal _ if @xmath76 with @xmath77 .
otherwise , we say @xmath63 is _ non - diagonal_. before stating our main theorem , theorem [ theo : main2 ] , we recall the two types of self - adjoint extensions found in @xcite . we introduce a class of vectors @xmath78 ( * ? ? ?
* definition 1 ) : we say @xmath79 is in a class ( class @xmath80 ) if and only if @xmath81 and @xmath82 for every vector @xmath80 in the class ( class @xmath80 ) , we give a matrix @xmath83 by @xmath84 , and call it the _ boundary matrix_. we recall that there are at least two types of boundary conditions for self - adjoint extensions of the the minimal schrdinger operator @xmath2 as shown in @xcite
. we will give a brief outline of its proof in
[ subsec : theo - main1 ] .
[ theo : main1 ] ( ( * ? ? ?
* theorem 1(ii)&theorem 2(i ) ) ) 1 . for every @xmath85 , set a subspace @xmath86 as @xmath87 where @xmath88 then , the restriction , @xmath89 , of the adjoint operator @xmath26 on @xmath86 is a self - adjoint extension of the minimal schrdinger operator @xmath2 .
2 . for every vector @xmath80 in the class ( class @xmath80 ) , define a subspace @xmath90 by @xmath91 where @xmath92 then , the restriction , @xmath93 , of the adjoint operator @xmath26 on @xmath90 is a self - adjoint extension of the minimal schrdinger operator @xmath2 .
we here recall the following lemma : ( ( * ? ? ?
* lemma 2 ) ) [ lem:1 ] if @xmath94 in in the class ( class @xmath80 ) , then @xmath95 for each @xmath96 .
we only have to show the case where @xmath97 . multiplying both sides of ( [ eq : lem-3 - 1 ] ) by @xmath98 ,
we have @xmath99 multiply both sides of this equation by @xmath100 . then , ( [ eq : lem-3 - 2 ] ) tells us that @xmath101 .
combining this fact with ( [ eq : lem-3 - 1 ] ) , we have @xmath102 . multiplying both sides of ( [ eq : lem-3 - 1 ] ) by @xmath100 leads to @xmath103 .
here we used @xmath104 .
multiplying both sides of the above representation of @xmath100 by @xmath105 brings us to @xmath106 .
here we used conditions in ( [ eq : lem-3 - 2 ] ) , especially , @xmath107 .
we reach @xmath108 by multiplying both sides of ( [ eq : lem-3 - 3 ] ) by @xmath109 and using the conditions in ( [ eq : lem-3 - 2 ] ) .
here we used @xmath110 .
we have proved our statement for @xmath111 , @xmath112 , @xmath113 , and @xmath114 .
it follows from these facts and the conditions in ( [ eq : lem-3 - 2 ] ) that @xmath115 for all the combinations of @xmath116 .
the following proposition shows how a phase factor appears in the boundary matrices @xmath83 with @xmath79 in the class ( class @xmath80 ) : [ prop : phase ] let @xmath83 be an arbitrary boundary matrix with the vector @xmath117 in the class ( class @xmath80 ) .
then , one of @xmath118 and @xmath100 is non - zero at least .
so , set @xmath119 , and @xmath120 as follows : let @xmath121 be @xmath118 if @xmath122 , and @xmath100 if @xmath123 .
define @xmath124 then , @xmath83 has the following representation : @xmath125 set @xmath126 as @xmath127 . since the vector @xmath80 is in the class ( class @xmath80 ) , lemma [ lem:1 ] says that @xmath115 for each @xmath116 .
moreover , ( [ eq : lem-3 - 1 ] ) says that @xmath122 or @xmath128 .
we can rewrite @xmath83 as : @xmath129 and @xmath130 we note @xmath131 by ( [ eq : lem-3 - 1 ] ) .
therefore , we can complete our proof . our main theorem in this paper
is concerned with the classification of all the boundary conditions . to state the classification , we prepare another lemma .
we denote by @xmath132 the hamilton quaternion field with determinant one , i.e. , @xmath133 the following lemma says that @xmath134 can be decomposed into the product of an element of @xmath135 and an element of @xmath132 .
although this lemma was already proved in ( * ? ? ?
* proposition 4.3 ) , we here give a simpler proof than that of ( * ? ? ?
* proposition 4.3 ) : [ lem : u(2 ) ] @xmath136 . since it is clear that @xmath137 , we show @xmath138 .
let @xmath139 be an arbitrary matrix in @xmath7 .
the equation @xmath140 implies @xmath141 similarly , the equation @xmath142 implies @xmath143 .
comparing this with ( [ eq : u(2)-3 ] ) , we have @xmath144 in the same way , comparing the equation @xmath145 derived from @xmath142 with ( [ eq : u(2)-3 ] ) leads to @xmath146 here we introduce argument @xmath126 of @xmath147 , i.e. , @xmath148 , and then @xmath149 , @xmath150 . by using @xmath142 , ( [ eq : u(2)-4 ] ) , and ( [ eq : u(2)-5 ] ) , we can reach @xmath151 in the case where @xmath152 , we set @xmath153 , and @xmath154 by @xmath155 , @xmath156 , and @xmath157 .
then , we have @xmath158 and @xmath159 . by ( [ eq : u(2)-4 ] ) we can compute @xmath160 as : @xmath161 which implies @xmath162 .
meanwhile , it follows from ( [ eq : u(2)-6 ] ) that @xmath163 using this , we compute @xmath164 as @xmath165 .
hence it follows that @xmath166 . in the case
@xmath167 , we define @xmath153 , and @xmath154 by @xmath168 , @xmath169 , and @xmath170 , and then , we have @xmath158 , @xmath171 .
since @xmath172 by ( [ eq : u(2)-4 ] ) , we have @xmath173 . finally , ( [ eq : u(2)-5 ] ) brings us to the computation , @xmath174 .
consequently , we obtain @xmath166 .
proposition [ prop : von - neumann ] based on von neumann s theory says that all the self - adjoint extensions of the minimal schrdinger operator @xmath2 are parameterized by @xmath134 .
our assertion is that there are only two types of boundary conditions for all the self - adjoint extensions of @xmath2 .
they are represented by ( [ schrodinger:*rho ] ) and ( [ schrodinger:*alpha ] ) .
therefore , the only thing we have to do for the classification of the boundary conditions is actually to construct the boundary condition ( [ schrodinger:*rho ] ) or ( [ schrodinger:*alpha ] ) from every @xmath134 .
in addition , proposition [ prop : von - neumann ] says that a diagonal @xmath134 leaves the set of the left - island functions ( resp .
the right - island functions ) invariant , on the other hand , a non - diagonal @xmath134 exchanges the left - island functions and the right - island functions . theorem [ theo : main2 ] shows that this situation is reflected in the boundary conditions .
[ theo : main2 ] for every self - adjoint extension @xmath51 , @xmath134 , of the minimal schrdinger operator @xmath2 , every boundary condition of the wave functions in @xmath175 is constructed as in ( a ) or ( b ) . 1 .
let @xmath176 be diagonal .
then , @xmath63 is represented as latexmath:[\[u= \begin{pmatrix } \gamma_{l } & 0 \\ 0 & \gamma_{r } \end{pmatrix } \,\,\ , \text{with}\,\,\ , the self - adjoint extensions @xmath51 parameterized by diagonal matrices @xmath134 and the self - adjoint extensions @xmath178 parameterized by vectors @xmath179 such that @xmath180 is given in the following : 1 . for every @xmath181 ,
set @xmath182 as @xmath183 .
then , the component @xmath184 is given by @xmath185 conversely , for every @xmath184 , the argument @xmath182 is determined by @xmath186 2 . the components @xmath187 and @xmath188 correspond to each other .
3 . for every @xmath189 ,
set @xmath190 as @xmath191 .
then , the component @xmath192 is given by @xmath193 conversely , for every @xmath192 , the argument @xmath190 is determined by @xmath194 4 . the components @xmath195 and @xmath196 correspond to each other .
2 . let @xmath176 be non - diagonal .
then , @xmath63 is represented as @xmath197 the one - to - one correspondence between the self - adjoint extensions @xmath51 parameterized by non - diagonal matrices @xmath134 and the self - adjoint extensions @xmath198 parameterized by vectors @xmath79 in the class ( class @xmath80 ) such that @xmath199 is given in the following : for every triple of components @xmath200 , @xmath201 , and @xmath154 , the vector @xmath79 in the class ( class @xmath80 ) is given by @xmath202 conversely , for every @xmath203 satisfying ( class @xmath80 ) , the triple of the components @xmath200 , @xmath201 , and @xmath154 is determined by @xmath204 theorems [ theo : main1 ] and [ theo : main2 ] show us the following correspondence : every diagonal @xmath134 corresponds to the boundary condition ( [ schrodinger:*rho ] ) , and every non - diagonal @xmath134 to the boundary condition ( [ schrodinger:*alpha ] ) .
in addition , theorem [ theo : main2 ] says that there is no boundary condition which makes a self - adjoint extension but the conditions ( [ schrodinger:*rho ] ) and ( [ schrodinger:*alpha ] ) . by proposition [ prop : von - neumann ] ,
the wave functions belonging to the domain of every self - adjoint extension @xmath51 are represented as in ( [ eq : von - neumann - representation ] ) . since @xmath205 for @xmath206 , the unitary operator
@xmath134 gives us the information about how the electron reflects at the boundary and how it passes through the junction . according to ( [ eq : tr - prob ] ) , the unitary operator u maps the eigenfunction @xmath207 living in the left island ( resp .
@xmath208 living in the right island ) to the eigenfunction @xmath209 ( resp .
@xmath210 ) staying in the same island with the probability @xmath211 ( resp .
@xmath212 ) and the eigenfunction @xmath210 ( resp .
@xmath209 ) coming from the opposite island with the probability @xmath213 ( resp .
@xmath214 ) .
theorems [ theo : main1 ] and [ theo : main2 ] show how the information from @xmath134 reflects in the boundary conditions .
the boundary condition ( [ schrodinger:*rho ] ) shows the solitariness : @xmath215 and @xmath216 both of the boundary conditions in the left island and the right one are independent of each other , which makes no interchange between the information of the individual wave functions living in the left island and right one .
in addition , no extra phase factor @xmath217 appears with the form @xmath218 in this boundary condition then . on the other hand ,
the boundary condition ( [ schrodinger:*alpha ] ) shows how the individual wave functions living in each island make interchange between each other at the boundary .
proposition [ prop : phase ] shows how a phase factor appears in the boundary condition : @xmath219 for some @xmath220 , @xmath221 , with @xmath222 .
( [ eq : tjf ] ) with ( [ eq : tjf - b ] ) and ( [ eq : tjf - u ] ) is our _ tunnel - junction formula _ for the schrdinger particle . using this tunnel - junction formula
, we try to consider a possibility of a quantum device .
that is , let us now consider a unit of a quantum device , consisting of a junction and two quantum wires as in fig .
[ fig : unit ] . in this unit ,
the junction is for controlling the phase factor , and the wires for transporting the schrdinger particle . in our argument
the two wires were respectively represented by the left island @xmath223 and the right island @xmath224 , and the junction by the segment @xmath225 $ ] .
we regarded the junction as a black box to give mathematical , physical arbitrariness to the junction .
as seen above , the self - adjointness of the hamiltonian of the schrdinger particle living in the two wires @xmath226 is mathematically determined by a boundary condition of the wave functions on which every self - adjoint extension acts . in real physics , actually , the boundary condition is uniquely determined by the quality and the shape of the boundary of a material of the wires .
thus , the wave functions have to satisfy the unit s own specific boundary condition to become the residents of the unit , otherwise the unit ejects them .
we then consider the combination of different two units as in fig.[fig : units ] .
we set unit@xmath227 with the boundary condition bc@xmath227 and unit@xmath24 with the boundary condition bc@xmath24 .
we connect the two junctions with each other by a phase - controller . with the boundary condition
bc@xmath227 , and the lower is unit@xmath24 with the boundary condition bc@xmath24 . the ellipsoid body is a phase - controller.,width=245 ] we employ @xmath228 for the phase factor in bc@xmath227 and @xmath229 ( @xmath230 ) for the phase factor in bc@xmath24 , respectively .
if we succeed in the phase - shift gate operation from @xmath228 to @xmath229 at the phase - controller , we can make a phase - based switching device for electron as the schrdinger particle .
this suggests that we could use unit@xmath227 and unit@xmath24 as a qubit in the same way as the flying qubit @xcite which is realized by the presence of an electron in either channel of the wires of an aharonov - bohm ring .
thus , our mathematical idea of the phase - based switching device may propose a possibility of the control over the transport of schrdinger particle so that the transport through either channel of unit@xmath227 and unit@xmath24 makes another flying qubit by using phase factor .
we here introduce the brief outline of the proof of theorem [ theo : main1 ] according to theorem 1(ii ) and theorem 2(i ) .
we will use the following lemma : [ lem:2 ] 1 .
let @xmath231 be arbitrary complex numbers
. 1 . for every @xmath232 with @xmath233 ,
there is a wave function @xmath234 such that @xmath235 .
2 . for every @xmath232 with @xmath236 and @xmath237 ,
there is a wave function @xmath234 such that @xmath238 and @xmath239 .
3 . for every @xmath232 with @xmath240 and @xmath241 ,
there is a wave function @xmath234 such that @xmath242 and @xmath243 .
4 . for every @xmath232 with @xmath244 ,
there is a wave function @xmath234 such that @xmath245 .
2 . for arbitrary complex numbers
@xmath246 , there is a wave function @xmath247 such that @xmath248 and @xmath249 .
we denote by @xmath250 and @xmath251 the characteristic functions on @xmath252\right.$ ] and @xmath253 , respectively .
\(i ) let @xmath254 be functions in @xmath255 so that @xmath256 , @xmath257 , and @xmath258 , @xmath259 .
we define the function @xmath260 by @xmath261 for part ( i-1 ) , @xmath262 for part ( i-2 ) , @xmath263 for part ( i-3 ) , @xmath264 for part ( i-4 ) .
then , we obtain our desired function @xmath265 .
\(ii ) let @xmath254 be functions in @xmath255 so that @xmath256 , @xmath266 , and @xmath258 , @xmath259 .
we define the function @xmath260 by @xmath267 where @xmath268 and @xmath269 .
then , the function @xmath270 satisfies our desired condition .
it follows from integration by parts that @xmath271 for every @xmath272 . since @xmath273 for @xmath274 , it is easy to check @xmath275 using ( [ eq : int - by - parts ] ) . on the other hand , @xmath276 , @xmath274 ,
is proved as follows : by the general definition of adjoint operator , we have @xmath277 for every @xmath278 and @xmath279 given in lemma [ lem:2 ] .
the arbitrariness of @xmath280 and @xmath281 , or @xmath282 and @xmath283 implies @xmath284 , that is , @xmath285 .
this is our path to prove the self - adjointness of @xmath286 , i.e. , @xmath287 , employed in @xcite .
for instance , we here show the proof of theorem [ theo : main1 ] ( b ) only . since @xmath288 , we have @xmath289 for every @xmath290 by using ( [ eq : int - by - parts ] ) , ( [ schrodinger:*alpha ] ) and conditions of ( class @xmath80 ) . hence it follows from this that @xmath198 is symmetric , i , e , @xmath291 . conversely , using the fact that @xmath292 along with the help of the general definition of adjoint operator and ( [ eq : int - by - parts ] ) , for every @xmath293 and @xmath247 given in lemma [ lem:2](ii ) we have @xmath294 so , the arbitrariness of @xmath282 and @xmath283 leads to @xmath295 we here note that the conditions of ( class @xmath80 ) leads to @xmath296 thus , these two equations imply that @xmath297 .
that is , @xmath298 .
therefore , we have proved that @xmath299 .
part ( a ) follows from ( * ? ? ?
* theorem 1 ) .
so , we prove part ( b ) only . in this proof , we set @xmath300 for simplicity .
let us give an arbitrary non - diagonal @xmath134 .
we know that @xmath134 has the representation in lemma [ lem : u(2 ) ] . by ( [ eq : von - neumann - representation ] ) we can write @xmath301 as @xmath302 where @xmath303 , and @xmath304 and @xmath305 run over @xmath306 arbitrarily . using this representation , ( [ eq : rel1 ] ) and
( [ eq : rel2 ] ) we can compute @xmath307 , @xmath308 , @xmath309 , and @xmath310 as : @xmath311 ( [ eq : representation - l - r ] ) says that @xmath312 where @xmath313 and @xmath314 since @xmath315 , we know @xmath316 exists . thus , our desired representation ( [ eq : tjf - b ] ) of @xmath317 , and @xmath109 by @xmath153 , and @xmath154 follows from the immediate computation of @xmath318 .
thus , every @xmath301 satisfies the boundary condition ( bc @xmath80 ) .
what we have to show next is that the vector @xmath79 given by ( [ eq : tjf - b ] ) is in the class ( class @xmath80 ) .
it is obvious that our @xmath118 , @xmath100 , @xmath319 , and @xmath109 satisfy ( [ eq : lem-3 - 2 ] ) .
we can compute @xmath320 as follows : @xmath321 \\ = &
\frac{1}{2|\gamma_{2}| } \bigl [ \left ( \eta\gamma_{1}+\eta^{*}\gamma_{1}^ { * } + \eta e^{-i\sqrt{2}\,\lambda}\gamma_{3}+\eta^{*}e^{i\sqrt{2}\,\lambda}\gamma_{3}^ { * } \right ) \\ & \qquad\qquad\times \left ( \eta^{*}\gamma_{1}+\eta\gamma_{1}^ { * } + \eta e^{-i\sqrt{2}\,\lambda}\gamma_{3 } + \eta^{*}e^{i\sqrt{2}\,\lambda}\gamma_{3}^ { * } \right ) \\ & \qquad -\left ( \gamma_{1}+\gamma_{1}^ { * } + e^{-i\sqrt{2}\,\lambda}\gamma_{3 } + e^{i\sqrt{2}\,\lambda}\gamma_{3}^ { * } \right ) \\ & \qquad\qquad\times \left ( \gamma_{1}+\gamma_{1}^ { * } + \eta^{2}e^{-i\sqrt{2}\,\lambda}\gamma_{3 } + \eta^{*2}e^{i\sqrt{2}\,\lambda}\gamma_{3}^ { * } \right ) \bigr ] \\ = & \frac{|\gamma_{3}|^{2}-|\gamma_{1}|^{2}}{|\gamma_{2}|^{2}}=1 . \end{aligned}\ ] ] thus , @xmath118 , @xmath100 , @xmath319 , and @xmath109 given by ( [ eq : tjf - b ] ) satisfy ( [ eq : lem-3 - 1 ] ) , and the vector @xmath322 is in the class ( class @xmath80 )
. therefore , we have constructed the boundary matrix @xmath83 with the vector @xmath79 in the class ( class @xmath80 ) from every non - diagonal @xmath134 .
conversely , let @xmath317 , and @xmath109 be arbitrary complex numbers in the class ( class @xmath80 ) .
it immediately follows from the definition of @xmath200 , @xmath201 , and @xmath323 in ( [ eq : tjf - u ] ) that @xmath324 . using this equation together with the conditions of ( class @xmath80 ) and the representation given in proposition [ prop : phase ] ,
we have @xmath325 here we note @xmath326 by its definition .
thus , ( [ eq : tjf - u ] ) gives us the unitary operator @xmath63 with the representation : @xmath327 we show from now on that the above @xmath63 satisfies @xmath199 . for arbitrarily given @xmath134 , insert @xmath307 , @xmath308 , @xmath309 , and @xmath310 with the representation ( [ eq : representation - l - r ] ) into the boundary conditions , @xmath328 then , since the coefficients @xmath304 and @xmath305 in @xmath175 are arbitrary and @xmath329 , we can show that the condition @xmath199 is equivalent to the system of the following system of equations : @xmath330 thus , we now show that our @xmath63 given by ( [ eq : tjf - u ] ) satisfies the system of equations : using the representation in proposition [ prop : phase ] , it is easy to check that our @xmath200 , @xmath201 , and @xmath154 given by ( [ eq : tjf - u ] ) satisfy ( [ eq : inverse-1 ] ) and ( [ eq : inverse-4 ] ) in the following .
@xmath331 we recall @xmath222 by proposition [ prop : phase ] .
this equation leads to @xmath332 we can show that our @xmath200 , @xmath201 , and @xmath154 satisfy ( [ eq : inverse-2 ] ) as @xmath333 with the help of ( [ eq : inverse-5 ] ) , and they satisfy ( [ eq : inverse-2 ] ) as @xmath334 with the help of ( [ eq : inverse-6 ] ) .
therefore , the unitary operator @xmath63 made from our @xmath200 , @xmath201 , and @xmath154 satisfies the equation @xmath199 , and we can complete the proof of our theorem .
we have completely characterized the boundary conditions for all the self - adjoint extensions of the minimal schrdinger operator .
we then found a tunnel - junction formula concerning the phase factor . in this formula
we can find the factor @xmath335 which depends on the length of the junction .
we have not yet clarified the physical reason why this factor appears . compared with the results in @xcite and this paper ,
we realize that such a factor concerning the parameter @xmath336 appears for the schrdinger particle , but it does not for the dirac particle @xcite .
that is , the schrdinger particle seems to feel the distance @xmath337 , though the dirac particle does not .
we conjecture that this situation physically depends on the speed of the particle .
one of the authors ( m.h . ) acknowledges the financial support from jsps , grant - in - aid for scientific research ( c ) 23540204 .
he also expresses special thanks to kae nemoto and yutaka shikano for the useful discussions with them . | we handle all the self - adjoint extensions of the minimal schrdinger operator for the non - relativistic electron living in the one - dimensional configuration space with a junction .
we are interested in every boundary condition corresponding to the individual self - adjoint extension .
thus , we clarify all the types of those boundary conditions of the wave functions of the non - relativistic electron .
we find a tunnel - junction formula for the non - relativistic electron passing through the junction .
using this tunnel - junction formula , we propose a mathematical possibility of a tunnel - junction device for qubit .
schrdinger operator , self - adjoint extension , tunnel - junction device , qubit 47b25 , 81p45 , 81p68 , 81q10 |
in the recent years the concept of lower ricci - curvature bounds has been generalized from the riemannian setting to the setting of geodesic metric measure spaces , most prominently in the works of sturm @xcite and lott and villani @xcite .
different authors have studied slightly different definitions of curvature bounds . what is common to all these definitions is that they strive to fulfill the same set of criteria : to naturally extend ricci - curvature bounds of riemannian manifolds , to be stable under the measured gromov - hausdorff convergence , and to provide enough structure for meaningful analysis on the metric space .
the first two criteria are well met @xcite , whereas the third one has so far been met only partially . from a classical result by buser @xcite we know that a riemannian manifold with nonnegative ricci - curvature supports a poincar inequality . moreover , in the case of measured gromov - hausdorff limits of riemannian manifolds with ricci - curvature bounded below we know that a local poincar inequality always holds @xcite . in order to prove a local poincar inequality for a larger class of metric spaces with ricci - curvature lower - bounds the stability under measured gromov - hausdorff convergence had to be sacrificed in @xcite by assuming the spaces to be nonbranching .
the purpose of this paper is to complete this part of the theory for some versions of curvature - dimension conditions by proving local poincar inequalities without the unnatural extra assumption of nonbranching .
the local poincar inequality together with a doubling measure constitute the very foundation of analysis on metric spaces which was pioneered by heinonen and koskela @xcite and cheeger @xcite . for an introduction to analysis on metric spaces we refer to the book by heinonen @xcite .
poincar inequalities have been proven in many classes of metric spaces , for example in locally linearly contractible ahlfors - regular metric spaces @xcite .
a metric measure space @xmath3 admits a weak local @xmath4-poincar inequality with @xmath5 if there exist constants @xmath6 and @xmath7 such that for any continuous function @xmath8 defined on @xmath9 , any point @xmath10 and radius @xmath11 such that @xmath12 and any upper gradient @xmath13 of @xmath8 we have @xmath14 where the barred integral denotes the average integral and @xmath15 denotes the average of @xmath8 in the ball @xmath16 .
recall that , as introduced in @xcite , a borel function @xmath17 $ ] is an upper gradient of @xmath8 if for any constant speed curve @xmath18 \to x$ ] with length @xmath19 we have @xmath20 the most studied poincar inequalities are @xmath21-poincar inequalities . from hlder s
inequality it follows immediately that a weak local @xmath21-poincar inequality implies a weak local @xmath22-poincar inequality for every @xmath23 . in this paper
we consider weak local @xmath24-poincar inequalities which we simply call weak local poincar inequalities .
the word _ weak _ in the weak local poincar inequality refers to the fact that we allow the ball on the right - hand side of to be larger than the one on the left .
if the balls on both sides of the inequality can be taken to be the same , meaning that we can take @xmath25 , the inequality is called a strong local poincar inequality . in a doubling geodesic metric space
the weak local poincar inequality implies the strong one , with possibly a different constant @xmath26 , see @xcite and also @xcite .
we say that a measure @xmath27 is doubling ( with a constant @xmath28 ) if for all @xmath10 and @xmath29 we have @xmath30 lott and villani gave in @xcite a definition for nonnegative @xmath31-ricci curvature with @xmath32 and a definition for @xmath33-ricci curvature being bounded below by @xmath34 . in their definition they required weak displacement convexity for a collection @xmath35 of suitable convex functionals .
( for the precise definitions , see section [ sec : definitions ] . )
more specifically they required that between any two probability measures that have bounded wasserstein distance between them there is at least one geodesic in the wasserstein space of probability measures along which all the functionals in @xmath35 satisfy a convexity inequality .
at the same time when lott and villani defined their curvature - dimension bounds another definition was given by sturm @xcite .
he defined spaces where @xmath31-ricci curvature is bounded from below by a constant @xmath34 . in sturm s definition
the displacement convexity is also required along only one geodesic but for only one critical functional .
later lott and villani combined the two different definitions by considering what we call here @xmath0 spaces with @xmath34 referring to the lower bound on curvature and @xmath36 $ ] to the upper bound on dimension . in this definition weak displacement convexity
is again required for the whole collection @xmath35 .
the assumptions in the definition by sturm are a priori weaker . however , in nonbranching metric spaces the two notions agree , see for example @xcite .
it should be emphasized that our point is to avoid making the nonbranching assumption .
therefore the results we get for @xmath0 spaces do not necessarily hold under the definitions of sturm .
let us present the poincar inequalities we are able to obtain .
the first result is for the large class of @xmath37 spaces .
[ thm : cd1 ] suppose that @xmath3 is a @xmath37 space with @xmath38 .
then the weak local poincar inequality @xmath39 holds for any continuous function @xmath8 defined on @xmath9 , any upper gradient @xmath13 of @xmath8 and for each point @xmath10 and radius @xmath11 .
we do not have average integrals in theorem [ thm : cd1 ] like we had in .
if we were to write the average integrals here we would have to multiply the right - hand side of the inequality by a factor @xmath40 .
this factor could well be unbounded as @xmath41 in an infinite dimensional space .
although for @xmath37 spaces the assumption for @xmath27 to be doubling is not very natural we always have doubling for @xmath0 spaces when @xmath42 : by the bishop - gromov inequality any @xmath0 space @xmath3 is doubling with a constant @xmath43 , see for example @xcite .
therefore for @xmath0 spaces we can write our local poincar inequalities in a more standard form . [ thm : cd2 ] any @xmath0 space with @xmath38 supports the weak local poincar inequality @xmath44 in particular , any @xmath45 space supports the uniform weak local poincar inequality @xmath46 notice that the constant @xmath47 here is better than the constant @xmath48 obtained in @xcite for nonbranching @xmath45 spaces .
so far we have considered @xmath0 spaces where the weak displacement convexity is required for a class of functionals .
we are unable to prove the previous two theorems from the curvature - dimension conditions of sturm where the weak displacement convexity is required for only one functional .
the nonnegativity of @xmath31-ricci curvature can also be generalized to metric measure spaces using gradient flows .
this is done by requiring the existence of a flow satisfying the so called evolution variational inequality ( e.v.i . )
@xmath49 for all @xmath50 and for any probability measure @xmath51 which is absolutely continuous with respect to the measure @xmath27 . here
@xmath52 is the wasserstein distance between measures which will be defined in section [ sec : definitions ] .
for example in compact alexandrov spaces with curvature bounded below an e.v.i .
condition is satisfied @xcite .
the functional @xmath53 in the e.v.i . is the critical entropy functional which depends on the dimension bound @xmath31 . for finite @xmath31
it is the rnyi entropy functional @xmath53 which corresponds to the @xmath45 spaces . for @xmath54 the critical functional
is the shannon entropy functional @xmath55 corresponding to @xmath56 spaces .
the e.v.i .
condition is also stable under the measured gromov - hausdorff convergence ( see for instance @xcite for this type of statement ) .
it is perhaps easier to relate the e.v.i .
condition to the previous two curvature - dimension conditions after noticing that it implies the strong displacement convexity of the entropy functional corresponding to the e.v.i .
, see @xcite .
this means that the functional is not only displacement convex along one geodesic between any two measures , as was required in the definitions by sturm , and lott and villani , but it is in fact displacement convex along any geodesic between any two measures .
this strong displacement convexity is not by itself stable under the measured gromov - hausdorff convergence and therefore it is a poor notion of curvature - dimension bound . clearly
the curvature - dimension condition of sturm is also weaker than strong displacement convexity and thus weaker than the e.v.i .
what is not so clear is how the e.v.i .
condition relates to the @xmath45 spaces .
the following diagram gathers the implications which we are aware of between the local poincar inequality and the different curvature - dimension conditions that are mentioned in this paper .
notice that in this paper we consider the e.v.i .
condition and strong displacement convexity of @xmath53 only for the curvature lower - bound @xmath57 .
we are not aware of a generalization of the e.v.i .
condition for all @xmath34 and @xmath36 $ ] .
the curvature - dimension condition @xmath0 however works for all @xmath58 and @xmath31 . @xmath59{\txt{\textsc{sturm
} : \\weak displacement\\ convexity for $ { \mathscr{e}}_n$ \\ \emph{(stable condition ) } } } & * + [ f- , ] { \txt{\textsc{lott - villani } : \\weak displacement\\ convexity for all $ { \mathscr{f}}\in \mathcal{dc}_n$ \\
\emph{(stable condition ) } } } \ar @{= > } [ l ] \ar @{= > } [ d]\\ * + [ f-,]{\txt{e.v.i . for $ { \mathscr{e}}_n$\\ \emph{(stable condition ) } } } \ar @{= > } [ r ] & * + [ f-,]{\txt{strong displacement\\ convexity for $ { \mathscr{e}}_n$\\ \emph{(unstable condition ) } } } \ar @{= > } [ u ] \ar @{= > } [ r ] & * + [ f]{\txt{local poincar\'e\\ inequality } } } \ ] ] as indicated by the previous diagram we have a local poincar inequality ( analogous to theorem [ thm : cd1 ] ) for spaces where the entropy functionals are strongly displacement convex , and therefore also for the spaces satisfying the corresponding e.v.i . condition .
[ thm : evi1 ] suppose that @xmath55 ( or @xmath53 ) is strongly displacement convex in a geodesic metric measure space @xmath3 .
then the weak local poincar inequality @xmath60 is satisfied .
when we combine the doubling property of the measure @xmath27 in the @xmath42 case with theorem [ thm : evi1 ] we get again a more standard version of local poincar inequality .
[ cor : evi2 ] suppose that @xmath53 is strongly displacement convex in a geodesic metric measure space @xmath3 .
then the space @xmath9 admits the weak local poincar inequality @xmath46 to once more emphasize that not all of the desired connections between local poincar inequalities and curvature - dimension bounds are known , we state the missing ones corresponding to @xmath61 here as a question .
[ q:1 ] does the conclusion of theorem [ thm : evi1 ] or corollary [ cor : evi2 ] remain valid if we require only weak displacement convexity instead of strong displacement convexity ?
let us now slightly change the perspective to the question of the nonbranching assumption . in the earlier proofs of local poincar inequalities nonbranching
was assumed in order to guarantee almost uniqueness of geodesics .
it is natural to ask to what extent in general we have uniqueness of geodesics under the curvature - dimension conditions .
we prove in the last section of this paper the following theorem , theorem [ thm : uniqueness ] , which says that the strong displacement convexity of @xmath53 , and therefore also the corresponding e.v.i .
condition , implies that from every point in the underlying space there is a unique geodesic to @xmath27-almost every point in the space . in connection with the local poincar inequalities above
it means that we could also prove corollary [ cor : evi2 ] with a larger constant following more directly the proof of lott and villani @xcite .
[ thm : uniqueness ] suppose that @xmath53 is strongly displacement convex in a geodesic metric measure space @xmath3 .
then for every @xmath10 @xmath62 we do not know if the conclusion of theorem [ thm : uniqueness ] holds also for the shannon entropy functional @xmath55 .
the proof of theorem [ thm : uniqueness ] does not seem to work in this situation . notice that in order to have a reasonable formulation of this result for @xmath55 the conclusion has to be weakened so that it does not involve any singular parts of the measures .
the question could be stated for example in the following way .
suppose that @xmath55 is strongly displacement convex in a geodesic metric measure space @xmath3 .
do we then have for every @xmath63 for which we have @xmath64 , and for every @xmath65 uniqueness of geodesics in the sense that @xmath66 another way to formulate the uniqueness is to rely on the measure @xmath27 without using optimal plans .
suppose that @xmath55 is strongly displacement convex in a geodesic metric measure space @xmath3 .
do we then have @xmath67 in the next section the relevant preliminaries will be given .
section [ sec : poincare ] contains the proofs of all the local poincar inequalities presented here . in the final section , section [ sec : uniqueness ] ,
we prove theorem [ thm : uniqueness ] on the almost uniqueness of geodesics under the strong displacement convexity of @xmath53 .
many thanks are due to luigi ambrosio for his lectures on optimal mass transportation in the spring of 2011 where i learned this topic , for the many discussions i have had with him around the subject and , in particular , for pointing out to me the question of uniqueness of geodesics under the strong displacement convexity .
thanks are also due to the anonymous referee for suggesting many improvements to the paper .
for the most parts we will follow the notation used in @xcite . the metric measure spaces @xmath3 we consider are always geodesic and complete .
furthermore , the support of the measure @xmath27 can always be assumed to be the whole space @xmath9 .
we denote the support of a measure @xmath68 by @xmath69 .
let us denote by @xmath70 the borel probability measures on @xmath9 and by @xmath71 the probability measures in @xmath9 that are absolutely continuous with respect to the measure @xmath27 .
recall that any geodesic in a metric space @xmath72 can be reparametrized to be a continuous mapping
@xmath18 \to x$ ] with @xmath73 we denote the space of all the geodesics of the space @xmath9 with such parametrization by @xmath74 .
an important concept for this paper is ( non)branching in geodesic metric spaces . by branching of geodesics we mean that there are some distinct geodesics starting from the same point which follow the same path for some initial time interval and then become disjoint .
let us recall what we mean by measured gromov - hausdorff convergence even though we will not need it in the proofs .
a sequence of compact metric measure spaces @xmath75 is said to converge to a compact metric measure space @xmath3 in the measured gromov - hausdorff sense , if there exists a sequence of borel maps @xmath76 and a sequence of positive numbers @xmath77 so that 1 .
for every @xmath78 we have @xmath79 .
2 . for every @xmath10 and @xmath80
there exists @xmath81 such that @xmath82 .
we have @xmath83 as @xmath84 in the weak- * topology of @xmath70 .
the convexity of the functionals will be considered along the geodesics in the wasserstein space @xmath85 .
the distance between two probability measures @xmath86 in this space is given by @xmath87 where the infimum is taken over all @xmath88 with @xmath68 as its first marginal and @xmath51 as the second , i.e. @xmath89 and @xmath90 for all borel subsets @xmath91 of the space @xmath9 .
notice that in the case where the distance @xmath92 is not bounded the function @xmath52 is strictly speaking not a distance as the above infimum can also take an infinite value .
now , if it happens that there is a geodesic @xmath93 between two measures @xmath86 in the space @xmath85 it must be that this geodesic can be realized as a measure @xmath94 so that @xmath95 , where @xmath96 for any geodesic @xmath97 and @xmath98 $ ] and @xmath99 denotes the push - forward of the measure @xmath68 under @xmath100 , see for example ( * ? ? ?
* corollary 7.22 ) .
this realization is convenient for us when we want to translate information from the geodesics on @xmath70 to the geodesics on @xmath9 .
the space consisting of all measures @xmath101 for which the mapping @xmath102 is a geodesic in @xmath70 from @xmath103 to @xmath104 is denoted by @xmath105 .
we refer to @xcite for a detailed account on the theory of optimal transportation which includes all the basic information on the wasserstein space .
all the curvature - dimension conditions we consider in this paper are defined using integral functionals . for @xmath32
the functionals are build using functions in the displacement convexity class @xmath35 consisting of all the continuous convex functions @xmath106 for which we have @xmath107 and for which the function @xmath108is convex on the interval @xmath109 .
for @xmath54 we use the class @xmath110 of continuous convex functions @xmath106 for which we have @xmath107 and for which the function @xmath111is convex on the interval @xmath112 . notice that @xmath113 for all @xmath114 .
this means that the requirement of satisfying some condition for all @xmath115 becomes more and more demanding as we decrease @xmath31 .
most of the functionals @xmath116 which we consider are of the form @xmath117 where the measure @xmath68 is decomposed to the absolutely continuous part @xmath118 and the singular part @xmath119 with respect to @xmath27 .
the derivative at infinity is defined as @xmath120 which is guaranteed to exist for all @xmath121 because of the convexity assumption .
particularly relevant integral functionals for the curvature - dimension conditions are defined via using as @xmath122 the functions @xmath123 where @xmath32 .
notice that @xmath124 and @xmath125 and that for these functions we have @xmath126 meaning that the rnyi entropy functional @xmath53 built from the function @xmath127 does not see the singular part of the measure @xmath68 whereas the shannon entropy functional @xmath55 , corresponding respectively to @xmath128 , in the presence of any singular part has value @xmath33 .
the notion of displacement convexity was first used by mccann in @xcite .
the functional @xmath129 is called _ strongly displacement convex _ if for any @xmath93 we have @xmath130 for all @xmath131 $ ] .
the functional @xmath132 is called _ weakly displacement convex _ if for any two measures @xmath133 there exists @xmath93 so that @xmath134 , @xmath135 and @xmath136 satisfies the inequality . in general only the implication strong displacement convexity @xmath137 weak displacement convexity holds . in the particular case of riemannian manifolds
the converse is also true @xcite .
the converse is also true if the space @xmath9 is nonbranching , see ( * ? ? ?
* theorem 30.32 ) . as was mentioned in the introduction one good @xmath31-dimensional condition , with @xmath36 $ ] , for nonnegative curvature is the requirement of the existence of a flow satisfying the so called _ evolution variational inequality _
( e.v.i . )
@xmath49 for all @xmath50 and @xmath138 .
see @xcite for a comprehensive introduction to e.v.i . and
gradient flows .
the e.v.i .
condition is stable under measured gromov - hausdorff convergence .
this can be proven in a similar way as ( * ? ? ?
* theorem 7.12 ) .
the existence of a flow satisfying the e.v.i .
implies that the entropy @xmath53 is strongly displacement convex .
this was shown in a very general setting for functionals in metric spaces in @xcite . in order to define the @xmath0 spaces we need some more notation . given @xmath34 and @xmath139 $ ] , we define @xmath140,\\ 1 & \text{if } n < \infty \text { and } k = 0,\\ \left(\frac{\sinh(t\alpha)}{t\sinh \alpha}\right)^{n-1 } & \text{if } n <
\infty \text { and } k < 0 , \end{cases}\ ] ] where @xmath141 for @xmath142 we define @xmath143 we say that @xmath3 is a @xmath0 space , with the interpretation that it has @xmath31-ricci curvature bounded below by @xmath58 , if for any two measures @xmath144 with @xmath145 there exists @xmath146 so that along the wasserstein geodesic @xmath147 for every @xmath148 and for every @xmath98 $ ] we have @xmath149 where we have written @xmath150 and @xmath151 to their absolutely continuous and singular parts with respect to the measure @xmath27 , and where @xmath152 denotes the disintegrated measure of @xmath153 with respect to @xmath154 and @xmath155 with respect to @xmath156 .
we will not be precise about this disintegration as it will not be needed here .
this is so because in the proofs we will use only measures @xmath157 and for such measures we know by ( * ? ? ?
* lemma 29.6 ) that the inequality can be written as @xmath158 important thing to notice from the definition of @xmath0 spaces is that any @xmath0 space is also a @xmath159 space for every @xmath160 and a @xmath161 space for any @xmath162 .
before proving the poincar inequalities mentioned in the introduction let us point out the main differences between this proof and the proof in @xcite for the case of nonbranching spaces . we know that in a nonbranching @xmath0 metric space with @xmath32 there exists a unique geodesic from every point @xmath163 to @xmath27-almost every point @xmath164 , see (
* theorem 30.17 ) .
therefore for every point @xmath163 and any ball @xmath165 we have a unique geodesic @xmath166 between @xmath167 and @xmath168 .
then for example in the @xmath45 case we can use the displacement convexity of @xmath53 along this geodesic to obtain for each @xmath169 $ ] a bound on the density @xmath170 of @xmath171 with respect to @xmath27 of the form @xmath172at @xmath27-almost every point @xmath164 . combining such estimates for all points @xmath173 to obtain a so called dynamical democratic transference plan and using the arguments of lott and villani , most notably the symmetry of the estimates in time , we deduce a poincar inequality @xmath174for the case @xmath45 , with @xmath32 .
this approach has also been used by other authors , see @xcite for a similar proof of the local poincar inequality in nonbranching spaces using a related measure contraction property @xcite .
the key observation for proving better local poincar inequalities is that it is sufficient to split the ball @xmath165 into two equal sized parts @xmath175 and @xmath176 using the median of the function @xmath8 in the ball @xmath165 and then to find a geodesic @xmath166 between @xmath177 and @xmath178 along which we have a good estimate from above to the density @xmath170 . in the case
@xmath61 the bound we obtain is the best possible one : for all @xmath98 $ ] @xmath179 at @xmath27-almost every point @xmath164 .
we have two rather standard ways of obtaining this bound . in the strong displacement convexity case
we can prove it by considering restrictions of the optimal plan @xmath166 .
this argument has been used for example in ( * ? ? ?
* chapter 19 ) . for the @xmath0 spaces
the bound can be proven by taking a sequence of more and more convex functionals in @xmath180 , just like in the proof of ( * ? ? ?
* theorem 30.20 ) .
we will write the density bounds as their own lemmas and then prove theorem [ thm : cd1 ] from the corresponding density bound .
the rest of the weak local poincar inequalities follow analogously .
let us start the proofs with the density bounds in the @xmath0 spaces .
[ lma : cd ] suppose that @xmath3 is a @xmath0 space with @xmath38 and @xmath36 $ ] .
let @xmath63 be measures with densities bounded from above by a constant @xmath181 and so that @xmath64 .
suppose also that @xmath182 .
then there exists @xmath65 so that for every @xmath98 $ ] we have @xmath183 at @xmath27-almost every @xmath10 , where @xmath170 is the density of @xmath171 with respect to @xmath27 .
take the geodesic @xmath65 along which every @xmath115 satisfies .
take @xmath184 and define @xmath185 . because @xmath115 for every @xmath36 $ ] and the measures @xmath68 and @xmath51 have no singular part with respect to @xmath27 , we get by @xmath186 where @xmath153 and @xmath187 , d(x_0,x_1 ) \le d\}$ ] .
we then have @xmath188 so it remains to estimate @xmath189 from below for different @xmath31 and @xmath58 .
if @xmath57 or @xmath142 we have @xmath190 . if @xmath54 we have @xmath191 finally , if @xmath192 and @xmath193 we get @xmath194 next we deduce the sharp bound from the strong displacement convexity of @xmath55 .
[ lma : evi ] suppose that @xmath55 ( or @xmath53 ) is strongly displacement convex in a geodesic metric measure space @xmath3 .
let @xmath63 be measures with densities bounded from above by a constant @xmath181 .
then for every @xmath65 we have for every @xmath98 $ ] @xmath195 at @xmath27-almost every @xmath10 , where @xmath170 is the density of @xmath171 with respect to @xmath27 .
we prove the claim for @xmath55 .
the proof for @xmath53 is essentially the same .
take any @xmath196 and let @xmath170 be the density of @xmath171 with respect to @xmath27 .
take @xmath197 and @xmath198 so that @xmath199 with @xmath200 .
define @xmath201 as @xmath202 because of the strong displacement convexity assumption @xmath55 is convex also along @xmath203 .
write @xmath204 for the density of @xmath205 with respect to @xmath27 . from the displacement convexity of @xmath55 along the geodesic
@xmath203 we get an upper bound @xmath206 on the other hand , from jensen s inequality we obtain the lower bound @xmath207 combining these bounds we get @xmath208 and thus we have proven the claim .
now we are ready to prove theorem [ thm : cd1 ] noting that the rest of the weak local poincar inequalities follow with a similar proof .
suppose that @xmath8 , @xmath13 , @xmath163 and @xmath209 are given .
let us abbreviate @xmath210 .
define @xmath211 as the median of @xmath8 in the ball @xmath165 .
in other words , @xmath212 using the median @xmath211 we split the ball @xmath165 into two borel sets @xmath175 and @xmath176 so that @xmath213 , @xmath214 , @xmath215 and @xmath216 these sets exist because the measure @xmath27 has no atoms and thus we can split the level set @xmath217 , if necessary , so as to make the sets @xmath175 and @xmath176 have the same measure . let @xmath218 be the geodesic given by lemma [ lma : cd ] and let @xmath170 be the density of @xmath171 with respect to @xmath27 . by lemma
[ lma : cd ] we have for all @xmath98 $ ] at @xmath27-almost every @xmath219 @xmath220 now observe that from we get an equality @xmath221 for @xmath166-almost every @xmath222 .
therefore @xmath223 the rest of the proof follows the same lines as the proof of ( * ? ? ?
* theorem 2.5 ) .
let us repeat the key steps here for the convenience of the reader .
notice that @xmath166-almost every @xmath222 is contained in the ball @xmath224 . thus @xmath225
\(i ) in the proof of theorem [ thm : cd1 ] above we obtained only a weak version of the local poincar inequality because in general it might be that a geodesic between two points @xmath226 does not stay inside the ball @xmath16 .
in a metric space @xmath72 where balls are convex in the sense that all the geodesics between any two points in a ball stay inside the ball the above proof immediately gives a strong local poincar inequality .
for example in those @xmath56 spaces where the balls are convex we have @xmath227 \(ii ) in a doubling geodesic metric space a weak local poincar inequality implies a strong local poincar inequality .
when we move from the weak inequality to the strong inequality the constant usually has to be enlarged .
however , from the assumption that the entropy @xmath53 is strongly displacement convex we can directly prove a strong local poincar inequality with the same constant as in corollary [ cor : evi2 ] : @xmath228 with two simple observations we can modify the proof of theorem [ thm : cd1 ] to give this inequality .
the first observation is that we do not have to use geodesics in the proof but instead we can take a collection of curves whose lengths are uniformly bounded from above . as we will shortly see , suitable curves can be constructed by combining geodesics .
the second observation is that a geodesic from a point @xmath229 to the center @xmath163 always stays inside the ball @xmath16 . with a similar proof as for lemma [ lma : evi
] we can show that for a geodesic @xmath230 from @xmath231 to @xmath167 we have for all @xmath232 $ ] at @xmath27-almost @xmath219 the estimate @xmath233 where @xmath234 is the density of @xmath171 with respect to @xmath27 .
the same argument works for a geodesic @xmath235 from @xmath236 to @xmath167 . by lemma [ lma : evi ] we know that for any @xmath237 we have for every @xmath98 $ ] at @xmath27-almost every @xmath238 the same bound for the density of @xmath239 with respect to @xmath27 .
if we combine these three geodesics to a rectifiable curve @xmath240 \to \mathcal{p}(x)$ ] by defining @xmath241 we get a curve in @xmath70 joining @xmath231 to @xmath236 so that the support of @xmath242 is always inside @xmath165 and we have for all @xmath243 $ ] the upper bound @xmath244 for the density of @xmath242 at @xmath27-almost all @xmath238 .
also , when we consider the curve @xmath136 as a measure on the rectifiable curves of @xmath9 , this measure is concentrated on curves in @xmath9 that have length bounded above by @xmath245 .
therefore , the same estimates as in the proof of theorem [ thm : cd1 ] give us the strong local poincar inequality .
let us now turn to the proof of theorem [ thm : uniqueness ] .
we explain the idea behind the proof with the help of figure [ fig : idea ] . by assumption
the entropy functional @xmath53 is convex along the geodesic in the space of measures which connects measure @xmath156 to @xmath246 using only geodesics in the upper part of the space .
the graph of @xmath53 along this geodesic is drawn on the right with a dashed line . now consider a new geodesic between @xmath156 and @xmath154 that moves half of the measure using the upper part and half of the measure using the lower part of the space . with this geodesic
the support of the transported measure is split to the upper and lower parts of the space . at the time when the measure is fully split
the entropy is @xmath247 times the entropy of the corresponding measure along the geodesic which uses only the upper part of the space . on the graph in the figure [ fig : idea ]
the entropy along this new geodesic is drawn with a solid line . on the time intervals
@xmath248 $ ] and @xmath249 $ ] , when the support of the measure travels past the branching points of the space , the entropy has a dramatic change .
this change contradicts the convexity of the entropy functional along the new geodesic proving that such branching space does not satisfy the strong convexity assumption . guided by the idea presented above we try to reduce the general case to a situation corresponding to that in figure [ fig : idea ] .
the first step is to show that if there are multiple geodesics joining many points in the space then there are also geodesics which agree on some initial time interval and then branch out .
next we choose a subset from these geodesics for which the branching happens roughly at the same time and so that all of these geodesics branch out sufficiently . the first requirement guarantees that the large change in the entropy happens on a small enough time interval , as in figure [ fig : idea ] .
the second requirement tells us that the supports of the two branches of the measure are really disjoint , justifying the calculations for the drop in the entropy .
let us now make these steps rigorous .
we already know that the measure @xmath27 is doubling so in particular it is outer regular .
let us use this to show that there are lots of points @xmath251 where the two distinct geodesics from @xmath164 to @xmath163 agree on a small initial time interval .
let @xmath252 and @xmath253 be so that @xmath254 and @xmath255 .
( actually , we know from ( * ? ? ?
* theorem 2.3 ) that either @xmath27 is a point mass or @xmath255 for all @xmath252 and @xmath253 . )
take a small @xmath256 and @xmath257 from the convexity of @xmath53 we get @xmath258 where @xmath259 is the absolutely continuous part of the measure @xmath260 with respect to @xmath27 . on the other hand , we always have by jensen s inequality @xmath261 combining the previous two estimates we see that @xmath262 as @xmath263
. we also have @xmath264 with @xmath265 as @xmath263 . recalling that @xmath255 we get @xmath266 as @xmath263 , implying that @xmath267 for sufficiently small @xmath268 .
therefore also the set @xmath269 has positive @xmath27-measure .
the open interval @xmath270 can be covered by a countable number of intervals of the form @xmath271 $ ] with the restriction @xmath272 therefore there exist some such @xmath273 and @xmath274 for which the set @xmath275\text { but not on } [ 0,t_2]\ } \end{aligned}\ ] ] has positive @xmath27-measure .
now let @xmath276 so that the set @xmath277\text { and } d(\gamma(t),\gamma'(t ) ) > 2\delta \text { for some } t \in [ t_1,t_2]\ } \end{aligned}\ ] ] has positive @xmath27-measure . because @xmath283 , there is some @xmath284 so that the set @xmath285 has positive @xmath27-measure .
let us then select for all @xmath286 a pair of geodesics @xmath287 , @xmath288 so that @xmath289 , @xmath290 , @xmath291 for all @xmath292 $ ] , @xmath293 and @xmath294 . using these pairs we write @xmath295 and @xmath296 .
next we define two measures @xmath297 .
the first one is defined as @xmath298 and the second one as @xmath299 the definition of @xmath300 and @xmath301 guarantee that @xmath302 . here
@xmath303 and @xmath304 are the absolutely continuous parts of @xmath305 and @xmath306 with respect to the measure @xmath27 . using the convexity of @xmath53 first to the geodesic @xmath307 between times @xmath308 and @xmath282 and then separately to the geodesics @xmath309 and @xmath310 between times @xmath281 and @xmath311 , and finally using the inequality we arrive at the contradiction @xmath312 thus proving the claim .
10 l. ambrosio and n. gigli , _ a user s guide to optimal transport _ , preprint , 2011 .
l. ambrosio , n. gigli and g. savar , _ gradient flows in metric spaces and in the space of probability measures _ ,
lectures in mathematics eth zrich , birkhuser verlag , basel , second ed . , 2008 .
p. buser , _ a note on the isoperimetric constant _ , ann .
sci . cole norm .
* 15 * ( 1982 ) , 213230 .
j. cheeger , _ differentiability of lipschitz functions on metric measure spaces _ ,
9 * ( 1999 ) , no .
3 , 428517 . j. cheeger and t. colding ,
_ on the structure of spaces with ricci curvature bounded below .
iii _ , j. differential geom . * 54 * ( 2000 ) , no .
1 , 3774 . s. daneri and g. savar _ eulerian calculus for the displacement convexity in the wasserstein distance _ , siam j. math
40 * ( 2008 ) , no .
3 , 11041122 . a. figalli and c. villani , _ strong displacement convexity on riemannian manifolds _ , math . z. , * 257 * ( 2007 ) , 251259 . n. gigli and s .-
ohta , _ first variation formula in wasserstein spaces over compact alexandrov spaces _
, canad . math . bull . , to appear , ( 2010 ) .
p. hajasz and p. koskela , _
sobolev meets poincar _ , c. r. acad .
paris sr .
* 320 * ( 1995 ) , no . 10 , 12111215 .
p. hajasz and p. koskela , _
sobolev met poincar _
* 145 * ( 2000 ) , no . 688 .
j. heinonen , _ lectures on analysis on metric spaces _ , springer - verlag , new york , 2001 . j. heinonen and p. koskela , _ quasiconformal maps in metric spaces with controlled geometry _ , acta math .
* 181 * ( 1998 ) , no .
1 , 161 . j. lott and c. villani , _ ricci curvature for metric - measure spaces via optimal transport _ , ann . of math . * 169 * ( 2009 ) , no .
3 , 903991 . j. lott and c. villani , _ weak curvature conditions and functional inequalities _ , j. funct .
* 245 * ( 2007 ) , no .
1 , 311333 .
r. j. mccann , _ a convexity principle for interacting gases _ , adv .
, * 128 * ( 1997 ) , 153179 .
ohta , _ gradient flows on wasserstein spaces over compact alexandrov spaces _ ,
amer . j. math .
, * 131 * ( 2009 ) , 475516 .
ohta , _ on the measure contraction property of metric measure spaces _
, comment .
* 82 * ( 2007 ) , no .
4 , 805828 . m .-
von renesse , _ on local poincar via transportation _ , math . z. * 259 * ( 2008 ) , no .
1 , 2131 .
g. savar , _ gradient flows and diffusion semigroups in metric spaces under lower curvature bounds _ , c. r. math .
paris , * 345 * ( 2007 ) , 151154 .
s. semmes , _ finding curves on general spaces through quantitative topology , with applications to sobolev and poincar inequalities _ , selecta math .
* 2 * ( 1996 ) , 155295 .
sturm , _ on the geometry of metric measure spaces .
i _ , acta math .
* 196 * ( 2006 ) , no .
1 , 65131 . k .-
sturm , _ on the geometry of metric measure spaces .
ii _ , acta math .
* 196 * ( 2006 ) , no .
1 , 133177 . c. villani , _ optimal transport .
old and new _ , vol .
338 of grundlehren der mathematischen wissenschaften , springer - verlag , berlin , 2009 . | we prove local poincar inequalities under various curvature - dimension conditions which are stable under the measured gromov - hausdorff convergence .
the first class of spaces we consider is that of weak @xmath0 spaces as defined by lott and villani .
the second class of spaces we study consists of spaces where we have a flow satisfying an evolution variational inequality for either the rnyi entropy functional @xmath1 or the shannon entropy functional @xmath2 .
we also prove that if the rnyi entropy functional is strongly displacement convex in the wasserstein space , then at every point of the space we have unique geodesics to almost all points of the space . |
in the zoo of the heavy - light @xmath5 mesons , the properties of radially excited @xmath9 and orbitally excited @xmath10 states have been explored for a long time .
however , no such higher excited @xmath5 state has been definitely established .
the observation of @xmath0 and @xmath1 has stimulated much more interest in these higher excited states .
@xmath0 was first observed by belle @xcite in @xmath11 with mass @xmath12 mev and width @xmath13 mev .
@xmath0 was also observed by babar in both @xmath14 and @xmath15 channels @xcite .
this state is included in pdg10 @xcite with mass @xmath16 mev , @xmath17 and width @xmath18 mev .
the branching ratio @xmath19 has also been measured .
@xmath1 was first reported by babar @xcite in @xmath20 with mass @xmath21 and width @xmath22 mev .
it was observed again in both @xmath14 and @xmath15 channels @xcite .
this state is included in pdg10 @xcite with mass @xmath23 mev , width @xmath24 mev and branching ratio @xmath25 .
its @xmath26 has not been measured or assigned .
the observation of @xmath27 rules out the possibility as a @xmath28 state since a @xmath29 @xmath5 is forbidden to decay into @xmath15 , and this state is supposed to have spin - parity : @xmath30 .
@xmath0 and @xmath1 have been explained within some models .
@xmath0 was identified with the first radial excitation of @xmath31 @xcite , or the orbitally excited d - wave @xmath6 @xmath5 @xcite , or the mixture of them @xcite .
the interpretation of @xmath32 as the @xmath33 @xmath5 @xcite was ruled out , and @xmath32 is interpreted as the @xmath7 @xmath5 @xcite . however , theoretical interpretations of these states in different assignments are not completely consistent with experiments either on their spectrum or on their decay properties .
where to place these two excited states ? obviously , the observations of @xmath0 , @xmath1 and some other resonances @xcite bring us some puzzles .
the study of the properties of strong decays is believed a good way to identify new observed states . among models for strong decays ,
the @xmath2 quark - pair - creation model has been employed successfully to evaluate the ozi - allowed strong decays of both mesons and baryons @xcite .
thorough understanding of the success of the @xmath2 model has also been investigated @xcite . within the @xmath2 model , @xmath0 and @xmath1
have been analyzed @xcite . however , there are different assignments and conclusions to these two states in these references . especially , some theoretical predictions of the branching ratios @xmath34 and @xmath35 are not consistent with experiments , and the predicted mixing angle @xmath36 is different from each other . therefore , it would be interesting to study the uncertainties relevant to the interpretations of @xmath0 and @xmath1 within the @xmath2 model .
it is believed that the p - wave @xmath37 and @xmath5 mesons have been established @xcite . in the p - wave multiplets of heavy - light mesons
, the @xmath38 and @xmath39 states may mix with each other , so the observed @xmath40 states should be the mixed ones . the mixing scheme has been explored and the mixing angle has been determined in refs . @xcite .
similarly , the radially excited @xmath4 and the orbitally excited @xmath6 may mix with each other through some mechanism , which is known as the excited - vector - meson puzzle " @xcite .
this mixing will complex our understanding of the higher excited @xmath17 states .
therefore , it will be useful to find out how large the mixing effect on the explanation of the excited states is .
for example , to see whether the theoretical interpretation of the @xmath17 states is mixing scheme dependent or not , or to determine how large the mixing angle is . in this paper , @xmath0 and @xmath1
are re - studied within the @xmath2 model .
the paper is organized as follows . in sec.ii ,
the strong decay widths and branching ratios @xmath41 of @xmath0 and @xmath1 are evaluated in different cases .
finally , we present our conclusions and discussions in sec.iii .
since different assignments to @xmath0 and @xmath1 have been suggested in literature , evaluations of their two - body open - flavor strong decays are made according to these assignments . in the conventional quark model , @xmath0 and @xmath1
may be pure @xmath42 states , or the mixed states of @xmath42 states .
if the @xmath4 and @xmath6 @xmath5 ( same @xmath26 and similar masses ) mix with each other , the physically observed states @xcite should be the mixed @xmath8 ones . in the mixed case ,
the two orthogonal partners are denoted as @xcite latexmath:[\[\begin{aligned } \label{mixing } where @xmath36 is the mixing angle . in our evaluation ,
@xmath0 and @xmath1 are first assumed as pure states , and subsequently assumed mixed states . within the @xmath2 model
, the evaluation of the strong decays widths ( formula in ref .
@xcite is adopted ) involves some parameters : the strength of quark pair creation from the vacuum @xmath44 , the @xmath3 value in the simple harmonic oscillator ( sho ) wave functions and the constituent quark masses .
@xmath44 represents the probability that a quark - antiquark pair is created from the vacuum , and it is supposed as a universal parameter .
all the partial widths are proportional to @xmath45 .
@xmath3 in the sho wave function is the harmonic oscillator strength parameter , which can be fixed to reproduce the realistic root mean square radius of the sho wave function .
there are often two ways for choices of @xmath3 .
one way is to determine @xmath3 individually for each meson @xcite , and the other way is to choose @xmath3 universal for all mesons @xcite .
our investigation indicates that the ways of choices of @xmath3 play an important role in the evaluation of the strong decays .
therefore , our study of the decay widths is presented in the next two subsections in these two different ways , respectively . to extract the flavor coefficient ,
the ideal meson wave functions @xmath46 and @xmath47 are used .
other parameters are given as follows .
@xmath48 @xcite is employed , which leads also to the right total decay width of @xmath49 .
the masses of relevant pure mesons are taken from the 2010 particle data group ( pdg ) @xcite or ref .
the constituent quarks masses are taken to be @xmath50 gev , @xmath51 gev and @xmath52 gev @xcite . in this subsection , @xmath3 is chosen as special value for each meson as that in ref .
@xmath3 values and relevant meson masses used for our evaluation of @xmath0 and @xmath53 are listed in table i. [ cols="^,^,^,^,^,^,^,^,^,^,^,^ " , ] finally , the uncertainty related the variation of @xmath3 as a universal parameter for all mesons is explored . to see the dependence of the results on @xmath3 , @xmath0 and @xmath1
are assumed as the pure states .
the branching ratios @xmath41 and the total decay widths of @xmath0 and @xmath1 are plotted in fig .
4 . from this figure ,
the variation of @xmath3 in the preferred region brings very small uncertainty to the total decay width of @xmath0 , but may bring @xmath54 uncertainty to the total decay width of @xmath1 .
the variation of @xmath3 bring considerable uncertainties to the branching ratios of @xmath0 , but smaller uncertainty to the branching ratios of @xmath1 .
[ pic3 ] as @xmath55 and @xmath1 as @xmath7 versus @xmath3,title="fig:",height=177 ] as @xmath55 and @xmath1 as @xmath7 versus @xmath3,title="fig:",height=162 ]
in this work , we have studied the strong decay properties of @xmath0 and @xmath1 in the @xmath2 model .
it is found that the interpretation of these two states depends on the mixing schemes and the ways of choices of the harmonic oscillator parameter @xmath3 . in the evaluation of
the strong decays widths , there are often two ways to choose the the harmonic oscillator parameter @xmath3 . one way is to choose @xmath3 individually for each meson , and the other way is to choose @xmath3 universal for all mesons .
the radially excited @xmath4 and the orbitally excited @xmath6 may mix with each other .
the interpretations of @xmath0 and @xmath1 depend on whether the mixing exists or not .
when the mixing exists , the interpretations of @xmath0 and @xmath1 depend on these two different ways of choices of @xmath3 . in the case of special @xmath3 for each meson
, @xmath0 can not be interpreted as the pure @xmath4 @xmath5 , but may be interpreted as the @xmath6 .
@xmath1 may be interpreted as the pure @xmath7 .
when the mixing scheme is employed , @xmath0 prefers to be identified with the mixed @xmath8 state of @xmath4 and @xmath56 , and @xmath1 prefers to be identified with the orthogonal partner of @xmath0 with a large mixing angle @xmath36 , which implies that @xmath6 is predominant . in the case of fixed @xmath3 for all mesons , @xmath0 can not be interpreted as a pure @xmath4 , but may be interpreted as the @xmath6 .
@xmath1 may be interpreted as the pure @xmath7 .
@xmath0 can also be interpreted as the mixed state of @xmath4 and @xmath7 with a small mixing angle @xmath36 ( @xmath4 is predominant ) .
however , it is hard to interpret @xmath1 as the orthogonal partner of @xmath0 . without mixing between the orbitally and radially excited states ,
the interpretations of @xmath0 and @xmath1 are the same for all @xmath3 no matter @xmath3 is universal or not . however , the conclusions are different for different ways of choices of @xmath3 when the mixing exists .
this uncertainty my be inherent within the @xmath2 model and appears accidently when the mixing exists .
this uncertainty blurs our understanding of the observed states . to draw conclusions for @xmath0 and @xmath1 through their strong decays within the @xmath2 model
, one must be careful when the mixing exists .
the analysis of the strong decays in other models is necessary .
of course , the study of the production and other ways of decays will provide other important understandings to them .
some other uncertainties within the @xmath2 model are explored .
the constituent quark masses bring small uncertainty to the results , but the phase spaces may bring considerable one .
this work is supported by the national natural science foundation of china under grant no .
bing chen is also supported by shanghai university under the graduates innovation fund shucx111023 ( no .
a. 16 - 0101 - 10 - 018 ) .
50 k. abe _
et al . _ ( belle collaboration ) , arxiv : hep - ex/0608031 ; j. brodzicka _ et al . _ ( belle collaboration ) , phys .
lett . * 100 * , 092001 ( 2008 ) .
b. aubert _
( _ babar _ collaboration ) , phys .
d * 80 * , 092003 ( 2009 ) .
k. nakamura _
( particle data group ) , j. phys .
g * 37 * , 075021 ( 2010 ) .
b. aubert _
( _ babar _ collaboration ) , phys .
lett . * 97 * , 222001 ( 2006 ) .
j. benitez _
et al . _ , in ichep2010 , paris , 2010 ; p.del amo sanchez _ et al . _
( babar collaboration ) , phys .
d * 82 * , 111101 ( 2010 ) . bo zhang , xing liu , wei - zhen deng and shi - lin zhu , eur .
j. c * 50 * , 617 ( 2007 ) .
p. colangelo , f.de fazio , s. nicotri and m. rizzi , phys .
d * 77 * , 014012 ( 2008 ) .
bing chen , deng - xia wang and ailin zhang , phys .
d * 80 * , 071502(r ) ( 2009 ) .
bing chen , ling yuan and ailin zhang , phys .
d * 83 * , 114025 ( 2011 ) .
ailin zhang , nucl .
a * 856 * , 88 ( 2011 ) .
de - ming li and bing ma , phys .
d * 81 * , 014021 ( 2001 ) .
f.e . close , c.e .
thomas , o. lakhina , and e. s. swanson , phys .
b * 647 * , 159 ( 2007 ) .
e.van beveren and g. rupp , phys .
lett . * 97 * , 202001 ( 2006 ) .
p. colangelo , f.de fazio and s. nicotri , phys .
b * 642 * , 48 ( 2006 ) .
xian - hui zhong and qiang zhao , phys .
d * 78 * , 014029 ( 2008 ) .
l. micu , nucl . phys .
b * 10 * , 521 ( 1969 ) .
a. le yaouanc , l. oliver , o. p@xmath57ne and j.c .
raynal , phys .
d * 8 * , 2223 ( 1973 ) ; * 9 * , 1415 ( 1974 ) ; * 11 * , 1272 ( 1975 ) .
a. le yaouanc , l. oliver , o. p@xmath57ne and j.c .
raynal , phys . lett .
b * 71 * , 57 ( 1977 ) ; * 71 * , 397 ( 1977 ) ; * 72 * , 57 ( 1977 ) . m. chaichian and r. kogerler , ann . phys . *
124 * , 61 ( 1980 ) .
f. stancu and p. stassart , phys .
d * 38 * , 233 ( 1988 ) .
t. barnes , f.e .
close , p.r . page and e.s .
swanson , phys .
d * 55 * , 4157 ( 1997 ) .
t. barnes , n. black , and p. r. page , phys .
d * 68 * , 054014 ( 2003 ) .
s. godfrey and n. isgur , phys .
d * 32 * , 189 ( 1985 ) .
r. kokoski and n. isgur , phys .
d * 35 * , 907 ( 1987 ) .
f. stancu and p. stassart , phys .
d * 39 * , 343 ( 1989 ) ; * 42 * , 1521 ( 1990 ) . s. capstick and n. isgur , phys . rev .
d * 34 * , 2809 ( 1986 ) . s. capstick and w. roberts , phys . rev .
d * 47 * , 1994 ; * 49 * , 4570 ( 1994 ) .
p. geiger and e.s .
swanson , phys .
d * 50 * , 6855 ( 1994 ) .
ackleh , t. barnes , and e.s .
swanson , phys .
d * 54 * , 6811 ( 1996 ) .
s. godfrey and r. kokoski , phys .
d * 43 * , 1679 ( 1991 ) f.e .
close and e.s .
swanson , phys .
d * 72 * , 094004 ( 2005 ) .
t. barnes , s. godfrey , e.s .
swanson , phys .
d * 72 * , 054026 ( 2005 ) . | the strong decays of @xmath0 and @xmath1 are investigated within the @xmath2 model .
it is found that the interpretation of these two states depends on the mixing schemes and the ways of choices of the harmonic oscillator parameter @xmath3 .
if @xmath0 and @xmath1 are two pure states , @xmath0 seems impossibly the @xmath4 @xmath5 , but may be the @xmath6 @xmath5 .
@xmath1 may be the @xmath7 .
if there is mixing between the @xmath4 and @xmath6 , @xmath0 may be the mixed @xmath8 state with a small mixing angle in the case of a special @xmath3 for each meson , and @xmath1 is the orthogonal partner of @xmath0 ; @xmath0 may also be the mixed @xmath8 state with a large mixing angle based on a universal @xmath3 for all mesons , and @xmath1 seems impossibly the orthogonal partner of @xmath0 .
other uncertainties related to the choices of constituent quark masses and phase spaces are also explored . |
99 donnelly t w , hitlin d , schwartz m , walecka j d and wiesner s j 1974 _ phys .
_ b * 49 * 8 | we report the results of a calculation of the neutrino- and antineutrino - induced @xmath0-ray production cross section for oxygen target .
our analysis is focused on the kinematical region of neutrino energy larger than @xmath1200mev , in which single - nucleon knockout is known to be the dominant reaction mechanism .
the numerical results have been obtained using a realistic model of the target spectral function , extensively tested against electron - nucleus scattering data .
we find that at neutrino energy 600 mev the fraction of neutral - current interactions leading to emission of @xmath0-rays of energy larger than 6 mev is @xmath141% , and that the contribution of the @xmath2 state is overwhelming .
+ + _ invited paper to nufact 11 , xiiith international workshop on neutrino factories , super beams and beta beams , 1 - 6 august 2011 , cern and university of geneva + ( submitted to iop conference series ) _ the observation of @xmath0 rays originating from nuclear deexcitation can be exploited to identify neutral - current ( nc ) neutrino - nucleus interactions in a broad energy range . following the pioneering studies of nuclear excitations by neutral weak currents of refs .
, theoretical calculations of the cross section of @xmath0-ray production from nc neutrino - oxygen interactions have been carried out in the neutrino energy range @xmath3500 mev @xcite .
neutrons , while providing @xmath150% of nc events , do not emit cherenkov light . as a consequence ,
the availability of an alternative signal allowing one to identify nc interactions is very important .
events with @xmath0 rays of energy above the observational threshold of 5 mev can be detected in a water cherenkov detector , like super - kamiokande , and contribute up to @xmath15% of the total event number @xcite , independent of neutrino oscillations .
note that in water @xmath190% ( 16 out of 18 ) of the nc interactions take place in oxygen . at low energy , elastic scattering and inelastic excitation of discrete nuclear states provide the main contribution to the neutrino - nucleus cross section .
however , at @xmath4 mev the cross section associated with these processes tends to saturate , and quasielastic ( qe ) nucleon knockout becomes the dominant reaction mechanism .
if the residual nucleus is left in an excited state , these processes can also lead to @xmath0-ray emission . in the qe regime ,
neutrino - nucleus scattering reduces to the incoherent sum of elementary scattering processes involving individual nucleons , the energy and momentum of which are distributed according to the target spectral function @xcite . a schematic representation of nc qe neutrino - nucleus scattering is given in fig .
[ fig1 ] , where the dashed line represents the threshold for nucleon emission in the continuum .
low - lying excited levels of the residual nuclei produced in @xmath5[8]{o}(\nu,\nu'n)$ ] scattering.,scaledwidth=72.0% ] low - lying excited levels of the residual nuclei produced in @xmath5[8]{o}(\nu,\nu'n)$ ] scattering.,scaledwidth=96.0% ] in this paper , we discuss the emission of @xmath0 rays arising from the decay of the residual nuclei of the reactions @xmath6[8]{o}\rightarrow \nu+p+\isotope[15][7]{n}^*$ ] , or @xmath6[8]{o}\rightarrow \nu+n+\isotope[15][8]{o}^*$ ] , the cross sections of which have been computed using a realistic model of the oxygen spectral function . note that the presented formalism applies also to antineutrino - induced @xmath0-ray production . in our approach , covered in detail in ref .
@xcite , the cross section of @xmath0-ray production following a nc qe interaction , @xmath7 , is written in the form @xmath8[8]{o } \to \nu + \gamma + y
+ n ) = \sum_\alpha\sigma(\nu + \isotope[16][8]{o } \to \nu+x_\alpha + n ) \:\textrm{br}(x_\alpha \to \gamma + y),\ ] ] where @xmath9 is the knocked out nucleon , @xmath10 denotes the residual nucleus in the state @xmath11 , and @xmath12 is the system resulting from the electromagnetic decay of @xmath10 , e.g. , , @xmath13[7]{n } + n$ ] , or @xmath13[6]{c } + p$ ] @xcite .
the energy spectrum of the states of the residual nuclei is schematically illustrated in fig .
[ fig2 ] .
the branching ratios br@xmath14 have been taken from ref . .
the nc qe cross section , @xmath15[8]{o } \to \nu+x_\alpha + n)$ ] , has been calculated within the approach discussed in ref .
@xcite for the case of charged - current interactions .
it may be expressed as @xmath16 where @xmath17 , @xmath18 being the nucleon mass , @xmath19 denotes the elementary neutrino - nucleon cross section and the spectral function @xmath20 yields the probability of removing a nucleon of momentum @xmath21 from the target leaving the residual nucleus with energy @xmath22 , @xmath23 being the target ground - state energy .
according to the shell model , nuclear dynamics can be described by a mean field . in the simplest implementation of this model , protons in the @xmath5[8]{o}$ ]
nucleus occupy three single - particle states , @xmath24 , @xmath25 , and @xmath26 , with removal energy 12.1 , 18.4 , and @xmath142 mev , respectively @xcite .
the neutron levels exhibit the same pattern , see fig .
[ fig1 ] , but are more deeply bound by 3.54 mev @xcite . as a consequence of a mean field dynamics , knock out of a target nucleon leaves the residual system in a bound state , and the spectral function can be conveniently written in the form @xmath27 where @xmath28 is the momentum - space wave function associated with the @xmath11-th shell - model state and the sum is extended to all occupied states belonging to the fermi sea @xmath29 . the _ occupation probability _ @xmath30 and the ( unit - normalized ) function @xmath31 , describing the energy width of the @xmath11-th state , account for the effects of nucleon - nucleon ( nn ) correlations , not included in the mean field picture . in the absence of correlations , @xmath32 and @xmath33 .
a realistic model of the proton spectral function of oxygen has been obtained within the local density approximation ( lda ) , combining the experimental data of ref .
@xcite with the results of theoretical calculations of the correlation contribution in uniform nuclear matter at different densities @xcite .
the results reported in ref .
@xcite show that the lda spectral function provides an accurate description of the inclusive electron - oxygen cross sections at beam energies around 1 gev .
in addition , it predicts a nucleon momentum distribution in agreement with that obtained from the data of ref .
@xcite . as pointed out in ref .
@xcite , nucleon - knockout experiments measure _ spectroscopic strengths _ , not occupation probabilities .
spectroscopic strengths are given by the area below the sharp peaks observed in the missing - energy spectra , corresponding to knockout of a nucleon occupying one of the shell - model states , corrected to take into account final - state interactions .
on the other hand , occupation probabilities include contributions corresponding to larger removal energy , arising from mixing of the one - hole state with more complex final states @xcite .
@l|lll@ @xmath11 & @xmath34 & @xmath2 & @xmath35 + @xmath36 & 0.632 & 0.703 & 0.422 + br@xmath37 & 0% & 100% & @xmath38% + cross section for @xmath0-ray production following nc qe interaction of neutrino ( solid line ) and antineutrino ( long - dashed line ) compared to the nc qe cross section of neutrino ( dotted line ) and antineutrino ( short - dashed line ) .
only the photons of energy larger that 6 mev are considered . ]
the @xmath34 , @xmath2 and @xmath35 spectroscopic strengths have been computed by integrating the oxygen spectral function of refs .
@xcite over the energy ranges @xmath39 mev , @xmath40 mev , and @xmath41 mev , respectively . dividing these numbers by the degeneracy of the shell - model states , one obtains the quantities @xmath36 listed in table [ tab1 ] .
the same spectroscopic strengths have been used for protons and neutrons .
the branching ratios @xmath42 , necessary to calculate the cross section @xmath7 according to eq . , are collected in table [ tab1 ] . in the case of the @xmath34-proton ( neutron )
knockout , the residual nucleus is ( ) produced in its ground state . hence , no @xmath0-rays are produced .
as the @xmath2-proton ( neutron ) hole lies below the nucleon - emission threshold , 10.21 mev ( 7.30 mev ) , it always deexcites through photon emission with half - life @xmath43 fs ( less than 1.74 fs ) @xcite .
when a proton ( neutron ) is knocked out from the deepest @xmath35 shell , the excitation energy is high enough for many deexcitation channels to open , of which only two , @xmath13[6]{c}+p$ ] and @xmath13[7]{n}+n$ ] ( @xmath13[6]{c}+p$ ] and @xmath44[6]{c}+\alpha$ ] ) , yield photons of energy higher than 6 mev ( see fig . [ fig2 ] ) . in fig .
[ fig4 ] our results for the neutrino- and antineutrino - induced @xmath0-ray production cross section are compared to the neutrino and antineutrino nc qe cross sections .
the error bands show the uncertainties arising form the determination of the spectroscopic strengths ( 5.4% ) , the treatment of pauli blocking ( 1% ) , and the branching ratio of the @xmath35 state ( 1% ) .
the @xmath7 s dependence on neutrino energy is very similar , although not identical , to that of the nc qe cross section .
the discrepancy arises from difference between the average removal energy associated with the whole spectral function and the energy of the @xmath2 shell , yielding the overwhelming contribution to @xmath7 .
the neutrino - induced @xmath0-production cross section reaches its maximum at @xmath45 gev and is slowly decreasing at larger energies . on the other hand ,
the corresponding antineutrino cross section is an increasing function of @xmath46 . in conclusion ,
we have computed the neutrino- and antineutrino - induced @xmath0-ray production cross section for oxygen target , focussing on the kinematical region in which single - nucleon knockout dominates .
considering photons of energy larger than 6 mev , we find that the @xmath2 state provides the overwhelming contribution , and that the ratio @xmath47 , exhibiting a significant energy - dependence at @xmath48 gev , is @xmath141% at @xmath49mev .
our results , obtained using a realistic model of the target spectral function , provide an accurate estimate of a signal that can be exploited to identify neutral - current events in water - cherenkov detectors .
this work was supported by infn ( grant mb31 ) , miur prin ( grant `` many - body theory of nuclear systems and implications on the physics of neutron stars '' ) , and in part by the grant - in - aid from the japan society for promotion of science ( nos .
21224004 and 23340073 ) .
a.m.a . was supported by the polish ministry of science and higher education under grant no . 550/mob/2009/0 . |
identification and measurement of muons in the inner tracker using its high spatial resolution and excellent momentum measurement in the 4 t field is a very important and demanding task , particularly for muons in jets . in the muon system
the occupancy is low and pattern recognition is relatively easy , provided the particle reaches the muon system at all , which is the case for the barrel is p@xmath4 4 gev . for non - isolated muons
the main difficulty is to match the track observed in the muon stations with the correct candidate in the inner tracker in case of the high track density inside a jet .
these situations arise for example in muon tagging of @xmath5-jets , in @xmath5 @xmath2 @xmath6(@xmath2 @xmath3 @xmath3 ) + @xmath7 cascades in @xmath5 and @xmath8 physics , in susy(@xmath9 , @xmath10 ) physics and most dramatically in very high e@xmath1 ( tev scale ) jets such as heavy flavour ( @xmath5 , @xmath11 ) qcd jets giving rise to `` muon bundles '' within jets .
the matching strategy is based on the use of the full parameter space , i.e. on both the reconstructed momentum and the position in a given plane for the two tracks , one in the muon system and the other in the inner tracker .
the quantities monitored are the efficiency of matching ( number of muons matched over number of muons reconstructed ) , the muon matching impurity ( 1 - number of muons matched over number of tracks matched ) and generation of `` ghost '' and `` fake '' tracks .
the results discussed here are obtained using a detailed geant description of the complete cms detector of tdr design . in all simulations the effects of the 4 t field are fully included .
we have performed a geant simulation of the response to @xmath0 jets with transverse momenta p@xmath12 of 50 , 100 and 200 gev ( which corresponds to the p@xmath1 range of @xmath5-jets in @xmath13 production ) and at least one muon inside each jet with p@xmath14 @xmath15 5 gev in the rapidity region -1.1
@xmath16 @xmath17 @xmath16 1.1 .
1000 events were generated and reconstructed in the barrel part at each energy point .
the results are obtained using the standard cms reconstruction procedure ( data card muon 2 and trak 3 ) existing in the cmsim version 115 .
pattern recognition in the muon stations is done by building track segments from the reconstructed hits separately in the ( r,@xmath18 ) and ( r , z ) planes ; ( r@xmath18,rz ) track segments in the same muon station and in the same or contiguous ( r,@xmath18 ) sectors are considered to form a pair used to perform the kalman filter .
this is done starting from the outermost stations and proceeding towards the inner stations with a constrained fit to the impact point ( ip ) with default values of @xmath19 = 15 @xmath20 and @xmath21 = 5.3 cm .
since the decay mean transverse length ( impact parameter ) of the @xmath5-quarks is about 500 @xmath20 another reconstruction was made using a constrained fit to the impact point with @xmath19 = 500 @xmath20 and @xmath21 = 5.3 cm .
the global track finder ( gtf ) was used for pattern recognition in the cms inner tracker . the gtf algorithm adopted for the cms tracker must process a large number of hits per event : typically 5.10@xmath22 hits at low luminosity and ten times more at high luminosity . to overcome the severe combinatorial problems , the concept of road preselection is used in the first stage of the algorithm . in the second stage a kalman filter
is used to carry out final hit selection and track fitting .
given two track segments , one in the muon system and the other in the inner tracker , one can extrapolate both segments to a common plane or point , check the difference in momentum and position and select the best combination among the candidates .
the correct combination is known from the simulation procedure and it is therefore possible to calculate the efficiency of matching .
we use as a matching criterion the minimal distance computed from @xmath17 , @xmath18 and/or p@xmath23 of the two track candidates at the outer plane ( trci - cmsim convention ) of the inner tracker and at the impact point .
association of reconstructed tracks to geant ones is done on the hit basis .
the reconstructed track is associated to a given geant particle number if more than half of the hits assigned to the track are from that particular geant track .
if there is no such dominating track , the reconstructed track is called a `` ghost '' .
if more than one reconstructed track is matched to a given geant number it is called a `` fake '' . the @xmath17 dependence of the reconstruction efficiency obtained for the barrel muon station , shown on fig .
1a ) is rather uniform except for the region between the central and next - to - central rings ( @xmath24 @xmath16 0.2 ) and for @xmath24 @xmath15 0.9 . to avoid the influence of the @xmath17 dependence of the reconstruction efficiency on the matching results the tracks with @xmath24 @xmath15 0.9
are removed . the p@xmath1 reconstruction efficiency ( fig .
1b ) ) is more than 95% for the tracks with p@xmath14 @xmath15 10 gev . in fig .
1c ) is given the `` type '' of reconstructed tracks in the muon stations ( solid line ) for which must be found the best candidate between the tracks reconstructed in the tracker ( error bars ) .
the comparison between reconstructed and generated variables : p@xmath1 , @xmath17 and @xmath18 is given in fig .
1d ) - fig .
1i ) both for the barrel muon stations and for the inner tracker , for p@xmath12 = 100 gev .
we find a p@xmath1 muon resolution of about 10% for the muon stations alone , and of @xmath16 1% for the inner tracker , in good agreement with the values given in the muon and tracker tdrs@xmath25 . to estimate the matching quality a @xmath26 is formed from the differences between reconstructed @xmath17 , @xmath18 and p@xmath1 in the two detectors for all reconstructed track combinations at the outer tracker plane and at the impact point : @xmath27 including in the @xmath26 , equation ( 1 ) , the contribution from p@xmath23 in the muon stations and the tracker improves the selection of the matching tracks mainly for the high transverse momenta .
a minimal @xmath26 of the matching tracks has been used as a matching criterion . the p@xmath1 dependence of the matching efficiency is shown on fig . 2 for p@xmath12 = 50 , 100 and 200 gev at the outer plane trci of the inner tracker and at the interaction point with the two assumptions for the constrained fit to the impact point with @xmath28 = 15 @xmath20 and @xmath28 = 500 @xmath20 .
the degradation of the matching efficiency at higher p@xmath12 is is due to the increase of charged particle multiplicity inside jets which makes more difficult the correct association of the tracks observed in the muon stations with the track candidates in the inner tracker .
the matching efficiency shows less variation when a constrained fit at the impact point is performed using @xmath28 = 500 @xmath20 .
results obtained from the matching of muons inside the @xmath0 jets at p@xmath12 of 50 , 100 and 200 gev are summarized in table 1 .
the `` type '' of the matched tracks indicates a higher purity at the impact point and at lower jet momenta . for the higher multiplicity and more collimated jets at p@xmath12 = 200 gev
, the impurity in matching degrades to a @xmath29 7% level . rejecting tracks with an increasing momentum threshold @xmath16 p@xmath30 ,
a high matching efficiency and purity of track matching can be obtained for high p@xmath12 , but at the expense of a substantial reduction in statistics .
the number of `` ghost '' tracks generated at reconstruction in the muon stations is negligible .
the number of `` fake '' tracks shows some tendency to increase with jets momenta , and is at level of 1% - 2% . after performing the matching procedure we can assign to the tracks observed in the muon station the precise measurements from the tracker .
the distributions of the differences between reconstructed and generated values of p@xmath1 , @xmath17 and @xmath18 at the impact point for the matched tracks at p@xmath12 = 100 gev are given in fig .
3 . these distributions are practically the same as in fig .
2 , where we compared reconstructed and generated variables for the inner tracker , confirming the good quality of the matching criterion . the bottom row in fig .
3 gives the `` type '' of the tracks reconstructed in the muon stations and the `` type '' of matched tracks at the impact point .
the quality of track matching can be measured by the value of the `` pull '' : @xmath31 where : z@xmath32 , z@xmath33 is the track charge in the tracker and the muon station ; p@xmath32 , p@xmath33 is the reconstructed momentum in tracker and muon station ; cov(1/p.1/p ) is the covariance matrix . a gaussian fit to the pull shown at the bottom of fig .
3 gives @xmath34 @xmath29 1.0 indicating a good accuracy for the selection and matching of the tracks .
the main results concerning the matching of muon tracks in ( @xmath0 ) jets between the muon stations and inner tracker are : using a simple minimal distance criterion between @xmath17 , @xmath18 and p@xmath1 of independent track measurements at the impact point one can obtain a high matching efficiency with a contamination of not more than 4% for p@xmath12 of 50 gev and 6% for 100 gev . for the more collimated jets with p@xmath12 = 200 gev ,
the impurity in matching is at a @xmath29 7% level .
the matching efficiency shows less variation when a constrained fit at the impact point is performed with @xmath28 = 500 @xmath20 . to further improve the matching efficiency and impurity a more sophisticated software than currently available
must be used , for example a constrain fir to the secondary @xmath0 jets vertices obtained from the inner tracker can be performed . | for most physical processes the tracks observed in the muon stations must be matched with the corresponding tracks in the inner tracker , the external muon system providing muon identification and triggering but the tracker points giving the precise momentum measurement at lower momenta .
for high momenta the momentum resolution is much improved by the interconnection of inner and outer measurements .
the matching of outer and inner measurements is more delicate in case of muons embedded in jets .
a study of the matching procedure was carried out using samples of @xmath0 jets at high p@xmath1 , requiring @xmath0 @xmath2 @xmath3 decays . * cms note 1999/057 * * muon track matching * + a. benvenuti + infn , bologna , italy + d. denegri + dapnia , saclay , france + v. genchev , p. vankov + inrne , sofia , bulgaria + a. khanov , n. stepanov + itep , moscow , russia
+ ( november 1 , 1999 ) |
in @xcite , c. huneke and r. wiegand propose the following conjecture regarding torsion submodules of tensor products .
[ c : hw ] let @xmath3 be a one - dimensional gorenstein domain .
let @xmath4 be a finitely generated @xmath3-module , which is not projective .
then the torsion submodule of @xmath5 is non - trivial .
recently , this still - open conjecture has spurred much subsequent work ( see @xcite ) .
of particular interest is @xcite , where p. garca - snchez and m. leamer study this conjecture in special cases related to numerical monoid algebras . given a numerical monoid @xmath1 and @xmath6 , they construct a monoid , which we denote as @xmath0 , whose elements correspond to arithmetic sequences in @xmath1 of step size @xmath2 and whose monoid operation is set - wise addition
. these monoids , which we refer to as _
leamer monoids _ in honor of @xcite , reduce a special case of the huneke - wiegand conjecture to finding irreducible elements of a certain type .
[ p : hw ] let @xmath1 be a numerical monoid and @xmath7 be a field .
the monoid algebra @xmath8 $ ] satisfies the huneke - wiegand conjecture for monomial ideals generated by two elements if and only if for each @xmath6 , there exists an irreducible arithmetic sequence of the form @xmath9 in @xmath1 .
thus , understanding the atomic structure of leamer monoids could provide progress towards proving the huneke - wiegand conjecture . in investigating these algebraic objects , we find that , beyond their importance in commutative algebra , leamer monoids possess a very interesting factorization theory with numerous notable properties .
we investigate several factorization invariants of @xmath0 , including a computation of elasticity and a bound on the delta set . in special cases ,
we provide exact formulae for length sets , delta sets , and catenary degrees . although elements in leamer monoids are arithmetic sequences in @xmath1 of a fixed step size , the atomic properties of @xmath0 differ greatly from the numerical monoid @xmath1 from which they are derived .
in particular , many of the computable invariants for numerical monoids ( e.g. delta sets , see @xcite ) are harder to establish in leamer monoids . thus , investigating special cases ( i.e. arithmetical leamer monoids ) becomes both necessary and fruitful .
as leamer monoids are a novel construction , this paper provides much of the foundational theory for studying these intriguing algebraic objects . in section [ s : defn ] , we provide several definitions and examples of leamer monoids and investigate their atomic structure , showing that although they possess infinitely many irreducible elements , these atoms are still fairly constrained . in section [ s : elasticitydelta ] , we show that leamer monoids have infinite elasticity and give an explicit bound for the maximal element of their delta sets . in section [
s : arithmetical ] , we investigate the special case of arithmetical leamer monoids , that is , leamer monoids @xmath0 where @xmath1 is generated by an arithmetic sequence with step size @xmath2 . for these leamer monoids
, we provide an exact description of the irreducible and reducible elements and use this to derive formulae for their delta sets and catenary degrees . in section [ s : future ] , we provide several open problems related to factorization invariants of leamer monoids and the study of @xmath0 when @xmath1 has certain types of generating sets .
in what follows , we let @xmath10 be a numerical monoid ( i.e. , co - finite additive submonoid of @xmath11 ) where @xmath12 constitutes its minimal generating set and @xmath13 . [
d : leamer ] given a numerical monoid @xmath1 , @xmath14 , define @xmath15 that is , @xmath0 is the collection of arithmetic sequences of step size @xmath2 contained in @xmath1 .
since @xmath0 is a subset of @xmath16 , it inherits its natural component - wise additive structure .
it is clear that @xmath0 is closed under this operation , so @xmath0 is an additive submonoid of @xmath16 . throughout this paper , we denote by @xmath17 the set of irreducible elements ( or atoms ) of @xmath0 . building upon functions in the ` gap ` package ` numericalsgps ` @xcite , we are able to plot the elements of @xmath0 and indicate ( by large red dots ) the irreducible elements
; see figures [ f : leamergraph ] and [ f : anotherleamergraph ] .
[ e:7_10_s=3 ] for @xmath18 and @xmath19 , the graph of the leamer monoid @xmath20 is given below in figure [ f : leamergraph ] .
in this example , @xmath1 is generated by an arithmetic sequence with step size @xmath21 , which is equal to @xmath2 .
the frobenius number of @xmath1 ( 53 in this case ) can be easily read off the graph as the rightmost column absent of any dots .
.,width=384 ] as in example [ e:7_10_s=3 ] , leamer monoids @xmath0 where @xmath1 is of the form @xmath22 for some @xmath23 have the best understood set of irreducibles ( see section [ s : arithmetical ] for a detailed discussion ) . when @xmath1 is not generated by an arithmetic sequence ( or even when it is , but @xmath2 is not equal to the step size between the generators ) , the leamer monoids have less predictable structure .
[ e:13_17_22_40_s=4 ] for @xmath24 and @xmath25 , the leamer monoid @xmath26 is given in figure [ f : anotherleamergraph ] .
unlike example [ e:7_10_s=3 ] , this monoid possesses numerous irreducibles with @xmath27 before the left - most infinite column .
these atoms make the analysis of such leamer monoids much more difficult .
.,width=384 ] motivated by these graphical representations , we present the following definitions that provide insight into the structure of leamer monoids . [
d : column ] for a leamer monoid @xmath28 and @xmath29 , the _ column at @xmath30 _ is the set @xmath31 if this set is empty , we say that no column exists at @xmath30 .
if a column exists at @xmath30 , the column at @xmath30 is said to be _ finite _ ( resp . ,
infinite ) if the column has finite ( resp . , infinite )
the _ height _ of the finite column at @xmath30 is @xmath32 definition [ d : leamer ] makes sense for any @xmath33 .
however , if @xmath34 , then the atomic structure of @xmath0 is inherited directly from @xmath1 .
furthermore , @xmath35 as monoids , and the latter corresponds to the ( equivalent ) embedding of @xmath0 in @xmath16 where the point @xmath36 represents the arithmetic sequence @xmath37 rather than @xmath38 .
many of the statements in lemma [ l : atoms ] would be different under this new embedding ; for instance , every column of @xmath39 is finite . the first non - identity element in a leamer monoid and the first infinite column play important computational roles in understanding the monoid structure of @xmath28 .
we provide their notation below . [
d : x0 ] given a leamer monoid @xmath28 , we use @xmath40 to denote the smallest @xmath30 such that @xmath41 .
we denote by @xmath42 the first infinite column of @xmath0 , that is , the smallest @xmath30 such that @xmath43 for all @xmath44 .
when @xmath0 is clear from the context , we refer to these as simply @xmath45 and @xmath46 .
if @xmath47 is the frobenius number of @xmath1 , then clearly there exists an infinite column at @xmath48 .
in particular , both @xmath45 and @xmath46 exist ; see lemma [ l : atoms ] below . since the connection between leamer monoids and the huneke - wiegand conjecture centers on the existence of irreducible elements of length @xmath49 , understanding the factorization structure of @xmath28 is of significant importance .
the first important observation is that leamer monoids are atomic .
we leave the proof of this to the reader .
[ t : atomic ] let @xmath1 be a numerical monoid with @xmath50 .
the leamer monoid @xmath0 is atomic .
that is , any @xmath43 can be written as a finite sum of irreducible elements .
the following observations are motivated by the graphs in figures [ f : leamergraph ] and [ f : anotherleamergraph ] .
[ l : atoms ] let @xmath28 be a leamer monoid . a. if @xmath41 , then @xmath51 b. for @xmath52 , @xmath53 . c. if @xmath54 , then @xmath55 for all @xmath56 .
d. if @xmath57 and @xmath54 for @xmath58 , then @xmath59 e. if @xmath57 and @xmath60 for @xmath58 , then @xmath61 f. the column at every @xmath62 is infinite .
g. for all @xmath63 and @xmath27 , @xmath36 is a reducible element in @xmath0 .
if @xmath64 with @xmath58 , then at least one of @xmath65 or @xmath66 is greater than @xmath67 , yielding a factorization for @xmath68 .
this proves the contrapositive of ( d ) . for ( g ) , notice that for @xmath27 and @xmath63 , we can write @xmath69 .
the remaining proofs are left to the reader .
the above lemma motivates the following definitions .
[ d : atoms ] fix a leamer monoid @xmath0 and an @xmath70 whose column in @xmath0 is non - empty . the element @xmath71 is called _ trivial_. the column at @xmath30 is called _ purely irreducible _ ( or _ purely atomic _ ) if it contains only atoms , _ mixed _ if it contains both reducible and non - trivial irreducible elements , and _
purely reducible _ otherwise . in figure
[ f : leamergraph ] , the red dots with height 1 are all trivial atoms , whereas the element @xmath72 is a nontrivial atom .
the column at 60 is purely atomic , since every element is irreducible , and the column at 52 is purely reducible , since its only atom has height 1 .
the column at 48 has 3 reducible elements , but every element with height at least 5 is irreducible , so this column is mixed .
[ r : bound ] the leamer monoid in example [ e:7_10_s=3 ] has @xmath73 and @xmath74 and has non - trivial irreducibles at @xmath75 .
thus , the bound presented in lemma [ l : atoms](g ) is sharp . on the other hand , the leamer monoid in example [ e:13_17_22_40_s=4 ] has @xmath76 with the last non - trivial irreducible occurring at @xmath77 .
thus , the bound is not always reached .
in atomic , cancellative , commutative monoids , various invariants have been used to measure how far an element is from have unique factorization into irreducible elements . in this section ,
we investigate several of these invariants for leamer monoids .
we begin by analyzing the elasticity of leamer monoids . for more detail on elasticity ,
see ( * ? ? ?
* definition 1.4.1 ) .
[ d : elasticity ] fix a leamer monoid @xmath0 .
the _ set of lengths _ of @xmath54 is given by @xmath78 let @xmath79 and @xmath80 denote the minimum and maximum factorization lengths of @xmath36 , respectively .
the _ elasticity _ of @xmath36 is given by @xmath81 , and the _ elasticity _ of @xmath0 is given by @xmath82 we begin by showing that every leamer monoid has infinite elasticity .
[ t : infelasticity ] for any leamer monoid @xmath0 , @xmath83 for @xmath84 , we have @xmath85 so @xmath86 . letting @xmath87 tend
to infinity completes the argument .
variants of elasticity also appear in factorization theory literature ( see @xcite and @xcite ) . for convenience ,
we define these here . [ d : fullyelastic ] the _ @xmath88-th refined elasticity of @xmath0 _ is defined by @xmath89 for @xmath90 .
we say @xmath0 is _ fully elastic _ if @xmath91 .
we now compute the refined elasticity for leamer monoids .
[ t : fullyelastic ] let @xmath0 be a leamer monoid . for all @xmath92 , @xmath93 .
moreover , @xmath0 is never fully elastic .
fix @xmath94 such that @xmath95 is an atom in @xmath0 . for @xmath96 ,
we have @xmath97 so @xmath98 and @xmath99 .
in fact , whenever @xmath100 and @xmath101 , we have @xmath102 , so @xmath103 .
this shows @xmath104 from which the second claim follows .
we continue this section with a discussion on the delta sets of leamer monoids .
in particular , we show that the delta set of any leamer monoid is finite , and we give a method to compute its maximal element . for more information on delta sets of commutative cancellative monoids , see ( * ? ? ?
* section 1.4 ) .
fix a leamer monoid @xmath0 .
delta set _ of @xmath54 is given by @xmath105 where @xmath106 .
delta set _ of @xmath0 is given by @xmath107 [ t : nstar ] fix a leamer monoid @xmath0 .
let @xmath108 that is , the set of reducible elements which lie in finite or mixed columns and have maximal height .
then @xmath109 , and in fact @xmath110 , where @xmath111 fix a reducible element @xmath54 . by
* lemma 4.1 ) , it suffices to show that @xmath112 .
if @xmath113 , then so is @xmath114 .
if @xmath115 , the column at @xmath30 is infinite , so we can write @xmath116 where there is a column at @xmath117 and an infinite column containing nontrivial irreducibles at @xmath118 . since @xmath119 is a sum of atoms , we have @xmath103 .
we now examine the length set @xmath120 for elements @xmath54 with @xmath121 . [
p : colstab ] fix a leamer monoid @xmath0 .
let @xmath122 for @xmath123 , and let @xmath124 for each @xmath125 with @xmath126 , we have @xmath127 for some integer @xmath128 . in particular , @xmath129 .
fix @xmath54 with @xmath126 , and a factorization @xmath130 .
then @xmath131 which gives @xmath132 , so some @xmath133 must reside in an infinite column with @xmath134 . by lemma [ l : atoms](d ) , we have @xmath135 for all @xmath136 .
this yields a factorization @xmath137 .
this means @xmath138 , and for @xmath139 , @xmath140 for distinct indices @xmath141 .
this proves both claims .
the proof of the following proposition is similar to that of proposition [ p : colstab ] and is left to the reader .
[ p : rowstab ] fix a leamer monoid @xmath0 . for each @xmath54 with @xmath142
, we have @xmath143 .
in particular , @xmath144 . to conclude this section ,
we give a method to find @xmath145 by giving a bounded region in @xmath16 in which it must occur .
this effectively gives an algorithm to compute the maximum value in the delta set for any given leamer monoid ; see remark [ r : deltaregion ] .
first , we give a technical lemma .
[ l : deltafactlen ] fix an atomic , cancellative monoid @xmath146 with @xmath147 finite . for @xmath148
, there exists @xmath149 with @xmath150 for some @xmath151 and @xmath152 .
fix @xmath153 with irreducible factorizations @xmath154 and no factorizations of length strictly between @xmath155 and @xmath156 .
if @xmath157 , then @xmath158 for some @xmath159 .
let @xmath160 .
this gives a factorization @xmath161 , so we must have @xmath162 .
in particular , @xmath151 , so for @xmath163 maximal , @xmath118 can not have a factorization with length strictly between @xmath164 and @xmath163 .
[ t : maxdelta ] let @xmath165 , let @xmath166 be defined as in proposition [ p : colstab ] and let @xmath167 denote a nontrivial irreducible in an infinite column
. then @xmath168 where @xmath169 .
fix @xmath54 .
if @xmath126 , then by proposition [ p : colstab ] , we have @xmath129
. now suppose @xmath170 . if @xmath171 , then by proposition [ p : rowstab ] , @xmath172 . if @xmath173 , then for @xmath174 , we can write @xmath175 , so @xmath176 .
thus , if @xmath177 , then by lemma [ l : deltafactlen ] some delta set value @xmath178 must occur in a column before @xmath30 .
[ r : deltaregion ] while the delta sets in the regions described in propositions [ p : colstab ] and [ p : rowstab ] are very well behaved , nontrivial delta set elements often occur throughout the remaining elements .
this makes it very difficult to find a region on which the entire delta set is obtained .
this problem is solved when the delta set is an interval ; see question [ q : deltainterval ] .
[ e : deltabound ] in many of the examples we computed , we found that @xmath179 .
however , this equality does not always hold .
for instance , when @xmath24 and @xmath25 , we have @xmath180 but @xmath181 ( see example [ e:13_17_22_40_s=4 ] ) .
in this section , we discuss a leamer monoid @xmath0 where @xmath1 is generated by an arithmetic sequence with step size @xmath2 . in particular , we give a complete characterization of the leamer monoids of this form , and use this to give a closed form for their delta sets .
[ d : arithmetical ] the leamer monoid @xmath0 is _ arithmetical _ if @xmath22 for some @xmath23 . for @xmath182 satisfying @xmath183 and @xmath184 ,
let @xmath185 , and let @xmath186 .
we begin by giving a closed form for the first infinite column of an arithmetical leamer monoid .
[ p : arithxf ] fix an arithmetical leamer monoid @xmath187 .
the first infinite column @xmath46 of @xmath187 is given by @xmath188 the second equality follows from ( * ? ? ?
* theorem 3.3.2 ) . for @xmath189 , we can write @xmath190 for some @xmath191 .
this means the column at @xmath192 is finite , so the column at @xmath30 is finite .
now let @xmath193 . by ( * ? ? ?
* theorem 3.1 ) , @xmath194 for @xmath195 .
notice that @xmath196 so @xmath197 .
this completes the proof .
we now give a complete characterization of the elements and atoms of arithmetical leamer monoids .
[ t : arith ] fix an arithmetical leamer monoid @xmath187 , fix @xmath198 with @xmath199 , and further let @xmath200 .
then we have the following . 1 .
@xmath187 has a finite column at @xmath30 if and only if @xmath201 and @xmath202 . in this case , the column at @xmath30 has height @xmath203 .
@xmath187 has an infinite column at @xmath30 if and only if @xmath204 .
3 . if the column at @xmath30 is finite , then it has nontrivial atoms if and only if @xmath205 and @xmath90 , in which case it consists entirely of atoms . 4 . if the column at @xmath30 is infinite , then it has nontrivial irreducibles if and only if @xmath206 . in this case
, the first nontrivial irreducible in @xmath30 has height @xmath207 . by ( * ?
* theorem 3.1 ) , each @xmath29 can be written as @xmath200 for some @xmath208 with @xmath199 . if @xmath209 , then @xmath210 , so @xmath211 it then follows by proposition [ p : arithxf ] that @xmath187 has an infinite column at @xmath30 . if @xmath201 , then by ( * ? ? ?
* theorem 3.1 ) , we must have @xmath212 .
in particular , we have @xmath213 since this is a unique factorization in @xmath214 .
this means @xmath0 has a finite column at @xmath215 of height @xmath203 for each @xmath202 . by (
* theorem 3.1 ) , we have considered every element of @xmath1 , and thus have accounted for every column of @xmath187 .
this proves ( a ) and ( b ) . now , suppose the column at @xmath200 is finite . if @xmath205 , then @xmath30 is a minimal generator for @xmath1 , so every element in the column at @xmath30 is irreducible . if @xmath216 and the column at @xmath30 does not have height 1 , then for @xmath217 , we can write @xmath218 so the top element in the column at @xmath30 is reducible , and by lemma [ l : atoms](d ) , so are the nontrivial elements below it .
finally , suppose the column at @xmath200 is infinite .
if @xmath219 , then there is an infinite column at @xmath220 , so the column at @xmath30 is purely reducible .
now suppose @xmath206 .
the column at @xmath220 is either empty or finite , so the column at @xmath30 must contain nontrivial irreducibles .
let @xmath221 denote the height of the lowest nontrivial irreducible element in the column at @xmath30 .
for any expression @xmath222 of @xmath30 in terms of finite columns @xmath223 , we have @xmath224 , which gives @xmath225 with equality as long as @xmath226 , as desired .
theorem [ t : arith ] shows that , in particular , each @xmath187 has an irreducible of height 2 .
this , together with proposition [ p : hw ] , yields the following .
let @xmath227 $ ] , and @xmath228 be a monomial ideal with @xmath229 .
then the ideal @xmath230 satisfies the huneke - wiegand conjecture .
we now give a closed form for the delta sets for this class of leamer monoids .
[ t : turtle ] fix an arithmetical leamer monoid @xmath187 .
then @xmath231 suppose @xmath232 .
let @xmath233 and fix @xmath234 .
we claim @xmath235 . by theorem [ t : arith ]
, we have factorizations @xmath236 so @xmath237 .
now , in order for a nontrivial irreducible in the column at @xmath200 to appear in a factorization of @xmath238 , we must have a column at @xmath239 if the column at @xmath30 is infinite , then by theorem [ t : arith ] , @xmath240 , so @xmath241 .
this means any factorization containing a nontrivial irreducible from @xmath30 must have length 2 .
thus , any nontrivial irreducible in a factorization of length greater than 2 must lie in a finite column .
we claim that @xmath242 has no factorization of length at most @xmath243 consisting entirely of irreducibles in columns of the form @xmath244 for some @xmath245 .
fix a sum @xmath246 of irreducibles with @xmath247 . by theorem [ t : arith ] ,
@xmath248 for each @xmath249 , so this sum can have height at most @xmath250 .
this means @xmath251 so if @xmath252 , this can be at most @xmath253 this means there can be no factorization of @xmath238 of this form . by the above
, every factorization of @xmath238 of length at most @xmath243 must contain a trivial irreducible .
but this means the second coordinates of the ( at most @xmath254 ) remaining factors must sum to @xmath255 , which is impossible since each such coordinate can be at most @xmath88 .
thus , there can be no factorization of @xmath238 of length strictly between @xmath49 and @xmath256 , meaning @xmath257 .
now , for @xmath258 , @xmath259 and @xmath260 , a similar argument shows that @xmath261 .
thus , for any @xmath262 , @xmath263 . by (
* lemma 4.1 ) and theorem [ t : arith ] , we have @xmath264 , and this completes the proof .
we have the following immediate corollary . for a full treatment on the catenary degree , see @xcite or @xcite .
[ c : catenary ] the arithmetical leamer monoid @xmath187 has catenary degree @xmath265 .
this follows from theorem [ t : turtle ] , ( * ? ? ?
* lemma 4.1 ) and ( * ? ? ?
* lemma 1.6.2 ) .
this paper develops many foundational theorems and tools related to leamer monoids , which have a very rich factorization theory .
thus , many open problems and questions about leamer monoids remain .
[ p : algorithm ] find an algorithm to compute @xmath266 for any leamer monoid .
[ q : deltainterval ] for every leamer monoid @xmath0 , does there exists an @xmath146 such that @xmath267 ?
[ p : catenary ] find the catenary degree @xmath268 for any leamer monoid @xmath0 .
a new invariant measuring how far an element is from being prime , called @xmath269-primality , has been studied in several different settings , including numerical monoids ( see @xcite ) .
a natural problem is to study @xmath269-primality in leamer monoids .
[ p : omega ] study the @xmath269-function @xmath270 for elements @xmath54 in any leamer monoid @xmath0 .
understanding leamer monoids associated to numerical monoids with special generators is also of interest .
see ( * ? ? ?
* lemma 1 ) and ( * ? ? ?
* theorem 3.1 ) for relevant membership criteria .
[ p : generatedgenarith ] a _ generalized arithmetic sequence _ is a sequence of the form @xmath271 where @xmath272 with @xmath273 and @xmath274 .
study leamer monoids @xmath0 where @xmath1 is generated by a generalized arithmetic sequence .
[ p : generatedinterval ] study leamer monoids @xmath0 where @xmath1 is generated by an interval of natural numbers .
much of this work was completed during the pacific undergraduate research experience in mathematics ( pure math ) , which was funded by national science foundation grants dms-1035147 and dms-1045082 and a supplementary grant from the national security agency .
the authors would like to thank scott chapman , pedro garca - snchez , and micah leamer for their numerous helpful conversations , as well as the anonymous referee for their very helpful comments .
paul baginski , scott t. chapman and george j. schaeffer .
_ on the delta set of a singular arithmetical congruence monoid _ , journal de thorie des nombres de bordeaux ( 2008 ) , volume : 20 , issue : 1 , page 4559 . | the huneke - wiegand conjecture has prompted much recent research in commutative algebra . in studying this conjecture for certain classes of rings , garca - snchez and leamer
construct a monoid @xmath0 whose elements correspond to arithmetic sequences in a numerical monoid @xmath1 of step size @xmath2 .
these monoids , which we call leamer monoids , possess a very interesting factorization theory that is significantly different from the numerical monoids from which they are derived . in this paper
, we offer much of the foundational theory of leamer monoids , including an analysis of their atomic structure , and investigate certain factorization invariants .
furthermore , when @xmath0 is an arithmetical leamer monoid , we give an exact description of its atoms and use this to provide explicit formulae for its delta set and catenary degree . |
in recent years , our knowledge of the high redshift universe has increased dramatically allowing us to start constructing a picture of how the different morphological types evolve ( e.g. steidel and hamilton 1992 ; madau 1995 ; lilly et al .
1995 ; cowie et al . 1996 ; ellis et al . 1996 ;
steidel et al . 1998 ) .
in particular , the comoving star formation history first depicted by madau et al .
( 1996 ) has resulted in a very useful way of quantifying galaxy evolution , under certain hypothesis .
since star formation directly translates into metal enrichment of the interstellar medium ( ism ) , observations of these two processes help astronomers to constrain models of the formation of the structure .
in particular , lyman alpha forests , damped lyman alpha systems and lyman alpha break galaxies have all contributed to estimate the cosmic metal ejection and star formation rates ( e.g. , lu et al . 1996 ; pettini et al .
1997 ) . however , we are still far from drawing a consistent picture since several points remain to be clarified . in this respect ,
the integration of the dust corrected cosmic star formation history of the universe up to @xmath0 can account for the entire stellar mass content in spirals and spheroids at present times . however , when we look at the metal content at @xmath1 , only @xmath2 of what is expected is actually measured ( renzini 1998 ) . regarding the chemical properties of galaxies ,
most of the available observations are restricted to the galaxy , and in particular , to the solar neighbourhood ( e.g. , edvardsson et al .
1993 ; gratton et al . 1996
; rocha - pinto et al .
extragalactic observations of hii regions provide information on the chemical content in other galaxies ( e.g. , pagel 1992 ; garnett et al .
1995 ; kennicutt and garnett 1996 ; kobulnicky and skillman 1996 ) .
chemical evolution models are a useful tool to attempt to study the physical processes that might determine the chemical characteristics of the different galaxies .
detailed modelling can be found in studies of the galaxy ( e.g. , ferrini et al .
1992 ; tossi 1996 ; chiappini , matteucci and gratton 1997 and references therein ) .
these analytical models carefully treat stellar evolution .
however , they can not account for dynamical evolution , and certainly , not for hierarchical clustering . from a numerical point of view
, metallicity enrichment mechanisms have been implemented in hydrodynamical simulations in different ways .
the first attempts were done by larson ( 1975 , 1976 ) .
other implementations came after using diverse techniques such as chemodynamical models ( e.g. , burkert et al . 1992 ; samland et al .
1997 ) which describe in more detail the interstellar medium evolution . on the other hand
, this approach uses a very simple prescription for galaxy formation where the dark matter halo is either not considered at all or assumed to be static . however , there are strong evidences from observations of normal spirals that the dark matter is dynamically important within the luminous radius ( e.g. , bottema 1992 ; courteau , de jong & broeils 1996 ; rhee 1996 ) , and that the response of the dark matter to the presence of baryons affect their evolution ( blumenthal et al .
1996 ; tissera & domnguez - tenreiro 1998 ) and , as a consequence , the star formation process ( mihos & hernquist 1996 ; barnes & hernquist 1996 ; tissera 2000 ; navarro & steinmetz 2000 ) . a first approach to include chemical evolution in smooth particle hydrodynamical ( sph ) code is described by steinmetz and mller ( 1994 , 1995 ) followed by raiteri , villata and navarro ( 1996 ) .
these works run prepared - cosmological initial conditions where the formation and evolution of only one object was studied .
these models have shed light on some physical mechanisms that may control metallicity in galaxies , and have also shown that hierarchical clustering scenarios may form galactic objects that resemble milky way - type galaxies from a chemical point of view .
however , the sph models with chemical implementations hitherto published are restricted to the galaxy
. it would be very important for the study of the formation and evolution of galaxies to be able to analyse a variety of galactic objects with different evolutionary histories .
hydrodynamical cosmological simulations are more suited to tackle this problem since the non - linear evolution of the matter is naturally accounted for , and , physical processes can be more consistently implemented .
these models can provide coherent well - described environments for all objects and a complete record of their formation and evolution .
the drawback is the treatment of numerical resolution effects . in this paper
, we concentrate on the description of the chemical model that has been implemented in a cosmological hydrodynamical ap3 m .
first results on galaxy formation , global chemical properties and comparison with observations are reported .
the detailed analysis of the chemical properties of the ism and stellar population in galaxy - like objects such as abundance gradients , age - metallicity relations , etc . , are given by tissera et al .
( 2000 , hereafter paper ii ) .
this paper is organized as follows .
section 2 describes the star formation process and the metal enrichment implementation . in section 3
we assess the performance of the chemical model . in section 4
we present the first results on galaxy formation and chemical evolution .
section 5 summarizes the results .
we use a cosmological numerical code based on the sph technique described by tissera et al .
the sph algorithm has been coupled to the ap3 m gravitational code ( thomas and couchman 1992 ) , in order to follow the gravitational and hydrodynamical evolution of particles within a cosmological context . for the sake of simplicity , and as first step in the development of this chemical cosmological code
, simulations have been run considering that the gas cools down by using the approximation given by dalgarno and mccray ( 1972 ) .
in general , the modelling of star formation ( sf ) in sph simulations is based on the schmidt law and a series of hypothesis to select suitable sf regions .
however , how it is actually implemented within each particular code depends on the different authors ( e.g. , katz 1992 ; navarro and white 1994 ; tissera et al .
1997 ) . in this paper
, we describe a modified version of the sf implemented by tissera et al .
( 1997 ) in order to be able to track the transformation of gas into stars in more detail , within a given gas particle .
we include the sf algorithm as follows .
gas particles are eligible to form stars if they are cold ( @xmath3 ) and satisfy a density criterium : @xmath4 .
this density criterium arises by requiring the cooling time of a gas particle to be smaller than its dynamical time .
the critical temperature @xmath5 is taken as the minimum provided value by the cooling functions ( @xmath6k ) .
finally , gas particles have to be part of a collapsing region .
this requirement is imposed by selecting particles in a convergent flow ( @xmath7 ) .
when a gas particle satisfies all these conditions , star formation occurs according to the schmidt law , @xmath8 where @xmath9 is the star formation efficiency and @xmath10 is a characteristic time - scale assumed to be proportional to the dynamical time of the particle ( @xmath11 ) .
then , each stellar mass formed in a particle at a certain sf episode is given by @xmath12 , where @xmath13 is the time step of integration ( @xmath14 yr ) and @xmath15 a new sf efficiency .
note that , according to equation ( 1 ) , in order to go from density to mass , the volume of the new - born stars has to be assumed .
taking into account that the process we are modelling happens _ within a gas particle _ and , that according to observations , sf occurs normally in clusters , not in isolation , we assume that the volume occupied by the fraction of new - born stars is constant for all them .
thus , it can be absorbed in the new constant , @xmath15 . according to this sf scheme ,
in a given baryonic particle there could be several sf episodes that had occurred at different times .
the gaseous mass of the baryonic particle is reduced by @xmath16 until its gas reservoir is depleted . a minimum gas mass equal to 5 @xmath17 of the initial gas mass component is left over in each particle .
when a particle reaches this minimum mass , it is transformed completely into a star particle and hereafter behaves as collisionless matter .
each @xmath16 formed can be followed up in time .
the number of stars of a given mass within @xmath16 is estimated by assuming an initial mass function ( imf ) .
hence , a baryonic particle may be formed by a gaseous and stellar components in different proportions according to its history of evolution ( hereafter , hybrid particle ) .
the stellar component can be made up of different stellar populations with different ages and chemical properties , which have been formed at different sf episodes .
the chemical properties of each stellar population ( @xmath16 ) reflect the chemical state of the ism at the time of its formation .
the decoupling between gas and stars depends on the actual rate at which stars are formed in each baryonic particle , that , on its turn , depends on each evolutionary path .
hybrid particles may introduce some numerical artifacts since the stellar populations follow the gas evolution when they should probably not . however , this is not a trivial problem that can be easily resolved .
an improvement of the decoupling process is under work and will be presented in a separate paper .
metals are produced and ejected to the interstellar medium at the end of the life of stars .
most of chemical elements are ejected by type ii supernovae ( sneii ) except for the iron that is mainly produced by type i supernovae ( snei ) .
particles are assumed to be initially formed by hydrogen and helium in primordial abundances ( @xmath18 , @xmath19 ) .
the first generations of stars with primordial abundances immediately enrich the ism from where the new generations are born .
it has to be stressed that we are not including the effects of ( thermal or kinetic ) energy injection into the ism due to sn explosions . for the sake of simplicity we have splitted the treatment of the feedback problem into two stages .
firstly , we include chemical enrichment ( this paper ) , and in a second step , we will intend to develop an energy feedback model .
there have been several attempts to implement energy feedback in sph codes by either injecting thermal ( e.g. , katz 1992 ) or kinetic ( e.g. , navarro & white 1994 ; metzler & evrard 1994 ; navarro & steinmetz 2000 ; springel 2000 ) energies to the gas component due to sn explosions . however , these methods are still controversial leaving this problem as an open question for galaxy formation .
let us now describe the nucleosynthesis prescriptions adopted in this model : _ type ii sne _
: we assume that stars more massive than @xmath20 end their lives as type ii sne . in order to estimate their number ,
we adopt a salpeter imf with a lower and upper mass cut - off of @xmath21 and @xmath22 , respectively .
the imf adopted gives the total number of sneii formed in a certain range of stellar masses in a given @xmath16 at a certain time .
supernovae are thought to eject their whole metal production within a few @xmath23 yr .
in particular , we assume that their life - time is equal to the integration time - step of the simulations . for comparison ,
we resort to both portinari , chiosi and bressan ( 1998 , p98 ) and woosley and weaver ( 1995 , ww95 ) metal ejecta models and follow the evolution of different elements according to the information provided by the authors .
p98 give metal yields for stars up to @xmath24 , while ww95 assume that stars larger than @xmath25 end up their lives as black holes .
we consider the following elements according to each author : @xmath26 , for both p98 and ww95 , and also , @xmath27 , for ww95 .
_ type i sne _ : following matteucci and francois ( 1989 ) , we assume that supernovae ia originate from carbon deflagration in c - o white dwarfs in binary systems .
it is assumed that the masses of these binary systems are likely to be in the range @xmath28 .
the adopted nucleosynthesis prescriptions are taken from thielemann , nomoto and hashimoto ( 1993 ) .
type ib sne are assumed to be half the total number of type i sne and to produce only iron ( @xmath29 per explosion ) .
the main interest in including sni events in the models is that they contribute with a substantial amount of iron .
this element is fundamental to attempt to reproduce several observational results ( see also chiappini et al .
. the relative rate of sni to snii explosions has been observationally estimated in the solar neighbourhood and from extragalactic sources ( van der bergh 1991 ) .
these estimations suggest a range of possible values for the total sni rate respect to that of snii of @xmath30 .
binary systems evolve for a certain period of time ( @xmath31 ) during which mass is transfered from the secondary to the primary star until the chandrasekhar mass is exceeded and a explosion is triggered .
it is generally assumed that a fair fraction of snei will explode after a period of @xmath32 yr ( greggio 1996 ) .
it is straightforward to estimate when snei will explode and eject metals to the neighbouring region , since the formation time of each @xmath16 , in a given baryonic particle , is known .
the effects of varying both the snrii / snri relative rate ( hereafter , @xmath33 ) and @xmath31 are analysed and results are confronted with observations in the following sections . to sum up ,
the free parameters of our chemical model are : star formation efficiency ( @xmath15 ) , the initial mass function ( imf ) , its upper and lower mass cut - offs , the relative rate of different types of supernovae ( @xmath33 ) , the life - time of binary systems that end as snei ( @xmath31 ) and the yields .
regarding the imf , we are going to adopt a constant and unique one throughout this paper . how the chemical elements are distributed and mixed in the interstellar medium
is a complex problem which is far from being solved from a theoretical point of view ( e.g. , tenorio - tagle et al .
furthermore , numerical simulations do not reach enough resolution to properly treat this small - scale mixing process together with the process of galaxy assembly .
hence , the ejection and mixing of elements have to be modelled in heuristic way .
we use the sph technique to distribute the metals within the neighbouring sphere of the @xmath34 particle where a @xmath16 is formed . according to this technique
, the value of a certain parameter at the location of @xmath34 particle can be estimated by using the information storaged in its neighbourhood , which is determined by its @xmath35 neighbouring particles within its smoothing length ( @xmath36 ) . using this concept
, we can define the total mass ( @xmath37 ) of a given @xmath38 chemical element ejected by a @xmath34 particle due to snei or sneii , as @xmath39 where @xmath40 is the number of @xmath35 neighbours of the @xmath34 particle ( @xmath41 ) , @xmath42 and @xmath43 the gas mass and density of the @xmath35 particle , respectively , @xmath44 and @xmath45 , the symmetrized kernel and smoothing length , respectively . in this scheme ,
each @xmath35 neighbouring particle gets a contribution of each metal defined by @xmath46 the @xmath34 particle where the stellar mass has just formed is also included in this process .
the masses of h and he are proportionally decreased according to the metal mass received by each particle , so that its total mass is always conserved . the distribution of metals by using the sph technique allows to spread out metals and enrich gas particles that may have not experienced sf but are close enough to other sf regions , working as an effective mixing mechanism .
note that gas particles can be enriched by more than one neighbour at each time - step .
it has to be mentioned that numerical resolution would affect the mixing of metals since it also affects the determination of the neighbours . in section 3
, we will discuss this point in more detail .
in the present implementation , metal mixing occurs only tided to the sf process , in the sense that , only new - born elements are distributed .
one of the advantage of this implementation is that particles move according to the equations of gravitation and hydrodynamics , leaving behind the closed - box hypothesis , among others .
particles with different astrophysical and chemical properties are mixed , particularly during violent events such as interaction and mergers , so that reproducing observations becomes a more challenging task .
we carry out a series of simple tests to assess the effects of both , numerical resolution and the variation of the different parameters ( @xmath15 , @xmath33 , @xmath31 ) on the chemical properties of baryons .
we also intend to evaluate the dependence of the results on the two adopted nucleosynthesis models , p98 and ww95 . for this purpose
, we follow the evolution of a homogeneous gaseous sphere .
the sphere is represented by @xmath47 gas particles initially distributed following a density profile @xmath48 , in a radius of 625 kpc , without a dark matter halo .
this simple experiment gives a clear idea of how each parameter affects the chemical properties of baryons and have the advantage of a low computational cost .
numerical problems may affect both , the process of star formation and specially the distribution of metals .
with respect to the former , if dark matter haloes in galactic objects are numerically well resolved , the star formation process may not be strongly affected by the numerical resolution ( see discussion in domnguez - tenreiro , tissera and siz 1998 ) .
this is due to the fact that dark matter determines the potential well on to which the gas settles on .
if dark matter haloes are well resolved , baryons are forced to distribute adequately . in this way
, the gas density profiles are well reproduced and the sf can be correctly followed ( tissera 2000 ) .
nevertheless , the gas component has to be resolved , at least , by few hundred particles ( navarro and white 1994 ) . in our simulations ,
baryonic and dark matter components are represented by particles of equal mass .
therefore , dark matter haloes are resolved by a much larger number of particles than the baryonic component .
hence , the potential wells of the larger galactic haloes are very well defined ( see tissera and domnguez - tenreiro 1998 ) .
consequently , we restrict our global analysis to galactic objects resolved with more than 250 baryonic particles . on the other hand , the radius of the neighbouring sphere in which the metals produced by the stellar component are distributed is affected by the numerical resolution of the experiment .
consequently , this problem may have a non - negligible effect on the metal distribution and the chemical properties of gas and stars , and requires a more detailed analysis .
for this purpose , we have studied the evolution of two gaseous spheres without dark matter haloes .
each of them represents a system with total gaseous mass of @xmath49 and initial radius of 615 kpc .
one of the experiments ( e1 ) resolves the sphere with @xmath47 particles ( @xmath50 ) , while the second experiment ( e1h ) has higher numerical resolution : @xmath51 gas particles ( @xmath52 ) .
since these spheres have no dark matter haloes , taking into account the former discussion , it is reasonable to expect that their star formation histories will be different .
in fact , the sphere in e1h starts forming stars before its counterpart in e1 .
however , the effects that this earlier sf has on their metallicity is not dramatic , as it can be observed from fig.1 , where we have plotted the mean values of [ o / fe ] vs. [ fe / h ] for both experiments .
the greatest differences are obtained for the low metallicity stars ( [ fe / h ] @xmath53 ) due to the fact that star formation begins before in e1h and has a greater intensity than sf in e1 . as stars with higher metallicities are formed , this difference diminishes .
note that error bars in e1h comprise the distribution of e1 for [ fe / h ] @xmath54 ( except for the first point ) .
the main difference observed between these two experiments is that low numerical resolution produces a reduction in the dispersion of the distribution as can be seen from the error bars in fig.1 .
this fact can be understood by considering that , in the experiment with lower resolution , the neighbouring sphere ( @xmath55 ) comprises a larger volume , thus producing a more effective mixing of the metals in the gas .
conversely , when numerical resolution is increased , the new elements produced by a certain particle are distributed in a smaller volume .
this leads to a less efficient mix at the scale of the system , producing an increase in the metallicity dispersion .
we can not theoretically assess at what extent this dispersion has any physical meaning since the mixing mechanisms in galaxies are not yet well understood .
it might be possible that sn explosions spread metals in larger volumes than that provided by the sph technique .
this fact would produce a more uniform - metal distributions in the interstellar medium , as it actually occurs in the low numerical resolution run .
taking into account these experiments , we conclude that the main effect of using the sph technique to distribute metals is that the better resolved objects ( i.e. , those with higher number of baryonic particles that , in our simulations , implies larger masses ) would have a larger dispersion in their chemical properties than the smaller ones .
mean values of the quantities will be considered as adequate estimations .
we are now going to analyse the effects that the different chemical model parameters have on the properties of the stellar populations of the spheres .
as can be seen from table 1 , experiments e1 , e2 and e3 differ among each other in the life - time of binary systems that end up as snei .
the values considered are @xmath56 , @xmath57 and @xmath58 yr , respectively . the sf efficiency and @xmath33 used
are the same in these three experiments .
the ratio [ o / fe ] vs. [ fe / h ] clearly illustrates the effects of varying the parameters .
fig.2 shows [ o / fe ] vs. [ fe / h ] for these three experiments at different stages of evolution . as can be seen from this figure , as soon as the first stars are formed , the differences among them are very important .
the smaller @xmath31 used in e1 allows a rapid production of fe and , consequently , the number of low metallicity stars ( [ fe / h ] @xmath53 ) is considerably lower than in e2 and e3 , for all times .
the larger @xmath31 value adopted in e3 produces a dex difference in the ratio [ o / fe ] compared to the values in e1 and e2 for @xmath59yr . as time goes on
, all experiments reach solar values , although , the variation of @xmath31 among them leads to appreciably different evolutionary paths . for e1 and e2 , the [ o / fe ]
is above observed values for the solar neighbourhood , and the steepness of the relation is much abrupt for [ fe / h]@xmath60 .
( @xmath61 yr ) gives a more consistent relation for stars with [ o / fe ] @xmath62 . in this case , the final shape and range of abundances values are set as soon as the first stars formed which implies that even at early stages of evolution , this combination of sn parameters can produce stellar populations with metal contents up to solar . in fig.3 , we show the same relations for experiments e4 , e5 and e6 . in e4
we have changed the relative ratio @xmath33 from 2 ( in e1 ) to 4 .
since snrii is the fraction of stars with masses greater than @xmath20 in a certain time - step , given by the imf adopted , an increase in the ratio @xmath33 implies a decrease of snri .
the effect of the variation of this parameter gives the expected differences in the abundance tracks .
note that there is no change in the slope of the [ o / fe ] vs. [ fe / h ] relation , but the abundances in e4 do not get to solar values because there are not enough snei to produce the necessary amount of fe .
in experiment e5 we use a higher sf efficiency than that in e1 . in this case , stars are formed faster and the gas is enriched very efficiently so that the new - born stars immediately evolve to higher metallicity regions . as a consequence , the low - metallicity tail in the [ o / fe ] vs. [ fe / h ] relation vanishes .
note that the higher the @xmath15 value used , the larger the stellar masses ( @xmath16 ) formed , so that the number of sf events in a given baryonic particle diminishes .
this is the reason why the number of points in e5 is smaller when compared to the distribution in e1 .
the important aspect to remark is that , in e5 , the range of metallicity covered by the stellar populations is considerably smaller .
this implies that a very efficient sf process can enrich the medium to solar values very quickly so that most of stars would tend to be located in the high metallicity region .
the comparison between experiments e1 and e6 clearly illustrates how the simulation results depend on the two yield models , p98 and ww95 .
the features of the [ o / fe ] vs. [ fe / h ] relation in e6 can be appreciated in the bottom raw of fig.3 .
the main characteristic observed for this experiment , where p98 yields were used , is the lack of the low - metallicity tail ( [ fe / h ] @xmath53 ) , which can be appreciated in e1 . on the other hand , the low - metallicity stars in e6 ( [ fe / h ] @xmath63 ) , have almost constant values of [ o / fe ] , in contrast with the results of e1 , where a gradient is observed .
this difference arises because p98 yields produce more iron with respect to the other elements than ww95 , up to an order of magnitude more , in some cases .
so , stars are rapidly brought to lower abundances ratios with respect to iron . despite this fact , these relations are quite similar , with exception that ww95 seems to reach slightly higher metallicities values for the same model parameters at @xmath0 .
this fact implies that p98 yields for enriched stars produce , in general , more elements than those of ww95 .
the comparison of other chemical elements shows that the major difference is in the [ c / fe ] vs. [ fe / h ] relation . in fig.4
, we show this relation for experiments e1 and e6 as a function of time .
as clearly seen in the case of e6 , there is an excess of carbon in relation to iron for [ fe / h ] @xmath64 that produces a positive gradient . [ cols="^,^,^,^,^ " , ] observations of galactic and extragalactic hii regions and ob associations provide information about the chemical properties of the forming stellar population and the ism .
these observations give relations for some primary elements such as s , ne , and c , and for the so - called secondary ones , like nitrogen , as a function of the ratio o / h . in this section ,
we resort to these observational results to assess the global chemical properties of the stellar populations and ism of the simulated gals . concerning primary elements , galactic and extragalactic hii regions
show that the ratios ( s / o ) and ( ne / o ) do not depend on ( o / h ) . however , the dispersion in these ratios is quite large ( pagel 1997 ) .
the simple model actually predicts a similar behavior since in this scheme , the ratios between primary elements are constant .
however , this model fails to reproduce the observed correlation between ( c / o ) vs. ( o / h ) .
a similar problem is detected for the observed ( n / o ) vs. ( o / h ) that exhibits a steeper correlation than that predicted by this model .
note that most observations are obtained from hii regions which are assumed good tracers of the metallicity of the galactic ism .
we analyse the isms in our gals , which are determined by the gas particle properties within @xmath65 .
the global metallicity , @xmath66 , for each @xmath38 chemical element in the ism of a gal is defined as follows , @xmath67 where @xmath68 is the total number of @xmath35 gas particles within @xmath65 , @xmath69 is the mass of the @xmath70 element in the @xmath71 particle and @xmath72 is the total gas mass within @xmath65 .
an equivalent relation can be defined for the stellar population , @xmath73 . in fig.9
, we plot log(s / o ) for the ism in the gals in simulations s2 , s3 , s4 and s5 . as it can be seen from this figure ,
the mean values obtained in the simulations are in very good agreement with observations ( taken from pagel 1997 ) .
the average log(s / o ) for gals shows a large dispersion that compares well with the observed values .
the larger s / o ratios are obtained for the simulation with the highest sf efficiency ( simulation s3 ) . however , by changing sn parameters in order to diminish the effects of sni explosions , and mantaining the same sf parameter in the simulation ( simulation s4 ) , we obtain substantially smaller abundance ratios and dispersions . in fig.10 , we plot log(c / o ) versus log(o / h ) for gals in simulations s2 , s3 , s4 and s5 ( ww95 ) and simulation s1 ( p98 ) . we include observations of hii regions ( pagel 1997 ) .
a remarkable fact that can be seen from this figure is the large difference between the results of ww95 and p98 models .
gals in s1 , which are exactly equivalent to those in s2 , except for the nucleosynthesis model adopted , have ratios almost half an order of magnitude larger than their ww95 counterparts , and are out of the observational range .
conversely , gals in any of the ww95 runs have isms with abundances similar to observed ones .
however , they do not show the gradient reported from observations of galactic and extragalactic hii regions ( e.g. , garnett et al .
1995 ; kennicutt and garnett 1996 ; kobulnicky and skillman 1996 ) , although the simulated abundances extend only in the range @xmath74log(o / h ) @xmath75 .
the simulated mean values are lower than the observed ones but it has to be recalled that we are not including pne yields that are thought to be important contributors of c ( and n ) .
observations of ( n / o ) in hii regions show that this ratio increases with ( o / h ) ( e.g. , pagel 1992 ) .
a gradient in the secondary element is actually predicted by the simple model , although the steepness of the predicted relation is much larger than that actually observed in hii regions .
there are several hypothesis that may explain the origin of these differences ( pagel 1997 ) .
we estimate this relation for gals in ww95 runs .
as shown in fig.11 , the simulated isms have abundances that are in good agreement with observations .
however , we only find gals with @xmath76 log(o / h ) @xmath77 , while the observed gradient in hii regions is @xmath76 log(o / h ) @xmath78 . we do not find gals with average ism metallicity corresponding to solar abundances , neither extremely low metallicity izw18 type objects .
it should be considered , however , that only massive gals are analysed in the simulations .
another important observational correlation is that between @xmath76 log(o / h ) and the gas fraction estimated for irregular and blue compact galaxies ( e.g. , axon et al .
1998 ) . this relation can be used to assess the metal content of the isms in relation to the left - over gas mass of the corresponding gals . in fig.12 , we plot @xmath76 log(o / h ) versus log @xmath79 for the ism in the simulated gals ( @xmath80 and @xmath72 are the total baryonic and gaseous mass within @xmath65 , respectively ) .
observations taken from pagel ( 1997 ) have been included for comparison .
the estimated abundance ratios in the models are consistent with observations .
it is notable that the observed relation is well reproduced in the sense that objects with a lower gas fraction tend to have larger metallicities without the need to introduce other physical mechanisms such as sn energy injection effects to the ism .
the simple model predicts a decrease of the metallicity with the gas richness of the system but it fails to correctly reproduce the observational relation .
we have implemented a chemical model in a sph cosmological code .
first results are reported together with the assessment of the effects of model parameters and numerical resolution .
it is found that the major problem introduced by a low numerical resolution is the artificial smoothing of the chemical properties of the objects .
however if a minimum number of baryonic particles is imposed , then average values can be considered reliable .
the mean metallicity is found nearly independent of the total stellar mass of the system indicating that the effects of mergers , interactions and gas infall on the mass distribution and metal mixing are very significant .
we find correlations between the stellar mass @xmath81 and the virial and optical circular velocities ( @xmath82 , @xmath83 ) .
however , the correlation with @xmath83 is significantly tighter , consistent with the tully - fisher relation
. nevertheless , the global metallicity of gals , @xmath84 shows no dependence on @xmath83 , in spite of the fact that these gals satisfy the tully - fisher relation and shows a strong correlation between the global metallicity of their stars and gas components with the left - over gas fraction of the systems .
we only find a significant correlation between the global gals metallicity of both stars and gas with the left - over gas fraction of the systems .
an analysis of the gals assuming they have behaved according to the one - zone simple model show important differences with the results of the numerical simulations .
the slope of @xmath85 versus @xmath86 and the systematic increase of stellar metallicity as a function of the stellar mass formed in the simple model contrasts with the behaviour of the numerical simulations .
we find that the simple model overestimates the metallicity of the galactic gaseous components if it is constrained to match the stellar chemical content .
according to the simple model , the global metallicity of a system continuously increases with the stellar mass in disagreement with the results found for the simulated gals that show higher metallicities for intermediate mass objects .
we find that the star formation efficiency , sn model parameters and the nucleosynthesis yields significantly affect the chemical properties of the gals . in those models where the sf process is gradual the mean stellar and gaseous metallicities are similar .
the observed abundance ratios for primary and secondary elements in galactic and extragalactic hii regions are naturally obtained .
however , the ranges of the simulated global metallicities are smaller than the observed ones .
also , the relation between the ( o / h ) and gas fraction obtained in the numerical simulations is consistent with the observational results .
the suitable agreement between the models and the observations for these relations suggests that hierarchical clustering scenarios are able to reproduce the chemical properties of galaxies .
this is a very encouraging fact taken into account that galactic objects formed by the accretion and mergers of substructures .
violent events are common and ubiquitous , and can affect the dynamical evolution of the matter , regulating the sf and chemical evolution in galactic objects . we intend to improve this chemical model in the future by allowing the gaseous component to cool according to its metallicity , and a further study of the mixing process of chemical element in the ism .
it is also under study a more efficient decoupling mechanism between stars and gas that could shorten the period of hybrid state of a baryonic particle .
finally , energy feedback remains an open question that we hope to address in a future work .
we thank s. woosley for kindly providing the yield tables and l. portinari for clarifying some aspects of their nucleosynthesis model .
we acknowledge the careful reading and useful suggestions of the anonymous referee .
we also thank m. abadi for useful discussions .
p.tissera is grateful to c. chiappini for introductory discussion in chemical evolution during the 1999 aspen summer workshop .
the authors thank the hospitality of the observatorio astronmico centroamericano de suyapa while finishing the writing of this paper .
p.tissera is grateful to the observatorio astronmico de crdoba and the research group iate for allowing the use of their computational facilities .
m. mosconi thanks iafe for their hospitality during the preparation of this work .
this work has been partially supported by conicet , conicor , secyt and fundacion antorchas .
axon d. j. , staveley - smith l. , fosbury r. a. e. , danziger j. , boksenberg a. , davies r. d. , 1988 , mnras , 231 , 1077 barnes j. , hernquist l. , 1996 , apj , 471 , 115 bottema a. h. , 1992 , a & a , 328 , 517 burkert a. , truran j. w. , hensler g. , 1992 , apj , 391 , 651 blumenthal g. r. , faber s. m. , flores r. a. , primack j. p. , 1986 apj , 301 , 27 chiappini c. , matteucci f. , gratton r. , 1997 , apj , 477 , 765 cora s. a. , tissera p. b. , lambas d. g. , mosconi m. b. , 2000 , proceeding of galaxy disks and disk galaxies rome , 12 - 16 june , 2000 .
cowie l. l. , songaila a. , hu e. m. , cohen j. g. , 1996 , apj , 112 , 839 courteau s. , de jong r. s. , broeils , a. h. , 1996 , apj , 457 , l73 dalgarno a. , mccray r.a . , 1972 , ar&aa , 10 , 375 dekel a. , silk j. , 1986 , apj , 303 , 39 domnguez - tenreiro , r. , tissera p. , siz a. , 1998 , apj , 508 , l123 edvardsson b. , andersen j. , gustafsson b. , lambert d. l. , nissen p. e. tomkin , j. , 1993 , a&as , 274 , 101 ellis r. s. , colless m. , broadhurst t. , heyl j. , glazebrook k. , 1996 , mn , 280 , 235 ferrini f. , matteucci f. , pardi c. , peuco u. , 1992 , apj , 387 , 138 garnett d. r. , sillman e. d. , dufour r. j. , peimbert m. , torres - peimbert s. , shields g. a. , terlevich r. j. , terlevich e. , 1995 , apj , 443 , 64 gratton r. , cannetta e. , matteucci f. , sneden c. , 1996 , morrison h. , sanajedini a. , eds . , formation of the galactic halo , inside out , vol.92 , p.307 greggio l. , 1996 , kunth d. , guiderdoni b. , heydari - malayeri m. , and thuan t. , eds . , the 11th iap astrophysics meeting : the interplay between massive star formation , the ism and galaxy evolution , p.89 katz n. 1992 , apj , 391 , 502 kennicutt r. , jr .
, garnett d. r. , 1996 , apj , 456 , 504 kobulnicky h. a. , skillman e. d. , 1996 , apj , 471 , 211 kunth d. , ostlin g. , 1999 . to appear in ara&a larson r. , 1975 , mnras , 173 , 671 larson r. , 1976 , mnras , 176 , 31 lilly s. j. , tiesse l. , hammer f. , crampton d. , le fevre o. , 1995 , apj , 441,18 lu l. , sargent l. w. , barlow t. , churchill c. , vogt s. , 1996 , apjs , 107 , 475 madau m. p. , 1995 ,
apj , 441 , 18 madau p. , ferguson h. , dickinson m. , giavalisco m. , steidel c. , fruchter a. , 1996 , mras , 283 , 1388 matteucci f. , francois p. , 1989 ,
mnras , 239 , 885 metzler c. a. , evrard a. e. , 1994 , apj , 437 , 564 mihos j. c. , hernquist l. , 1996 , apj , 464 , 641 navarro j.f . , white s.d.m .
1994 , mnras , 267 , 401 navarro j. f. , steinmetz m. , 2000 , astro - ph/0003384 pagel b. e. j. , 1992 . , alloin d. , stansiska g. , eds . , the feedback of chemical evolution on the stellar content of galaxies , p.87 pagel b. e. j. , 1997 , nucleosynthesis and chemical evolution of galaxies , cambridge university press , united kingdom pettini m. , smith l. , king d. , hunstead r. , 1997 , apj , 998 , 665 portinari l. , chiosi c. , bressan a. , 1998 , a&as , 334 , 505 ( p98 ) raiteri c. m. , villata m. , navarro j. f. , 1996 , a&as , 315 , 105 renzini a. , 1998 , dodorico , fontana a. , giallongo e. , eds . , asp conference series , the young universe : galaxy formation and evolution at intermediate and high redshift , vol.146 , p.298 rhee m. h. , 1996 , ph.d .
thesis , university of groningen rocha - pinto h. j. , maciel w. j. , flynn c. , 2000 , a & a , submitted ( astro - ph/0001383 ) samland , m. , hensler , g. , theis , ch . , 1997 , apj , 476 , 544 schmidt m. , 1963 , apj , 137 , 758 springel , v. , 2000 , mnras , 312 , 859 steidel c. c. , hamilton d. , 1992 , aj , 104 , 941 steidel c. c. , adelberger k. l. , giavalisco m. , dickinson m. , pettini m. , 1998 , apj , submitted ( astro - ph/9811399 ) steinmetz , m. , & mller , e. 1994 , a&a , 281 , l97 steinmetz , m. , & mller , e. 1995 , mnras , 276 , 549 tenorio - tagle g. , sergey a. s. , kunth d. , terlevich e. , roberto terlevich r. , 1999 , mnras , in press thielemann f. k. , nomoto k. , hashimoto m. , 1993 , prantzos n. , vangoni - flam e. , cass n. , eds . , origin and evolution of the elements , p.299 thomas p. a. couchman h. m. p. , 1992 ,
mnras , 257 , 11 tinsley b. m. , 1980 , fundamental of cosmic physics , vol.5 , p.287 tissera p. b. , 2000 , apj , 534 , 636 tissera p. b. , domnguez - tenreiro r. , 1998 , mnras , 297 , 177 tissera p.b .
, lambas d.g . ,
abadi m.g .
, 1997 , mnras , 286 , 384 tissera p. b. , mosconi m. b. , cora s. a. , lambas d. g. , 2000 , in preparation tosi m. , 1996 , from stars to galaxies : the impact of stellar physics on galaxy evolution , asp , conf.ser . , vol.198 , p.229 van den bergh s. , 1991 , woosley s. e. , ed . , the tenth santa cruz workshop in astronomy and astrophysics : supernovae , p.711 van den bergh s. , 1962 , a&as , 67 , 486 white s.d.m .
, frenk c.s .
1991 , apj , 379 , 25 white s. d. m. , 1981 , mnras , 191 , 1 woosley s. e. , weaver t. a. , 1995 , apjs , 101 , 181
figure 1 : comparison of the [ o / fe ] versus [ fe / h ] relation obtained from the low ( solid line ) and high resolution tests ( dashed lines ) .
figure 2 : [ o / fe ] vs. [ fe / h ] for experiments e1 , e2 and e3 where the life - time of binary systems has been varied : @xmath56 yr , @xmath87 yr and @xmath88 yr , respectively .
time @xmath89 is given in units of @xmath90 yr .
figure 3 : [ o / fe ] vs. [ fe / h ] for experiments e4 , e5 and e6 where the relative ratio of snii / sni ( @xmath33 ) , the star formation efficiency ( @xmath15 ) and the nucleosynthesis yields have been changed with respect to e1 .
time @xmath89 is given in units of @xmath90 yr .
figure 5 : global metallicities of the stellar population and gas component in gals as a function of their gas fraction , @xmath91 , ( a , c ) , and their total stellar mass , @xmath81 , ( b , d ) in simulations s2 ( open pentagons ) , s6 ( filled squares ) and s7 ( open squares ) .
lines represent the relations given by the simple model for @xmath92 ( dotted - dashed lines ) , @xmath93 ( solid lines ) and @xmath94 ( dashed lines ) .
@xmath81 is given in units of @xmath95 .
figure 6 : the total stellar mass of gals ( @xmath81 ) as a function of a ) the virial velocity ( @xmath82 ) and b ) the optical velocity ( @xmath83 ) for the same gals shown in fig.5 .
velocities are given in km@xmath96 and @xmath81 is given in units of @xmath95 .
figure 7 : global metallicities of the stellar populations in gals ( @xmath85 ) in simulation s1 ( p98 ) as a function of a ) the gas fraction ( @xmath91 ) , and b ) the total stellar mass ( @xmath81 ) of gals .
@xmath81 is given in units of @xmath97 .
lines represent the relations given by the simple model as in fig.5 .
figure 8 : global metallicities of the stellar populations ( @xmath85 ) as a function of a ) the gas fraction ( @xmath91 ) , and b ) the total stellar mass ( @xmath81 ) of gals in simulations s3 ( filled pentagons ) , s4 ( open triangles ) and s5 ( open stars ) .
@xmath81 is given in units of @xmath97 .
figure 10 : ( c / o ) versus ( o / h ) for the ism of gals in simulations s2 , s3 , s4 and s5 ( see fig.5 and 8 for feature code ) .
gals in s1 ( filled triangles ) have been included .
observations of hii regions have been included ( small filled circles ) .
figure 12 : logarithm of the oxygen abundances versus the logarithm of their gas fraction for the same gals shown in fig.9 .
observations of hii regions in blue compact and irregular galaxies from pagel ( 1997 ) have been included ( small filled circles ) . | we develop a model to implement metal enrichment in a cosmological context based on the hydrodynamical ap3msph code described by tissera , lambas and abadi ( 1997 ) .
the star formation model is based on the schmidt law and has been modified in order to describe the transformation of gas into stars in more detail .
the enrichment of the interstellar medium due to supernovae i and ii explosions is taken into account by assuming a salpeter initial mass function and different nucleosynthesis models .
the different chemical elements are mixed within the gaseous medium according to the smooth particle hydrodynamics technique .
gas particles can be enriched by different neighbouring particles at the same time .
we present tests of the code that assess the effects of resolution and model parameters on the results .
we show that the main effect of low numerical resolution is to produce a more effective mixing of elements , resulting in abundance relations with less dispersion .
we have performed cosmological simulations in a standard cold dark matter scenario and we present results of the analysis of the star formation and chemical properties of the interstellar medium and stellar population of the simulated galactic objects .
we show that these systems reproduce abundance ratios for primary and secondary elements of the interstellar medium , and the correlation between the ( o / h ) abundance and the gas fraction of galaxies .
we find that star formation efficiency , the relative rate of supernovae ii to supernovae i and life - time of binary systems as well as the stellar nucleosynthesis model adopted affect the chemical properties of baryons .
we have compared the results of the simulations with an implementation of the one - zone simple model , finding significant differences in the global metallicities of the stars and gas as well as their correlations with dynamical parameters of the systems .
the numerical simulations performed provide a detailed description of the chemical properties of galactic objects formed in hierarchical clustering scenarios and proved to be useful tools to deepen our understanding of galaxy formation and evolution .
= = = = = = = = # 1 # 1 # 1 # 1 @mathgroup@group @mathgroup@normal@groupeurmn @mathgroup@bold@groupeurbn @mathgroup@group @mathgroup@normal@groupmsamn @mathgroup@bold@groupmsamn = `` 019 = ' ' 016 = `` 040 = ' ' 336 = " 33e = = = = = = = = # 1 # 1 # 1 # 1 = = = = = = = = [ firstpage ] cosmology : theory - galaxies : formation - galaxies : evolution - galaxies : abundances . |
in this chapter , we give a brief sketch about the field of complexity science : the questions that led to the emergence of this new approach to understand real - world structures and their dynamics .
we discuss in particular the emergence of the new field of research - _ network science _ , and the various network models , their characteristic properties and their time - dependent dynamic behaviour . in relation to the theme of the thesis ,
we outline the significance of complex networks approach to understand public transportation infrastructures .
further , we review the literature in the area of structurally constrained network of bus routes in urban cities , which forms the core of the thesis .
we end this section by elaborating on dynamical processes , like information percolation and phase - transition in networks in general , and epidemic spreading in transportation networks in particular .
nonlinearity is the essence of reality .
the emergence of the field of complexity science is often attributed to the fact that real - world processes are indeterministic due to the presence of numerous variables and their nonlinear combinations .
real - world systems are also hard to comprehend because the agents constituting those systems show non - trivial interactions between them . in the context of these systems
, it is always observed that the system as a whole is greater than its constituting parts @xcite .
therefore , these systems are complex not only because of their scale but also because of their functionality .
the growing interest to understand the underlying machinery of these _ complex systems _ has given rise to numerous mathematical and simulation techniques .
some of the widely used techniques include agent - based modelling , time series analysis , ant - colony optimization , cellular automata , nonlinear differential equations , information theory and network theory @xcite . amongst these existing mathematical methods ,
network theory in particular has been immensely successful in describing real world systems and processes in the recent times .
the underlying mathematical concepts of network theory or commonly , network science , is the theory of graphs .
one of the oldest examples of using graph theory to analyze real - world problem dates back to as early as 1736 , when the resolution to the famous knigsberg bridge problem by euler actually laid the foundations of graph theory ( see figure 1 ) .
the abstract approach by euler made the geographical intricacies present in the problem seem totally irrelevant .
his although simple , yet abstract formulation of the problem in terms of vertices ( nodes ) and edges ( links ) provided an elegant approach to model real - world structures as graphs or networks .
the first physical application of graph - theoretical ideas was discovered by gustav kirchhoff in 1845 for calculating the voltage and current in electric circuits . since then
, the use of graph theory as a modelling technique has found innumerable applications across diverse disciplines .
one of the very important properties that these network models capture is the way different agents interact with each other in a connected system , which gives rise to non - trivial ( emergent ) properties in the system .
be it social ties or technological interactions , socio - economic infrastructures or interaction between biological entities , almost every thing can be modelled as a graph containing nodes and edges @xcite .
complex network science today has established itself as a mainstream field of research in the physical sciences .
although , the fundamental breakthroughs in network science are often attributed to statistical physicists and mathematicians , the origin of this field of study surprisingly lies in the social sciences .
two spectacular examples of studies that gave birth to network science as we know it today are that of milgram s ` small - world experiment ' of 1967 and granovetter s theory on the spread of information in social networks in 1973 @xcite . the results from milgram s experiment helped us to understand how closely connected we are through our social ties whereas granovetter s theory stressed on the importance of the individual connection that a node shares in the network .
both the observations led to an increased surge of interest in this new field of study by mathematicians and social scientists . with the advancement in information technology that enabled availability of large amounts of real - world data and improved computational resources ,
physicists eventually came into the picture in the early 1990 s .
it was during this time that the structure and topology of large - scale complex networks were being studied and fundamental questions like the nature of connections and in general , the statistical properties of the network s building blocks or the nodes were being asked .
the field which initially branched out from the mathematical theory of graphs suddenly found its applications and similarities to statistical physics .
no longer were the properties of the single node a question of interest .
the group behaviour of the nodes - how they connect to each other - became a significant non - trivial question of academic pursuit .
the field of network science has emerged out of the mathematical theory of graphs .
therefore , the underlying mathematical structure of the networks and its various characteristic properties use the language of graph theory for theoretical and computational purposes .
before we go into the details of the theoretical questions of interests ( those we pointed out earlier ) , it is necessary to first mathematically describe a network and discuss certain characteristic features of these graphs , which are of interest in the particular context of the thesis .
we will also observe later that these are the properties which help in differentiating between various network models .
adjacency matrix is diagonally - symmetric.,scaledwidth=80.0% ] we define a graph , @xmath0 where the set @xmath1 where each @xmath2 is a node , and the set @xmath3 where each @xmath4 is a link connecting the node pair @xmath5 .
the set of nodes belong to the n - dimensional euclidean space , @xmath6 , and the set of links form the cartesian product over @xmath6 .
a graph can be directed or undirected .
if a graph is directed , then @xmath7 , whereas if a graph is undirected , we have @xmath8 . another important mathematical structure associated with graphs
is the adjacency matrix , @xmath9 .
the adjacency matrix is a representation of the network as a @xmath10 square matrix .
each element , @xmath11 either takes a value of @xmath12 or @xmath13 when the graph is unweighted . when @xmath14 the link @xmath4 does not exist , whereas @xmath15 or @xmath13 signifies the directionality of the link ( with @xmath16 in first case and @xmath17 in the latter ) . if the graph in question is a weighted graph , then @xmath18 can take any real value .
the weights in the graph can represent diverse quantities , such as strength of connections ( social network ) , frequency of interactions ( call networks ) , travel times ( transportation networks ) , distances ( road networks ) , etc .
for an undirected graph , the directionality of the links does not play any role ( @xmath8 ) , and the adjacency matrix takes on a diagonally - symmetric form .
also , note that @xmath19 represents self - loops in the network ( see figure 2 ) @xcite .
an important measure is the characteristic path length , @xmath20 which is defined as the average number of nodes crossed along the shortest paths for all possible pairs of network nodes .
the average distance from a certain vertex to every other vertex is given by @xmath21 .
then , @xmath20 is calculated by taking the median of all the calculated @xmath22 @xmath23 .
the characteristic path length helps in identifying whether a small - world phenomenon exists in the network or not .
the significance of the small - world property is two - fold : one , it signifies very low characteristic path length , which simply means that given any pair of randomly chosen nodes , the number of ` hops ' needed to reach one from the other will be very low as compared to the network size ; second , it also characterizes the change in the metric , @xmath20 as the network size changes ( the number of nodes present in a network is called its size ) @xcite .
milgram s experiment showed that on an average any randomly chosen pair of individuals are separated by six acquaintances ( hence , the popular term : _ six - degrees of separation _ ) .
another important network metric is the the clustering coefficient which measures the extent to which nodes tend to cluster in a network .
the concept of clustering coefficient originated mainly from studying social networks , where a pair of individuals are connected by a link if they are friends .
in social networks , close friends tend to have close - knit communities where each individual in the group knows every other individual . in figure 3
, we can observe that the maximum number of triangles possible for the blue node is three .
in the first case , since every node is connected to every other node , the magnitude of clustering coefficient is one . whereas , in the second and the last case when the dotted ( red ) links are removed , the possible number of triangles gets reduced to one and subsequently to zero .
hence , the magnitude of clustering coefficient in the second case is one - third and zero in the latter .
the above example denotes the local clustering coefficient , which is in reference to a particular node in the network .
the local clustering coefficient is given by : @xmath24 in the above expression , @xmath25 are the neighbours of the node @xmath2 and the neighbourhood , @xmath26 , for a node , @xmath2 is defined as the set of its immediately connected neighbours , as @xmath27 . for the complete network ,
watts and strogatz defined a global clustering coefficient , @xmath28 @xcite .
hence , the clustering coefficient measures the extent to which nodes tend to form close - knit groups in a network .
a network with a small characteristic path length and a high clustering coefficient is called a small - world network . ) in an undirected graph . in the first case , @xmath29 as every node is connected to every other node . in the second case , the removal of two links ( red , dotted ) causes the number of triangles for the blue node to reduce to @xmath30 , hence @xmath31 . in the last case ,
an additional removal of link causes the number of triangles for the blue node to reduce to zero , therefore @xmath32.,scaledwidth=80.0% ] although the above metrics are crucial in identifying small - world phenomenon in a network , they do not help us in understanding the network topology .
understanding the network topology holds the key to several interesting and non - trivial questions . in order
to answer questions about network robustness or node centrality , we need to look at the pattern by which the nodes are connected in the network .
before we go into the details of various degree - distribution patterns , we define the quantity - degree ( of a node ) in a network .
the degree of a node is the number of neighbours to which it is directly connected to . in an undirected graph ,
the degree of a node is mathematically expressed as @xmath33 . in a directed graph
, the degree of a node will depend upon the directionality of the links which either emerge _ out _ of the node or converge _ into _ the node .
therefore , in a directed graph , a node will have an out - degree ( @xmath34 ) and an in - degree ( @xmath35 ) , with @xmath36 , where the index @xmath37 runs over all the positive @xmath18 and vice - versa .
when the network as a whole is studied , the significance of an individual node takes the backseat ( unless we are interested in node - specific properties ) , and the way the degree of every single node is distributed throughout the network becomes a question of extreme interest .
we define degree - distribution ( @xmath38 ) of a network as the probability that a node , @xmath2 has a degree of atleast @xmath25 .
the notion of degree - distribution holds a central role in network science .
networks that follow a similar degree - distribution law tend to show similar network characteristics .
therefore , the degree - distribution function can be used as a signature to differentiate between different network classes . among all the possible degree - distribution laws ,
the ones that are most frequently encountered are the poisson , exponential and power - law patterns .
the poissonion form of degree - distribution is given by : @xmath39 similarly , the exponential degree - distribution is given by : @xmath40 and finally , the power - law degree distribution is given by : @xmath41 in the above expressions , @xmath42 denotes the average node - degree , @xmath43 and @xmath44 , the exponential and power - law degree exponents .
although all the distribution functions @xmath45 ( in eqn . 2 , 3 , 4 ) decay for large magnitudes of @xmath46 , a special feature of the distributions in eqn . 2 and 3
is that they contain a typical scale .
it is either the location of the maximum for the poisson distribution , or the characteristic decay length for the exponential .
on the contrary , the power - law distribution in eqn .
4 does not contain such a scale .
networks with a power - law degree - distribution are therefore called scale - free networks . when the degree - exponent , @xmath47 , the average node degree diverges and when @xmath48 , the standard deviation of the degree diverges @xcite .
2 , 3 and 4 are probability density functions , therefore : @xmath49 for the continuous case and @xmath50 for the discrete case .
the constant , @xmath51 is known as the normalization constant . clearly , the distribution in eqn .
4 diverges as @xmath52 so eqn .
4 can not hold for all @xmath53 , _
i.e. _ , there must be some lower - bound to the power - law behavior @xcite .
we will denote this bound by @xmath54 .
then , provided @xmath55 , it is straightforward to calculate the normalizing constant , and we find that : @xmath56 for the discrete case , the distribution in eqn .
4 diverges when @xmath57 , so there must be a lower bound @xmath58 on the power - law behavior . on calculating the normalizing constant , we find that : @xmath59 where the hurwitz zeta function , @xmath60 is given by : @xmath61 the poisson distribution is strongly peaked about the mean @xmath42 , and has a tail that decays very rapidly as @xmath62 .
this rapid decay is completely different from the heavy - tailed degree - distribution that is observed in many real - world complex networks .
real - world networks tend to show a degree heterogeneity , _
i.e. _ , very a small fraction of nodes tend to hold majority of the connections in the network which are called as ` hubs ' .
instead of having a normally distributed degree in the network about the mean degree ( poissonian distribution function ) , real - world networks show tails which are heavily skewed towards the right . also , it is hard to find real - world networks that show a perfect power - law or an exponential distribution , therefore majority of the networks tend to show a combination of them , such as power - law degree - distribution with exponential cut - offs or a combination of different power exponents .
whatever the case may be , it is important that we find proper explanations to these non - trivial characteristics of the real - world complex networks . in the following sections , we will elaborate on the important network models that will answer some of the interesting features of the networks that we saw earlier .
also , these are not the exhaustive list of network properties to be analyzed .
we will see in the later sections that there are numerous other network metrics that describe both the global network properties and local nodal characteristics . in this section ,
we discuss the three prominent network models .
starting with the random graph model by paul erds ( see figure 4 ) and alfred rnyi in 1959 that laid the fundamental ideas of network science , we look into the small - world properties present in a network by duncan watts and steven strogatz in 1998 . finally , in order to understand the presence of heavy tails in degree - distribution patterns of real - world networks , we look into the scale - free network model proposed by albert - lszl barabsi and rka albert also in 1998 .
the erds - rnyi ( er ) model is either of the two closely related network models in graph theory for generating random graphs @xcite . in the model introduced by erds and rnyi , @xmath63 , the number of vertices ( @xmath64 ) and the number of edges ( @xmath65 ) are fixed , implying that all possible graph combinations are equally likely .
whereas in the model introduced by gilbert , @xmath66 , each edge has a fixed probability @xmath67 , of being present or absent , independently of the other edges . therefore , in the @xmath66 model , a graph is constructed by connecting nodes randomly with each edge having a fixed probability @xmath67 independent from every other edge .
therefore , all graphs with @xmath64 nodes and @xmath65 links have equal probability of @xmath68 .
as the parameter @xmath67 in this model increases from @xmath69 , the model becomes more likely to include graphs with more edges and less likely to include graphs with fewer edges .
in particular , when @xmath70 all @xmath71 graphs on @xmath64 vertices are chosen with equal probability .
one of the interesting features of the er random graph is the degree - distribution patterns of the nodes which follow the binomial form , @xmath72 and approaches the poissonian form as in eqn . 2 when @xmath73 and @xmath74 constant .
some of the important features about random graphs that erds and rnyi pointed out are @xcite : * if @xmath75 , then a graph in @xmath66 will almost surely have no connected components of size larger than @xmath76 * if @xmath77 , then a graph in @xmath66 will almost surely have a largest component whose size is of order @xmath78 * if @xmath79 , then a graph in @xmath66 will almost surely have a unique giant component containing a positive fraction of the vertices and no other component will contain more than @xmath76 vertices although these models ( @xmath63 and @xmath80 ) can be used in the probabilistic method to prove the existence of graphs satisfying various properties , or to provide a rigorous definition of what it means for a property to hold for almost all graphs , they hardly capture the essence of the real - world network properties . , with increasing wiring probability @xmath81 and @xmath82 ( left to right ) and @xmath83 . for @xmath84 the number of links present @xmath85 .
similarly for @xmath86 and @xmath87 . observe how a ten - fold increase in probability causes the number of links to increase roughly by ten times.,scaledwidth=100.0% ] the er model provides a good mathematical description of graphs and the various properties associated with them .
although there may be instances when er graphs show exceptionally small characteristic path lengths , they however do not show two important properties observed in many real - world networks : * presence of clusters or triadic closures @xmath88 er graphs have low clustering coefficient * presence of exceptionally high degree nodes or hubs @xmath88 degree - distributions in er graphs converge to a poisson form rather than a power - law pattern in order to address the first of the above two limitations , duncan watts and steven strogatz devised a simple model that kept the characteristic path length small ( similar to a random graph ) but added the additional attribute of clustering .
the model interpolated between a regular ring lattice and an er random graph , and was able to explain the small - world phenomenon to some extent @xcite . given a fixed number of nodes @xmath89 , mean degree @xmath42 ( assumed to be an even integer ) , and a special parameter @xmath67 , satisfying @xmath90 and @xmath91 , the algorithm constructs an undirected graph with @xmath89 nodes and @xmath92 edges in the following way ( see figure 6 ) : * construct a regular ring lattice , a graph with @xmath89 nodes each connected to @xmath42 neighbors , @xmath93 on either side , _
i.e. _ , if the nodes are labeled @xmath94 , there is an edge @xmath95 if and only if @xmath96 * for every node @xmath97 take every edge @xmath95 with @xmath98 , and rewire it with probability @xmath67 .
rewiring is done by replacing @xmath95 with @xmath99 where @xmath65 is chosen with uniform probability from all possible values that avoid self - loops @xmath100 and link duplication ( there is no edge @xmath101 with @xmath102 at this point in the algorithm ) plays a crucial role .
when @xmath103 , we have a regular ring lattice , whereas at @xmath104 , we have a random graph , @xmath80 with @xmath105 and @xmath106 @xcite.,scaledwidth=75.0% ] we list out some network characteristics of the watts - strogatz small - world network model ( ws ) .
the degree distribution in the case of the ring lattice is the dirac delta function centered at @xmath42 , which assumes the poisson form in the limiting case of @xmath107 ( eqn .
2 ) , similar to the classical er random graphs @xcite .
the degree distribution for @xmath108 can be written as : @xmath109 where @xmath25 is the degree of the @xmath110 node , @xmath111 , and @xmath112 .
the shape of the degree - distribution curve is similar to that of a random graph and has a pronounced peak at @xmath113 , which decays exponentially for large @xmath114 .
the topology of the network is relatively homogeneous , and all nodes have more or less the same degree .
the additional feature that the ws model exhibits is the property of forming triadic closures or clusters .
the clustering coefficient is given as the ratio of the number of triangles present among the nodes to the total number of connected triplets among them ( see figure 3 ) . in terms of @xmath67 , @xmath115 where , @xmath116 is the clustering coefficient of a regular ring lattice given by @xmath117 and @xmath118 as @xmath119 .
the ws model preserves the short characteristic path lengths exhibited by random graphs . for a regular ring lattice ,
the characteristic path length is given by @xmath120 , which in the limiting case @xmath121 approaches to @xmath122 or @xmath123 for a random graph .
thus , the number of hops required to visit all the nodes in a network scales as the logarithm of the network size , which still remains a very low number even though the network size is large ( or grows with time ) .
the small - world property has been reported in various real - world networks such as electric power grids , www , internet , social - networks , protein - yeast ( metabolite ) interaction networks , citation networks , movie - actors collaboration networks @xcite .
although these networks show small characteristic path lengths and high clustering , they also exhibit a heavy - tailed degree - distribution that can not be explained either by the er or the ws models . in order to understand the mechanism of growth and evolution of real - world networks ,
we analyze the barabsi - albert scale - free model in the following section .
albert - lszl barabsi , along with his doctoral student rka albert published two important articles in the late 90 s @xcite . in both the articles , they pointed out the peculiarity in the patterns of degree - distribution in real - world networks .
they observed that the degree - distribution laws did not follow the poissonian form nor the usually expected normal form .
the density functions decayed slower than exponentials and log - normals , and showed a heavy - tail towards their end .
interestingly , their studies also confirmed the small - world property in these real - world datasets . in order to understand the growth and evolution of real - world networks exhibiting scale - free nature , barabsi and albert proposed an algorithm ( ba model ) that generates random networks with power - law degree - distribution patterns .
the two salient features of the ba model are : _ growth _ and _ preferential attachment _ , both of which widely exist in real - world networks .
the presence of preferential attachment allowed networks to exhibit degree - heterogeneity .
a small fraction of nodes tend to attract more connections as compared to others , for example , highly influential people in a social network , important airports in an airline network or web pages , like google . in a way ,
preferential attachment acted as a positive feedback loop for the network or initiated the ` rich getting richer ' phenomenon in the network .
the earlier models failed to provide an explanation to this particularly observed phenomenon .
barabsi and albert in their model incorporated this property and came up with the class of scale - free networks .
the ba model , with growth and preferential attachment can be mathematically described using continuum theory , which calculates the time - dependence of the degree of a node @xcite .
incorporating growth causes the network size to increase with time due to new incoming nodes .
the preferential attachment is taken care by assigning a probability , @xmath124 for the new node to attach to an existing node in the network , such that @xmath125 .
it is assumed that @xmath25 is a continuous real variable and the rate at which @xmath25 changes is proportional to @xmath124 . starting with a small number ( @xmath126 ) of nodes , we add at every time - step a new node with @xmath127 edges .
the real variable , @xmath25 satisfies the dynamical equation : @xmath128 where @xmath129 leading to : @xmath130 the solution to the above equation with the initial condition that every node introduced at time @xmath131 has @xmath127 edges @xmath132 is given by @xmath133 , with @xmath134 . using eqn .
11 , the probability that a node has a degree @xmath135 smaller than @xmath46 , @xmath136 can be written as : @xmath137 since the nodes are added at equal time intervals , the parameter @xmath131 has a constant probability density , @xmath138 . substituting the value of @xmath139 into eqn .
12 , we obtain : @xmath140 the degree - distribution @xmath45 can be obtained by : @xmath141 the above equation can be written in the asymptotic ( @xmath142 ) limit as @xmath143 , with @xmath144 .
the above result is independent of the number of links ( @xmath127 ) an incoming node has .
therefore , ba model produces scale - free networks with degree - distribution following pure power - laws with degree exponent , @xmath145 ( see figure 7 ) .
although ba model generates networks obeying pure power - law degree - distributions , real - world networks on the contrary mostly show a heavy tail .
the reason for this can be attributed to several factors , such as non - linear preferential attachment , node aging , node s competitiveness to acquire incoming links , or presence of exponential cut - offs @xcite .
even though real - world networks may not follow strict power - laws , the presence of preferential attachment is extremely crucial . without the presence of the preferential attachment
, rule networks can not exhibit heavy - tailed degree - distribution patterns because the preferential attachment rule acts as a positive reinforcement in the network .
ba models , without growth have also been proposed . however , network models without the attachment rule will result into random graphs @xcite . the characteristic path length for ba networks
is given as @xmath146 .
for @xmath147 , the characteristic path length has been found to decay as @xmath148 , thus making the topology of scale - free networks ultra - small in nature @xcite . and @xmath149 .
the node size reflects the node degree , @xmath150 , nodes which have bigger size have higher degrees and vice versa . in the right panel , we plot the degree - distribution of the network on a double logarithmic scale .
the degree - distribution follows a power - law pattern , @xmath151 , with @xmath145 .
the parameter , @xmath44 is calculated from the slope of the degree - distribution plot in the right panel ( red , dotted).,scaledwidth=100.0% ] the availability of huge amounts of real - time data and sudden surge of interest in the field of networks have attracted numerous researchers working in applied sciences as well .
particular in this context are those civil engineers whose research is primarily focused in the domain of transportation networks .
although the study of transportation networks have a long history , the questions of interests have significantly changed after the introduction of network science techniques . the traditional network flow formulation ( in transportation networks )
has answered many interesting engineering questions related to optimality of cost , maximality of flows and the classical shortest path determination @xcite .
questions , those related to the topological structure of the network , such as the presence of small - world property or heavy - tailed degree - distribution patterns , or questions which are primarily concerned with the inter - nodal connectivity and static ( dynamic ) evolution of the network which the traditional formulation failed to address , found answers in the network science domain .
one particular type of transportation network that found widespread interest in the network science community is that of the public transit network or ptn .
a topological drawback of ptns is that they are structurally constrained in a two - dimensional space as compared to other networks , such as the internet , social networks or the airline networks .
airline and metro - networks , specifically have been reported to show scale - free degree distribution patterns , whereas degree - distribution in bus and railway networks tend more towards exponential forms ( or to power - law patterns with larger magnitudes of @xmath44 ) . yet ,
the above mentioned properties - small - world phenomenon and scale - free topology - have been reported in them as well @xcite . in some studies ,
specific subsets of ptns were analyzed , for example , the boston subway network , the vienna subway network , or bus networks of three cities in china @xcite .
it is important to note that each separate type of public transport ( bus , subway , trams or mono - rail network ) is not a closed system : these are only sub - networks of a much wider city transport system .
an interesting observation was reported when network characteristics obtained while separately analyzing subway networks were compared to those of the network combination of subway and buses @xcite .
statistical analysis of complete ptns for cities , such as berlin , paris , dsseldorf and 22 polish cities showed that power - law degree - distribution pattern is a common feature @xcite .
they also concluded that the degree - distribution patterns for ptns can have both exponential as well as power - law forms , depending upon the topology of the network representation chosen and presence of geographical constraints .
the small - world phenomenon in transportation networks makes sense as transportation facilities in a city are planned to provide maximum convenience to its people by allowing them to travel between places in minimum possible time .
most transportation networks are pre - planned networks , where the initial design of the network decides the presence of hubs . also , the size of transportation networks is not as large compared to social - networks or the internet , and are subjected to geographical as well as socio - economical constraints .
the reason for the contrasting behaviour of the airline and metro networks from bus and rail networks can be attributed to the two following observations : ( * i * ) airline - networks ( like the internet or social - networks ) are not bounded by geographical constraints , and ( * ii * ) metro - networks are _ local _ often catering to a part of the city , whereas bus and railway - networks are _ global _ as they are spread throughout the entire state and sometimes across the entire country . specific to indian scenarios , exhaustive studies on public transit networks as a whole are yet to be conducted .
previous works , have shown that the pattern of nodal connectivity of the indian railway network ( irn ) drastically differs from that of the airport network of india ( ani ) , while the nature of indian bus networks still remains an unsolved problem @xcite .
the central theme of the thesis lies around the topological structure and dynamics of bus transport networks or btns in india .
the presence of comparatively less number of studies on btns among other modes of transportation is one of the fundamental reasons that their characteristic properties and their topological structures are yet inconclusive . in this section
, we present the currently available literature on the studies pertaining to btns in specific .
analysis of the statistical properties of btns in china have revealed scale - free degree - distribution patterns and small - world properties in them .
the presence of nontrivial clustering , _ i.e. _ , the variation of clustering with degree , indicated a hierarchical and modular structure in those btns .
weighted analysis of the network revealed a heavy - tailed power law with the strength ( weighted - degree ) and degree showing a linear dependency @xcite . in another study , the btns of four major cities of china , namely
, hangzhou , nanjing , beijing and shanghai were analyzed using @xmath152-space topology .
the degree - distribution was reported to follow an exponential form , indicating a tendency for random attachment of the nodes .
the authors also evaluated two new statistical properties of the btns : the distribution of number of stops in a bus route ( @xmath153 ) , and the number of bus routes a stop joins ( @xmath154 ) .
while the former had an exponential functional form , the latter had an asymmetric unimodal functional form @xcite . in a separate study ,
the urban public bus networks of two chinese cities , beijing and chengdu were analyzed .
their analysis revealed small world characteristics and a scale - free topology , although with exceptionally high values for the degree - exponent , @xmath44 .
the presence of more hubs in the beijing network yielded a smaller @xmath44 as compared to chengdu ; while both showed large clustering coefficients and small characteristic path lengths .
the location of bus stops in a similar fashion in both the cities have led to a hierarchical structure , which is denoted by a power - law behaviour ( with nearly same exponents ) between the degree strength ( characterizing the passenger flows ) and clustering coefficient @xcite . in a recent study ,
the combination of rail ( rtns ) and bus transportation systems ( btns ) in singapore were studied with respect to their topological as well as dynamical perspectives .
the stations in rtns had high average degree indicating high connectivity amongst them , while the btns had a small average degree .
both networks had an exponential degree - distribution indicative of randomly evolved connectivity .
strength - distribution of the nodes ( weighted degree - distribution ) however showed scale - free topology for both networks , indicating the existence of high traffic hubs .
both btns and rtns exhibited small - world characteristics .
the btn in particular had a hierarchical star like topology .
degree - assortativity ( @xmath155 ) , which measures the inter - connectivity between hubs , revealed rtns to be slightly disassortative , while the btns displayed strong disassortative nature @xcite . with the availability of geo - located data , an extended space ( es ) model with information on geographical location of bus stops and routes
was recently used to analyze the spatial characteristics of btns in china . the @xmath156 model consisted of
directed weighted variations of the @xmath157- and @xmath152-space networks , designated as es-@xmath157 and es-@xmath152 networks respectively . often , two bus stops which are geographically close to each other
may not have any direct bus route link between them .
however , such stops are at walkable distances from each other .
these are defined as short - distance station pairs or ssps .
the ssps greatly influence the btns by reducing the transfer times as well as the number of bus routes . the symmetry - weighted esw network model stored information of the ssps .
the average clustering coefficient of the esw network was considerably large , denoting a nearly circular location of the ssps around a station .
majority of the route sections in the bus routes were short , while a few route sections connecting cities downtowns and satellite towns or special purpose routes were long , leading to a power - law edge length distribution of the es-@xmath157 network @xcite .
it may seem at first that the complexity of a bus transportation network is much lesser than that of other large - scale networks , however it is the nature of growth and the penetrative effect of these networks that makes them not only complex , but interesting and worthwhile to investigate . in the earlier sections , we discussed about the topological characteristics of the networks in detail . in order to get a deeper insight into the network characteristics , it is important to look at the dynamical process on networks as well . in this section ,
we discuss the two dynamical processes on networks in detail : network diffusion ( si model ) and network contagion ( sir model ) .
the spatial characteristics of the networks along with numerical simulations of the dynamical processes will help us understand the in - depth significance of various network metrics .
the earliest accounts of mathematical modelling to capture the spread of diseases dates back to as early as the @xmath158 century .
bernoulli used mathematical equations to defend his stand on vaccination against the outbreak of smallpox .
works following bernoulli earliest formulation of epidemic modelling helped in understanding germ theory in detail .
however , this was not until the works of mckendrick and kermack , which first proposed a deterministic model that predicted epidemic outbreaks very similar to the ones that were recorded during those times @xcite .
since then our understanding of the mathematical models in epidemology has evolved over the years , the accounts of which can be found in the extensive works of anderson and may @xcite .
all the above formulations focused on modelling epidemics over a set of population in which uniform ties between agents were assumed _ a priori_. contrary to this , the field of network science asserts the fact that ties , their strength and types in a system or a population , are not uniform and they play a significant role to describe the system s dynamics @xcite . also , a network model is not exhaustive to the study of population , rather it is a universal framework which can be used to understand numerous complex systems in general . over the years
, the term _ epidemic modelling _ has evolved into a common metaphor for a wide array of dynamical processes on these networks .
various complex phenomena such as percolation , the spreading of blackouts to the spreading of memes , ideas and opinions in a social network have been modelled under the common framework of epidemic modelling @xcite . in this context ,
transportation networks play a vital role due to their widespread outreach across cities , countries and continents @xcite .
hould people be worried about getting ebola on the subway@xmath159 was one of the numerous similar headlines that made the front pages of the newspapers around the world during the 2014 ebola scare @xcite . in this particular incident however , nobody was infected because the subject did not show symptoms of ebola while using public transportation .
therefore not only airline networks , that can transmit pathogens across continents , even modes of public transport operating within cities , such as buses and subways , pose a serious threat as well as a source of panic during desperate times .
although epidemic spreading in airline networks have been studied extensively , similar studies on bus networks are relatively rare @xcite .
epidemological models have been simulated on bus network datasets ; however , the results were only used to validate the numerical models .
also , a recent study on city - wide integrated travel networks ( itns ) , has found both the traveling speed and frequency to be important factors in epidemic spreading @xcite .
thus , the effect of network structure and constraints in epidemic spreading is yet to be studied in these networks .
the si model is the most basic representation of an epidemic spreading model that captures diffusion in complex networks . in this model , there are two states that an agent or a node can exist in : s ( susceptible ) or i ( infected ) .
the si model describes the status of individuals or agents switching from susceptible to infected at every instant of time .
it is assumed that the population is homogeneous and closed , _
i.e. _ , no new entity is either created due to birth or removed due to death , and also , no new entity enters the system , thus preserving homogeneous mixing in the system .
the si model also implies that each individual has the same probability to transfer disease , innovation or information to its neighbors .
thus , the si model helps to capture the diffusion or percolation process in the entire network .
the si model is formulated using the following differential equation . since an agent in the entire population
can either be in state s or i , @xmath160 the si model is governed by a single parameter , @xmath161 , the infection transmission rate or simply , the infection rate .
the growth in the number of agents in either of the sates is given by : @xmath162 substituting the value of s from eqn . 15 to eqn .
16 , we get the following differential equation describing the growth rate of i : @xmath163 the solution of the above equation with the initial condition at @xmath164 , @xmath165 is given by the logistic form : @xmath166 contrary to the si model , the agents in sir model have access to three states : s ( susceptible ) , i ( infected ) and r ( recovery ) . although the earlier assumptions of a closed population and homogeneous mixing also hold in this case , the complexity of the dynamical process increases due to the addition of one more state .
the agents , instead of only switching between susceptible and infected ( as in si model ) , tend to recover in the sir epidemic model .
the dynamics of the sir model is controlled by two parameters : the infection rate , @xmath161 , and the recovery rate , @xmath44 .
the sir model can be mathematically represented by the set of the following differential equations : @xmath167 and @xmath168 .
the @xmath169axis represents the population fraction of @xmath153 , @xmath170 and @xmath154 species respectively and @xmath171axis represents simulation time.,scaledwidth=75.0% ] the population of susceptible nodes decreases in proportion to the number of encounters multiplied by the probability that each encounter results in an infection .
the negative sign denotes that the population of s is decreasing .
similarly , we can describe the evolution of the other two states , i and r. nodes become infected at a rate proportional to the number of encounters , and the probability of infection controlled by the parameter , @xmath161 .
nodes recover at a rate proportional to the number of infected individuals , and the probability of recovery controlled by the parameter , @xmath44 : @xmath172 it would be interesting to analyze the spread of infection with respect to the susceptible individuals when there is a constant recovery ( from eqn .
we calculate the variation in i with respect to s : @xmath173 the solution to the above equation with the initial conditions at @xmath164 , @xmath174 ( negligible as compared to the population ) and @xmath175 , is given by : @xmath176 in order to understand the rate of spread of infection in the population , we look at the rate equation for i from eqn . 20 :
@xmath177 the above equation implies that the infection spreads if and only if @xmath178 . the epidemic dies out ( the number of infected individuals decreases ) if the above quantity is less than zero .
bifurcation occurs at the stationary state , when @xmath179 , which separates the above two regimes and corresponds to the epidemic threshold . for a network with average node degree @xmath42
, the rate of change of the susceptible population is given by : @xmath180 the rate of change of the infected and simultaneously recovered individuals is similarly given by : @xmath181 note that the rate of change of the recovered individuals remains the same as before .
substituting the value of i from eqn .
25 into eqn .
24 , and solving for s gives us the following expression for s in terms of r , @xmath182 at time @xmath142 , @xmath183 and @xmath184 .
population of recovered individuals , @xmath185 is given as @xmath186 .
recovery of the individuals occur in the network if and only if the slope of @xmath187 or @xmath188 .
in this section , we present our main results on the studies of btns in india .
first , we discuss the motivation of our study , and then we describe the datasets used for analysis . before we discuss the primary results of this work in detail , we outline the network representation techniques specific to transportation terminology .
we also saw earlier that any physical , chemical , biological or social system can be visualized as a complex network with constituting elements known as nodes , and the interactions between them identified as links . based on the nature of the links ,
these networks can be broadly classified into virtual and spatial networks . in the former category , the links are physically absent , e.g. , social networks or collaboration networks , whereas in the later case
, the links are physically present , e.g. , geographically embedded road or railway networks . in between these two broad classes , there exist networks in which the links , although physically absent , are still geographically constrained . the structure of the real - world networks such as bus or electric power grid are dependent upon the structure of the physically constrained , geographically embedded networks on which they grow and evolve .
therefore , our study would be incomplete if we do not explain the role constraints play in geographical embeddedness .
we analyze this aspect by calculating the ` dimensions ' of these networks and check for self - similar patterns in them . in this study , we present the statistical analysis of the bus networks of six major indian cities as graphs in @xmath157- and @xmath152-spaces , using concepts from network science .
although public transport networks such as airline and railway networks have been extensively studied , a comprehensive study on the structure and growth of bus networks is still lacking . in india , where btns play an important role in day - to - day commutation , it is of significant interest to analyze their topological structures and answer some of the basic questions on their growth , evolution , robustness and resiliency .
therefore , we do a comparative study of the bus networks of some of the major indian cities , namely ahmedabad ( abn ) , chennai ( cbn ) , delhi ( dbn ) , hyderabad ( hbn ) , kolkata ( kbn ) and mumbai ( mbn ) . in order to understand the structure of these networks , we calculate various metrics , such as clustering coefficients , characteristic path lengths , degree - distribution and assortativity . we also simulate network robustness and resiliency by first removing nodes at random , followed by targeted removal based on degree , closeness and betweenness .
simulating node removals and simultaneously capturing the variation in their characteristic path lengths helps us in understanding nodal redundancy in these networks .
few studies on btns have although looked into the structural aspects in detail , the esw network that we described earlier , is the only model that has looked into the aspect of network redundancy albeit due to geographical placements of the nodes . for this study
, we use the bus routes as network datasets by considering bus stops as nodes and bus routes as links .
the route details were obtained from the government websites of amts ( abn ) , mtc ( cbn ) , dtc ( dbn ) , apsrtc ( hbn ) , cstc ( kbn ) , best ( mbn ) and ahmedabad brts ( bus rapid transit system ) .
it can be seen from table 1 that the network sizes of all the cities are comparable to each other , except that of kbn , because cstc is localized and operates as a subdivision of the west bengal surface transport corporation ( wbstc ) that operates buses in the entire state . for computational and visualization purposes
we parse the datasets as edge - lists , where the two adjacent columns are labeled as ` source ' and ` target ' respectively . in @xmath157-space representation , the values in the respective columns represent neighboring bus stops in a given route , whereas in @xmath152-space representation , the adjacent columns represent all possible transfers in a route for a fixed value in the column of ` source ' ( see figure 9 ) .
-space and @xmath152-space representation for transportation networks .
if @xmath189 represents a route in @xmath157-space , then in @xmath152-space representation we have all the possible transfers for the route , _
i.e. _ , @xmath190.,scaledwidth=100.0% ] before we analyze the statistical properties of the different btns , we need to understand the relationship between the various network representation forms and their respective advantages .
the most common is the @xmath157-space representation , where each bus stop is a node and a link between the nodes indicates that there is at least one route that services these two corresponding nodes consecutively ( see figure 9 ) . in this representation , no multiple links are allowed between a pair of nodes , or @xmath15 . in cases of route overlaps ,
the element @xmath18 is multiplied with a weight element @xmath191 , thus denoting strength of that link ( or node ) : @xmath192 .
a different network representation is that of a bipartite graph that has been found to be useful in the analysis of cooperation networks . in this representation , also called @xmath193-space , both routes and stops are represented by nodes and each route node is linked to all bus stops that it services , and no direct links between nodes of same type occur .
the neighbors of a given route node in this representation are all those stops that it services , while the neighbors of a given bus stop are all the routes that service it @xcite .
there are two one - mode projections of the bipartite graph of @xmath193-space .
the projection to the set of station nodes is the @xmath152-space graph ( see figure 9 ) , and the complementary projection to route nodes leads to the @xmath51-space graph .
the @xmath152-space network representation has proven to be particularly useful in the analysis of ptns @xcite .
the nodes of this graph are bus stops , and they are linked if they are serviced by at least one common route . in this way , the neighbors of a @xmath152-space node are all those stops that can be reached without changing means of transport . in order to get the essence of different network representations and their significance , we try to visualize the metric : characteristic length ( @xmath20 ) , which in an @xmath157-space graph is the number of ` hops ' one has to make to travel between any two randomly chosen bus stops . when the network is represented in @xmath152-space
, @xmath20 signifies the number of bus changes one has to make in order to travel between any two randomly chosen bus stops . from a transportation perspective
, less number of bus changes will imply a small - world property .
therefore , our calculation of the network metrics will strongly depend upon the network representation that is chosen .tabular representation of the statistical data for the ahmedabad brts and bus routes of six major indian cities as graphs in @xmath157-space .
@xmath127 represents the number of communities in these networks . for ahmedabad brts ,
the dataset is very limited , which disables us from concluding its exact topological structure . [
cols="^,^,^,^,^,^,^,^,^,^",options="header " , ] in figure 25 , we simulate the si and sir models on the us airline network for 500 of the busiest airports ( @xmath194 ) @xcite .
statistical analysis of the network reveals scale - free degree - distribution pattern between the nodes , with the characteristic path length @xmath195 and average degree @xmath196 .
the si plot of the us airline network and abn show similar pattern of growth as both of them exhibit scale - free behaviour .
however , the sir plot is similar to that of hbn due to extremely low characteristic path length and high average node degree . in table 3
, we present our findings for the networks studied in the paper .
the first column represents the simulation time ( in seconds ) for @xmath197 percolation threshold from the si model . in the second and third column ,
we present the epidemic thresholds for the various networks studied by computing the values of the plots from the sir model ( figure 24 ) ( as a fraction of network size ) and the corresponding simulation times ( in seconds ) respectively . in the final column
, we present the characteristic path lengths for the various networks .
node removal ( @xmath198 degree - biased , @xmath199 betweenness - biased and @xmath200 closeness - biased ) .
the y - axis denotes the cdf of the infection probability of the nodes , and the x - axis represents simulation time.,scaledwidth=100.0% ] finally , in figure 26 , we plot the variation in the rate of percolation by removing nodes from the network based upon their centralities and degrees . in transportation networks ,
other than the degree of a node , closeness and betweenness centralities play a crucial role . in order to capture their effects on information diffusion , we simulate the si model on modified networks generated after directed removal of nodes .
since cbn and mbn are strongly assortative , we remove only two percent of the nodes ( higher number of node removal will cause cbn and mbn to disintegrate into disconnected components ) .
we find that the removal of nodes does not significantly affect the diffusion in abn .
however for cbn , dbn and hbn , we observe that when nodes are removed based upon their closeness centrality , the diffusion curve shifts towards right , thus signifying a delay in the diffusion process .
this can be explained due to the fact that the removal of nodes based upon closeness centrality has a direct effect on the characteristic path length .
a node with high closeness allows every other node in the network to be reached along the shortest paths .
the removal of such a node affects / delays diffusion until the next central node is encountered . for mbn , we observe that degree - biased removal causes the diffusion rate to increase steeply , signifying the presence of redundant nodes that simply increase the characteristic path length of the network .
a removal of @xmath201 of such nodes causes the diffusion to improve significantly , as can be compared from the simulation times recorded in figure 23 .
the present work places before us numerous questions of both academic as well as practical pursuit . one important practical scope of this work is in the planning of large - scale transportation networks for the future
. it would be interesting to see the functionality of those networks which are planned using network science tools and techniques .
another question of practical importance lies in optimizing these networks for efficient transportation and communication purposes . one important finding that could be of practical interest is the strong positive correlation between network assortativity and characteristic path length : will it be more efficient to travel on longer routes to reach hub nodes and from there to other parts of the city , or to travel though shorter routes , not reach the hub nodes but travel to other parts of the city through intermediate nodes ? from academic pursuit , there are numerous scopes for in - depth analysis based upon this study .
the current study takes into account one subset of the large - scale public transit networks
. it would be interesting to do an integrated study involving other modes of transport as well .
the availability of high quality geo - located data will help in actually identifying redundant nodes in the network , thus making the network more efficient .
since transportation plays an important role in the economic development of a city , the present study can be extended to incorporate other networks as well , such as road networks , supply chain networks and economic networks .
a holistic approach to all these networks will help us in understanding each layer of society s complexity .
in this study , we analyzed the statistical properties of the bus routes of six indian cities , namely ahmedabad , chennai , delhi , hyderabad , kolkata and mumbai .
our analysis suggests that the bus networks show a wide spectrum of topological structure from power - law to exponential , with varying magnitude of the power - law exponent @xmath44 .
we also observe that these networks show small - world behaviour in terms of either the placement of the bus stops ( @xmath157-space ) or in terms of transfers ( @xmath152-space ) .
for example , cbn and mbn do not show the small - world property in @xmath157-space .
they , however , do show the small - world property in terms of transfers , as the majority of places can be visited by making as little as 2 to 3 bus changes .
the redundancy in the network structure , as seen from the variation in @xmath20 and the presence of exponential cutoffs in the degree - distribution plots , validate our findings regarding the randomness associated with the growth and evolution of the bus networks with time .
recently , wang et al .
simulated exponential growth models for networks , with growth and adjacent node attachment as underlying processes .
the growth equation according to their model is as follows : @xmath202 , where @xmath203 .
the variable @xmath204 marks the vertices , and @xmath205 represents the probability that a vertex @xmath206 has a degree @xmath46 at time @xmath207 . the stationary degree distribution , @xmath208
is given as @xmath209 .
in the continuum limit , the above equation approaches the form @xmath210 , giving rise to the decay equation @xmath211 , with @xmath212 @xcite .
our findings on the weak correlations between degree and centrality plots and self - similar structures of sub - networks motivate us to investigate the fractal nature of these networks .
the degree - degree correlation matrix gives a rough idea on the degree - assortativity of these networks .
strong assortativity in networks relate to strong connections between hubs , thus hubs tend to come in - between short - range connections .
whereas , in disassortative mixing we observe hub - repulsion .
the presence of hub - inter connectedness causes the characteristic path lengths of the networks to increase and also the fractal nature to diminish .
presence of repelling hubs generate fractal topology in complex networks .
an interesting aspect of the fractal topology is that they generate local small - worlds in well - defined communities . from a transportation perspective
, it would be beneficial to have local small - worlds connected to a central core , as such a structure will reduce the characteristic path length and make the network efficient .
the high values of characteristic path lengths for cbn and mbn can be attributed to the geographical structures of the two cities , chennai and mumbai .
the routes in these two cities are exceptionally longer because the btns in these two cities have evolved more in a linear fashion .
the reason for this may lie in the geographical limitation imposed by the presence of a water body on one side ( see figure 27 ) .
also , most of our results on ahmedabad brts are inconclusive as the dataset is small . at present , the brts operates across only 13 routes with 129 bus stops .
therefore , it would be interesting to see the evolution of this network in future . finally , we simulate si and sir dynamical processes on these networks in @xmath157-space . since experiments with epidemic outbreaks in a population ( or a network ) is not a viable option , we resort to mathematical modelling to understand diffusion of information and contagion spreading .
we therefore study the effect of percolation and epidemic spreading on these networks using si and sir epidemic models through numerical simulations . while it is observed that the characteristic path length plays a crucial role in information diffusion and epidemic spreading , several other network metrics also play important roles .
their importance is however restricted to their relative contribution to the topological structures of the networks .
small - world property , while an extremely desirable property in transportation networks , is highly subjective in its role of information diffusion , solely due to the diffusing entity .
finally , bus networks form a specific class of complex networks that grow and evolve over physically constrained spatial networks .
interesting in this regard is the city of ahmedabad and the abn
. statistical analysis of road networks ( by considering intersections as nodes and roads as links ) has shown that the topological structure of the road networks in the city of ahmedabad exhibits a scale - free degree distribution with @xmath213 and @xmath214 , which is very much similar to abn @xcite .
road intersections are usually separated by a distance which is geographically much smaller as compared to the distance between bus stops .
therefore , our results emphasize the fact that transportation undoubtedly brings the world closer .
weibing deng , wei li , xu cai , and qiuping a wang .
the exponential degree distribution in complex networks : non - equilibrium network theory , numerical simulation and empirical data .
, 390(8):14811485 , 2011 .
yihan zhang , qingnian zhang , and jigang qiao .
analysis of guangzhou metro network based on l - space and p - space using complex network . in
_ geoinformatics ( geoinformatics ) , 2014 22nd international conference on _ , pages 16 .
ieee , 2014 .
o woolley - meza , c thiemann , d grady , jj lee , h seebens , b blasius , and d brockmann .
complexity in human transportation networks : a comparative analysis of worldwide air transportation and global cargo - ship movements .
, 84(4):589600 , 2011 .
roger guimera , stefano mossa , adrian turtschi , and la nunes amaral .
the worldwide air transportation network : anomalous centrality , community structure , and cities global roles .
, 102(22):77947799 , 2005 .
harold soh , sonja lim , tianyou zhang , xiuju fu , gary kee khoon lee , terence gih guang hung , pan di , silvester prakasam , and limsoon wong .
weighted complex network analysis of travel routes on the singapore public transportation system .
, 389(24):58525863 , 2010 .
william o kermack and anderson g mckendrick .
a contribution to the mathematical theory of epidemics . in _ proceedings of the royal society of london a : mathematical , physical and engineering sciences _ , volume 115 ,
pages 700721 . the royal society , 1927 . atanu chatterjee and gitakrishnan ramadurai .
scaling laws in chennai bus network . in _
4th international conference on complex systems and applications , france _ , pages 137141 .
https://halshs.archives-ouvertes.fr/halshs-01060875/document , 2014 . | in recent times , the domain of network science has become extremely useful in understanding the underlying structure of various real - world networks and to answer non - trivial questions regarding them . in this study
, we rigourously analyze the statistical properties of the bus networks of six major indian cities as graphs in _ l_- and _ p_-space , using tools from network science .
although public transport networks , such as airline and railway networks have been extensively studied , a comprehensive study on the structure and growth of bus networks is lacking . in india , where bus networks play an important role in day - to - day commutation ,
it is of significant interest to analyze their topological structure , and answer some of the basic questions on their evolution , growth , robustness and resiliency .
we start from an empirical analysis of these networks , and determine their principle characteristics in terms of the complex network theory .
the common features of small - world property and heavy tails in degree - distribution plots are observed in all the networks studied .
our analysis further reveals a wide spectrum of network topologies arising due to an interplay between preferential and random attachment of nodes . unlike real - world networks , like the internet , www and airline , which are virtual ,
bus networks are physically constrained in two - dimensional space by the underlying road networks . in order to understand the role of constraints in the evolution of these networks ,
we calculate their fractal dimensions that reveal a three - dimensional space - like evolution in a constrained two - dimensional plane .
we also extend our study to understand the complex dynamical processes of epidemic outbreaks and information diffusion in these networks using si and sir models .
our findings , therefore throw light on the evolution and dynamics of such geographically and socio - economically constrained networks , which will help us in designing more efficient networks in the future . + * keywords * : complex networks , power - laws , self - similarity , small - world phenomenon , transportation networks |
outflows in the form of jets and winds are observed from many disc accreting objects ranging from young stars to systems with white dwarfs , neutron stars , and black holes .
a large body of observations exists for outflows from young stars at different stages of their evolution , ranging from protostars , where powerful collimated outflows - jets - are observed , to classical t tauri stars ( cttss ) where the outflows are weaker and often less collimated ( see review by ray et al .
correlation between the disc s radiated power and the jet power has been found in many cttss ( cabrit et al .
1990 ; hartigan , edwards & gandhour 1995 ) . a significant number of cttss show signs of outflows in spectral lines , in particular in he i where two distinct components of outflows had been found ( edwards et al .
2003 , 2006 ; kwan , edwards , & fischer 2007 ) .
outflows are also observed from accreting compact stars such as accreting white dwarfs in symbiotic binaries ( sokoloski & kenyon 2003 ) , or from the vicinity of neutron stars , such as from circinus x-1 ( heinz et al .
2007 ) .
different theoretical models have been proposed to explain the outflows from protostars and cttss ( see review by ferreira , dougados , & cabrit 2006 ) .
the models include those where the outflow originates from a radially distributed disc wind ( blandford & payne 1982 ; knigl & pudritz 2000 ; casse & keppens 2004 ; ferreira et al .
2006 ) or from the innermost region of the accretion disc ( lovelace , berk & contopoulos 1991 ) .
the latter model is related to the x - wind model ( shu et al . 1994 ; 2007 ; najita & shu 1994 ; cai et al .
2008 ) where the outflow originates from the vicinity of the disc - magnetosphere boundary .
progress in understanding the theoretical models has come from mhd simulations of accretion discs around rotating magnetized stars as discussed below .
laboratory experiments are also providing insights into jet formation processes ( hsu & bellan 2002 ; lebedev et al .
2005 ) but these are not discussed here .
outflows or jets from the disc - magnetosphere boundary were found in early axisymmetric mhd simulations by hayashi , shibata & matsumoto ( 1996 ) and miller & stone ( 1997 ) . a one - time episode of outflows from the inner disc and inflation of the innermost field lines connecting the star and the disc
were observed for a few dynamical time - scales .
somewhat longer simulation runs were performed by goodson et al .
( 1997 , 1999 ) , hirose et al .
( 1997 ) , matt et al .
( 2002 ) and kker , henning & rdiger ( 2003 ) where several episodes of field inflation and outflows were observed .
these simulations hinted at a possible long - term nature for the outflows .
however , the simulations were not sufficiently long to establish the behavior of the outflows .
mhd simulations showing long - lasting ( thousand of orbits of the inner disc ) outflows from the disc - magnetosphere have been obtained by our group ( romanova et al .
2009 ; lii et al .
we obtained these outflows / jets in two main cases : ( 1 ) where the star rotates slowly but the field lines are bunched up into an x - type configuration , and ( 2 ) where the star rotates rapidly , in the `` propeller regime '' ( illarionov & sunyaev 1975 ; alpar & shaham 1985 ; lovelace , romanova & bisnovatyi - kogan 1999 ) .
figure 1 shows the equatorial angular rotation rate @xmath3 of the plasma in the two cases . here
, @xmath4 is the radius of the star ; @xmath5 is the magnetospheric radius where the kinetic energy density of the disc matter is about equal to the energy density of the magnetic field ; and @xmath6 is the co - rotation radius where the angular rotation rate of the star @xmath7 equals that of the keplerian disc @xmath8 . for a slowly rotating star @xmath9 whereas for a rapidly rotating star in the propeller regime @xmath10 .
is the angular rotation rate of the star , @xmath11 is the keplerian rotation rate of the disc , @xmath4 is the star s radius , @xmath5 is the radius of the magnetosphere , and @xmath12 is the co - rotation radius.,width=480 ] figure 2 shows examples of the outflows in the two cases . in both cases , two - component outflows are observed : one component originates at the inner edge of the disc near @xmath5 and has a narrow - shell conical shape close to the disc and therefore is termed a conical wind " .
it is matter dominated but can become collimated at large distances due to its toroidal magnetic field .
the other component is a magnetically dominated high - velocity `` axial jet '' which flows along the open stellar magnetic field lines .
the axial jet may be very strong in the propeller regime .
a detailed discussion of the simulations and analysis can be found in romanova et al .
( 2009 ) and lii et al .
( 2012 ) .
2 describes the simulations .
3 discusses the conical winds and axial jets , the driving and collimation forces , and the variability of the winds and jets .
4 discusses simulation results on one - sided and lop - sided jets . sec .
5 gives the conclusions .
, the arrows are the poloidal velocity vectors , and the lines are sample magnetic field lines .
the labels point to the main outflow components . ]
we simulate the outflows resulting from disc - magnetosphere interaction using the equations of axisymmetric mhd . outside of the disc
the flow is described by the equations of ideal mhd . inside the disc
the flow is described by the equations of viscous , resistive mhd . in an inertial reference frame
the equations are : @xmath13 @xmath14 @xmath15 @xmath16 here , @xmath17 is the density , @xmath18 is the specific entropy , @xmath19 is the flow velocity , @xmath20 is the magnetic field , @xmath21 is the magnetic diffusivity , @xmath22 is the momentum flux - density tensor , @xmath23 is the rate of change of entropy per unit volume , and @xmath24 is the gravitational acceleration due to the star , which has mass @xmath25 .
the total mass of the disc is assumed to be negligible compared to @xmath25 . here
, @xmath26 is the sum of the ideal plasma terms _ and _ the @xmath27viscosity terms discussed in the next paragraph .
the plasma is considered to be an ideal gas with adiabatic index @xmath28 , and @xmath29 .
we use spherical coordinates @xmath30 with @xmath31 measured from the symmetry axis .
the equations in spherical coordinates are given in ustyugova et al .
( 2006 ) .
both the viscosity and the magnetic diffusivity of the disc plasma are considered to be due to turbulent fluctuations of the velocity and the magnetic field .
both effects are non - zero only inside the disc as determined by a density threshold .
the microscopic transport coefficients are replaced by turbulent coefficients .
the values of these coefficients are assumed to be given by the @xmath32-model of shakura and sunyaev ( 1973 ) , where the coefficient of the turbulent kinematic viscosity is @xmath33 , where @xmath34 is the isothermal sound speed and @xmath35 is the keplerian angular velocity .
we take into account the viscous stress terms @xmath36 and @xmath37 ( lii et al . 2012 ) . similarly , the coefficient of the turbulent magnetic diffusivity @xmath38 . here , @xmath39 and @xmath40 are dimensionless coefficients which are treated as parameters of the model .
the mhd equations are solved in dimensionless form so that the results can be readily applied to different accreting stars . [ cols="<,<,<,<,<,<,<",options="header " , ] the system of mhd equations ( 1 - 4 ) have been integrated numerically in spherical @xmath41 coordinates using a godunov - type numerical scheme .
the calculations were done in the region @xmath42 , @xmath43 .
the grid is uniform in the @xmath31-direction with @xmath44 cells .
the @xmath45 cells in the radial direction have @xmath46 ( @xmath47 ) so that the poloidal - plane cells are curvilinear rectangles with approximately equal sides .
this choice results in high spatial resolution near the star where the disc - magnetosphere interaction takes place while also permitting a large simulation region . we have used a range of resolutions going from @xmath48 to @xmath49 .
a large number of simulations were done in order to understand the origin and nature of conical winds .
all of the key parameters were varied in order to ensure that there is no special dependence on any parameter .
we observed that the formation of conical winds is a common phenomenon for a wide range of parameters .
they are most persistent and strong in cases where the viscosity and diffusivity coefficients are not very small , @xmath50 , @xmath51 .
another important condition is that @xmath52 ; that is , the magnetic prandtl number of the turbulence , @xmath53 .
this condition favors the bunching of the stellar magnetic field by the accretion flow .
figure 3 shows a snapshot from our simulations at time @xmath54 for the propeller regime .
the figure shows the dimensionless density and velocity at sample points .
one can see that the velocities in the conical wind component are similar to those in conical winds around slowly rotating stars .
matter launched from the disc initially has an approximately keplerian azimuthal velocity , @xmath55 .
it is gradually accelerated to poloidal velocities @xmath56 and the azimuthal velocity decreases .
the flow has a high density and carries most of the disc mass into the outflows .
the situation is the opposite in the axial jet component where the density is @xmath57 times lower , while the poloidal and total velocities are significantly higher .
thus we find a _ two - component outflow _ : a matter dominated conical wind and a magnetically dominated axial jet .
we observe conical winds in both slowly and rapidly rotating stars . in both cases ,
matter in the conical winds passes through the alfvn surface ( and shortly thereafter through the fast magnetosonic point ) , beyond which the flow is matter - dominated in the sense that the energy flow is carried mainly by the matter .
the situation is different for the axial jet component where the flow is sub - alfvnic within the simulation region . for this component
the energy flow is carried by the poynting flux and the angular moment flow is carried by the magnetic field .
* collimation and driving of the outflows : * figure 4 shows the long - distance development of a conical wind from a slowly rotating star . at large distances the conical wind becomes collimated . to understand the collimation we analyzed total force ( per unit mass ) perpendicular to a poloidal magnetic field line ( lii et al .
2012 ) . for distances beyond the alfvn surface of the flow
this force is approxiamtely @xmath58 ( ustyugova et al .
1999 ) . here , @xmath59 is the angle between the poloidal magnetic field and the symmetry axis , @xmath60 is the arc length along the poloidal field line , @xmath61 is a coordinate normal to the poloidal field , and the @xmath62subscripts indicate the poloidal component of a vector .
once the jet begins to collimate , the curvature term @xmath63 also becomes negligible .
the magnetic force may act to either collimate or decollimate the jet , depending on the relative magnitudes of the toroidal @xmath64 gradient ( which collimates the outflow ) and poloidal @xmath65 gradient ( which `` decollimates '' ) . in our simulations
, the collimation of the matter implies that the magnetic hoop stress is larger than the poloidal field gradient .
thus the main perpendicular forces acting in the jet are the collimating effect of the toroidal magnetic field and the decollimating effect of the centrifugal force and the gradient of @xmath66 .
the collimated effect of @xmath67 dominates .
note that in mks units @xmath68 is the poloidal current flowing through a surface of radius @xmath69 from colatitude zero to @xmath31 .
for the jets from young stars this current is of the order of @xmath70 a. the driving force for the outflow is simply the force parallel to the poloidal magnetic field of the flow @xmath71 .
this is obtained by taking the dot product of the euler equation with the @xmath72 unit vector which is parallel to the poloidal magnetic field line @xmath73 .
the derivation by ustyugova et al .
( 1999 ) gives @xmath74.\ ] ] here , the terms on the right - hand side correspond to the pressure , gravitational , centrifugal and magnetic forces , respectively denoted @xmath75 .
the pressure gradient force , @xmath76 , dominates within the disk .
the matter in the disk is approximately in keplerian rotation such that the sum of the gravitational and centrifugal forces roughly cancel ( @xmath77 ) . near the slowly rotating star , however ,
the matter is strongly coupled to the stellar magnetic field and the disk orbits at sub - keplerian speeds , giving @xmath78 .
the magnetic driving force ( the last term of eq .
6 ) can be expanded as @xmath79 ( lovelace et al .
1991 ) .
figure 5 shows the variation of the total force @xmath80 , the gravitational plus centrifugal force , and the magnetic force along a representative field line .
this analysis establishes that the predominant driving force for the outflow is the magnetic force ( eq .
7 ) and not the centrifugal force .
this in agreement with the analysis of lovelace et al .
( 1991 ) .
* variability : * for both rapidly and slowly rotating stars the magnetic field lines connecting the disc and the star have the tendency to inflate and open ( lovelace , romanova & bisnovatyi - kogan 1995 ) .
quasi - periodic reconstruction of the magnetosphere due to inflation and reconnection has been discussed theoretically ( aly & kuijpers 1990 ) and has been observed in a number of axisymmetrtic simulations ( hirose et al . 1997 ; goodson et al . 1997 , 1999 ; matt et al .
2002 ; romanova et al .
goodson & winglee ( 1999 ) discuss the physics of inflation cycles .
they have shown that each cycle of inflation consists of a period of matter accumulation near the magnetosphere , diffusion of this matter through the magnetospheric field , inflation of the corresponding field lines , accretion of some matter onto the star , and outflow of some matter as winds , with subsequent expansion of the magnetosphere .
there simulations show @xmath81 cycles of inflation and reconnection .
our simulations often show @xmath82 cycles of inflation and reconnection .
figure 6 shows the time evolution of the accretion rates for a of a slowly rotating star ( romanova et al . 2009 ) .
kurosawa and romanova ( 2012 ) have calculated spectra from modeled conical winds using the radiative transfer code torus and have shown that conical winds may explain different features in the hydrogen spectral lines , in the he i line and also a relatively narrow , low - velocity blue - shifted absorption components in the he i @xmath83 which is often seen in observations ( kurosawa et al .
there is clear evidence , mainly from hubble space telescope ( hst ) observations , of the asymmetry between the approaching and receding jets from a number of young stars .
the objects include the jets in hh 30 ( bacciotti et al .
1999 ) , rw aur ( woitas et al .
2002 ) , th 28 ( coffey et al . 2004 ) , and lkh@xmath32 233 ( perrin & graham 2007 ) .
specifically , the radial speed of the approaching jet may differ by a factor of two from that of the receding jet .
for example , for rw aur the radial redshifted speed is @xmath84 km / s whereas the blueshifted radial speed is @xmath85 km / s .
the mass and momentum fluxes are also significantly different for the approaching and receding jets in a number of cases . of course , it is possible that the observed asymmetry of the jets could be due to say differences in the gas densities on the two sides of the source . here
, we discuss the case of intrinsic asymmetry where the asymmetry of outflows is connected with asymmetry of the star s magnetic field .
substantial observational evidence points to the fact that young stars often have _ complex _ magnetic fields consisting of dipole , quadrupole , and higher order poles misaligned with respect to each other and the rotation axis ( jardine et al . 2002 ;
donati et al . 2008 ) .
analysis of the plasma flow around stars with realistic fields have shown that a fraction of the star s magnetic field lines are open and may carry outflows ( e.g. , gregory et al . 2006 ) .
it is evident that the complex magnetic field of a star will destroy the commonly assumed symmetry of the magnetic field and the plasma about the equatorial plane .
figure 7 shows an illustrative complex magnetic field consisting of the combination of a dipole and a quadrupole field both of which are axisymmetric .
the figure includes the suggested locations of the funnel flow to the star ( romanova et al .
2002 ) and the conical wind outflows .
the mhd simulations fully support the qualitative picture suggested in figure 7 ( lovelace et al .
the time - scale during which the jet comes from the upper hemisphere is set by the evolution time - scale for the stellar magnetic field .
this is determined by the dynamo processes responsible for the generation of the field . remarkably , once the assumption of symmetry about the equatorial plane is dropped , we find that the conical winds alternately come from one hemisphere and then the other _ even _ when the stellar magnetic field is a centered axisymmetric dipole ( lovelace et al .
2010 ) . an illustrative case of this spontaneous symmetry breaking is shown in figure 8 .
the time - scale for the ` flipping ' is the accretion time - scale of the inner part of the disc which is expected to be much less than the evolution time of the star s magnetic field .
detailed magnetohydrodynamic simulations have established that long - lasting outflows of cold disc matter into a hot low - density corona from the disc - magnetosphere boundary in cases of slowly and rapidly rotating stars .
the main results are the following : for slowly rotating stars a new type of outflow a conical wind has been discovered .
matter flows out forming a _ conical wind _ which has the shape of a thin conical shell with a half - opening angle @xmath86 .
the outflows appear in cases where the magnetic flux of the star is bunched up by the inward accretion flow of the disc .
we find that this occurs when the turbulent magnetic prandtl number ( the ratio of viscosity to diffusivity ) @xmath87 , and when the viscosity is sufficiently high , @xmath50 .
winds from the disc - magnetosphere boundary have been proposed earlier by shu and collaborators and referred to as x - winds ( shu et al .
1994 ) . in this model
, the wind originates from a small region near the corotation radius @xmath12 , while the disc truncation radius @xmath88 ( or , the magnetospheric radius @xmath5 ) is only slightly smaller than @xmath12 ( @xmath89 , shu et al .
it is suggested that excess angular momentum flows from the star to the disc and from there into the x - winds .
the model aims to explain the slow rotation of the star and the formation of jets . in the simulations discussed here we have obtained outflows from both slowly and rapidly rotating stars
both have conical wind components which are reminiscent of x - winds . in some respects
the conical winds are similar to x - winds : they both require _ bunching _ of the poloidal field lines and show outflows from the inner disc ; and they both have high rotation and show gradual poloidal acceleration ( e.g. , najita & shu 1994 ) .
the main differences are the following : ( 1 ) the conical / propeller outflows have _ two components _ : a slow high - density conical wind ( which can be considered as an analogue of the x - wind ) , and a fast low - density jet .
no jet component is discussed in the x - wind model .
( 2 ) conical winds form around stars with _ any rotation rate _ including very slowly rotating stars .
they do not require fine tuning of the corotation and truncation radii .
for example , bunching of field lines is often expected during periods of enhanced or unstable accretion when the disc comes closer to the surface of the star and @xmath90 . under this condition
conical winds will form .
in contrast , x - winds require @xmath91 .
( 3 ) the base of the conical wind component in both slowly and rapidly rotating stars is associated with the region where the field lines are bunched up , and not with the corotation radius .
( 4 ) x - winds are driven by the _
centrifugal force _ , and as a result matter flows over a wide range of directions below the dead zone " ( shu et al . 1994 ; ostriker & shu 1995 ) . in conical winds
the matter is driven by the
_ magnetic force _
( lovelace et al . 1991 ) which acts such that the matter flows into a _ thin shell _ with a cone half - angle @xmath92 .
the same force tends to collimate the flow . for rapidly rotating stars in the propeller regime where @xmath93 and where the condition for bunching , @xmath94 , is satisfied we find two distinct outflow components ( 1 ) a relatively low - velocity conical wind and ( 2 ) a high - velocity axial jet .
a significant part of the disc matter and angular momentum flows into the conical winds . at the same time a significant part of the rotational energy of the star flows into the magnetically - dominated axial jet .
this regime is particularly relevant to protostars , where the star rotates rapidly and has a high accretion rate .
the star spins down rapidly due to the angular momentum flow into the axial jet along the field lines connecting the star and the corona . for typical parameters a protostar spins down in @xmath95 years .
the axial jet is powered by the spin - down of the star rather than by disc accretion .
the matter fluxes into both components ( wind and jet ) strongly oscillate due to events of inflation and reconnection .
most powerful outbursts occur every @xmath96 months .
the interval between outbursts is expected to be longer for smaller diffusivities in the disc .
outbursts are accompanied by higher outflow velocities and stronger self - collimation of both components .
such outbursts may explain the ejection of knots in some cttss every few months .
when the artificial requirement of symmetry about the equatorial plane is dropped , mhd simulations reveal that the conical winds may alternately come from one side of the disc and then the other even for the case where the stellar magnetic field is a centered axisymmetric dipole ( lovelace et al .
2010 ) . in recent work
we have studied the disc accretion to rotating magnetized stars in the propeller regime using a new code with very high resolution in the region of the disc ( lii et al .
2013 ) . in this code
_ no _ turbulent viscosity or diffusivity is incorporated , but instead strong turbulence occurs due to the magneto - rotational instability .
this turbulence drives the accretion and it leads to episodic outflows .
the authors thank g. v. ustyugova and a. v. koldoba for the development of the codes used in the reported simulations .
this research was supported in part by nsf grants ast-1008636 and ast-1211318 and by a nasa atp grant nnx10af63 g ; we thank nasa for use of the nasa high performance computing facilities .
alpar , m.a . , & shaham , j. 1985 , nature , 316 , 239 donati , j .- f . , jardine , m. m. , gregory , s. g. , petit , p. , paletou , f. , bouvier , j. , dougados , c. , mnard , f. , cameron , a. c. , harries , t. j. , hussain , g. a. j. , unruh , y. , morin , j. , marsden , s. c. , manset , n. , aurire , m. , catala , c. , alecian , e. 2008 , mnras , 386 , 1234 | we discuss recent progress in understanding the launching of outflows / jets from the disc - magnetosphere boundary of slowly and rapidly rotating magnetized stars . in most of the discussed models
the interior of the disc is assumed to have a turbulent viscosity and magnetic diffusivity ( as described by two `` alpha '' parameters ) , whereas the coronal region outside of the disc is treated using ideal magnetohydrodynamics ( mhd ) .
extensive mhd simulations have established the occurrence of long - lasting outflows in both the cases of slowly and rapidly rotating stars .
( * 1 * ) in the case of _ slowly rotating stars _ , a new type of outflow , _ a conical wind _ , is found and studied in simulations .
the conical winds appear in cases where the magnetic flux of the star is bunched up by the inward motion of the accretion disc . near their region of origin , the winds have the shape of a thin conical shell with a half - opening angle of @xmath0 . at large distances
these outflows become magnetically collimated by their toroidal magnetic field and form matter dominated jets .
that is , the jets are current carrying . about @xmath1 of the disc matter flows from the inner disc into the conical winds .
the conical winds may be responsible for episodic as well as long - lasting outflows in different types of stars .
the predominant driving force for the conical winds is the magnetic force proportional to the negative gradient of the square of the toroidal magnetic field and not the centrifugal force .
( * 2 * ) in the case of _ rapidly rotating stars _ in the `` propeller regime , '' a two - component outflow is observed .
the first component is similar to the matter dominated conical winds .
a large fraction of the disc matter may be ejected into the winds in this regime .
the second component is a high - velocity , low - density magnetically dominated _ axial jet _ where matter flows along the opened polar field lines of the star .
the axial jet has a mass flux about @xmath2 that of the conical wind , but its energy flux due to the poynting flux can be larger than the energy flux of the conical wind .
the jet s angular momentum flux is carried by the magnetic field and causes the star to spin - down rapidly .
such propeller - driven outflows may be responsible for the jets in protostars and for their rapid spin - down .
when the artificial requirement of symmetry about the equatorial plane is dropped , we show that the conical winds may alternately come from one side of the disc and then the other even for the case where the stellar magnetic field is a centered axisymmetric dipole .
recent mhd simulations of disc accretion to rotating stars in the propeller regime have been done with _ no _ turbulent viscosity and _ no _ diffusivity .
the strong turbulence we observe is due to the magneto - rotational instability .
this turbulence drives accretion in the disc and leads to episodic outflows . |
a natural question about the hamiltonian used in the main text is whether errors might be induced by writing the quantum spin hall edge as spin - polarized along some direction , with right - moving electrons having spin up and left - moving electrons having spin down .
it is often emphasized that the quantum spin hall edge survives even when no spin direction is conserved , and that the edge should be viewed as a `` kramers pair '' of states conjugate under the time - reversal operation .
this supplement reviews what it means to say that the edge does not have a well - defined spin direction , and why in principle that could lead to a reduction in the tunneling amplitude @xmath105 , although the hamiltonian will still have the same form . in practice ,
the degree of spin polarization at the quantum spin hall edge is thought to be high , and @xmath105 is usually taken as a measured quantity rather than computed theoretically , but the considerations here may be useful for other problems .
first , it is simple to show that the form of the hamiltonian remains correct even if the edge is not made up of spin eigenstates .
consider the action of the tunneling hamiltonian that takes the low - lying right - moving state into some superposition in the two - level system made of the spin states of one electron on the dot : @xmath106 the action of the tunneling hamiltonian on the left - moving state is then obtained by acting with the time - reversal operation : @xmath107 where we have used the convention that the time - reversal operator is @xmath108 , with @xmath96 denoting complex conjugation
. then an @xmath109 unitary transformation of spin basis on the dot @xmath110 gives the form assumed in the main text , with @xmath111 : @xmath112 the tunneling term thus remains the same .
it is also easy to see that even though the interaction term on the dot @xmath113 is not invariant under the transformation , the additional terms can be readily absorbed by a redefinition of @xmath27 .
therefore , the full hamiltonian has exactly the same form if the helical edge is not fully spin - polarized , albeit with ( slightly ) renormalized coefficients . in particular
, the coupling to the effective magnetic impurity in the kondo limit remains isotropic even in the presence of spin - orbit coupling .
a similar conclusion was reached in @xcite in a different language .
this result seems perhaps counterintuitive if the edge is not fully spin - polarized : the dot with one electron is a two - level system and hence can be interpreted as a maximally polarized spin - half state along some direction . to give a physical picture of what is the `` rotated '' spin basis on the dot when the edge state is not a spin eigenstate to start with
, it may be simplest to work with a simple example of a non - spin - polarized bloch state and think about a photoemission or tunneling experiment .
consider for simplicity the following ( not yet normalized ) bloch state of crystal momentum zero , obtained by superposing a spin - up gaussian centered on locations @xmath114 , @xmath115 , with a spin - down gaussian cented on sites
@xmath116 : @xmath117 as @xmath118 , this becomes the uniform state with @xmath119 and spin state @xmath120 + @xmath121 , i.e. , perfectly polarized along the @xmath122 spin direction .
however , at finite @xmath123 the expansion of the above wavefunction over plane waves ( momentum eigenstates ) , as would be detected in an idealized photoemission experiment , becomes more interesting : the expansion of the above wavefunction in plane waves is @xmath124 where @xmath125 is a normalization constant .
we see that the plane - wave expansion now includes all momenta differing from the crystal momentum ( zero ) by a reciprocal lattice vector , and that plane waves with odd @xmath126 are spin - polarized along @xmath127 .
the absolute normalization can be set by letting @xmath128 at @xmath129 ( a single plane wave appears ) .
then , @xmath130 will be consistently normalized for all @xmath123 if @xmath131 now the meaning of the statement that this state is not spin - polarized is evident : while @xmath132 is a pure quantum state , it involves multiple momenta , and consequently a measurement that only probes spin will give probabilities consistent with a mixed spin state rather than a pure spin state .
( in modern parlance the state shows `` entanglement '' between the spin and momentum hilbert spaces . )
the average spin along the @xmath133 direction is @xmath134 which drops from 1 at @xmath129 , to 0.17 at @xmath135 , to zero at @xmath136 .
the average of a measurement in the other spin directions is always zero , proving that ( except at @xmath137 where only a single momentum state is involved ) the spin is not fully polarized if momentum is not resolved .
the tunneling problem above can now be analyzed similarly and shows the same effect . to start with ,
assume that the fundamental tunneling is spin - independent : then the magnitude of the effective tunneling strength @xmath105 to the dot can be reduced if the different momenta contained in the bloch state interfere destructively because they tunnel into the dot with opposite spin directions
. if the tunneling has its own spin - orbit coupling , then the details of this interference will be modified depending on the specific momentum dependence of the spin direction induced in the tunneling . to summarize
, the incomplete spin polarization in a bloch state can reduce the total tunneling matrix element @xmath105 , but the form of the hamiltonian in the main text remains valid , and the reduction of @xmath105 is only expected to be large if the spin polarization of the edge state is small , which is not believed to be the case for ( hg , cd)te . | the gapless edge modes of the quantum spin hall insulator form a helical liquid in which the direction of motion along the edge is determined by the spin orientation of the electrons . in order to probe the luttinger liquid physics of these edge states and their interaction with a magnetic ( kondo ) impurity ,
we consider a setup where the helical liquid is tunnel - coupled to a semiconductor quantum dot which is excited by optical absorption , thereby inducing an effective quantum quench of the tunneling . at low energy , the absorption spectrum is dominated by a power - law singularity .
the corresponding exponent is directly related to the interaction strength ( luttinger parameter ) and can be computed exactly using boundary conformal field theory thanks to the unique nature of the quantum spin hall edge .
recent experiments on optical absorption of a quantum dot coupled to a metallic reservoir offer a new window into the correlated electron state that underlies the kondo effect @xcite .
the absorption of a single photon can be treated as a sudden change of the hamiltonian ( a `` quench '' ) : the exchange interaction between the dot and reservoir is abruptly changed by the absorption . over a range of ( shifted ) photon energies between the observation temperature @xmath0 and kondo temperature @xmath1 , a power - law in the absorption spectrum is observed .
this power law is a consequence of the anderson orthogonality catastrophe @xcite : the overlap between two metallic states differing by the presence or absence of a scattering potential goes to zero algebraically in the number of electrons @xmath2 .
it is natural to ask whether other kinds of metals that also support kondo effects can be probed with this type of experiment , and what such an experiment would reveal .
the main goal of this letter is to explain the optical absorption of a few - electron quantum dot in the kondo regime when it is coupled to the helical metal of electrons at the edge of a quantum spin hall effect ( qshe ) droplet .
the qshe edge @xcite is a one - dimensional metal where spin plays a fundamental role : as a result of spin - orbit coupling , there is a single time - reversal - related `` kramers pair '' of low - energy propagating modes , which can be pictured as a right - moving mode of electrons with spin up along some axis and a left - moving mode with spin down .
this edge is robust to disorder and interactions as long as the original symmetries of time - reversal and charge are unbroken @xcite .
when time - reversal is broken by a static magnetic perturbation , the conductance goes to zero and can be computed via integrability of point tunneling in a luttinger liquid @xcite .
the anderson impurity problem , which is our starting point , is more complex as the impurity is dynamical and time - reversal is preserved .
qshe edges have been probed through transport on ( hg , cd)te quantum wells @xcite , inas / gasb quantum wells @xcite , and ( in the former material ) through squid imaging of the generated magnetic flux @xcite . the kondo effect along the qshe edge has previously been studied theoretically for its effects on transport @xcite ( see also _
e.g. _
@xcite for other studies involving transport ) , where it has little effect at least for a single impurity : the dc conductance remains @xmath3 , just as in the absence of the impurity .
this might suggest that kondo effects in the qshe are subtle and difficult to observe . to the contrary , the kondo impurity s effects on optical absorption
are much clearer than in transport : again a power - law is observed in the absorption spectrum , but now the power - law is determined by the interaction strength ( luttinger parameter ) along the edge because special properties of the qshe edge fix a scaling dimension in the field theory of this problem . unlike an ordinary luttinger liquid @xcite , the helical liquid plus kondo impurity can be mapped exactly onto the kondo problem in an ordinary fermi liquid , where the interactions in the qshe edge generate an effective anisotropic exchange coupling .
in fact a different version of this mapping was noted @xcite before the helical liquid was understood to arise in physical systems .
this remarkable property of the qshe edge enables us to solve exactly the absorption problem at low energy . in the same regime realized in existing experiments with fermi liquids , we find that the absorption spectrum shows a power - law tail with an exponent given by the luttinger parameter , thus providing a direct measurement of interaction strength in helical liquids . [
[ physical - setup ] ] physical setup + + + + + + + + + + + + + + we start with the description of the quantum spin hall edge as a helical liquid ( hl ) with counterpropagating modes with opposite spins along some axis , with forward scattering interactions @xcite @xmath4 with @xmath5 a two - dimensional spinor .
we consider interactions at the edge only and ignore bulk interactions that would lead to more exotic behaviors @xcite .
single - particle backscattering terms that would open a gap in the luttinger liquid are not allowed by time - reversal symmetry . following @xcite
, this hl is tunnel - coupled to a semiconductor quantum dot ( qd ) whose charge state is controlled by an external gate voltage @xmath6 .
this gate voltage can be tuned such that the topmost occupied level @xmath7 ( `` valence level '' ) lies far below the fermi energy , so that it can be considered as occupied . on the other hand ,
the conduction level @xmath8 can be considered as unoccupied initially .
we then apply a circularly polarized light beam ( say with polarization @xmath9 ) with frequency @xmath10 .
this will excite an electron with spin @xmath9 from the valence level @xmath7 into the conduction level @xmath8 , thus leaving a positively charged hole behind with spin @xmath11 in the @xmath7 level ( see @xcite for a related protocol in the case of a dot coupled to a fermi reservoir ) .
this effectively induces an attractive coulomb interaction @xmath12 between the excited electron and the hole this can also be thought of as the energy difference between the state with both @xmath8 and @xmath7 levels occupied ( with coulomb repulsion between the electrons ) , and the state with only the @xmath8 level occupied .
this hole is assumed to be stable and static compared to the other time scales of the problem .
let @xmath13 and @xmath14 be the electron / hole creation operators in the @xmath8 and @xmath7 levels , respectively .
the quantum dot / light interaction is thus described by @xmath15 .
[ [ anderson - impurity - model . ] ] anderson impurity model .
+ + + + + + + + + + + + + + + + + + + + + + + + the relevant part of the hamiltonian of the quantum dot is then given by @xmath16 with @xmath17 , @xmath18 , and @xmath19 .
all the energies are measured with respect to the fermi energy @xmath20 and are positive .
the energy cost for creating a hole in the @xmath7 level @xmath21 is assumed to be very large ( compared to the bulk gap ) , so the @xmath7 level can be considered as occupied initially _
i.e. _ , before the light perturbation is turned on . the hole degree of freedom will be integrated out in the following but it is important to keep track of it in order to trigger the quantum quench .
finally , the tunneling between the hl and the qd reads @xmath22 this gives a ( bare ) hybridization width @xmath23 to the @xmath8 level , where @xmath24 is the density of states ( per spin ) of the helical liquid . in the following , we will neglect the two - particle backscattering by the impurity .
the full hamiltonian is @xmath25 where @xmath26 .
we remark that even if the two modes of the helical edge are not fully spin polarized , the coupling to the impurity in the anderson hamiltonian will have the same form @xcite .
we have assumed that @xmath27 does not depend on the spin of the @xmath8 electrons , which can be achieved by adding a term that accounts for an applied bias on the hl .
let us hence assign a different chemical potential @xmath28 to the right- and left - moving electrons @xmath29 .
because the total number of electrons with a given spin is conserved by the hamiltonian , this term can be replaced by an effective magnetic field acting on the dot , thereby inducing an effective zeeman splitting @xmath30 . in the following
, we will assume that this bias term is tuned such that @xmath31 . a genuine magnetic field ( except along the @xmath32 axis )
would open a gap in the hl .
[ [ absorption - spectrum - and - loschmidt - echo . ] ] absorption spectrum and loschmidt echo .
+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + we remark that the hamiltonian conserves @xmath33 , so we can easily integrate out the hole degree of freedom . using fermi s golden rule , the absorption spectrum of the photons can be expressed as @xmath34 where @xmath35 is a proportionality constant that will depend on the precise experimental setup .
the labels @xmath36 correspond to the initial and final hamiltonians @xmath37 with @xmath38 , and @xmath39 is the initial thermal density matrix of the system . with the parameters described previously ( in particular , @xmath40 ) , the @xmath8 level can initially be considered as completely empty and decoupled from the helical liquid , and the absorption of the photon induces an effective quantum quench of the tunneling between the dot and the helical liquid .
we can thus write the eigenstates of @xmath41 as product states over the helical liquid and the unoccupied dot @xmath42 .
the quantity @xmath43 then corresponds to the distribution of the work done during a quantum quench @xcite starting from a decoupled hl at temperature @xmath0 ( supposed to be smaller than all the other energy scales ) and a qd in the state @xmath44 , with the quench corresponding to suddenly turning on the tunneling between the helical liquid and the dot ( see fig .
[ fig1 ] ) .
the fourier transform of the work distribution is known @xcite as the loschmidt echo @xmath45 where @xmath46 refers to a thermal average over the initial density matrix @xmath47 , with the dot in the state @xmath44 .
we shall show in the following that the large time behavior of this loschmidt echo can be computed exactly .
[ [ from - anderson - to - kondo . ] ] from anderson to kondo .
+ + + + + + + + + + + + + + + + + + + + + + + because we are dealing with a strongly interacting many - body problem , the real - time dynamics of the system after the quantum quench is extremely complicated , and is controlled by several different energy scales . in what follows , we will restrict ourselves initially to the particle - hole symmetric case @xmath48 of the anderson impurity model , and consider the case @xmath49 .
we will also shift the frequencies by @xmath50 , the groundstate energy difference between @xmath51 and @xmath52 , which corresponds to the minimal work needed at zero temperature to perform the quantum quench ( @xmath53 ) .
let us assume that we are in a frequency regime such that @xmath54 , so that one can effectively integrate out the charge degrees of freedom on the dot to go from the anderson problem to a reduced kondo setup @xcite although the initial quantum dot setup is crucial to trigger physically the quantum quench using optical absorption .
the resulting effective hamiltonian density at low energy is @xcite @xmath55 with @xmath56 the local spin on the dot .
( there is no potential scattering term induced at the symmetric point . ) in terms of the anderson model parameters , the bare kondo coupling reads @xmath57 at the particle - hole symmetric point .
this tunneling term should be considered energy - dependent and will be renormalized by the interactions in the hl .
we also point out that the coupling to the effective kondo impurity remains isotropic even in the presence of spin - orbit coupling @xcite ( see also @xcite ) .
[ [ bosonization . ] ] bosonization .
+ + + + + + + + + + + + + to proceed , we bosonize the hl electrons by introducing @xmath58 . using standard bosonization formulas ,
one obtains a ( spinless ) luttinger liquid hamiltonian for the hl @xmath59 $ ] where we have introduced the non - chiral boson @xmath60 , and its dual @xmath61 with @xmath62 .
the renormalized velocity reads @xmath63 , and the luttinger liquid parameter is given by @xmath64 . using standard canonical transformations and taking the scaling limit
, the bosonized form of the kondo interaction is @xmath65 , mixing right and left - movers @xmath66 .
this hamiltonian is purely chiral in terms of the left - moving fields @xmath67 @xcite .
the odd boson then decouples , while the chiral hamiltonian for @xmath68 reads @xmath69 this is the bosonized form of the one - channel anisotropic kondo effect in a fermi liquid .
put differently , the kondo problem in a helical liquid can be mapped onto the usual kondo effect in a fermi liquid , but the coulomb interactions in the hl induce an effective anisotropy in the kondo coupling after this transformation @xcite .
this mapping is due to the unique nature of the hl , and does not apply to the case of a kondo impurity in the usual spinful luttinger liquid @xcite .
it is worth pointing out at this point that up to another canonical transformation , eq .
is related to the interacting resonant level model , a problem that has attracted a lot of attention recently due to the development of exact methods out of equilibrium @xcite .
computed from bcft ( with @xmath70 in this example ) .
inset : finite - temperature smearing of the edge singularity . ]
[ [ low - energy - kondo - physics . ] ] low energy kondo physics .
+ + + + + + + + + + + + + + + + + + + + + + + + + the impurity perturbation has dimension @xmath71 , so it is always relevant for repulsive coulomb interactions .
at lowest - order , the kondo temperature reads @xmath72 . a more precise form of the kondo temperature in our problem is given by @xcite @xmath73 with @xmath74 the bandwidth and @xmath75 the coulomb - induced anisotropy parameter .
we expect this formula to be valid only at low energy compared to the dot coulomb interaction @xmath76 . in the absence of interaction in the qsh
helical liquid ( @xmath77 ) , one recovers the well - known form of the kondo temperature in the anderson / kondo problem .
we will focus on the low - temperature regime @xmath78 that is appropriate for the existing experiments with fermi liquid leads . in the crossover regime
where @xmath79 are comparable to @xmath1 but still less than @xmath76 , numerical renormalization group methods might be useful to obtain the intermediate behavior .
[ [ zero - temperature - edge - singularity . ] ] zero temperature edge singularity .
+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + in the zero temperature limit @xmath80 , there is no energy scale remaining in the problem so we can tackle this low energy ( large time @xmath81 ) limit using boundary conformal field theory ( bcft ) . folding the chiral anisotropic kondo hamiltonian , the kondo impurity then becomes a boundary condition on a non - chiral boson . at low energy , this flows to a conformally invariant boundary condition and the loschmidt echo can be argued to be related to the two - point function of a boundary condition changing operator ( bcc ) @xcite ( see also @xcite ) .
this operator @xmath82 has dimension @xmath83 so we recover the @xmath84 phase shifts of the electrons in the usual case of noninteracting leads ( @xmath77 ) as @xmath85 @xcite , while the value for @xmath86 can be interpreted as twice the dimension of the spin operator in the ising model @xcite .
the loschmidt echo thus behaves as @xmath87 at large time . in terms of the absorption spectrum
, this means that we expect the following edge singularity @xmath88 the low energy part of the absorption spectrum thus contains a clear signature of the luttinger physics of the hl , very different from what one would obtain in the case of an ordinary ( spinful ) luttinger liquid @xcite . for @xmath77 ,
one recovers the @xmath89 behavior observed experimentally for a fermi liquid reservoir @xcite .
[ [ finite - temperature - crossover . ] ] finite temperature crossover .
+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + at non - zero temperature @xmath90 , the loschmidt echo can still be thought of as a two - point function of bcc operators .
indeed , introducing @xmath91 with the normalization @xmath92 , one can recast as @xmath93 , where the time evolution is performed with the initial hamiltonian @xmath51 .
this finite temperature two - point function can readily be computed using a conformal mapping @xmath94 with @xmath1 playing the role of a uv cutoff .
the absorption spectrum can then be obtained by fourier transform ( see fig .
[ fig2 ] ) .
the edge singularity is smeared at finite temperature , so it is important to work in the regime @xmath95 to have the power - law behavior .
however , we emphasize that as long as @xmath90 , the full finite temperature crossover is captured by cft , providing a wider frequency range to measure the exponent @xmath96 . in that low energy regime , the finite temperature absorption spectrum could also be used to test non - equilibrium fluctuation theorems @xcite .
[ [ discussion . ] ] discussion .
+ + + + + + + + + + + our proposal to measure the interaction strength along the qshe edge using optical absorption requires one to work in the kondo regime @xmath97 . for
estimated ( hg , cd)te qd and hl parameters @xmath98 mev , @xmath99 mev and @xmath100 mev , one finds @xmath101 k depending on the luttinger parameter @xmath96 , with larger values of @xmath1 corresponding to strong coulomb interactions .
the luttinger liquid nature of the helical liquid thus makes it easier to reach the kondo regime ( large @xmath1 with @xmath102 ) , which was already accessed experimentally for a fermi liquid reservoir @xcite . for the temperatures used in the ( hg , cd)te quantum well experiment @xcite
, one expects to have @xmath103 so that the low energy bcft results should hold . to our knowledge
few - electron quantum dots have not yet been created in this material , so it may be more practical to use inas / gasb quantum wells @xcite , in which the qshe also survives to higher temperatures .
although we have derived eq . at the particle - hole symmetric point @xmath48 , we expect this formula to remain valid whenever the dot is approximately occupied by one electron only @xcite , in part because potential scattering terms @xmath104 that would change the exponent in eq
. are not allowed by time - reversal symmetry .
our results suggest that optical absorption could be a reasonable alternative to transport in order to probe the edge physics of topological phases of matter
. it would be interesting to see whether similar quantum dot setups could provide new insights on the physics of topological kondo systems @xcite , or help in probing majorana modes at the edge of topological superconductors @xcite .
[ [ acknowledgments . ] ] acknowledgments .
+ + + + + + + + + + + + + + + + this work was supported by the quantum materials program of lbnl ( r.v ) and nsf dmr-1206515 ( j.e.m . ) .
we thank j. dahlhaus and d. goldhaber - gordon for insightful discussions .
rv also wishes to thank d. kennes , v. meden and h. saleur for collaborations on related matters .
99 h. e. treci _
et al . _ ,
phys .
rev .
lett . * 106 * , 107402 ( 2011 ) . c. latta _ et al .
_ , nature * 474 * , 627630 ( 2011 ) .
p.w .
anderson , phys .
rev .
lett . * 18 * , 10491051 ( 1967 ) .
c.l . kane and e.j .
mele , phys .
rev .
lett . * 95 * , 226801 ( 2005 ) .
c.l . kane and e.j .
mele , phys .
rev .
lett . * 95 * , 146802 ( 2005 ) . c. wu , b.a . bernevig and s - c .
zhang , phys .
rev .
lett . * 96 * , 106401 ( 2006 ) . c. xu and j.e .
moore , phys .
rev .
b * 73 * , 045322 ( 2006 ) .
r. ilan , j. cayssol , j.h .
bardarson and j.e .
moore , phys .
rev .
lett . * 109 * , 216602 ( 2012 ) .
b. a. bernevig , t. l. hughes , and s. c. zhang , science * 314 * , 1757 ( 2006 ) .
m. knig _
et al . _ , science * 318 * , 766 ( 2007 ) .
i. knez , r .-
r .
du , and g. sullivan , phys .
rev .
lett .
* 107 * , 136603 ( 2011 ) .
k.c .
nowack _
et al . _ , nature materials * 12 * , 787791 ( 2013 ) .
j. maciejko , c. liu , y. oreg , x - l .
qi , c. wu and s - c .
zhang , phys .
rev .
lett .
* 102 * , 256803 ( 2009 ) .
y. tanaka , a. furusaki and k.a .
matveev , phys .
rev .
lett . * 106 * , 236402 ( 2011 ) .
b.l .
altshuler , i.l .
aleiner and v.i .
yudson , phys .
rev .
lett * 111 * , 086401 ( 2013 ) .
c .- y .
hou , e .- a .
kim and c. chamon , phys .
rev .
lett .
* 102 * , 076602 ( 2009 ) .
g. dolcetto , s. barbarino , d. ferraro , n. magnoli and m. sassetti , phys .
rev .
b * 85 * , 195138 ( 2012 ) . j
r .
souquet and p. simon , phys .
rev .
b * 86 * , 161410(r ) ( 2012 ) . y .-
l .
lee and y .- w .
lee , phys .
rev .
b * 88 * , 035112 ( 2013 ) .
d - h .
lee and j. toner , phys .
rev .
lett . * 69 * , 33783381 ( 1992 ) .
a. schiller and k. ingersent , phys .
rev .
b * 51 * , 46764679 ( 1995 ) .
m. hohenadler , t. c. lang , and f. f. assaad , phys .
rev .
lett . * 106 * , 100403 ( 2011 ) .
e .- g .
moon , c. xu , y.b .
kim and l. balents , phys .
rev .
lett . *
111 * , 206401 ( 2013 ) .
r.w .
helmes , m. sindel , l. borda and j. von delft , phys .
rev .
b * 72 * , 125301 ( 2005 ) .
m. heyl and s. kehrein , phys .
rev .
b * 85 * , 155413 ( 2012 ) .
a. silva , phys .
rev .
lett .
* 101 * , 120603 ( 2008 ) .
m. heyl and s. kehrein , phys .
rev .
lett . * 108 * , 190601 ( 2012 ) .
see supplemental material .
a. hewson , _ the kondo problem to heavy fermions , cambridge studies in magnetism _ ( cambridge university press , cambridge , england , 1997 ) .
l. shekhtman , o. entin - wohlman and a. aharony , phys .
rev .
lett . * 69 * , 836 ( 1992 ) .
i. affleck and a.w.w .
ludwig , j. phys . a : math
. gen .
* 27 * , 5375 ( 1994 ) .
p. fendley , a.w.w .
ludwig and h. saleur , phys .
rev .
lett . * 74 * , 3005 ( 1995 ) .
b. doyon , phys .
rev .
lett . * 99 * , 076806 ( 2007 ) .
e. boulat and h. saleur , phys .
rev .
b * 77 * , 033409 ( 2008 ) . e. boulat , h. saleur and p. schmitteckert , phys .
rev .
lett .
* 101 * , 140601 ( 2008 ) .
r. vasseur , k. trinh , s. haas and h. saleur , phys .
rev .
lett . * 110 * , 240601 ( 2013 ) .
j - m .
stphan and j. dubail , j. stat .
mech .
( 2011 ) p08019 .
i. affleck , nucl .
phys .
b * 336 * , 517532 ( 1990 ) .
m. heyl and s. kehrein , phys .
rev .
lett . * 108 * , 190601 ( 2012 ) .
b. bri and n.r .
cooper , phys .
rev .
lett . * 109 * , 156803 ( 2012 ) .
m. cheng , m. becker , b. bauer and r.m .
lutchyn , _ interplay between kondo and majorana interactions in quantum dots _
, arxiv:1308.4156 . * supplementary material * |
understanding living cells at a systemic level is an increasingly important challenge in biology and medicine @xcite .
regulatory interactions between intracellular molecular agents ( e.g. dna , rna , proteins , hormones , trace elements ) , form so - called _ genetic regulatory networks _ ( grn ) , which orchestrate gene expression and replication , coordinate metabolic activity , and cellular development , respond to changes in the environment , or stress .
grn coordinate regulatory dynamics on all levels from cell - fate @xcite to stress response @xcite .
qualitative understanding of grn topology is for instance obtained from promoter sequences @xcite , gene - expression profiling @xcite or protein - protein interactions ( proteome ) @xcite .
however qualitative information on grn topology alone is insufficient to understand grn dynamics .
it has been recognized that quantitative information is required to understand the complex dynamical properties of regulatory interactions in living cells @xcite , mainly because dynamics on interaction networks with identical topology still depends on the strength of interactions ( links ) between agents ( nodes ) .
models of grn dynamics aid the task of understanding properties of grn at various levels of detail available in experimental data and therefore provide valuable tools for integrating information from different sources into unifying pictures and for reverse engineering grn from experimental data .
any model should _ adequately _ reproduce grn dynamics and _ sufficiently _ exhibit systemic properties of the grn , including homeostasis , multi - stability , periodic dynamics , alternating activity , self - organized critical dynamics ( soc ) and differentiability
. _ homeostatic dynamics _ regulates the equilibrium concentration levels of agents , e.g. @xcite , _ multi - stability _ shows switching between multiple steady states @xcite .
examples for _ periodic dynamics _ are e.g. the cell - cycle @xcite , circadian - clock @xcite , i@xmath0b - n@xmath0b signaling @xcite , her dynamics @xcite etc .
some molecular agents show _ alternating activity _ , i.e. their concentrations alternate between being detectable ( on ) and below detection threshold ( off ) , see e.g. @xcite .
_ self - organized critical _ ( soc ) dynamics corresponds to details of regulatory dynamics ensuring ( approximate ) stability within a fluctuating environment through various mechanisms of adaptation .
finally the property of _ differentiability _ means that cells of multicellular organisms can differentiate into various cell - types ( liver , muscle , blood , kidney , cancer , ... ) .
the differentiated cells possess identical grn but express distinguishable patterns of regulatory activity .
the same grn therefore can be expressed in different _ modes _ so that some agents become expressed in one mode but not in another @xcite .
recently it has been reported that both regulation of transcription and mrna decay rates ( i.e. the mrna turnover ) are necessary to understand experimentally observed expression values @xcite .
moreover it has been demonstrated that decay rates of mrna are cell - type specific @xcite .
analogously for proteins , where the dominant mechanism is the ubiquitin driven proteolyse in the proteasom @xcite , protein abundance and therefore their degradation has to be tightly controlled @xcite .
also the abundance of proteins and whether certain proteins are produced or not is again cell - type specific @xcite .
this indicates that decay - rates and their control play a crucial role in cell - differentiation .
variable decay rates however and the property of differentiability are hardly ever considered in grn models where decay rates of agents are usually kept constant .
understanding the effects of changes of decay rates of agents therefore is a crucial step towards a deeper understanding of grn dynamics and the role decay rates play in cell - differentiation .
the grn is the set of all possible interactions of molecular reactions and bindings .
the grn captures all possible features of cells and are responsible for the immense levels of adaptation characteristic to living systems .
what happens when different cell - types express the same grn in alternative ways ? at any point in time only small subsets of the grn are active .
any active subset of the grn leads to the expression of particular sets of molecules ( expression modes ) .
the _ active regulatory network _ at time @xmath1 is the regulatory sub - network of the grn , governing the molecular ( auto - catalytic ) dynamics of all agents which exist at time @xmath1 .
the set of existing molecules forms the _ active agent set _ at time @xmath1 .
the active network changes over time and typical sequences of active sets represent what we call the _ expression modes _ of a specific cell - type and their cell - cycle .
expression modes themselves can be modified , either locally as a reaction to an external signal , or fundamentally through further cell differentiation .
active sets of molecules are transient and what is observed in experiments is a superposition of subsequent active sets , which we call the _ expressed set of agents _ and the regulatory interactions between the expressed agents the _ expressed regulatory network_. to find the property of differentiability in a regulatory network model therefore requires that one network is capable of producing different expression modes while perturbations ( external signal ) only modify active sets locally and the particular expression mode can be restored .
the six dynamical properties we have listed above have been addressed with a variety of conceptually different models .
the essence of all these models is that they try to capture the dynamics induced by positive and negative feed back loops within the grn .
the choice of model depends largely on the type and resolution ( coarse graining ) of experimental data . at the single cell level cellular activity ( e.g. concentrations @xmath2 of biochemical agents @xmath3 )
can be modeled by non - linear ( stochastic ) differential equations @xcite which can explain homeostasis , periodic and multi - stable behavior .
the dynamics governed by a grn is given by a set of coupled non - linear differential equations @xmath4 where @xmath5 is a ( non - linear ) function capturing the grn .
it depends on the vector of concentrations of all the possible @xmath6 molecular agents in a cell , @xmath7 .
@xmath8 is the time derivative of the concentrations @xmath2 .
note that @xmath5 can have stochastic components .
analysis of such systems is often complicated by the interplay between fluctuations and non - linearities @xcite .
differential equation models can be approximated by cellular automata , boolean or piecewise - linear models .
the property of soc dynamics , or dynamics at the `` edge of chaos '' @xcite , has been studied mainly in the context of cellular automata and boolean models @xcite .
soc dynamics was also discussed in continuous differential equation based models @xcite .
boolean and piecewise - linear models share common origins in the work of glass and kauffman , @xcite , and have extensively been used for modeling and analyzing grn @xcite . for their superior properties in approximating non - linear systems ( in principle to any suitable precision ) piecewise - linear models also are applied in different disciplines , for instance for modeling highly non - linear electronic circuits @xcite . in the context of grn both boolean and piecewise - linear models
usually are used for describing non - linear dynamics with switch - like regulatory elements frequently observed in biological regulatory processes @xcite .
such switches react if the concentration of an agent ( the signal ) crosses a specific threshold level . to model such switches in regulation networks of @xmath6 molecular agents with concentrations @xmath2 the space of concentrations @xmath9 is cut into segments defined by the threshold values where the concentration @xmath2 can trigger a regulatory switch .
these segments are called _ regulatory domains _ ( e.g. @xcite ) . in each such domain eq .
( [ nlde ] ) gets approximated by a linear equation of the form @xmath10 where the @xmath11 are production rates and @xmath12 are interaction matrices between agents . if @xmath13 , then @xmath14 promotes the production of @xmath3 .
if @xmath15 , then @xmath14 suppresses @xmath3 . if @xmath16 @xmath14 has no influence on @xmath3
the diagonal elements @xmath17 are _ decay rates _ , @xmath18 .
non - linear effects purely come from concentration @xmath2 passing threshold levels , where the dynamics of @xmath19 switches from one to another regulatory domain with different values of @xmath20 and @xmath12 .
equation ( [ linearisation ] ) is a slight generalization of the glass - kauffman plm , @xcite , where @xmath16 except for the ( usually ) fixed decay rates @xmath21 , so that only the production rates change with the regulatory domain . given that the interaction matrix @xmath22 of the regulatory network is invertible ( which is almost certainly true for the biologically relevant range of connectivities of grn ) eq . ( [ linearisation ] )
can be rewritten @xmath23 with @xmath24 being the solution of the equation @xmath25 .
the fixed - point @xmath24 is stable ( unstable ) and @xmath2 will be attracted ( repelled ) by @xmath26 .
if @xmath24 is stable and @xmath27 for all @xmath3 then @xmath28 is a stationary solution of eq .
( [ linearisation ] ) .
not all models approximating nonlinear differential equation descriptions of grn are equally suited to capture all grn properties discussed above simultaneously depending on whether discrete ( boolean , cellular automata ) or smooth ( differential equation ) features dominate the model .
however there exists a surprisingly simple class of models which exhibits _ all _ desired grn properties . here
we present such a simple model that captures all of the above dynamical properties .
we find that the alternating dynamics plays a key role for the stability of regulatory systems and for the formation of soc dynamics in particular @xcite .
most importantly we are able to show that even unspecific control over decay rates , changing the magnitude of all decay rates simultaneously by a ( small ) factor , leads to `` cell differentiation '' , i.e. the same regulatory network enters different expression modes , displaying different sequences of active regulatory networks .
we show that experimental facts , linking variations of decay rates observed between different cell - types of an organism to variations of the abundance of intra - cellular biochemical agents in these cell - types , correspond to ( a ) differences in the _ expressed _ genetic regulatory network , and ( b ) these differences can be controlled via decay rates of intracellular agents .
in other words typical expression modes ( cyclical sequences of successive active sub - networks of the grn ) can be altered and switched by controlling decay rates .
glass - kauffman systems , @xcite , produce positive concentrations @xmath29 for all times @xmath1 given positive initial conditions @xmath30 .
this however makes it impossible to produce alternating activity of agents since zero - concentrations @xmath31 can not appear .
therefore we have to generalize glass - kauffman systems to more general forms of invertible interaction matrices @xmath12 where the positivity of solutions of eq .
( [ linearisation ] ) is not implicitly guaranteed , but where positivity ( non - negativity ) is ensured as a constraint to the system , @xmath32 this constraint alters the linear dynamics of eq .
( [ linearisation ] ) in the following way .
whenever a concentration @xmath2 becomes zero at time @xmath1 then @xmath33 remains zero for @xmath34 for as long as @xmath35 , according to eq .
( [ linearisation ] ) . if @xmath36 for @xmath37 then @xmath38 is no longer subject to the positivity constraint and continues to evolve according to eq .
( [ linearisation ] ) again .
agent @xmath3 is said to be _ active _ at time @xmath1 , if @xmath29 and _ inactive _ , if @xmath31 .
the positivity constraint eq .
( [ eq : pc ] ) implies the following consequences . at any point in time
there will be a sub - set of agents with non - vanishing concentrations which we call the _ active set _ of agents .
the remaining agents have zero concentration , and therefore do not actively influence the concentrations of any of the non - vanishing agents .
there exist @xmath39 different active sets , i.e. @xmath39 combinations in which @xmath6 agents can be active or inactive .
each active set can be uniquely identified by an index @xmath40 .
in the course of time @xmath1 some agents will vanish while others re - appear , so that one effectively observes a sequence of sets of active agents @xmath41 @xmath42 being the initial active set .
the active set @xmath43 switches to active set @xmath44 at time @xmath45 . in each time interval
@xmath46 $ ] of duration @xmath47 it is thus possible to only consider the regulatory sub - network acting on the set of active agents @xmath44 .
this sub - network is described by the part of the full interaction matrix @xmath12 , where @xmath3 and @xmath14 are restricted to the set of active agents @xmath44 .
these sub - matrices we call _ active networks _ and denote them by @xmath48 . the concentration vector of active agents we call @xmath49 .
active agents also `` feel '' a modified _ effective _ fixed point @xmath50 , such that finally for @xmath51 the concentrations of the active agents follow a linear equation @xmath52 we refer to such systems as _ sequentially linear _ systems .
the attractiveness of this description arises through the fact that it becomes possible to understand the dynamics by considering the sequences of active networks @xmath53 which allows to analyze dynamical properties in terms of eigenvalues and eigenvectors of the active sub - matrices @xmath48 ( see materials and methods ) .
this model can be shown to be mathematically equivalent to @xcite .
@xmath54 in the picture of sequentially linear dynamics it becomes possible to identify operational modes of a cell as a particular sequence of active networks .
cell types in ordinary operational modes may be classified by specific sequences . as a hypothetical example a liver cell under typical conditions
might be characterized by a _ periodic _
sequence @xmath55 , whereas an endothelial cell is given by @xmath56 .
note that all types share the same full regulatory network @xmath22 .
this separates timescales of the dynamics : on the fast timescale the dynamics is continuous and characterized by linear changes of the concentrations @xmath2 . on the slower time - scale the dynamics is characterized by discrete changes of active sets . the change from one sequence of active sets to another
can be interpreted as the expression modes of different cell - types ( cell differentiation ) and we show that changes in decay rates of molecular species trigger switches between expression modes . as an example for sequentially linear dynamics we consider a system with @xmath57 molecular agents , @xmath58 , @xmath59 for all agents @xmath60 , and a regulatory network given by @xmath61 and the fixed point @xmath62 for all agents @xmath3 simulated with time - increment @xmath63 .
periodic time - series organized into a sequence of four domains with three different active sets . for each time - domain the associated spectrum of eigenvalues for the active sets is shown in ( b ) . in ( c ) a 3d poincare map of the limit cycle is plotted together with the projection of @xmath24 in the center .
the domains are marked with bold numbers and switching events with dots .
( d ) the eigenvalue spectra of the different subsystems are plotted in the imaginary plane .
the shift of the spectrum along the real axis depending on the decay rate @xmath64 is indicated .
[ figure_main1 ] , title="fig:",width=384 ] + and the fixed point @xmath62 for all agents @xmath3 simulated with time - increment @xmath63 . periodic time - series organized into a sequence of four domains with three different active sets . for each time - domain the associated spectrum of eigenvalues for the active sets is shown in ( b ) .
in ( c ) a 3d poincare map of the limit cycle is plotted together with the projection of @xmath24 in the center .
the domains are marked with bold numbers and switching events with dots .
( d ) the eigenvalue spectra of the different subsystems are plotted in the imaginary plane .
the shift of the spectrum along the real axis depending on the decay rate @xmath64 is indicated .
[ figure_main1 ] , title="fig:",width=230 ] and the fixed point @xmath62 for all agents @xmath3 simulated with time - increment @xmath63 .
periodic time - series organized into a sequence of four domains with three different active sets . for each time - domain the associated spectrum of eigenvalues for the active sets is shown in ( b ) . in ( c ) a 3d poincare map of the limit cycle is plotted together with the projection of @xmath24 in the center .
the domains are marked with bold numbers and switching events with dots .
( d ) the eigenvalue spectra of the different subsystems are plotted in the imaginary plane .
the shift of the spectrum along the real axis depending on the decay rate @xmath64 is indicated .
[ figure_main1 ] , title="fig:",width=153 ] the dynamics of this system ( over one period ) is shown in fig .
[ figure_main1 ] a. the property describing the stability of an active set @xmath44 is the maximal real part of the eigenvalues @xmath65 of the active matrix @xmath48 denoted @xmath66 .
the number @xmath67 denotes the number of time - domains in a periodic sequence of active networks and @xmath68 is the number of different sub - networks that are activated in a sequence ( see also materials and methods ) . in this example
there are four time - domains ( @xmath69 ) associated with three different active sets ( @xmath70 ) which are periodically repeated .
the sequence starts in time - domain @xmath71 with active set @xmath72 with maximum real eigenvalue @xmath73 .
positive @xmath74 means that the fixed point of the active set is unstable and the associated leading eigenvalue implies that the concentration of one agent ( green ) is decaying to zero .
the positivity condition deactivates this agent as its concentration becomes zero and the system enters time - domain @xmath75 as the active set switches to @xmath76 with @xmath77 .
negative @xmath78 means that the fixed point @xmath79 is stable and @xmath80 tries to approach @xmath81 .
this leads to the deactivated agent ( green ) becoming produced again and the system switches back to @xmath72 entering the third time - domain . in time - domain
@xmath82 the initial conditions differs from the one in time - domain @xmath71 and a different node ( magenta ) gets deactivated .
the system switches to @xmath83 with @xmath84 at the beginning of the fourth time - domain .
this means @xmath81 is a stable fixed - point and the inactive node ( magenta ) eventually gets produced again as the system switches back to the beginning ( @xmath72 ) and enters the next period .
the system is thus precisely characterized by the sequence @xmath85 .
the eigenvalue spectra of the sub - matrices @xmath48 associated with subsequent time - domains @xmath86 are shown in fig .
[ figure_main1 ] b. fig .
[ figure_main1 ] c shows a projection of the trajectory into a three dimensional poincare map .
[ figure_main1 ] d shows the eigenvalue spectra of the different active sub - systems of the dynamics .
.*properties of 4-node example system*. some characteristics of the four node system shown in fig [ figure_main1 ] are listed , including the index of the time domain , the index of the sub - system @xmath87 , the number of active nodes @xmath88 , the time the system spends in the @xmath87th sub - system , the real - part of the leading eigenvalue of @xmath87 , and whether sub - system @xmath87 is stable or not .
[ cols="^,^,^,^ , > , > " , ] some details of the dynamics , like the existence of multiple stable fixed - points , the periodicity of bounded attractors and temporal self - organization , can be mathematically fully understood . in @xcite
it was already shown mathematically that sequentially linear models exhibit _ homeostasis _ , _ multi - stability_. this has been demonstrated for a wide range of system size @xmath6 , and a number of interactions ( connectivity ) and fixed decay rates . _
periodic dynamics _ , and _ self - organized critical _ dynamics have been noted in @xcite but were not clarified and require further explanation which is given in detail in the materials and methods , where also a simple temporal balance condition is described and derived .
the temporal balance condition states that the time - average over the real parts of the leading eigenvalues @xmath89 of the matrices @xmath48 in a sequence of active networks approximate the lyapunov exponent @xmath90 .
the lyapunov exponent @xmath90 measures the overall stability of a system ( @xmath91 stable , @xmath92 instable , @xmath93 critical ) and for sequences following a periodic attractor @xmath90 can be shown to be exactly zero . inserting the values for @xmath94 and @xmath95 from table ( [ table_1 ] ) into the balance condition , eq .
( [ eq : balance ] ) gives the value @xmath96 as an approximation of @xmath90 ( which has an exact value of zero ) .
although the balance equation gives only a crude approximation of the lyapunov exponent it allows to understand why the example - system spends more time in the weakly instable time - domain @xmath71 and @xmath82 , than in the stable time - domains @xmath75 and @xmath97 which is obviously true from fig .
[ figure_main1 ] .
strong convergence needs less time to compensate for weak divergence .
temporal balance is a consequence of the mechanism of self - organization that fine - tunes switching times such that stable parts of the dynamics compensate instable parts of the dynamics exactly .
this mechanism can be understood in the following way .
sequentially linear systems try to converge to a fixed point .
if it is reached the system becomes static .
the fixed point might not be _ accessible _ however , meaning that the trajectory on the way toward the fixed point hits a boundary ( fig .
[ figure_main1 ] c ) causing a switching event which changes the dynamics so that the system now is attracted by a different effective fixed point , which it tries to reach .
if the system does not converge to an accessible fixed point it is either unstable and some concentrations @xmath2 diverge , or the system circles through some of the @xmath39 possible active sets and converges onto an effective attractor - characterized in the sequence of active networks . in the later case small perturbations of @xmath98 on the attractor will vanish with time . this allows to show that bounded dynamics that does not converge to a fixed - point has to be periodic ( materials and methods ) .
switching times are not static but react to perturbations of concentrations @xmath2 .
perturbations shift the occurrence of switching times proportional to the magnitude of the perturbation .
this has the effect that switching events act like sliding `` focal planes '' allowing the perturbed dynamics to `` refocus '' onto the periodic attractor .
while the perturbed dynamics returns to the attractor switching times cumulate small time - shifts resulting in a phase - shift of the periodic dynamics .
a perturbation is remembered as a phase - shift of the periodic dynamics which neither grows exponentially nor dies out .
the lyapunov exponent therefore is zero and the systems self - organizes to the `` edge of chaos '' by adaptation of switching times .
stable adaptive dynamics is a result of this `` temporal self - organization '' . for all agents and fixed point concentrations @xmath99 $ ] .
correlation coefficients for simulated and measured time - series are @xmath100 for time larger @xmath101 and agents @xmath3 in order of the legend .
the model simulation uses zero concentrations for all agents as initial condition and a time increment @xmath102 . for matching the simulation with experiment time in the model
is shifted by @xmath103 .
[ figure_main2 ] , width=384 ]
we first show that the model is able to explain actual empirical data , including alternating dynamics .
figure [ figure_main2 ] shows data of molecular concentrations @xmath104 ( her@xmath105 ( black ) , pol ii ( red ) , trip1 ( blue ) , hdac1 ( green ) ) over three periods of about 40 minutes time .
these four agents are all part of the human estrogen nuclear receptor dynamics .
the source of the data is metivier et .
al . @xcite .
data points were taken from pigolotti et al . @xcite and the actual values of the matrix elements @xmath106 are bests fits with identical decay rates for optimal explanation of the data .
the trip1 data ( blue ) shows _ alternating activity _ which is reproduced perfectly by our sequential linear model . in the following we show how the change of decay rates induces changes from one cell - type to another . in particular we show how changes of the overall strength of the decay rates results in differentiated dynamics , i.e. in distinct sequences of active expressed networks .
this allows to understand recent experimental observations which indicate correlations between cell - type , expressed sets of agents , and decay - rates @xcite . for a fixed interaction network temporal self - organization
can be maintained for a wide range of decay rates @xmath64 .
we show this in the same @xmath97-node system considered in fig .
[ figure_main1 ] by only varying the decay rate @xmath107 from eq .
( [ eq : matrix ] ) . of the four node system ,
( [ eq : matrix ] ) , is shown in ( a ) as a function of the decay rate @xmath64 , which exhibits a `` plateau '' with @xmath93 in the range @xmath108 . in
( b ) the length @xmath67 of the periodic sequence of domains is plotted in green triangles and the number of different active sets @xmath68 as red squares
. in ( c ) the sequences of active sets are shown for decay rates @xmath109 , @xmath110 and @xmath111 .
the limit circles for decay rates @xmath109 ( short sequence ) and @xmath112 ( long sequence ) are visualized in ( d ) in a poincare map using three out of four phase - space dimensions . with decreasing @xmath64
the radius of the limit circle becomes wider and additional sets ( marked with colors ) become active . in ( e ) the spectra of eigenvalues are shown for all the appearing active sets with @xmath112 .
[ figure_main3 ] , title="fig:",width=384 ] + of the four node system , eq . ( [ eq : matrix ] ) , is shown in ( a ) as a function of the decay rate @xmath64 , which exhibits a `` plateau '' with @xmath93 in the range @xmath108 . in
( b ) the length @xmath67 of the periodic sequence of domains is plotted in green triangles and the number of different active sets @xmath68 as red squares . in ( c ) the sequences of active sets are shown for decay rates @xmath109 , @xmath110 and @xmath111 .
the limit circles for decay rates @xmath109 ( short sequence ) and @xmath112 ( long sequence ) are visualized in ( d ) in a poincare map using three out of four phase - space dimensions . with decreasing @xmath64
the radius of the limit circle becomes wider and additional sets ( marked with colors ) become active . in ( e ) the spectra of eigenvalues are shown for all the appearing active sets with @xmath112 .
[ figure_main3 ] , title="fig:",width=230 ] of the four node system , eq . ( [ eq : matrix ] ) , is shown in ( a ) as a function of the decay rate @xmath64 , which exhibits a `` plateau '' with @xmath93 in the range @xmath108 . in
( b ) the length @xmath67 of the periodic sequence of domains is plotted in green triangles and the number of different active sets @xmath68 as red squares . in ( c )
the sequences of active sets are shown for decay rates @xmath109 , @xmath110 and @xmath111 .
the limit circles for decay rates @xmath109 ( short sequence ) and @xmath112 ( long sequence ) are visualized in ( d ) in a poincare map using three out of four phase - space dimensions . with decreasing @xmath64
the radius of the limit circle becomes wider and additional sets ( marked with colors ) become active . in ( e ) the spectra of eigenvalues are shown for all the appearing active sets with @xmath112 .
[ figure_main3 ] , title="fig:",width=153 ] figure [ figure_main3 ] a shows the lyapunov exponent @xmath90 as a function of @xmath64 .
a plateau , where @xmath113 , is clearly visible .
if the decay rate is larger than a critical value @xmath114 , the lyapunov exponent becomes negative ( @xmath91 ) and the system stable .
if the decay rate is smaller than a critical value of @xmath115 , temporal balance can not be achieved any more , refocusing breaks down , and the system becomes chaotic and trajectories diverge exponentially with @xmath92 . in fig .
[ figure_main3 ] b the length of the periodic sequences @xmath67 ( green triangles ) , which is the number of time - domains in a sequence , and the number @xmath68 of different active sets activated in this sequence ( red squares ) is depicted .
figure [ figure_main3 ] b also shows that at several critical values of @xmath116 in the plateau region the sequences of active regulatory sub - networks changes when temporal balance can no longer be established merely by adapting the switching times of a sequence .
sequences do not usually change completely at critical values of @xmath64 and are only expanded by additional active subsets .
this can be seen clearly in the 3d poincare map of the dynamics fig .
[ figure_main3 ] d , where the sequence of subsystems @xmath87 given by @xmath117 ( for @xmath109 ) gets expanded to the sequence @xmath118 ( for @xmath112 ) . in the materials and methods , fig .
( 1 ) , the longer sequence is also shown in the space of all possible active sets .
the mathematical reason why such critical decay rates exist is that changes of @xmath64 shift the eigenvalue spectra of the active interaction matrix @xmath119 , shown in fig .
[ figure_main3 ] e , along the real axis .
the real part of the leading eigenvalues , @xmath95 , is becoming smaller ( larger ) than zero and @xmath120 becomes an attractor ( repellor ) of @xmath121 .
the stable fixed point then either is accessible and the dynamic changes from periodic to stationary or inaccessible and the dynamic changes qualitatively but remains periodic . which agents become active in a given active set @xmath87
is depicted in fig .
[ figure_main3 ] b for three different sequences of active sets associated with three different ranges of the decay rate @xmath64 indicated by gray lines .
if node @xmath3 is active in active set @xmath87 then the associated field is white and black otherwise .
and @xmath122 and identical initial conditions for all values of @xmath64 . (
a ) the lyapunov exponent , ( b ) the number of active sets @xmath68 in a period ( if @xmath123 then the the sequence is not periodic but a steady state ! ) , and ( c ) the fraction of expressed nodes are plotted as functions of the uniform decay rates @xmath124 . for @xmath125 @xmath24
is stable . in the range @xmath126
the @xmath24 has become unstable but the plateau ( @xmath93 ) can not form since the dynamic finds active sets @xmath87 with stable and accessible @xmath120 .
the inset in ( b ) shows that in the plateau region a small window , @xmath127 , exists where again an active set @xmath87 contains an accessible @xmath120 attracting the dynamics . in the range @xmath128
the plateau forms and dynamics gets periodic . for @xmath129
the system gets unstable .
[ figure_main4 ] , width=384 ] the number of _ expressed _
agents @xmath130 is the number of agents that are active at least once during a period of the dynamics . to demonstrate that not only the periodic activation of agents depends on @xmath64 but also the number of expressed nodes @xmath130 itself
, we consider a larger sequentially linear system with @xmath131 agents .
the interaction matrix of the system is a random matrix with average connectivity @xmath132 , meaning for each node @xmath133 interactions with other agents have been randomly chosen with equal probability .
each non - zero entry , describing such an interaction , is drawn from a normal distribution with mean zero and a standard deviation of @xmath134 .
this means that the interaction strength is of magnitude @xmath71 on average and has positive or negative sign with equal probability . in fig .
[ figure_main4 ] a the lyapunov exponent @xmath90 , in fig .
[ figure_main4 ] b the number @xmath68 of sets that become active during a cycle and in fig .
[ figure_main4 ] c the fraction of expressed agents @xmath135 is plotted as a function of @xmath64 . for large decay rates ( @xmath125 )
the system is stable and @xmath24 is a fixed - point of the dynamics .
as @xmath64 decreases @xmath24 becomes unstable for @xmath136 .
however for @xmath137 the system ends up in some stable accessible fixed point @xmath120 so that @xmath98 approaches a stationary state and @xmath123 . in this range
@xmath130 increases with @xmath64 .
the @xmath138 plateau with stable self - organized critical dynamics ( @xmath139 ) only emerges in the range @xmath128 where number of active sets @xmath68 and expressed network size @xmath135 vary strongly .
@xmath135 varies between @xmath71 and @xmath140 which means that changes of the decay rate can induce changes of the size of the expressed network comparable to the magnitude of the full interaction network .
a small window of stability exists for @xmath127 ( see inset ) .
the strong dependence of @xmath135 on the decay - rate @xmath64 ( up to @xmath141 of the total regulatory network ) demonstrates clearly that decay - rates alone massively influence sequences of active systems without changing the interaction strength between agents in the regulatory network at all .
moreover , decay rates can also cause switches between fixed - point dynamics and periodic dynamics . while fixed points favor larger decay - rates ( in the example @xmath142 ) there can also exist fixed points for smaller decay rates ( window of stability @xmath127 ) where systems favor periodic dynamics .
we presented a model which de - composes the dynamics of molecular concentrations governed by the full molecular regulatory networks into a temporal sequence of active sub - networks .
this novel type of model allows not only to reduce the vast complexity of the full regulatory network into sub - networks of managable size but further to approximate the complicated dynamics by linear methods .
the intrinsic non - linearities in the system which lead to alternating dynamics in concentrations ( as found in countless experiments ) are absorbed into switching events , where the dynamics of one linear system switches to another one . in this view
different cell types correspond to different sequences of active sub - networks over time .
these sequentially linear models allow not only for the first time to describe all the relevant dynamical features of the gnr ( homeostasis , multi - stability , periodic dynamics , alternating activity , differentiability , and self - organized criticality ) , but also offers the handle to understand the role of molecular decay rates .
the fact that sequentially linear dynamics properly models homeostasis , multi - stability and periodic behavior was shown in @xcite .
here we have shown how self - organized criticality ( lyapunov exponent self - regulates to zero ) arises as a consequence of temporal balance of switching events .
this requires agents to show alternating activity ( being repeatedly on and off ) , which is a natural property by construction of sequentially linear models , and which has posed an unresolved problem of previous models such as the glass - kauffman @xcite model and its many variants .
the mechanism behind self - organized criticality is based on adaptive switching times which effectively lead to refocusing of perturbed dynamics onto the attractor of sequences of active sub - networks .
such a temporal self - organization causes long time memory of perturbations in terms of phase - shifts of the otherwise unchanged periodic dynamics , causing the lyapunov exponent to become zero . in other words
slight perturbations , e.g. noise , only cause time - shifts of the sequence of regulatory reactions but do not change the underlying sequence .
perturbations are `` remembered '' by the system by non vanishing phase - shifts and the dynamics gets `` refocused '' onto the periodic attractor merely accumulating a time - shift .
this has the consequence that the lyapunov exponent is zero and the system self - organizes its criticality by adapting switching - times .
practically this means that a system balances the time it spends in its active sub networks with stable and unstable dynamics ( temporal balance ) .
applying the sequentially linear model to the problem of cell - differentiation we demonstrate that different levels of decay rates are one to one related with transitions from one active sub - network sequence ( cell type ) to another .
this might be a key ingredient to understand a series of recent experimental facts reported on the role of decay - rate regulation systems and the role of noise in cell differentiation @xcite .
we found that by varying the decay rates only , while keeping the complete regulatory network fixed over time , substantially modifies the temporal organization of regulatory events .
in particular the decay rate controls the number of expressed agents , the sequence of active sub - networks , and sometimes even the type of solution ( stable , stationary , periodic ) .
the changes occur at critical levels of decay rates and changes can be drastic . for example we find situations where a 5% variation of the decay rate causes an approximate doubling of the number of expressed agents .
this demonstrates that different expression modes , which distinguish different cell - types from each other , can be very efficiently obtained by controlling the decay rates of agents without altering any interactions between agents in the regulatory network , which is very costly in an evolutionary sense .
these findings highlight the importance of intracellular decay rate control mechanisms and the role of noise in cell differentiation .
for system shown in article fig .
( 3 ) . in set
@xmath71 all @xmath143 , yellow background stand for complex leading eigenvalues of the active interaction matrix .
black indicates that the agent associated with that index is not active .
the gray lines indicate to all possible switching events where the number of active agents @xmath144 changes @xmath145 .
blue arrows mark the observed sequence of the dynamics for the examples eq .
( 8) with @xmath146 .
[ fig - systree ] , width=384 ] the eigenvalues @xmath147 and eigenvectors @xmath148 of a matrix @xmath22 are defined as solutions of the matrix equation @xmath149 .
the solution of a linear differential equation @xmath150 is of the form @xmath151 . for large times
the @xmath98 will therefore point into the direction of the eigenvector @xmath152 with the eigenvalue @xmath153 with the largest real part and @xmath154 as @xmath1 gets large .
if the largest real part of @xmath153 is larger ( smaller ) than zero @xmath155 will grow ( decay ) exponentially and @xmath24 is an unstable ( stable ) fixed point of the differential equation . let @xmath156 be the maximal real part of the leading eigenvalue of the active interaction matrix @xmath119 associated with the active subset @xmath87 .
the effective fixed point @xmath120 is _ stable _ and perturbations of concentrations vanish if @xmath157 .
the fixed point is _ accessible _ if @xmath121 approaching @xmath120 does not cause a switching event and _ inaccessible _ otherwise .
stationary solutions of a sequentially linear system therefore require fixed points that are both stable and accessible .
suppose a bounded attractor exists for a sequentially linear system with @xmath6 agents @xmath3 .
the perturbation @xmath158 at time @xmath159 also effects later switching times of agents @xmath3 , i.e. @xmath160 such that @xmath161 for some constant @xmath162 , where @xmath163 . since @xmath164 sufficiently fast as @xmath165 ( there exists an attractor ) the cumulated time shift @xmath166 of switching times remains finite for all times .
this shows that the perturbed @xmath167 behaves ( after some time ) just like the unperturbed @xmath19 only shifted in time .
perturbation neither vanishes nor grow exponentially , and the lyapunov exponent can only be zero ( @xmath93 ) .
moreover , since the number of active sets is finite ( @xmath39 ) and the dynamics is bounded the concentrations have to return to values on the attractor with arbitrary precision within some finite return - time .
the remaining concentration difference can be seen as a perturbation so that the attractor can only be a periodic cycle .
the time - shift produces a phase - shift of the periodic dynamics .
while eigenvalues tell us something about the stability of a fixed point the lyaponov exponent @xmath90 tells something about the stability of the dynamics @xmath98 itself .
the lyapunov exponent @xmath168 measures how a small perturbation @xmath169 grows with time . if @xmath91 the perturbation vanishes exponentially with time or grow exponentially if @xmath92 .
system with @xmath92 are chaotic ( in - stable dynamics extremely sensitive to noise or perturbations ) while @xmath91 indicates stable dynamics insensitive to perturbations and noise .
systems with @xmath93 are special as their dynamics is sensitive to noise and perturbations without `` overreacting '' like chaotic systems .
these systems at the `` edge of chaos '' adapt to fluctuations but remain close to their unperturbed dynamics .
here we derive a simple approximation of the lyapunov exponent of sequentially linear dynamics which explains temporal self - organization quantitatively .
this is necessary for understanding why switching in general happens between active networks with stable and unstable dynamics and not from one stable stable ( unstable ) to another stable ( unstable ) active network .
qualitative analysis of bounded attractors of sequentially linear dynamics has shown that the attractor is periodic and the lyapunov exponent @xmath93 .
characteristic information on the dynamics gets encoded by periodic sequences @xmath170 , @xmath171 with a period of some length @xmath67 such that @xmath172 and @xmath173 ( for large enough @xmath174 ) as in the example shown in fig .
( [ figure_main1 ] ) in the main article . if the dynamics of the system would remain in an active network @xmath119 the lyapunov exponent would be identical with the largest real part @xmath95 of the eigenvalues of @xmath22 .
the lyapunov exponent @xmath90 of the sequentially linear system therefore is well approximated or into the direction of the leading possibly complex eigenvector , if @xmath120 is unstable , remains incomplete since convergence is always interrupted by a switching event .
] by the time average over @xmath89 , i.e. @xmath175 since the dynamics is periodic the time average only needs to be taken over one period and since @xmath93 one gets @xmath176 for @xmath177 large enough .
the `` refocusing '' mechanism discussed above qualitatively therefore is also `` balancing '' the times @xmath178 specific active sets @xmath44 remain active by fine tuning switching times approaches zero consistently as the time increment is made smaller and orbits become periodic again . ]
such that contributions from time - domains with stable ( @xmath179 ) and unstable dynamics ( @xmath180 ) compensate each other .
_ temporal balance _ and _ refocusing _ are two aspects of the temporal self - organizing principle manipulating switching times .
this work has been funded by the _ forum integrativmedizin _ an initiative of the _ hilde umdasch privatstiftung_. hood l , heath jr , phelps me , lin b ( 2004 ) systems biology and new technologies enable predictive and preventative medicine . _
gavin ac , aloy p , grandi p , krause r , boesche m , et al . ( 2006 ) proteome survey reveals modularity of the yeast cell machinery .
_ nature _ 440:631636 .
kashtan n , alon u ( 2005 ) spontaneous evolution of modularity and network motifs .
_ proc natl acad sci usa _ 102:1377313778 .
mtivier r , penot g , hbner mr , reid g , brand h , et al .
( 2003 ) estrogen receptor- directs ordered , cyclical , and combinatorial recruitment of cofactors on a natural target promoter . _ cell _ 115:751763 .
ciechanover a ( 2006 ) intracellular protein degradation : from a vague idea thru the lysosome and the ubiquitin - proteasome system and onto human diseases and drug targeting . _
exp biol med _ 231:11971211 .
bossi a , lehner b ( 2009 ) tissue specificity and the human protein interaction network .
_ mol syst biol _ 5:260 .
burkard tr , planyavsky m , kaupe i , breitwieser fp , brckstmmer t , et al .
( 2011 ) initial characterization of the human central proteome .
_ bmc syst biol _ 5:17 .
kauffman sa ( 1993 ) _ the origins of order- self - organization and selection in evolution_. ( oxford university press ) .
bhattacharjya a , liang s ( 1996 ) power - law distributions in some random boolean networks .
_ phys rev lett _ 77:16441647 .
rewieski m , white j ( 2003 ) a trajectory piecewise - linear approach to model order reduction and fast simulation of nonlinear circuits and micromachined devices .
_ ieee / acm transactions on computer - aided design of integrated circuits and systems _ 252 - 257 .
yagil g , yagil e ( 1971 ) on the relation between effector concentration and the rate of induced enzyme synthesis .
_ biophys j _ 11:1127 .
ptashne m ( 1992 ) _ a genetic switch : phage @xmath90 and higher organisms_. _ blackwell science & cell press_. casey r , de jong h , gouz jl ( 2006 ) piecewise - linear models of genetic regulatory networks : equilibria and their stability .
_ j math biol _ 52:2756 . | systemic properties of living cells are the result of molecular dynamics governed by so - called genetic regulatory networks ( grn ) .
these networks capture all possible features of cells and are responsible for the immense levels of adaptation characteristic to living systems . at any point in time only small subsets of these networks are active .
any active subset of the grn leads to the expression of particular sets of molecules ( expression modes ) .
the subsets of active networks change over time , leading to the observed complex dynamics of expression patterns .
understanding of this dynamics becomes increasingly important in systems biology and medicine .
while the importance of transcription rates and catalytic interactions has been widely recognized in modeling genetic regulatory systems , the understanding of the role of degradation of biochemical agents ( mrna , protein ) in regulatory dynamics remains limited .
recent experimental data suggests that there exists a functional relation between mrna and protein decay rates and expression modes . in this paper
we propose a model for the dynamics of successions of sequences of active subnetworks of the grn .
the model is able to reproduce key characteristics of molecular dynamics , including homeostasis , multi - stability , periodic dynamics , alternating activity , differentiability , and self - organized critical dynamics .
moreover the model allows to naturally understand the mechanism behind the relation between decay rates and expression modes .
the model explains recent experimental observations that decay - rates ( or turnovers ) vary between differentiated tissue - classes at a general systemic level and highlights the role of intracellular decay rate control mechanisms in cell differentiation . |
the case that the jets in low - luminosity , fri radio galaxies @xcite are initially relativistic and decelerate to sub - relativistic speeds on kiloparsec scales rests on several independent lines of evidence , as follows . 1 .
proper motions corresponding to speeds comparable with or exceeding @xmath4 have been observed directly on pc and kpc scales in fri jets @xcite .
2 . fri sources must be the side - on counterparts of at least a subset of the bl lac population , in which evidence for bulk relativistic flow on pc scales is well established ( e.g. @xcite ) .
fri jets show side - to - side asymmetries which decrease with distance from the nucleus , most naturally explained by doppler beaming of emission from an intrinsically symmetrical , decelerating , relativistic flow @xcite , in which case the brighter jet is approaching us
. there is continuity of sidedness from pc to kpc scales and superluminal motion , if observed , is in the brighter jet .
the lobe containing the brighter jet is less depolarized than the counter - jet lobe , consistent with faraday rotation in the surrounding halo of hot plasma if the brighter jet is approaching @xcite .
theoretical work has demonstrated that initially relativistic jets can decelerate from relativistic to sub - relativistic speeds without disruption provided that they are not too powerful , the surrounding halo of hot plasma providing a pressure gradient large enough to recollimate them @xcite .
we have developed a method of modelling jets on the assumption that they are intrinsically symmetrical , axisymmetric , relativistic , decelerating flows . by fitting to deep radio images in total intensity and linear polarization
, we can determine the three - dimensional variations of velocity , emissivity and magnetic - field ordering .
we first applied the technique to the jets in the radio galaxy 3c31 ( * ? ? ? * lb ) . with minor revisions ,
the same method proved capable of modelling the jets in b20326 + 39 and b21553 + 24 ( * ? ? ?
* cl ) and in the giant radio galaxy ngc315 ( * ? ? ?
we used the derived velocity field for 3c31 , together with _
observations of the hot plasma surrounding the parent galaxy , to derive the run of pressure , density , mach number and entrainment rate along its jets via a conservation - law analysis @xcite .
we also examined adiabatic models for 3c31 in detail @xcite .
we now present a model for the jets in 3c296 , derived by fitting to new , sensitive vla observations at 8.5ghz . in section [ obs ]
, we introduce 3c296 and summarize our vla observations and their reduction .
images of the source at a variety of resolutions are presented in section [ descrip ] , where we also describe the distributions of spectral index , rotation measure and apparent magnetic field derived by combining our new images with lower - frequency data from the vla archive .
we briefly recapitulate our modelling technique in section [ model ] and compare the observed and model brightness and polarization distributions in section [ results ] .
the derived geometry , velocity , emissivity and field distributions are presented in section [ physical ] .
we compare the result of model fitting for the five sources we have studied so far in section [ sourcecomp ] and summarize our results in section [ summary ] .
we adopt a concordance cosmology with hubble constant , @xmath5 = 70@xmath6 , @xmath7 and @xmath8 .
spectral index , @xmath9 , is defined in the sense @xmath10 and we use the notations @xmath11 for polarized intensity and @xmath12 for the degree of linear polarization .
we also define @xmath13 , where @xmath14 is the flow velocity .
@xmath15 is the bulk lorentz factor and @xmath16 is the angle between the jet axis and the line of sight .
for our modelling technique to work , both radio jets must be detectable at high signal - to - noise ratio in linear polarization as well as total intensity , separable from any surrounding lobe emission , straight and antiparallel .
3c296 satisfies all of these criteria , although the lobe emission is relatively brighter than in the other sources we have studied .
this is a potential source of error , as we shall discuss .
3c296 was first imaged at 1.4 and 2.7ghz by @xcite .
vla observations showing the large - scale structure at 1.5ghz and the inner jets at 8.5ghz were presented by @xcite and @xcite respectively .
the large - scale radio structure of 3c296 has two radio lobes with well - defined outer boundaries and weak diffuse `` bridge '' emission around much of the path of both jets , at least in projection @xcite .
3c296 is therefore more typical of a `` bridged twin - jet '' fri structure @xcite than the `` tailed twin - jet '' type exemplified by 3c31 and other sources we have studied so far .
@xcite point out that the `` lobed , bridged '' fri population is more abundant in a complete sample of low - luminosity radio galaxies than the `` plumed , tailed '' population .
it is therefore particularly interesting to compare the inferred properties of the jets in 3c296 with those of the sources we have modelled previously .
of the two lobes , that containing the counter - jet depolarizes more rapidly with decreasing frequency between 1.7 and 0.6ghz @xcite , consistent with the idea that it is further away from us .
the parsec - scale structure , imaged by @xcite , is two sided , but brighter on the same side as the kpc - scale jet emission .
non - thermal x - ray and ultraviolet emission have been detected from the brighter jet @xcite and the nucleus @xcite .
the source is associated with the giant elliptical galaxy ngc5532 ( @xmath17 ) , the central , dominant member of a small group @xcite .
the linear scale is 0.498kpcarcsec@xmath18 for our adopted cosmology .
the galaxy has a slightly warped nuclear dust ellipse in pa @xmath19 with an inclination of @xmath20 assuming intrinsic circularity ( * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* parameters are taken from the last reference ) .
x - ray emission from hot plasma associated with the galaxy was detected by _
rosat _ and imaged by _
chandra _ @xcite .
new , deep vla observations of 3c296 at 8.5ghz are presented here .
we were interested primarily in the inner jets , and therefore used a single pointing centre at the position of the nucleus and the maximum bandwidth ( 50mhz ) in each of two adjacent frequency channels . in order to cover the full range of spatial scales in the jets at high resolution , we observed in all four vla configurations .
we combined our new datasets with the shorter observations described by @xcite , which were taken with the same array configurations , centre frequency and bandwidth . in order to correct the observed @xmath21-vector position angles at 8.5ghz for the effects of faraday rotation and to estimate the spectrum of the source
, we also reprocessed earlier observations at 4.9ghz and 1.4 1.5ghz ( l - band ) .
the 4.9-ghz data were taken in the a configuration . at l - band , we used two a - configuration observations ( the later one also from @xcite ) and the combined b , c and d - configuration dataset from @xcite .
a journal of observations is given in table [ tab : journal ] .
.journal of vla observations . @xmath22 and @xmath23 are the centre frequencies and bandwidth , respectively , for the one or two frequency channels observed and t is the on - source integration time scaled to an array with all 27 antennas operational .
references to previous publications using the same data : 1 @xcite , 2 @xcite .
[ cols="^,<,<,<,^ , < " , ] the angle to the line of sight from our model fits is @xmath24degree .
the axis of the nuclear dust ellipse therefore differs from that of the jets by @xmath25 in inclination and @xmath26 in the plane of the sky @xcite .
the orientation of the dust is such that the counter - jet would be on the receding side of the galaxy if the dust and jet axes are even very roughly aligned .
note , however , that @xcite suggest that dust ellipses such as that seen in 3c296 are preferentially aligned with the major axes of the galaxies and statistically unrelated to the jet orientations .
the outermost isophote of the jet emission is well fitted by our assumed functional form .
the jets flare to a maximum half - opening angle of 26@xmath1 at a distance of 8.3kpc from the nucleus , thereafter recollimating .
the outer region of conical expansion starts at @xmath27 = 15.8kpc from the nucleus and has a half - opening angle of @xmath28 .
the outer envelope of the model jet emission is shown in fig . [
fig : profiles](a ) . note that our models fit to faint emission outside the 5@xmath29 isophote of 20@xmath30jybeam@xmath18 at 0.75-arcsec resolution .
contours of the derived velocity field are shown in fig .
[ fig : velcont ] .
the positions of the velocity regions , with their uncertainties , and the profiles of velocity along the on - axis and edge streamlines are plotted in figs [ fig : profiles](b ) , ( d ) and ( f ) . the initial on - axis velocity
is poorly constrained .
the best - fitting value , @xmath31 , is very uncertain , and the upper limit is unconstrained by our @xmath32 analysis .
the reason for this is that the fractional edge velocity @xmath33 is very low : the emission from the main jet close to the brightening point is dominated by slower material .
emission from any high - speed component is doppler - suppressed and therefore makes very little contribution to the overall @xmath32 .
in contrast , after the jets decelerate the on - axis velocity is well - constrained and consistent with a constant value of @xmath34 .
the deceleration is extremely rapid in the best - fitting model ( fig .
[ fig : profiles]d ) , but the uncertainties in the boundary positions are also large . the low fractional edge velocity at larger distances , @xmath35 , is required to fit the transverse sidedness - ratio profile ( fig .
[ fig : transside ] ) and is well - determined .
the fact that the sidedness ratio is less than unity in places implies that there is some intrinsic asymmetry ( section [ badfit ] ) , but the qualitative conclusion that the edge velocity is low remains valid .
x - ray and ultraviolet emission is detected between 2 and @xmath3610arcsec from the nucleus @xcite , corresponding to 1.2 5.9kpc in the frame of the jet .
this corresponds to the bright part of the flaring region , up to and possibly including the rapid deceleration .
we have evaluated the jet / counter - jet sidedness ratio outside the modelled region by integrating flux densities in boxes of size 10arcsec along the jet axis and @xmath3715arcsec transverse to it .
the mean ratio drops from @xmath361.7 between 40 and 70arcsec from the nucleus to @xmath361.05 between 70 and 100arcsec .
it is therefore possible that further deceleration to @xmath38 occurs at @xmath3640kpc in the jet frame . as noted in section [ outline - method
] , however , the jets bend and lobe contamination becomes significant , so we can not model the velocity variation in this region with much confidence .
the positions of the emissivity regions , with their uncertainties , and the profiles of emissivity along the on - axis and edge streamlines are plotted in figs [ fig : profiles](c ) , ( e ) and ( g ) . as in all of the fri sources we have examined so far ,
the on - axis emissivity profile of 3c296 is modelled by three power - law sections separated by short transitions .
close to the nucleus , before the brightening point at 1.8kpc , the power - law index , @xmath39 .
the errors on @xmath40 are estimated by eye because the jets are faint and any change to the index has very little effect on the value of @xmath32 , even over a small area close to the nucleus .
we model the brightening point as a discontinuous increase in emissivity by a factor of @xmath41 .
the short region between 1.8 and 2.5kpc is introduced primarily to model the first knots in the main and counter - jets ; its index is poorly constrained because of our inability to fit the main and counter - jet simultaneously : high values fit the counter - jet knot well but overestimate the main jet brightness ; small values have the opposite problem .
the main section of the bright jet base from 2.5 8.9kpc is described by a power law of index @xmath42 .
this joins smoothly onto the outer region ( @xmath43 ) without the need for a rapid transition .
the transverse emissivity profile is fairly flat at the brightening point ( albeit with large uncertainties ) but evolves to a well - defined gaussian with a fractional edge emissivity of 0.12 at large distances .
the inferred magnetic field components are shown as profiles along the on - axis and edge streamlines in fig .
[ fig : bprofs ] and as false - colour plots in fig .
[ fig : bcol ] . in the former figure ,
the shaded areas define the region which the profile could occupy if any one of the six free parameters defining it is varied up to the error quoted in in table [ tab : results ] .
we were able to fit the polarization structure in 3c296 accurately using a model with only toroidal and longitudinal magnetic - field components .
the best fitting model has field - component ratios which vary from the nucleus out to a distance of 14.6kpc , consistent with the boundary between flaring and outer regions , and thereafter remain constant .
the field at the edge of the jet is almost purely toroidal everywhere ( figs [ fig : bprofs]c , [ fig : bcol]a ) , although there is a marginally significant longitudinal component close to the nucleus ( figs [ fig : bprofs]d , [ fig : bcol]b ) . on - axis
, in contrast , the longitudinal field is initially dominant , but decreases over the flaring region to a smaller , but still significant value at larger distances ( figs [ fig : bprofs]b , [ fig : bcol]b ) .
we have described the magnetic field structure as disordered on small scales , but anisotropic .
a field of this type in an axisymmetric model always generates symmetrical transverse profiles of total intensity , degree of polarization and apparent field position angle .
in contrast , a globally - ordered helical field , unless observed at 90@xmath1 to the line of sight in its rest frame , will always generate asymmetric profiles @xcite .
the observed profiles in 3c296 ( figs [ fig : transi ] , [ fig : transpol ] ) are very symmetrical , even in regions where the longitudinal and transverse field components have comparable amplitudes , so we conclude that a simple , globally - ordered field configuration can not fit the data .
our calculations would , however , be unchanged if one of the two field components were vector - ordered ; in particular , a configuration in which the toroidal component is ordered but the longitudinal one is not would be entirely consistent with our results .
the polarization in the high - signal regions closer to the jet axis requires a mixture of longitudinal and toroidal field components , implying in turn that @xmath44 at the edge of the jet with an apparent field along the jet axis .
our models include this emission , but at a level below the blanking criteria in total intensity and linear polarization at 0.75-arcsec resolution ( fig .
[ fig : pcomplo]c ) .
we demonstrated in section [ badfit ] that this emission is indeed present and that it has the correct polarization in the counter - jet . in the main jet , contamination by surrounding lobe emission
confuses the issue , but there is a real discrepancy ( fig.[fig : transpol]g ) : the observed apparent field remains transverse at the edges .
the difference between the main and counter - jets can not be an effect of aberration as we infer velocities @xmath45 in these regions .
there must , therefore , be a significant radial field component at the edge of the main jet which is not present at the corresponding location in the counter - jet . although the low - resolution 8.5-ghz image shown in fig .
[ fig : ivecdeep ] gives the impression that the jet might propagate within its associated lobe while the counter - jet is in direct contact with interstellar medium , the l - band images ( fig .
[ fig : ifull ] ) show diffuse emission surrounding _ both _ jets , so this is unlikely to be the case . ; ( b ) longitudinal @xmath46 . @xmath47 .
[ fig : bcol],width=309 ] given the assumption of flux freezing in a jet without a transverse velocity gradient , the magnetic field components evolve according to : @xmath48 in the quasi - one - dimensional approximation , where @xmath49 is the radius of the jet @xcite .
the field - component evolution predicted by these equations is shown by the dashed lines in fig . [
fig : bprofs ] , arbitrarily normalised to the model values at a distance of 14kpc from the nucleus .
the evolution of the longitudinal / toroidal ratio towards a toroidally - dominated configuration is qualitatively as expected for flux - freezing in an expanding and decelerating flow , but the model shows a much slower decrease . in the absence of any radial field , the ratio should not be affected by shear in the laminar velocity field we have assumed , but the field evolution is clearly more complex than such a simple picture would predict . a small radial field component could still be present , and this would allow the growth of longitudinal field via the strong shear that we infer in 3c296 .
we will investigate this process using the axisymmetric adiabatic models developed by @xcite .
a second possibility is that the flow is turbulent in the flaring region
. two - dimensional turbulence ( with no velocity component in the radial direction ) would lead to amplification of the longitudinal component without generating a large radial component , as required .
after the jets recollimate , the model field ratios have constant values on a given streamline , consistent with the variation expected from flux freezing in a very slowly expanding flow ( fig . [
fig : bprofs ] ) .
note , however , the problems in fitting the main jet polarization at large distances from the axis ( section [ field - structure ] ) .
if the radiating electrons suffer only adiabatic losses , the emissivity is : @xmath50 in the quasi - one - dimensional approximation @xcite .
@xmath51 can be expressed in terms of the parallel - field fraction @xmath52 and the radius @xmath53 , velocity @xmath54 and lorentz factor @xmath55 at some starting location using equation 8 of @xcite : @xmath56^{1/2}\end{aligned}\ ] ] the resulting profiles for the on - axis and edge streamlines are plotted as dashed lines in fig .
[ fig : profiles](e ) and ( g ) , normalized to the model values at the brightening point ( 1.8kpc ) and 14kpc .
their slopes are grossly inconsistent with the model emissivity in the flaring region , but match well beyond @xmath5714kpc essentially where the jet recollimates at @xmath27 = 15.8kpc . as pointed out in section [ images ] , there are discrete , narrow features ( `` arcs '' ) in the brightness distributions of both jets in 3c296 .
strikingly similar structures are seen in the jets of 3c31 ( lb ) , as shown in fig .
[ fig:3c31 ] . in both sources ,
the arcs have systematically different shapes in the main and counter - jets : centre - brightened arcs with well - defined intensity gradients near the jet axis [ type ( i ) ] are found in the main jets of both sources and in the outer counter - jet of 3c296 , while more elongated arcs with well - defined intensity gradients near the jet edges [ type ( ii ) ] are found in both counter - jets .
we will present models of the arc structures elsewhere : here we demonstrate that relativistic aberration can plausibly account for the systematic difference .
suppose that an arc is a thin , axisymmetric shell of enhanced emissivity , concave towards the nucleus , and that at any point its speed is roughly that of the local average flow .
close to the centre - line of the jet , the shell can be approximated as a planar sheet of material orthogonal to the axis in the observed frame .
aberration causes a moving object to appear at the angle to the line of sight appropriate to the rest frame of the flow @xcite . for the the normals to the sheets in the main and counter - jets these angles are @xmath58 and @xmath59 , where : @xmath60^{-1 } \\
\sin\theta_{\rm cj}^\prime & = & \sin\theta [ \gamma(1+\beta\cos\theta)]^{-1 } \\\end{aligned}\ ] ] just as in section [ outline - method ] . in 3c296 ,
the arcs are visible as far as @xmath3690arcsec from the nucleus outside the region we model
so we do not have a direct measure of the average velocity field in their vicinity . given that our model indicates a roughly constant velocity on - axis from 10 40arcsec from the nucleus we will assume for the purposes of a rough estimate that the parameters at the outer edge of the model approximate the average flow parameters near the arcs between 40 and 70arcsec , where the mean jet / counter - jet sidedness ratio remains significantly larger than unity . in fig .
[ fig : aberrate ] we plot @xmath58 and @xmath61 against @xmath62 for @xmath63 ( table [ tab : results ] ) . for a velocity @xmath64 , the angles to the line of sight in the fluid rest frame are @xmath65 and @xmath66 ( the main jet is close to the maximum boost condition @xmath67 which corresponds to edge - on emission in the rest frame ) .
a sheet of enhanced emissivity which is normal to the flow on the jet axis is therefore observed nearly edge - on in the main jet , appearing narrow , with a high brightness gradient . in the counter - jet , on the other hand , the same sheet would appear to be significantly rotated about a line perpendicular to the jet axis in the plane of the sky and therefore less prominent both in intensity and in brightness gradient .
close to the edges of the jets , however , the model flow velocities are essentially zero , so the effects of aberration should be negligible . if the shells are roughly tangential to the surface , then we expect to see narrow brightness enhancements orientated parallel to the edges in both jets .
this picture provides a reasonable qualitative description of the differences in arc structure for the main and counter - jets in 3c296 within @xmath3670arcsec of the nucleus ( fig .
[ fig : i1.5 ] ) .
the evolution of the mean jet / counter - jet ratio suggests that the jets become sub - relativistic at larger distances ( section [ results : vel ] ) and this may explain why we see type ( i ) arcs in both jets between 70 and 90arcsec .
the hypothesis that the main difference in appearance of the arcs between the main and counter - jets is an effect of differential relativistic aberration also allows us to predict that differences should be observed between main and counter - jet arcs in other fri sources if they are orientated at moderate angles to the line of sight .
this explains why we observe systematic differences in the arcs of 3c31 , which has similar model parameters to 3c296 ( fig .
[ fig:3c31 ] , lb ) , but we would not expect differences between the main and counter - jet arcs if @xmath58 and @xmath61 are similar .
the latter condition holds if the jets are close to the plane of the sky ( e.g. 3c449 and pks1333 - 33 ; @xcite ) or slow ( e.g. the outer region of b20326 + 39 ; cl ) .
it will therefore be interesting to search for arcs in deep images of these sources .
we have modelled five fri sources : 3c31 ( lb ) , b20326 + 39 , b21553 + 24 ( cl ) , ngc315 ( clbc ) and 3c296 ( this paper ) .
we will compare and contrast their properties in detail elsewhere , but in this section we summarize their similarities and differences and draw attention to a few potential problems with our approach .
the five fri sources we have studied in detail so far all exhibit the following similarities : 1 . a symmetrical , axisymmetric
, decelerating jet model fits the observed brightness and polarization distributions and asymmetries well .
the outer boundaries of the well - resolved parts of the jets can be divided into two main regions : a _ flaring _ region with a rapidly increasing expansion rate followed by recollimation and an _
region with a uniform expansion rate .
3 . close to the nucleus ,
the on - axis velocity is initially consistent with @xmath68 . on - axis
, rapid deceleration then occurs over distances of 1 10kpc , after which the velocity either stays constant or decreases less rapidly .
all except b21553 + 24 have low emissivity at the base of the `` flaring '' region , so even the main jet is initially very faint ( this region may be present but unresolved in b21553 + 24 ) .
the faint region ends at a `` brightening point '' where the emissivity profile flattens suddenly or even increases .
this , coupled with the rapid expansion of the jets , leads to a sudden increase in surface brightness in the main jets ( not necessarily in the counter - jets , whose emission is doppler suppressed ) .
the brightening point is always _ before _ the start of rapid deceleration , which in turn is completed before the jets recollimate .
immediately after the brightening point , all of the jets contain bright substructure with complex , non - axisymmetric features which we can not model in detail .
optical / ultraviolet or x - ray emission has been detected in all of the jets observed so far with _
chandra _ ( results for b20326 + 39 are not yet available ) .
high - energy emission is associated with the bright regions of the main jets before the onset of rapid deceleration in all cases , and also with the faint inner region in 3c31 .
the longitudinal emissivity profile may be represented by three power - law segments separated by short transitions .
the indices of the power laws decrease with distance from the nucleus .
all of the jets are intrinsically centre - brightened , with fractional emissivities at their edges @xmath360.1 0.5 where these are well - determined .
11 . on average ,
the largest single magnetic field component is toroidal .
the longitudinal component is significant close to the nucleus but decreases with distance .
the radial component is always the smallest of the three .
12 . for those sources we have fit with functional forms allowing variation of field - component ratios across the jets ( 3c31 , ngc315 and 3c296 )
, we find that the toroidal component is stronger relative to the longitudinal component at the edge of the jet ( this is likely to be true for the other two sources )
the evolution of the field component ratios along the jets before they recollimate is not consistent with flux freezing in a laminar flow , which requires a much more rapid transition from longitudinal to transverse field than we infer .
a possible reason for this is two - dimensional turbulence ( with no radial velocity component ) in the flaring regions .
after recollimation , flux freezing is consistent with our results . 14
. the inferred evolution of the radial / toroidal field ratio ( in those sources where the radial component is non - zero ) is qualitatively inconsistent with flux freezing even if shear in a laminar velocity field is included .
the quasi - one - dimensional adiabatic approximation ( together with flux freezing ) is grossly inconsistent with the emissivity evolution required by our models before the jets decelerate and recollimate .
earlier attempts to infer fractional changes in velocity from surface - brightness profiles @xcite , therefore required much more rapid deceleration than we deduce in these regions .
16 . after deceleration and recollimation ,
the emissivity evolution along the jets is described quite well by flux freezing and a quasi - one - dimensional adiabatic approximation ( our analysis does not cover this region in ngc315 ) .
as relativistic effects are slight , the brightness evolution in the outer regions of 3c296 , b20326 + 39 and b21553 + 24 is quite close to the constant - speed , perpendicular - field prediction originally derived by @xcite .
previous estimates of fractional velocity changes derived on the adiabatic assumption [ references as in ( xv ) , above ] are likely to be more reliable on these larger scales . .
the angle between the jet axis and the line of sight is @xmath69 , as we infer for 3c296 .
full line : main jet ; dashed line : counter - jet .
the vertical dotted line marks the on - axis velocity of @xmath70 we derive for the outer parts of the modelled region in 3c296 .
[ fig : aberrate],width=321 ] 3c296 exhibits the following differences from the other sources : 1 .
the transverse velocity profile falls to a low fractional value @xmath71 at the edge of the jets in 3c296 , whereas all the other sources are consistent with an edge / on - axis velocity ratio @xmath360.7 everywhere , and also with a top - hat velocity profile at the brightening point .
this could be related to the differences in lobe structure : as noted earlier , 3c296 is the only source we have observed with a bridged twin - jet structure .
its jets may then be embedded ( almost ) completely within the lobes rather than propagating in direct contact with the interstellar medium of the host galaxy .
modelling of other bridged fri sources would then be expected to show similarly low edge velocities .
whether the lower - velocity material is best regarded as part of the jets or the lobes ( and , indeed , whether this distinction is meaningful ) remains unclear .
2 . some support for the idea of an interaction between jets and lobes comes from our detection of anomalous ( i.e. inconsistent with our model predictions ) polarization at the edges of the main jet in its outer parts .
we have argued that this can not be a simple superposition of unrelated jet and lobe emission .
rapid deceleration is complete before the end of the bright region in 3c296 .
in contrast , the bright region ends within the deceleration zone for b20326 + 39 , b21553 + 24 and ngc315 and the end of the bright region roughly coincides with the end of rapid deceleration in 3c31 .
4 . when deciding where in the jets our hypothesis of intrinsic symmetry first becomes inappropriate , we have successfully used the criterion that the jets must remain straight while selecting the region to be analysed . in 3c296
however there are hints of intrinsic asymmetries on a still smaller scale because the counter - jet appears to be slightly wider than the main jet in the flaring region .
our results are generally consistent with the conclusions of a statistical study of a larger sample of fri sources with jets @xcite , but the velocity range at the brightening point derived from the distribution of jet / counter - jet ratios assuming an isotropic sample has a maximum @xmath72 and a minimum in the range @xmath73 .
such a range would be expected if low edge velocities ( as in 3c296 ) are indeed typical of bridged twin - jet sources , since these form the majority of the sample @xcite .
we have made deep , 8.5-ghz images of the nearby fri radio galaxy 3c296 with the vla at resolutions ranging from 0.25 to 5.5arcsec fwhm , revealing new details of its twin jets .
in particular we see several thin , discrete brightness enhancements ( `` arcs '' ) in both jets .
these appear to have systematically different morphologies in the main and counter - jets .
a comparison with lower - frequency images from archive data shows that the flat - spectrum jets are surrounded by a broad sheath of steeper - spectrum diffuse emission , possibly formed by a backflow of radiating plasma which has suffered significant synchrotron losses as in models of frii sources .
the spectral index of the jets initially flattens slightly with distance from the nucleus ( from @xmath74 to @xmath75 ) as in other fri jets .
we have also imaged the faraday rotation and depolarization over the source .
the counter - jet is more depolarized than the main jet , but the differences in depolarization and faraday rotation variance between the two lobes are small .
the rms fluctuations in rotation measure are larger close to the nucleus in both lobes .
we have shown that many features of the synchrotron emission from the inner @xmath3740arcsec of the jets in 3c296 can be fit accurately and self - consistently on the assumption that they are intrinsically symmetrical , axisymmetric , decelerating , relativistic flows .
the functional forms we use for the model are similar to those developed in our earlier work , as are most of the derived physical parameters . in this sense ,
a strong family resemblance is emerging between the physical properties that we derive for the outflows in these fri sources . * geometry * : : the jets can be divided into a flaring region , whose radius @xmath49 is well fitted by the expression @xmath76 , where @xmath77 is the distance from the nucleus along the axis , and a conical outer region .
we infer an angle to the line of sight of @xmath78degree .
the boundary between the two regions is then at 15.8kpc from the nucleus . * velocity structure * : : where the jets first brighten , their on - axis velocity is @xmath79 , but this value is poorly constrained .
rapid deceleration occurs around 5kpc to an on - axis velocity of @xmath80 which thereafter remains constant .
3c296 differs from all of the other objects we have studied in having a very low velocity ( essentially consistent with 0 ) at the edges of its jets . *
emissivity * : : the longitudinal profile of rest - frame emissivity is modelled as three principal power - law sections @xmath81 with indices of @xmath82 ( 0 1.8kpc ) , @xmath83 ( 2.5 8.9kpc ) and @xmath84 ( 8.9 24.5kpc , the end of the modelled region ) .
the first two regions are separated by a short transition zone in which the emissivity increases discontinuously by a factor of @xmath367 and then has a power - law slope @xmath572.1 ( 1.8 2.5kpc ; representing the first knots in the main and counter - jets ) .
the rest - frame emissivity is centre - brightened . *
magnetic - field configuration * : : we model the magnetic field as disordered , but anisotropic , with toroidal and longitudinal components , but our results are equally consistent with an ordered toroidal component .
we rule out a globally - ordered helical field .
the ratio of longitudinal to toroidal field decreases with distance from the nucleus ; the toroidal component is also more dominant at the edges of the jets . * flux freezing * : : the field - component evolution is qualitatively consistent with flux freezing , but much slower than expected in the flaring region , where the adiabatic approximation also fails to fit the emissivity variation . in the outer region , after the jets decelerate , the adiabatic approximation is a reasonable fit . *
arc asymmetry * : : if the arcs are narrow shells of enhanced emissivity moving with the underlying flow , then the differences in their appearance in the main and counter - jets can be understood as the effect of relativistic aberration , as required for a fully self - consistent model of the jets .
jrc acknowledges a research studentship from the uk particle physics and astronomy research council ( pparc ) and mjh thanks the royal society for a university research fellowship .
the national radio astronomy observatory is a facility of the national science foundation operated under cooperative agreement by associated universities , inc .
we thank paddy leahy for providing the self - calibrated uv data from @xcite and , in his capacity as referee , for constructive comments . [
baum et al.1997]baum97 baum s.a . ,
odea c.p . , giovannini g. , biretta j. , cotton w.b . , de koff s. , feretti l. , golombek d. , lara l. , macchetto f.d .
, miley g.k .
, sparks w.b . ,
venturi t. , komissarov s.s . , 1997 , apj , 483 , 178 ( erratum apj , 492 , 854 ) [ begelman , blandford & reesbegelman et al.1984]bbr begelman m.c . ,
blandford r.d .
, rees m.j . , 1984 , rev .
phys . , 56 , 255 [ bicknell 1984]bick84 bicknell , g.v . , 1984 , apj , 286 , 68 [ bicknell 1986]bick86 bicknell , g.v . , 1986 , apj , 305 , 109 [ bicknell 1994]bic94 bicknell , g.v . , 1994 ,
apj , 422 , 542 [ bicknell et al.1990]bick90 bicknell g.v .
, de ruiter h.r . ,
parma p. , morganti r. , fanti , r. , 1982 , apj , 354 , 98 [ biretta , zhou & owen1995]biretta95 biretta j.a .
, zhou f. , owen f.n . , 1995 , apj , 447 , 582 [ birkinshaw , laing & peacock1981]blp birkinshaw m. , laing r.a . , peacock j.a . , 1981 , mnras , 197 , 253 [ bondi et al.2000]bondi00 bondi m. , parma p. , de ruiter h.r . , laing r.a . , fomalont e.b . , 2000 ,
mnras , 314 , 11 [ bowman , leahy & komissarov1996 ] blk96 bowman m. , leahy j. p. , komissarov s. s. , 1996 , mnras , 279 , 899 [ bridle1982]b82 bridle , a.h .
1982 , in extragalactic radio sources , eds heeschen d.s .
, wade c.m . , iau symp .
97 , kluwer , dordrecht , p. 121
[ burch1979]burch79 burch s.f . , 1979 , mnras , 187 , 187 [ canvin & laing2004]cl canvin j.r .
, laing r.a . , 2004 , mnras , 350 , 1342 ( cl ) [ canvin et al.2005]clbc canvin , j.r . , laing , r.a . , bridle , a.h . , cotton , w.d . , 2005 , mnras , 363 , 1223 ( clbc ) [ chiaberge , capetti & celotti1999]ccc chiaberge m. , capetti a. , celotti a. , 1999 , a&a 349 , 77 [ de koff et al.2000]dekoff00 de koff s. , best p. , baum s.a . , sparks w. , rttgering h. , miley g. , golombek d. , macchetto f. , martel a. , 2000 , apjs , 129 , 33 [ dineen & coles2005]dc05 dineen p. , coles p. , 2005 , mnras , 362 , 403 [ fanaroff & riley1974]fr74 fanaroff b.l . , riley j.m . , 1974 ,
mnras , 167 , 31p [ fanti et al.1982]fanti82 fanti r. , lari c. , parma p. , bridle a.h . , ekers r.d .
, fomalont e.b , 1982 , a&a , 110 , 169 [ feretti et al.1999]fer99 feretti l. , perley r. , giovannini g. , andernach h. , 1999 , a&a , 341 , 29 [ garrington , holmes & saikiagarrington et al.1996]ghs garrington s.t . , holmes g.f .
, saikia d.j . , 1996 , in extragalactic radio sources , eds ekers r.d . ,
fanti c. , padrielli l. , iau symp .
175 , kluwer , dordrecht , p. 397
[ giovannini et al.2001]giov01 giovannini g. , cotton w.d .
, feretti l. , lara l. , venturi t. , 2001 , apj , 552 , 508 [ giovannini et al.2005]giov05 giovannini g. , taylor g.b .
, feretti l. , cotton w.d .
, lara l. , venturi t. , 2005 , apj , 618 , 635 [ hardcastle et al.1997]hard97 hardcastle m.j . , alexander p. , pooley g.g .
, riley j.m . , 1997 , mnras , 288 , l1 [ hardcastle & worrall1999]hw99 hardcastle m.j . , worrall d.m . , 1999 , mnras , 309 , 969 [ hardcastle et al.2003]hard03 hardcastle m.j .
, worrall d.m . , kraft r.p . , forman w.r . , jones c. , murray s.s .
, 2003 , apj , 593 , 169 [ hardcastle et al.2005]hard05 hardcastle m.j .
, worrall d.m . , birkinshaw m. , laing r.a . , bridle a.h . , 2005 , mnras , 358 , 843 [ katz - stone & rudnick1997]k - sr katz - stone d.m . , rudnick l. , 1997 , apj , 488 , 146 [ katz - stone et al.1999]ksetal katz - stone d.m . , rudnick l. , butenhoff c. , odonoghue a.a . , 1999 , apj , 516 , 716 [ killeen , bicknell & ekerskilleen et al.1986]kbe killeen n.e.b . , bicknell g.v . , ekers r.d . , 1986 , apj , 302 , 306 [ komissarov1994 ] kom94 komissarov s.s . , 1994 , mnras , 269 , 394 [ knigl 1980]kon80knigl , a. , 1980 , phys . fluids , 23 , 1083 [ laing1981]laing81 laing r.a . , 1981 , apj , 248 , 87 [ laing1988]l88 laing r.a
, 1988 , nature , 331 , 149 [ laing1996]l96 laing r.a .
, 1996 , in extragalactic radio sources , eds ekers r.d .
, fanti c. , padrielli l. , iau symp .
175 , kluwer , dordrecht , p. 147
[ laing & bridle2002a]lb02a laing r.a .
, bridle a.h . , 2002a , mnras , 336 , 328 ( lb ) [ laing & bridle2002b]lb02b laing r.a . , bridle a.h . , 2002b , mnras , 336 , 1161 [ laing & bridle2004]lb04 laing r.a .
, bridle a.h . , 2004 ,
mnras , 348 , 1459 [ laing , canvin & bridle2006]lcb laing r.a . , canvin j.r .
, bridle a.h . , 2006 , an , 327 , 523 [ laing et al.2006]ngc315ls laing r.a . , canvin j.r .
, cotton w.d . ,
bridle a.h . , 2006 , mnras , 368 , 48 [ laing et al.1999]lpdrf laing r.a . ,
parma p. , de ruiter h.r .
, fanti , r. , 1999 , mnras , 306 , 513 [ laing & peacock1980]lp80 laing r.a . , peacock j.a . , 1980 , mnras , 190 , 903 [ leahy & perley1991]lp91 leahy j.p .
, perley r.a . , 1991 , aj , 102 , 537 [ leahy , pooley & riley1986]lpr86 leahy j.p .
, pooley g.g .
, riley j.m . , 1986 ,
mnras , 222 , 753 [ martel et al.1999]martel99 martel a. , et al . 1999 ,
apjs , 122 , 81 [ miller et al.2002]miller02 miller n.a .
, ledlow m.j . ,
, hill j.m . , 2002 , aj , 123 , 3018 [ miller et al.1999]miller99 miller n.a . , owen f.n . , burns j.o .
, ledlow m.j .
, voges w. , hill j.m .
, 1999 , aj , 118 , 1988 [ morganti et al.1997]morganti97 morganti r. , parma p. , capetti a. , fanti r. , de ruiter h.r . , 1997 , a&a , 326 , 919 [ parma et al.1996]pdf96 parma , p. , de ruiter , h.r . ,
fanti , r. , 1996 , in extragalactic radio sources , eds ekers r.d .
, fanti c. , padrielli l. , iau symp . 175 , kluwer , dordrecht , p. 137
[ penrose1959]penrose penrose r. , 1959 , proc .
soc . , 55 , 137 [ phinney1983]phi83 phinney e. s. , 1983 , ph.d .
thesis , university of cambridge [ pratt1991]pratt pratt , w.k .
, 1991 , digital image processing , wiley , new york [ simard - normandin , kronberg & button1981]skb simard - normandin m. , kronberg p.p . , button s. , 1981 , apjs , 45 , 97 [ taylor & perley1993]hydraa taylor g.b . , perley r.a .
, 1993 , apj , 416 , 554 [ terrell1959]terrell terrell j. , 1959 , phys .
, 116 , 1041 [ taylor , ulvestad & perley2004]obsss taylor , g. b. , ulvestad , j.s .
, perley , r.a . , 2004 , the very large array observational status summary , nrao , socorro , http://www.vla.nrao.edu/astro/guides/vlas/current/ [ urry & padovani1995]up95 urry c. m. , padovani p. , 1995 ,
pasp , 107 , 803 [ verdoes kleijn & de zeeuw2005]vkdz05 verdoes kleijn g. , de zeeuw p.t . , 2005 , a&a , 435 , 43 [ wardle & kronberg1974]wk wardle j.f.c . ,
kronberg p.p . , 1974 ,
apj , 194 , 249
in table [ tab : param ] , we give for reference the functional forms for the spatial variations of velocity , emissivity and magnetic - field ordering used in this paper .
they are identical to those used by clbc except for the elimination of any radial field component and a slight change to the form of the longitudinal emissivity profile .
we use the streamline coordinate system @xmath85 in the outer region , the streamline index @xmath86 , where @xmath87 is the angle between the ( straight ) streamline and the axis . in the flaring region , the distance of a streamline from the jet axis at a distance @xmath77 from the nucleus is @xmath88 , the coefficients being defined by continuity of @xmath89 and its first derivative across the boundary between the outer and flaring regions . the coordinate @xmath90 increases monotonically along a streamline and is defined by : @xmath91 on - axis ( @xmath92 ) , @xmath93 in both regions .
llcl & & & + quantity & functional form & range & free parameters + & & & + & & & + + & & & + @xmath94 & @xmath95 & & + & @xmath96\exp[c_{1}(\rho - \rho_{\rm v_1})]$ ] & @xmath97 & distances @xmath98 , @xmath99 + & & & + & @xmath100 & @xmath101 & velocities @xmath102 , @xmath103 + & & & + & @xmath104 & & + & @xmath105\exp[c_{4}(\rho_{\rm v_0 } - \rho)]$ ] & @xmath106 & fractional edge velocities @xmath107 , @xmath108 + & & & + & & & + + & & & + @xmath109 & @xmath110 & @xmath111 & distances @xmath112 , @xmath113 , @xmath114 + & & & + & @xmath115 $ ] & @xmath116 & indices @xmath40 , @xmath117 , @xmath118 , @xmath119 + & & & + & @xmath120 & @xmath121 & fractional edge emissivities @xmath122 , @xmath123 + & & & + & @xmath124 & @xmath125 & fractional jump at inner boundary , @xmath126 + & & & + & & & + + & & & + @xmath127 & @xmath128 & @xmath129 & ratios @xmath130 , @xmath131 , @xmath132 , @xmath133 + & & & + & @xmath134 & @xmath135 & distances @xmath136 , @xmath137 + & where @xmath138 & & + & @xmath139 & & + & & & + & @xmath140 & @xmath141 & + & & & + | we present new , deep 8.5-ghz vla observations of the nearby , low - luminosity radio galaxy 3c296 at resolutions from 0.25 to 5.5arcsec .
these show the intensity and polarization structures of the twin radio jets in detail .
we derive the spectral - index distribution using lower - frequency vla observations and show that the flatter - spectrum jets are surrounded by a sheath of steeper - spectrum diffuse emission .
we also show images of faraday rotation measure and depolarization and derive the apparent magnetic - field structure .
we apply our intrinsically symmetrical , decelerating relativistic jet model to the new observations .
an optimized model accurately fits the data in both total intensity and linear polarization .
we infer that the jets are inclined by @xmath0 to the line of sight .
their outer isophotes flare to a half - opening angle of 26@xmath1 and then recollimate to form a conical flow beyond 16kpc from the nucleus . on - axis , they decelerate from a ( poorly - constrained ) initial velocity @xmath2 to @xmath3 around 5kpc from the nucleus , the velocity thereafter remaining constant .
the speed at the edge of the jet is low everywhere .
the longitudinal profile of proper emissivity has three principal power - law sections : an inner region ( 0 1.8kpc ) , where the jets are faint , a bright region ( 1.8 8.9kpc ) and an outer region with a flatter slope .
the emission is centre - brightened .
our observations rule out a globally - ordered , helical magnetic - field configuration . instead , we model the field as random on small scales but anisotropic , with toroidal and longitudinal components only .
the ratio of longitudinal to toroidal field falls with distance along the jet , qualitatively but not quantitatively as expected from flux freezing , so that the field is predominantly toroidal far from the nucleus .
the toroidal component is relatively stronger at the edges of the jet .
a simple adiabatic model fits the emissivity evolution only in the outer region after the jets have decelerated and recollimated ; closer to the nucleus , it predicts far too steep an emissivity decline with distance .
we also interpret the morphological differences between brightness enhancements ( `` arcs '' ) in the main and counter - jets as an effect of relativistic aberration .
[ firstpage ] galaxies : jets radio continuum : galaxies magnetic fields polarization |
the origin of mass is one of the fundamental and fascinating subjects in physics .
about 99% of mass of matter in the world originates from the strong interaction , which provides the large constituent quark mass @xmath11 .
such a dynamical fermion - mass generation in the strong interaction can be interpreted as spontaneous chiral - symmetry breaking ( @xmath12sb).@xcite in the strong interaction , however , there is other type of dynamical mass generation than @xmath12sb .
for instance , gluons , which are massless in perturbation qcd , seem to have a large effective mass as @xmath13,@xcite due to non - perturbative effects .
actually , glueballs , which are ideally composed only by gluons , have a large mass , e.g. , about @xmath14.@xcite the same holds for charm quarks .
whereas the current mass of charm quarks is about 1.2gev at the renormalization point @xmath15 = 1gev , the constituent charm - quark mass in the quark model is set to be about 1.6gev .
the about 400mev difference between the current and the constituent charm - quark masses could be explained by dynamical mass generation without @xmath12sb , since there is no chiral symmetry for such heavy quarks .
these examples imply mass generation without @xmath12sb in the strong interaction .
we therefore conjecture that large dynamical mass generation generally occurs even without @xmath12sb in the strong - interaction world , i.e. , _ all colored particles have a large effective mass generated by dressed gluon effects._@xcite in this study , we investigate the system of colored scalar particles , which do not have chiral symmetry .
we consider light 3c - colored scalar - quarks " @xmath0 .
the light scalar - quarks can be also regarded as idealized point - like `` diquarks '' at the scale of @xmath16gev .
we investigate `` scalar - quark mesons '' @xmath17 and `` scalar - quark baryons '' @xmath3 as the bound states of scalar quarks @xmath0 .
we also investigate the bound states of scalar - quarks @xmath0 and quarks @xmath4 , i.e. , @xmath18 , @xmath6 and @xmath7 , which we name `` chimera hadrons''.@xcite to include scalar - quarks @xmath0 together with quarks @xmath4 and gluons in qcd , we adopt the generalized qcd lagrangian density,@xcite @xmath19 where @xmath20 denotes the quark part and @xmath21 the bare mass of scalar - quarks @xmath0 . in the actual calculation
, we use a discretized euclidean action on the @xmath22 lattice at @xmath23=5.7,@xcite i.e. , lattice spacing @xmath241.1gev.@xcite the parameters employed in the analysis are summarized in table 1 .
the gauge - invariant local operators @xmath25 of scalar - quark hadrons and chimera hadrons are summarized in table 2 .
we introduce scalar - quark flavor " denoted by @xmath26 and investigate the scalar - quark flavor non - singlet mesons , which do not have disconnected diagrams in their correlators .
note that , without the `` scalar - quark flavor '' degrees of freedom , the baryonic local operators of @xmath3 and @xmath7 inevitably vanish due to the anti - symmetric tensor @xmath27 .
we calculate the temporal correlator @xmath28 , where the total momentum is projected to be zero .
the mass @xmath29 of these hadrons are obtained as @xmath30 for large @xmath31 . here
, we show the lattice results for the masses of new - type hadrons .
figure 1 shows the squared scalar - quark - meson mass @xmath32 and the squared scalar - quark - baryon mass squared @xmath33 , plotted against the bare scalar - quark mass squared @xmath34 at @xmath35 . even for zero bare scalar - quark mass ,
scalar - quark hadrons have a large mass as @xmath36 and @xmath37 .
we find the `` constituent scalar - quark picture '' , i.e. , @xmath38 and @xmath39 , where @xmath40gev is the constituent scalar - quark mass .
the calculation can be performed even in the region @xmath41 due to large quantum corrections on @xmath0 .
we also find the relations , @xmath42 and @xmath43 from the figure . together with the `` constituent scalar - quark picture '' , we reach the relation @xmath44 , where @xmath45 is the self - energy of @xmath0 and is expected to be insensitive to @xmath21 .
this is a natural relation between the renormalized mass and the bare mass for scalar particles . and
( b ) `` scalar - quark baryon '' mass squared @xmath33 plotted against the bare scalar - quark mass squared @xmath34 at @xmath35.,width=377 ] around the heavy scalar - quarks in the chimera hadrons .
similar to chimera mesons , the wave - function of @xmath4 in chimera baryons distributes around the two @xmath0 s localized near the center of mass .
, width=230 ] chimera hadrons also have a large mass even at @xmath46 , i.e. , @xmath47gev for chimera mesons @xmath5 , and @xmath48gev , @xmath49gev for chimera baryons ( @xmath50 ) .
we find a `` constituent scalar - quark / quark picture '' , i.e. , an approximate relation as @xmath51 with the constituent quark mass @xmath52 , and the large constituent scalar - quark mass @xmath53.@xcite from the @xmath10-dependence of chimera hadron masses , we conjecture a similar structure between chimera mesons @xmath5 and chimera baryons @xmath7.@xcite the wave - function of @xmath4 in a chimera meson @xmath5 is distributed around the heavy scalar - quark @xmath54 due to the large mass of @xmath0 , and , similarly , the wave - function of @xmath4 in a chimera baryon @xmath7 is distributed around the point - like di - scalar - quark " @xmath55 .
( see fig .
[ fig2 ] . )
we have performed the first study of light scalar - quarks " @xmath0 ( colored scalar particles or idealized diquarks ) and their color - singlet hadronic states in quenched su(3)@xmath1 lattice qcd in terms of dynamical mass generation .
we have investigated the mass of `` scalar - quark mesons '' @xmath17 , `` scalar - quark baryons '' @xmath3 and `` chimera hadrons '' ( @xmath18 , @xmath6 , @xmath7 ) , which are composed of quarks @xmath4 and scalar - quarks @xmath0 .
we have observed the large dynamical mass generation of scalar - quarks @xmath0 about 1.5gev at @xmath56 due to large quantum corrections by gluons , even at the zero bare scalar - quark mass @xmath8 .
this lattice result also indicates that plausible diquarks used in effective hadron models can not be described as the point - like particles and should have a much larger size than @xmath57fm.@xcite this study indicates that , even without @xmath12sb , large dynamical mass generation in the strong interaction occurs for the scalar - quark systems .
together with the large glueball mass and the large difference between the current and the constituent charm - quark masses , this type of mass generation would generally occur in the strong interaction , and therefore we conjecture that all colored particles generally acquire a large effective mass due to dressed gluon effects,@xcite as shown in fig . 3 . | we perform the first study for the bound states of colored scalar particles @xmath0 ( `` scalar quarks '' ) in terms of mass generation with quenched su(3)@xmath1 lattice qcd .
we investigate the bound states of @xmath0 , @xmath2 and @xmath3 ( scalar - quark hadrons " ) , as well as the bound states of @xmath0 and quarks @xmath4 , i.e. , @xmath5 , @xmath6 and @xmath7 ( chimera hadrons " ) .
all these new - type hadrons including @xmath0 have a large mass of several gev due to large quantum corrections by gluons , even for zero bare scalar - quark mass @xmath8 at @xmath9 .
we find a similar @xmath10-dependence between @xmath5 and @xmath7 , which indicates their similar structure due to the large mass of @xmath0 . from this study
, we conjecture that all colored particles generally acquire a large effective mass due to dressed gluons . |
the investigation of the qcd phase diagram is a very important subject both theoretically and experimentally nowadays . the ongoing and future heavy ion experiments such as rhic , and cern / lhc study the low density part of the phase diagram which can also be investigated theoretically by lattice qcd , at cbm / fair the high density part will be studied , which is still not settled theoretically , so it is worth to investigate the phase diagram thoroughly . our starting point is the ( axial)vector meson extended linear sigma model with additional constituent quarks and polyakov - loop variables .
the previous version of the model , without constituent quarks and polyakov - loops , was exhaustively analyzed at zero temperature in @xcite term ) .
this , however , does not influence the results much . ] .
the lagrangian of the model is given by , @xmath0-m_{0 } ^{2}{\mathop{\mathrm{tr}}}(\phi^{\dagger}\phi)-\lambda_{1}[{\mathop{\mathrm{tr}}}(\phi^{\dagger } \phi)]^{2}-\lambda_{2}{\mathop{\mathrm{tr}}}(\phi^{\dagger}\phi)^{2}\nonumber \\ & -\frac{1}{4}{\mathop{\mathrm{tr}}}(l_{\mu\nu}^{2}+r_{\mu\nu}^{2})+{\mathop{\mathrm{tr}}}\left [ \left ( \frac{m_{1}^{2}}{2}+\delta\right ) ( l_{\mu}^{2}+r_{\mu}^{2})\right ] + { \mathop{\mathrm{tr}}}[h(\phi+\phi^{\dagger})]\nonumber \\ & + c_{1}(\det\phi+\det\phi^{\dagger})+i\frac{g_{2}}{2}({\mathop{\mathrm{tr}}}\{l_{\mu\nu}[l^{\mu},l^{\nu}]\}+{\mathop{\mathrm{tr}}}\{r_{\mu\nu}[r^{\mu},r^{\nu } ] \})\nonumber \\ & + \frac{h_{1}}{2}{\mathop{\mathrm{tr}}}(\phi^{\dagger}\phi){\mathop{\mathrm{tr}}}(l_{\mu } ^{2}+r_{\mu}^{2})+h_{2}{\mathop{\mathrm{tr}}}[(l_{\mu}\phi)^{2}+(\phi r_{\mu } ) ^{2}]+2h_{3}{\mathop{\mathrm{tr}}}(l_{\mu}\phi r^{\mu}\phi^{\dagger})\\ & + g_{3}[{\mathop{\mathrm{tr}}}(l_{\mu}l_{\nu}l^{\mu}l^{\nu})+{\mathop{\mathrm{tr}}}(r_{\mu}r_{\nu } r^{\mu}r^{\nu } ) ] + g_{4}[{\mathop{\mathrm{tr}}}\left ( l_{\mu}l^{\mu}l_{\nu}l^{\nu } \right ) \nonumber\\ & + { \mathop{\mathrm{tr}}}\left ( r_{\mu}r^{\mu}r_{\nu}r^{\nu}\right ) ] + g_{5}{\mathop{\mathrm{tr}}}\left ( l_{\mu}l^{\mu}\right ) \,{\mathop{\mathrm{tr}}}\left ( r_{\nu}r^{\nu}\right ) + g_{6 } [ { \mathop{\mathrm{tr}}}(l_{\mu}l^{\mu})\,{\mathop{\mathrm{tr}}}(l_{\nu}l^{\nu})\nonumber \\ & + { \mathop{\mathrm{tr}}}(r_{\mu}r^{\mu})\,{\mathop{\mathrm{tr}}}(r_{\nu}r^{\nu } ) ] + \bar{\psi}i { \mbox{$\,\slash$ \hspace{-0.9em}$\partial$}}\psi - g_{f}\bar{\psi}\left(\phi_{s } + i\gamma_5\phi_{ps}\right)\psi,\nonumber \ ] ] where @xmath1 , \nonumber\\ l^{\mu\nu } & = \partial^{\mu}l^{\nu}-iea_{e}^{\mu}[t_{3},l^{\nu } ] - \left\{\partial^{\nu}l^{\mu } - iea_{e}^{\nu}[t_{3},l^{\mu}]\right\ } , \nonumber \\ r^{\mu\nu } & = \partial^{\mu}r^{\nu } - iea_{e}^{\mu}[t_{3},r^{\nu } ] - \left\ { \partial^{\nu}r^{\mu}-iea_{e}^{\nu}[t_{3},r^{\mu}]\right\ } .
\nonumber\end{aligned}\ ] ] here @xmath2 stands for the scalar and pseudoscalar fields , @xmath3 and @xmath4 for the left and right handed vector fields , @xmath5 for the constituent quark fields , while @xmath6 for the external field .
in order to go to finite temperature / chemical potential , parameters of the lagrangian have to be determined , which is done at @xmath7 . for this
we calculate tree - level masses and decay widths of the model and compare them with the experimental data taken from the pdg @xcite . for the comparison we use a @xmath8 minimalization method @xcite to fit our parameters ( for more details see @xcite ) .
it is important to note that in the present work we also included in the scalar and pseudoscalar masses the contributions coming from the fermion vacuum fluctuations by adapting the method of @xcite .
we have @xmath9 unknown parameters , namely @xmath10 , @xmath11 , @xmath12 , @xmath13 , @xmath14 , @xmath15 , @xmath16 , @xmath17 , @xmath18 , @xmath19 , @xmath20 , @xmath21 , @xmath22 , and @xmath23 . here
@xmath23 is the coupling of the additionally introduced yukawa term , which can be determined from the constituent quark masses through the equations @xmath24 , @xmath25 .
it is worth to note that if we do not consider the very uncertain scalar - isoscalar sector @xmath10 , and @xmath11 always appear in the same combination @xmath26 in all the expressions , thus we can not determine them separately .
additionally a similar combination appears for @xmath14 and @xmath17 in the vector sector as @xmath27 ( see details in @xcite ) .
the parameter values of the fit without scalars are given in table [ tab - param ] .
.parameters determined by @xmath8 minimalization [ cols="<,<,<,<",options="header " , ] since @xmath11 and @xmath28 are undetermined they can be tuned to select the @xmath29 ( a.k.a .
@xmath30 ) from the scalar spectrum ( by its mass and decay widths ) and its mass has , as we will see , a huge effect on the thermal properties of the model .
in our approach we have four order parameters , which are the @xmath31 non - strange and @xmath32 strange condensates , and the @xmath2 and @xmath33 polyakov - loop variables .
the condensates arise due to the spontaneous symmetry breaking and @xmath34 , while the polyakov - loop variables naturally emerge in mean field approximation , if one calculates free fermion grand canonical potential on a constant gluon background .
the effect of fermions propagating on a constant gluon background in the temporal direction formally amounts to the appearance of imaginary color dependent chemical potentials ( for details see @xcite ) . at finite temperature / baryochemical potential
we can set up four coupled field equations for the four fields , which are just the requirements that the first derivatives of the grand canonical potential according to the fields must vanish .
as a first approximation we apply a hybrid approach in which we only consider vacuum and thermal fluctuations for the fermions , but not for the bosons .
we use a mean field polyakov - loop potential @xmath35 of a polynomial form with coefficients determined in @xcite . within this simplified treatment
the equations are the following : @xmath36 where @xmath37 and @xmath38 with the modified distribution functions @xmath39
solving eqs . [ eq_phi]-[eq_phis ] we get the temperature dependence of the order parameters , which can be seen in fig . [ fig - fields_high_sigma ] .
gev , scaledwidth=110.0% ] gev , scaledwidth=110.0% ] in @xcite it was shown that the @xmath40 scalar nonet most probably contains @xmath41 s with masses higher than @xmath42 gev .
if we set @xmath43 we get @xmath44 gev , which is in agreement with @xcite .
however in this case we get a very high pseudocritical temperature , @xmath45 mev , for @xmath31 , which is much larger than earlier results ( e.g. on lattice @xmath46 mev @xcite ) . now , if we tune @xmath11 to get @xmath47 mev ( which corresponds to the physical particle @xmath48 ) , than @xmath49 goes down to @xmath50 mev , which can be seen in fig .
[ fig - fields_low_sigma ] .
this finding is in line with the results of @xcite , where they used a similar model , but without vector mesons .
this suggests that in order to get a good pseudocritical temperature we would need a scalar - isoscalar particle with low mass ( @xmath51 mev ) , which is probably not a @xmath40 state according to @xcite .
gev , scaledwidth=110.0% ] gev , scaledwidth=110.0% ] in fig .
[ fig - mass_high_sigma ] and [ fig - mass_low_sigma ] we show the temperature dependence of the scalar meson masses .
the mass of the parity partners ( @xmath52 and @xmath29 ) reaches the same value above the phase transition temperature .
authors were supported by the hungarian otka fund k109462 and by the hic for fair guest funds of the goethe university frankfurt .
d. parganlija , p. kovcs , gy .
wolf , f. giacosa and d. h. rischke , phys .
d * 87 * , 014011 ( 2013 ) .
p. kovcs and gy .
wolf , acta phys.polon.supp . 6
( 2013 ) 3 , 853 - 858 p. kovcs , a. lukcs , j. vrczy , gy
. wolf and m. ztnyi , phys.rev .
d89 ( 2014 ) 054004 s. chatterjee and k. a. mohan , phys .
d * 85 * , 074018 ( 2012 ) . c. p.
korthals altes , r. d. pisarski and a. sinkovics , phys .
d * 61 * , 056007 ( 2000 ) .
g. mark and zs .
szp , phys . rev .
d * 82 * , 065021 ( 2010 ) . c. ratti , m. a. thaler and w. weise , phys .
d * 73 * ( 2006 ) 014019 .
y. aoki _ et al .
_ , phys . lett .
b * 643 * , 46 ( 2006 ) . | we investigate the scalar meson mass dependence on the chiral phase transition in the framework of an su(3 ) , ( axial)vector meson extended linear sigma model with additional constituent quarks and polyakov loops .
we determine the parameters of the lagrangian at zero temperature in a hybrid approach , where we treat the mesons at tree - level , while the constituent quarks at 1-loop level .
we assume two nonzero scalar condensates and together with the polyakov - loop variables we determine their temperature dependence according to the 1-loop level field equations . |
the theory of open quantum systems allows a description of the open dynamics in terms of the features of the external environment @xcite .
usually , the environment is modeled by a collection of field modes or harmonic oscillators .
the interaction between the open system and the external environment consists in a frequency dependent linear coupling between each mode and the degrees of freedom of the open system .
the open dynamic is fully characterized by the spectral density ( sd ) of the system . for the sake of clarity we report below a brief description of these quantities and refer to @xcite for details .
the total hamiltomian @xmath0 of the open system and the external environment is usually described by the following form , @xmath1 , where @xmath2 is the microscopic hamiltonian of the open system that acts on the corresponding hilbert space @xmath3 @xcite .
the hamiltonian @xmath4 acts on the hilbert space @xmath5 of the external environment and is represented as follows , @xmath6 . the rising operator @xmath7 and the lowering operator @xmath8 act on the hilbert space of the @xmath9th mode or oscillator of frequency @xmath10 .
if the interaction between the open system and external environment is linear , the term @xmath11 usually takes the form @xmath12 .
the operator @xmath13 acts on the hilbert space @xmath3 of the open system and @xmath14 represents the coupling constant between the open system and the @xmath9th frequency mode @xcite .
the spectral density ( sd ) of the system @xmath15 is defined via the coupling constants as follows , @xmath16 this function encodes information about the environment , via the mode frequencies , and about the interaction between open system and environment , via the coupling constants .
the open dynamics is influenced by the temporal correlations among the degrees of freedom of the environment .
these effects are stored in the bath correlation ( bcf ) .
if the external environment is a thermal bath at temperature @xmath17 , if no correlation exists initially between the open system and the thermal bath , and if the stationarity condition holds , the bcf is expressed in terms of the sd by the following form , @xmath18 in units where the planck constant @xmath19 and boltzmann constant @xmath20 equal unity , @xmath21 .
we refer to @xcite for details . in ref .
@xcite a powerful technique has been developed for the analytical evaluation of the bcf by fitting the sd via a class of simple functions . in this way , the bcf is represented as a sum of damped harmonic oscillations .
this form is useful for the description of the reduced dynamics of open quantum systems .
the method allows the evaluation of the bcf for a large variety of sds that include the ohmic and super - ohmic condition and forms appearing in models of photosynthetic light harvesting complexes , to name a few . as a continuation of the scenario described above , here , we consider a general class of sds that exhibit at low frequencies logarithmic factors in addition to the ohmic - like structure .
these terms are represented by arbitrarily positive or negative powers of logarithmic functions .
the corresponding logarithmic singularities are removable and provide sds that are arbitrarily shaped at higher frequencies .
we analyze the short and long time behavior of the bcf in dependence on the low and high frequency structured of these sds , that include and continuously depart from the ohmic - like condition .
the paper is organized as follows .
section [ 2 ] is devoted to the definition of the sds under study . in section
[ 3 ] , the short and long time behavior of the bcf is evaluated analytically in terms of the low or high frequency structure and integral properties of the sd .
conclusions are drawn in section [ 5 ] and details of the calculations are provided in the appendix .
for the sake of convenience , the sds under study are described via the dimensionless auxiliary function @xmath22 .
this function is defined for every @xmath23 by the following scaling property , @xmath24 , where @xmath25 is a typical scale frequency of the system . the first class of sds under study is defined by auxiliary functions that are continuous for every @xmath26 and exhibit the following asymptotic behavior @xcite as @xmath27 , @xmath28 where @xmath29 , @xmath30 , @xmath31 for every @xmath32 , and @xmath33 as @xmath34 .
notice that ohmic - like sds @xcite are obtained as a special case for @xmath35 .
for this reason , the power @xmath36 is referred as the ohmicity parameter . if @xmath35 , the corresponding sds are super - ohmic as @xmath37 for @xmath38 , ohmic for @xmath39 and sub - ohmic for @xmath40 .
the logarithmic singularity in @xmath41 is removed by defining @xmath42 .
the summability of the sd is guaranteed by requiring as @xmath43 the following asymptotic behaviors , @xmath44 , where @xmath45 .
let the complex valued functions @xmath46 and @xmath47 represent the mellin transform @xcite of the auxiliary function @xmath22 and of the function @xmath48 , that is defined as follows , @xmath49 .
the functions @xmath46 and @xmath47 and the meromorphic continuations @xcite are required to decay sufficiently fast as @xmath50 .
details are provided in the appendix . in light of the asymptotic analysis performed in ref .
@xcite , the second class of sds under study is described by auxiliary functions with the following asymptotic expansion as @xmath27 , @xmath51 the powers @xmath52 fulfill the constraints reported above , while the powers @xmath53 are considered to be just real valued , either positive , vanishing or negative .
again , the logarithmic singularity in @xmath41 is removed by setting @xmath42 .
let the parameter @xmath54 be the least natural number such that @xmath55 , where the index @xmath9 is a non - vanishing natural number .
the function @xmath56 is required to be continuous on the interval @xmath57 .
the integral @xmath58 must converge uniformly for all sufficiently large values of the variable @xmath59 and the integral @xmath60 has to converge at @xmath61 uniformly for all sufficiently large values of the variable @xmath59 .
the auxiliary function is required to be differentiable @xmath9 times and to fulfill the following asymptotic expansion as @xmath27 , @xmath62 for every @xmath63 .
furthermore , for every @xmath64 , the function @xmath65 has to vanish as @xmath66 .
@xcite for details .
if compared to the first class , the second class of sds under study has to fulfill more constraints but includes arbitrary positive , vanishing or negative powers of logarithmic forms . in both the classes under
study the auxiliary functions @xmath22 are non - negative , bounded and summable , due to physical grounds , and , apart from the above constraints , arbitrarily shaped .
for the sake of shortness , we refer to the above classes as logarithmic sds in order to highlight the presence of removable logarithmic singularities in the low frequency structure .
the bcf , given by eq .
( [ ct ] ) , is uniquely determined by the sd and the temperature of the thermal bath .
the low frequency and integral properties of the sd characterize the long time behavior of the bcf . for the sake of simplicity
, we define the functions @xmath67 and @xmath68 as the real part and the opposite of the imaginary part of the bcf , respectively , @xmath69 and @xmath70 , this means @xmath71 .
the functions @xmath72 and @xmath68 vanish over long times since the constraint @xmath73 holds for the sds under study , and due to the summability of the sd . over short times ,
the behavior of bcf depends on the high frequency structure and integral properties of the sd .
in fact , the function @xmath72 departs from the finite value @xmath74 in various ways that are determined by the decay of the sd at high frequencies .
a simplified form is obtained for sds that are tailored at high frequency , @xmath75 , as follows , @xmath76 with @xmath77 . in this case
the function @xmath72 decreases quadratically for @xmath78 , @xmath79 where @xmath80 , @xmath81 for @xmath82 , and @xmath83 .
similarly , over short times the function @xmath68 exhibits various behaviors , depending on the high frequency decay of the sd .
consider high frequency profiles where @xmath76 for @xmath75 with @xmath84 . in this case
the function @xmath68 grows linearly for @xmath78 , @xmath85 where @xmath86 . at non - vanishing temperatures , @xmath82 ,
if the sds belong to the first class under study , the function @xmath72 vanishes over long times according to relaxations that involve logarithmic and power laws . if the ohmicity parameter is not an odd natural number we find for @xmath87 the following logarithmic decays , @xmath88 where @xmath89 .
the above form provides inverse power laws for @xmath35 , @xmath90 faster logarithmic relaxations are obtained for @xmath91 , where @xmath92 is a natural number and @xmath93 is a non - vanishing natural number , @xmath94 the parameter @xmath95 reads @xmath96 .
the above asymptotic expression gives inverse power laws for @xmath97 , @xmath98 even faster relaxations are found if @xmath91 and @xmath35 .
consider the least positive index @xmath99 such that either @xmath100 is not an odd natural number , or @xmath101 , where @xmath102 and @xmath103 are non - vanishing natural numbers .
the long time behavior of the function @xmath72 is obtained from eqs .
( [ c1tinft ] ) and ( [ c1tinftpl ] ) by substituting the parameter @xmath36 with @xmath100 and @xmath93 with @xmath103 , in the former case , or by replacing in eqs .
( [ c1tinft1 ] ) and ( [ c1tinft1pl ] ) the number @xmath92 with @xmath102 and @xmath93 with @xmath103 , in the latter case .
we consider sds such that the index @xmath99 exists with the required properties .
if the sds belong to the second class under study , the function @xmath72 exhibits for @xmath87 a variety of relaxations that involve arbitrarily positive , vanishing or negative powers of logarithmic forms , @xmath104 where @xmath105 .
if the ohmicity parameter does not take odd natural values , the dominant part of the above relaxation is @xmath106 , and turns into the inverse power laws @xmath107 for @xmath108 . if the ohmicity parameter is an odd natural number , eq .
( [ c1tinftsd2 ] ) provides faster logarithmic relaxations , @xmath109 , that become the inverse power laws @xmath110 for @xmath111 .
the function @xmath112 vanishes over long times according to various relaxations that involve logarithmic and power laws for sds belonging to the first class under study . if the ohmicity parameter is not an even natural number we find for @xmath87 the following form , @xmath113 where @xmath114 .
the above relaxations become inverse power laws for @xmath35 , @xmath115 faster logarithmic decays appear for @xmath87 if @xmath116 , where @xmath117 and @xmath93 are non - vanishing natural numbers , @xmath118 the parameter @xmath119 reads @xmath120 .
the above asymptotic relaxations become inverse power laws for @xmath97 , @xmath121 even faster relaxations are found if @xmath116 and @xmath35 .
consider the least non - vanishing index @xmath122 such that either @xmath123 is not an even natural number or @xmath124 , where @xmath125 and @xmath126 are non - vanishing natural numbers .
the function @xmath112 is obtained , in the former case , from eqs . ( [ c1tinf0 ] ) and ( [ c1tinf0pl ] ) by substituting the ohmicity parameter with @xmath123 and @xmath93 with @xmath126 , and , in the latter case , from eqs . ( [ c1tinf0m1 ] ) and ( [ c1tinf0m1pl ] ) by substituting the parameter @xmath117 with @xmath125 and @xmath93 with @xmath126 .
if the sds belong to the second class under study , we obtain for @xmath87 a variety of relaxations that involve arbitrary positive or negative powers of logarithmic forms , @xmath127 where @xmath128 .
if the ohmicity parameter does not take even natural values , the dominant part of the above asymptotic form is @xmath129 , and gives the inverse power laws @xmath130 for @xmath108 . if the ohmicity parameter is an even natural number , eq .
( [ c1tinf0sd2 ] ) provides faster logarithmic relaxations , @xmath131 , that result in the inverse power laws @xmath132 for @xmath111 .
if the ohmicity parameter is not an odd natural number , the function @xmath133 vanishes for @xmath87 as below , @xmath134 where @xmath135 .
the above expression describes inverse power laws for @xmath35 , @xmath136 we find faster logarithmic relaxations for @xmath87 if @xmath137 , where @xmath138 is a natural number and @xmath93 is a non - vanishing natural number , @xmath139 the parameter @xmath140 reads @xmath141 .
the above asymptotic form provides inverse power laws for @xmath97 , @xmath142 even faster relaxations are obtained if @xmath143 and @xmath35 .
consider the least non - vanishing index @xmath144 such that either @xmath145 is not an odd natural number or @xmath146 , where @xmath147 is a natural number , and @xmath148 is a non - vanishing natural number .
the function @xmath133 is obtained , in the former case , from eqs .
( [ c2tinf0 ] ) and ( [ c2tinf0pl ] ) by substituting the parameter @xmath36 with @xmath145 and @xmath93 with @xmath148 , and , in the latter case , from eqs .
( [ c2tinfm2 ] ) and ( [ c2tinfm2pl ] ) by substituting the parameter @xmath138 with @xmath147 and @xmath93 with @xmath148 .
for the second class of sds under study , we obtain for @xmath87 a variety of relaxations that involve arbitrarily positive , vanishing or negative powers of logarithmic forms , @xmath149 where @xmath150 .
if the ohmicity parameter does not take odd natural values , the dominant part of the above asymptotic form is @xmath151 , and provides the inverse power laws @xmath152 for @xmath108 . if the ohmicity parameter is an odd natural number , eq .
( [ c2tinf0sd2 ] ) gives faster logarithmic relaxations , @xmath153 , that turn into the inverse power laws @xmath154 for @xmath111 .
numerical computations concerning the real part of the bcf are displayed in figures [ fig1sd ] , [ fig2sd ] and [ fig5sd ] . the short time algebraic behavior is suggested by the initial linear growth of the curves plotted in figure [ fig1sd ] and confirmed by the parallel asymptotic lines appearing in the @xmath155-@xmath155 plot of figure [ fig2sd ] . the long time logarithmic relaxations are confirmed by the asymptotic lines plotted in figure [ fig5sd ] .
numerical computations regarding the imaginary part of the bcf are displayed in figures [ fig3sd ] , [ fig4sd ] and [ fig6sd ] . the short time linear growth is suggested by the curves plotted in figure [ fig3sd ] and confirmed by the parallel asymptotic lines appearing in the @xmath155-@xmath155 plot of figure [ fig4sd ] . the long time logarithmic relaxations
are confirmed by the asymptotic lines plotted in figure [ fig6sd ] . versus @xmath156 for @xmath157 , @xmath158 , @xmath159 and different values of the parameters @xmath160 , @xmath161 and @xmath162 .
the curve @xmath163 corresponds to @xmath164 , @xmath165 and @xmath166 ; @xmath167 corresponds to @xmath168 , @xmath169 and @xmath170 , @xmath171 corresponds to @xmath172 , @xmath173 and @xmath170 ; @xmath174 corresponds to @xmath175 , @xmath173 and @xmath170 ; @xmath176 corresponds to @xmath177 , @xmath169 and @xmath170 ; @xmath178 corresponds to @xmath175 , @xmath169 and @xmath170 ; @xmath179 corresponds to @xmath180 , @xmath181 and @xmath170 ; @xmath182 corresponds to @xmath164 , @xmath165 and @xmath183 ; @xmath184 corresponds to @xmath185 , @xmath186 and @xmath170.,width=10,height=6 ] versus @xmath187 for @xmath188 , @xmath158 , @xmath159 and different values of the parameters @xmath160 , @xmath161 and @xmath162 .
the curve @xmath163 corresponds to @xmath189 , @xmath169 and @xmath166 ; @xmath167 corresponds to @xmath164 , @xmath165 and @xmath166 , @xmath171 corresponds to @xmath175 , @xmath173 and @xmath170 ; @xmath174 corresponds to @xmath189 , @xmath190 and @xmath170 ; @xmath176 corresponds to @xmath191 , @xmath192 and @xmath170 ; @xmath178 corresponds to @xmath164 , @xmath165 and @xmath193 ; @xmath179 corresponds to @xmath194 , @xmath165 and @xmath170 ; @xmath182 corresponds to @xmath185 , @xmath186 and @xmath195 ; @xmath184 corresponds to @xmath164 , @xmath196 and @xmath170 , @xmath197 corresponds to @xmath175 , @xmath198 and @xmath170 . over short times
each curve tends to an asymptotic line with slope @xmath183.,width=10,height=6 ] versus @xmath199 for @xmath200 , @xmath158 , @xmath159 and different values of the parameters @xmath160 , @xmath161 and @xmath162 .
the curve @xmath163 corresponds to @xmath201 , @xmath202 and @xmath166 ; @xmath167 corresponds to @xmath203 , @xmath204 and @xmath166 , @xmath171 corresponds to @xmath205 , @xmath204 and @xmath166 ; @xmath174 corresponds to @xmath203 , @xmath206 and @xmath166 ; @xmath176 corresponds to @xmath207 , @xmath173 and @xmath193 ; @xmath178 corresponds to @xmath168 , @xmath173 and @xmath170 ; @xmath179 corresponds to @xmath164 , @xmath165 and @xmath170.,width=10,height=6 ] versus @xmath187 for @xmath188 , @xmath158 , @xmath159 and different values of the parameters @xmath160 , @xmath161 and @xmath162 .
the curve @xmath163 corresponds to @xmath208 , @xmath209 and @xmath193 ; @xmath167 corresponds to @xmath210 , @xmath206 and @xmath193 , @xmath171 corresponds to @xmath203 , @xmath165 and @xmath193 ; @xmath174 corresponds to @xmath203 , @xmath204 and @xmath166 ; @xmath176 corresponds to @xmath189 , @xmath165 and @xmath170 ; @xmath178 corresponds to @xmath164 , @xmath165 and @xmath170 ; @xmath179 corresponds to @xmath180 , @xmath165 and @xmath170 ; @xmath182 corresponds to @xmath164 , @xmath204 and @xmath211 ; @xmath184 corresponds to @xmath212 , @xmath213 and @xmath166 , @xmath197 corresponds to @xmath180 , @xmath214 and @xmath193 . over short times
each curve tends to an asymptotic line with slope @xmath215.,width=10,height=6 ] versus @xmath216 for @xmath217 , @xmath158 , @xmath159 and different values of the parameters @xmath160 , @xmath161 and @xmath162 .
the curve @xmath163 corresponds to @xmath189 , @xmath165 and @xmath166 ; @xmath167 corresponds to @xmath218 , @xmath206 and @xmath166 , @xmath171 corresponds to @xmath191 , @xmath214 and @xmath166 ; @xmath174 corresponds to @xmath219 , @xmath206 and @xmath193 ; @xmath176 corresponds to @xmath220 , @xmath206 and @xmath170 ; @xmath178 corresponds to @xmath189 , @xmath165 and @xmath170 ; @xmath179 corresponds to @xmath221 , @xmath165 and @xmath193 ; @xmath182 corresponds to @xmath218 , @xmath222 and @xmath183 ; @xmath184 corresponds to @xmath189 , @xmath165 and @xmath193 , @xmath197 corresponds to @xmath223 , @xmath214 and @xmath193 . over long times
each curve tends to an asymptotic line.,width=10,height=6 ] versus @xmath216 for @xmath224 , @xmath158 , @xmath159 and different values of the parameters @xmath160 , @xmath161 and @xmath162 .
the curve @xmath163 corresponds to @xmath189 , @xmath165 and @xmath166 ; @xmath167 corresponds to @xmath164 , @xmath225 and @xmath166 , @xmath171 corresponds to @xmath203 , @xmath206 and @xmath170 ; @xmath174 corresponds to @xmath164 , @xmath226 and @xmath170 ; @xmath176 corresponds to @xmath189 , @xmath227 and @xmath193 ; @xmath178 corresponds to @xmath194 , @xmath222 and @xmath193 ; @xmath179 corresponds to @xmath164 , @xmath186 and @xmath193 . over long times
each curve tends to an asymptotic line.,width=10,height=6 ]
the bcf is a fundamental quantity in the analysis of open quantum systems .
this function depends on the sd of the system and on the temperature of the thermal bath .
usually , the sds under study in literature are ohmic - like at low frequencies with exponential cut - off at high frequencies .
ohmic - like sds can describe an open quantum system that is immersed in a cold environment .
this is the case , for example , of an impurity qubit that is trapped in a double - well potential and is immersed in a bose - einstein - condensed reservoir @xcite .
a detailed analysis of the bcf in terms of the sd is performed in ref .
@xcite , where a powerful method has been developed to express the bcf a sum of damped harmonic oscillations .
this approach applies , in particular , to ohmic - like sds @xcite . in various open quantum systems
the sds depart from the ohmic - like structure .
for example , the analysis of bacteriochlorophyll molecules in pigment - protein complexes requires the introduction of sds of log - normal form @xcite .
these sds are characterized by logarithmic singularities at low frequencies .
recently , logarithmic singularities have been considered in the energy spectrum of unstable quantum systems @xcite .
the corresponding survival amplitude vanishes and the instantaneous energy tends to the minimum energy of the spectrum according to logarithmic relaxations . in light of the scenario described above , here , we have concentrated on sds with removable logarithmic singularities at low frequencies .
the sds are arbitrarily shaped at higher frequencies .
the singularities consist in arbitrary positive , vanishing , or negative powers of logarithmic functions .
these forms are introduced as additional factors in the terms of the power series expansion that describes the low frequency structure of the sds . in this way ,
the sds under study include and continuously depart from the power laws of the ohmic - like condition at low frequencies , and are approximately proportional to the form @xmath228 for@xmath229 .
we remind that @xmath25 represents a typical scale frequency of the system .
the low frequency power law profiles are enhanced ( reduced ) or unchanged by positive ( negative ) or vanishing values of the logarithmic power @xmath230 . over high frequencies , @xmath231 ,
the sds are tailored as follows , @xmath232 with @xmath45 . for the sds
characterized above , we have provided an analytical description of the short and long time behavior of the bcf . over short times , @xmath233 ,
a common algebraic behavior is found , @xmath234 , both at zero and arbitrarily non - vanishing temperature , on condition that the sd decays sufficiently fast over high frequencies , @xmath76 for @xmath75 with @xmath77 .
the involved coefficients @xmath235 and @xmath236 depend on integral properties of the sd .
as far as the long time behavior is concerned , at non - vanishing temperatures the real part of the bcf shows regular dependence on the low frequency structure of the sd .
in fact , the function @xmath72 is approximately proportional to the form @xmath237 for @xmath87 , if the ohmicity parameters @xmath36 differs from odd natural values .
faster long time relaxations appear for odd natural values of the ohmicity parameter . at zero temperature ,
the real part of the bcf exhibits regular dependence on the low frequency behavior of the sd .
in fact , the function @xmath112 is approximately proportional to the expression @xmath238 for @xmath87 , on condition that the ohmicity parameter is not an even natural number .
faster long time relaxations are obtained for even natural values of the ohmicity parameter . in the transition from vanishing to arbitrary non - vanishing temperature , the power law term of the dominant long time decay of the real part of the bcf changes from @xmath239 into the slower decay @xmath240 .
consequently , the decay of the whole bcf slows down and becomes arbitrarily slow for vanishing values of the ohmicity parameter , @xmath241 , and non - negative values of the logarithmic power @xmath230 .
the imaginary part of the bcf shows regular dependence on the low frequency profile of the sd .
in fact , the function @xmath133 is approximately proportional to the form @xmath238 for @xmath87 , if the ohmicity parameter differs from odd natural values .
faster long time relaxations appear for odd natural values of the ohmicity parameter .
the ohmic - like condition is studied as a particular case by making the logarithmic power @xmath230 vanish .
the short time algebraic behavior of the bcf still holds for ohmic - like sds with the required sufficiently fast decay , @xmath77 , at high frequencies . instead ,
the real and imaginary parts of the bcf vanish over long times according to inverse power laws . in conclusion
, a variety of long time relaxations , that are slower than exponential decays and arbitrarily faster or slower than inverse power laws , can be interpreted in terms of removable logarithmic singularities in the sds of open quantum systems that interact with thermal baths .
the bcf , given by eq .
( [ ct ] ) , is represented in terms of cosine and sine transforms of the sd .
the function @xmath72 vanishes asymptotically due to the riemann - lebesgue lemma if the relationship ( [ cond1 t ] ) is fulfilled , while the function @xmath68 vanishes over long times since the sd is summable .
the short and long time behaviors of the functions @xmath72 and @xmath68 are studied in the dimensionless variables @xmath242 and @xmath243 , that are defined as follows , @xmath244 and @xmath245 . for the sake of convenience ,
we introduce the functions @xmath246 and @xmath247 via the following forms , @xmath248 and @xmath249 .
notice the dependence of the function @xmath246 on the temperature @xmath17 . according to eq .
( [ ct ] ) , the functions @xmath246 and @xmath247 read as below , @xmath250 where the auxiliary function @xmath48 is defined as @xmath251 .
the mellin transforms @xcite of the functions @xmath246 and @xmath247 are given by the following forms , @xmath252 the fundamental strips @xcite of the mellin transforms @xmath253 and @xmath254 depend on the asymptotic behavior of the auxiliary function @xmath22 and are discussed below .
the short time behavior of the function @xmath72 is given by the function @xmath246 for @xmath255 .
let the auxiliary function @xmath22 belong to the first class of sds introduced in section [ 2 ] .
the fundamental strip of the function @xmath253 is @xmath256 .
let the strip @xmath257 exist such that the function @xmath258 , or the meromorphic continuation , vanishes for @xmath82 in the strip as follows , @xmath259 for @xmath260 , where @xmath261 .
let the parameters @xmath262 and @xmath263 fulfill the following constraints , @xmath264 with @xmath77 , and @xmath265 . since the following asymptotic relationship @xcite holds for @xmath266 , @xmath267 eq .
( [ cond0omegast ] ) induces a sufficiently fast decay of the function @xmath253 as @xmath266 , and the singularities of the function @xmath253 in @xmath268 and @xmath269 provide eq .
( [ c1tt0 ] ) . if @xmath270 , the fundamental strip of the mellin transform @xmath271 is @xmath272 .
let the strip @xmath273 exist such that the function @xmath274 , or the meromorphic continuation , vanishes for @xmath260 in the strip as follows , @xmath275 for @xmath260 , where @xmath276 .
let the parameters @xmath277 and @xmath278 fulfill the following constraints , @xmath279 with @xmath77 , and @xmath280 . under the above conditions the singularities of the function @xmath271 in @xmath268 and @xmath269
provide eq .
( [ c1tt0 ] ) , for @xmath270 .
we study the short time behavior of the function @xmath68 by analyzing the function @xmath281 for @xmath282 .
the fundamental strip of the mellin transforms @xmath254 is @xmath283 .
let the strip @xmath284 exist such that the function @xmath274 , or the meromorphic continuation , vanishes in the strip as follows , @xmath285 for @xmath260 , where @xmath261 .
let the parameters @xmath286 and @xmath287 fulfill the following constraints , @xmath288 with @xmath84 , and @xmath289 . under the above conditions the singularity of the function @xmath254 in @xmath290
provide eq .
( [ c2t0 ] ) .
the long time behavior of the function @xmath72 is obtained from the asymptotic expansion of the function @xmath246 for @xmath291 .
consider sds that belong to the first class under study . at non - vanishing temperature , @xmath82 ,
let the strip @xmath292 exist such that the function @xmath258 , or the meromorphic continuation , vanishes in the strip as follows , @xmath293 for @xmath260 , where @xmath294 .
let the parameters @xmath295 and @xmath296 fulfill the following constraints , @xmath297 and @xmath298 .
the parameter @xmath299 coincides with @xmath36 , if @xmath36 is not an odd natural number , or if @xmath91 and @xmath300 , otherwise @xmath299 coincides with the parameter @xmath100 that is defined in section [ 3 ] .
the index @xmath301 is the least natural number that is larger than @xmath302 and such that @xmath303 does not take odd natural values , or such that @xmath303 is an odd natural number and @xmath304 . under the above conditions the singularity of the function @xmath305 in @xmath306 provides eqs .
( [ c1tinft])-([c1tinft1pl ] ) . for the second class of sds under study
the analysis performed in refs .
@xcite leads to eq .
( [ c1tinftsd2 ] ) .
the long time behavior of the function @xmath112 is provided by the function @xmath307 for @xmath291 .
consider sds that belong to the first class under study .
let the strip @xmath308 exist such that the function @xmath274 , or the meromorphic continuation , vanishes in the strip as follows , @xmath309 for @xmath260 , where @xmath310 .
let the parameters @xmath311 and @xmath312 fulfill the following constraints , @xmath313 and @xmath314 .
the power @xmath315 coincides with @xmath36 , if @xmath36 is not an even natural number , or if @xmath316 and @xmath300 , otherwise @xmath315 coincides with the power @xmath123 that is defined in section [ 3 ] .
the index @xmath317 is the least natural number that is larger than @xmath318 and such that @xmath319 is not an even natural number , or such that @xmath319 is an even natural number and @xmath320 . under the above conditions the singularity of the function @xmath321 in @xmath322 provides eqs .
( [ c1tinf0])-([c1tinf0m1pl ] ) . for the second class of sds under study
the analysis performed in refs .
@xcite leads to eq .
( [ c1tinf0sd2 ] ) .
the long time behavior of the function @xmath133 is given by the function @xmath247 for @xmath291 .
consider sds that belong to the first class under study .
let the strip @xmath323 exist such that the function @xmath274 , or the meromorphic continuation , vanishes in the strip as follows , @xmath324 for @xmath260 , where @xmath325 .
let the parameters @xmath326 and @xmath327 fulfill the following constraints , @xmath328 and @xmath329 .
the power @xmath330 coincides with @xmath36 , if @xmath36 is not an odd natural number , or if @xmath331 and @xmath300 , otherwise @xmath330 coincides with the power @xmath145 that is defined in section [ 3 ] .
the index @xmath332 is the least natural number that is larger than @xmath333 and such that @xmath334 does not take odd natural values , or @xmath334 is an odd natural number and @xmath335 . under the above conditions the singularity of the function @xmath336 in @xmath337 provides eqs .
( [ c2tinf0])-([c2tinfm2pl ] ) .
for the second class of sds under study the analysis performed in refs .
@xcite leads to eq .
( [ c2tinf0sd2 ] ) .
this concludes the demonstration of the present results . | we study the bath correlation functions ( bcfs ) of open quantum systems interacting with thermal baths , in case the spectral densities ( sds ) exhibit removable logarithmic singularities at low frequencies and are arbitrarily shaped at higher frequencies .
the singularities consist in arbitrarily positive or negative powers of logarithmic functions , as additional factors for the power laws of the ohmic - like sds .
if the sd vanishes sufficiently fast at high frequencies the short time behavior of the bcf is algebraic .
the long time behavior of the bcf exhibits a variety of relaxations that involve inverse power laws and arbitrary powers of logarithmic forms .
the imaginary part of the bcf shows over long times regular dependence on the low frequency structure of the sd , except for certain conditions where the ohmicity parameter takes odd natural values .
same dependence holds for the real part of the bcf at non - vanishing temperatures . at zero temperature the real part of the bcf exhibits over long times the same regular relationship with the low frequency structure of the sd , except for certain conditions involving even natural values of the ohmicity parameter .
the exceptional conditions provide relaxations that are faster than those obtained via the regular dependence . in this way
, various long time relaxations of the bcf that are slower than exponential decays and arbitrarily faster or slower than inverse power laws , can be interpreted in terms of removable logarithmic singularities in the low frequency structure sd of an open quantum system .
pacs : 03.65.yz , 03.65.ta |
suspensions of spherical colloidal particles have proven a valuable model system for understanding many complex phenomena .
perhaps most notably , the model has provided insight into dynamical processes within different phases of matter @xcite , as well as various mechanisms involved during phase transitions @xcite .
a combination of digital video microscopy and computerized particle tracking algorithms @xcite has allowed for the direct visualization of such colloidal systems and measurement of static and dynamical properties under many experimental conditions @xcite .
given the spherical symmetry of the particles , most previous studies have understandably focused only on understanding translational dynamics , though there have been a few studies on rotational dynamics in dilute systems of spheres @xcite , and on the translational and rotational dynamics of anisotropic colloidal particles @xcite . within the past decade
, researchers have developed a variety of techniques for synthesizing clusters of colloidal particles with a wide range of reproducible morphologies @xcite , several of which are shown in fig .
[ fig : clusters ] .
colloidal clusters have the potential to extend the colloidal model past one of simple spherical particles and into a realm where the collective dynamics of particles with complex shapes , more representative of molecules , can be studied @xcite .
and @xmath1 from 3d confocal micrographs .
images have been filtered and enhanced to allow easy visualization of the 3-dimensional structures .
the cases @xmath2 and @xmath3 are accurate representations of the simulated tetrahedra and pentahedra discussed in the text .
individual particles are approximately 2 @xmath4 m in diameter . ]
[ fig : clusters ] conventional particle tracking methods @xcite are designed to follow the translational motions of individual particles , and so are immediately applicable to colloidal clusters , provided the particles can be reliably distinguished . however
, a description of dynamics within systems composed of or including clusters is incomplete without knowledge of how rotational degrees of freedom are explored .
such an understanding could provide further insight into fundamental behaviors of systems with orientational order , such as liquid crystals , or systems subjected to external fields and anisotropic flows @xcite .
to our knowledge , there has been only one study that combined video microscopy with particle tracking to measure the rotational motion of colloidal clusters @xcite , using different methods than described in this work .
that study focused on dilute systems of planar clusters undergoing two - dimensional diffusion near a boundary , and revealed a decoupling of translational and rotational diffusion due to hydrodynamic effects .
we present here a simple and generalized method to track and calculate the two- or three - dimensional rotational motions of clusters of colloidal particles .
our method uses existing particle tracking routines and rigid body transformations to measure the changes in orientation of a cluster over time . to demonstrate the effectiveness and accuracy of the method
, we simulate the motion of colloidal clusters under conditions that are relevant to common microscopy experiments .
we then use our method to measure the rotational diffusion coefficient of a real colloidal cluster , and compare the results of both simulations and experiments to other methods of measuring rotational diffusion .
our procedure for calculating rotational displacements of clusters is based on a method by challis for determining rigid body transformations between reference frames @xcite .
challis procedure was originally intended for comparing osteometric measurements in biomechanical analyses .
as we will show , the method can be intuitively adapted to measure rotational dynamics .
first , we give a brief reprise of challis method , and then describe how it is used to study the systems mentioned here .
given a set of points which have coordinates @xmath5 measured in one reference frame and coordinates @xmath6 measured in a second frame , there exists a transformation @xmath7 where subscript @xmath8 refers to the @xmath8-th point in the set , @xmath9 is a scale factor , @xmath10 is a 3@xmath113 rotation matrix , and @xmath12 is the vector separation of the two reference frames . for our purposes , we may set the scale factor to unity and assume that both coordinate frames share a common origin , thereby setting all elements of @xmath12 to zero .
we are therefore left with @xmath13 and only the rotation matrix to describe the transformation between coordinate frames .
the rotation matrix is an orthonormal matrix with the properties @xmath14 @xmath15 where @xmath16 is the identity matrix and @xmath17 denotes the determinant . for a set of @xmath18 points
, @xmath10 can be calculated using a least squares approach .
this method minimizes the quantity @xmath19^\mathrm{t } \left [ \mathbf{y}_i - \mathbf{r}\mathbf{x}_i\right].\ ] ] ignoring the factor of @xmath20 , expansion of eq .
( [ eqn : leastsquares ] ) yields @xmath21 given that all @xmath22 and @xmath23 are fixed , minimizing eq .
( [ eqn : leastsquares2 ] ) is therefore equivalent to maximizing @xmath24 where @xmath25 is the cross - dispersion matrix calculated from @xmath26 at this point , a singular value decomposition is performed on @xmath25 such that @xmath27 where @xmath28 is a diagonal matrix containing the singular values of @xmath29 , and @xmath30 and @xmath31 are orthogonal matrices . as was shown in @xcite , upon maximizing eq .
( [ eqn : maximize ] ) , @xmath10 is given by @xmath32 \textbf{v}^\mathrm{t}.\ ] ] this procedure is applicable to all non - colinear sets of points with @xmath33 .
particle tracking @xcite yields vector coordinates @xmath34 for every particle @xmath8 over a distinct set of times .
hence , the first step in tracking rotational motion of a cluster is to track the translational motion of each particle within the cluster . for each cluster ,
we first determine the center of mass @xmath35 at a given time and subtract this quantity from the coordinates of particles belonging to the cluster , thereby removing any translational motion .
we are left with new coordinates @xmath22 in the center of mass frame , @xmath36 this step is equivalent to setting the elements of @xmath12 to zero in eq .
( [ eqn : chalreftran ] ) . with translational motion removed , we may apply eq .
( [ eqn : chalrtran ] ) with a slightly different interpretation . rather than representing a transformation between coordinate frames ,
we may understand @xmath10 as describing the rotational trajectory of a particle with inital position @xmath37 to a final position @xmath22 such that @xmath38 therefore , we may use challis procedure to calculate a rotation matrix for each pair of successive times @xmath39 $ ] . with the complete set of rotation matrices , @xmath40
, we may reconstruct the entire trajectory of a particle about the cluster center of mass by computing the product of successive rotations .
given @xmath37 , the position of a particle at some later time @xmath41 can be calculated as @xmath42 where the index @xmath43 enumerates the rotation between successive times .
the advantage of calculating @xmath44 is that it describes the collective behaviors of particles within a cluster , rather than a property of any individual particle .
for example , knowledge of @xmath44 for a cluster allows for immediate calculation of the motions of any particular particle about the center of mass , or the motion of the cluster about any arbitrary axis of rotation .
diffusive anisotropic clusters with large aspect ratios rotate more freely about a long axis than about a short axis . given @xmath45
, however , one needs only the initial orientation of these axes to compute and compare the motions around them .
for simplicity and to better reproduce the data collection process in typical microscopy experiments , we adopt a length scale of microns and a time scale measured in timesteps ( ts ) , which is equivalent in microscopy to video frames or image stacks . in conventional particle tracking experiments , it is important to minimize and understand the uncertainty , i.e. the noise , inherent in locating a particle .
typical microscopy experiments combine high magnification optics with ccd cameras to record raw digital images . in the absence of other sources of noise ,
the uncertainty in particle position , i.e. the minimum noise level , depends on the optical resolution [ pixels / distance ] of the instrumentation and the size [ pixels ] of the object being tracked .
standard image processing and particle tracking techniques can locate the centers of particles to within @xmath46 of a pixel , where @xmath47 is the width of the object in pixels .
optical resolution varies between experimental set - ups but is typically in the range of @xmath48 .
the minimum uncertainty in particle position is the product of these factors .
for example , observing a @xmath49 pixel wide object with an optical resolution of @xmath48 leads to a lower limit of @xmath50 nm uncertainty in particle position .
other sources of noise , such as stray light entering the microscope , noise within the ccd camera itself , etc . , slightly increase the uncertainty in particle position and further limit particle tracking resolution . to test our method
, we simulate the rotational brownian motion of tetra- and pentahedral clusters with different rotational diffusion coefficients , @xmath51 , and different levels of noise , @xmath52 .
we first generate noise - free cluster trajectories . for tetrehedra , we place particles at initial coordinates @xmath53 where @xmath54 is the distance from a particle center to the cluster center of mass . for pentahedra
, we use initial coordinates @xmath55 once initialized , we evolve each simulation for @xmath56 time steps . at each time step , we select three random angles , @xmath57 , from a gaussian distribution with a standard deviation of @xmath58 .
this distribution ensures that the simulated dynamics will be in agreement with the stokes - einstein - debye relation discussed later [ eq .
( [ eqn : sed ] ) ] . each particle in a tetrahedron
is rotated by an angle @xmath59 about the @xmath60-axis , then by an angle @xmath61 about the @xmath62-axis , and finally by angle @xmath63 about the @xmath64-axis to produce the tetrahedron at the subsequent time .
the rotation matrices used are @xmath65 @xmath66 thus , given an initial position vector @xmath67 , the subsequent position vector is @xmath68 .
trajectories generated in this way are noise - free in the sense that they are absent of uncertainty in particle position to within machine precision . to mimic the type of experimental noise previously mentioned , we post - process the trajectories by adding gaussian random numbers , with standard deviation @xmath52 , to the particle coordinates .
the levels of noise presented here correspond to uncertainties of @xmath69 nm in the @xmath64- , @xmath62- , and @xmath60-directions .
experimental uncertainties are typically within the range of 20 - 60 nm , and so the levels studied here are relevant to microscopy experiments .
after the noise is added , we apply our method of measuring rotational motion in order to gauge the effect of experimental noise on results .
in this section , we calculate the rotational motions of simulated colloidal clusters using the matrix methods described above .
we focus on a method that uses the rotation matrices to determine the motion of a fictional orientation vector attached to the cluster .
the rotational displacements of one or more such vectors about the center of mass can be used to compare and contrast motions about different axes , which is especially useful when the objects studied are anisotropic and motions about axes are expected to differ @xcite .
we note that a second method exists to calculate rotational displacements using solely the rotation matrix .
this method determines the rotational axis from @xmath10 and can then be used to calculate the magnitude of an angular displacement . while a mathematically direct and general approach , it describes only the average cluster dynamics and can provide no insight into how motions about different axes vary .
given these limitations , we relegate a description of this method to the appendix . to analyze the rotations of a cluster , we consider the motions of an initial orientational unit vector @xmath70 fixed to the cluster .
we determine its orientation at a later time @xmath41 by applying the set of rotations such that @xmath71 in a similar way as in eq .
( [ eqn : rotprod ] ) .
although there are no constraints on what one may select @xmath70 to be , some choices may be more enlightening than others .
for example , the cluster @xmath3 shown in fig .
[ fig : clusters ] has distinct long and short axes , and so one expects slower diffusion about the short axis . to quantify how rotational dynamics about these axes differ ,
one could choose two @xmath70 to study separately : one choice of @xmath70 perpendicular to the long axis ; a second perpendicular to the short .
such a procedure would yield information relating to motions about the long and short axes , respectively . in line with @xcite
, we define a vector rotational displacement @xmath72 in the time interval @xmath73 $ ] .
the vector @xmath74 has a direction given by @xmath75 and magnitude @xmath76 $ ] , which is the angle subtended by @xmath77 during this time interval . to illustrate the meaning of @xmath78 ,
consider an object with constant angular velocity @xmath79 . over a time @xmath80 ,
the rotational vector displacement is given by @xmath81 .
therefore in general , @xmath78 has components in each of the cartesian axes , @xmath82 , corresponding to cumulative rotations about those axes . in fig .
[ fig : orient_traj](a ) , we show the orientational trajectories of particles within a simulated tetrahedron projected onto the surface of a unit sphere .
[ fig : orient_traj](b ) shows 2d projections of trajectories for two @xmath70 through the rotation space described above . as shown in fig .
[ fig : orient_traj ] , even though the cluster is a solid body , trajectories of individual particles differ due to rotations of the cluster about random axes . .
( b ) 2d projection of two trajectories through rotation space @xmath83 .
colors correspond to the same particles in ( a ) .
both trajectories in ( b ) begin at @xmath84 and end at open circles . ] given the definition of @xmath85 in eq .
( [ eqn : vecdtheta ] ) , we may define an unbounded mean square angular displacement ( msad ) , akin to a translational mean square displacement , as @xmath86 ^ 2 \rangle,\ ] ] where the angle brackets indicate an average over all equivalent lag times @xmath80 . in three dimensions
the stokes - einstein - debye relation states that the msad grows as @xmath87 where @xmath88 is the rotational diffusion coefficient . in this paper , we focus on measuring the msads of clusters , but we point out that other techniques exist to quantify rotational dynamics .
for example , by observing the decay of an orientational correlation function @xmath89 one can measure @xmath51 @xcite .
our method can be applied , in this case , to compute @xmath90 as in eq .
( [ eqn : rphat ] ) . in fig .
[ fig : msadnoise](a ) & ( b ) , we show the msad of two simulated tetrahedral clusters with diffusion constants @xmath91 and @xmath92 respectively and different levels of noise .
the influence of noise is apparent in fig .
[ fig : msadnoise ] as deviations from linearity at small @xmath80 .
eventually , the msad recovers the true diffusive behavior because the cluster has made rotations large enough to distinguish from the noise . and diffusion coefficients of ( a ) @xmath93 and ( b ) @xmath94 .
open circles are the theoretical msad based on eq .
( [ eqn : sed ] ) . deviations from linearity at small @xmath80 demonstrate the effect of noise when resolving small rotations
. deviations at large @xmath80 , however , are the result of low statistics at these lag times . ] with no experimental noise , a log - log plot of msad as a function of @xmath80 will be a straight line with a slope of unity , as indicated by the open circles in figs .
[ fig : msadnoise](a ) & ( b )
. however , in the presence of noise in particle positions , rotations can not be accurately resolved below a certain threshold , @xmath95 .
for example , a stationary cluster will appear to make small , but fictional , rotations as a result of this noise , and measurements of the msad will yield @xmath96 where @xmath97 is independent of @xmath80 due of the lack of correlations in noise . in the case of translational diffusion ,
one expects that the noise in each direction will contribute an error of @xmath98 to the mean - square displacement @xcite . in terms of an angular uncertainty ,
this contribution is diminished by a factor of @xmath99 , where @xmath54 is the average distance from a particle to the cluster center of mass .
our matrix method reduces this uncertainty further by a factor of @xmath18 .
thus , when only static noise is present , one expects for a diffusing cluster @xmath100 where @xmath101 to test this assertion , we add noise to simulations of stationary ( @xmath102 ) tetrahedral clusters and calculate the msads . as in eq .
( [ eqn : msadnoise ] ) , the msads are constant over time .
we take @xmath97 to be the value of the msad as @xmath103 .
shown in fig .
[ fig : chi2](a ) , eq .
( [ eqn : chidefined ] ) accurately describes the static angular uncertainty for a wide range of tetrahedral cluster sizes , @xmath54 , and noise levels .
[ fig : chi2](b ) shows all of the previously measured msads in fig .
[ fig : msadnoise ] plotted with the noise subtracted .
this precisely collapses the msads to the true values in each case .
values of @xmath97 for pentahedral clusters are also predicted by eq .
( [ eqn : chidefined ] ) , and the msads for these clusters can be similarly collapsed . for non - diffusing tetrahedral clusters .
colors indicate noise levels and symbols indicate cluster size .
solid line is the prediction @xmath104 .
( b ) same data in fig .
[ fig : msadnoise ] with the appropriate @xmath97 for each noise level subtracted .
@xmath97 for pentahedral clusters collapses in the same way . ]
knowing how uncertainty in particle positions affects measurements of dynamics quantities , such as the msad , is clearly important .
however , one is typically unable to define the noise level so precisely in experiments .
as stated earlier , particle tracking resolution depends on various experimental factors including particle size , optics , and the type of camera used to record images . as shown in fig .
[ fig : msadnoise](a ) , for slowly diffusing clusters and moderate - to - high noise levels , the measurements of the msad may not be representative of the true dynamics until fairly large lag times . at these large @xmath80 ,
statistics are poorer than small @xmath80 , and confident measurements are harder to obtain .
therefore , this type of oversampling can be detrimental to microscopy experiments where data sets consist of a few hundred to a few thousand images .
thus it is important to determine an imaging rate that yields the most information from a data set . in microscopy experiments , one can effectively increase the rotational signal - to - noise ratio between each time step by imaging less frequently . determining the appropriate sampling rates in experiments
can be done by estimating the noise level @xmath52 , the diffusion constant @xmath51 and calculating an approximate @xmath97 .
diffusive motion will begin exceeding noise when @xmath105 . solving for @xmath80
then yields a reasonable sampling lag time .
it is also important to avoid undersampling , that is , too long a lag time .
doing so will make diffusive motion between images appear erroneously slow .
we find that a good rule of thumb for the upper limit on sampling time should be the time when the cluster has diffused @xmath1061 radian@xmath107 .
thus , an estimate for the upper limit is @xmath108 .
however , as in all particle tracking experiments , the time between images must be small enough that individual particles can be confidently identified .
this typically means that particles must be imaged before moving a distance of one interparticle spacing @xcite .
measurements of rotational motion of diffusive tetrahedral clusters have been performed using the described method @xcite .
real fluorescent tetrahedral clusters are synthesized as in @xcite .
a cluster is composed of individual poly(methyl methacrylate ) ( pmma ) spheres , each with a diameter 2.45 @xmath4 m as measured by static light scattering ( sls ) .
the particles within a cluster are irreversibly bound together , but are sterically stabilized to prevent the possibility of aggregation to other clusters .
dilute suspensions of clusters are prepared in a mixture of cyclohexyl bromide ( cxb ) and _ cis_-decalin ( dcl ) at a ratio of 85/15 ( w / w ) that closely matches both the density and index of refraction of the particles .
clusters are imaged in 3d over time with a leica tcs sp5 confocal microscope .
we track locations of the individual particles within a tetrahedron using standard particle tracking routines @xcite .
the uncertainties in particle position for these experiments are @xmath109 nm in the @xmath64- and @xmath62-directions , and @xmath110 nm in the @xmath60-direction . given these tracking resolutions and assuming a maximally packed tetrahedron , from eq .
( [ eqn : chidefined ] ) we estimate the angular resolution in this experiment as @xmath111 radians ( @xmath112 ) . once tracked , we calculate the translational msd and the msad and determine the translational and rotational diffusion coefficients , @xmath113 and @xmath51 , respectively . in three dimensions ,
the translational msd is described by the stokes - einstein - sutherland equation @xmath114 while the msad is described by eq .
( [ eqn : sed ] ) .
hydrodynamically , tetrahedral clusters can be accurately modeled as spheres @xcite given the relation @xmath115 where @xmath116 is the effective hydrodynamic diameter of the cluster , and @xmath117 is the diameter of the particles within the cluster .
theoretical translational and rotational diffusion coefficients , @xmath113 and @xmath51 respectively , can be calculated using @xmath118 @xmath119 where @xmath120 is boltzmann s constant , @xmath121 is the temperature , and @xmath122 is the viscosity of the suspending solvent .
the viscosity of the cxb / dcl mixture was measured at @xmath123 mpas and experiments were performed at @xmath124 k. in fig .
[ fig : expmsds ] , we show the msd and msad of a diffusing tetradedral cluster .
the msd corresponds to motions of the center of mass of the cluster , and the msad is an average over the individual particle msads ( i.e. , using the initial orientations of the particles as four separate @xmath70 ) .
solid lines are fits to the data over the indicated range @xmath125 . as can be seen ,
both the msd and msad are approximately linear only at small lag times .
this is because the data set used to make these measurements consists of only a single cluster for less than 300 timesteps ( @xmath126 220 s ) .
diffusion coefficients are measured by performing linear fits to the msd and msad shown in fig .
[ fig : expmsds ] over the indicated range and extracting the slopes of the lines .
.measurements from tracking tetrahedral cluster [ cols="<,^,^ , > " , ] [ tab : expmeasure ] shown in table [ tab : expmeasure ] are the measured translational and rotational diffusion coefficients .
values of @xmath116 are calculated from eqs .
( [ eqn : theotransd ] ) & ( [ eqn : theorotd ] ) using known experimental conditions , and values for @xmath117 then follow trivially using eq .
( [ eqn : tetsphere ] ) .
ideally , the sizes calculated from translational motions would be identical to those calculated from rotational motions .
as shown , these values agree to within 4% .
such good agreement between these two measurements demonstrates our ability to track translational and rotational motions of clusters simultaneously .
previous work using standard particle tracking and our matrix method reported particle diameters @xmath127% larger than the 2.45 @xmath4 m measured by sls @xcite .
a reanalysis of the data , presented above , improves the particle diameter measured using our matrix method to within @xmath128% .
we attribute the better accuracy of using rotational motion , as opposed to translational , to the least squares minimization process , which incorporates the rotational motions of all particles within a cluster .
the remaining disagreement is likely due to multiple factors , including the fitting routines used in sls and swelling of particles when in a solvent of cxb / dcl .
sls measurements were performed with suspensions in pure dcl , which is not known to swell particles .
we have presented a simple method of tracking the rotational motions of colloidal clusters .
our method implements conventional particle tracking routines to determine the locations of individual particles within a cluster and uses this information to compute rigid body transformations that describe changes in a cluster s orientation over time .
the set of matrix transformations constitutes a global description of a cluster s motion during the course of an experiment or simulation and allows one to calculate rotational dynamics about any arbitrary axis .
the least squares minimization used in this method considers the motions of each individual particle in calculating the rotation matrix , and therefore measurements of rotational displacements are less sensitive to tracking noise .
additionally , precision in determining angular displacements increases with cluster size @xmath54 .
when combined , the resolution of measuring angular displacements scales with 1/@xmath129 . though constructed specifically for tracking colloidal clusters , we emphasize that our method is not limited to these systems , but is applicable to following the rotational motion of any body over time , providing that at least three distinct noncolinear points in the body can be reliably distinguished .
because the accuracy of tracking a rotation depends on the number of particles and the size of the body , the possibility exists that this method can be adapted to many - body systems where each particle diffuses independently of the rest while the system itself also undergoes bulk rotations . in this scenario ,
diffusive motions are random and can therefore be treated as noise while calculating @xmath130 for the system . the rotation matrices can be inverted and , in a manner similar to eq .
( [ eqn : rotprod ] ) , the bulk rotations can be removed , leaving only uncorrelated diffusive motion .
thus , this method is also applicable to determining particle motions in rotating coordinate frames .
g.l.h . , k.v.e . , and e.r.w .
acknowledge financial support from nsf grant che-0910707 , and m.t.e .
acknowledges support from nsf grant dmr-0706453 .
we thank d.j .
pine for helpful discussions .
given a rotation matrix @xmath10 , one can calculate at each time the axis of rotation @xmath131 and angular displacement @xmath132 . in this notation
, a rotational displacement can be described by a vector @xmath133 , where @xmath131 has components in each of the cartesian axes . from eq .
( [ eqn : ronu ] ) , we see that the axis of rotation is an eigenvector of the matrix @xmath10 with an eigenvalue of 1 . for a set of rotational displacements @xmath130
, one may determine the axes of rotation by calculating the eigenvectors and eigenvalues of the rotation matrices , searching for the eigenvalues equal to 1 , and taking the corresponding eigenvectors . to determine the size of the displacement about the axis of rotation
, one defines an arbitrary vector @xmath135 perpendicular to @xmath131 . for simplicity , we choose @xmath135 to be perpendicular to the @xmath64-axis ( denoted by @xmath136 ) , calculating displacements relative to the axis of rotation always results in displacements greater than or equal to those measured relative to an arbitrary @xmath70 .
for example , measuring a diffusion coefficient of a spherically symmetric body with this method will yield a value that is a factor of @xmath141 of the actual diffusion coefficient in eq .
( [ eqn : sed ] ) .
rad@xmath142ts and @xmath143 rad@xmath142ts and @xmath144 50 nm .
dashed lines show the uncorrected msad described below and solid lines are corrected data , obtained by multiplying the dashed lines by 2/3 .
open circles are the theoretical msad from eq .
( [ eqn : sed ] ) . ] to understand the origin of this difference , consider the rotational axis @xmath145 , a perpendicular vector @xmath146 , and an arbitrary vector @xmath70 located at spherical coordinates @xmath147 . if @xmath135 is rotated by an amount @xmath148 , the angle between @xmath135 and @xmath139 is identically @xmath148 .
however , the angle @xmath149 between @xmath77 and @xmath150 can be shown to be the factor of @xmath155 arises for the same reasons in perrin s original derivation of rotational diffusion @xcite .
[ fig : pmsad ] shows the msads measured in this way for two pentahedral clusters .
as stated , one can correct for the overestimation of motion by simply multiplying the msad by 2/3 .
the average msad for any body can be corrected in the same manner , however we stress that this method returns only the average dynamics . while more direct than using an orientation vector , characterizing anisotropic bodies in this way will convolute motions about separate axes , and so will require some care when interpreting results . + | we describe a novel method of tracking the rotational motion of clusters of colloidal particles .
our method utilizes rigid body transformations to determine the rotations of a cluster and extends conventional proven particle tracking techniques in a simple way , thus facilitating the study of rotational dynamics in systems containing or composed of colloidal clusters .
we test our method by measuring dynamical properties of simulated brownian clusters under conditions relevant to microscopy experiments .
we then use the technique to track and describe the motions of a real colloidal cluster imaged with confocal microscopy . 10 a. kose , m. ozaki , k. takano , y. kobayashi , and s. hachisu , `` direct observation of ordered latex suspension by metallurgical microscope , '' j. colloid interf .
sci . * 44 * , 330338 ( 1973 ) .
d. j. w. aastuen , n. a. clark , l. k. cotter , and b. j. ackerson , `` nucleation and growth of colloidal crystals , '' phys .
rev .
lett .
* 57 * , 17331736 ( 1986 )
. c. a. murray and d. h. van winkle , `` experimental observation of two - stage melting in a classical two - dimensional screened coulomb system , '' phys .
. lett . *
58 * , 12001203 ( 1987 ) .
e. b. sirota , h. d. o. yang , s. k. sinha , p. m. chaikin , j. d. axe , and y. fujii , `` complete phase diagram of a charged colloidal system : a synchrotron x - ray scattering study , '' phys .
rev .
lett . * 62 * , 15241527 ( 1989 ) .
p.
n. pusey and w. van megen , `` phase behaviour of concentrated suspensions of nearly hard colloidal spheres , '' nature * 320 * , 340342 ( 1986 ) .
p.
n. pusey and w. van megen , `` observation of a glass transition in suspensions of spherical colloidal particles , '' phys .
rev .
lett . * 59 * , 20832086 ( 1987 ) .
p. n. pusey , w. c. k. poon , s. m. ilett , and p. bartlett , `` phase behaviour and structure of colloidal suspensions , '' j. phys .
: cond .
matt . * 6 * , a29a36 ( 1994 ) .
e. r. weeks , j. c. crocker , a. c. levitt , a. schofield , and d. a. weitz , `` three - dimensional direct imaging of structural relaxation near the colloidal glass transition , '' science * 287 * , 627631 ( 2000 ) . j. hernndez - guzmn and e. r. weeks , `` the equilibrium intrinsic crystal - liquid interface of colloids , '' proc .
nat .
acad .
sci .
* 106 * , 1519815202 ( 2009 ) .
u. gasser , e. r. weeks , a. schofield , p. n. pusey , and d. a. weitz , `` real - space imaging of nucleation and growth in colloidal crystallization , '' science * 292 * , 258262 ( 2001 ) .
d. g. grier and c. a. murray , `` the microscopic dynamics of freezing in supercooled colloidal fluids , '' j. chem .
phys . * 100 * , 90889095 ( 1994 ) . c. a. murray , w. o. sprenger , and r. a. wenk , `` comparison of melting in three and two dimensions : microscopy of colloidal spheres , '' phys .
rev .
b * 42 * , 688703 ( 1990 ) . d. h. van winkle and c. a. murray , `` layering transitions in colloidal crystals as observed by diffraction and direct - lattice imaging , '' phys . rev .
a * 34 * , 562573 ( 1986 )
. j. r. savage , d. w. blair , a. j. levine , r. a. guyer , and a. d. dinsmore , `` imaging the sublimation dynamics of colloidal crystallites , '' science * 314 * , 795798 ( 2006 ) .
a. d. dinsmore , e. r. weeks , v. prasad , a. c. levitt , and d. a. weitz , `` three - dimensional confocal microscopy of colloids , '' appl . opt . * 40 * , 41524159 ( 2001 ) .
j. c. crocker and d. g. grier , `` methods of digital video microscopy for colloidal studies , '' j. colloid interf .
sci . *
179 * , 298310 ( 1996 ) .
r. besseling , e. r. weeks , a. b. schofield , and w. c. k. poon , `` three - dimensional imaging of colloidal glasses under steady shear , '' phys .
rev . lett . * 99 * , 028301 ( 2007 ) . c. r. nugent , k. v. edmond , h. n. patel , and e. r. weeks , `` colloidal glass transition observed in confinement , '' phys . rev . lett . * 99 * , 025702 ( 2007 ) .
e. r. dufresne , t. m. squires , m. p. brenner , and d. g. grier , `` hydrodynamic coupling of two brownian spheres to a planar surface , '' phys .
rev .
lett . * 85 * , 33173320 ( 2000 ) .
d. chen , d. semwogerere , j. sato , v. breedveld , and e. r. weeks , `` microscopic structural relaxation in a sheared supercooled colloidal liquid , '' phys .
rev .
e * 81 * , 011403 ( 2010 ) .
p. schall , d. a. weitz , and f. spaepen , `` structural rearrangements that govern flow in colloidal glasses , '' science * 318 * , 18951899 ( 2007 ) .
s. m. anthony , l. hong , m. kim , and s. granick , `` single - particle colloid tracking in four dimensions , '' langmuir * 22 * , 98129815 ( 2006 ) .
s. martin , m. reichert , h. stark , and t. gisler , `` direct observation of hydrodynamic rotation - translation coupling between two colloidal spheres , '' phys .
rev .
lett .
* 97 * , 248301 ( 2006 ) .
d. mukhija and m. j. solomon , `` translational and rotational dynamics of colloidal rods by direct visualization with confocal microscopy , '' j. colloid interf .
sci . * 314 * , 98106 ( 2007 ) .
y. han , a. m. alsayed , m. nobili , j. zhang , t. c. lubensky , and a. g. yodh , `` brownian motion of an ellipsoid , '' science * 314 * , 626630 ( 2006 ) .
l. hong , s. m. anthony , and s. granick , `` rotation in suspension of a rod - shaped colloid , '' langmuir * 22 * , 71287131 ( 2006 ) .
p.
j. yunker , k. chen , z. zhang , w. g. ellenbroek , a. j. liu , and a. g. yodh , `` rotational and translational phonon modes in glasses composed of ellipsoidal particles , '' phys .
rev .
e * 83 * , 011403 ( 2011 ) .
m. t. elsesser , a. d. hollingsworth , k. v. edmond , and d. j. pine , `` large core - shell poly(methyl methacrylate ) colloidal clusters : synthesis , characterization , and tracking , '' langmuir * 27 * , 917927 ( 2011 ) . v. n. manoharan , m. t. elsesser , and d. j. pine , `` dense packing and symmetry in small clusters of microspheres , '' science * 301 * , 483487 ( 2003 ) . s. c. glotzer and m. j. solomon , `` anisotropy of building blocks and their assembly into complex structures , '' nature materials * 6 * , 557562 ( 2007 ) .
a. van blaaderen , `` materials science : colloids get complex , '' nature * 439 * , 545546 ( 2006 ) .
a. van blaaderen , `` chemistry : colloidal molecules and beyond , '' science * 301 * , 470471 ( 2003 ) .
s. m. anthony , m. kim , and s. granick , `` translation - rotation decoupling of colloidal clusters of various symmetries , '' j. chem .
phys . * 129 * , 244701 ( 2008 ) .
j. challis , `` a procedure for determining rigid body transformation parameters , '' j. biomech .
* 28 * , 733737 ( 1995 ) . m. g. mazza , n. giovambattista , h. e. stanley , and f. w. starr , `` connection of translational and rotational dynamical heterogeneities with the breakdown of the stokes - einstein and stokes - einstein - debye relations in water , '' phys .
rev .
e * 76 * , 031203 ( 2007 ) .
m. g. mazza , n. giovambattista , f. w. starr , and h. e. stanley , `` relation between rotational and translational dynamic heterogeneities in water , '' phys .
rev .
lett . *
96 * , 057803 ( 2006 ) .
s. kmmerer , w. kob , and r. schilling , `` dynamics of the rotational degrees of freedom in a supercooled liquid of diatomic molecules , '' phys .
rev .
e * 56 * , 54505461 ( 1997 ) .
b. j. berne , p. pechukas , and g. d. harp , `` molecular reorientation in liquids and gases , '' j. chem .
phys . * 49 * , 31253129 ( 1968 ) .
g. williams , `` time - correlation functions and molecular motions , '' chem
. soc .
rev .
* 7 * , 89131 ( 1978 ) .
m. hoffmann , c. s. wagner , l. harnau , and a. wittemann , `` 3d brownian diffusion of submicron - sized particle clusters , '' acs nano * 3 * , 33263334 ( 2009 ) .
f. perrin , `` tude mathmatique du mouvement brownien de rotation , '' annales scientifiques de l.n.s .
@xmath0 srie * 45 * , 151 ( 1928 ) . |
a wealth of data@xcite exists on charge - exchange processes in soft hadronic reactions , where an initial - state proton is transformed into a final - state neutron , @xmath60 .
a successful phenomenological description of these results has been obtained with the concept of the exchange of virtual isovector mesons , such as @xmath61 , @xmath62 , and @xmath63 , using regge theory@xcite .
since the pion is by far the lightest hadron , its exchange dominates the @xmath60 transition , particularly at small values of the squared momentum transfer , @xmath64 , between the proton and the neutron .
the assumption of factorization , namely that the partonic nature of a hadron is independent of the hard scattering process in which it participates , has been shown to be valid in the case of the nucleon , whose partonic structure has been probed extensively in jet - production processes as well as in deep inelastic scattering .
the idea of factorization may be extended to the exchanged objects in charge - exchange reactions . under this assumption ,
hard processes occurring in charge - exchange reactions , such as the production of high-@xmath65 jets , provide a means of investigating the partonic nature of the exchanged objects .
charge - exchange processes have been studied in deep inelastic scattering at hera@xcite .
this paper reports the first observation of the photoproduction of dijets in association with an energetic forward neutron : @xmath66 where @xmath67 denotes the remainder of the final state .
the virtuality of the exchanged photon , @xmath68 , was less than @xmath69 gev@xmath70 , with a median value of about @xmath71 gev@xmath70 .
neutrons with energy @xmath72 gev and produced at an angle @xmath2 mrad with respect to the direction axis pointing in the proton beam direction , hereafter referred to as forward , and the @xmath73 axis horizontal , pointing towards the center of hera .
pseudorapidity is defined as @xmath74 , where the polar angle @xmath75 is taken with respect to the proton beam direction . ] of the hera proton beam ( @xmath76 gev ) were detected in a forward neutron calorimeter .
these results extend previous zeus photoproduction dijet studies@xcite .
the present data are compared to an inclusive sample of dijet events selected without the requirement of a forward neutron .
cross sections are presented both as a function of the kinematic variables of the jet and of the @xmath60 transition .
the contributions to the cross section of direct processes , where the photon acts as a point particle , and resolved , where the photon acts as a source of partons , are compared in both the neutron - tagged and the inclusive samples .
in addition , the fraction of the inclusive dijet sample with a leading neutron is given as a function of the jet transverse energy ( @xmath77 ) and pseudorapidity ( @xmath78 ) .
these fractions are used to study the factorization properties of the processes .
the results are compared to predictions of one - pion exchange .
the dijet processes under consideration here are characterized by an initial state consisting of a positron @xmath79 , and a proton @xmath80 , and a final state consisting of the scattered positron , the scattered neutron @xmath81 , and a hadronic system @xmath82 : @xmath83 where @xmath84 , @xmath85 , @xmath86 and @xmath87 are the four - momenta of the initial and scattered positron , and the proton and neutron , respectively .
the process is described by four lorentz invariants .
two describe the positron - photon vertex and can be taken to be the virtuality of the exchanged photon ( @xmath68 ) and the electron s inelasticity ( @xmath88 ) , defined by : @xmath89 in photoproduction , where @xmath68 is small , @xmath90 , where @xmath91 is the energy of the initial ( scattered ) positron , and @xmath92 is the incident photon energy .
the other two variables , which describe the proton - neutron vertex , are the fraction of the energy of the initial - state proton carried by the neutron ( @xmath93 ) , and the square of the momentum transfer ( @xmath64 ) between the initial proton and the produced neutron , defined by : @xmath94 where @xmath95 is the energy of the incident proton .
the transverse momentum , @xmath96 , of the neutron is related to @xmath64 and @xmath93 by : @xmath97 where @xmath98 is the mass of the proton ( neutron ) . in the photoproduction of dijets tagged with a leading neutron ,
the hadronic system h contains at least two jets : @xmath99 in @xmath100 scattering of massless partons , the fractions of the four - momenta @xmath101 and @xmath102 carried into the hard scattering by the initial - state partons are given by : @xmath103 respectively , where @xmath104 is the four - momentum of the @xmath105th final - state parton , and the approximation @xmath106 has been used .
the energy fraction contributed by the exchanged photon to the production of the dijets is @xmath107 ; in regge models , where @xmath60 is the result of the @xmath61 , @xmath62 or @xmath63 trajectory coupling to the @xmath108 vertex , the corresponding contribution of the exchanged meson is @xmath109 .
a further relationship is : @xmath110 where @xmath111 is the fraction of the proton energy participating in the production of the dijets .
the observables @xmath112 , @xmath113 and @xmath114 , defined in terms of jets , are introduced@xcite : @xmath115 where the sum runs over the two jets of highest @xmath77 in an event .
the variables @xmath112 and @xmath114 are estimators of @xmath107 and @xmath109 , respectively . in leading - order ( lo ) qcd ,
two types of processes contribute to jet photoproduction @xcite : either the entire photon interacts with a parton in the target ( the _ direct _ process ) , or the photon acts as a source of partons which scatter off those in the target ( the _ resolved _ process ) .
figure [ fig : feynman ] illustrates these processes for reaction ( [ eqn : gpn ] ) with an assumed meson - exchange contribution .
the observable @xmath112 is sensitive to which type of process occurs@xcite .
direct processes are concentrated at high values of @xmath112 , resolved processes at low values .
the data sample used in this analysis was collected in 1995 with the zeus detector using @xmath116 interactions . in this period hera
operated with 174 colliding bunches of @xmath117 820 gev protons and @xmath118 27.5 gev positrons , corresponding to a center - of - momentum - frame energy @xmath119 gev . additionally , 21 unpaired bunches of protons or positrons allowed the beam related backgrounds to be determined .
the integrated luminosity used in this analysis is .
the zeus detector is described in detail elsewhere@xcite .
the principal components used in the present analysis were the central tracking detector ( ctd)@xcite positioned in a 1.43 t solenoidal magnetic field , the uranium - scintillator sampling calorimeter ( cal)@xcite , and the forward neutron calorimeter ( fnc)@xcite . the tracking system was used to establish an interaction vertex with a typical resolution along ( transverse to ) the beam direction of 0.4(0.1 ) cm .
energy deposits in the cal were used to find jets and to measure their energies and angles .
the cal is hermetic and consists of 5918 cells , each read out by two photomultiplier tubes . under test beam conditions , the cal has energy resolutions of @xmath120 for electrons and @xmath121 for hadrons ( @xmath122 in gev)@xcite .
jet energies were corrected for the energy lost in inactive material ( typically one radiation length ) in front of the cal .
the forward neutron calorimeter@xcite was installed in the hera tunnel at degrees and at @xmath123 m from the interaction point in the proton - beam direction , and used for the 1995 data taking .
the layout of the calorimeter is shown in fig .
[ fig : fnc ] .
the fnc is a sampling calorimeter with 134 layers of 1.25 cm thick lead as the absorber and 0.26 cm thick scintillator as the active material .
the scintillator is read out on each side with wavelength - shifting light - guides coupled to photomultiplier tubes .
it is segmented longitudinally into a front section , seven interaction - lengths deep , and a rear section , three interaction - lengths deep .
the front section is divided vertically into 14 towers , allowing the separation of electromagnetic and hadronic showers using the energy - weighted vertical width of the showers .
the energy resolution for hadrons , as measured in a beam test , is @xmath124 ( @xmath122 in gev)@xcite .
neutrons are easily distinguished from protons , which are deflected upwards by the beam magnets and deposit most of their energy in the top four towers of the fnc .
three planes of veto counters , each @xmath125 @xmath126 , are located 70 , 78 , and 199 cm in front of the calorimeter .
these counters , which completely cover the bottom front face of the calorimeter , were used offline to identify charged particles and so reject particles which interacted in the inactive material in front of the fnc .
magnet apertures limit the fnc acceptance to neutrons with production angles less than , that is to transverse momenta @xmath127 gev .
only about one quarter of the corresponding azimuth is free of obstruction , as can be seen from the outline of the aperture in fig .
[ fig : fnc](a ) . the @xmath128-axis intersection with the fnc
is also indicated .
the overall acceptance of the fnc , which includes beam - line geometry , absorbing material , and the angular distribution of the neutrons , is about 30% for neutrons with energy and .
the kinematic region covered by the fnc in longitudinal and transverse variables is shown in fig .
[ fig : kine ] .
although the acceptance extends to @xmath129 gev@xmath70 , the mean value of @xmath130 for the accepted data is less than 0.05 gev@xmath70@xcite .
the @xmath64 acceptance is strongly affected by the minimum value of @xmath131 , @xmath132 .
the calibration and monitoring@xcite of the fnc follow the methods developed for the fnc test calorimeters@xcite .
the gain of the photomultiplier tubes is obtained by scanning the fnc with a @xmath133co radioactive source .
changes in gain during data taking are monitored using energy deposits from interactions of the hera proton beam with residual gas in the beam pipe .
the overall energy scale is set from the kinematic end point of 820 gev by fitting proton beam - gas interaction data with energy greater than 600 gev to that expected from one - pion exchange@xcite .
the zeus detector uses a three - level trigger system . at the first level , events were selected by a coincidence of a regional or transverse energy sum in the calorimeter , and at least one track from the interaction point measured in the ctd . at the second level , at least 8 gev total transverse energy , excluding the eight calorimeter towers immediately surrounding the forward beam pipe , was required and cuts on calorimeter energies and timing were used to suppress events caused by beam - gas interactions @xcite . at the third level , a cone algorithm used the calorimeter cell energies and positions to identify jets .
events were required to have at least two jets , each of which satisfied the requirement @xmath134 gev and @xmath135 , or @xmath136 gev and @xmath137 .
additional tracking cuts were made to reject proton beam - gas interactions and cosmic - ray events .
no requirement on the fnc was made at any trigger level .
further selection criteria were applied offline . charged
current scattering events were rejected by a cut on the missing transverse momentum measured in the calorimeter . to reject remaining beam - gas and cosmic - ray backgrounds ,
tighter cuts were applied .
these used the final @xmath128-vertex position , other tracking information and timing information .
two additional cuts were also made @xcite : * events with a well - identified positron candidate in the uranium calorimeter were removed .
* a cut was made on the jacquet - blondel estimator of @xmath88 @xcite , @xmath138 , where @xmath139 , and @xmath140 is the energy deposited in the calorimeter cell @xmath105 with a polar angle @xmath141 with respect to the measured @xmath128-vertex of the event .
the sum runs over all calorimeter cells . for any event where the scattered positron entered the uranium calorimeter and was not well identified ,
the value of @xmath142 is close to one .
proton beam - gas events have low values of @xmath142 .
to reduce further the contamination from this source , it was required that @xmath143 .
this range corresponds approximately to a true @xmath88 range of @xmath144 , and so to a energy range of in the @xmath145 center - of - momentum frame .
these cuts restricted the range of @xmath68 to be less than @xmath69 gev@xmath146 , with a median value of about @xmath71 gev@xmath70 .
dijet candidates were selected using the ktclus@xcite jet algorithm ( details can be found in a previous zeus publication@xcite ) .
cone@xcite algorithms were also used as a check : the conclusions did not change . the jet transverse energy measured in the zeus detector
was corrected as a function of pseudorapidity and transverse energy to account for energy lost in inactive material .
this correction was derived from the monte carlo ( mc ) simulation described in the next section . after all cuts , the kinematic region under study is defined by : , , and .
events with a leading neutron were selected from the inclusive dijet sample by requiring a large energy deposit ( @xmath147 gev ) in the fnc .
the segmentation of the fnc permits the identification of protons , photons , and neutrons .
scattered protons are bent into the top towers ( 11 - 14 ) by the hera dipole magnets . as seen in fig .
[ fig : fnc ] , the geometric aperture of the fnc for neutral particles at normal incidence is centered on towers 7 and 8 .
scattered protons were eliminated from the sample by requiring that the tower with the maximum energy deposit be either 6,7,8 or 9 . although both photons and neutrons produce large energy deposits in the bottom section ( towers 1 - 10 ) , the vertical spread of electromagnetic showers is much less than that of hadronic showers .
photons were removed by eliminating showers with a small vertical spread ( @xmath148 cm ) . finally , neutrons that started showering in front of the fnc were removed by requiring that the scintillator veto counter farthest from the fnc had an energy deposit below that of a minimum - ionizing particle .
only the farthest counter was used , to minimize false vetoes due to calorimeter albedo .
showers with spreads greater than 7.5 cm were also removed since they are inconsistent with originating from a single high - energy hadron . after these cuts , 1921 events with a neutron remained , comprising approximately 1% ( before correction for the fnc acceptance ) of the inclusive dijet sample of @xmath149 events .
monte carlo simulations were used to correct the data for acceptance and for smearing of the measured quantities due to the finite resolution of the detector . for all generated events ,
the zeus detector response was simulated in detail using a program based on geant3.13 @xcite , and the monte carlo events were subjected to the same analysis chain as the data . for the inclusive dijet analysis ,
the data were compared to monte carlo simulations based on pythia5.7 @xcite and herwig5.9 @xcite , which include leading - order qcd calculations . a minimum cut - off value @xmath150 of 2.5 gev was applied at the mc generator level to the transverse momenta of the outgoing partons in the hard scattering process .
the herwig event generator was used to check the pythia results . in pythia
, the photon flux is calculated using the weizscker - williams approximation .
the parton densities used were grv lo @xcite for the photon and cteq4 lo @xcite for the proton .
the hadronization in pythia was performed using the lund string model as implemented in jetset @xcite . in herwig
, the hadronization of partons is based on a cluster model . for comparison ,
the lac1 @xcite parameterization for the photon and mrsa lo @xcite for the proton were also used .
previous studies have shown that including a simulation of multiparton interactions ( mi ) in the parton shower programs significantly improves the description of jet production in the forward region@xcite .
this option , which applies only to resolved processes , adds interactions between the partons in the remnants of the proton and photon , calculated as lo qcd processes , to the hardest scattering process of the event .
it was implemented in the herwig simulation@xcite .
the energy corrections for jets were determined from the monte carlo samples by comparing the true transverse energy of a jet ( found by applying the algorithm to the final state particles ) to the transverse energy measured in the calorimeter simulation .
the correction to the jet energies was on average @xmath151 , varying between @xmath152 and @xmath153 depending upon @xmath154 .
the largest corrections occurred at boundaries within the calorimeter .
no correction was applied to the jet pseudorapidity , since the average shift in @xmath155 between the true and detected jets was less than @xmath1560.05 for all @xmath155 values in the range used for the cross section measurements . in each event , the two jets with the highest transverse energies were selected . for the dijet events with a neutron tag , the data were corrected using monte carlo programs based on pompyt2.5 @xcite and rapgap2.06 @xcite .
these simulations include pion - exchange processes where a virtual pion is emitted from the incoming proton ( see appendix ) .
as discussed later , such a model reproduces the characteristics of the data .
the pompyt generator makes use of the program pythia to simulate @xmath157 interactions via resolved and direct photon processes .
these programs simulate higher order effects through the use of leading - order parton showers .
hadronization processes are implemented by jetset ( the lund string model ) . in both programs grv - lo
was used for the parton densities of the photon and smrs - p3 @xcite for the parton densities of the pion .
the conclusions are independent of the monte carlo model used for the corrections : the jet energy corrections and acceptance calculations , both inclusive and neutron - tagged , can be performed with any of the four monte carlo programs ( pompyt , rapgap , herwig , pythia ) without significant change in the results .
a detailed study of the sources contributing to the systematic uncertainties of the measurements was carried out@xcite .
those associated with the ctd and cal , which impact on the jet measurement , and those associated with the fnc , which impact on the neutron measurement , are considered separately . for the jet measurements ,
the uncertainties are grouped into the following classes : * absolute energy scale of the cal : the energy scale uncertainty of the low @xmath65 cal jets used in this study is 5% , leading to an uncertainty of 15 to 20% on the cross section ; * model dependence : for the inclusive dijet cross section , the jet - acceptance correction was performed using herwig instead of pythia .
for the neutron - tagged sample , the acceptance correction was changed from pompyt to rapgap .
the associated uncertainties were at the 5% level ; * parton parameterization : changing the parton densities of the proton ( cteq to mrsa ) contributed an uncertainty of 0.4% ; changing the parton densities of the photon ( grv to lac1 ) 2% ; * event selection : variation of each selection cut by one standard deviation of the resolution gave uncertainties of about 5% .
the main systematic uncertainties associated with the fnc were : * absolute energy scale of the fnc : the uncertainty on the fnc energy scale is @xmath158%@xcite .
this introduced a 1.5% normalization uncertainty on the cross section ; * beam - gas background : charge - exchange processes for beam - gas interactions can produce a high - energy nucleon which might overlap with a dijet event .
the uncertainty in the correction of 2% is less than 1% ; * event selection : veto - counter noise ( from beam halo and calorimeter albedo ) and veto - counter inefficiencies resulted in a 2.5% uncertainty ; * angular distribution of neutrons : the acceptance of the fnc was uncertain due to uncertainties in the angular distribution of the neutrons .
the uncertainty in the acceptance was estimated by using different parameterizations of the pion flux . to obtain a model - independent estimate ,
the angular distribution at fixed @xmath93 was assumed to fall exponentially with @xmath130 , @xmath159 , and the acceptance and its systematic uncertainty were determined by varying the slope parameter @xmath160 as a function of @xmath93 within the limits allowed by the data@xcite .
the resulting acceptance uncertainty for neutrons with energy @xmath72 gev was 5.5% ; * sensitivity of the description of the beam - line : the calculated acceptance of the fnc depends on a complete and accurate description of the proton beam - line between the interaction point and the calorimeter .
both the amount of inactive material and the alignment must be known .
the model was tested by comparing the fraction of neutrons in the fnc surviving the selection cuts , according to rapgap , to that observed .
the discrepancy between the expected and observed fraction gave a normalization uncertainty of 6% .
the dominant systematic uncertainty is that associated with the energy scale of the cal jets .
the statistical errors and the systematic uncertainties were added in quadrature and are shown as error bars in the figures .
the systematic uncertainty associated with the absolute energy scale is shown as a shaded band .
the systematic uncertainties from the fnc give a 9% normalization uncertainty on the neutron - tagged cross section .
since this uncertainty does not affect the shape of any distribution , it is not included in the figures .
in addition , a normalization uncertainty of 1.5% from the luminosity determination was not included ; this is not relevant in the measurement of the neutron - tagged to inclusive event - rate ratios .
the usual simulation models for hard photoproduction processes such as herwig or pythia contain a fraction of events with a leading neutron , although it is not , a priori , expected that such simulations will properly describe particle production in the proton fragmentation region . for leading neutrons produced with high longitudinal momentum and low transverse momentum , particle - exchange models may be more appropriate . in this case
the lo cross section can be expressed as : @xmath161 \label{eqn : fac}\end{aligned}\ ] ] where the exchanged meson is denoted by @xmath61 , and a sum over all exchanged mesons is implied .
the dijet cross section in charge - exchange photoproduction contains contributions from both the direct and resolved processes . in eq .
( [ eqn : fac ] ) , @xmath162 is the splitting function of a positron into a photon and positron , @xmath163 is the splitting function of a proton into a meson and neutron ( i.e. , the flux of mesons in the proton ) , @xmath164 is the density of partons of type @xmath105 in the photon , and @xmath165 is the density of partons of type @xmath166 in the meson .
the sum in @xmath167 runs over all possible types of partons @xmath105 present in the photon and @xmath166 in the meson .
the sum in @xmath168 runs over all possible types of final state partons .
the term @xmath169 is the cross section for the two - body collision @xmath170 and depends on the square of the center - of - momentum - frame energy , @xmath171 , the transverse momentum of the two outgoing partons ( @xmath172 ) , and the momentum scale ( @xmath173 ) at which the strong - coupling constant ( @xmath174 ) is evaluated . for the direct process , eq .
( [ eqn : fac ] ) also holds except that @xmath175 is replaced by the dirac delta function , @xmath176 , and there is no sum over partons @xmath105 in the photon .
equation ( [ eqn : fac ] ) incorporates the assumption of factorization .
in particular , the resolved cross section depends on four `` parton '' densities and a two - body scattering cross section .
the kinematic variables @xmath88 , @xmath93 , @xmath177 and @xmath109 are coupled only through the @xmath178 dependence of the hard - scattering cross section , @xmath179 . a priori
, the shape of the neutron - tagged jet cross sections depends on the kinematic variables of the neutron . in a complete factorization of the baryon and the photon vertices , however , the shape of @xmath177 and the jet - variable distributions would not depend either on the presence of a neutron or explicitly on its kinematic variables
similarly , the energy spectrum of the neutrons would be independent of the photon and jet variables .
meson - exchange models , in the context of regge theory , are often used to describe nucleon charge - exchange reactions . since the mass of the pion is small compared to all other mesons , pion exchange ( see appendix )
is expected to dominate the amplitude for the @xmath60 transition , with small contributions from @xmath62 , @xmath63 , etc .
in fact , lo one - pion - exchange models account for both the shape and normalization of the distributions for the neutron - tagged data .
figure [ fig : fig4](a ) compares the shape of the uncorrected neutron energy spectrum to the prediction of the one - pion - exchange ( ope ) model pompyt for the monopole and light - cone pion form - factors .
the pompyt ope model with the light - cone form - factor agrees well with the measured energy spectrum .
rapgap gives essentially the same prediction .
the figure also shows that the monopole choice for the form factor is disfavored .
it gives a distribution which is both shifted in energy and too narrow .
for this comparison , the pion parton densities have been used ; however , the predicted shape of the neutron spectrum is insensitive to the choice of parton density of the pion@xcite ( not shown ) .
these parton densities are constrained by dimuon and prompt - photon production data from fixed - target experiments that are sensitive mainly to the valence quark distributions , and the parameterizations are similar in the @xmath109 range studied here . figure [ fig : fig4](b ) shows a comparison of the shape of the neutron energy distribution with the predictions of the monte carlo programs pythia and herwig .
neither simulation provides a good description of the data .
moreover , pythia ( herwig ) predicts a leading neutron ( @xmath72 gev ) in 2% ( 0.5% ) of the dijet events in comparison with the @xmath180% ( after correction for the acceptance of the fnc ) observed in the data .
the monte carlo models considered here do not contain an explicit diffractive component ; however , neutrons can also be produced by the diffractive dissociation of the incoming proton through pomeron exchange .
monte carlo studies indicate that such neutrons will have an energy spectrum which agrees qualitatively with that observed in the data .
diffractive processes give rise to a large rapidity gap between the hadronic system and the remnant of the proton , which is either a single proton or a low - mass system with the quantum numbers of the proton .
the pseudorapidity of the most - forward hadron ( @xmath181 ) in the central detector was used to select diffractive events .
meson exchange can also give rise to events with a large rapidity gap via the deck mechanism@xcite in which the exchanged meson itself scatters diffractively off the incoming photon and escapes undetected down the beam pipe .
diffractively dissociating protons are expected to account for only a small fraction of the dijet events , both inclusive and neutron - tagged , because the @xmath65 of diffractive events is severely limited by the small fraction of the proton energy carried by the pomeron ( @xmath182% ) .
in contrast , in neutron production there is on average a much larger fraction of the initial proton energy ( @xmath183% ) available for jet production .
figure [ fig : figeta](a ) shows the @xmath181 distribution for both inclusive and neutron - tagged dijet photoproduction , normalized to equal area .
the shapes of the two distributions are similar .
although there are differences between the two distributions at large @xmath181 , this region is insensitive to diffractive processes .
less than 1% of each sample has a large rapidity gap ( @xmath184 ) .
in addition , the distributions of the jet variables , @xmath77 , @xmath154 , and @xmath112 for events with a large rapidity gap@xcite are different from those of the neutron - tagged sample , which strongly resemble the inclusive sample ( see section [ sec : factorization ] ) .
pompyt and rapgap also reproduce the @xmath181 distribution of neutron - tagged events , as seen in fig .
[ fig : figeta](b ) , although neither monte carlo contains an explicit diffractive component .
neutrons can also be produced indirectly through the production and decay of baryonic resonances .
most prominent of these is the @xmath185 , which is itself produced directly through @xmath61 , @xmath62 or @xmath63 exchange , and which can decay via or .
monte carlo studies indicate that such decay neutrons will also have an energy spectrum which agrees qualitatively with the data ; however , the study of hadronic interactions@xcite shows that the @xmath186 contribution to the fock state of the proton is approximately half that of @xmath187@xcite .
when account is taken of the branching ratio for @xmath188 , such indirect neutron production is small compared to the direct @xmath60 transition .
it can therefore be concluded that standard fragmentation processes , diffractively dissociating protons , and the decay of the @xmath185 resonance are ruled out as the dominant source of dijet events tagged with a leading neutron .
the data will be further compared with the one - pion - exchange model in section [ sec : ope ] .
the differential dijet cross sections as a function of @xmath77 and @xmath154 are shown in fig .
[ fig : fig6](a , b ) for the inclusive sample , and in fig .
[ fig : fig6](c , d ) for the neutron - tagged sample .
the predictions of pythia and herwig describe the shape of the @xmath77 distribution reasonably well .
the predictions have been normalized to the measured cross section at high in order to facilitate the shape comparison . for the @xmath154 distribution ,
the monte carlo is normalized at small @xmath154 , .
herwig with multiparton interactions is in better agreement with the data than pythia without such interactions .
the study of ratios of neutron - tagged to inclusive cross sections is advantageous since many systematic uncertainties , especially those related to the jet measurements , are greatly reduced .
in addition , the ratio provides a quantitative comparison of the shapes of the cross sections , and so tests factorization . that the neutron - tagged and inclusive differential cross sections have similar shapes as a function of @xmath77 is evident in fig .
[ fig : fig6](e ) .
the cross sections fall by over two orders of magnitude in the range @xmath189 gev , but the ratio is approximately constant as a function of @xmath77 .
figure [ fig : fig6](f ) shows that the ratio as a function of @xmath154 falls slightly with increasing @xmath78 .
the ratios of rapgap to pythia and rapgap to herwig with multiparton interactions are also shown in figs . [
fig : fig6](e , f ) .
the ratio of rapgap to herwig is in better agreement with the data , as expected from the cross section comparisons .
the agreement of rapgap and pompyt with the tagged data supports factorization , which is built into these mc models . to test further the factorization properties of the dijet cross section
, the neutron - tagged and inclusive samples were divided into bins of @xmath112 and @xmath77 .
the shape of the observed energy spectrum of the neutron is approximately independent of @xmath112 and @xmath190 , as seen in fig .
[ fig : fig7](a ) and ( b ) .
moreover , in a given bin of @xmath112 , the fraction of events with a leading neutron is approximately independent on @xmath77 , as seen in fig .
[ fig : fig7](c )
. the @xmath112 distribution is determined by the parton densities in the colliding particles , kinematic factors , and a possible presence of interactions between the hadron remnants .
figures [ fig : fig8](a , b ) show the uncorrected @xmath112 distribution for inclusive and neutron - tagged events , respectively .
the corresponding ratio of neutron - tagged to inclusive production is shown in fig .
[ fig : fig8](c ) , corrected only for the acceptance of the fnc . in sharp contrast to the results shown in fig .
[ fig : fig6 ] , the ratio rises with increasing @xmath112 .
the rise is only partially explained by the monte carlo models .
the monte carlo predictions are shown area normalized to the data in figs . [
fig : fig8](a , b ) ; the ratio of rapgap to herwig normalized at @xmath191 is shown in fig .
[ fig : fig8](c ) . figure [ fig : fig8](d ) shows the ratio of the resolved ( @xmath192 ) and direct ( @xmath193 ) photoproduction cross sections as a function of @xmath77 for inclusive events while fig .
[ fig : fig8](e ) show the same quantity for neutron - tagged events .
the size of the direct component increases with @xmath194 for both the neutron - tagged and the inclusive samples .
overall the direct component is approximately twice as large in the neutron - tagged sample .
figure [ fig : fig8](f ) shows that the ratio of the two ratios is constant within errors as a function of @xmath77 indicating that , although the proportion changes , the shape in @xmath77 is the same for both samples . in summary ,
the shape of the neutron energy spectrum is approximately independent of the the jet variables @xmath77 , @xmath78 and @xmath112 ; the jet variables @xmath77 and @xmath78 are relatively insensitive to the presence of a neutron ; however , the ratio of neutron - tagged to inclusive dijet events rates rises with increasing @xmath112 .
in the previous two sections , it was shown that standard fragmentation processes , diffractively dissociating protons , and the decay of the @xmath185 resonance are not the dominant source of dijet events tagged with a leading neutron , but that the data are consistent with one - pion exchange and approximately satisfy factorization .
it is therefore appropriate in this section to make further comparisons to ope under the assumption that it is the dominant mechanism for the production of the neutron - tagged events .
the predictions of rapgap simulations using the smrs - p3 parton densities for the pion and the light - cone pion form - factor are in good agreement with the @xmath77 spectrum for the neutron - tagged events of fig .
[ fig : fig6](c )
. pompyt , which is not shown , gives a similar result .
both ope monte carlo models also adequately describe the shape of the @xmath154 distribution of fig .
[ fig : fig6](d ) , where only rapgap is shown .
the leading - order ope model is also able to account for the normalization of the data , as shown by the rapgap prediction .
in contrast to the inclusive case ( fig .
[ fig : fig6](b ) ) , the monte carlo simulations reproduce the forward @xmath154 region without a simulation of multiparton interactions .
this comparison suggests that hadron remnant interactions are less important for the neutron - tagged sample , in agreement with naive expectations based on their lower effective center - of - momentum - frame energy compared to the inclusive sample .
the rapgap and pompyt simulations are also in fair agreement with the @xmath112 distribution for the neutron - tagged events ( fig .
[ fig : fig8](b ) ) . as in the case of the @xmath78 distribution ,
the ope monte carlo models describe the data without simulation of mi , which is needed in the case of the inclusive sample ( fig .
[ fig : fig8](a ) ) . according to the monte carlo models
, the mi contribution in the inclusive sample strongly increases at low values of @xmath112@xcite .
therefore the ratio of neutron - tagged to inclusive events , which increases with @xmath112 ( fig .
[ fig : fig8](c ) ) , can be interpreted to be at least partially caused by the difference in remnant - remnant interactions of the two samples .
the differential cross section for neutron - tagged events as a function of @xmath114 ( see eq .
[ eqn : xpi ] ) is shown in fig . [
fig : tag_pion ] .
the measured cross section is compared in fig .
[ fig : tag_pion ] to the predictions of the rapgap monte carlo model using the light - cone pion form - factor and the smrs - p3 pion parton densities .
the shape of the @xmath114 distribution disfavors the monopole form - factor ( not shown ) , but both the shape and normalization are reproduced by both rapgap and pompyt ( not shown ) .
the results for other parameterizations of the pion parton densities , which are determined from hadron - hadron interactions , are indistinguishable .
the bands show the systematic uncertainty due to the calorimeter energy scale .
there is an additional normalization error of 9% which is not shown .
differential cross sections for the inclusive reaction and the neutron - tagged reaction have been measured in photoproduction for @xmath195 gev@xmath70 and @xmath196 , with jet transverse energy @xmath197 gev , neutron energy @xmath198 gev and neutron production angle @xmath199 0.8 mrad .
such neutrons are observed in @xmath180% of the events .
the shape of the neutron energy spectrum is observed to be independent of both @xmath77 and @xmath112 .
in addition , the ratio of neutron - tagged to inclusive dijet production rates is approximately independent of the jet transverse energy and pseudorapidity .
these observations support the idea of factorization of the positron and proton vertices
. the above ratio does depend on @xmath112 , the fraction of the initial photon momentum participating in the hard interaction ; the direct to resolved fraction is approximately twice as large in the neutron - tagged sample as in the inclusive sample .
the @xmath112 distribution for the inclusive dijet process can be reproduced in a simulation which includes multiparton interactions .
in contrast the neutron - tagged data are well simulated by lo monte carlo models including one - pion exchange but without multiparton interactions .
these comparisons suggest that the rising ratio of neutron - tagged to inclusive dijet rates as a function of @xmath112 is at least partially due to the difference in remnant - remnant interactions in the two samples .
the standard photoproduction monte carlo models , pythia and herwig , fail to reproduce either the neutron production rate or the neutron energy spectrum .
in contrast , lo monte carlo models such as pompyt and rapgap , which include one - pion exchange and which assume factorization of the pion flux and pion structure , reproduce all aspects of the neutron - tagged data for both the neutron and jet kinematic variables .
we are especially grateful to the desy directorate whose encouragement and financial support made possible the construction and installation of the fnc .
we are also happy to acknowledge the efforts of the desy accelerator group and the support of the desy computing group .
we thank h. jung and g. ingelman for their help with the monte carlo programs rapgap and pompyt .
although there are many choices found in the literature for the splitting function of a proton into a pion and neutron , they all can be summarized conveniently in the form@xcite : @xmath200 where @xmath201 is the coupling at the @xmath202 vertex , @xmath203 is the mass of the pion , and @xmath204 is the regge trajectory of the pion .
the trajectory is often omitted by setting the slope constant @xmath205 , instead of to the measured value , @xmath206 gev@xmath207 .
@xmath208 is a form - factor which accounts for final state rescattering of the neutron .
possible choices for the form - factor @xmath208 are : @xmath209 where @xmath160 , @xmath210 and @xmath211 are constants@xcite .
the choice of form - factor depends on whether the pion s regge trajectory is included or excluded . for the flux with the regge trajectory
the form factor is usually chosen to be an exponential@xcite .
the light - cone form - factor is usually associated with the flux without the regge trajectory@xcite .
these choices are associated with the experimental observation that in hadronic interactions the shape of the @xmath212 distribution depends on @xmath64@xcite .
the experimental data are consistent with the flux with the trajectory term included and a constant form factor ( that is , @xmath213)@xcite .
beam - gas data at hera are also consistent with a small value for @xmath160@xcite ; kopeliovich et al@xcite use @xmath214 gev@xmath207 . for the light - cone form - factor and the flux without the pion s regge trajectory , holtmann et al.@xcite take @xmath215 gev@xmath1 .
a calculation@xcite of the pion form - factor for the one - boson - exchange potential ( obep ) suggests that the monopole with @xmath216 gev is a good approximation ; however , rapgap ( see section [ sec : montecarlo ] ) takes @xmath217 gev@xcite and the flux without the regge trajectory . in the region of the peak of the energy spectrum , @xmath218
, the regge and light - cone splitting functions differ by @xmath219% .
it should be noted that choices other than those given in eq .
( [ eqn : form ] ) are possible .
in particular , recent studies@xcite suggest that the the experimental data is best described by a form factor that is hard for small momentum transfers and soft for large momentum transfers .
other isovector meson exchanges , such as the @xmath62 or @xmath63 , can also contribute to direct neutron production .
recent theoretical studies of neutron production in @xmath220 collisions show that processes other than direct pion exchange are expected to contribute @xmath221% of neutron production@xcite . these backgrounds to ope , which increase the rate of neutron production in the fnc phase space , are offset by absorptive rescattering of the neutron , which decreases the rate by approximately the amount of the increase@xcite .
also absorptive rescattering preferentially removes neutrons with larger @xmath96 , increasing the pion contribution relative to the @xmath62 and @xmath63 .
therefore these effects are also neglected in the present analysis .
zeus fnc group , s. bhadra et al .
, in _ proceedings of the seventh international conference on calorimetry in high energy physics _ , tucson , arizona , usa , november 9 - 14 , 1997 , eds .
e. cheu et al , ( world scientific , singapore , 1998 ) .
j. huth et al . , in _ proceedings of the 1990 dpf study on high energy physics _ , snowmass , colorado , ed .
e. l. berger ( world scientific , singapore , 1992 ) p. 134 ; + ua1 collaboration , g. arnison et al .
b 123 ( 1983 ) 115 . , and
( b ) @xmath130 covered by the angular acceptance of the fnc ( @xmath223 mrad ) are shown as shaded bands .
the solid lines show the average @xmath64 and @xmath130 , respectively , as function of the neutron energy @xmath224 . ]
distribution for neutron - tagged and inclusive dijet photoproduction data .
( b ) the @xmath181 distribution for neutron - tagged dijet photoproduction compared to the predictions of pompyt and rapgap .
the inclusive distribution is area normalized to the tagged distribution .
only statistical errors are shown . ] and @xmath154 for both inclusive ( a , b ) and neutron - tagged ( c , d ) photoproduction .
the kinematic region studied is , , , , , and . for the inclusive cross sections the monte carlo predictions have been normalized to the measured cross sections at high @xmath190 and negative @xmath154 , respectively , in order to facilitate the comparison of the shapes .
the tagged data are compared with the predictions of the one - pion - exchange model rapgap using the smrs - p3 pion parton densities and the light - cone pion form - factor .
the bands show the systematic uncertainty due to the calorimeter energy scale .
the error bars show the statistical error added in quadrature with the remaining systematic error .
the tagged cross sections have an additional normalization uncertainty of 9% which is not shown .
also shown are the ratio of the cross sections of neutron - tagged to inclusive dijet photoproduction as a function of ( e ) @xmath77 , i.e. @xmath225 $ ] , and ( f ) @xmath154 , i.e. @xmath226 $ ] ; and monte carlo predictions for the ratio using rapgap to pythia and rapgap to herwig . ] and ( b ) @xmath190 .
the spectra are area - normalized .
( c ) the ratio of the cross - sections of neutron - tagged to inclusive dijet photoproduction as a function of @xmath77 for different ranges of @xmath112 ( applied to both the inclusive and tagged samples ) .
the points are offset slightly to improve visibility .
only statistical errors are shown .
there is an overall normalization uncertainty ( not shown ) of 9% for ( c ) . ]
distribution for ( a ) inclusive events compared to the expectations of herwig with multiple interactions , and ( b ) for neutron - tagged events compared to the expectations of pompyt and rapgap .
the ratio of the neutron - tagged to inclusive @xmath112 distributions is shown in ( c ) together with the ratio of rapgap to herwig mi .
the monte carlo predictions are area normalized to the data in ( a , b ) and normalized at @xmath191 in ( c ) .
the ratio of the resolved to direct cross sections as a function of @xmath77 is shown in ( d ) and ( e ) for inclusive and neutron - tagged photoproduction , respectively . resolved ( direct events )
are defined by @xmath192 ( @xmath193 ) .
the ratio of the ratios shown in ( d ) and ( e ) , i.e. @xmath227 $ ] , is shown in ( f ) .
only statistical errors are shown . ] , the fraction of the exchanged pion s momentum participating in the production of the dijet system for the neutron - tagged sample .
the measured cross section is compared to the prediction of the monte carlo model rapgap using the light - cone pion form - factor and the smrs - p3 pion parton densities .
the bands show the systematic uncertainty due to the calorimeter energy scale .
the error bars show the statistical error added in quadrature with the remaining systematic error .
there is an additional normalization error of 9% which is not shown . ] | differential cross sections for dijet photoproduction in association with a leading neutron using the reaction @xmath0 have been measured with the zeus detector at hera using an integrated luminosity of 6.4 pb@xmath1 .
the fraction of dijet events with a leading neutron in the final state was studied as a function of the jet kinematic variables .
the cross sections were measured for jet transverse energies , neutron energy , and neutron production angle @xmath2 mrad .
the data are broadly consistent with factorization of the lepton and hadron vertices and with a simple one - pion - exchange model .
desy 00 - 142 + october 2000 + -1.cm -0.3 cm -0.3
cm 16.cm 0.cm 0.3 cm plus0.05 cm minus0.05 cm the zeus collaboration j. breitweg , s. chekanov , m. derrick , d. krakauer , s. magill , b. musgrave , a. pellegrino , j. repond , r. stanek , r. yoshida + _ argonne national laboratory , argonne , il , usa _ @xmath3 m.c.k .
mattingly + _ andrews university , berrien springs , mi , usa _ p. antonioli , g. bari , m. basile , l. bellagamba , d. boscherini@xmath4 , a. bruni , g. bruni , g. cara romeo , l. cifarelli@xmath5 , f. cindolo , a. contin , m. corradi , s. de pasquale , p. giusti , g. iacobucci , g. levi , a. margotti , t. massam , r. nania , f. palmonari , a. pesci , g. sartorelli , a. zichichi + _ university and infn bologna , bologna , italy _ @xmath6 c. amelung@xmath7 , a. bornheim@xmath8 , i. brock , k. cobken@xmath9 , j. crittenden@xmath4 , r. deffner@xmath10 , h. hartmann , k. heinloth@xmath11 , e. hilger , p. irrgang , h .- p .
jakob , a. kappes@xmath12 , u.f .
katz , r. kerger , e. paul , j. rautenberg , + h. schnurbusch , a. stifutkin , j. tandler , k.c .
voss , a. weber , h. wieber + _ physikalisches
institut der universitt bonn , bonn , germany _
@xmath13 d.s .
bailey , o. barret , n.h .
brook@xmath14 , b. foster@xmath4 , g.p .
heath , h.f .
heath , e. rodrigues@xmath15 , j. scott , r.j .
tapper + _ h.h .
wills physics laboratory , university of bristol , bristol , u.k . _ @xmath16 m. capua , a. mastroberardino , m. schioppa , g. susinno + _ calabria university , physics dept.and infn , cosenza , italy _
@xmath6 h.y .
jeoung , j.y .
kim , j.h .
lee , i.t .
lim , k.j .
ma , m.y .
pac@xmath17 + _ chonnam national university , kwangju , korea _
@xmath18 a. caldwell , w. liu , x. liu , b. mellado , s. paganis , s. sampson , w.b .
schmidke , f. sciulli + _ columbia university , nevis labs .
, irvington on hudson , n.y . , usa _ @xmath19 j. chwastowski , a. eskreys , j. figiel , k. klimek , k. olkiewicz , k. piotrzkowski@xmath7 , m.b .
przybycie , p. stopa , l. zawiejski + _ inst . of nuclear physics , cracow , poland _ @xmath20 b. bednarek , k. jele , d. kisielewska , a.m. kowal , t. kowalski , m. przybycie , e. rulikowska - zarbska , l. suszycki , d. szuba + _ faculty of physics and nuclear techniques , academy of mining and metallurgy , cracow , poland
_ @xmath20 a. kotaski + _ jagellonian univ . ,
dept . of physics , cracow , poland _
@xmath21 l.a.t .
bauerdick , u. behrens , j.k .
bienlein , k. borras , v. chiochia , d. dannheim , k. desler , g. drews , , u. fricke , f. goebel , s. goers , p. gttlicher , r. graciani , t. haas , w. hain , g.f .
hartner , k. hebbel , s. hillert , w. koch@xmath22@xmath23 , u. ktz , h. kowalski , h. labes , b. lhr , r. mankel , j. martens , m. milite , m. moritz , d. notz , m.c .
petrucci , a. polini , m. rohde@xmath11 , a.a .
savin , , f. selonke , m. sievers@xmath24 , s. stonjek , g. wolf , u. wollmer , c. youngman , + _ deutsches elektronen - synchrotron desy , hamburg , germany _ c. coldewey , , a. meyer , , p.b .
straub + _ desy zeuthen , zeuthen , germany _ g. barbagli , e. gallo , a. parenti , p. g. pelfer + _ university and infn , florence , italy _ @xmath6 a. bamberger , a. benen , n. coppola , s. eisenhardt@xmath25 , p. markun , h. raach , s. wlfle + _ fakultt fr physik der universitt freiburg i.br . ,
freiburg i.br .
, germany _
@xmath13 p.j .
bussey , m. bell , a.t .
doyle , c. glasman , s.w .
lee@xmath26 , a. lupi , n. macdonald , g.j .
mccance , d.h .
saxon , l.e .
sinclair , i.o .
skillicorn , r. waugh + _ dept . of physics and astronomy , university of glasgow , glasgow ,
u.k .
_ @xmath16 i. bohnet , n. gendner , u. holm , a. meyer - larsen , h. salehi , k. wick + _ hamburg university , i. institute of exp .
physics , hamburg , germany _
@xmath13 t. carli , a. garfagnini , i. gialas@xmath27 , l.k .
gladilin@xmath28 , d. kira@xmath29 , r. klanner , e. lohrmann + _ hamburg university , ii .
institute of exp .
physics , hamburg , germany _ @xmath13 r. gonalo@xmath15 , k.r .
long , d.b .
miller , a.d .
tapper , r. walker + _ imperial college london , high energy nuclear physics group , london , u.k . _ @xmath16 p. cloth , d. filges + _ forschungszentrum jlich , institut fr kernphysik , jlich , germany _
t. ishii , m. kuze , k. nagano , k. tokushuku@xmath30 , s. yamada , y. yamazaki + _ institute of particle and nuclear studies , kek , tsukuba , japan _
@xmath31 s.h .
ahn , s.b .
lee , s.k .
park + _ korea university , seoul , korea _
@xmath18 h. lim@xmath26 , d. son + _ kyungpook national university , taegu , korea _
@xmath18 f. barreiro , g. garca , o. gonzlez , l. labarga , j. del peso , i. redondo@xmath32 , j. terrn , m. vzquez + _ univer .
autnoma madrid , depto de fsica terica , madrid , spain _
@xmath33 m. barbi , f. corriveau , d.s .
hanna , a. ochs , s. padhi , d.g .
stairs , m. wing + _ mcgill university , dept . of physics , montral , qubec , canada _ @xmath34 @xmath35 t. tsurugai + _ meiji gakuin university , faculty of general education , yokohama , japan _ a. antonov , v. bashkirov@xmath36 , m. danilov , b.a .
dolgoshein , d. gladkov , v. sosnovtsev , s. suchkov + _ moscow engineering physics institute , moscow , russia _ @xmath37 r.k .
dementiev , p.f .
ermolov , yu.a .
golubkov , i.i .
katkov , l.a .
khein , n.a .
korotkova , + i.a .
korzhavina , v.a .
kuzmin , o.yu .
lukina , a.s .
proskuryakov , l.m .
shcheglova , a.n .
solomin , n.n .
vlasov , s.a .
zotkin + _ moscow state university , institute of nuclear physics , moscow , russia _
@xmath38 c. bokel , m. botje , n. brmmer , j. engelen , s. grijpink , e. koffeman , p. kooijman , s. schagen , a. van sighem , e. tassi , h. tiecke , n. tuning , j.j .
velthuis , j. vossebeld , l. wiggers , e. de wolf + _ nikhef and university of amsterdam , amsterdam , netherlands _
@xmath39 b. bylsma , l.s .
durkin , j. gilmore , c.m .
ginsburg , c.l .
kim , t.y . ling + _ ohio state university , physics department , columbus , ohio , usa _
@xmath3 s. boogert , a.m. cooper - sarkar , r.c.e .
devenish , j. groe - knetter@xmath40 , t. matsushita , o. ruske , + m.r .
sutton , r. walczak + _ department of physics , university of oxford , oxford u.k .
_ @xmath16 a. bertolin , r. brugnera , r. carlin , f. dal corso , s. dusini , s. limentani , a. longhin , m. posocco , l. stanco , m. turcato + _ dipartimento di fisica dell universit and infn , padova , italy _ @xmath6 l. adamczyk@xmath41 , l. iannotti@xmath41 , b.y .
oh , j.r . okrasiski , p.r.b .
saull@xmath41 , w.s .
toothacker@xmath22@xmath23 , + j.j .
whitmore + _ pennsylvania state university , dept . of physics ,
university park , pa , usa _ @xmath19 y. iga + _ polytechnic university , sagamihara , japan _
@xmath31 g. dagostini , g. marini , a. nigro + _ dipartimento di fisica , univ .
la sapienza and infn , rome , italy _ @xmath42 c. cormack , j.c .
hart , n.a .
mccubbin , t.p .
shah + _ rutherford appleton laboratory , chilton , didcot , oxon , u.k .
_ @xmath16 d. epperson , c. heusch , h.f .- w .
sadrozinski , a. seiden , r. wichmann , d.c .
williams + _ university of california , santa cruz , ca , usa _ @xmath3 i.h .
park + _ seoul national university , seoul , korea _
n. pavel + _ fachbereich physik der
universitt - gesamthochschule siegen , germany _
@xmath13 h. abramowicz@xmath43 , s. dagan@xmath44 , s. kananov@xmath44 , a. kreisel , a. levy@xmath44 + _ raymond and beverly sackler faculty of exact sciences , school of physics , tel - aviv university , tel - aviv , israel _ @xmath45 t. abe , t. fusayasu , t. kohno , k. umemori , t. yamashita + _ department of physics , university of tokyo , tokyo , japan _ @xmath31 r. hamatsu , t. hirose , m. inuzuka , s. kitamura@xmath46 , k. matsuzawa , t. nishimura + _ tokyo metropolitan university , dept . of physics , tokyo , japan
_ @xmath31 m. arneodo@xmath47 , n. cartiglia , r. cirio , m. costa , m.i .
ferrero , s. maselli , v. monaco , c. peroni , m. ruspa , r. sacchi , a. solano , a. staiano + _ universit di torino , dipartimento di fisica sperimentale and infn , torino , italy _ @xmath6 d.c .
bailey , c .- p .
fagerstroem , r. galea , t. koop , g.m .
levman , j.f .
martin , a. mirea , a. sabetfakhri + _ university of toronto , dept . of physics , toronto , ont .
, canada _
@xmath48 j.m .
butterworth , c.d .
catterall , m.e .
hayes , e.a .
heaphy , t.w .
jones , j.b . lane , b.j . west + _ university college london , physics and astronomy dept . , london , u.k .
_ @xmath16 j. ciborowski , r. ciesielski , g. grzelak , r.j .
nowak , j.m .
pawlak , r. pawlak , b. smalska , + t. tymieniecka , a.k .
wrblewski , j.a .
zakrzewski , a.f .
arnecki + _ warsaw university , institute of experimental physics , warsaw , poland _ @xmath20 m. adamus , t. gadaj + _ institute for nuclear studies , warsaw , poland
_ @xmath20 o. deppe , y. eisenberg , d. hochman , u. karshon@xmath44 + _ weizmann institute , department of particle physics , rehovot , israel _ @xmath49 w.f .
badgett , d. chapin , r. cross , c. foudas , s. mattingly , d.d .
reeder , w.h .
smith , a. vaiciulis@xmath50 , t. wildschek , m. wodarczyk + _ university of wisconsin , dept . of physics , madison ,
wi , usa _
@xmath3 a. deshpande , s. dhawan , v.w .
hughes + _ yale university , department of physics , new haven , ct , usa _
@xmath3 s. bhadra , c. catterall , j.e .
cole , w.r .
frisken , r. hall - wilton , m. khakzad , s. menary + _ york university , dept . of physics , toronto , ont . ,
canada _
@xmath48 @xmath51 now visiting scientist at desy + @xmath52 now at univ . of salerno and infn napoli , italy + @xmath53
now at cern + @xmath54 now at caltech , usa + @xmath55 now at sparkasse bonn , germany + @xmath56 now at siemens icn , berlin , germanny + @xmath57 retired + @xmath58 supported by the gif , contract i-523 - 13.7/97 + @xmath59 pparc advanced fellow + @xmath15 supported by the portuguese foundation for science and technology ( fct ) + @xmath17 now at dongshin university , naju , korea + @xmath22 deceased + @xmath24 now at netlife ag , hamburg , germany + @xmath25 now at university of edinburgh , edinburgh , u.k .
+ @xmath26 partly supported by an icsc - world laboratory bjrn h. wiik scholarship + @xmath27 visitor of univ . of crete ,
greece , partially supported by daad , bonn - kz .
+ @xmath28 on leave from msu , supported by the gif , contract i-0444 - 176.07/95 + @xmath29 supported by daad , bonn - kz . a/98/12712 + @xmath30 also at university of tokyo
+ @xmath32 supported by the comunidad autonoma de madrid + @xmath36 now at loma linda university , loma linda , ca , usa + @xmath40 supported by the feodor lynen program of the alexander von humboldt foundation + @xmath41 partly supported by tel aviv university + @xmath43 an alexander von humboldt fellow at university of hamburg + @xmath44 supported by a minerva fellowship + @xmath46 present address : tokyo metropolitan university of health sciences , tokyo 116 - 8551 , japan + @xmath47 now also at universit del piemonte orientale , i-28100 novara , italy + @xmath50 now at university of rochester , rochester , ny , usa + [ cols= " > , < " , ] -1.5 cm -0.3 cm -0.3 cm 16.cm 0.cm 0.3 cm plus0.05 cm minus0.05 cm |
since the simple modules of a ring and its semi - simple quotient are the same , for many purposes it suffices to consider the latter ring . in this note
we study the problem of calculating the semi - simple quotient of the group ring @xmath16 for the automorphism group of a finite @xmath1-group .
the usual method is to consider the maximal elementary abelian quotient @xmath17 , where @xmath18 is the frattini subgroup .
the induced map @xmath19 has a @xmath1-group as kernel by the hall basis theorem .
hence the map of algebras @xmath20 has a nilpotent kernel and thus suffices to compute the semi - simple quotient .
however this map is not necessarily onto and one is left with the still considerable problem of determining the image . for @xmath21 ,
when @xmath1 is an odd prime , this has been done by dietz @xcite giving a complete determination of @xmath5 .
we adopt an inductive approach via extensions ; that is , we assume @xmath2 is given as an extension @xmath22 with @xmath4 , @xmath6 under control .
then there is an exact sequence relating the automorphism groups of @xmath13 and @xmath14 with that of @xmath2 , depending on the cohomology class of the extension @xmath7 \in h^2(q , z(n))$ ] where @xmath23 denotes the center of @xmath13 .
our motivation comes from stable homotopy theory .
let @xmath24 be a finite group and @xmath1 be a prime .
the classifying space of @xmath24 completed at @xmath1 , @xmath25 , decomposes stably into a wedge of indecomposable summands @xmath26 each summand @xmath27 is the mapping telescope of a primitive idempotent @xmath28 in the ring of stable self - maps of @xmath25 , @xmath29 .
thus there is a one - to - one correspondence between the indecomposable summands and the simple modules of the ring of stable self - maps .
this correspondence is explored in both @xcite and @xcite ( see also @xcite ) .
it turns out that modular representation theory plays a crucial role : if @xmath2 is a sylow @xmath1-subgroup of @xmath24 then each indecomposable summand of @xmath25 originates in @xmath30 for some subgroup @xmath31 and corresponds to a simple @xmath32 module . of course , the automorphism group of a group is of intrinsic interest in its own right , and our methods shed some light on its structure .
an outline of the paper follows : section 2 covers the preliminaries on extensions , @xmath33 , including the fundamental exact sequence , theorem [ thmwb ] , relating @xmath4 , @xmath34 , @xmath6 , and the extension class @xmath7 $ ] . in section 3
we define and identify a group structure , @xmath8)$ ] , on the intersection of the two orbits @xmath35 $ ] and @xmath10 $ ] where @xmath4 and @xmath6 act on @xmath36 in the usual way .
this group , which we call the intersection orbit group , is useful in computing @xmath37 .
the case of trivial action ( or twisting ) of @xmath14 on @xmath23 is considered in section 4 .
extensions with @xmath13 , @xmath14 , elementary abelian @xmath1-groups are studied in section 5 . in case @xmath38
this involves the theory of quadratic forms over @xmath39 and the arf invariant .
we recall browder s classification theorem @xcite and give several results describing the order of a quotient of @xmath34 by a normal @xmath1-subgroup in section 6 . for more complicated @xmath1-groups @xmath24 we describe an inductive procedure for extending these results using the mod-@xmath1 lower central series .
section 7 is devoted to several applications of the theory . in what follows
all groups are assumed finite , except as noted in section 3 .
we begin by recalling the results of c. wells @xcite as extended by j. buckley @xcite . because their notation is now non - standard , e.g. , functions written on the right , we re - couch these results in more standard notation .
let @xmath40 be an extension of the group @xmath13 by the group @xmath14 and let @xmath41 be the group of automorphisms of @xmath24 mapping @xmath13 to itself .
the obvious homomorphism @xmath42 provides a means of studying @xmath41 . as usual two extensions @xmath43 ,
@xmath44 are _ equivalent _ @xmath45 if there is an isomorphism @xmath46 restricting to the identity on @xmath13 and inducing the identity on @xmath14 .
the set of such equivalent extensions is denoted @xmath47 .
the _ twisting _ ( or _ coupling _ ) @xmath48 of @xmath49 is the homomorphism defined as usual by @xmath50 where @xmath51 , @xmath52 .
equivalent extensions have the same twisting .
the center @xmath53 of @xmath13 has the structure of a @xmath14 module via a homomorphism @xmath54 defined by the composite @xmath55 where @xmath56 is induced by @xmath57 .
it is well - known that we may identify @xmath58 where @xmath59 ranges over all twistings @xmath60 .
now consider @xmath61 and form the extension @xmath62 then @xmath63 acts on @xmath47 from the left by @xmath64 = [ \sigma e \tau^{-1 } ] .
\label{eq : a}\ ] ] one checks @xmath65 = ( \sigma\sigma',\tau\tau')[e]$ ] and @xmath66= [ e]$ ] .
the twisting of @xmath67 is given by @xmath68 where @xmath69 denotes conjugation by @xmath70 , the image of @xmath71 in @xmath72 .
for a given @xmath59 define the subgroup @xmath73 by @xmath74 that is , the following diagram commutes [ eq : b ] @xmath75 the subgroup @xmath76 consists of all ordered pairs @xmath77 that preserve the twisting .
if @xmath59 is trivial then clearly @xmath78 .
we note that @xmath79 plays no role in the commutativity of the diagram and @xmath80 .
thus if the sequence @xmath81 splits , e.g. , if @xmath14 is elementary abelian , then @xmath82 can be an arbitrary linear isomorphism .
then @xmath76 acts on @xmath83\ |\ e \ has \ twisting \ \chi\}$ ] .
it is trivial to check that @xmath84 so we consider @xmath85 as a homomorphism @xmath86 .
let @xmath87 denote the group of derivations , i.e. , functions @xmath88 satisfying @xmath89 for @xmath90 . then there is a homomorphism @xmath91 defined by @xmath92 . finally we define a function @xmath93 by restricting the action of @xmath94 on the extension class @xmath7 $ ] , that is , @xmath95 $ ]
. in general @xmath96 is not a homomorphism .
the following is the principal result of @xcite as extended by @xcite : [ thmwb ] for a given extension @xmath33 with twisting @xmath97 there is an exact sequence @xmath98 with @xmath99}$ ] , the isotropy subgroup of @xmath76 fixing @xmath7 $ ] .
the map @xmath96 is not onto and is only a set map .
an alternate exact sequence results by replacing @xmath87 with the cohomology group @xmath100 and @xmath41 by @xmath101 where @xmath102 is group of inner automorphisms of @xmath24 induced by elements of @xmath53 , ( see @xcite , proposition iv 2.1 )
. we shall be interested in @xmath103 so we will not need this refinement . in case @xmath53
is a @xmath1-group theorem [ thmwb ] shows @xmath104 is a normal @xmath1-subgroup of @xmath41 which in turn implies @xmath105})$ ] suffices to compute simple modules and idempotents : [ corb ] if @xmath106 is an extension with @xmath53 a p - group , then @xmath107})$ ] is surjective with nilpotent kernel .
_ proof : _ since @xmath108 is a @xmath1-group it is well known that @xmath109 is nilpotent
. we shall be interested in determining when @xmath41 is a @xmath1-group ; clearly we have [ corbaa ] if @xmath6 and @xmath4 are @xmath1-groups then @xmath41 is a @xmath1-group .
[ corba ] suppose @xmath24 is a @xmath1-group with @xmath1 odd , and @xmath13 is generated by the elements of @xmath24 of order @xmath1 .
then @xmath34 is a @xmath1-group if @xmath4 is a @xmath1-group . _
proof : _ by hypothesis @xmath110 is a characteristic subgroup , hence @xmath111 . if @xmath112 has @xmath113 order then @xmath114 . since @xmath1 is odd , the automorphisms of @xmath24 which have @xmath113 order and fix @xmath13 are trivial by theorem 5.3.10 , @xcite .
thus @xmath34 is a @xmath1-group .
let @xmath115 be a left @xmath116 set where @xmath117 , @xmath118 are groups , not necessarily finite .
equivalently @xmath117 acts on the left , @xmath118 acts on the right such that @xmath119 , and @xmath120 , where @xmath121 , @xmath122 , @xmath123 . as usual
let @xmath124 denote the orbit of @xmath125 under the action of @xmath117 ( respectively @xmath126 denote the orbit of @xmath125 under the action of @xmath118 ) .
we define the _ intersection orbit group at @xmath125 _ @xmath127 if @xmath128 , @xmath129 are elements of @xmath130 , their product is defined by @xmath131 it is straightforward to check that this pairing is well - defined giving @xmath130 the structure of a group .
although left and right actions commute , @xmath130 is not necessarily abelian ; however , if either @xmath117 or @xmath118 is a @xmath1-group , then @xmath130 is a @xmath1-group .
also if @xmath117 or @xmath118 is trivial then obviously @xmath132 , the trivial group .
again , as usual , let @xmath133 and @xmath134 denote the respective isotropy subgroups .
[ propaz ] there is an isomorphism of groups @xmath135 given by @xmath136 .
_ proof : _
@xmath137 if and only if @xmath138 .
similarly @xmath139 if and only if @xmath140 .
thus @xmath141 is well - defined and bijective .
it is a homomorphism by the definition of the product in @xmath130 .
[ corab ] if @xmath117 and @xmath118 are finite groups , then @xmath142 divides @xmath143 is the largest subset of @xmath144 which is a group .
thus @xmath145 is a subgroup and so @xmath142 divides @xmath146 and hence @xmath147 .
similarly for @xmath118 .
thus @xmath142 divides @xmath148 . returning to automorphism groups
, the following result often simplifies computing @xmath149 .
[ corabb ] for a given extension of finite groups @xmath150 , \i ) @xmath151}|\cdot |{aut(n)}_{[e]}| \cdot |\omega([e])|$ ] \ii ) @xmath103 is a @xmath1-group if and only if @xmath152}$ ] , @xmath153}$ ] , and @xmath154)|$ ] are @xmath1-groups . _
proof : _ ii ) follows from i ) which follows immediately from proposition [ propaz ] with @xmath155 , @xmath156 .
there exists extensions that do not depend on the intersection orbit group .
[ propo ] suppose @xmath13 is an abelian @xmath1-group , @xmath6 is a @xmath1-group , and @xmath157 with a trivial twisting .
then there exists a non - split extension @xmath7 $ ] such that @xmath158 _ proof : _ by corollary [ corab ] , it suffices find @xmath7 $ ] with @xmath154)| = 1 $ ] .
let @xmath159 be a sylow @xmath1-subgroup of @xmath94 .
the @xmath94-module @xmath160 has cardinality a power of @xmath1 .
thus there is an @xmath159 fixed point , @xmath7 \neq 0 $ ] by the fixed point result @xcite , p. 64 , which generalizes to this case ) .
thus @xmath161 } = \{[e]\}$ ] and @xmath162 .
the kernel of the exact sequence in theorem [ thmwb ] is @xmath163 , which explains the presence of that term in the proposition .
[ propa ] if @xmath33 has trivial twisting the exact sequence of theorem [ thmwb ] reduces to @xmath164 _ proof : _ @xcite , th . 3.1 .
clearly @xmath165 and @xmath78 .
before proceeding , it is instructive to consider the simplest case @xmath166 . the extension class is trivial thus @xmath167 is surjective .
in fact @xmath85 is split by the usual inclusion @xmath168 . for a more complete discussion of @xmath169 see @xcite . in what follows we are interested in @xmath170 , a @xmath1-group . as an immediate corollary of [ propa ] we have [ corcb ]
let @xmath2 be a @xmath1-group and define @xmath171 , @xmath172 .
suppose @xmath173 is a @xmath1-group for each @xmath174 .
then @xmath5 is a @xmath1-group . in particular
this hypothesis holds if @xmath38 and for each @xmath174 the summands of @xmath175 have distinct exponents .
_ proof : _ the first statement is clear from corollary [ corbaa ] . similarly the second follows from proposition 4.5 of @xcite which implies each @xmath176 is a @xmath15-group .
another particularly tractable case is that of @xmath177 , with @xmath13 an abelian , characteristic subgroup . since @xmath178 tends to be relatively large in this case , the quotient @xmath103 can be significantly smaller than @xmath34 .
we examine this phenomenon in more detail .
let @xmath179 be the _ mod - p lower central series _
, that is @xmath180 and @xmath181 where @xmath182 is the @xmath183-fold iterated commutator .
then @xmath184 is the frattini subgroup and @xmath185 is the largest elementary abelian quotient of @xmath2 .
for some @xmath186 , @xmath187 .
[ exa ] for @xmath1 an odd prime we consider the group @xmath188 , trivial higher commutators @xmath189 we shall determine @xmath190 using the extension @xmath191 where @xmath192 and @xmath193 are elementary abelian .
this extension has trivial twisting and its extension cocycle is easily seen to be @xmath7 = ( x\wedge y ) \otimes c \in h^2(v , q ) = h^2(v ) \otimes \mathbf{z}/p \langle z \rangle$ ] where @xmath194 are dual to @xmath195 respectively . by inspection @xmath196 thus @xmath197 .
we conclude @xmath198 is surjective with nilpotent kernel .
finally we note @xmath199 , thus @xmath200 and @xmath201 .
[ exb ] let @xmath202 be the unipotent subgroup of upper triangular matrices over @xmath39 .
these matrices necessarily have ones on the diagonal .
thus @xmath203 is generated by @xmath204 where @xmath205 and @xmath206 is the standard elementary matrix with @xmath207 in the @xmath208 position and zeros elsewhere .
let @xmath209 .
then there is a central extension @xmath210 since @xmath13 is the commutator subgroup of @xmath2 , it is characteristic .
the extension cocycle is @xmath211 = y_{12}y_{23}\otimes \overline{x}_{13 } + y_{23}y_{34 } \otimes \overline{x}_{24 } + y_{34}y_{45 } \otimes \overline{x}_{35}\ ] ] where @xmath212 and @xmath213 is dual to @xmath214 . since the twisting is trivial , @xmath215 .
direct calculation shows @xmath216 } = \langle a , b \
a^4 = b^2 = 1 , bab = a^{-1 } \rangle\ ] ] the dihedral group of order @xmath217 generated by @xmath218 @xmath219 moreover @xmath220 , @xmath5 is a @xmath15 group of order @xmath221 .
let @xmath222 be a @xmath1-group . since @xmath223 is characteristic there is an induced homomorphism @xmath224 which factors as @xmath225 @xmath226 we shall describe an inductive procedure for lifting elements in the image of this map .
consider the extensions @xmath227 @xmath228 where @xmath44 and @xmath229 have trivial twisting . in each case
the kernel is a characteristic subgroup .
let @xmath230 . by theorem [ thmwb ] , @xmath231 if and only if there exist @xmath232 such that @xmath233 fixes @xmath234 \in h^2(v , \gamma^1(p)/\gamma^2(p)).\ ] ] then there exists @xmath235 lifting @xmath236 . since @xmath237 this completes the initial step .
now suppose inductively that we have found elements @xmath238 , @xmath239 such that @xmath240 are lifts of @xmath241 , respectively .
we need to find @xmath242 such that @xmath243 fixes @xmath244 \in h^2_{\overline{\chi}}(v , z ( \gamma^1(p)/\gamma^{i+1}(p))).\ ] ] then by thm [ thmwb ] there exists @xmath245 lifting @xmath236 . to find @xmath246
we apply the same technique to the extension @xmath247 noting that the twisting is trivial so condition ( 1 ) is trivially satisfied .
thus we must find @xmath248 such that @xmath249 fixes @xmath250 \in h^2 ( \gamma^1(p)/\gamma^i(p ) , \gamma^i(p)/\gamma^{i+1}(p))\ ] ] since @xmath251 is its own center .
then applying theorem [ thmwb ] again there exists @xmath252 lifting @xmath253 as desired .
the induction terminates when @xmath254 .
the procedure described in this section is demonstrated in the examples in section 7 .
we consider extensions @xmath255 where @xmath256 are elementary abelian @xmath1-groups with @xmath13 central . in this case
the twisting is trivial , @xmath257 and @xmath258 .
our aim is to use corollary [ corabb ] to compute @xmath149 .
let @xmath259 .
then the extension cocycle @xmath7 \in h^2(v;n)$ ] has the form @xmath7 = ( x_1 , x_2 , \dots , x_n)$ ] where @xmath260 .
we recall that @xmath261 acts diagonally on @xmath7 $ ] , @xmath262 = ( \sigma x_1 , \sigma x_2 , \dots , \sigma x_n)$ ] for @xmath263 .
thus the isotropy subgroup @xmath264 } = aut(v)_{x_1 } \cap \cdot \cdot \cdot \cap aut(v)_{x_n}\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ ( 3)\ ] ] the action of @xmath4 on @xmath7 $ ] is induced from that on @xmath13 . at this point
we restrict our attention to the case @xmath38 .
let @xmath265 then each @xmath27 is a quadratic form in @xmath266 the generators of @xmath267 $ ] , @xmath268 .
we recall some classical facts about quadratic forms @xmath269 , @xcite , @xcite , @xcite .
the defining property is that the associated form @xmath270 is alternate bilinear .
the _ bilinear radical _ of @xmath118 , @xmath271 as usual , @xmath118 is called _ non - degenerate _ if @xmath272 , i.e. its matrix is non - singular .
the _ radical _ of @xmath14 , @xmath273 @xmath14 is said to be _ non - degenerate _ if @xmath274 . by a theorem of dickson @xcite ( section 199 ) a ( non - zero ) quadratic form over @xmath39 in @xmath275 variables
which is not equivalent ( by a change of basis ) to one in fewer variables must be equivalent to one of the following standard non - degenerate quadratic forms @xmath276 @xmath277 @xmath278 if @xmath272 , then @xmath279 is even and one can define the _ arf invariant _ of @xmath14 with respect to a symplectic basis @xmath280 by @xmath281 this is invariant of the choice of symplectic basis and determines @xmath14 up to equivalence .
it is convenient to write @xmath282 multiplicatively .
then with this notation , w. browder has shown that @xmath283 if and only if @xmath14 sends the majority of elements of @xmath284 to @xmath285 respectively @xcite . for @xmath275
even one finds @xmath286 , @xmath287 .
the arf invariant can be extended to the @xmath275 odd case ( where @xmath118 is degenerate ) as follows .
it is easy to see that @xmath288 sends the same number of elements to @xmath207 and @xmath289 thus one can define @xmath290 .
it is clear that browder s definition ( also known as the democratic invariant " ) is invariant under any basis change .
this leads to the following classification theorem .
[ thmwy]@xcite , theorem iii.1.14 a quadratic form @xmath291 is determined up to equivalence by the triple @xmath292 the action of @xmath293 on an @xmath186-tuple @xmath294 of quadratic forms is linear and thus involves the sum of forms .
unfortunately it is impossible , in general , to determine the sum from the arf invariant .
for example if @xmath295 and @xmath296 then @xmath297 .
however , on a direct sum of vector spaces then it follows easily from the definition that @xmath298 for the rest of this subsection we restrict attention to @xmath275 even and consider the special case @xmath7 = ( x_1 , x_2 , \dots , x_n)$ ] where the @xmath27 are in standard form .
we write @xmath299 or @xmath300 depending on whether the arf invariant is @xmath207 or @xmath289 . in the following theorem we use @xmath301 to denote conjugacy .
[ thmxy ] suppose @xmath279 and @xmath7 = ( x_1 , x_2 , \dots , x_n)$ ] with @xmath302 , @xmath303
. then @xmath8 ) = [ e]$ ] and @xmath304}|\cdot |aut(n)_{[e]}|$ ] .
furthermore \1 ) if @xmath305 then @xmath264 } = o_{m}^+(\mathbf{f}_2)\ ] ] the orthogonal group of order @xmath306 of matrices preserving the form @xmath115 .
@xmath307}\sim \left ( \begin{matrix } 1 & 0\\ * & gl_{n-1}(\mathbf{f}_2)\\ \end{matrix } \right)\ ] ] of order @xmath308 .
@xmath103 is a @xmath15-group if and only if @xmath309 .
\2 ) if @xmath310 then @xmath264 } = o_{m}^-(\mathbf{f}_2)\ ] ] the orthogonal group of order @xmath311 of matrices preserving the form @xmath300 and @xmath307 } \sim \left ( \begin{matrix } 1 & 0\\ * & gl_{n-1}(\mathbf{f}_2)\\ \end{matrix } \right).\ ] ] @xmath103 is a @xmath15-group if and only if @xmath312 and @xmath313 .
\3 ) if @xmath314 then @xmath264 } = o_{m}^+(\mathbf{f}_2 ) \cap o_{m}^-(\mathbf{f}_2)\ ] ] and @xmath307 } \sim \left ( \begin{matrix } i_2 & 0\\ * & gl_{n-2}(\mathbf{f}_2)\\ \end{matrix } \right)\ ] ] of order @xmath315 .
@xmath103 is a @xmath15-group if and only if @xmath316 .
_ proof : _ the calculation of @xmath317}$ ] follows from ( 3 ) .
the intersection orbit group @xmath8)= ( aut(v)\cdot [ e ] ) \cap ( aut(n)\cdot [ e ] ) = \ { [ e ] \}$ ] . to see this suppose @xmath262 = [ e]{\tau}^{-1}$ ] differs from @xmath7 = ( x_1,x_2,\dots , x_n)$ ] in the @xmath174-th coordinate @xmath299 , say .
now @xmath318{\tau}^-1)_i = ax + by$ ] , @xmath319 .
thus @xmath320 since @xmath321 and @xmath300 has arf invariant @xmath289 .
however @xmath322 contradicting the fact that @xmath115 is not equivalent to a quadratic form in fewer than @xmath275 variables . similarly if @xmath323 . since @xmath115 and @xmath300 are linearly independent polynomials , @xmath324 is equivalent to @xmath325 by a change of basis
. thus the descriptions of @xmath326}$ ] follow immediately .
we now turn to the more general case where the forms @xmath27 of @xmath7 = ( x_1 , x_2 , \dots , x_n)$ ] are not in standard form .
one can still determine @xmath149 ; we illustrate this by considering the case @xmath327 , @xmath328 .
thus we consider pairs of quadratic forms @xmath329 in variables @xmath330 each equivalent to ( but not necessarily equal to ) one of the standard four forms : @xmath331 the respective isotropy groups , as subgroups of @xmath332 , are @xmath333 non - abelian of order @xmath334 ; @xmath335 elementary abelian of rank @xmath336 ; @xmath337 of order @xmath338 ; @xmath339 of order @xmath338 .
first we consider the case where @xmath115 and @xmath300 are simultaneously equivalent to a standard form , i.e. , there is an invertible linear transformation @xmath117 of @xmath284 such that @xmath340 and @xmath341 are each in standard form .
case 1 , @xmath342 : then @xmath343 . it is also easy to see @xmath344 and the intersection of orbits @xmath8 ) = \{(x , x)\}$ ] in this case .
\a ) @xmath115 is equivalent to @xmath345 as above the isotropy subgroup @xmath346 is elementary abelian of rank @xmath336 .
the intersection of the orbits is just @xmath347 so the order of @xmath348 .
\b ) @xmath115 is equivalent to @xmath349 @xmath350 is of order @xmath338 .
thus @xmath351 .
\c ) @xmath115 is equivalent to @xmath352 @xmath353 .
thus @xmath354 .
\d ) @xmath115 is equivalent to @xmath355 : as above @xmath350 is of order @xmath338 .
thus @xmath356 .
case 2 , @xmath357 : in this case @xmath358 .
further @xmath359 .
there are several possibilities ; we give only four in detail since the rest follow the same general pattern . by considering the arf invariant
we see that @xmath8 ) = 1 $ ] except in the third example .
\1 ) @xmath329 equivalent to @xmath360 : then @xmath361 is @xmath362 the intersection of orbits is @xmath363 hence @xmath364 .
\2 ) @xmath329 equivalent to @xmath365 : then @xmath366 is @xmath367 i.e. , dihedral of order @xmath217 .
the intersection of orbits is @xmath363 thus @xmath368 .
similarly for @xmath369 and @xmath370 .
\3 ) @xmath329 equivalent to @xmath371 : then @xmath372 is trivial and @xmath8 ) = \{(x , y ) , ( x+y , y ) \ } = \mathbf{z}/2 $ ] .
thus @xmath373 .
\4 ) @xmath329 equivalent to @xmath374 : then @xmath361 is trivial .
direct calculation shows the intersection of the orbits has order @xmath334 .
thus @xmath375 .
similarly for @xmath376 , @xmath377 , and @xmath378 . by this we mean @xmath115 and @xmath300 are not simultaneously equivalent to a pair of standard forms . since @xmath357 , @xmath379 in all cases . to illustrate this phenomena the following table gives a complete computation of @xmath149 in case @xmath115 and @xmath300 are equivalent ( non - simultaneously ) to @xmath380 .
then @xmath115 may be assumed to be @xmath380 and we list only the relevant @xmath300 s and the corresponding values of @xmath381 . *
@xmath149 * [ cols="^,^,^",options="header " , ] in more detail , if @xmath373 , then @xmath382 and @xmath8 ) = \{(x , y ) , ( y , x)\ } = \mathbf{z}/2 $ ] if @xmath383 , then @xmath384 and @xmath8 ) = \{(x , y ) , ( x+y , x ) , ( y , x+y ) \ } = \mathbf{z}/3 $ ] .
if @xmath385 then @xmath386 and @xmath8 ) = \{(x , y ) , ( y , x ) \ } = \mathbf{z}/2 $ ] . as a final example we consider @xmath387 , @xmath388 . to analyze @xmath389 we separately reduce @xmath115 , @xmath300 to standard forms @xmath390 respectively @xmath391 with @xmath392 .
let @xmath393 , then @xmath394 hence @xmath395 \sigma\ ] ] this determines @xmath389 up to conjugacy since @xmath396\sigma_1 $ ] .
in this example @xmath397 , @xmath398 with @xmath399 , @xmath400
. then @xmath401 \sigma = \mathbf{z}/2 \bigg\langle \left ( \begin{matrix } 1 & 0 & 0\\ 0 & 1 & 0\\ 0 & 1 & 1\\ \end{matrix } \right)\bigg\rangle\ ] ] further @xmath8 ) = \{(x , y),(x , x+y)\ } = \mathbf{z}/2 $ ] hence @xmath402 .
let @xmath2 denote the extraspecial group of order @xmath403 and exponent @xmath404 defined by the central extension @xmath405 where @xmath406 , @xmath407 .
the twisting @xmath59 is trivial thus @xmath408 .
complete information about the automorphism group of @xmath2 as well as all other extensions of elementary abelian @xmath1-groups by @xmath409 is known @xcite,@xcite . here
we apply our results to obtain a quick derivation of @xmath410 .
@xmath2 is generated by elements @xmath411 of order @xmath1 satisfying @xmath412 = 1\ ] ] @xmath413 = \zeta\ ] ] @xmath414 = 1 , \ \
j \neq 2i\ ] ] @xmath415 = 1 , \ \
j \neq 2i-1\ ] ] thus @xmath416 and @xmath417 .
it is immediate from these relations that the extension cocycle is @xmath211 = b \otimes \zeta\ ] ] where @xmath418 , @xmath419 is dual to @xmath420 . since @xmath1 is an odd prime , @xmath421 .
thus @xmath118 is exactly the skew - symmetric form for the symplectic group @xmath422 .
we shall need a slightly more general version @xmath423 , @xcite , the transformations which fix @xmath118 up to a scalar .
it is easy to see that @xmath424 where @xmath425 is the linear transformation @xmath426 and @xmath427 is a generator of @xmath428 .
let @xmath435 be the _
universal w - group _
@xcite on @xmath186 generators defined as the central extension @xmath436 where @xmath437 .
these groups arise , for instance , in the study of the cohomology of galois groups .
the extension class @xmath7 = ( x_1 , x_2 , \dots , x_{n + \binom n2})$ ] where the @xmath438 form an ordered basis for @xmath439 , the second symmetric power of the dual of @xmath14 .
any order will do .
the twisting is trivial , hence @xmath440 .
now @xmath441 induces linear isomorphisms on @xmath442 which are determined by their values on squares in @xmath443 .
thus given @xmath444 we can find a @xmath445 such that @xmath446 ) = ( \sigma [ e]){\tau}^{-1 } = [ e]$ ] .
thus we see that @xmath447 by an injection which projects to an isomorphism on the first factor , i.e. @xmath448 . * 3 . * we consider @xmath5 for the unipotent group @xmath449 of upper triangular matrices over the finite field @xmath39 .
these matrices necessarily have ones on the diagonal .
thus @xmath2 is generated by @xmath450 .
then @xmath451 , @xmath452 , and @xmath453 .
the corresponding extensions have @xmath454 where @xmath455 , @xmath456 and @xmath457 the twisting @xmath59 is trivial for @xmath44 but not for @xmath458 . in this application
we shall not need the auxiliary extensions @xmath459 $ ] .
_ proof : _ ( sketch ) from the commutator relations @xmath461
= x_{ik}$ ] for @xmath462 , one sees that the extension class @xmath463 \in h^2(v ; \gamma^1(p)/\gamma^2(p))$ ] is @xmath464 where @xmath465 @xmath466 \otimes \mathbf{z}/2 \langle { x}_{13 } , { x}_{24 } \rangle\ ] ] here @xmath467 are classes dual to @xmath468 respectively .
direct calculation shows the @xmath463 $ ] is fixed the subgroup generated by the involution defined by @xmath469 and the map of order three @xmath470 @xmath471 one finds @xmath472 . turning to the extension @xmath458 , we can extend @xmath71 , @xmath473 to @xmath474 by letting them act identically on @xmath475
. then a simple calculation shows @xmath476 .
we note that @xmath477 = [ e_2]$ ] considered as an element of @xmath478 .
this follows from the definition of the cocycle @xmath479 for @xmath463 $ ] .
recall that given a set theoretic section @xmath480 then @xmath481 if we use one of the sections not involving the center , then it lifts to a section @xmath482 and the corresponding cocycle @xmath483 is a lift of @xmath484 .
( for example using the ordered basis @xmath485 for @xmath284 let @xmath486 . for products
@xmath487 let @xmath488 where @xmath489 is ordered lexicographically . thus if @xmath490 then @xmath491
. hence @xmath492 implies @xmath493 in general we observe that no two - fold products among the elements of @xmath494 involve the center i.e. , two - fold products among the elements @xmath495 are on the first and second diagonals , not in the center . ) we shall study @xmath5 for @xmath501 , the smallest prime for which @xmath2 is regular .
we consider the extensions of @xmath502 @xmath234 : \ \ \
n_2 : = \gamma^1(p)/\gamma^2(p)\to p/\gamma^2(p ) \to v\ ] ] @xmath503 : \ \ \ \ \ \ \ \ n_3
: = \gamma^1(p ) \to p \to v\ ] ] since @xmath504 , we also have @xmath505 : \ \ \ \widetilde{n}_{3 } : = \gamma^2(p ) \to \gamma^1(p ) \to n_2\ ] ] each @xmath13 is abelian : @xmath506 extension @xmath44 has trivial twisting and is quite similar to that of example 1 . using the same notation and arguing analogously
we can determine the extension class @xmath463 = xy\otimes z \in h^2(v , n_2)$ ] .
hence @xmath509 } = \langle ( a , b)\in gl_2(\mathbf{f}_p)\times gl_1(\mathbf{f}_p ) \
\ | \ \ b = det(a ) \rangle\ ] ] extension @xmath510 splits , @xmath511 .
thus every @xmath512 lifts to @xmath513 and we proceed to study extension @xmath458 .
@xmath284 on @xmath515 is given by @xmath516 , @xmath517 .
hence @xmath518 .
since @xmath519 , it must have the form @xmath520 now solving @xmath521 where @xmath522 we find @xmath523 where @xmath524 .
thus @xmath525 and @xmath526 .
hence @xmath527 to determine @xmath528}$ ] it is easier in this case to observe that @xmath529 ( using magma ) and argue directly instead of using the extension class .
first @xmath530 contains a subgroup of order @xmath531 generated by the inner automorphisms of order @xmath532 , @xmath533 together with the automorphism @xmath534 on @xmath535 .
next we show that @xmath103 contains a subgroup of order @xmath536 which must equal @xmath103 . | we use extensions to study the semi - simple quotient of the group ring @xmath0 of a finite @xmath1-group @xmath2 . for an extension @xmath3 , our results involve relations between @xmath4 , @xmath5 , @xmath6 and the extension class @xmath7\in h^2(q , zn)$ ] . one novel feature is the use of the _ intersection orbit group _ @xmath8)$ ] , defined as the intersection of the orbits @xmath9 $ ] and @xmath10 $ ] in @xmath11 .
this group is useful in computing @xmath12 . in case @xmath13 , @xmath14 are elementary abelian @xmath15-groups our results involve the theory of quadratic forms and the arf invariant
. [ section ] [ thm]lemma [ thm]corollary [ thm]proposition [ thm]example [ thm]definition [ thm]remark [ multiblock footnote omitted ] |
compressed sensing ( cs ) @xcite is an emerging sparse sampling theory which received a large amount of attention in the area of signal processing recently . consider a @xmath1-sparse signal @xmath3 which has at most @xmath1 nonzero entries .
let @xmath4 be a measurement matrix with @xmath5 and @xmath6 be a measurement vector .
compressed sensing deals with recovering the original signal @xmath7 from the measurement vector @xmath8 by finding the sparsest solution to the undetermined linear system @xmath6 , i.e. , solving the the following _ @xmath9-optimization _ problem @xmath10 where @xmath11 denotes the @xmath9-norm or ( hamming ) weight of @xmath7 .
unfortunately , it is well - known that the problem ( [ l0 ] ) is np - hard in general @xcite . in compressed sensing , there are essentially two methods to deal with it .
the first one pursues greedy algorithms for @xmath9-optimization , such as the orthogonal matching pursuit ( omp ) @xcite and its modifications @xcite .
the second method considers a convex relaxation of ( [ l0 ] ) , or the _ @xmath12-optimization _ ( basis pursuit ) problem , as follows @xcite @xmath13 where @xmath14 denotes the @xmath12-norm of @xmath7 .
note that the problem ( [ l1 ] ) could be turned into a linear programming ( lp ) problem and thus tractable .
the construction of measurement matrix @xmath15 is one of the main concerns in compressed sensing . in order to select an appropriate matrix ,
we need some criteria . in their earlier and fundamental works ,
donoho and elad @xcite introduced the concept of _ spark_. the spark of a measurement matrix @xmath15 , denoted by @xmath16 , is defined to be the smallest number of columns of @xmath15 that are linearly dependent , i.e. , @xmath17 where @xmath18 furthermore , @xcite obtained several lower bounds of @xmath16 and showed that if @xmath19 then any @xmath1-sparse signal @xmath7 can be exactly recovered by the @xmath9-optimization ( [ l0 ] ) .
in fact , we will see in the appendix that the condition ( [ proeq1 ] ) is also necessary for both the @xmath9-optimization ( [ l0 ] ) and the @xmath12-optimization ( [ l1 ] ) .
hence , the spark is an important performance parameter of the measurement matrix .
other useful criteria include the well - known restricted isometry property ( rip ) @xcite and the nullspace characterization @xcite .
although most known constructions of measurement matrix rely on rip , we will use spark instead in this paper since the spark is simpler and easier to deal with in some cases . generally , there are two main kinds of constructing methods for measurement matrices : random constructions and deterministic constructions .
many random matrices , e.g. , fourier matrices @xcite , gaussian matrices , rademacher matrices @xcite , _ etc _ , have been verified to satisfy rip with overwhelming probability .
although random matrices perform quite well on average , there is no guarantee that a specific realization works .
moreover , storing a random matrix may require lots of storage space .
on the other hand , a deterministic matrix is often generated on the fly , and some properties , e.g. , spark , girth and rip , could be verified definitely .
there are many works on deterministic constructions @xcite . among these , constructions from coding theory @xcite
attract many attentions , e.g. , amini and marvasti @xcite used bch codes to construct binary , bipolar and ternary measurement matrices , and li _ et al . _ @xcite employed algebraic curves to generalize the constructions based on reed - solomon codes . in this paper , we usually use @xmath15 to denote a real measurement matrix and @xmath20 a binary measurement matrix .
recently , connections between ldpc codes @xcite and cs excite interest .
dimakis , smarandache , and vontobel @xcite pointed out that the lp decoding of ldpc codes is very similar to the lp reconstruction of cs , and further showed that parity - check matrices of good ldpc codes can be used as _ provably _ good measurement matrices under basis pursuit .
ldpc codes are a class of linear block codes , each of which is defined by the nullspace over @xmath21 of a binary sparse @xmath22 parity - check matrix @xmath20 .
let @xmath23 and @xmath24 denote the sets of column indices and row indices of @xmath20 , respectively .
the _ tanner graph _
@xmath25 @xcite corresponding to @xmath20 is a bipartite graph comprising of @xmath26 variable nodes labeled by the elements of @xmath27 , @xmath28 check nodes labelled by the elements of @xmath29 , and the edge set @xmath30 , where there is an edge @xmath31 if and only if @xmath32 . the _ girth _
@xmath33 of @xmath25 , or briefly the girth of @xmath20 , is defined as the minimum length of circles in @xmath25 .
obviously , @xmath33 is always an even number and @xmath34 .
@xmath20 is said to be @xmath35-_regular _ if @xmath20 has the uniform column weight @xmath36 and the uniform row weight @xmath37 .
the performance of an ldpc code under iterative / lp decoding over a binary erasure channel is completely determined by certain combinatorial structures , called stopping sets . a _ stopping set _
@xmath38 of @xmath20 is a subset of @xmath27 such that the restriction of @xmath20 to @xmath38 , say @xmath39 , does nt contain a row of weight one .
the smallest size of a nonempty stopping set , denoted by @xmath40 , is called the stopping distance of @xmath20 . lu _ et al . _
@xcite verified that binary sparse measurement matrices constructed by the well - known peg algorithm @xcite significantly outperform gaussian random matrices by a series of experiments .
similar to the situation in constructing ldpc codes , matrices with girth 6 or higher are preferred in the above two works @xcite . in this paper
, we manage to establish more connections between ldpc codes and cs .
our main contributions focus on the following two aspects .
* _ lower bounding the spark of a binary measurement matrix @xmath20 .
_ as an important performance parameter for ldpc codes , the stopping distance @xmath40 plays a similar role that the spark does in cs .
firstly , we show that @xmath41 , which again verifies the fact that good parity - check matrices are good measurement matrices . a special case of the lower bound is the binary corollary of the lower bound for real matrices in @xcite .
then , a new general lower bound of @xmath42 is obtained , which improved the previous one in most cases .
furthermore , for a class of binary matrices from finite geometry , we give two further improved lower bounds to show their relatively large spark . * _ constructing binary measurement matrices with relatively large spark .
_ ldpc codes based on finite geometry could be found in @xcite . with similar methods ,
two classes of deterministic constructions based on finite geometry are given , where the girth equals 4 or 6 .
the above lower bounds on spark ensure that the proposed matrices have relatively large spark .
simulation results show that the proposed matrices perform well and in many cases significantly better than the corresponding gaussian random matrices under the omp algorithm . even in the case of girth 4 ,
some proposed constructions still manifest good performance . moreover ,
most of the proposed matrices could be put in either cyclic or quasi - cyclic form , and thus the hardware realization of sampling becomes easier and simpler .
the rest of the paper is organized as follows . in section [ fgldpc ]
we give a brief introduction to finite geometries and their parallel and quasi - cyclic structures , which result in the two classes of deterministic constructions naturally .
section [ mainresults ] obtains our main results , or the two lower bounds of spark for general binary matrices and two further improved lower bounds for the proposed matrices from finite geometry .
simulation results and related remarks are given in section [ simulation ] .
finally , section [ conclusion ] concludes the paper with some discussions .
finite geometry was used to construct several classes of parity - check matrices of ldpc codes which manifest excellent performance under iterative decoding @xcite @xcite .
we will see in the later sections that most of these structured matrices are also good measurement matrices in the sense that they often have considerably large spark and may manifest better performance than the corresponding gaussian random matrices under the omp algorithm . in this section ,
we introduce some notations and results of finite geometry @xcite@xcite .
let @xmath43 be a finite field of @xmath44 elements and @xmath45 be the @xmath46-dimensional vector space over @xmath43 , where @xmath47 .
let @xmath48 be the @xmath46-dimensional euclidean geometry over @xmath43 .
@xmath48 has @xmath49 points , which are vectors of @xmath45 .
the @xmath50-flat in @xmath48 is a @xmath50-dimensional subspace of @xmath45 or its coset .
let @xmath51 be the @xmath46-dimensional projective geometry over @xmath43 .
@xmath51 is defined in @xmath52 .
two nonzero vectors @xmath53 are said to be equivalent if there is @xmath54 such that @xmath55 .
it is well known that all equivalence classes of @xmath52 form points of @xmath51 . @xmath51
has @xmath56 points .
the @xmath50-flat in @xmath51 is simply the set of equivalence classes contained in a @xmath57-dimensional subspace of @xmath58 . in this paper , in order to present a unified approach , we use @xmath59 to denote either @xmath48 or @xmath51 .
a point is a @xmath60-flat , a _ line _ is a @xmath61-flat , and a @xmath62-flat is called a _
hyperplane_. for @xmath63 , there are @xmath64 @xmath65-flats contained in a given @xmath66-flat and @xmath67 @xmath68-flats containing a given @xmath69-flat , where for @xmath48 and @xmath51 respectively @xcite @xmath70 @xmath71 let @xmath72 and @xmath73 be the numbers of @xmath69-flats and @xmath66-flats in @xmath59 respectively .
the @xmath69-flats and @xmath66-flats are indexed from @xmath61 to @xmath26 and @xmath61 to @xmath74 respectively .
the incidence matrix @xmath75 of @xmath66-flat over @xmath69-flat is a binary @xmath76 matrix , where @xmath32 for @xmath77 and @xmath78 if and only if the @xmath79th @xmath66-flat contains the @xmath80th @xmath69-flat .
the rows of @xmath20 correspond to all the @xmath66-flats in @xmath59 and have the same weight @xmath64 .
the columns of @xmath20 correspond to all the @xmath69-flats in @xmath59 and have the same weight @xmath67 .
hence , @xmath20 is a @xmath35-regular matrix , where @xmath81 the incidence matrix @xmath20 or @xmath82 will be employed as measurement matrices and called respectively the _ type i or type ii finite geometry measurement matrix_. moreover , by puncturing some rows or columns of @xmath20 or @xmath82 , we could construct a large amount of measurement matrices with various sizes . to obtain submatrices of @xmath20 or @xmath82 with better performance ,
the property of parallel structure in euclidean geometry are often employed as follows . in this class of constructions , an important rule of puncturing the rows or columns of @xmath20 or @xmath82
is to make the remained submatrix as regular as possible .
a possible explanation may come from theorem [ preth ] in the next section .
this rule can be applied since the euclidean geometry has the parallel structure and all @xmath66-flats ( or @xmath69-flat ) can be arranged by a suitable order . since a projective geometry
does not have the parallel structure , we concentrate on @xmath48 only . recall that a @xmath50-flat in @xmath48 is a @xmath50-dimensional subspace of @xmath45 or its coset .
a @xmath50-flat contains @xmath83 points .
two @xmath50-flats are either disjoint or they intersect on a flat with dimension at most @xmath84 .
the @xmath50-flats that correspond to the cosets of a @xmath50-dimensional subspace of @xmath45 ( including the subspace itself ) are said to be parallel to each other and form a parallel bundle .
these parallel @xmath50-flats are disjoint and contain all the points of @xmath48 with each point appearing once and only once .
the number of parallel @xmath50-flats in a parallel bundle is @xmath85 .
there are totally @xmath86 @xmath87-flats which consist of @xmath88 parallel bundles in @xmath48 .
we index these parallel bundles from @xmath89 .
consider the @xmath76 incidence matrix @xmath20 of @xmath87-flat over @xmath90-flat .
all @xmath74 rows of @xmath20 could be divided into @xmath91 bundles each of which contain @xmath92 rows , i.e. , by suitable row arrangement , @xmath20 could be written as @xmath93 where @xmath94 ( @xmath95 ) is a @xmath96 submatrix of @xmath20 and corresponds to the @xmath80-th parallel bundles of @xmath66-flat .
clearly , the row weight of @xmath94 remains unchanged and its column weight is 1 or 0 .
similar to the ordering of rows , the columns of @xmath20 can also be ordered according to the parallel bundles in @xmath97 .
hence , by deleting some row parallel bundles or column parallel bundles from @xmath20 , and transposing the obtained submatrix if needed , we could construct a large amount of measurement matrices with various sizes .
these will be illustrated by several examples in section iv . in this paper , we call @xmath20 or @xmath82 the first class of binary measurement matrices from finite geometry , and their punctured versions the second class of binary measurement matrices from finite geometry . apart from the parallel structure of euclidean geometry , most of the incidence matrices in euclidean geometry and projective geometry also have cyclic or quasi - cyclic structure @xcite .
this is accomplished by grouping the flats of two different dimensions of a finite geometry into cyclic classes .
for a euclidean geometry , only the flats not passing through the origin are used for matrix construction .
based on this grouping of rows and columns , the incidence matrix in finite geometry consists of square submatrices ( or blocks ) , and each of these square submatrices is a circulant matrix in which each row is a cyclic shift of the row above it and the first row is the cyclic shift of the last row .
note that by puncturing the row blocks or column blocks of the incidence matrices , the remained submatrices are often as regular as possible . in other words , this skill is compatible with the parallel structure of euclidean geometry .
hence , the sampling process with these measurement matrices is easy and can be achieved with linear shift registers . for detailed results and discussions ,
we refer the readers to ( * ? ? ?
* appendix a ) .
the definition of spark was introduced by donoho and elad @xcite to help to build a theory of sparse representation that later gave birth to compressed sensing . as we see from ( [ proeq1 ] ) , spark of the measurement matrix
can be used to guarantee the exact recovery of @xmath1-sparse signals . as a result , while choosing measurement matrices , those with large sparks are prefered . however , the computation of spark is generally np - hard @xcite . in this section ,
we give several lower bounds of the spark for general binary matrices and the binary matrices constructed in section ii by finite geometry .
these theoretical results guarantee the good performance of the proposed measurement matrices under the omp algorithm to some extent .
firstly , we give a relationship between the spark and stopping distance of a general binary matrix . for a real vector @xmath98
, the support of @xmath7 is defined by the set of non - zero positions , i.e. , @xmath99 . clearly , @xmath100 .
traditionally , a easily computable property , _ coherence _ , of a matrix is used to bound its spark . for a matrix @xmath4 with column vectors
@xmath101 , the coherence @xmath102 or the matrix @xmath15 has a all - zero column . ]
is defined by @xcite : @xmath103 where @xmath104 denotes the inner product of vectors .
furthermore , it is shown in @xcite that @xmath105 note that this lower bound applies to general real matrices . for the general binary matrix @xmath20
, the next theorem shows that the spark could be lower bounded by the stopping distance .
[ preth0 ] let @xmath20 be a binary matrix .
then , for any @xmath106 , the support of @xmath107 must be a stopping set of @xmath20 .
moreover , @xmath108 assume the contrary that @xmath109 is not a stopping set .
by the definition of stopping set , there is one row of @xmath20 containing only one ` 1 ' on @xmath109 .
then the inner product of @xmath110 and this row will be nonzero , which contradicts with the fact that @xmath111 .
hence , @xmath112 according to the definitions of stopping distance and spark .
let @xmath20 be the @xmath113 parity - check matrix of a binary @xmath114 $ ] hamming code @xcite , which consists of all @xmath28-dimensional non - zero column vectors .
it is easy to check that @xmath115 , which implies that the lower bound ( [ lbpre ] ) could be achieved .
this lower bound verifies again the conclusion that good parity - check matrices are also good measurement matrices .
in particular , consider a binary @xmath22 matrix @xmath20 .
suppose the minimum column weight of @xmath20 is @xmath116 , and the maximum inner product of any two different columns of @xmath20 is @xmath117 . by ( [ defcoh ] ) , we have @xmath118 thus the lower bound ( [ generalspark ] ) from @xcite implies @xmath119 on the other hand , it was proved that @xmath120 @xcite .
hence , the bound ( [ binarysparkold ] ) is a natural corollary of theorem [ preth0 ] . as a matter of fact , for the general binary matrix @xmath20
, we often have a tighter lower bound of its spark . [ preth ]
let @xmath20 be a binary @xmath22 matrix @xmath20 .
suppose the minimum column weight of @xmath20 is @xmath116 , and the maximum inner product of any two different columns of @xmath20 is @xmath117 .
then @xmath121 for any @xmath122 , we split the non - empty set @xmath123 into two parts @xmath124 and @xmath125 , @xmath126 without loss of generality , we assume that @xmath127 . for fixed @xmath128 , by selecting the @xmath79-th column of @xmath20 and all the columns in @xmath125 of @xmath20 , we get a submatrix @xmath129
. since the column weight of @xmath20 is at least @xmath36 , we could select @xmath36 rows of @xmath129 to form a @xmath130 submatrix of @xmath20 , say @xmath131 , where the column corresponds to @xmath79 is all 1 column .
now let s count the total number of 1 s of @xmath131 in two ways .
* from the view of columns , since the maximum inner product of any two different columns of @xmath20 is @xmath132 , each of the columns of @xmath131 corresponds to @xmath125 has at most @xmath132 1 s .
so the total number is at most @xmath133 . * from the view of rows , we claim that there is at least two 1 s in each row of @xmath131 , which implies the total number is at least @xmath134 1 s .
the claim is shown as follows .
let @xmath135 be a row of @xmath131 and @xmath136 be its corresponding row in @xmath20 .
note that @xmath137 . since @xmath138 , @xmath139 which implies that @xmath140 so there are at least one 1 s in @xmath141 and @xmath135 has at least two 1 s .
therefore , @xmath142 , which implies that @xmath143 . since @xmath144 , @xmath145 and the conclusion follows .
note that the matrix in theorem [ preth ] has girth at least 6 if @xmath146 .
when @xmath147 , it is clear that the lower bound ( [ thgeneral ] ) of theorem [ preth ] is tighter than ( [ binarysparkold ] ) .
combining theorem [ preth ] with ( [ proeq1 ] ) , we have that any @xmath1-sparse signal @xmath7 can be exactly recovered by the @xmath9-optimization ( [ l0 ] ) if @xmath148 .
consider a complete graph on 4 vertices .
the incidence matrix is @xmath149 clearly , @xmath150 and the lower bound ( [ thgeneral ] ) is 4 .
moreover , it is easy to check that @xmath151 and @xmath152 is a stopping set , which implies that @xmath153 and @xmath154 .
this result could be generalized to the complete graph on @xmath26 nodes . then spark and stopping distance are also 4 and 3 respectively , which implies that the lower bound ( [ thgeneral ] ) of theorem [ preth ]
could be achieved and may be tighter than one in theorem [ preth0 ] .
clearly , the lower bound of theorem [ preth ] for general binary matrices applies to all ones constructed in section ii by finite geometry . in this subsection
, we will show that for these structured matrices based on finite geometry , more tighter lower bound could be obtained .
let @xmath20 be the @xmath76 incidence matrix of @xmath87-flats over @xmath90-flats in @xmath155 , where @xmath156 , @xmath157 and @xmath158 .
recall that @xmath20 or @xmath82 are called respectively the type - i or tyep - ii finite geometry measurement matrix .
the following lemma is needed to establish our results .
@xcite[lem1 ] let @xmath159 and @xmath160 .
given any @xmath161 different @xmath69-flats @xmath162 in @xmath59 and for any @xmath163 , there exists one @xmath164-flat @xmath165 such that @xmath166 and @xmath167 for all @xmath168 .
[ thm1 ] let @xmath169 be integers , @xmath170 and @xmath20 be the type - i finite geometry measurement matrix .
then @xmath171 where @xmath172 let @xmath173 and assume the contrary that @xmath174 select a @xmath122 such that @xmath175 . by ( [ split1 ] ) and ( [ split2 ] ) , we split the non - empty set @xmath123 into two parts @xmath124 and @xmath125 , and assume @xmath127 without loss of generality .
thus by the assumption @xmath176 for fixed @xmath128 , by selecting @xmath79-th column of @xmath20 and all the columns in @xmath125 of @xmath20 , we get a submatrix @xmath129 .
the number of columns in @xmath129 is @xmath177 and not greater than @xmath178 .
let @xmath179 and @xmath180 be the @xmath69-flats corresponding to the columns of @xmath129 . by lemma [ lem1 ]
, there exists one @xmath164-flat @xmath165 such that @xmath166 and @xmath167 for all @xmath181 .
there are exactly @xmath178 @xmath68-flats containing @xmath165 .
note that among these @xmath66-flats , any two distinct @xmath66-flats have no other common points except those points in @xmath165 ( see @xcite ) .
hence , each of these @xmath178 @xmath68-flats contains the @xmath69-flat @xmath179 and for any @xmath181 , there exist at most one of these @xmath178 @xmath68-flats containing the @xmath69-flat @xmath182 . in other words ,
there exist @xmath178 rows in @xmath129 such that each of these rows has component @xmath61 at position @xmath79 and for any @xmath183 , there exists at most one row that has component @xmath61 at position @xmath80 .
let @xmath184 be the @xmath185 submatrix of @xmath129 by choosing these rows , where the column corresponds to @xmath79 is all 1 column .
now let s count the total number of 1 s of @xmath184 in two ways .
the column corresponds to @xmath79 has @xmath178 @xmath61 s while each of the other columns has at most one @xmath61 .
thus from the view of columns , the total number of @xmath61 s in @xmath184 is at most @xmath186 . on the other hand ,
suppose @xmath187 is the number of rows in @xmath184 with weight one . then , there are @xmath188 rows with weight at least two .
thus from the view of rows , the total number of @xmath61 s in @xmath184 is at least @xmath189 .
hence , @xmath190 , which implies that @xmath191 by the assumption . in other words , @xmath129 contains a row with value @xmath61 at the position corresponding to @xmath79 and @xmath60 at other positions .
denote this row by @xmath135 and let @xmath136 be its corresponding row in @xmath20 .
note that @xmath137 and @xmath192 . since @xmath138 , @xmath193 which leads to a contradiction
therefore , the assumption is wrong and the theorem follows by ( [ fg3 ] ) .
combining theorem [ thm1 ] with ( [ proeq1 ] ) , we have that when the type - i finite geometry measurement matrix is used , any @xmath1-sparse signal @xmath7 can be exactly recovered by the @xmath9-optimization ( [ l0 ] ) if @xmath194 .
[ remark2 ] for the type - i finite geometry measurement matrix @xmath20 , it is known that @xmath195 @xcite .
thus , by theorem [ preth0 ] , @xmath196 obviously , @xmath20 has uniform column weight @xmath197 .
the inner product of two different columns equals to the number of @xmath87-flats containing two fixed @xmath90-flats .
it is easy to see that the maximum inner product is @xmath198 .
thus , by theorem [ preth ] , @xmath199 it is easy to verity by ( [ fg3 ] ) that @xmath200 where the last inequality always holds because of @xmath201 and @xmath202 .
this implies that the lower bound ( [ lb2 ] ) is strictly tighter than the lower bound ( [ lb1 ] ) . on the other hand ,
the lower bound ( [ eqthm1 ] ) of theorem [ thm1 ] is tighter than the lower bound ( [ lb2 ] ) .
this is because @xmath203 in other words , for @xmath201 , the three lower bounds ( [ eqthm1 ] ) , ( [ lb2 ] ) , ( [ lb1 ] ) satisfies @xmath204 where the inequality becomes an equality if and only if @xmath205 .
similarly , for the type - ii finite geometry measurement matrix , we obtain the following results .
@xcite [ lem2 ] let @xmath206 and @xmath207 .
given any @xmath161 different @xmath66-flats @xmath162 in @xmath59 and for any @xmath163 , there exists one @xmath208-flat @xmath165 such that @xmath209 and @xmath210 for all @xmath168 .
[ thm2 ] let @xmath169 be integers , @xmath170 and @xmath82 be the type - ii finite geometry measurement matrix .
then @xmath211 where for euclidean geometry ( eg ) and projective geometry ( pg ) respectively @xmath212 note that the columns of @xmath82 are rows of @xmath20 .
let @xmath213 and assume the contrary that @xmath214 select a @xmath215 such that @xmath216 .
we split @xmath123 into @xmath124 and @xmath125 , and assume @xmath127 without loss of generality .
thus @xmath176 for fixed @xmath128 , by selecting @xmath79-th column of @xmath82 and all the columns in @xmath125 of @xmath82 , we get a submatrix @xmath217 .
the number of columns in @xmath217 is @xmath177 and not greater than @xmath178 .
let @xmath179 and @xmath180 be the @xmath66-flats corresponding to the columns of @xmath217 . by lemma [ lem2 ]
, there exists one @xmath208-flat @xmath165 such that @xmath218 and @xmath210 for all @xmath181 .
there are exactly @xmath178 @xmath65-flats contained in @xmath165 .
now , we claim that @xmath179 contains these @xmath178 @xmath65-flats and @xmath182 ( @xmath181 ) contains at most one @xmath69-flat among these @xmath178 @xmath65-flats .
otherwise , if @xmath182 ( @xmath181 ) contains at least two distinct @xmath69-flats among these @xmath178 @xmath65-flats , then @xmath182 must contain @xmath165 since @xmath165 is the only @xmath208-flat containing these two distinct @xmath69-flats .
this contradicts to the fact that @xmath165 is not contained in @xmath182 .
hence , there exist @xmath178 rows in @xmath217 such that each of these rows has component 1 at position @xmath79 and for any @xmath181 , there exists at most one row that has component 1 at position @xmath80 . using the same argument in the proof of theorem [ thm1 ]
, it leads to a contradiction .
therefore , the assumption is wrong and the theorem follows by ( [ fg1 ] ) and ( [ fg2 ] ) . combining theorem [ thm2 ] with ( [ proeq1 ] ) , we have that when the type - ii finite geometry measurement matrix is used , any @xmath1-sparse signal @xmath7 can be exactly recovered by the @xmath9-optimization ( [ l0 ] ) if @xmath219 . [ remark3 ] for the type - ii finite geometry measurement matrix @xmath82
, it is known that @xmath220 @xcite .
thus , by theorem [ preth0 ] , @xmath221 obviously , @xmath82 has uniform column weight @xmath222 .
the inner product of two different columns equals to the number of @xmath90-flats contained in two fixed @xmath87-flats at the same time .
it is easy to see that the maximum inner product is @xmath223 .
thus , by theorem [ preth ] , @xmath224 using the same argument in remark [ remark2 ] , we have that for @xmath201 , the three lower bounds ( [ eqthm2 ] ) , ( [ lb2 t ] ) , ( [ lb1 t ] ) satisfies @xmath225 where the inequality becomes an equality if and only if @xmath205 or @xmath226 for euclidean geometry and @xmath205 for projective geometry , respectively .
in this section , we will give some simulation results on the performances for the proposed two classes of binary measurement matrices from finite geometry .
the theoretical results on the sparks of these matrices in last section could explain to some extent their good performance .
afterwards , we will show by examples how to employ the parallel structure of euclidean geometry to construct measurement matrices with flexible parameters .
all the simulations are performed under the same conditions as with @xcite .
the upcoming figures show the percentage of perfect recovery ( snr@xmath227db ) when different sparsity orders are considered . for the generation of the @xmath1-sparse input signals , we first select the support uniformly at random and then generate the corresponding values independently by the standard normal distribution @xmath228 . the omp algorithm is used to reconstruct the @xmath1-sparse input signals from the compressed measurements and the results are averaged over 5000 runs for each sparsity @xmath1 . for the gaussian random matrix each entry is chosen _ i.i.d . _ from @xmath228 .
the percentages of perfect recovery of both the proposed matrix ( red line ) and the corresponding gaussian random matrix ( blue line ) with the same size are shown in figures for comparisons . from theorem
[ thm1 ] and theorem [ thm2 ] , the two types of finite geometry measurement matrices have relatively large sparks and thus we expect them to perform well under the omp . for the type - i finite geometry measurement matrix , we expect to recover at least @xmath229-sparse signals ; while for the type - ii finite geometry measurement matrix , we expect to recover at least @xmath230-sparse signals .
[ example1 ] let @xmath231 and @xmath232 .
the @xmath233 consists of @xmath234 3-flats and @xmath235 1-flats .
let @xmath20 be the incidence matrix of 3-flat over 1-flat in @xmath233 .
then @xmath20 is a @xmath236 type - i euclidean geometry measurement matrix .
@xmath20 has girth 4 and is @xmath35-regular , where @xmath237 and @xmath238 .
moreover , @xmath239 according to theorem [ thm1 ] . from fig .
[ figeg4213 ] , it is easy to find that the performance of the proposed matrix is better than that of the gaussian random matrix . in particular , for all signals with sparsity order @xmath240 the recovery are perfect .
this example shows that some girth 4 matrices from finite geometry could also perform very well . with @xmath241 and the corresponding gaussian random matrix.,scaledwidth=50.0% ] with @xmath242 and the corresponding gaussian random matrix.,scaledwidth=50.0% ] [ example2 ]
let @xmath243 and @xmath244 .
the @xmath245 consists of @xmath246 lines and @xmath247 points and @xmath20 is the @xmath248 incidence matrix of line over point in @xmath245 .
then @xmath82 is an @xmath249 type - ii projective geometry measurement matrix .
@xmath82 has girth 6 and is @xmath35-regular , where @xmath250 and @xmath251 .
moreover , @xmath252 by theorem [ thm2 ] . it is observed from fig .
[ figpg3401 ] that @xmath82 performs better than the gaussian random matrix , and the sparsity order with exact recovery may exceed the one ensured by the proposed lower bound .
for @xmath253 , exact recovery is obtained and the corresponding points are not plotted for clear comparisons , and the similar methods are used in the following figures . with @xmath242 and the corresponding gaussian random matrix .
the step size of @xmath1 is 4.,scaledwidth=50.0% ] let @xmath254 , @xmath255 , @xmath256 , and @xmath226 .
the @xmath257 consists of @xmath258 lines and @xmath259 points , and @xmath20 is the @xmath260 incidence matrix of line over point in @xmath257 .
then @xmath82 is a type - ii euclidean geometry measurement matrix .
@xmath82 has girth 6 and is @xmath35-regular , where @xmath261 and @xmath262 .
moreover , @xmath263 by theorem [ thm2 ] . note that the step size of sparsity order @xmath1 is 4 . with @xmath264 and the corresponding gaussian random matrix .
the step size of @xmath1 is 6.,scaledwidth=50.0% ] let @xmath254 , @xmath265 , @xmath266 , and @xmath267 .
the @xmath268 consists of @xmath269 2-flats and @xmath270 1-flats .
let @xmath20 be the @xmath271 incidence matrix of 2-flat over 1-flat in @xmath268 .
then @xmath20 is a type - i euclidean geometry measurement matrix .
@xmath20 has girth 6 and is @xmath35-regular , where @xmath272 and @xmath273 .
moreover , @xmath274 by theorem [ thm1 ] .
[ figeg3812 ] shows that some matrices from finite geometry have very good performance for the moderate length of input signals ( about 5000 ) .
parallel structure of euclidean geometry is very useful to obtain various measurement matrices .
next , we show how to puncture rows or columns from the incidence matrix @xmath20 or @xmath82 by several examples .
[ example - parallel-256 ] let @xmath275 and @xmath244 .
the euclidean plane @xmath276 consists of @xmath277 points and @xmath278 lines .
let @xmath20 be the @xmath76 incidence matrix . since @xmath74 is close to @xmath26 , both @xmath20 and @xmath279 are not suitable to be measurement matrices directly .
however , according to the the parallel structure of @xmath20 described in section [ fgldpc ] , all the 272 lines can be divided into @xmath280 parallel bundles and each bundle consists of 16 lines . by ( [ hpara ] ) , @xmath281 , where for @xmath282 , @xmath94 consists of the 16 lines in the @xmath80-th parallel bundle . by choosing the first @xmath36 submatrices @xmath94 , we get an @xmath22 measurement matrix with uniform column weight @xmath36 . fig .
[ figpara256 ] shows the performance of the @xmath283 , @xmath284 , @xmath285 , @xmath286 submatrices of @xmath20 which correspond to the first 4 , 5 , 6 and 7 parallel bundles of lines in @xmath287 , respectively . with @xmath242 and their corresponding gaussian random matrices .
the rows of the 4 submatrices from left to right are chosen according to the first 4 , 5 , 6 and 7 parallel bundles of lines in @xmath287 , respectively .
the step size of @xmath1 is 2.,scaledwidth=50.0% ] from fig .
[ figpara256 ] we can see that all of the proposed submatrices perform better than their corresponding gaussian random matrices , and the more parallel bundles are chosen , the better the submatrix performs , and its gain over the corresponding gaussian random matrix becomes larger .
[ example - parallel-1024 ] let @xmath288 and @xmath244 .
the euclidean plane @xmath289 consists of @xmath290 points and @xmath291 lines .
let @xmath20 be the @xmath76 incidence matrix .
all the 1056 lines can be divided into @xmath292 parallel bundles and each bundle consists of 32 lines . by ( [ hpara ] ) , @xmath293 , where @xmath94 consists of the 32 lines in the @xmath80-th parallel bundle . by choosing the first @xmath36 parallel bundles , we get an @xmath22 measurement matrix with uniform column weight @xmath36 .
[ figpara1024 ] shows the performance of the @xmath294 , @xmath295 , @xmath296 , @xmath297 submatrices of @xmath20 which correspond to the first 6 , 8 , 10 and 12 parallel bundles of lines in @xmath298 , respectively . from fig .
[ figpara1024 ] it is observed that all of the submatrices perform better than their corresponding gaussian random matrices , and the more parallel bundles are chosen , the better the submatrix performs , and its gain over the corresponding gaussian random matrix becomes larger . with @xmath242 and their corresponding gaussian random matrices .
the rows of the 4 submatrices from left to right are chosen according to the first 6 , 8 , 10 and 12 parallel bundles of lines in @xmath298 , respectively .
the step size of @xmath1 is 8.,scaledwidth=50.0% ] and their corresponding gaussian random matrices , where @xmath299 is the @xmath300 submatrix of @xmath20 in example 6 .
the 4 submatrices ( the red lines from left to right ) are obtained by deleting 0 , 128 , 256 , 384 columns of @xmath299 , respectively .
the step size of @xmath1 is 4.,scaledwidth=50.0% ] [ example - parallel-1024-del ] consider the @xmath300 submatrix in @xmath301 , say @xmath299 , in the last example .
we will puncture its columns to obtain more measurement submatrices . recall that @xmath293 and the first 10 submatrices are chosen to obtain @xmath302 . for the fixed submatrix @xmath303 ,
its corresponding 32 lines are paralleled to each other and partition the geometry .
hence , when selecting the first @xmath79 lines from @xmath303 , the points on these @xmath79 lines are different pairwise and the total number of points is @xmath304 since each line contains 32 points . by deleting the @xmath304 columns corresponding to these @xmath304 points from @xmath299
, we obtain a @xmath305 submatrix , where is still regular . the 4 red lines from left to right in fig .
[ figpara1024-del ] show the performance of the @xmath296 , @xmath306 , @xmath307 , @xmath308 submatrices of @xmath299 which correspond that @xmath309 , respectively .
it is observed that all of the submatrices perform better than their corresponding gaussian random matrices ( the 4 blue lines from left to right ) , but its gain becomes slightly smaller when more columns are deleted .
in this paper , by drawing methods and results from ldpc codes , we study the performance evaluation and deterministic constructions of binary measurement matrices .
the spark criterion is used because its similarity to the stopping distance of an ldpc code and the fact that a matrix with large spark may perform well under the approximate algorithms of @xmath2-optimization , e.g. , the well - known omp algorithm .
lower bounds of spark were proposed for real matrices in @xcite many years ago . when the real matrices are changed to binary matrices , better results may emerge .
firstly , two lower bounds of spark are obtained for general binary matrices , which improve the one derived from @xcite in most cases .
then , we propose two classes of deterministic binary measurement matrices based on finite geometry .
one class is the incidence matrix @xmath20 of @xmath66-flat over @xmath69-flat in finite geometry @xmath59 or its transpose @xmath82 , which are called respectively the type i or type ii finite geometry measurement matrix .
the other class is the submatrices of @xmath20 or @xmath82 , especially those obtained by deleting row parallel bundles or column parallel bundles from @xmath20 or @xmath82 in euclidean geometry . in this way
, we could construct a large amount of measurement matrices with various sizes .
moreover , most of the proposed matrices have cyclic or quasi - cyclic structure @xcite which make the hardware realization convenient and easy . for the type i or ii finite geometry measurement matrix , two further improved lower bounds of spark are given to show their relatively large spark .
finally , a lots of simulations are done according standard and comparable procedures .
the simulation results show that in many cases the proposed matrices perform better than gaussian random matrices under the omp algorithm .
the future works may include giving more lower or upper bounds of sparks for general binary measurement matrices , determining the exact value of sparks for some classes of measurement matrices , and constructing more measurement matrices with large sparks .
we only need to show the necessity .
clearly , the measurement matrix @xmath15 does not have an all-0 column , which implies that @xmath311 .
assume the contrary that @xmath312 .
select a @xmath313 such that @xmath314 .
let @xmath315 , where @xmath316 is the floor function .
then @xmath317 let @xmath318 be the @xmath319-th non - zero position of @xmath110 and set @xmath320 let @xmath321 . clearly , @xmath322 in other words , both @xmath7 and @xmath323 are @xmath1-sparse vectors , and @xmath323 may be sparser .
however , since @xmath324 , we have that @xmath325 which implies that @xmath7 can not be exactly recovered by the @xmath9-optimization ( [ l0 ] ) .
this finishes the proof .
it is known @xcite that any @xmath1-sparse signal @xmath7 can be exactly recovered by the @xmath12-optimization ( [ l1 ] ) if and only if @xmath15 satisfies the so - called _ nullspace property _ , or for any @xmath111 and any @xmath326 with @xmath327 , @xmath328 where @xmath329 .
assume the contrary that @xmath312 . by selecting a @xmath313 such that @xmath314
, it is easy to see that @xmath107 does not satisfy ( [ kspcon1 ] ) for some @xmath1-subset @xmath91 , e.g. , letting @xmath91 be the set of positions with the largest @xmath1 @xmath330 s .
this leads to a contradiction , which implies the conclusion .
99 e. j. cand@xmath331s , j. romberg , and t. tao , `` robust uncertainty principles : exact signal reconstruction from highly incomplete frequency information , '' _ ieee trans .
inf . theory _ ,
2 , pp . 489509 , feb . 2006 .
w. xu and b. hassibi , `` compressed sensing over the grassmann manifold : a unified analytical framework , '' in _ proc .
46th allerton conf .
commun . , control , comput .
_ , monticello , il , sep .
2008 , pp . 562567 .
m. stojnic , w. xu , and b. hassibi , `` compressed sensing - probabilistic analysis of a null - space characterization , '' in _ proc .
acoust . , speech signal process _ , lasvegas , nv , mar.31-apr.4 , 2008 , pp .
33773380 .
l. applebauma , s. d. howardb , s. searlec , and r. calderbank , `` chirp sensing codes : deterministic compressed sensing measurements for fast recovery , '' _ appl .
2 , pp . 283290 , mar .
m. a. iwen , `` simple deterministically constructible rip matrices with sublinear fourier sampling requirements , '' in _ proc .
information sciences and systems _ , baltimore , md , usa , 2009 , pp . 870875 .
s. d. howard , a. r. calderbank , and s. j. searle , `` a fast reconstruction algorithm for deterministic compressive sensing using second order reed - muller codes , '' in _ proc .
information sciences and systems _ , princeton , nj , usa , 2008 , pp . 1115 . n. kashyap and a. vardy , `` stopping sets in codes from designs , '' in _ proc .
inf . theory _
, yokohama , japan , june 29-july 4 , 2003 , p. 122 .
the full version is available online via http://www.mast.queensu.ca/ nkashyap / papers / stopsets.pdf . | for a measurement matrix in compressed sensing , its spark ( or the smallest number of columns that are linearly dependent ) is an important performance parameter . the matrix with spark greater than @xmath0 guarantees the exact recovery of @xmath1-sparse signals under an @xmath2-optimization , and the one with large spark may perform well under approximate algorithms of the @xmath2-optimization .
recently , dimakis , smarandache and vontobel revealed the close relation between ldpc codes and compressed sensing and showed that good parity - check matrices for ldpc codes are also good measurement matrices for compressed sensing . by drawing methods and results from ldpc codes
, we study the performance evaluation and constructions of binary measurement matrices in this paper
. two lower bounds of spark are obtained for general binary matrices , which improve the previously known results for real matrices in the binary case .
then , we propose two classes of deterministic binary measurement matrices based on finite geometry .
two further improved lower bounds of spark for the proposed matrices are given to show their relatively large sparks .
simulation results show that in many cases the proposed matrices perform better than gaussian random matrices under the omp algorithm . compressed sensing ( cs ) , measurement matrix , @xmath2-optimization , spark , binary matrix , finite geometry , ldpc codes , deterministic construction . |
the data used in this analysis were collected by icecube @xcite , an ice cherenkov neutrino detector located at depths between 1450 m and 2450 m at the geographic south pole .
icecube consists of 5160 downward - facing 10-inch pmts , enclosed in glass pressure spheres , known as digital optical modules ( doms ) @xcite .
the detector is an array of 86 strings , each holding 60 doms . of these
, 78 strings are arranged in a triangular grid with a typical distance of 125 m between the strings and a vertical distance of 17 m between doms on the strings .
the lower center region of the array , from 1760 m down to 2450 m , houses a region of denser instrumentation ( 7 m dom spacing ) known as deepcore @xcite with eight strings at string - to - string distances between @xmath12 m . some 50% of the pmts in this region have 35% higher quantum efficiency than the standard icecube pmts .
the result is a neutrino energy threshold in deepcore an order of magnitude smaller than in icecube , of about 10gev .
the icecube detector records more than @xmath13 atmospheric neutrinos every year , a large fraction of them in the deepcore sub - array @xcite . these neutrinos cover path lengths through the earth ranging from 10 km to about 12700 km depending on their arrival zenith angle , @xmath14 . above gev energies
they follow a steeply falling spectrum @xcite that covers several orders of magnitude up to a few hundred tev .
neutrino oscillations modify the flavor ratio of the flux of atmospheric neutrinos measured at the detector site .
the strongest effect to which deepcore is sensitive is the disappearance of @xmath15 , modulated by the large ( atmospheric ) mass splitting , with @xmath16 , and the mixing angle @xmath10 @xcite . in this analysis ,
these parameters are derived from the distortions on the expected @xmath15 flux .
muon neutrino charged current ( cc ) interactions in the ice with energies between 10gev and 100gev , the primary signal event in this analysis , typically produce a minimum - ionizing muon track and initiate a hadronic shower , both of which emit cherenkov light .
the signature of these interactions in deepcore are individual cherenkov photons that are partially scattered due to the optical properties of the ice .
figure[fig : event_view ] shows the detector s
response for one such interaction .
the dominant sources of background for this measurement are muons from cosmic ray showers , cc interactions of electron and tau neutrinos , and neutral current interactions of all neutrino flavors .
atmospheric muons trigger the detector at a rate @xmath13 times higher than the @xmath15 cc rate , which itself is three times higher than the combined rate of all other neutrino interactions .
muons from cosmic ray showers are generated using the corsika package @xcite .
the atmospheric neutrino simulation follows the flux predicted by honda et al @xcite , while neutrino - nucleon interactions in ice are simulated using the genie software @xcite .
muons are propagated according to the parameterization presented in @xcite , while all other particles are passed to the geant4 package @xcite .
the event selection , described in detail in @xcite , has the goal of identifying events that start in the detector volume with a clear muon track to reduce the background . to avoid contamination from atmospheric muons ,
the data analyzed consists only of events reconstructed as passing through the earth ( @xmath17 ) .
however , while atmospheric muons enter the detector only from above ( @xmath18 ) , the small probability of a mis - reconstruction combined with the large number of events detected results in a significant pollution of the neutrino sample .
the event selection starts by rejecting atmospheric muons using the dedicated deepcore trigger and filter @xcite . the atmospheric muon background which remains after the deepcore filter is removed by searching for muon tracks that enter the deepcore volume from outside and pass near the doms that triggered the detector .
this uses the outer part of the deepcore detector as an effective veto region , similar to that described in @xcite .
atmospheric muon simulation is used to understand the basic characteristics of the background and develop methods to remove it .
the statistics available are , however , not enough to provide a complete description of the background at the final level of the analysis and detector data is used instead . in this analysis , cuts on the position of the earliest dom involved in the trigger , the total charge observed in the doms above and prior to the trigger , the charge collected as a function of time @xmath19 , and the number of doms above threshold in a narrow time window @xmath20 $ ] in coincidence with the photons expected from an atmospheric muon hypothesis are applied .
events reconstructed with @xmath21 by a fast track reconstruction algorithm @xcite and a maximum likelihood reconstruction @xcite are also tagged as atmospheric muons .
the veto selection cuts reduce the atmospheric muons to similar rates as the neutrino events while keeping about 40% of the original muon neutrino sample .
the last veto method listed , which uses a muon track hypothesis , is particularly sensitive to muons which enter the fiducial volume through the corridors formed by the detector geometry , leaving very little detected light .
the number of photons observed in an event therefore depends primarily on its azimuth arrival direction , and is largely independent of the event characteristics inside the fiducial volume ( with variations of less than 10% ) . by selecting events above the noise threshold of the search
, a sample of atmospheric muons which fulfill the quality criteria , outlined next , is obtained .
these events are used to create the template for the muon background at the final selection level .
figure[fig : muon_background ] shows the zenith angle distribution of a subsample of the data at three steps in the selection process , where the contributions from neutrinos and atmospheric muons are given separately . in the figure , the region where @xmath22 is also shown , even though it is not used for obtaining the final result .
the transition from a muon to neutrino ( assuming the best known oscillation parameters ) dominated sample as additional selection criteria are applied can be seen in the three steps .
the normalization of both components is fit for each figure , and the results are consistent with those obtained from the fit of the oscillation parameters . the neutrino interactions of interest result in a small number of doms with photon hits ( see fig.[fig : event_view_a ] ) .
it is likely that most photons will have scattered before detection and requiring a minimum number of _ direct _ photons preferentially selects events that occur close to a string .
this reduces the impact of optical scattering in the ice and ensures a well reconstructed event .
direct photons are identified by exploiting the fact that cherenkov light is emitted at a fixed angle relative to the direction of the charge particle and thus the depth at which a photon arrives at a dom on a string as a function of its time of arrival is described by a hyperbola ( see fig.[fig : event_view_b ] and @xcite ) .
when photons scatter they follow a longer path length , resulting in a delay that makes them fall outside the expected hyperbola .
direct photons from multiple cherenkov emitters are also well approximated by a single hyperbola .
the direct photon identification procedure looks for signals that match a hyperbolic pattern while iterating over one string at a time . in such a search
there is no need to assume a track or hadronic shower hypothesis .
each dom is characterized by the time of arrival of the earliest photon in the event and the total observed charge .
the dom with the highest charge is used as the starting point .
a time window is defined for accepting a photon in the dom directly above or below , given by @xmath23 , where @xmath24 is the distance between the doms , @xmath25 is the speed of light in ice and @xmath26 is the permitted time delay , set to 20ns for this analysis .
the selected doms of a given string are considered directly hit only if three or more are found .
direct photons identified by this method have a mean arrival time delay , due to minimal scattering , of 18ns , compared to a typical mean delay time of 230ns .
an event is selected for subsequent processing if at least a total of five doms with direct photons are found .
this keeps about @xmath27 of the muon neutrinos in the relevant energy and zenith angle range .
the agreement between data and simulation after this cut is shown in fig.[fig : pulse_delay_exp ] .
difference in arrival time between direct photons and the expected hyperbola from the track fit , comparing simulation with 5% of the final data sample . ] following @xcite , the direct photons of an event are used to fit two topology hypotheses for cherenkov emission , a single point ( hadronic shower ) and along a track ( muon ) , using a @xmath28 optimization where no scattering is assumed .
the @xmath28 ratio of the track - like and point - like hypotheses is used to select events with a muon track . requiring a track
assures that the event has directional information but retains only 30% of the @xmath15 cc interactions with direct photons , as not all interactions produce a sufficiently long ( 20 m ) muon track .
the zenith angle obtained from the fitted track hypothesis is used as one of the observables in the measurement , and the reduced-@xmath28 of the fit is used as a cut variable .
the method used for estimating the total energy of a neutrino event makes the assumption that all interactions produce a hadronic shower at the interaction vertex , the brightness of which scales with energy , and a minimum - ionizing muon , assuming constant energy loss , that are emitted in the same direction .
the total neutrino energy is then determined by the range of the produced muon @xmath29 and the energy of the hadronic shower @xmath30 , @xmath31 where @xmath32 is the constant energy loss for muons ( in ice @xmath33gev / m ) .
the directions of the muon and the hadronic shower are held fixed in the reconstruction .
expectations for light from the tracks and electromagnetic showers at any given dom are obtained from a multi - dimensional parameterization of many different source configurations , as explained and used in @xcite .
the light expected from a hadronic shower is obtained by scaling - down the expectation from an electromagnetic shower . unlike the directional fit , here both direct and scattered photons
are used to reconstruct the neutrino energy .
the energy estimation is completed in two steps .
the goal of the first step is to determine the range of the muon by assuming all the light present in the detector is explained by a single muon .
the vertex and decay point of the muon are fit by maximizing the likelihood of a finite muon track , to explain the pattern observed , normalized to the probability obtained from an infinite muon track hypothesis .
the fit starts by finding the projection of the first and last dom with a signal along the track , serving as first guesses for the vertex and decay point .
two independent likelihood functions , which are formed by multiplying the probability for a dom to observe zero photons given an expectation of @xmath34 photons from a track , @xmath35 , are maximized for each of these points .
the doms considered are those that observed no light and are situated before the first guess vertex in one likelihood , and after the first guess decay point in the second ( within a distance of 200 m from the infinite track hypothesis ) .
doms that detected light are not included in the calculation .
the likelihood functions are given by @xmath36 where @xmath37 runs over all doms that fulfill the criteria outlined above for the vertex and decay point .
this procedure has a typical accuracy of 25 m .
the vertex point found in the first step above is used as a seed in the second step of the energy reconstruction where the aim is to describe the light in the vicinity of the interaction vertex taking into account that a hadronic shower might have also been produced . while the light output of the muon is taken to be constant along its range , the light expected from the hadronic shower depends on its energy .
both the energy of the hadronic shower and the position of the interaction vertex are obtained by maximizing a likelihood function similar to that in eq.[eq : llh_finitereco ] , but which also includes doms that have detected light , a probability given by @xmath38 .
the likelihood function for the energy deposited at the vertex is then @xmath39 \prod_j \exp(-x_j ) , \label{eq : llh_leera}\ ] ] where @xmath34 is the sum of the light expectation from a muon track , a hadronic shower and noise , and the subscripts @xmath40 and @xmath37 run over the doms within a 300 m radius of the vertex which have observed some or no light , respectively . the energy is finally obtained from evaluating eq.[eq : full_energy ] with the decay point of the muon fit in the first step and the vertex position and hadronic shower energy obtained in the second step .
the energy distribution of the final analyzed neutrino sample , as given by the simulation , is shown in fig.[fig : sample_energy ] .
the sample is composed of 74% @xmath15 cc , 13% @xmath41 cc , 8% neutral current interactions and 5% @xmath42 cc
. deep inelastic scattering ( dis ) events constitute 80% of the sample , followed by resonant and quasi - elastic interactions . the expected atmospheric muon contribution to the final sample , from simulation , is less than 5% .
the median zenith angle resolution obtained for @xmath15 events is 12@xmath43 at @xmath44gev and improves to 5@xmath43 at @xmath45gev .
the median energy resolution is 30% at 8gev and improves to 20% at 15gev .
true energy distribution of simulated neutrino events in the final sample , with hatched areas representing each component .
only events used for the final result are considered ( @xmath46\,$]gev , and @xmath47 ) .
the missing @xmath15 component is also shown . ]
to determine the oscillation parameters , a binned maximum likelihood is used that includes nuisance parameters to account for systematic uncertainties @xcite .
the data are binned in an 8@xmath488 two - dimensional histogram as a function of @xmath49 , between 6gev and 56gev , and @xmath50 between -1 and 0 .
the physics parameters of the fit are the mixing angle @xmath10 and the mass splitting @xmath11 .
oscillation probabilities are calculated using a full three - flavor scheme @xcite , including the effects of the earth s matter distribution @xcite .
the mixing angle @xmath51 is treated as a nuisance parameter using the constraints from @xcite .
the remaining oscillation parameters are fixed to the values given in @xcite , as their uncertainties have a negligible impact on the result . the atmospheric neutrino flux uncertainties , including the normalization , varied by 20% , the spectral index , with an uncertainty of 0.04 , and the ratio of @xmath41/@xmath15 , varied by 20% , are considered .
the template for the atmospheric muon contribution is obtained from inverting the veto cuts on the data , and its normalization is unconstrained in the fit .
cross - section uncertainties are estimated from @xcite . for dis events
the total cross - section is modified by 5% and its energy dependence is changed by @xmath52 .
both uncertainties are smaller than the atmospheric neutrino flux uncertainties which impact the expectation in the same way .
non - linear energy dependencies occur at @xmath53gev . with less than 3% of the sample below that threshold
, the impact was found to be negligible .
the effects of the uncertainty on resonant and quasi - elastic axial masses on the cross - section of non - dis events , of @xmath54 and @xmath55 respectively , were tested and found to modify the result by less than 1% .
the energy scale of hadronic showers , a 5% bias in the energy estimation for minos @xcite , was assessed by varying the reconstructed hadronic energy , known to a resolution of 30% , and was found to have no impact on the result .
the optical efficiency , or energy scale , of the doms has been calibrated in - situ using leds integrated into the doms and minimum - ionizing muons , and is known with an accuracy better than 10% @xcite .
the angular acceptance of the doms , defined by the properties of the surrounding refrozen ice columns , is obtained from fitting the led data and its uncertainty is estimated to be between 30% and 10% , depending on the incident angle .
the impact of the description of the optical properties of the pristine glacial ice is tested by analyzing the data with two independent models @xcite .
proper inclusion of the effect of uncertainties related to the detection process in the minimization proccedure would require events to be re - simulated with the modified values at each step .
full re - simulation of the sample at each minimization step is too computationally intensive , thus an alternative method has been introduced . the most important detector related systematics are the optical efficiency , or energy scale , of the doms and the optical scattering in the ice columns where the doms have been deployed .
the latter modifies the dom angular acceptance and both are parametrized as a function of a single variable .
complete simulation sets were produced for several values of each parameter within the expected uncertainties .
the sets are passed through the same selection and reconstruction steps as the data .
when a fit is performed , the different sets are used to fill the bins of a histogram used to compare data and simulation . for each bin ,
the deviation in the number of expected events with respect to the nominal value is parameterized as a linear function of the quantity varied , e.g. the optical efficiency .
the bin expectation for a set of physics and nuisance parameters is then given by weighting the standard simulation , creating the two - dimensional histogram in energy and direction , and multiplying the bin content by the variation expected for the detector parameters demanded .
the re - weighting scheme at the histogram level was found to succesfully reproduce simulation generated with modified detector settings , and is used in the fit to the data .
the analysis was applied to the data taken between may 2011 and april 2014 , corresponding to 953 days of detector livetime .
a total of 5174 events were observed , compared to an expectation of 6830 events assuming no oscillations .
the parameters that describe the data best are , for a normal mass ordering , @xmath2 and @xmath0ev@xmath1 , and for an inverted mass ordering , @xmath56 and @xmath57ev@xmath1 .
there is no significant preference found for the mass ordering .
the errors solely due to statistical uncertainties are @xmath58 for @xmath59 , and @xmath60ev@xmath1 for @xmath11 .
statistical and systematic uncertainties have an almost equal contribution to the errors of the final result .
the two ice models tested return best - fit points which differ by 0.04 in @xmath59 and by @xmath61ev@xmath1 in @xmath11 . the model that yields the most conservative errors , 30% larger in @xmath59 and about 7% larger in @xmath11 ,
is taken for the final reported results .
for both models , the values of the nuisance parameters at the best fit point are within the assumed uncertainties and the atmospheric muon contribution to the final sample is fit to 1% , consistent with the expectation obtained from simulation .
a cross check was performed by fitting the data in an unbounded two - flavor scheme , which yielded consistent results .
the simulation is in good agreement with the data , with a @xmath28/d.o.f.@xmath62 for the energy - zenith angle histogram used in the fit , shown in fig.[fig : full_histogram ] .
there , the zenith angle distribution of data and simulation is given for different energy bands .
below each histogram the ratio of data and the best - fit to the case of no - oscillations is also included .
the maximum disappearance can be seen in the panel containing @xmath63$]gev .
the simulation agrees with the data at all energies considered .
figure[fig : lefit ] shows the agreement between data and simulation as a function of reconstructed baseline over energy @xmath64 , a variable that does not directly enter the analysis .
the 90% confidence contours on the atmospheric oscillation parameters derived from this analysis , compared to the results from other experiments , are shown in fig.[fig : contours ] . while this measurement is made at higher energies than other experiments ( see fig.[fig : sample_energy ] ) , the results are compatible and the precision achieved is comparable . higher statistics , ongoing improvements in veto algorithms , and the inclusion of cascade - like events will enhance the sensitivity of the oscillation studies with icecube in the near future .
this could be further improved by deploying additional instrumentation within the deepcore array to collect more light per event and thus increase the statistics below @xmath44gev , as proposed in the precision icecube next generation upgrade ( pingu ) @xcite .
we acknowledge the support from the following agencies : u.s .
national science foundation - office of polar programs , u.s .
national science foundation - physics division , university of wisconsin alumni research foundation , the grid laboratory of wisconsin ( glow ) grid infrastructure at the university of wisconsin - madison , the open science grid ( osg ) grid infrastructure ; u.s .
department of energy , and national energy research scientific computing center , the louisiana optical network initiative ( loni ) grid computing resources ; natural sciences and engineering research council of canada , westgrid and compute / calcul canada ; swedish research council , swedish polar research secretariat , swedish national infrastructure for computing ( snic ) , and knut and alice wallenberg foundation , sweden ; german ministry for education and research ( bmbf ) , deutsche forschungsgemeinschaft ( dfg ) , helmholtz alliance for astroparticle physics ( hap ) , research department of plasmas with complex interactions ( bochum ) , germany ; fund for scientific research ( fnrs - fwo ) , fwo odysseus programme , flanders institute to encourage scientific and technological research in industry ( iwt ) , belgian federal science policy office ( belspo ) ; university of oxford , united kingdom ; marsden fund , new zealand ; australian research council ; japan society for promotion of science ( jsps ) ; the swiss national science foundation ( snsf ) , switzerland ; national research foundation of korea ( nrf ) ; danish national research foundation , denmark ( dnrf ) 46ifxundefined [ 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 link:\doibase 10.1103/physrevlett.81.1562 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.100.221803 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.89.011301 [ * * , ( ) ] @noop * * , ( ) , link:\doibase 10.1103/physrevlett.108.171803 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.108.191802 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.108.131801 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.112.181801 [ * * , ( ) ] , link:\doibase 10.1103/physrevlett.110.251801 [ * * , ( ) ] , in @noop _ _ ( ) link:\doibase 10.1088/1742 - 6596/408/1/012001 [ * * , ( ) ] , link:\doibase 10.1142/s0217732313300243 [ * * , ( ) ] , link:\doibase 10.1103/physrevd.82.093011 [ * * , ( ) ] , link:\doibase 10.1007/jhep02(2013)082 , 10.1007/jhep07(2013)026 [ * * , ( ) ] , link:\doibase 10.1103/physrevd.88.013013 [ * * , ( ) ] , link:\doibase 10.1016/j.astropartphys.2006.06.007 [ * * , ( ) ] , link:\doibase 10.1016/j.nima.2009.01.001 [ * * , ( ) ] , link:\doibase 10.1016/j.astropartphys.2012.01.004 [ * * , ( ) ] , @noop * * , ( ) link:\doibase 10.1088/1126 - 6708/2006/10/075 [ * * , ( ) ] , link:\doibase 10.1103/physrevd.79.102005 [ * * , ( ) ] , link:\doibase 10.1140/epjc / s10052 - 013 - 2606 - 4 [ * * , ( ) ] , @noop ( ) , @noop _ _ , ( ) , link:\doibase 10.1103/physrevd.83.123001 [ * * , ( ) ] , link:\doibase 10.1016/j.nima.2009.12.009 [ * * , ( ) ] , @noop ( ) , link:\doibase 10.1016/s0168 - 9002(03)01368 - 8 [ * * , ( ) ] _ _ , http://edoc.hu-berlin.de/docviews/abstract.php?id=40895[ph.d . thesis ] , ( ) link:\doibase 10.1103/physrevlett.110.151105 [ * * , ( ) ] , link:\doibase 10.1016/j.nima.2013.10.074 [ * * , ( ) ] , link:\doibase 10.1016/j.nima.2004.01.065 [ * * , ( ) ] , link:\doibase 10.1016/j.astropartphys.2011.01.003 [ * * , ( ) ] , link:\doibase 10.1088/1748 - 0221/9/03/p03009 [ * * , ( ) ] , link:\doibase 10.1088/1674 - 1137/38/9/090001 [ * * , ( ) ] link:\doibase 10.1103/physrevd.22.2718 [ * * , ( ) ] @noop ( ) \doibase http://dx.doi.org/10.1016/0031-9201(81)90046-7 [ * * , ( ) ] link:\doibase 10.1103/physrevd.86.013012 [ * * , ( ) ] , @noop link:\doibase 10.1007/jhep08(2011)042 [ * * , ( ) ] , @noop ( ) , link:\doibase 10.1103/physrevlett.101.131802 [ * * , ( ) ] , link:\doibase 10.1016/j.nima.2013.01.054 [ * * , ( ) ] , in @noop _
_ ( ) @noop ( ) , | we present a measurement of neutrino oscillations via atmospheric muon neutrino disappearance with three years of data of the completed icecube neutrino detector .
deepcore , a region of denser icecube instrumentation , enables the detection and reconstruction of atmospheric muon neutrinos between 10gev and 100gev , where a strong disappearance signal is expected .
the icecube detector volume surrounding deepcore is used as a veto region to suppress the atmospheric muon background .
neutrino events are selected where the detected cherenkov photons of the secondary particles minimally scatter , and the neutrino energy and arrival direction are reconstructed .
both variables are used to obtain the neutrino oscillation parameters from the data , with the best fit given by @xmath0ev@xmath1 and @xmath2 ( normal mass ordering assumed ) .
the results are compatible , and comparable in precision , to those of dedicated oscillation experiments . in the 1990s
, super - kamiokande s measurements of atmospheric neutrinos @xcite led to the acceptance of the mass - induced oscillation model where neutrinos are massive particles whose interaction eigenstates do not have an exact correspondence to their mass eigenstates .
this property gives neutrinos produced in one flavor eigenstate , @xmath3 , a non - zero probability of interacting in a different flavor , @xmath4 , after traveling for some distance @xmath5 . in the simplest scenario , with only two neutrino flavors ,
the transition probability is given by @xmath6 where @xmath7 defines the mixing between mass and flavor eigenstates , @xmath8 is the difference in the squared masses and @xmath9 is the neutrino energy ( in natural units ) . considering the existence of three neutrinos , as done in this letter , provides an oscillation probability that consists of a sum of terms of the form of eq.[eq : osc_prob ] , but involving three mixing angles , two mass - squared differences and a complex phase .
currently , the mixing angles , the solar mass splitting and the absolute value of the atmospheric mass splitting have been measured @xcite while the existence of cp - violation and the ordering of the masses remain open questions @xcite .
addressing these questions requires improving the measurement precision on the known parameters and improving the measurements sensitive to the parameters that modify the oscillation probabilities as neutrinos traverse matter @xcite . in the following , we focus on the measurement of the oscillation parameters @xmath10 and @xmath11 .
the measurement presented here achieves a precision comparable to the leading experiments in the field @xcite using a sample of atmospheric high energy neutrinos , from 10gev to 100gev , recorded with a sparsely instrumented detector located in a natural medium . |
this paper was inspired by the 1965 work of padnos , as reported in reference @xcite . using the variation ansatz with normalized wavefunctions of a particle in a box as trial wavefunctions , he obtained approximate values for the ground state energy and the first excited state energy of the one dimensional quantum harmonic oscillator .
padnos work was preceded by and is an improvement upon the work of rich @xcite , who approximated the oscillator by a particle in a one dimensional box whose size was the classical range of the oscillator .
one of the reasons padnos was discouraged from extending the calculations to higher energy levels was the belief that the calculation becomes more tedious as one must introduce extra terms to orthogonalize the new function to the ones already found ( for example , the second excited state must be chosen to be orthogonal to both the ground state and the first excited state ) .
the authors of this present paper have no need to introduce any extra terms since the eigenstates of the particle in a box can all be chosen to be mutually orthogonal , as simple trigonometric functions ( see section [ sec.pf1lpk4 ] ) . another notable paper which employed the variational method for harmonic oscillator quantization is reference @xcite , where different sets of basis functions , built from non - orthogonal monomials , were used as trial wavefunctions .
a particle of mass @xmath0 , free to move _ only _ in a ` box ' of size @xmath1 , so that the potential , @xmath2 , is @xmath3 is decribed by the hamiltonian @xmath4 the eigenfunctions of @xmath5 are non - degenerate and are given by @xmath6 } , & r=0,2,4,\ldots\\ \\
\sqrt{\frac 1l}\sin{\left[\frac{(r+1)\pi x}{2l}\right ] } , & r=1,3,5,\ldots\ , , \end{cases}\ ] ] with corresponding eigenvalues @xmath7 where @xmath8 .
the eigenfunctions @xmath9 can be written more compactly as @xmath10,\quad r=0,1,2,3,\ldots\ ] ] if the particle is not free in the box but is instead under the influence of a potential , @xmath2 , where @xmath11 then the system becomes a ` confined ' quantum harmonic oscillator ( cho ) , and is now described by the hamiltonian @xmath12 where @xmath13 is the classical frequency of the oscillator . although the confined quantum harmonic oscillator has been studied for a long time ( see references @xcite and the references therein ) , it does not seem to enjoy the same popularity as its @xmath14 limit , the ` free ' quantum harmonic oscillator ( fho ) , with @xmath15 it is our aim in this paper to apply the variation procedure to quantize @xmath16 through @xmath17 .
since @xmath5 and @xmath17 live in the same hilbert space we will use the complete set of functions @xmath9 ( eigenstates of @xmath5 ) as trial wavefunctions in the variation ansatz .
the size of the box , @xmath18 , then enters , naturally , as the variation parameter .
the functions @xmath9 as given in , being non - degenerate eigenstates of a hermitian operator , @xmath5 , are a complete set of vectors spanning an infinite dimensional hilbert space .
in particular @xmath9 being mutually orthogonal in the interval and satisfying the boundary conditions are a suitable choice of trial wavefunctions for the confined harmonic oscillator hamiltonian , @xmath17 , whose eigenfunctions are also required to satisfy the same boundary conditions . the box size , @xmath18
, then serves as the variation parameter which will be optimized to get the approximate eigenfunctions and eigenvalues of the free harmonic oscillator .
henceforth , any quantity derived for the free harmonic oscillator by optimization will be indicated with an asterik on its symbol .
the optimized box size , @xmath19 , is obtained by solving @xmath20 for @xmath19 , where @xmath21 the optimized approximate energy eigenvalues of the free harmonic oscillator are then given by @xmath22 actually , the @xmath23 introduced in ( ground state energy of the free particle in a box ) is more convenient to use as the variation parameter , with its optimum value @xmath24 being related to the optimum value of the box size @xmath19 by @xmath25 @xmath24 is obtained by expressing the @xmath26 of in terms of @xmath23 and then solving @xmath27 for @xmath24 .
the optimized approximate eigenenergies of the free harmonic oscillator are then obtained from @xmath28 performing the integration in we have @xmath29 where @xmath30 .
thus @xmath31 where @xmath32 from and we have @xmath33 so that @xmath34 from and we have @xmath35 so that the approximate eigenenergies of the free harmonic oscillator , @xmath36 , are given by @xmath37 writing as @xmath38 where @xmath39 , and using we have @xmath40 so that the approximate eigenstates of the free harmonic oscillator are given by @xmath41,\quad r=0,1,2,3,\ldots}\ ] ] the optimized eigenfunctions @xmath42 are expected to approximate the eigenstates of the free harmonic oscillator in the interval : .
the free harmonic oscillator is one of the few quantum mechanical systems that can be quantized exactly .
expressions for the eigenenergies and the corresponding wavefunctions are derived in every book on quantum mechanics . for quantum numbers @xmath43 ,
the energy eigenvalues are given by @xmath44 with corresponding normalized wavefunctions @xmath45 where @xmath46 is the degree @xmath47 hermite polynomial in @xmath48 and , as before , and .
the first few hermite polynomials are : @xmath49 we recall expressions and for the approximate and exact eigenenergies of the free one - dimensional linear harmonic oscillator : @xmath50 @xmath51 and @xmath52 are plotted in figure [ fig.zsxuwpe ] as functions of the harmonic oscillator quantum number @xmath53 .
the agreement between @xmath36 and @xmath26 is quite remarkable , especially for low quantum numbers . , with the exact eigenenergies , @xmath26 , for the free harmonic oscillator .
] to illustrate the agreement between the variation results and the exact wavefunctions for the free harmonic oscillator , it is instructive to plot the approximate and the exact wavefunctions together , as functions of position .
here we do this for the ground state and the first excited state . from and
we have @xmath54 and @xmath55 the good correlation between @xmath56 and @xmath57 is already obvious from the taylor series expansion of both functions : @xmath58 and @xmath59 with @xmath60 the approximate ground state wavefunction @xmath61 and the exact ground state wavefunction @xmath62 of the free harmonic oscillator , with @xmath63 are shown in figure [ fig.q96gfpe ] as functions of position . , and exact ground state , @xmath62 as functions of position , for the free harmonic oscillator , for @xmath63 . ] as for the first excited state , we have from and we have @xmath64 and @xmath65 and exact first excited state , @xmath66 , as functions of position , for the free harmonic oscillator , for @xmath63 . ] the variation of @xmath67 and @xmath68 with respect to position are as shown in figure [ fig.wipsqc4 ] .
by using the normalized mutually orthogonal wavefunctions of the free - particle - in - a - box model as trial wavefunctions in variation calculation we have obtained approximate energy eigenvalues and the corresponding eigenstates for the one dimensional free harmonic oscillator of mass @xmath0 and classical frequency @xmath13 .
we obtained , for quantum numbers @xmath69 , @xmath70 and @xmath71\,,\ ] ] where @xmath72 and @xmath73 with @xmath39 . | the main purpose of this paper is to demonstrate and illustrate , once again , the potency of the variational technique as an approximation procedure for the quantization of quantum mechanical systems . by choosing particle - in - a - box wavefunctions as trial wavefunctions , with the size of the box as the variation parameter , approximate eigenenergies and the corresponding eigenfunctions are obtained for the one dimensional free harmonic oscillator . |
this work is based on observations obtained with _ xmm - newton _ , an esa science mission with instruments and contributions directly funded by esa member states and nasa .
ek thanks the gates cambridge scholarship .
acf thanks the royal society .
the authors thank the anonymous referee for helpful comments . | we use archival _ xmm - newton _ observations of ark 564 and mrk 335 to calculate the frequency dependent time - lags for these two well - studied sources .
we discover high - frequency fe k lags in both sources , indicating that the red wing of the line precedes the rest frame energy by roughly 100 s and 150 s for ark 564 and mrk 335 , respectively . including these two new sources , fe k reverberation lags
have been observed in seven seyfert galaxies .
we examine the low - frequency lag - energy spectrum , which is smooth , and shows no feature of reverberation , as would be expected if the low - frequency lags were produced by distant reflection off circumnuclear material .
the clear differences in the low and high frequency lag - energy spectra indicate that the lags are produced by two distinct physical processes . finally , we find that the amplitude of the fe k lag scales with black hole mass for these seven sources , consistent with a relativistic reflection model where the lag is the light travel delay associated with reflection of continuum photons off the inner disc .
[ firstpage ] black hole physics galaxies : active x - rays : galaxies [ cols="^,^,^,^,^,^,^,^",options="header " , ] lastly , we comment briefly on the apparent weak non - stationarity in the light curves ( as noted by * ? ? ?
* through the lag - frequency spectrum and psd ) .
fig .
[ ark_redblue ] shows that the first half of the 500 ks observation peaks at the rest frame of the fe k line , presumably from a larger emitting region , while at slightly higher frequencies , the second half shows a broader lag peaking at 45 kev , the red wing of the fe k line from a smaller emitting region .
while the non - stationarity of the light curves is not currently well understood , it does provide tantalizing evidence for further behavior that will be accessible either through much deeper observations or future observatories .
the case for relativistic reflection continues to grow with recent discoveries of the fe k reverberation lag @xcite that is now seen in seven sources , flux - dependent reverberation lags @xcite , the black hole mass scaling relation with lag @xcite , and reverberation lags in a black hole x - ray binary @xcite and possibly a neutron star @xcite . the work presented here on ark 564 and mrk 335 support a relativistic reflection model and are not consistent with partial covering .
the study of x - ray reverberation lags is quickly developing , and is revealing a new perspective through which to probe strong gravity . |
after the first run of the lhc at a center of mass ( c.m . ) energy of mostly 8 tev , no significant excesses have been observed in searches for physics beyond the standard model @xcite .
these searches cover a wide range of possible signatures , notably various combinations of jets , missing transverse energy ( @xmath1 ) , @xmath4-jets and leptons ( electrons or muons ) .
same - flavour opposite - sign dileptons can be classified into `` off-@xmath0 '' leptons ( typically with an invariant mass @xmath5 gev or @xmath6 gev ) , and `` on-@xmath0 '' leptons with @xmath7 gev .
often , leptons and in particular on-@xmath0 dileptons are vetoed in order to suppress standard model ( sm ) backgrounds . on the other hand ,
some decay cascades of supersymmetric ( susy ) particles could be particularly rich in off-@xmath0 dileptons ( in the presence of light sleptons ) , or on-@xmath0 dileptons if @xmath0 bosons appear particularly frequently in these cascades .
recently , results of searches for susy particles in events with dileptons , jets and @xmath1 have been published by the cms and atlas collaborations @xcite .
the aim was to test scenarios of gluino pair production in which the gluinos @xmath8 decay via sleptons ( leading to off-@xmath0 dileptons ) , and scenarios of gauge mediated susy breaking ( gmsb ) or generalised gauge mediation ( ggm ) where the gluinos undergo 3-body decays into quark pairs and a neutralino @xmath9 .
the latter may decay subsequently into a nearly massless gravitino @xmath10 and a @xmath0 boson , leading to on-@xmath0 dileptons .
the corresponding gluino decay chain is then @xmath11 .
relevant parameters are the gluino mass @xmath12 , the neutralino mass @xmath13 , and the branching fractions of the involved decays . whereas no significant excesses were observed by cms in @xcite ( up to an excess of 2.6 standard deviations in the dilepton mass window @xmath14 gev ) , an excess of 3.0 standard deviations was reported by atlas in @xcite in the on-@xmath0 signal region : summing electron and muon pairs , 29 events passing the cuts
were observed versus @xmath15 background events expected .
no attempt was made in @xcite to explain the excess in terms of a specific model ; instead , weaker exclusion limits than expected were shown in the @xmath16 plane of ggm models .
various studies of scenarios which could contribute to this excess have recently been published @xcite .
@xmath0 bosons decay dominantly hadronically .
thus , whenever gluinos are pair produced , in most cases each of the two gluino cascades will produce no dileptons , but two hard jets : either from @xmath17 if @xmath18 , or from hadronic @xmath0 decays if @xmath19 implying a neutralino much heavier than the gravitino , i.e. energetic @xmath0 bosons .
hence , both scenarios are subject to constraints from `` standard '' searches for susy in events with hard jets and @xmath20 @xcite , even if one considers simplified models where squarks are assumed to be decoupled and gluino pair production is the only process taken into account . in order to study the impact of these constraints on gmsb - like scenarios , we simulated various configurations of gluino and @xmath9 masses . using the latest version 1.2.0 of checkmate @xcite we found that constraints from @xcite ( a preliminary version of @xcite ) on final states with jets and @xmath20 are very restrictive , and supersede even the recent cms constraints from @xcite in the @xmath16 plane .
exceptions are scenarios with reduced branching fractions for the considered decay chain , without allowing for alternative final states leading to jets and @xmath20 . in the present paper
we study to which extent a scenario with _ two heavy _ neutralinos in the gluino decay cascade can contribute to the atlas signal region , circumventing constraints from searches for jets and @xmath20 .
the gluino decay cascade considered subsequently is of the form [ eq:1.1 ] q+|q+_2 ^ 0 q+|q+z+_1 ^ 0 with m__2 ^ 0 m_,m__1 ^ 0 ~m__2 ^ 0 -100 and sketched in fig . 1 .
now jets from both steps of the gluino decay cascade ( including the jets from the @xmath0 boson ) are relatively soft , and constraints from searches for jets and @xmath20 are easier to satisfy unless the mass splitting @xmath21 is too large . such a scenario has been considered recently also in @xcite
. we will compare their results to ours in the conclusions . in the following we consider first simplified models with 100% branching fractions for both steps of the gluino decay cascade .
we simulated corresponding events , verified which scenarios satisfy the constraints from the cms @xcite and other susy searches , and applied the cuts of atlas @xcite .
we will compare the signal rates and various distributions to the data given in @xcite , and to a gmsb - like scenario ( the latter with reduced branching fractions in order to comply with constraints ) . constraints from cms @xcite prevent an excess as large as 3.0 standard deviations in the atlas signal region , but about 14 signal events on top of the background are possible .
however , the question arises in which susy scenario such a neutralino spectrum and , notably , such a dominant gluino decay cascade are possible : what can prevent a dominant @xmath22 decay which is favored by phase space ? in gmsb the rle of @xmath9 is played by the nearly massless gravitino , which has tiny couplings to the mssm - like sparticles and is not produced unless , due to r - parity conservation , it is the only decay channel .
a heavier neutralino @xmath9 with small couplings to the mssm - like sparticles , as required in the present scenario , is possible in the nmssm @xcite in the form of the singlino , the fermionic partner of the singlet superfield @xmath23 whose vacuum expectation value generates dynamically a @xmath24-term ( a susy mass term for the two higgs doublets in the mssm ) of the order of the susy breaking scale .
we find that there exist indeed scenarios within the parameter space of the nmssm for which the gluino decay cascade in eq .
( [ eq:1.1 ] ) is dominant . in the next section
we describe details of the simulation and cuts .
results for simplified models and the description of a nmssm scenario are given in section 3 .
we conclude in section 4 .
we have simulated events at the lhc at 8 tev using madgraph / madevent @xcite which includes pythia 6.4 @xcite for showering and hadronisation .
the emission of one additional hard jet was allowed in the simulation in order to obtain realistic distributions for kinematical variables .
the production cross sections were obtained by prospino at nlo @xcite .
first , the output was given to checkmate version 1.2.0 @xcite which includes the detector simulation delphes @xcite and compares the signal rates to constraints from various search channels of atlas and cms .
all searches present in checkmate version 1.2.0 have been verified ; the most relevant ones ( with the largest ratio for the event yield to @xmath25 where @xmath25 is the observed 95% cl upper bound ) are obtained from the atlas search for jets and @xmath20 in @xcite .
second , the pythia output was given directly to delphes and analysed according to the object identification and selection criteria given in @xcite , respectively , and finally the corresponding cuts were applied .
for the atlas on-@xmath0 searches @xcite these were as follows : @xmath26 gev ; @xmath27 jets with @xmath28 gev ; two same - flavour opposite - sign leptons with @xmath29 gev for the leading , @xmath30 gev for the sub - leading lepton ; @xmath31 gev where @xmath32 ( including jets with @xmath28 gev ) ; and finally @xmath7 gev . 29 events passing the cuts were observed , whereas @xmath15 background events were expected . for the events passing the cuts , distributions of @xmath33 , @xmath1 , @xmath2 and the jet multiplicity @xmath34
were shown in @xcite separately for the electron and muon channels .
these distributions were compared with those expected from two ggm benchmark points with gluino masses and neutralino masses of @xmath35gev , @xmath36gev , respectively .
we found , however , that both points violate constraints from @xcite on final states with jets and @xmath1 .
we compared the expected properties of the two ggm benchmark points in @xcite to the results of our simulation and found that they agree within @xmath37 .
we conclude that the results of our simulations deviate by a systematic error of up to @xmath37 from the more realistic ( detector- ) simulation of the experimental collaboration .
we can expect that this systematic error cancels to a large extent when comparing the properties of different simulated scenarios , but should be taken into account when comparing to the actual data from @xcite . since it is of the same order ( actually somewhat larger ) than the difference in the acceptances of dielectrons and dimuons in @xcite , we found it reasonable to consider the sum of the data of dielectron and dimuon events not only for the signal rate , but also for the kinematical distributions and the expected sm background in order to obtain a larger statistics . in the cms on-@xmath0 searches
@xcite , no cuts on @xmath2 were applied .
signal jets were required to have @xmath38 gev .
six @xmath20- and @xmath34-dependent on-@xmath0 signal regions were defined : @xmath39 gev and @xmath40 , respectively .
finally the cms and atlas analyses differ slightly in the jet algorithms and in the lepton acceptances . comparing the signal rates obtained by our simulations of the two gmsb - like benchmark points to the simulations in @xcite we found again that they agree within @xmath37 .
as already stated above , no significant excesses were observed in the on-@xmath0 signal regions by cms .
hence the event yields in the six on-@xmath0 signal regions lead to constraints on any scenarios which attempt to explain the atlas excess . in the next section
we discuss by means of benchmark points to which extent the atlas excess can be matched in consideration of these constraints , as well as constraints from @xcite on final states with jets and @xmath20 .
first we considered gmsb - like simplified models with a branching fraction of 100% for the @xmath11 decay chain .
then , however , constraints from the search for jets and @xmath20 in @xcite as tested by checkmate @xcite require @xmath41 gev for small @xmath42 gev , and larger gluino masses for larger @xmath13 .
accordingly contributions to the atlas signal region can not exceed @xmath43 events . moreover the distribution of @xmath20 and @xmath2 peak towards large values ( most events have @xmath44 gev ) in sharp contrast to the data in @xcite . in realistic models ,
the branching fractions for the steps of the above gluino decay chain can well be below 100% .
below we will consider a gmsb - like benchmark point `` gmsb '' with @xmath45 gev and a branching fraction of 10% for the above decay chain ; such a small branching fraction makes it compatible with the cms constraints .
( a heavy @xmath9 was chosen in order to shift the peak of the @xmath2 distribution towards lower values . ) for the remaining 90% of the gluino decays one has to expect that , depending on the complete spectrum and branching fractions , they contribute to the signal regions in the search for jets and @xmath20 in @xcite .
one can make the somewhat optimistic assumption that these contributions do not exceed 50% of the contributions of the @xmath11 decay chain .
then this point remains within the constraints from @xcite , but contributes about 10 events to the atlas on - z signal region .
next we consider simplified models with two heavy neutralinos whose decay chain is depicted in fig . 1 .
assuming a branching fraction of 100% for this decay chain , gluinos can be as light as 800 gev without conflict with constraints from the search for jets and @xmath20 in @xcite under the condition , however , that @xmath46 and @xmath47 are relatively large such that all jets remain relatively soft .
we studied two benchmark points p1 and p2 with @xmath48 gev and @xmath49 gev , respectively .
such scenarios belong to the few exceptions allowing for gluinos with a mass below 1 tev , see the study in @xcite .
for p1 with @xmath50 gev the jets from the first step @xmath51 of the decay cascade are very soft , as are the jets from @xmath0 decays from the second step @xmath52 .
practically all energy of a single gluino decay cascade goes into @xmath20 .
however , for typical kinematical configurations the momenta of @xmath9 tend to be back - to - back in the transverse plane , leading to a reduction of @xmath20 of the complete event .
only for relatively rare kinematical configurations ( and/or extra jets from initial state radiation as included in our simulation ) , @xmath20 of the complete event can assume large values .
for p2 with @xmath53 gev the jets from the first step @xmath51 of the decay cascade are harder , leading to less @xmath20 .
one aim is to study the impact of this difference on the distributions of kinematical variables . for all benchmark points we assumed practically decoupled squarks with masses of 3 tev ;
then the gluino pair production cross section from prospino at nlo is 128 fb .
( since stops and sbottoms are assumed to have masses of 3 tev as well their pair production does not contribute to the signal . )
we deliberately chose identical gluino masses for all points in order to maintain a common production cross section ; therefore all differences in contributions to signal regions and kinematical distributions originate from the neutralino sector .
the masses of the latter are recalled in table 1 below .
.sparticle masses of the benchmark points gmsb , p1 and p2 ( in gev ) , event yields including 30% systematic errors from the simulation of the benchmark points gmsb , p1 and p2 in the six signal regions of the cms on-@xmath0 searches in @xcite , and in the most constraining signal regions ct , em and et of the atlas search @xcite .
the ranges of @xmath20 for the six cms signal regions are given in gev .
the last line indicates the contributions to the atlas on-@xmath0 signal region . [ cols="^,^,^,^,^",options="header " , ] in addition we indicate in the table 1 in how far the benchmark points gmsb , p1 and p2 satisfy constraints from the six signal regions of the cms on-@xmath0 searches in @xcite ( including 30% systematic errors from the simulation ) .
the 95% cl upper limits for the six signal regions of the cms on-@xmath0 searches had already been obtained in @xcite .
we find that the central values of event yields of the benchmark points are below these 95% cl upper limits with the exception of p2 in the bins @xmath54 , @xmath55 and @xmath56 , @xmath55 .
however , taking the systematic errors from the simulation into account , the @xmath57 values for p2 in these bins are 0.11 and 0.09 , respectively , i.e. well above the 95% cl exclusion limit of 0.05 . out of the 10 signal regions in the atlas search @xcite for jets and @xmath20
we show the event yields for the signal regions ct , em and et which give the largest ratio event yield/@xmath25 for the points p1 , p2 and gmsb , respectively .
all these signal regions require @xmath58 gev , @xmath59 gev for the leading jet , and @xmath60 gev for 3 additional jets ( ct ) , @xmath60 gev for 5 additional jets ( em and et ) .
em and et differ by @xmath61 gev , respectively ( see @xcite for more details ) .
we recall that the event yields for the point gmsb assume only a branching fraction of 10% into the considered gluino decay chain . in table 1 , only the contributions from the simulated decay chain are shown . within the systematic error bars , @xmath62 more events from other gluino decays are allowed to contribute to the signal regions in the atlas search @xcite for jets and @xmath20 in order to saturate the bound from the signal region et .
finally we compare the contributions of the benchmark points gmsb , p1 and p2 to the atlas on-@xmath0 signal region , summing dielectrons and dimuons , in the last line of table 1 .
we see that a price has to be paid for the very compressed gluino@xmath63 spectrum in p1 : due to the softness of the jets , not enough jets satisfy the cut @xmath64 .
the gmsb point seems to do quite well , despite its gluino branching fraction being reduced by a factor @xmath65 .
the best fit is given by p2 with its less compressed gluino@xmath63 spectrum .
next we consider the distributions of kinematical variables .
as stated above we combine the atlas dielectron and dimuon data ( despite the different acceptances ) in order to enhance the visibility of possible trends .
we only show the ( dominant ) statistical error of the data ; we are not in a position to combine the partially correlated systematic errors . in the figures below we show the data with the expected sm background contribution subtracted , with the aim to expose possible desirable features of signal contributions ( see @xcite for the error attributed to the expected background ) .
we start with @xmath1 in fig . 2 where we compare the data with the expected background subtracted to the gmsb scenario and with the two heavy - neutralino benchmark points p1 and p2 .
we simulated 500.000 events for each scenario .
each expected event for the lhc run i as shown in fig .
[ fig:2 ] corresponds to 10 simulated events , which allows to estimate the statistical errors .
these are smaller than the estimated systematic errors from our simulation , and much smaller than the statistical error of the data .
the measured event numbers seem to decrease continuously with @xmath1 ( within the error bars , and note that the rightmost bin includes the overflow ) , whereas the @xmath1 distributions of the gmsb and p1 points are nearly flat : in these scenarios , nearly all energy is transformed into missing energy which prefers accordingly large values of @xmath1 .
( note that @xmath20 is shown after the application of all cuts , notably on @xmath66 gev .
for p1 with its compressed spectrum this cut selects atypical kinematical configurations with particularly large @xmath20 . ) for a quantitative comparison we compute the reduced @xmath67 statistic ^2 _ = _ i=1^n _ for each benchmark point , where @xmath68 is the data with the expected background subtracted ( as shown in fig . 2 ) .
@xmath69 combines the statistical error of the data shown in fig . 2 and the systematic error of 30% of our simulation ( with respect to which the systematic error of the background is negligible ) .
we obtained @xmath70 for gmsb , @xmath71 for p1 and @xmath72 for p2 .
hence the scenario p2 with its larger splitting between the gluino and the @xmath73 masses describes best the shape of the @xmath1 distribution .
of course , the scenario p2 profits also from its larger total event rate . in fig .
[ fig:3 ] we compare the data on @xmath2 ( with the expected background subtracted ) with the gmsb scenario and the benchmark points p1 and p2 . since @xmath2 represents most of the visible transverse energy , the point p1 with its compressed spectrum peaks at low values of @xmath2 .
this coincides with the trend of the data , but the total signal rate ( limited by constraints from cms ) is small , as indicated in table 1 . for the reduced @xmath67 statistic
we find @xmath74 for gmsb , @xmath70 for p1 and @xmath75 for p2 .
again , the benchmark point p2 provides the best agreement with the shape of the distribution despite its somewhat less compressed spectrum .
finally we turn to the distribution of the jet multiplicity in fig .
[ fig:4 ] .
the trend of the data towards low jet multiplicities is reproduced only by p1 with its excessively low signal rate .
the jet multiplicity of simulations is sensitive , amongst others , to the matching between soft and hard qcd radiation , accordingly this quantity has to be considered with some reserve .
for the reduced @xmath67 statistic we find @xmath76 for gmsb , @xmath77 for p1 and @xmath78 for p2 . in this case
the trend of the data is not well reproduced by the point p2 .
but since the scenario p2 provides the best fit to the atlas signal rate and the @xmath1 and @xmath2 distributions , it would be interesting to know about susy extensions of the standard model which share the features of this simplified model . as discussed in the introduction , this is possible within the nmssm . using the spectrum generator nmssmtools @xcite with decay branching fractions computed by nmsdecay @xcite ( based on hdecay @xcite ) we found that the following region of the parameter space of the @xmath79-invariant nmssm shares the following properties with the point p2 : * heavy ( decoupled ) squarks in order to satisfy constraints from searches for events with jets and @xmath1 in the presence of a gluino with a mass of 800 gev . * a bino - like neutralino @xmath73 with a mass of 600 gev , but winos and higgsinos slightly heavier than the gluino .
( the running gaugino masses do not satisfy su(5)-like relations at the gut scale . ) * a singlet - like neutralino @xmath9 with a mass of 500 gev .
then the branching fraction for the decay @xmath80 is 100% . *
the loop - induced gluino two - body decay @xmath81 should be suppressed , since it would not contribute to the signal .
it is induced by the higgsino component of @xmath9 , and can be of similar order of the desired gluino three - body decay @xmath82 .
the singlino - higgsino mixing is proportional to the nmssm - specific yukawa coupling @xmath83 @xcite , and @xmath83 should not exceed @xmath84 .
( the loop - induced gluino two - body decay @xmath85 leads to similar signals as @xmath82 , but it can be expected that it would improve the jet multiplicity distribution shifting it towards smaller values . )
the remaining parameters can be chosen to obtain a standard model - like higgs boson with a mass of @xmath86 gev .
we have checked that a corresponding point in the parameter space with all squark masses of 2.5 tev ( leading to a gluino pair production cross section of @xmath87 fb in order to compensate for a gluino br into @xmath88 slightly below 100% ) , @xmath89 , @xmath90 gev , @xmath91 , @xmath92 , @xmath93 tev and @xmath94 gev ( see @xcite for the definitions of the latter parameters ) has the properties of p2 and would not be distinguishable from p2 regarding the different observables shown above in figs .
[ fig:2][fig:4 ] .
we studied to which extent susy extensions of the sm can describe the excess of events observed by atlas in the on-@xmath0 signal region , respecting constraints by cms on similar signal channels as well as constraints from searches for jets and @xmath1 . for viable scenarios we compared the distribution of kinematical variables to the data , combining dielectron and dimuon events .
due to hadronic @xmath0-decays , gmsb - like scenarios are typically in conflict with constraints from searches for jets and @xmath1 . assuming a 100% branching fraction for the gluino decay cascade including the @xmath95 decay , these scenarios become viable only if the gluino has a mass above @xmath96 tev implying a small contribution ( @xmath97 events ) to the atlas signal region .
reducing the branching fraction ( including the @xmath95 decay ) to @xmath98 , lighter gluinos with @xmath99 gev may contribute significantly to the signal region , remaining within the 95% cl limit of cms .
however , the @xmath2 and notably the @xmath1 distributions do not coincide well with the trends of the data .
therefore we studied alternative scenarios with two massive neutralinos @xmath73 , @xmath9 . in order to compare the impact of different neutralino spectra to gmsb - like scenarios and among themselves for fixed gluino pair production cross sections and gluino masses , we fixed the latter also to 800 gev .
a very compressed @xmath100 spectrum reproduces somewhat better the trend of the @xmath2 distribution , but does not improve the shape of the @xmath1 distribution . in particular , the contribution to the atlas signal region can not be enhanced significantly while remaining within the 95% cl limit of cms .
a less compressed @xmath100 spectrum provides the best fit to the total number of events in the atlas in the on-@xmath0 signal region , as well as to the @xmath2 and @xmath1 distributions ; only the jet multiplicity is still not well reproduced .
( larger @xmath101 mass splittings as assumed here become again sensitive to constraints from searches for jets and @xmath1 . )
we found that such a scenario can be realised within the nmssm .
a somewhat different approach has recently been persued in @xcite , where the space of the two lightest neutralino masses within the nmssm was scanned systematically in order to maximise the contribution to the atlas on-@xmath0 signal region respecting existing constraints .
the shapes of the kinematical variables have not been studied , however .
still , their main results coincide with ours : whereas compressed spectra make it easier to satisfy constraints from other susy searches , the contributions to the atlas on-@xmath0 signal region are suppressed as well . for gluino masses below @xmath102
gev , only a small corner in the plane of the two lightest neutralino masses survives the 95% cl limits of cms . within this corner ( for @xmath8 , @xmath73 , @xmath9 masses of 650 , 565 and 465 gev ,
respectively ) the authors found a maximal contribution of about 11 events to the atlas on-@xmath0 signal region .
moreover , the authors of @xcite considered constraints from signal regions in @xcite ( 2jw and 4jw ) which are not implemented in the checkmate version 1.2.0 @xcite used here .
the authors applied these constraints to our benchmark point p2 and obtained a ratio for the event yield/@xmath25 of 1.19 , i.e. about 20% too large , but within the systematic errors from the simulation .
a similar excess holds for this point actually also for two cms signal regions considered in table 1 .
we recall that identical gluino masses of 800 gev were chosen for all points to simplify comparisons .
a slightly heavier gluino mass of @xmath103 gev would reduce the gluino pair production cross section by @xmath104 , but with little changes in the decay kinematics if all mass splittings remain the same .
then this modified point would pass all constraints without the help of systematic error bars , but its contribution to the atlas on-@xmath0 signal region would drop to @xmath105 events .
this number coincides with the maximum found in @xcite for the slightly different point above . clearly ,
if the excess observed by atlas indicates the presence of particles beyond the standard model , it should become more visible in both atlas and cms experiments at the run ii of the lhc .
but since it is present in the atlas analysis of the available data from run i we found it appropriate to discuss possible interpretations . within the class of models considered here ,
fits to the event numbers and shapes of the atlas on - z can be improved with respect to the gmsb scenarios considered in @xcite . however
, perfect fits would lead to unacceptable tensions with constraints from other searches .
we acknowledge support from european union initial training network invisibles ( pitn - ga-2011 - 289442 ) , the erc advanced grant higgs@lhc , and from the european union initial training network higgstools ( pitn - ga-2012 - 316704 ) .
g. aad _ et al . _ [ atlas collaboration ] , `` search for supersymmetry in events containing a same - flavour opposite - sign dilepton pair , jets , and large missing transverse momentum in @xmath106 tev @xmath107 collisions with the atlas detector , '' arxiv:1503.03290 [ hep - ex ] . the atlas collaboration , `` search for squarks and gluinos with the atlas detector in final states with jets and missing transverse momentum and 20.3 fb@xmath108 of @xmath106 tev proton - proton collision data , '' atlas - conf-2013 - 047 . | we study to which extent susy extensions of the standard model can describe the excess of events of 3.0 standard deviations observed by atlas in the on-@xmath0 signal region , respecting constraints by cms on similar signal channels as well as constraints from searches for jets and @xmath1 .
gmsb - like scenarios are typically in conflict with these constraints , and do not reproduce well the shape of the @xmath1 distribution of the data .
an alternative scenario with two massive neutralinos can improve fits to the total number of events as well as to the @xmath2 and @xmath1 distributions .
such a scenario can be realised within the nmssm .
lpt orsay 15 - 27 + * ulrich ellwanger@xmath3 * + _ @xmath3 lpt , umr 8627 , cnrs , universit de paris sud , 91405 orsay , france , and + _ school of physics and astronomy , university of southampton , + _ highfield , southampton so17 1bj , uk _ _ _ |
the conventional leading order ( lo ) and next to leading order ( nlo ) charm - anticharm production diagrams at the quark level show little or no asymmetry in the @xmath2 or @xmath9 behaviour of the quark and antiquark , as viewed in the laboratory frame .
previous studies , primarily using pion beams at a variety of energies up to 500 gev/@xmath6 , have shown that in many cases charm mesons that share a quark in common with the beam hadron , so - called _ leading charm _ , are produced more copiously in the forward hemisphere than those states which do not have a shared quark , i.e. , _ non - leading charm _ @xcite . recently ,
this has been extended to charm baryon production by a @xmath5 beam , where asymmetry for @xmath1 compared to @xmath0 as a function of @xmath2 was reported to be large @xcite .
there are limited - statistics proton data available for central production , where only a 90@xmath10 confidence level lower limit is reported @xcite . because the quark - level processes are thought to be symmetric , the asymmetries are viewed as features of the hadronization process @xcite or as a manifestation of an intrinsic charm content of the beam hadron @xcite . in this experiment the @xmath1 shares a quark in common with all three beam hadrons , while the antibaryon has an antiquark in common with the @xmath4 beam only .
we shall compare the hadroproduction characteristics for these charm baryon and antibaryon states to illuminate the role of the beam fragments in hadronization induced asymmetries .
the selex spectrometer is a three - stage magnetic spectrometer designed to cover charm production in the forward hemisphere .
the beam hadron was identified by a transition radiation detector as pion or not - pion and tracked in 8 beam silicon strip detectors before interacting in a set of 5 target foils , 2 copper and 3 carbon totalling 5@xmath10 of a proton interaction length .
interaction products were tracked in a set of 20 vertex silicon detectors arranged in 4 sets of planes rotated by @xmath11 .
single - track efficiencies for each system exceeded 98@xmath10 .
data were taken with negative beam at 615 gev/@xmath6 ( half @xmath12 , half @xmath13 ) totalling 12.3 @xmath14 inelastic interactions and positive beam at 540 gev/@xmath6 ( 92@xmath10 protons ) totalling 2.8 @xmath14 inelastic interactions .
the layout of the spectrometer can be found elsewhere @xcite .
+ two proportional wire chamber ( pwc ) magnetic spectrometer systems analyzed tracks with momenta above 3 gev/@xmath6 ( m1 spectrometer , 12 planes ) or 15 gev/@xmath6 ( m2 spectrometer , 14 planes ) .
the beam region of the pwcs was covered with double - sided silicon detectors to give high - resolution momentum measurement up to full beam energy .
the combination of excellent vertex - region tracking resolution ( typically 65 @xmath15rad at 150 gev/@xmath6 ) and excellent momentum resolution ( @xmath16 for a typical 100 gev/@xmath6 track ) gave a @xmath17 mass resolution ( @xmath18 ) of about 9 mev/@xmath19 at all @xmath2 .
tracks above 22 gev/@xmath6 traversed a ring - imaging herenkov ( rich ) and were identified as @xmath12 , k , or p @xcite .
the rich distinguished @xmath20 from @xmath21 up to 165 gev/@xmath6 .
the proton identification efficiency was greater than 95 @xmath10 above proton threshold ( @xmath22 90 gev/@xmath6 ) and greater than 90 @xmath10 above kaon threshold ( @xmath22 45 gev/@xmath6 ) .
an important innovation in selex was the use of an online topological trigger to identify charm .
the hardware trigger was loose , requiring @xmath23 charged hadrons in the forward 150 mrad cone and @xmath24 hits from positive track candidates in a hodoscope after m2 .
about 1/3 of all inelastic interactions satisfied this trigger .
the software trigger made a full vertex reconstruction of the beam track and all tracks in the m2 spectrometer ( high momentum tracks ) to test that hypothesis that they all came from a single primary vertex .
a software - adjustable @xmath25 cut of 8.5 selected candidate events that were inconsistent with a single primary vertex , i.e. , events that _ might _ have a downstream decay .
this gave a rejection factor of about 8 in the data volume .
studies with the software trigger turned off show that this costs about a factor of 2 in charm efficiency , for a net filter enhancement of 4 .
a general data reduction pass found reconstructable charged tracks and identified the interacting beam track .
primary and secondary vertices were found by geometric reconstruction without regard to particle identification .
the high resolution of the beam and vertex systems , along with the thin production targets , gave typical primary vertex resolution of 270 @xmath15 m and secondary vertex resolution for these @xmath1 events of 560 @xmath15 m for a mean @xmath1 momentum of 230 gev/@xmath6 .
all the @xmath26 candidates used in this analysis were selected by the following requirements : ( 1 ) primary and secondary vertex fits each had @xmath25/dof @xmath27 5 ; ( 2 ) the primary - secondary vertex separation significance ( l/@xmath18 ) had to be greater than 8 , where @xmath18 is the quadrature sum of the primary and secondary vertex errors ; ( 3 ) the total momentum vector of the decay tracks had to point back to the primary vertex within errors ; ( 4 ) ( @xmath0 ) @xmath1 events were required to have positive rich identification for the ( anti)proton and the ( @xmath28)@xmath29 . the pion could be any track ; ( 5 ) secondary vertices which occured inside any target were removed ; ( 6 ) each secondary track was extrapolated back to the primary vertex to determine its transverse miss distance .
the second - largest transverse miss distance had to exceed 20 @xmath15 m .
events were selected by the above cuts and sorted by the primary beam tag : @xmath5 or @xmath4 from the negative beam and protons from the positive beam .
figure [ fig : mass ] shows the full selex statistics for the @xmath30 and @xmath31 mass plots where signals for all beams are combined .
gaussian fits are shown for illustration only .
the background shapes are smooth on the low mass and high mass side of the signal regions .
the signal yield for bayons and antibaryons are determined by a sideband subtraction method which assumes a linear mass resolution dependence for the momentum spectrum .
the number of background events estimated outside the signal window is subtracted from the number of events in the signal window .
the signal window corresponds to a 3.33-@xmath18 mass width and the background is 10-@xmath18 centered at 2.285 gev/@xmath19 .
we find @xmath32 signal events for @xmath1 and @xmath33 signal events for @xmath0
the data yields were corrected in each @xmath2 bin by the total acceptance ( geometrical acceptance and reconstruction efficiency ) .
the simulation used monte carlo generated @xmath1 ( or c.c . ) events with phase space decay to the @xmath26 final state .
the simulated trajectories were converted to hits in all tracking detectors , including multiple coulomb scattering but not secondary interaction losses .
the hits were embedded in the hit banks of real data and all tracks were reconstructed .
this overestimates slightly the tracking inefficiency , but this effect is small . the rich simulation has been tested against a set of @xmath34 events taken without a rich cut .
agreement is excellent ; the proton identification efficiency averaged over the momentum spectrum is 95@xmath10 .
the simulated events were analysed with the same tracking and reconstruction procedure used in the data .
the whole set of cuts used to extract the signal was applied here as well .
the acceptance is the ratio between the number of events reconstructed by the number of embedded events generated .
selex acceptance does not depend on the transverse momentum of the charm particle up to @xmath35 ( gev/@xmath6)@xmath36 .
therefore we consider only the acceptance as a function of @xmath2 .
an important issue for this analysis is the relative efficiency for the baryon and antibaryon decays . as seen in figure [ fig : accep ] , there is no discernable difference in the acceptances . the @xmath2 distribution of the acceptance - corrected number of events from each beam is shown in figure [ fig : xf ] .
because the beam flux is the same for baryon and antibaryon , these figures compare the relative production cross sections for these states .
the curves shown are fits of the standard parametrization @xmath37 to the data .
the results are summarized in table [ tab xf ] .
.summary of measured n - value from the fit @xmath37 for @xmath1 and @xmath0 .
the errors are statistical only . [ cols="^,^,^,^,^ " , ] the relative sizes of the differential cross sections for baryon and anti - baryon production are clearly different for pion and baryon beams .
the pion beam valence quark content is baryon / antibaryon symmetric .
relative cross sections for @xmath1 and @xmath0 production from pions are comparable in shape and magnitude over the entire @xmath2 range .
such behaviour is consistent with the leading particle picture , since both charm states share a valence quark in common with the beam . for @xmath5 and proton beams @xmath1
production is much stronger than @xmath0 production , especially at high @xmath2 . in this case
there are no valence antiquarks , and antibaryon production appears to be strongly disfavored .
the four distributions that show leading behaviour are consistent with a common @xmath37 dependence having n @xmath38 2.5 , much harder than the n for leading meson production from pion beams @xcite .
note that the high - statistics @xmath5 distribution shows some interesting structures .
clearly @xmath1 production is favored over @xmath0 , and the enhancement increases with @xmath2 . the @xmath3 spectrum for @xmath1 production by the three different beam hadrons is shown in fig .
[ fig : pts ] . for @xmath39 2 ( gev/@xmath6)@xmath36 all three beams are consistent with a smooth exponential spectrum @xmath40 with the same parameter b @xmath22 1.1 ( gev/@xmath6)@xmath41 . at larger @xmath3 values ,
seen only in the high - statistics @xmath5 data , the exponential gives way to a power - law behavior as expected in qcd and as seen in @xmath42 production @xcite .
there is no indication in the large-@xmath2 events of an enhancement at small @xmath3 as suggested by the intrinsic charm picture .
however , when we look at the @xmath2 behaviour of events after the slope change at @xmath43 2 ( gev/@xmath6)@xmath36 , we see a significant difference in the @xmath2 behaviour .
the events with large @xmath3 have a softer @xmath2 spectrum than those at small @xmath3 , as seen in figure [ fig : xf ] ( d ) . the hadroproduction asymmetry a is defined as : @xmath44 where @xmath45 ( @xmath46 ) is the production cross - section for the charm particle ( anti - charm particle ) in study .
we calculated the values of the asymmetry and the errors using the maximum likelihood method with no limit at @xmath47 .
events in each @xmath2 bin follow the poisson distribution . therefore , the likelihood function is a product between two poisson distributions , one for the background region and the other for the signal region .
the results are presented for the three beam hadrons in figure [ fig : asy ] .
as is already clear from the @xmath2 distributions , @xmath8 production in @xmath5 and proton beams shows a strong asymmetry favoring baryons over antibaryons at all @xmath2 .
the proton asymmetry rises sharply toward unity already at small @xmath2 , reflective of the paucity of @xmath0 production by protons .
the @xmath5 asymmetry has a smoother approach to the limit at @xmath48 . in the case of the @xmath4 beam
we see only a slightly positive and constant asymmetry throughout the @xmath2 range .
our result agrees well in the overlap region with recent e791 data @xcite .
systematic errors in kinematic analysis may arise from several sources .
we list here the ones that have been considered . in all cases
systematic effects are found to be small compared to statistical errors . * @xmath2 binning effects + the @xmath2 distribution is relatively smooth , but there are structures in the high statistics data .
we have shifted the bin limits for all @xmath2 distributions by 0.025 ( half a bin ) and refit the data .
no effect in the character of the distributions is observed .
for example , all the structures seen in fig.3(c ) are also present in the shifted distribution .
changes in the n - value are consistent with purely - statistical effects .
therefore , we quote a conservative @xmath49n for the fit of 0.1 , even though no systematic effect has been seen . + * mass window size + the signal window in the mass distribution is taken as @xmath5030 mev/@xmath19 around 2.285 gev/@xmath19 .
we have changed the window up to @xmath5035 mev/@xmath19 and down to @xmath5025 mev/@xmath19 .
the systematic error for changing the mass window compared to statistical error is negligible .
* acceptance corretion + the acceptance shown in fig .
2 is computed using the method of adding hits from simulated charm decay tracks into the data banks from an observed event .
the modified event is reanalysed to see if the charm state is reconstructed
. there could be a shift in the corrected @xmath2 distribution if the acceptance calculation does not represent truly the @xmath2 variation of the apparatus efficiency . in e781 the dominant efficiency variation at low @xmath2
is purely geometric : both proton and kaon pass through the fiducial volume of the rich counter .
a second geometric effect is the related one : the charm particle live long enough to pass the vertex significance cut .
the tracking characteristics of the spectrometer match those predicted from the simulation very well over all momenta studied here .
the @xmath51 mass is the same for all observed momenta , both in data and simulation .
we believe that there is no systematic bias from the simulation and assign no error .
* trigger bias in asymmetry + the @xmath2-dependent effects tend to cancel in the asymmetry measurement .
however , a new systematic effect must be considered : trigger bias .
the e781 trigger required counts on the postive - bend side of a trigger hodoscope after the second bending magnet .
this _ might _ favor @xmath1 production over @xmath52 .
we looked at the asymmetry of @xmath51 production , for times when the @xmath53 is found in the high - momentum spectrometer compare to the converse .
we see a maximum 3% asymmetry .
we also look at the charm baryon data itself .
if we @xmath54 the charm tracks , then in 97% of the triggers there are still enough hits in the hodoscope to trigger the event without the charm decay tracks .
we conservatively assign a 3% systematic error to the asymmetry from trigger bias .
the selex experiment reconstructed about 2500 @xmath55 from 600 gev/@xmath6 @xmath5 and @xmath4 beams and a 540 gev/@xmath6 proton beam .
we observed that the @xmath2-dependence of @xmath1 production is similar for all three beams , much stiffer than for meson production .
both baryon beams show a strong enhancement of the production of charm baryons over antibaryons , while the two are produced comparably from a pion beam .
these results are in general agreement with color - drag models of hadronization .
the differences in behaviour for protons and @xmath5 illustrate significant non - perturbative effects in production dynamics at these energies .
the authors are indebted to the staff of fermi national accelerator laboratory and for invaluable technical support from the staffs of collaborating institutions .
this project was supported in part by bundesministerium fr bildung , wissenschaft , forschung und technologie , consejo nacional de ciencia y tecnologa ( conacyt ) , conselho nacional de desenvolvimento cientfico e tecnolgico , fondo de apoyo a la investigacin ( uaslp ) , fundao de amparo pesquisa do estado de so paulo ( fapesp ) , the israel science foundation founded by the israel academy of sciences and humanities , istituto nazionale di fisica nucleare ( infn ) , the international science foundation ( isf ) , the national science foundation ( phy # 9602178 ) , nato ( grant cr6.941058 - 1360/94 ) , the russian academy of science , the russian ministry of science and technology , the turkish scientific and technological research board ( tbitak ) , the u.s .
department of energy ( doe grant de - fg02 - 91er40664 and doe contract number de - ac02 - 76cho3000 ) , and the u.s .- israel binational science foundation ( bsf ) .
deceased present address : infinion , mnchen , germany now at imperial college , london sw7 2bz , u.k .
now at instituto de fsica da universidade estadual de campinas , unicamp , sp , brazil now at physik - department , technische universitt mnchen , 85748 garching , germany present address : the boston consulting group , mnchen , germany now at instituto de fsica terica da universidade estadual paulista , so paulo , brazil present address : lucent technologies , naperville , il present address : spss inc .
, chicago , il now at university of alabama at birmingham , birmingham , al 35294 present address : doe , germantown , md present address : siemens medizintechnik , erlangen , germany present address : deutsche bank ag , eschborn , germany selex collaboration , j.s .
et al . _ , in _ proceedings of the 29th international conference on high energy physics _ , 1998 , edited by a. astbury _
( world scientific , singapore , 1998 ) , vol .
ii , p. 1259 ; hep - ex/9812031 . | we present data from fermilab experiment e781 ( selex ) on the hadroproduction asymmetry for @xmath0 compared to @xmath1 as a function of @xmath2 and @xmath3 distributions for @xmath1 .
these data were measured in the same apparatus using incident @xmath4 , @xmath5 beams at 600 gev/@xmath6 and proton beam at 540 gev/@xmath6 .
the asymmetry is studied as a function of @xmath2 . in the forward hemisphere with @xmath7 both baryon beams exhibit very strong preference for producing charm baryons rather than charm antibaryons , while the pion beam asymmetry is much smaller . in this energy regime
the results show that beam fragments play a major role in the kinematics of @xmath8 formation , as suggested by the leading quark picture . |
the entanglement entropy , being a measure of the correlations between two subsystems , depends on the geometry of the separating surface , becoming for this reason a quantity with a strong geometric character .
this has inspired a holographic interpretation within the ads - cft correspondence framework @xcite , which provides a purely geometric way of calculating the entropy of conformal field theories .
there are several results in the literature where the geometrical properties for different sets , dimensions and theories are explored both , from the quantum field theory and holographic approaches . among these , the coefficient @xmath1 of the logarithmically divergent term in the entanglement entropy @xmath2 , where @xmath3 is the ultraviolet cut - off , for general conformal field theories in four dimensions has been found to be proportional to the type @xmath4 and @xmath0 conformal anomaly coefficients @xcite @xmath5 here @xmath6 is the euler number of the surface , @xmath7 is the second fundamental form , @xmath8 with @xmath9 are a pair of unit vectors orthogonal to @xmath10 , and @xmath11 is the induced metric on the surface . from ( [ general ] ) , it can be seen that the sphere and the cylinder are sensitive only to one type of anomaly coefficient , @xmath4 or @xmath0 , respectively .
this is @xcite @xmath12 for the sphere , while for a cylinder of length @xmath13 and radius @xmath14 it is @xmath15 for spherical sets , this result was later recalculated , numerically and analytically @xcite , using different technics , and recently extended to any theory @xcite . instead , the validity of the result ( [ coefs ] ) for the cylinder is more subtle .
it has been also studied from the holographic point of view in @xcite where @xmath1 is found to be proportional to the @xmath0 conformal anomaly in four dimensions as in ( [ coefs ] ) , but it can not be generalized to higher dimensions where other conformally invariant terms can be added to ( [ general ] ) . in this paper , we study numerically the logarithmic contributions to the entropy for a cylinder in @xmath16 dimensions for free massless scalar and dirac fields .
the method we use consists first , in dimensionally reducing the problem of the three dimensional cylinder to the one of an infinite set of massive fields living in a two dimensional spatial disk .
the entanglement entropy is then calculated numerically by the real time approach @xcite where the reduced density matrix is written in terms of free correlators .
this is based , on one hand , on the method developed by sredniki for calculating the entropy for scalars in spherical sets @xcite where the field is discretized in the radial direction in polar coordinates , and on the other , on the work by peschel , whereby the reduced density matrix can be written in terms of correlators for both , bosonic and fermionic , solvable lattice systems @xcite .
thus , for the numerical evaluation of the entropy of a dirac field in the disk , we extend the method described in @xcite , originally applicable only to scalar fields .
then , from the expansion of the entropy in powers of @xmath17 valid in the large @xmath18 limit , we find the coefficient @xmath1 for the cylinder ( [ coefs ] ) .
this is directly related to the coefficient of the term @xmath19 . in the same expansion
, we also identify the coefficient of the linear term @xmath18 .
this last , has been recently calculated analytically in @xcite , where the same type of expansion is considered for general smooth geometries and massive scalar theories . from the holographic point of view
, new contributions to the entanglement entropy were also found when considering theories deformed by a relevant operator @xcite , consistently to that reported in @xcite .
we find , for the cylinder , within a porcentual error @xmath20 for scalars and @xmath21 for dirac fields , @xmath22 and for the disk , within a porcentual error @xmath23 , @xmath24 in agreement with @xcite and @xcite .
coming from the same expansion , each coefficient is cross checked by the other testing both at the same time
. this concordance sums support to the general validity of the method used in @xcite .
we will discuss this issue in more detail , later in the discussion .
the paper is organized as follows : in the second section , we discuss the dimensional reduction procedure and show the problem of massless fields in the cylinder can be reduced to the one of massive fields in a disk . in the third section , we discuss the approach we use for the numerical evaluation of the entanglement entropy in the disk , for both , scalar and dirac massive fields . in section 4
we present our results and finally , its interpretation in the discussion .
for free fields , some universal terms in the entanglement entropy in high dimensions can be obtained via dimensional reduction technics from results calculated in lower dimensions @xcite .
let us consider a set in three spatial dimensions of the form @xmath25 , where @xmath26 is a line on the first coordinate @xmath27 , of length @xmath13 , and @xmath28 is a sphere in two dimensions ( disk ) .
we are interested in the entropy of @xmath29 in the limit of large @xmath13 .
the direction @xmath27 can be compactified by imposing periodic boundary conditions @xmath30 , without changing the result of the leading extensive term . for a free massless field
we fourier decompose it into the corresponding field modes in the compact direction @xmath31 the problem then reduces to a two dimensional one with an infinite tower of massive fields .
for example , for a free scalar we obtain the tower of fields @xmath32 , @xmath33 from the point of view of the non compact @xmath34 directions , these fields have masses given by @xmath35 summing over the contributions of all the decoupled @xmath36 dimensional fields we have @xmath37 in the limit of large @xmath13 we can convert this sum into integral @xmath38 the universal terms in @xmath39 will then come from the ones of @xmath40 after integrating over the mass . for a dirac field , the spin multiplicity factor @xmath41}$ ]
has to be incorporated . in the present case ,
it is @xmath42 expanding the entanglement entropy @xmath43 in powers of @xmath18 for large @xmath18 @xmath44 and inserting ( [ expansion ] ) in ( [ ala ] ) and ( [ ala1 ] ) for scalar and dirac fields respectively , we obtain for the @xmath16 dimensional theory that the logarithmic coefficient @xmath1 in @xmath39 is directly related to @xmath45 by @xmath46 on the other hand , for the dimensionally reduced theory , the contribution in the entropy proportional to the mass is given by the term proportional to the coefficient @xmath47 .
this contribution was calculated in @xcite for massive scalar fields in any dimension for a waveguide geometry with specified boundary conditions using heat kernel methods .
the terms extensive in the area @xmath48 in even spatial dimensions are given by @xmath49 from ( [ heatkernel ] ) it follows , @xmath50 where @xmath51^{-1}$ ] for @xmath52 even . for dirac fermions with @xmath41}$ ] components ,
the same calculation can be done taking into account the extra factor @xmath41 - 1}$ ] @xcite , which relates the scalar and dirac @xmath53 coefficients , @xmath54 - 1}\gamma_d^{s}.\ ] ] in our case , @xmath55 and @xmath56 both for dirac and scalar fields .
we evaluate numerically the entanglement entropy for a two dimensional disk . in this section
we describe the numerical method and models in the lattice for scalar and dirac massive fields .
consider the quadratic hamiltonian for a massive scalar in @xmath57 dimensions @xmath58 following @xcite , we separate variables in polar coordinates and introduce new ones @xmath59 such that @xmath60=i \delta_{nn^{\prime}}\delta(r - r^{\prime } ) \,.\ ] ] then , the hamiltonian @xmath61 takes the form @xmath62\ , .
\label{hscalar}\ ] ] for a general quadratic discrete hamiltonian @xmath63 , ( dropping the @xmath64 index temporarily ) , the vacuum ( ground state ) two point correlators @xmath65 and @xmath66 @xmath67 are given by @xcite @xmath68 in terms of the @xmath69 matrix .
here , @xmath70 , where @xmath71 is the size of the lattice and acts as an infrared regulator . then , the entropy of the disk is given by @xmath72 with @xcite @xmath73 the superscript @xmath28 in ( [ for ] ) means the indices of the matrices are restricted to the region @xmath28 : @xmath74 , with @xmath75 the radius of the disk . in this case , after discretization we find the matrix @xmath76 corresponding to ( [ hscalar ] ) is @xmath77 summarizing , the numerical evaluation of the entropy for massive scalar fields in a disk starts with the calculation of the @xmath78 matrix @xmath76 for a given mass @xmath79 , angular momentum @xmath64 , and infrared regulator @xmath71 which gives the size of the unidimensional lattice . from ( [ x ] ) and ( [ p ] ) we calculate the two point correlators @xmath26 and @xmath80 .
then , we reduce them to the disk and calculate the contribution @xmath81 ( [ for ] ) . finally , the entropy @xmath82 is given by the sum ( [ sscalar ] ) .
the hamiltonian for a massive dirac field in @xmath57 dimensions can be written as @xmath83 with @xmath84 the one particle hamiltonian @xmath85 choosing @xmath86 with @xmath87 the pauli matrices , the relations @xmath88 and @xmath89 are satisfied . in polar
coordinates @xmath90 the hamiltonian takes the form @xmath91 where @xmath92 , @xmath93 and @xmath94 are @xmath95 separating variables , we propose the following two component dirac spinor @xmath96 where @xmath97 and @xmath98 are the eigenvector components of the angular momentum operator @xmath99 with half integer eigenvalue @xmath64 @xmath100 note that @xmath101 commutes with @xmath102 .
then , we can express the hamiltonian as a sum over @xmath64 such that @xmath103 where @xmath104 and @xmath105 in general , in the discrete free case , the reduced density matrix can be written in terms of a hermitian operator @xmath106 @xmath107 this can be expressed in terms of the correlators @xmath108 @xmath109 via the identification @xcite @xmath110 here , the @xmath111 are fermion operators canonically normalized .
then , the entropy can be written in terms of @xmath112 as @xmath113 . for a general quadratic case , with discrete hamiltonian @xmath114 , the correlator is directly related to @xmath115 by @xmath116 in ( [ cfe ] ) , the indices run over the complete space .
then , we introduce an infrared regulator @xmath71 which is the size of the lattice . therefore , the entropy @xmath117 for the disk @xmath28 , is given by a sum over the angular momentum @xmath64 as @xmath118 the correlators @xmath119 for fixed angular momentum @xmath64 are restricted to the disk region @xmath28 . in order to identify the @xmath120 matrix in the case of the disk , we first introduce the operators @xmath121 .
these are normalized such that they satisfy the canonical anticonmutation relations required for the application of ( [ entd ] ) .
the discrete hamiltonian in these variables for a @xmath71 lattice , takes the form ( dropping the angular momentum index temporarily ) @xmath122 where the indices @xmath123 are the discrete variables corresponding to the continuum radial coordinate @xmath75 .
@xmath124 , for fixed @xmath123 , is a @xmath125 matrix such that @xmath126 with @xmath127 .
finally , we define a @xmath128 matrix @xmath129 , for @xmath130 and @xmath131 .
the non zero entries of @xmath132 ( for each angular momentum @xmath64 ) are @xmath133 the @xmath134 matrix satisfies @xmath135 and it has real eigenvalues symmetric with respect to the origin .
there is also a symmetry related to the mapping @xmath136 which can be seen also in the continuum limit .
[ tp ] summarizing , the numerical evaluation of the entropy for massive dirac fields in a disk starts with the calculation of the @xmath137 matrix @xmath138 for a given mass @xmath79 , angular momentum @xmath64 , and @xmath71 , the size of the unidimensional lattice . from ( [ cfe ] )
we calculate the two point correlator @xmath112 ( [ corf ] ) .
then , we reduce it to the disk @xmath119 and calculate the contribution @xmath81 in ( [ entd ] ) . finally , the entropy @xmath82 is given by the sum ( [ entd ] ) .
we calculate the entanglement entropy for a disk in a lattice of @xmath139 points .
we consider regions of radii @xmath75 going from @xmath140 to @xmath141 in lattice units for the scalar and from @xmath142 to @xmath140 for the fermionic field .
the sum over the angular momentum @xmath64 is done exactly up to @xmath143 and the corrections coming from contributions @xmath144 are added by fitting the exact entropy for some large values of @xmath64 .
the fit we use is of the form @xmath145 since for large @xmath64 ( @xmath146 ) both @xmath112 and @xmath147 defined in ( [ cfe ] ) , ( [ x ] ) and ( [ p ] ) have expansions in even inverse powers of @xmath64 @xcite .
this is done for different masses in the range @xmath148 such that @xmath149 ( we fix @xmath150 ) and @xmath151 . for each mass
, we fit the entropy in terms of the size @xmath75 of the disk , @xmath152 once the coefficients @xmath153 and @xmath154 are identified , we expand in powers of @xmath79 , @xmath155 we obtain @xmath47 and @xmath156 of ( [ expansion ] ) as the @xmath79 and @xmath157 coefficients in ( [ c1 m ] ) and ( [ c-1 m ] ) respectively . for example , in the scalar case , for @xmath158 we calculate the entanglement entropy for disks with radii from @xmath159 to @xmath141 ( in lattice units ) . from the best fit ,
we extract the coefficients of the terms proportional to @xmath75 and @xmath160 . in this example , this gives @xmath161 .
then , @xmath162 and @xmath163 .
the values of @xmath153 and @xmath154 are shown in figures ( 1),(2 ) for scalars and figures ( 3),(4 ) for the dirac field . in order to compare the lattice results with the continuum expectations for fermions , a factor @xmath164 has to be incorporated in the entropy formula due to the fermion doubling in the unidimensional radial lattice .
we obtain @xmath165 and @xmath166 in agreement to the analytical results .
our numerical results agree with the ones predicted analytically within porcentual errors @xmath167 or less . since both results
, the one for the cylinder and the one for the disk , come from the identification of different coefficients in the same expansion ( [ expansion ] ) , they cross check each other giving a solid basement to our numerical test . on the other hand , by the analytical side within the qft approach ,
the general grounds the result for the cylinder is based on , rely first on the application of the replication method , then on demanding conformal invariance of the logarithmic contribution in the entropy derived from the effective action of the replicated manifold , and finally on calibrating a free parameter by using the holographic correspondence @xcite .
the replica trick is a standard approach to calculate the entanglement entropy @xcite . in this construction
, the trace of the integers powers of the reduced density matrix corresponds to the partition function of @xmath64 copies of the system connected consistently through the cuts along the boundary .
the replicated manifold is non trivial due to the conical singularities of deficit angle @xmath168 placed on the boundary . from this
, the entropy associated to the set @xmath29 can be calculated as @xmath169=-\lim_{n \rightarrow 1}(\frac{\partial}{\partial n}-1)\log z_n\ , .
\label{ee}\ ] ] in the above formula we assume that not only the partition function @xmath170 is calculable , but it is also possible to analytically continue it to non integer @xmath64 .
the method presented in @xcite differs subtly from the above : instead of continuing analytically in @xmath64 the partition function @xmath170 , they first analytically continue the cover manifold assuming the corresponding geometry can be defined for an arbitrary and small angular deficit in the limit @xmath171 and then calculate the corresponding @xmath170 @xcite .
taking into account this is not possible in general , then , the identification of the logarithmic coefficient with linear combinations of the stress tensor anomalies is not fully justified @xcite . in the case of spherical sets ,
the objections to the applicability of the replication method disappear since there is an extra rotational symmetry in the transverse space about the entangling surface .
in fact , the result for the sphere in four dimensions can be extended to any dimensions and any field theory @xcite .
these results complete a solid proof for the term proportional to the type @xmath4 anomaly in ( [ general ] ) .
on the other hand , for the cylinder , the situation is more obscure .
very recently , it was found in @xcite for holographic models , that in four dimensions the logarithmic coefficient is proportional to the anomaly as expected , but that the same is not true for higher dimensions , where new conformally invariant terms could be added to ( [ general ] ) . in this scenario ,
we conclude our results give support to the validity of the solodukhin s calculation for the cylinder in four dimensions .
i thank h.casini for useful discussions and comments on the manuscript .
i also acknowledge support from the eplanet programme grant and dr .
g. mussardo for his hospitality at sissa where the initial stages of this work were done .
this work was partially supported by conicet , anpcyt and universidad nacional de cuyo , argentina .
j. s. dowker , `` entanglement entropy for even spheres , '' arxiv:1009.3854 [ hep - th ] , j. s. dowker , `` entanglement entropy for odd spheres , '' arxiv:1012.1548 [ hep - th ] . h. casini and m. huerta , `` entanglement entropy for the n - sphere , '' phys . lett .
b * 694 * , 167 ( 2010 ) [ arxiv:1007.1813 [ hep - th ] ]
. s. n. solodukhin , `` entanglement entropy of round spheres , '' phys .
b * 693 * , 605 ( 2010 ) [ arxiv:1008.4314 [ hep - th ] ] . j. s. dowker , `` hyperspherical entanglement entropy , '' j. phys .
a * 43 * , 445402 ( 2010 ) [ arxiv:1007.3865 [ hep - th ] ] .
h. casini , m. huerta and r. c. myers , `` towards a derivation of holographic entanglement entropy , '' jhep * 1105 * , 036 ( 2011 ) [ arxiv:1102.0440 [ hep - th ] ] .
l. y. hung , r. c. myers and m. smolkin , `` on holographic entanglement entropy and higher curvature gravity , '' jhep * 1104 * , 025 ( 2011 ) [ arxiv:1101.5813 [ hep - th ] ] .
l. y. hung , r. c. myers and m. smolkin , `` some calculable contributions to holographic entanglement entropy , '' jhep * 1108 * , 039 ( 2011 ) [ arxiv:1105.6055 [ hep - th ] ] .
d. n. kabat , `` black hole entropy and entropy of entanglement , '' nucl .
b * 453 * , 281 ( 1995 ) [ arxiv : hep - th/9503016 ] .
see for example , c. holzhey , f. larsen and f. wilczek , `` geometric and renormalized entropy in conformal field theory , '' nucl .
b * 424 * , 443 ( 1994 ) [ arxiv : hep - th/9403108 ] . s. n. solodukhin , phys .
d * 51 * ( 1995 ) 609 , [ arxiv : hep - th/9407001 ] ; d. v. fursaev , s. n. solodukhin , phys .
b * 365 * ( 1996 ) 51 , [ arxiv : hep - th/ 9412020 ] ; d. v. fursaev , s. n. solodukhin , phys . rev .
d * 52 * ( 1995 ) 2133 , [ arxiv : hep - th/ 9501127 ] ; r. b. mann , s. n. solodukhin , nucl .
b * 523 * ( 1998 ) 293 , [ arxiv : hep - th/ 9709064 ] ; s. n. solodukhin , phys .
d * 57 * ( 1998 ) 2410 , arxiv : hep - th/9701106 .
r. c. myers and a. sinha , `` holographic c - theorems in arbitrary dimensions , '' jhep * 1101 * , 125 ( 2011 ) [ arxiv:1011.5819 [ hep - th ] ] .
a. schwimmer and s. theisen , `` entanglement entropy , trace anomalies and holography , '' nucl .
b * 801 * , 1 ( 2008 ) [ arxiv:0802.1017 [ hep - th ] ] .
s. n. solodukhin , `` entanglement entropy of black holes , '' arxiv:1104.3712 [ hep - th ] . | we calculate numerically the logarithmic contribution to the entanglement entropy of a cylindrical region in three spatial dimensions for both , free scalar and dirac fields . the coefficient is universal and proportional to the type @xmath0 conformal anomaly in agreement with recent analytical predictions .
we also calculate the mass corrections to the entanglement entropy for scalar and dirac fields in a disk .
these apparently unrelated problems make contact through the dimensional reduction procedure valid for free fields whereby the entanglement entropy for the cylinder can be calculated as an integral over masses of the disk entanglement entropies . coming from the same numerical evaluation in the lattice
, each coefficient is cross checked by the other , testing in this way the two results simultaneously . |
even though the standard model(sm)@xcite has been successful in describing the strong , weak and electromagnetic interaction phenomena at the energy scale up to @xmath7 gev , it has been known for a long time that it is still considered as a low - energy effective theory suffering from a number of theoretical difficulties , such as hierarchy problem .
it is likely that at the higher energy scale the minimal supersymmetric standard model ( mssm)@xcite is the most attractive candidate among various extensions of the sm , which is expected to resolve theoretical difficulties existing in the sm . by adopting two higgs doublets to preserve the supersymmetry in the mssm , five higgs bosons are predicted .
they are two cp - even higgs scalar bosons @xmath8 and @xmath9 , one cp - odd higgs boson @xmath10 and a charged higgs pair @xmath11 .
the charged higgs boson plays an important role in the extensions of the sm , its discovery in experiment will unequivocally imply the existence of the physics beyond the sm .
if the charged higgs boson really exists in nature then it is mostly possible to be discovered in the coming few years at the cern large hadron collider ( lhc ) due to its tev scale colliding energy , but the further accurate measurements of masses of higgs bosons and investigations of the properties of higgs bosons@xcite , which is crucial for verification of electroweak ( ew ) symmetry breaking mechanism , need be best performed in a clean environment of @xmath12 collisions , such as at the future international linear collider ( ilc ) @xcite .
the study would be of the utmost importance for establishment of the mssm or some of more general supersymmetric models as the theory of the ew interactions .
in addition , further study of the complete higgs spectrum is the cornerstone of the definitive test of mechanism of electroweak symmetry breaking . in discovery and detailed study of the properties for these higgs bosons , there are some important processes for neutral and charged higgs bosons at linear colliders , such as , @xmath13 , @xmath14 , @xmath15 , @xmath16 and so on@xcite .
the lightest neutral higgs boson @xmath8 in the mssm has the mass upper limit of about @xmath17 , and it decays dominantly into @xmath18 pairs . for the heavier neutral higgs bosons @xmath19 , as a result of the strong enhancement of the coupling between higgs boson and down - type fermions , both @xmath9 and @xmath10 will decay almost exclusively into @xmath18 ( @xmath20 ) and @xmath21 ( @xmath22 ) pairs ( the decays @xmath23 are possible only when they are kinematically allowed ) , and on the contrary all other decays are strongly suppressed@xcite . for the mass of charged higgs boson being blow @xmath24 ( i.e. , the @xmath25 decay threshold ) , the dominant decay mode of the charged higgs boson is @xmath26 .
but when the mass of charged higgs boson is above the @xmath25 decay threshold , the charged @xmath11 boson mainly decays into @xmath27 pair leading to @xmath28 final states@xcite . if @xmath29 pair production at @xmath12 linear colliders(lc ) is kinematically allowed , the dominant production process of charged higgs bosons is the pair production via exchanging a virtual photon and @xmath30 boson @xmath31@xcite . for a fixed @xmath12 colliding energy ,
its cross section is related only with the charged higgs boson mass .
another important production mode is the production of single charged higgs boson , especially when the charged higgs boson is too heavy to be produced in pair at linear colliders . among various single
charged higgs boson production channels , the bottom quark associated production process of @xmath32@xcite is experimentally interesting since it can be yielded by many topologies , such as the @xmath33 and @xmath29 pair productions followed by their corresponding sequential decays .
the @xmath34 process can be also induced from following three single charged higgs boson production processes ( [ process ] ) and the sequential decays @xmath35 , @xmath36 s. kanemura , s. moretti , _ et al , _ studied the processes of single charged higgs boson production channels in the mssm at lcs , including the processes ( [ process ] ) at tree - level in ref.@xcite .
it was found that there are regions of parameter space where it is accessible beyond the kinematic limit for pair production , and these processes are excellent channels in studying the interactions among neutral higgs boson(@xmath8 , @xmath9 or @xmath10 ) , charged higgs boson and gauge boson(@xmath1 ) , and is a useful alternative channel for studying the phenomenology of charged higgs boson . in present work
, we will study the ew corrections to the processes ( [ process ] ) including complete one - loop diagrams in the mssm .
the paper is organized as follows : in sect.2 , we present the analytical calculation of the complete one - loop ew radiative corrections to @xmath37 processes in the mssm as well as the treatment of charged higgs boson resonance . the numerical results and discussions are given in sect.3 .
finally , we give a short summary .
we denote the processes as @xmath38 the lowest order cross section reads @xmath39 where @xmath40 is the c.m.s .
kinematical momentum of the incoming positron , the integration is performed over the three - body phase space of final particles @xmath41 , the bar over summation indicates averaging over initial spins .
the phase - space element @xmath42 is defined by @xmath43 in our calculation , we use dimensional reduction(dr ) regularization scheme , which is supersymmetric invariant at least one - loop level , to regularize uv divergences and adopt on - shell(os ) conditions(neglecting the finite widths of the particle ) to renormalize the relevant fields@xcite .
we adopt the t hooft - feynman gauge and the definitions of one - loop integral functions in ref.@xcite in the calculation of one - loop diagrams .
the feynman diagrams and their corresponding amplitudes are created by @xmath44 @xcite automatically , and the feynman amplitudes are subsequently reduced by @xmath45@xcite .
we depict the lowest - order feynman diagrams in fig.1 .
the numerical calculation of the two- , three- and four - point integral functions are done by using ff package@xcite .
the implementations of the scalar and the tensor five - point integrals are carried out exactly by using the fortran programs as used in our previous works@xcite with the approach presented in ref.@xcite . due to the fact that the yukawa coupling of higgs / goldstone to the fermions is proportional to the mass of the fermion
, we ignore the contributions of the feynman diagrams which involve the yukawa coupling between any higgs / goldstone boson and electrons . .
] the @xmath46 virtual correction to the cross section can be expressed as @xmath47 where @xmath48 is the renormalized amplitude involving all the one - loop electroweak feynman diagrams and corresponding counter - terms@xcite . ]
the disposal of @xmath56 resonance in the calculation of the ew correction is similar with the method described in ref.@xcite . for the process
@xmath57 in @xmath58 scale colliding energy , the amplitude of the feynman diagram with internal line of @xmath56(shown in fig.2(a ) , denoted as @xmath59 ) suffers from resonance effect when @xmath60 due to the real charged higgs boson decay @xmath61 .
we have to introduce the complex @xmath56 mass to do the following replacement in the resonance propagator of the amplitude @xmath59 , @xmath62 where @xmath63 is @xmath56 decay width
. then we get the amplitude at the tree - level as @xmath64 the modified amplitude @xmath65 is safe amplitude being free of the @xmath56 resonance singularity . we know that at the tree - level calculation this replacement does not induce the gauge invariant and double - counting problems@xcite , but we should remember that after this treatment , there includes the contribution from the imaginary part of charged higgs boson self - energy(@xmath66 ) at higher order . in order to avoid the double - counting in calculation up to one - loop level , we should subtract the higher order contribution from the tree - level amplitude @xmath65 . in our practical calculation
, we separate some special feynman diagrams(shown in fig.2(b ) ) from entirety .
each of these diagrams in fig.2(b ) contains a one - loop of charged higgs boson self - energy , or the corresponding counterterm .
then we can express the one - loop level amplitude as below : @xmath67_{\frac{1}{s_{34}-m_{h^-}^2 } \to \frac{1}{s_{34}-m_{h^-}^2+i m_{h^-}\gamma_{h^- } } } { \nonumber}\\ & & -\left[\frac{\sigma^{h^-h^-}(s_{34})-\left[\sigma^{h^-h^-}(m_{h^-}^2)-b_{a^0a^0}\right]}{s_{34}-m_{h^-}^2 } + \delta z_{h^-h^- } \right]{\cal m}_{tree}^{h^-},\end{aligned}\ ] ] where the first term on right - handed side is just the amplitude at one - loop level after subtracting the the contributions of fig.2(b ) .
the one - loop(or higher ) order contributions are included by doing the replacement shown in eq.([replace ] ) .
the real part of @xmath68 is just equal to @xmath69 , the mass counterterm of charged higgs boson , while @xmath68 receives its imaginary part in the resonance from the subtraction of the @xmath70 contribution already contained in @xmath71 . in the calculation ,
charged higgs boson width was evaluated using the hdecay package@xcite , and uv divergence cancelled numerically .
the @xmath46 virtual correction suffers from both ultraviolet(uv ) and infrared(ir ) divergences .
it is checked numerically that the uv divergence vanished after renormalization .
ir divergence comes from virtual photonic loop correction .
it can be cancelled by the contribution of the real photon emission processes .
we denote the real photon emission as @xmath72 we adopt the general phase - space - slicing method @xcite to separate the soft photon emission singularity from the real photon emission processes . by using this method ,
the bremsstrahlung phase space is divided into singular and non - singular regions by the soft photon cutoff @xmath73 , then the relative correction of the real photon emission is broken down into corresponding soft term @xmath74 with radiated photon energy @xmath75 and hard terms @xmath76 with @xmath77 , where @xmath78 and @xmath79 is the mass of photon , which is used to regulate the ir divergences existing in the soft term .
the ir divergence from the soft contribution cancels that from the virtual correction exactly@xcite .
therefore , the sum of the virtual and soft cross section is independent of the ir regulator @xmath79 . in the calculation ,
the phase space integration of the processes @xmath80 with hard photon emission is performed by using the program @xmath81 @xcite .
the hard photon emission cross section @xmath82 is uv and ir finite with @xmath83 .
finally , we can get the uv and ir finite @xmath84 correction .
in the numerical calculation , we use the following sm parameters @xcite @xmath85 here we take the effective values of the light quark masses @xmath86 and @xmath87 which can reproduce the hadronic contribution to the shift in the fine structure constant @xmath88 @xcite . as a numerical demonstration , we generally adopt the relevant supersymmetric parameters from the cp and r - parity invariant mssm reference point @xmath2 which is defined in the spa project@xcite .
this point satisfies all the available high- and low - energy precision data and both the bounds for the masses of the susy particles and the bounds from cosmology .
as we use the on - shell renormalization scheme , we have to transform the @xmath2 @xmath89 input parameters @xmath90 into on - shell parameters @xmath91 .
this transformation is simply performed by subtracting the corresponding counterterms i. e. @xmath92 .
the input parameters for the reference point @xmath2 are defined in the @xmath89 scheme at the scale @xmath93 , which are a set of msugra parameters listed below : @xmath94 the relevant parameters of @xmath2 point are listed in table [ table ] . .mass spectrum of supersymmetric particles and higgs bosons in the reference point @xmath95 . [ cols="^,^,^,^",options="header " , ] since the difference between @xmath89 and os mass values for sparticle and higgs bosons is of high order , either of them can be used . in our paper
we generally adopt the @xmath89 masses of sparticle and higgs boson , but when we study the cross sections as the functions of @xmath96 or take @xmath97 , we still adopt the @xmath2 mssm spectrum and parameters except higgs sector . in the later case
the masses of the mssm higgs bosons are fixed by taking into account the significant radiative corrections in the mssm by using formcalc package with the input parameters : @xmath96 and os @xmath98 value(i.e .
, @xmath99 at @xmath2 point)@xcite .
we checked in our calculation that the total cross section is independent on the ir regulator parameter @xmath79 and the soft cutoff @xmath100 . in following
we take @xmath101 and @xmath102 .
the born and @xmath46 electroweak corrected cross sections at the @xmath2 point for process @xmath103 as the functions of c.m.s energy in the energy range from @xmath104 to @xmath105 are illustrated in fig.3(a ) .
the tree - level total cross section is drawn in solid line , while the full one - loop ew corrected cross section in dashed line .
we can read from the figure that at the position of @xmath106 , the born and ew corrected cross sections are @xmath107 and @xmath108 , respectively .
the corresponding relative corrections are depicted in fig.3(b ) and we can see that the absolute relative correction is very large when the colliding energy goes close to the vicinity of the threshold for process @xmath109 .
fig.3(b ) shows that when @xmath110 , @xmath111 is about @xmath112 , but when the @xmath113 goes to @xmath114 , the absolute relative correction declines to the value of @xmath115 .
electroweak corrected cross sections as well as the corresponding relative corrections @xmath111 as the functions of the c.m.s energy @xmath113 at @xmath2 point for process @xmath103 .
( a ) cross section versus @xmath113 .
( b ) @xmath111 versus @xmath113.,title="fig : " ] electroweak corrected cross sections as well as the corresponding relative corrections @xmath111 as the functions of the c.m.s energy @xmath113 at @xmath2 point for process @xmath103 .
( a ) cross section versus @xmath113 .
( b ) @xmath111 versus @xmath113.,title="fig : " ] in fig.4(a ) , we present the born cross section @xmath116 and one - loop electroweak corrected cross section @xmath117 with @xmath118 and related mssm parameters being taken from @xmath2 scenario except higgs sector for reaction @xmath109 . in this case the mass of the lightest neutral higgs boson @xmath119 is equal to @xmath120 , which is fixed in the mssm by using formcalc program with the input parameters : @xmath121 and os @xmath122 .
in contrast with the plots in fig.3(a - b ) , the complete ew one - loop corrections enhance born cross sections as shown in fig.4(a - b ) , and both the @xmath116 and the ew corrected @xmath117 in fig.4(a ) are at least one order larger than in fig.3(a ) .
@xmath117 can reach maximal value @xmath123 at c.m.s energy near @xmath124 and decreases smoothly with the increment of @xmath113 .
the corresponding relative correction is depicted in fig.4(b ) in the same parameter space , the correction varies gently from @xmath125 to @xmath126 .
electroweak corrected cross sections as well as the corresponding relative corrections @xmath111 as the functions of the c.m.s energy @xmath113 with @xmath118 and related mssm parameters being taken from the @xmath2 scenario except higgs sector for process @xmath109 .
( a ) cross section versus @xmath113 , ( b ) @xmath111 versus @xmath113.,title="fig : " ] electroweak corrected cross sections as well as the corresponding relative corrections @xmath111 as the functions of the c.m.s energy @xmath113 with @xmath118 and related mssm parameters being taken from the @xmath2 scenario except higgs sector for process @xmath109 . ( a ) cross section versus @xmath113 , ( b ) @xmath111 versus @xmath113.,title="fig : " ] in fig.5(a - b ) we depict the born and ew corrected cross sections and corresponding relative corrections as the functions of mass @xmath96 with @xmath127 and the relevant mssm parameters adopting from the @xmath2 point except higgs sector for process @xmath109 .
the other higgs boson masses are determined in the mssm by using formcalc program with @xmath96 and os @xmath122 being its input parameters .
the values of @xmath119 and @xmath128 corresponding to different @xmath96 values are also scaled on x - axes in fig.5(a - b ) . in fig.5(a ) , we see there is an abrupt change on each curves due to the resonance effect of @xmath129 at the position around @xmath130(where @xmath131 and @xmath132 ) . after reaching its maximal value at position of @xmath133 ,
the born electroweak corrected cross sections decrease gently .
the dependence of relative correction on @xmath96(or @xmath119 , @xmath128 ) is displayed in fig.5(b ) .
@xmath111 curve bends downward in the vicinity of @xmath134 because of resonance effect , and then decreases as @xmath96 goes up .
the relative correction varies from positive to negative value region and the absolute value exceeds @xmath135 for both light and heavy higgs bosons .
electroweak corrected cross sections as well as the corresponding ew relative corrections @xmath111 as the functions of @xmath96 with @xmath127 and relevant mssm parameters taken from @xmath2 point except higgs sector being obtained by running formcalc program with input parameters : @xmath136 and os @xmath99 , for process @xmath109 .
( a ) cross section versus @xmath96 , ( b ) relative correction versus @xmath96 .
, title="fig : " ] electroweak corrected cross sections as well as the corresponding ew relative corrections @xmath111 as the functions of @xmath96 with @xmath127 and relevant mssm parameters taken from @xmath2 point except higgs sector being obtained by running formcalc program with input parameters : @xmath136 and os @xmath99 , for process @xmath109 .
( a ) cross section versus @xmath96 , ( b ) relative correction versus @xmath96 . , title="fig : " ] we depict the born and full one - loop electroweak corrected cross sections as the functions of c.m.s energy @xmath113 with the mssm parameters at the @xmath2 point in fig.6(a ) and fig.6(b ) for process @xmath137 and @xmath138 , separately . in the @xmath2 scenario , the value of heavier neutral and charged higgs boson masses(@xmath136 , @xmath96 and @xmath128 )
have the values beyond @xmath139 , accordingly the threshold energy of this parameter choice can reach @xmath140 . therefore , in fig6(a - b ) we plot the c.m.s .
energy @xmath113 in the range between @xmath141 and @xmath142 .
we see both of born and @xmath46 corrected cross sections increase with the increment of the colliding energy .
fig.6(c ) and fig.6(d ) show the dependence of the relative corrections @xmath111 of processes @xmath143 and @xmath138 on the c.m.s energy @xmath113 respectively , with the same mssm parameters as taken in fig.6(a - b ) .
there we see that when the colliding energy @xmath113 goes up , both curves in fig.6(c - d ) for relative correction @xmath111 increase rapidly in the vicinity of threshold energy , and then becomes increasing smoothly with @xmath111 value being in the range of @xmath144 $ ] when c.m.s energy is beyond @xmath145 . electroweak corrected cross sections as well as
the corresponding relative corrections @xmath111 as the functions of the c.m.s energy @xmath113 in the @xmath2 scenario for process @xmath146 .
( a ) cross section versus @xmath113 for @xmath147 , ( b ) cross section versus @xmath113 for @xmath148 , ( c ) @xmath111 versus @xmath113 for @xmath147 , ( d)@xmath111 versus @xmath113 for @xmath148 .
, title="fig : " ] electroweak corrected cross sections as well as the corresponding relative corrections @xmath111 as the functions of the c.m.s energy @xmath113 in the @xmath2 scenario for process @xmath146 . ( a ) cross section versus @xmath113 for @xmath147 , ( b ) cross section versus @xmath113 for @xmath148 , ( c ) @xmath111 versus @xmath113 for @xmath147 , ( d)@xmath111 versus @xmath113 for @xmath148 .
, title="fig : " ] electroweak corrected cross sections as well as the corresponding relative corrections @xmath111 as the functions of the c.m.s energy @xmath113 in the @xmath2 scenario for process @xmath146 .
( a ) cross section versus @xmath113 for @xmath147 , ( b ) cross section versus @xmath113 for @xmath148 , ( c ) @xmath111 versus @xmath113 for @xmath147 , ( d)@xmath111 versus @xmath113 for @xmath148 .
, title="fig : " ] electroweak corrected cross sections as well as the corresponding relative corrections @xmath111 as the functions of the c.m.s energy @xmath113 in the @xmath2 scenario for process @xmath146 .
( a ) cross section versus @xmath113 for @xmath147 , ( b ) cross section versus @xmath113 for @xmath148 , ( c ) @xmath111 versus @xmath113 for @xmath147 , ( d)@xmath111 versus @xmath113 for @xmath148 .
, title="fig : " ] the born and full @xmath46 corrected cross sections as the functions of c.m.s energy @xmath113 with @xmath149 and relevant @xmath2 mssm parameters except higgs sector for processes @xmath147 and @xmath150 are plotted in fig.7(a ) and fig.7(b ) , separately .
again , the values of other higgs boson masses are obtained from running formcalc program with input parameter values of @xmath149 and os @xmath122 .
both born cross section and corrected section increase gently when @xmath113 goes beyond the threshold , while when @xmath113 approaches to @xmath145 , the @xmath46 ew corrected cross section can achieve the maximal value of about @xmath151 .
the corresponding ew relative corrections with the same conditions as in fig.7(a ) and fig.7(b ) , are described in fig.7(c ) and ( d ) , respectively . as we can see in these figures , the line - shapes of born and ew corrected cross sections(relative correction ) for @xmath147 process are very similar with the corresponding ones for @xmath148 process .
electroweak corrected cross sections as well as the corresponding relative corrections @xmath111 as the functions of the c.m.s energy @xmath113 with @xmath118 and relevant @xmath152 mssm parameters except higgs sector for processes @xmath153.(a ) cross section versus @xmath113 for @xmath147 , ( b ) cross section versus @xmath113 for @xmath148 , ( c ) @xmath111 versus @xmath113 for @xmath147 , ( d)@xmath111 versus @xmath113 for @xmath148 .
, title="fig : " ] electroweak corrected cross sections as well as the corresponding relative corrections @xmath111 as the functions of the c.m.s energy @xmath113 with @xmath118 and relevant @xmath152 mssm parameters except higgs sector for processes @xmath153.(a ) cross section versus @xmath113 for @xmath147 , ( b ) cross section versus @xmath113 for @xmath148 , ( c ) @xmath111 versus @xmath113 for @xmath147 , ( d)@xmath111 versus @xmath113 for @xmath148 . , title="fig : " ] electroweak corrected cross sections as well as the corresponding relative corrections @xmath111 as the functions of the c.m.s energy @xmath113 with @xmath118 and relevant @xmath152 mssm parameters except higgs sector for processes @xmath153.(a ) cross section versus @xmath113 for @xmath147 , ( b ) cross section versus @xmath113 for @xmath148 , ( c ) @xmath111 versus @xmath113 for @xmath147 , ( d)@xmath111 versus @xmath113 for @xmath148 . , title="fig : " ] electroweak corrected cross sections as well as the corresponding relative corrections @xmath111 as the functions of the c.m.s energy @xmath113 with @xmath118 and relevant @xmath152 mssm parameters except higgs sector for processes @xmath153.(a ) cross section versus @xmath113 for @xmath147 , ( b ) cross section versus @xmath113 for @xmath148 , ( c ) @xmath111 versus @xmath113 for @xmath147 , ( d)@xmath111 versus @xmath113 for @xmath148 .
, title="fig : " ] the born and ew corrected cross sections as the functions of input parameter @xmath96 are drawn in fig.8(a ) for process @xmath154 , and the corresponding relative corrections versus @xmath96 are depicted in fig.8(b ) , with c.m.s energy @xmath155 and the relative mssm parameters taken from @xmath2 point except the higgs sector being fixed by using formcalc program . in fig.8(a ) the one - loop electroweak corrected cross section decreases from @xmath156 to @xmath157 with the increment of @xmath158 from @xmath159 to @xmath160 for process @xmath137 .
fig.8(b ) shows the relative correction for process @xmath137 goes from @xmath161 to @xmath162 when @xmath158 goes up from @xmath159 to @xmath160 . for channel @xmath163 , the cross section and ew relative correction versus @xmath96
are depicted in fig.8(c , d ) respectively , with @xmath113 being @xmath164 and other mssm parameters taken from @xmath2 point except higgs sector being determined by using formcalc package .
we can read from fig.8(c - d ) that the ew corrected cross section goes down from @xmath156 to @xmath165 and the relative correction varies from @xmath166 to @xmath167 , when input value of @xmath96 increases from @xmath168 to @xmath169 . in these four figures of fig.8
, we can see the behavior of these two processes @xmath170 are very similar except tiny different in the numerical value .
electroweak corrected cross sections as well as the corresponding relative corrections @xmath111 as the functions of @xmath96 with c.m.s energy @xmath155 and relevant mssm parameters in @xmath2 except higgs sector , for processes @xmath153 .
( a ) cross section versus @xmath96 for @xmath147 , ( b ) @xmath111 versus @xmath96 for @xmath147 , ( c ) cross section versus @xmath96 for @xmath148 , ( d)@xmath111 versus @xmath96 for @xmath148.,title="fig : " ] electroweak corrected cross sections as well as the corresponding relative corrections @xmath111 as the functions of @xmath96 with c.m.s energy @xmath155 and relevant mssm parameters in @xmath2 except higgs sector , for processes @xmath153 .
( a ) cross section versus @xmath96 for @xmath147 , ( b ) @xmath111 versus @xmath96 for @xmath147 , ( c ) cross section versus @xmath96 for @xmath148 , ( d)@xmath111 versus @xmath96 for @xmath148.,title="fig : " ] electroweak corrected cross sections as well as the corresponding relative corrections @xmath111 as the functions of @xmath96 with c.m.s energy @xmath155 and relevant mssm parameters in @xmath2 except higgs sector , for processes @xmath153 .
( a ) cross section versus @xmath96 for @xmath147 , ( b ) @xmath111 versus @xmath96 for @xmath147 , ( c ) cross section versus @xmath96 for @xmath148 , ( d)@xmath111 versus @xmath96 for @xmath148.,title="fig : " ] electroweak corrected cross sections as well as the corresponding relative corrections @xmath111 as the functions of @xmath96 with c.m.s energy @xmath155 and relevant mssm parameters in @xmath2 except higgs sector , for processes @xmath153 .
( a ) cross section versus @xmath96 for @xmath147 , ( b ) @xmath111 versus @xmath96 for @xmath147 , ( c ) cross section versus @xmath96 for @xmath148 , ( d)@xmath111 versus @xmath96 for @xmath148.,title="fig : " ] from above numerical results we can see that the experimental statistic errors for processes @xmath171 , @xmath172 and @xmath138 can reach @xmath173 , @xmath174 and @xmath174 , respectively .
if we assume @xmath175 , @xmath176 , and do not consider the backgrounds to the @xmath177 signals and the systematic errors of the experiment(due to lacking their studies at ilc ) , the ew corrections to the cross sections of these three processes should be taken into account at a linear collider with realistic accuracy better than @xmath178 .
in this paper , we present the calculation of the full @xmath84 electroweak corrections to the processes @xmath179 at a lc in the mssm . in the numerical calculation , we generally adopt the mssm parameters at the reference point @xmath2 defined in the spa project . for channel @xmath3 , we present the details for treating the charged higgs boson resonance in the calculation up to one - loop level .
our numerical results show that the ew relative correction for the process @xmath180 can be either positive or negative with the increment of higgs boson mass @xmath96 , and the relative correction varies in the range of @xmath181 $ ] as the @xmath96 goes from @xmath168 to @xmath169 . for channels @xmath5 ,
the ew correction generally reduce the born cross sections and the relative correction is typically of order @xmath6 .
we conclude that the complete @xmath46 electroweak corrections to the process @xmath182 are generally significant and can not be neglected in the precise experiment analysis .
99 s. l. glashow , nucl .
* b 22 * ( 1961 ) 579 ; s. weinberg , phys .
rev . lett .
* 19 * ( 1967 ) 1264 ; a. salam , in _ proceedings of the 8th nobel symposium _ ,
stockholm , 1968 , edited by n. svartholm ( almquist and wiksells , stockholm , 1968 ) , p.367 ; h. d. politzer , phys .
* 14 * ( 1974 ) 129 .
p. w. higgs , phys .
lett * 12 * ( 1964 ) 132 , phys .
* 13 * ( 1964 ) 508 ; phys .
* 145 * ( 1966 ) 1156 ; f. englert and r. brout , phys .
* 13 * ( 1964 ) 321 ; g. s. guralnik , c. r. hagen , and t. w. b. kibble , _
ibid_. * 13 * ( 1964 ) 585 ; t. w. b. kibble , phys . rev . * 155 * ( 1967 ) 1554 .
aguilar - saavedra _ et al .
_ [ ecfa / desy lc physics working group collaboration ] , arxiv : hep - ph/0106315 ; t. abe _ et al .
_ , [ american linear collider working group collaboration ] , arxiv : hep - ex/0106056 ; k. abe _ et al .
_ [ acfa linear collider working group collaboration ] , arxiv : hep - ph/0109166 .
s. komamiya , phys . rev .
* d38 * ( 1998 ) 2158 ; m.beccaria , f.m.renard , s.trimarchi , and c.verzegnassi , phys . rev .
* d68 * ( 2003 ) 035014 ; m.beccaria , h.eberl , f.m.renard , and c.verzegnassi , phys . rev .
* d69 * ( 2004 ) 091301(r ) .
p. janot , orsay report no .
lal 91 - 61 , 1991 ( unpublished ) ; j - f grivas , orsay report no .
lal 91 - 63 , 1991 ( unpublished ) ; a. brignole , j. ellis , j. f. gunion , m. guzzo , f. olnes , g. ridolfi , l. roszkowski and f. zwirner , @xmath188 : @xmath189 , munich , annecy , hamburg workshop , desy 92 - 123a , desy 92- 123b , desy 93 - 123c , ed .
p. zerwas , p. 613 ; a. djouadi , j. kalinowski , p. m. zerwas , _ ibid .
_ ; p.83 and z. phys . *
c57 * , ( 1993 ) 56 ; p.janot , @xmath190
@xmath191 @xmath192 @xmath193 @xmath194 , eds . f. harris , s. olsen , s. pakvasa and x. tata , waikoloa , hi ( 1993 ) , world scientific publishing , p. 192 ; @xmath195 @xmath196 @xmath197 @xmath198 @xmath199 , bnl 52 - 502 , fnal - pub -96/112 , lbnl - pub-5425 , slac report 485 , ucrl - jd-124160 ; j. f. gunion , @xmath200 @xmath201 @xmath202 , arxiv : hep - ph/9705282 , ucd-97 - 13,1997 . e. accomando , _ et al . , _ ( ecfa / desy lc physics working group ) @xmath203 @xmath204 @xmath205 ,
report no .
desy 97 - 100 , phys .
( 1998 ) 1 - 78 ; a. djouadi , talk at @xmath206 , report no .
lpt-04 - 133 , czech .
j. phys . 55
( 2005 ) b23-b44 ; a. djouadi , report no .
lpt - orsay-05 - 18 ( unpublished ) .
j. guasch , w. hollik , j. sola , j. high energy phys .
10 ( 2002 ) 040 , arxiv : hep - ph/0207364 ; w. hollik , e. kraus , m. roth , c. rupp , k. sibold and d. stoecklinger , nucl.phys .
b639 ( 2002 ) 3 - 65 , arxiv : hep - ph/0204350 .
b. c. allanach _ et al .
_ , in proc . of the aps / dpf / dpb summer study on the future of particle physics ( snowmass 2001 ) ed
n. graf , eur .
j. c 25 ( 2002 ) 113 [ econf c010630 ( 2001 ) p125 ] , arxiv : hep - ph/0202233 . | we study the complete one - loop electroweak(ew ) corrections to the processes of single charged higgs boson production associated with a neutral higgs boson@xmath0 and a gauge boson @xmath1 in the framework of the minimal supersymmetric standard model(mssm ) .
numerical results at the @xmath2 benchmark point as proposed in the spa project , are presented for demonstration .
we find that for the process @xmath3 the ew relative correction can be either positive or negative and in the range of @xmath4 in our chosen parameter space . while for the processes @xmath5 the corrections generally reduce the born cross sections and the ew relative corrections are typically of order @xmath6 .
= 0.3 in |
graphene is a two - dimensional honeycomb lattice of carbon atoms that has attracted considerable attention recently because of experimental progress@xcite that has raised hopes for applications in nanoelectronics and because of exotic chiral features@xcite in its electronic structure . in the absence of spin - orbit interactions ,
the energy bands of graphene are described at low energies by a two - dimensional dirac equation with linear dispersion centered on the hexagonal corners of the honeycomb lattice brillouin zone .
the recent advances in fabrication techniques have made it possible to produce graphitic systems with only a few layers or even a single monolayer of graphene@xcite .
one of the most remarkable properties of graphene is its half integer quantum hall effect , confirmed by recent experiments@xcite .
this electronic property follows directly from the system s dirac - like band structure@xcite . in a recent paper , kane and
mele@xcite showed that symmetry allowed spin - orbit interactions can generate an energy gap and convert graphene from a two dimensional zero gap semiconductor to an insulator with a quantized spin hall effect@xcite .
the quantized spin hall conductivity can be zero or nonzero , depending on the relative strength of intrinsic and rashba spin - orbit interactions .
the temperature at which the spin hall effect can be observed , and the sample quality requirements for its occurrence , depend on the absolute magnitude of these two spin - orbit interaction terms in the band structure .
( kane and mele@xcite argued on the basis of rough estimates of the spin - orbit interaction scale , that the quantum spin hall effect in graphene should be observable at relatively accessible temperatures of the order of 1@xmath3 . )
motivated by the fundamental interest associated with the spin hall effect and spin - orbit interactions in graphene , we have attempted to estimate , on the basis of microscopic considerations , the strength of both interactions . in order to allow for a rashba interaction ,
we account for the presence of an external gate electric field @xmath1 of the type used experimentally in graphene to move the fermi energy away from the dirac point .
( importantly this electric field explicitly removes inversion through the graphene plane from the symmetry operations of the system . ) then , starting from a microscopic tight - binding model with atomic spin - orbit interactions of strength @xmath0 , we use perturbation theory to derive expressions for the spin - orbit coupling terms that appear in the low - energy hamiltonian . at
leading order in @xmath0 only the rashba spin - orbit interaction term ( @xmath4 ) appears .
the intrinsic ( @xmath2 ) spin - orbit coupling has a leading contribution proportional to @xmath5 .
both terms have the form proposed by kane and mele@xcite on the basis of symmetry considerations . according to our theory
the respective coupling constants are given by the following expressions : @xmath6 and @xmath7 where @xmath8 and @xmath9 are tight - binding model parameters explained more fully below , @xmath1 is a perpendicular external electric field , and @xmath10 is proportional to the atomic size of carbon .
the coupling constants @xmath11 and @xmath12 have numerical values @xmath13 times smaller and @xmath13 times larger , respectively , than the estimates of kane and mele@xcite with @xmath14 at the largest reasonable values of @xmath1 . together , these estimates suggest that the quantum spin hall effect will be observable in ideal samples only at temperatures below @xmath15 in a zero - field limit .
our paper is organized as follows . in section [ sec : tight_binding ] we briefly summarize the tight - binding model used to represent graphene in this paper .
section [ sec : perturbation ] describes some details of the perturbation theory calculation . in section [ sec : abinitio ] we discuss _ ab initio _ density functional theory calculations we have carried out as a partial check on the tight - binding model and on the atomic approximation for spin - orbit interactions used in the perturbation theory calculations .
we conclude in section [ sec : summary ] with a brief summary and present our conclusions .
for our analytic perturbation theory calculations we choose the simplest possible tight - binding model with carbon @xmath16 and @xmath17 orbitals , a two - center slater - koster approximation@xcite for nearest - neighbor hopping , and orthogonality between wannier functions centered on different sites assumed .
this gives a tight - binding hamiltonian of the form @xmath18 where @xmath19 label the four orbitals on each site , @xmath20 and @xmath21 represent the two distinct sites in the honeycomb lattice unit cell , and @xmath22 is one of the three vectors connecting a lattice site and its near neighbors .
we choose a coordinate system in which the honeycomb s bravais lattice has primitive vectors @xmath23 where @xmath24 is the lattice constant of graphene .
the corresponding reciprocal lattice vectors are @xmath25 and the near - neighbor translation vectors are : @xmath26 the site - diagonal matrix elements @xmath27 are the atomic energies of @xmath16 and @xmath17 orbitals , with the latter chosen as the zero of energy . in table
[ tab : two_center ] we reproduce for completeness the relationship between the required nearest - neighbor hopping matrix elements @xmath28 and the four independent slater - koster parameters @xmath29 , @xmath9 , @xmath30 , and @xmath31 whose numerical values specify this model quantitatively . if the graphene lattice is placed in the @xmath32 plane , @xmath33 for hops on the graphene lattice and the atomic @xmath34 orbitals decouple from other orbitals .
this property is more general than our model , since it follows from the graphene plane inversion symmetry that orbitals which are even and odd under this symmetry operation will not be coupled , and is key to the way in which weak spin - orbit interactions influence the low - energy bands .
.two - center matrix elements for hoping between @xmath16 and @xmath17 orbitals along a direction specified by the unit vector @xmath35 . [ cols="^,^,^,^ " , ] these values imply a graphene energy gap at @xmath36 equal to @xmath37 where we used the nonorthogonal tight - binding parameters neglecting the overlap for simple estimations .
our estimates of @xmath12 are discussed later .
we have performed realistic _
ab initio _ electronic structure calculations@xcite for inversion symmetric ( @xmath38 ) graphene sheets using the projector augmented wave ( paw)@xcite method with a perdew - burke - ernzerhof ( pbe ) generalized gradient approximation ( gga)@xcite density functional in order to partly test the quantitative accuracy of the conclusions reached here about spin - orbit interaction gaps based on a simplified electronic structure model .
the calculations were performed using vasp ( vienna _ ab initio _ simulation package)@xcite . in vasp ,
spin - orbit interactions are implemented in the paw method which is based on a transformation that maps all electron wave functions to smooth pseudo wave functions .
all physical properties are evaluated using pseudo wave functions .
the spin - orbit interaction is evaluated taking into account only the spherical part of the potential inside muffin tins surrounding the carbon nuclei : @xmath39 in order to make the gaps induced by spin - orbit interaction exceed the accuracy of vasp eigenvalues , we have artificially increased the strength of @xmath40 by up to 300 times by decreasing the speed of light @xmath41 .
figure [ fig : band_so ] shows the tight - binding band structure of graphene for @xmath42 and @xmath43 , where @xmath44 mev .
the spin - orbit gap is not large on the scale of the full band width , even when enlarged by a factor of 300 .
figure [ fig : gap_so ] compares the _ ab initio _ calculation and tight - binding model low - energy gaps at the hexagonal brillouin - zone corners for @xmath36 , finding close agreement .
both approximations find a gap that grows as the second power of the spin - orbit coupling strength .
the close agreement is perhaps not surprising given that vasp also makes an atomiclike approximation for the spin - orbit coupling strength . in our opinion , however , the neglected contributions from interstitial regions and from aspherical potentials inside the muffin - tin sphere are small and their contributions to energy levels tends toward even smaller values due to spatial averaging by the bloch wave functions .
we believe that these calculations demonstrate that the tight - binding model spin - orbit gap estimates are accurate .
the intrinsic and rashba spin - orbit interactions arise from mixing between @xmath45 and @xmath46 bands due to atomic spin - orbit interactions alone in the case of @xmath11 ( eq.([eq : intrinsic ] ) ) and due to a combination of atomic spin - orbit and stark interactions in the case of @xmath47 ( eq.([eq : rashba ] ) ) .
these expressions for @xmath11 and @xmath47 follow directly from eq.([eq : h_second ] ) and from the eigenvectors and eigenenergies listed in table [ tab : eigenvector_k ] .
( the energetic ordering in table [ tab : eigenvector_k ] applies for @xmath48 which holds for the tight - binding parameters in table [ tab : hopping ] . )
the pure @xmath49 hybridized bonding and antibonding states ( energies @xmath50 in table [ tab : eigenvector_k ] ) are symmetrically spaced with respect to the undoped fermi level and do not make a net contribution to either @xmath47 or @xmath11 .
the @xmath51 hybridized bonding states ( energy @xmath52 in table [ tab : eigenvector_k ] ) , on the other hand , are further from the fermi energy than the corresponding antibonding states ( energy @xmath53 in table [ tab : eigenvector_k ] ) because of the difference between atomic @xmath16 and @xmath17 energies .
their net contribution to @xmath11 is proportional to @xmath16 and inversely related to @xmath9 , which sets the scale of the energy denominators .
similar considerations explain the expression for @xmath47 which is proportional to @xmath0 and @xmath54 and inversely proportional to @xmath55 .
@xmath47 vanishes at @xmath2 because of the inversion symmetry of an isolated graphene plane .
the numerical value of the rashba interaction parameter @xmath12 obviously depends on the electric field perpendicular to the graphene plane which varies as the carrier density is modulated by a gate voltage .
a typical value can be crudely estimated from eq .
( [ eq : rashba ] ) , by assuming a typical electric field @xmath56@xcite , and using the value @xmath57 obtained by scaling the hydrogenic orbital stark matrix element by the ratio of the atomic radii@xcite of carbon and hydrogen : @xmath58 the value of @xmath12 is influenced by screening of the electric field at one graphene atom by the polarization of other graphene atoms and by dielectric screening in the substrate and cap layers , but these correction factors are expected to be @xmath59 . note that our estimate for @xmath11 is 100 times smaller than kane s estimate,@xmath60 , whereas @xmath12 is 100 times larger than kane s estimate , @xmath61 . if our estimates are accurate , @xmath14 at large gate voltages . for undoped samples , however , the requirement for a quantized spin hall effect gap@xcite , @xmath62 , should still be achievable if accidental doping in the substrate and cap layer can be limited .
when @xmath11 is smaller than @xmath12 , the energy gap closes and graphene becomes a zero gap semiconductor with quadratic dispersion@xcite .
our estimates suggest that the quantum spin hall effect in graphene should occur only below @xmath63 , a temperature that is still accessible experimentally but not as convenient as @xmath64 .
in addition , it seems likely that disorder will dominate over the spin - orbit couplings in current samples , so further progress in increasing the mobility of graphene sheets may also be necessary before the quantum spin hall effect can be realized experimentally .
we emphasize , however@xcite , that the spin hall effect survives , albeit with a reduced magnitude , even when the spin - orbit gap is closed by disorder . in summary , we have derived analytic expressions for the intrinsic and rashba spin - orbit interaction coupling constants that appear in the low - energy hamiltonian of a graphene sheet under a perpendicular external electric field .
the rashba interaction parameter is first order in the atomic carbon spin - orbit coupling strength @xmath0 and the perpendicular external electric field @xmath1 , whereas the intrinsic spin - orbit interaction is second order in @xmath0 and independent of @xmath1 .
the estimated energy gap for @xmath2 is of the order of 0.01@xmath3 and agrees with realistic _ ab initio _ electronic structure calculations .
this work was supported by the welch foundation ( houston , tx ) under grant no .
f-1473 and f-0934 , by the texas advanced computing center ( tacc ) , university of texas at austin , by seagate corporation , and by the department of energy under grant no . de - fg03 - 96er45598 .
the authors gratefully acknowledge helpful conversations with a.k .
geim , c.l .
kane , and p. kim .
999 k.s.novoselov and a.k.geim and s.v.morozov and d.jiang and y.zhang and s.v.dubonos and i.v.grigorieva and a.a.firsov , science * 306 * , 666 ( 2004 ) .
k.s.novoselov and d.jiang and f.schedin and t.j.booth and v.v.khotkevich and s.v.morozov and a.k.geim , proc .
natl acad .
usa * 102 * , 10451 ( 2005 ) . y.zhang and joshua p.small and michael e.s.amori and philip kim , phys .
* 94 * , 176803 ( 2005 ) .
c.berger and z.song and t.li and x.li and a.y.ogbazghi and r.feng and z.dai and a.n.marchenkov and e.h.conrad and p.n.first and w.a.de heer , j. phys .
chem b * 108 * , 19912 ( 2004 ) .
c.l.kane and e.j.mele , phys .
lett . * 95 * , 226801 ( 2005 ) .
c.l.kane and e.j .
mele , phys .
lett . * 95 * , 146802 ( 2005 ) .
semenoff , phys .
lett . * 53 * , 2449 ( 1984 ) .
f.d.m.haldane , phys .
rev . lett .
* 61 * , 2015 ( 1988 ) .
v.p.gusynin and s.g .
sharapov , phys .
. lett . * 95 * , 146801 ( 2005 ) .
n.a.sinitsyn , j.e.hill and hongki min and jairo sinova and a.h.macdonald , phys .
lett . * 97 * , 106804 ( 2006 ) .
r.saito and g.dresselhaus and m.s.dresselhaus , _ physical properties of carbon nanotubes _
( imperial college press , london,1998 ) .
mark wilson , phys . today
* 59*(1 ) , 21 ( 2006 ) . k.s.novoselov and a.k.geim and s.v.morozov and d.jiang and m.i.katsnelson and i.v.grigorieva and s.v.dubonos and a.a.firsov , nature * 438 * , 197 ( 2005 ) . y.zhang and yan - wen tan and horst l.stormer and philip kim , nature * 438 * , 201 ( 2005 ) .
j.c.slater and g.f.koster , phys . rev . *
94 * , 1498 ( 1954 ) .
leonard i. schiff , _ quantum mechanics _ ( mcgraw - hill company , new york , 1968 ) . w.kohn and l.j.sham , phys
. rev . * 140 * , a1133 ( 1965 ) .
p.e.blchl , phys .
b * 50 * , 17953 ( 1994 ) . j.p.perdew and k.burke and m.ernzerhof , phys . rev
. lett . * 77 * , 3865 ( 1996 ) . g.kresse and j.furthmller , phys .
b * 54 * , 11169 ( 1996 ) .
r.saito and m.fujita and g.dresselhaus and m.s.dresselhaus , phys .
b * 46 * , 1804 ( 1992 ) . j.t.waber and don.t.cromer , j. chem . phys . * 42 * , 4116 ( 1965 ) .
yugui yao , fei ye , xiao - liang qi , shou - cheng zhang and zhong fang , cond - mat/0606350 ; d. huertas - hernando , f. guinea and a. brataas , cond - mat/060658 , phys . rev .
b ( to be published ) . | starting from a microscopic tight - binding model and using second order perturbation theory , we derive explicit expressions for the intrinsic and rashba spin - orbit interaction induced gaps in the dirac - like low - energy band structure of an isolated graphene sheet .
the rashba interaction parameter is first order in the atomic carbon spin - orbit coupling strength @xmath0 and first order in the external electric field @xmath1 perpendicular to the graphene plane , whereas the intrinsic spin - orbit interaction which survives at @xmath2 is second order in @xmath0 .
the spin - orbit terms in the low - energy effective hamiltonian have the form proposed recently by kane and mele . _
ab initio _ electronic structure calculations were performed as a partial check on the validity of the tight - binding model . |
here we provide parameters of the single - qubit clifford gates for @xmath1-supcode , which is obtained from solving three coupled nonlinear equations ( involving @xmath32 , @xmath33 , and @xmath34 respectively ) relevant to the charge noise only .
note that @xmath13-rotations @xmath69 , @xmath70 , and @xmath71 do not need to turn on @xmath18 at all ( therefore having no @xmath1 error ) and when charge noise is the only error source , the nave pulses suffice .
.parameters of the @xmath1-supcode sequence , eq . , appropriate for the identity operation and the hadamard gate .
strengths of the exchange interactions are given in units of @xmath15 . [ cols="^,^,^,^,^,^,^",options="header " , ] 42ifxundefined [ 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 * * , ( ) link:\doibase 10.1038/nature10707 [ * * , ( ) ] @noop * * , ( ) @noop * * , ( ) @noop ( ) link:\doibase 10.1103/physrevlett.89.147902 [ * * , ( ) ] @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) link:\doibase 10.1103/physreva.88.052326 [ * * , ( ) ] @noop ( ) @noop * * , ( ) link:\doibase 10.1103/physreva.89.022310 [ * * , ( ) ] @noop * * , ( ) @noop * * , ( ) link:\doibase 10.1103/physreva.85.042311 [ * * , ( ) ] link:\doibase 10.1103/physrevx.4.021044 [ * * , ( ) ] @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) link:\doibase 10.1103/physreva.77.012307 [ * * , ( ) ] @noop _ _ ( , ) | we study the performance of supcode a family of dynamically correcting pulses designed to cancel simultaneously both overhauser and charge noise for singlet - triplet spin qubits adapted to silicon devices with electrostatic control .
we consider both natural si and isotope - enriched si systems , and in each case we investigate the behavior of individual gates under static noise and perform randomized benchmarking to obtain the average gate error under realistic @xmath0 noise .
we find that in most cases supcode pulses offer roughly an order of magnitude reduction in gate error , and especially in the case of isotope - enriched si , supcode yields gate operations of very high fidelity .
we also develop a version of supcode that cancels the charge noise only , `` @xmath1-supcode '' , which is particularly beneficial for isotope - enriched si devices where charge noise dominates overhauser noise , offering a level of error reduction comparable to the original supcode while yielding gate times that are @xmath2 to @xmath3 shorter .
our results show that the supcode noise - compensating pulses provide a fast , simple , and effective approach to error suppression , bringing gate errors well below the quantum error correction threshold in principle . ever since kane s original proposal for a silicon - based quantum computer,@xcite si has been a prime candidate as a host material for solid - state quantum computing.@xcite a decisive advantage of si
is that it is available in an isotope - enriched form ( @xmath4si ) that has zero nuclear spin .
this allows for the nearly complete removal of decoherence due to the hyperfine interaction between the qubit and surrounding nuclear spins ( overhauser noise ) , leading to remarkably long coherence times and high control fidelity.@xcite recent years have witnessed substantial experimental progress in the fabrication , initialization , readout and control of spins in both phosphorous donors and gate - defined quantum dots in si systems.@xcite for a p donor qubit on si , both the donor electron spin@xcite and the nuclear spin@xcite have been explored as potential qubits , exhibiting coherence times up to 30 seconds and control fidelities exceeding 99.99% in recent experiments operating the qubit as a quantum memory,@xcite both of which are the highest achieved for any solid state qubit . in laterally defined si quantum dot systems , coherent operation of a singlet - triplet spin qubit@xcite has also been demonstrated.@xcite while si qubits have been shown repeatedly to possess long information storage ( i.e. quantum memory ) times , it remains an open question on how to extend this long - lived coherence to qubits undergoing quantum gate operations .
experiments have progressed to such a stage where high fidelity quantum gates are now within reach .
while quantum devices built on isotope - enriched si already enjoy many advantages , some noise remains .
in particular , the residual @xmath5si impurities distort the wave function of the donor qubit or the lateral confinement potential of the quantum dot , which either broadens the esr resonance in the former case@xcite or shifts the exchange interaction in the latter,@xcite both causing inaccuracies in control . more importantly
, additional errors can arise through electrical fluctuations from either the leads or nearby charged impurities , and from variations in control parameters.@xcite while a vast literature on dynamically corrected gates ( dcgs ) has existed long before the conception of spin qubits , it was only recently realized that the special constraints imposed by solid state spin systems often require brand - new approaches to developing dcgs.@xcite while one of these approaches , known as supcode,@xcite has already been experimentally demonstrated for nitrogen - vacancy centers in diamond@xcite and is also applicable to single donor spins in si , it is particularly suitable for singlet - triplet qubits subject to overhauser and charge noise because it is designed to yield comparatively simple pulse sequences that cancel both while respecting all experimental constraints , including the restriction to positive , single - axis control . in the case of isotope - enriched si , however , canceling overhauser noise is no longer necessary , leaving open the possibility that robust quantum operations on si devices might be realized with even simpler and faster control sequences . in this paper
, we show that supcode pulses can substantially improve gate fidelities in silicon singlet - triplet spin qubits .
we begin by quantifying the extent to which charge noise corrupts the application of individual quantum gates and demonstrate the cancelation of these errors effected by supcode pulses .
we then provide a more systematic analysis of the performance of supcode by studying its effectiveness for long sequences of quantum gates subject to @xmath0 noise , which describes both overhauser and charge noise.@xcite through randomized benchmarking,@xcite we extract the average error per gate for the set of single - qubit clifford gates , and we find that supcode pulses remain robust against noise in this realistic situation .
we also develop a version of supcode that works particularly well for isotope - enriched si while being about 30% to 50% faster , and we demonstrate that it offers a level of error reduction comparable to the original version .
additionally , we discuss the applicability of supcode to @xmath0 noise with different exponents .
these results illustrate that si quantum devices can achieve long coherence times not only when the qubit is idle but even while it is undergoing operations .
we start with the model hamiltonian governing a singlet - triplet qubit,@xcite which can be realized in si either by gating exchange - coupled p donors@xcite or by a gate - defined double dot in a heterostructure:@xcite @xmath6}{2}\sigma_z.\ ] ] the computational bases are @xmath7 and @xmath8 , where @xmath9 with @xmath10 creating an electron with spin @xmath11 in the @xmath12th dot .
rotations around the @xmath13-axis are performed with a magnetic field gradient across the double - dot system , which in energy units reads @xmath14 .
this magnetic field gradient can be generated by a micromagnet , as demonstrated in si / sige quantum dot systems,@xcite where @xmath15 is approximately @xmath16 nev .
rotations around the @xmath17-axis are done using the exchange interaction @xmath18 , the energy level splitting between @xmath19 and @xmath20 .
the magnitude of @xmath18 is controlled by the detuning @xmath21 , namely the tilt of the double - well confinement potential , through the gate voltages .
experimentally , @xmath18 can be tuned from close to zero up to tens of @xmath22ev.@xcite one therefore has fast , all - electrical control over the rotation rate around the @xmath17-axis . on the other hand , the magnetic field gradient @xmath15
can not be changed efficiently during a gate operation , and we will take it to be constant throughout the computation . in general , decoherence can arise through both @xmath15 and @xmath18 .
fluctuations arising from overhauser noise add a small , but unknown error term @xmath23 to the hamiltonian : @xmath24 . for natual si , @xmath23 is on the scale of nev , @xcite corresponding to a few percent of @xmath15 . however , for isotope - enriched silicon , @xmath23 can be reduced by up to three orders of magnitude,@xcite down to @xmath25 . on the other hand , impurities on the substrate
lead to deformations of the confinement potential and consequently the energy level structure , thereby causing charge noise.@xcite this noise produces fluctuations in the exchange energy , @xmath1 , through fluctuations in the detuning @xmath21 . in this work ,
we assume a phenomenological form of @xmath26 , implying @xmath27,@xcite but our method applies to other forms also.@xcite estimates of @xmath28 vary , ranging from 10% ( @xmath29ev ) @xcite to @xmath30 ( cf .
ref . ) .
we study a range of @xmath31 values in this work .
we previously developed a family of composite pulses ( supcode ) that cancels both overhauser noise@xcite and charge noise.@xcite the basic idea is as follows : assuming the noise is quasi - static , one expands the evolution operator to first order in both @xmath23 and @xmath31 . in order to cancel these error terms ,
the original rotation is supplemented by an ( imperfect ) identity operation , carefully designed such that the error arising from performing this identity operation exactly cancels that of the target rotation .
this formalism reduces to a multi - dimensional optimization problem involving six coupled nonlinear equations ( corresponding to error terms for @xmath32 , @xmath33 and @xmath34 , each of which has @xmath23 and @xmath31 components ) , from which the pulse sequence can be obtained .
a detailed description of the theoretical framework is presented in ref . , and we shall refer to these sequences as `` full supcode '' sequences in this work because they cancel both overhauser noise and charge noise . while undoubtedly full supcode works for the problem at hand , further optimizations are possible for an isotope - enriched si system . because the overhauser noise is negligible in this case , one can completely ignore the @xmath23 component of the error , leading to a simplification of the problem since only three coupled nonlinear equations need to be solved , and the resulting sequences are much shorter and simpler . in fig .
[ fig1 ] we show pulse profiles comparing the full supcode and @xmath1-supcode for a hadamard gate . detailed sequences and pulse parameters for each of the single - qubit clifford gates are given in the appendix . in obtaining these results , we restricted the strength of the exchange to @xmath35 so that the resulting sequences are more feasible for experimental implementation .
in the appendix we give explicit numerical results of the gate durations of @xmath1-supcode , and also show that @xmath1-supcode is about 30% to 50% shorter than the full supcode .
we shall also show , in the remainder of this paper , that for parameter regimes of interest , the error reduction of @xmath1-supcode is comparable to that of full supcode for isotope - enriched si , making it particularly suitable for experimental realization in these systems . .
( a ) full supcode canceling both overhauser and charge noise .
( b ) @xmath1-supcode canceling charge noise only.,width=302 ] figure [ fig2 ] shows the infidelity ( cf . ref . ) of the hadamard gate versus quasi - static charge noise strength .
we model natural si as having @xmath36 , while isotope - enriched si has @xmath37 , consistent with experimental values.@xcite fig .
[ fig2](a ) shows the result for natural si .
while the curve for the nave pulse saturates at an infidelity of about @xmath38 , the full supcode has a residual error of only @xmath39 for charge noise of less than @xmath40 , and if the quantum error correction threshold is set at @xmath30 , the range of charge noise that can be tolerated is as large as 5% .
this is as expected because full supcode cancels both @xmath23 and @xmath31 errors .
turning to the curve for @xmath1-supcode , we see that for most of the @xmath31 range ( @xmath41 ) , its performance is worse than the nave pulse .
this is understandable because @xmath1-supcode completely ignores the overhauser noise , and this is precisely the range where overhauser noise is dominant .
figure [ fig2](b ) shows the case of isotope - enriched si . from the different scale of the @xmath42-axis
compared with fig .
[ fig2](a ) one immediately notices that the gate error is substantially reduced for all three curves as expected for negligible overhauser noise .
the exceedingly low errors achieved by the full supcode are consistent with the remarkably long coherence times measured in experiments.@xcite at a typical quantum error correction threshold of @xmath30 , the nave pulse tolerates charge noise up to @xmath40 , whereas the full supcode tolerates as much as @xmath43 . while this range does not differ from the case of natural si , the gate error for charge noise less than @xmath44 is substantially smaller due to the absence of overhauser noise .
note that for the isotope - enriched si , @xmath1-supcode has low error comparable with full supcode for a wide range of noise amplitudes . even in the regime
where it is not as effective as full supcode ( @xmath45 ) , the error is still greatly reduced compared to that of the nave pulse .
similar behavior is observed for all single - qubit clifford gates ( not shown here ) with the exception of @xmath13-rotations , where @xmath1-supcode has a constant error because charge noise does not come into play . for ( a ) @xmath36 relevant to natual si and ( b ) @xmath46 relevent to isotope - enriched si .
infidelity is shown for the nave pulse ( black solid ) , full supcode ( red dashed ) , and @xmath1-supcode ( blue dotted ) .
note the different scales of the @xmath42-axis.,title="fig:",width=302 ] for ( a ) @xmath36 relevant to natual si and ( b ) @xmath46 relevent to isotope - enriched si .
infidelity is shown for the nave pulse ( black solid ) , full supcode ( red dashed ) , and @xmath1-supcode ( blue dotted ) .
note the different scales of the @xmath42-axis.,title="fig:",width=302 ] while individual gate errors such as those shown in fig .
[ fig2 ] are strongly indicative of the improvements afforded by supcode , a more complete characterization of this improvement is obtained by studying the error over long sequences of gates through randomized benchmarking.@xcite we have so far assumed a static noise model , which is a good approximation for single gates,@xcite but a more realistic model taking into account temporal variations of the noise is needed for longer gate sequences . to include these effects ,
we employ a @xmath0 noise model in which both the overhauser and charge noise have a power spectrum of the form @xmath47 , where @xmath48 denotes overhauser @xmath49 and charge noise @xmath50 , and @xmath51 and @xmath52 denote the noise amplitudes .
we choose the exponent to be @xmath53 , which is consistent with the upper bound estimated from experiments.@xcite we have considered other values of @xmath54 , and the results will be presented later in the paper .
we implement randomized benchmarking by averaging the fidelity over random sequences of single - qubit clifford gates and over different noise realizations for a varying number of gates in the sequences .
the fidelity of a sequence behaves as @xmath55 $ ] , where @xmath56 is the average error per gate , and @xmath57 is the number of clifford gates applied.@xcite we fit our results to the exponentially decaying function @xmath58 , where the average error per gate is related to the fitted exponent via @xmath59 .
the realizations of @xmath0 noise are generated using a weighted sum of random telegraph signals.@xcite noise for charge noise with amplitude @xmath60 for ( a ) @xmath61 relevant to natual si and ( b ) @xmath62 relevant to isotope - enriched si . the nave pulse , full supcode and @xmath1-supcode are shown as black , red and blue lines respectively , and the thin black dot - dashed , red dashed and blue dotted lines are fits of the corresponding data to @xmath63 .
note the different scales of both the @xmath13 and @xmath42 axes.,title="fig:",width=302 ] noise for charge noise with amplitude @xmath60 for ( a ) @xmath61 relevant to natual si and ( b ) @xmath62 relevant to isotope - enriched si .
the nave pulse , full supcode and @xmath1-supcode are shown as black , red and blue lines respectively , and the thin black dot - dashed , red dashed and blue dotted lines are fits of the corresponding data to @xmath63 .
note the different scales of both the @xmath13 and @xmath42 axes.,title="fig:",width=302 ] for ( a ) @xmath61 relevant to natual si and ( b ) @xmath62 relevant to isotope - enriched si .
the gate error is shown for the nave pulses ( black solid ) , full supcode ( red dashed ) , and @xmath1-supcode ( blue dotted ) .
note the different scales of the @xmath42-axis.,title="fig:",width=302 ] for ( a ) @xmath61 relevant to natual si and ( b ) @xmath62 relevant to isotope - enriched si .
the gate error is shown for the nave pulses ( black solid ) , full supcode ( red dashed ) , and @xmath1-supcode ( blue dotted ) .
note the different scales of the @xmath42-axis.,title="fig:",width=302 ] figure [ fig3 ] shows the randomized benchmarking results for charge noise with @xmath64 ( note that one should not directly compare this result to that of constant noise with @xmath65 since the noise spectra are completely different ) .
fig .
[ fig3](a ) shows the result for @xmath61 , relevant to natural si .
these results are generated using the average of 200 independent realizations of noise / gate sequences , each of which contains up to 300 clifford gates .
all curves asymptote to fidelity @xmath66 as expected , but the nave pulse and @xmath1-supcode drops to @xmath66 much more steeply , reaching the asymptotic value after less than 20 gate operations .
full - supcode pulses , on the other hand , maintain the fidelity much longer , with 70% fidelity after 60 gate operations .
this clearly demonstrates that supcode pulses play a crucial role in error reduction .
the result is consistent with that of fig .
[ fig2 ] in that when nuclear noise is present , @xmath1-supcode does not work well , even worse than a nave pulse .
turning to fig .
[ fig3](b ) , the results for @xmath62 relevant to isotope - enriched si , we see a huge reduction in error . here
we average over 100 independent realizations of the noise and gate sequences and increase the number of gates in each sequence to 500 .
we see that even with the nave pulses , the fidelity remains larger than 80% after 400 gate operations .
most remarkably , when supcode pulses are used , the fidelity remains greater than 95% even after 500 gate operations .
this fact reinforces the main point of this paper : while isotope - enriched si is already superior to natural si , application of supcode reduces the error much further .
we see again that @xmath1-supcode has an error reduction ability comparable with the full supcode , so it is particularly useful for isotope - enriched si since it is typically shorter , less complicated , and works almost perfectly when charge noise plays a much more significant role than overhauser noise .
we have also performed randomized benchmarking for different magnitudes of the charge noise , and these results are summarized in fig .
[ fig4 ] , where we show the average error per gate , @xmath56 , obtained from the fitted decaying exponent .
we note from fig .
[ fig4](a ) that , for charge noise below 3% in natural si , the nave pulse has a gate error of approximately 10% .
while this is a large error , it can be reduced by roughly one order of magnitude by the application of full supcode .
@xmath1-supcode carries larger error than the naive pulse as expected .
fig .
[ fig4](b ) shows the result for isotope - enriched si .
the gate errors are again much lower than those of natural si : on a log - log plot the gate error of the nave pulses reduces linearly from around 1% for 3% charge noise , to @xmath67 for a charge noise of 0.1% . in this range , the gate error is further reduced by an order of magnitude when supcode pulses are used .
interestingly , the @xmath1-supcode reduces error more than the full supcode which presumably arises from the simplicity of those pulses . when nuclear noise is absent .
the result of full - supcode is shown as the red dashed line and that of @xmath1-supcode is shown as the blue dotted line .
the yellow area shows the regime where supcode reduces error.,width=302 ] to understand better the applicability of supcode for different 1/f noises we show in fig .
[ fig5 ] the results of the supcode improvement ratio as a function of charge noise exponent , @xmath54 , when nuclear noise is absent .
the improvement ratio is defined as the ratio between the average gate errors of nave and supcode pulses at asymptotically small charge noise . for a given exponent @xmath54 ,
this ratio is consistent for several different levels of small charge noise as indicated in the numerical calculation .
although the results shown in fig . [ fig4](b ) has a very small amount of nuclear noise , we can still see that the improvement ratio is approximately 10 and is consistent with the result of fig .
[ fig5 ] . for large @xmath54
, the noise is concentrated at low frequency which is precisely what supcode pulses are designed to correct , thus the improvement ratio increases exponentially as @xmath54 increases . on the other hand , for smaller @xmath54 the noises are more like white noise and performing supcode would not only be useless for noise correction , but also introduces more error due to the longer gate time which allows the error to accumulate more .
we therefore conclude that the supcode pulses are useful for @xmath68 and the improvement is much more pronounced if @xmath54 is further increased . in conclusion
, we studied the performance of supcode for both static noise and @xmath0 noise in si ( natural and isotope - enriched ) .
we found that supcode is useful for @xmath0 noise with exponent @xmath68 in which case noise is concentrated at low frequency . in that regime ,
application of supcode reduces the gate error in a substantial way .
we also demonstrated that @xmath1-supcode can perform comparably to the full supcode , so it is particularly useful for isotope - enriched si since it typically generates simpler and faster sequences .
we note that the same protocol applies to two - qubit gates for single - spin qubits as well.@xcite our results clearly show that the extremely long coherence times of si qubits pertain not only to information storage but also to information processing , if supcode is applied .
thus , si spin qubits may now be close to quantum error corrected fault tolerant quantum computing applications .
this work is supported by lps - cmtc and iarpa .
jpk acknowledges support from the umbc office of research through an srais award . |
in experiments on current induced switching of magnetization ( see e.g. ref . ) , current passing through a thick polarizing magnet ( pm ) becomes spin polarized .
the spin polarized current ( spin current ) then flows through a nonmagnetic layer ( the spacer layer ) and becomes partially or fully absorbed by a switching magnet ( sm ) .
the absorbed spin current exerts a spin - transfer torque on the switching magnet and this torque can be used to switch the direction of the magnetization of the switching magnet between the parallel ( p ) and antiparallel ( ap ) orientations relative to the magnetization of the polarizing magnet . in this traditional setup
the current is perpendicular to both the pm / spacer and spacer / sm interfaces .
this setup is referred to as current perpendicular to plane ( cpp ) geometry and is shown schematically in fig.1 .
the switching process relies on the scenario in which one of the configurations ( p or ap ) becomes unstable , at a critical current , the other configuration is stable and , therefore , available for switching into .
however , in the presence of an external magnetic field stronger than the coercive field of the switching magnet , it is found experimentally @xcite that , for current greater than a critical value and with the correct sense , neither the p nor the ap configuration is stable .
the magnetization of the switching magnet then precesses continually and becomes a source of microwave generation .
it was also proposed @xcite that microwave generation can occur even in the absence of an applied field provided the spin - transfer torque has both the in - plane and out - of plane components of appropriate relative sign .
both the switching and microwave generation scenarios have potentially very important applications .
however , to limit the current to acceptable values and to minimize the oersted fields generated by the current , experiments are performed on cpp nanopillars with a very small diameter of the order of 100 nm .
such nanopillars are difficult to prepare . moreover , to achieve a usable microwave power , large arrays of cpp nonopillars would have to be manufactured , and this is even more difficult to achieve .
we have , therefore , investigated theoretically two alternative geometries , shown in fig.2 , which may have interesting applications since they offer much more flexibility for design of current - induced switching and microwave generation devices . in the first geometry shown in fig.2a ,
the current is perpendicular to the pm / spacer interface but parallel to the spacer / sm interface ( cpip ) . in the second geometry shown in fig.2b
the current is parallel to both the pm / spacer and spacer / sm interfaces ( cip ) .
it is clear from fig.2 that switching magnets in the cpip and cip geometries are arrays of either magnetic dots or wires deposited on the surface of a nonmagnetic substrate .
it should be noted that our cpip geometry in which the current flows parallel to the switching magnet / nonmagnet interface is closely related to that used in the so called pure - spin - current - induced magnetization switching which was recently demonstrated experimentally @xcite .
this is because , just like in the pure spin current switching , no net charge current flows in the cpip and cip geometries through the switching magnet in the direction perpendicular to its interface with the spacer .
nevertheless we shall see that a spin current is absorbed by the switching magnet and this gives rise to a nonzero spin - transfer torque .
this effect is sometimes called nonlocal spin - transfer torque ( for detailed discussion of spintronics circuits see ref . ) .
while the potential advantages of the cpip and cip geometries are obvious the crucial question is whether these alternative geometries are as efficient for switching / microwave generation as the traditional cpp geometry . to address this question
we have applied the nonequilibrium keldysh formalism @xcite to calculate from first principles the spin - transfer torques in the cpip and cip geometries .
we assume in all our calculations that the spin diffusion length is much longer than the dimensions of our system ( spin is conserved ) .
we performed calculations of the spin - transfer torque for perfect cpip and cip systems ( ballistic limit ) and also in the case of a rough nonmagnet / magnet interface to check that our results remain valid beyond the ballistic limit .
rather surprisingly both our single - orbital model calculations and fully realistic calculations for co / cu show that the spin current flowing parallel to the spacer / sm interface can be absorbed by the switching magnet as efficiently as in the traditional cpp geometry .
spin polarization of the current in the cip geometry is not as large as in the cpp geometry but remains sizable , of the same order of magnitude as in the cpp geometry .
the keldysh formalism had been applied previously by edwards _ et al . _
@xcite to calculate the spin - transfer torque in the cpp geometry .
an essential requirement for the implementation of the keldysh formalism is that a sample with an applied bias can be cleaved into two noninteracting left ( l ) and right ( r ) parts by passing a cleavage plane between two neighboring atomic planes .
it follows that , initially , neither charge nor spin current flows in the cleaved system although the left and right parts of the sample have different chemical potentials .
this is most easily achieved for a tight - binding ( t .- b . ) band structure since the t .- b . hopping matrix between the l and r parts
can be switched off .
we shall , therefore , describe our systems by a tight - binding model , in general multiorbital with s , p , and d orbitals whose one - electron parameters are fitted to first - principle band structure , as described previously @xcite .
the hopping between the l and r parts is then turned on adiabatically and the system evolves to a steady state .
the nonequilibrium keldysh formalism provides a prescription for calculating the steady - state charge and spin currents flowing between the l and r parts of the connected sample in terms of local one - electron green functions for the equilibrium cleaved system . in the cpp geometry , considered by edwards _
@xcite , the sample is translationally invariant in the direction parallel to all the interfaces and , therefore , the relevant quantity is the total spin current flowing between any two neighboring atomic planes .
in particular , the spin - transfer torque acting on the switching magnet is obtained as the difference between the spin currents entering and leaving the switching magnet ( the spin current is naturally conserved in the nonmagnetic spacer and leads ) .
@xcite showed that the local spin current is expressed entirely in terms of the one - electron surface green functions @xmath0 and @xmath1 for the cleaved sample . here , @xmath2 is the wave vector parallel to the interface .
the green functions at the surfaces of the cleaved system are obtained from the surface green functions of the nonmagnetic leads by the method of adlayers @xcite . in this method one
`` grows '' the sample by depositing , one by one , all its atomic planes on the leads and , after each deposition , the surface green function is updated using dyson s equation .
the surface green function of semi - infinite leads is obtained by the method of umerski @xcite .
we now wish to apply the keldysh method to the cpip and cip geometries . referring to fig.2 ,
it is clear that the translational invariance is broken in the z and y directions but @xmath3-space description remains valid in the x direction .
we , therefore , need to work in a representation that is atomic - like in the z and y directions but bloch - like in the x direction .
the method for modelling cpip and cip systems is shown schematically in fig.3 for the cpip geometry .
the whole system is built up from chains of atoms parallel to the z axis which are repeated periodically in the x direction .
we shall label the position of each chain by @xmath4 and the position of atoms within a chain by @xmath5 .
although we shall frequently refer to chains , in reality each chain stands for a sheet of atoms since the chains are repeated periodically in the x direction .
the tight - binding on - site potentials depend on the location of each atom in the sample and those for magnetic atoms include an interaction between electrons in d orbitals which leads to an exchange splitting of the bands in the ferromagnets .
the region which lies outside the sample is modelled by fictitious atoms with an infinite on - site potential which prevents electrons from hopping to these vacant sites .
all chains can be thus regarded as having the same length of @xmath6 atoms , where @xmath7 and @xmath8 are , respectively , the numbers of atoms in the lead and in the switching magnet ( vacuum ) in the vertical z direction .
it follows that we can create the whole sample by depositing all its chains one by one on semi - infinite left and right leads .
the surface green functions on the chains located immediately to the left and right of a cleavage plane , that are required in the calculation of the spin current , are obtained by updating the green function from the dyson equation after each chain deposition .
since the deposition of chains of atoms takes place in real space in the z direction the green function is a matrix of dimension @xmath9 , where @xmath10 is the number of atoms in a chain and @xmath11 is the number of orbitals .
the factor 2 appears because the green function has two components corresponding to two spin projections on the spin quantization axis . to calculate the spin and charge currents we assume that a bias @xmath12 is applied between the left and right leads .
our goal is to determine the spin and charge currents between any two neighboring chains of atoms parallel to the the z axis , i.e. to the interface between the left ( polarizing ) magnet and the lead .
if the cleavage line is first passed to the left of the switching magnet and then to the right of the magnet , as indicated in fig.3 , the spin - transfer torque acting on the switching magnet is obtained as the difference between the total spin currents in these two locations .
following edwards _ et al .
_ @xcite and assuming the linear - response case of a small bias , it is straightforward to show that the thermal average of the total spin current @xmath13 flowing between the chains @xmath14 and @xmath4 is given by @xmath15{\mbox{\boldmath $ \sigma$}}\}_{mm}v_{b } , \label{eq1}\end{aligned}\ ] ] where @xmath16^{-1}$ ] , @xmath17^{-1}$ ] are defined in terms of retarded surface green function matrices @xmath18 , @xmath19 for the decoupled equilibrium system .
the subscript l(r ) refers to the chains on the left ( right ) of the cleavage line .
the green functions depend on the wave vector @xmath3 labelling bloch states in the x direction and on the indices @xmath5 , @xmath20 labelling the atoms in a chain .
the matrix @xmath21 is the tight - binding interchain hopping matrix .
the components of @xmath22 are direct products of the 2@xmath232 pauli matrices @xmath24 , @xmath25 , @xmath26 and @xmath27 unit matrix .
finally , the trace in eq.(1 ) is taken over all the orbital and spin indices which are suppressed .
equation ( 1 ) yields the charge current if @xmath28 is replaced by a unit matrix multiplied by @xmath29 , where @xmath30 is the electronic charge .
it follows from eq.(1 ) that the total spin current ( charge current ) between the chains @xmath31 is the sum of partial currents flowing between pairs of atoms which are located on the opposite sides of the cleavage plane and connected by the t .- b . hopping matrix . by evaluating the individual partial currents we can ,
therefore , obtain a detailed information about the local current flow .
equation ( 1 ) yields , of course , only information about current flow in the y direction , which is perpendicular to the cleavage line .
however , by applying locally kirchhoff s law , the current components in the direction parallel to the cleavage line ( z axis ) can also be determined .
the current vector describing the flow of charge current between any two neighboring atoms in the ( y , z ) plane can be thus reconstructed . while local currents are not conserved , the total charge current between any two neighboring chains anywhere in the system is , of course , conserved .
the total spin current between neighboring chains is conserved in the nonmagnetic parts of the system but can be absorbed in the magnets , which gives rise to spin - transfer torque .
the application of eq.(1 ) to specific cpip and cip structures will be discussed in section 3 .
to gain some insight , we have first applied the keldysh formalism to the cpip and cip geometries using a single - orbital tight - binding model with atoms on a simple cubic lattice and nearest - neighbour hopping @xmath32 . in this model
the relevant parameters are the on - site potentials @xmath33 , @xmath34 which are measured in the units of @xmath35 .
the fermi level is always set at zero .
we begin with the cpip geometry illustrated in fig.2a . for a meaningful comparison of the cpip geometry with the traditional cpp setup , we also need to determine the cpp spin current for a system which is finite in the z direction .
we , have therefore , applied to the cpp geometry the same real space method described in section 2 for the cpip and cip geometries .
we choose the total number @xmath10 of atoms in a chain to be the same in the cpp and cpip geometries and make all the spin currents dimensionless by dividing them by the total charge current multiplied by @xmath36 , where @xmath30 is the electronic charge .
the magnetization of the polarizing magnet is assumed to be parallel to the x axis and that of the switching magnet is parallel to the z axis . for simplicity ,
we choose the polarizing magnet to be semi - infinite in the y direction .
the switching magnet should , of course , be finite since the torque is calculated by taking the difference between the spin currents before and after the switching magnet .
however , it has been demonstrated for the cpp geometry @xcite that the dependence of the outgoing spin current on the switching magnet thickness is almost exactly the same as the dependence of the spin current on the distance from the spacer / switching magnet interface in a semi - infinite magnet .
we checked that this is also true for the cpip geometry .
we may , therefore , determine the spin - transfer torque using a semi - infinite switching magnet .
the advantage of using a semi - infinite magnet is a faster convergence of the @xmath3-space sum since small and physically unimportant interference effects which occur in a ferromagnet of a finite thickness are eliminated . placing a cleavage plane in the position ( 1 ) in fig.3
, we first determine from eq.(1 ) the spin current in the nonmagnetic spacer , i.e. the spin current incident on the switching magnet .
we then place a cleavage plane between any two neighboring atomic chains in the switching magnet and determine again from eq.(1 ) the local spin current in the magnet .
the spin current @xmath13 flowing between the chains @xmath14 and @xmath4 can be then plotted as a function of the position @xmath4 of the cleavage plane in the switching magnet .
such plots are shown in fig.4 for @xmath37 and for three different aspect ratios @xmath38 , @xmath39 , and @xmath40 corresponding to the height of the switching magnet in the cpip geometry of one atom , ten atoms , and nineteen atoms .
the dependence of the spin current on @xmath4 in the cpp geometry is also shown in fig.4 .
the spin current curves in fig.4a and 4b correspond to different tight - binding on - site potentials in the polarizing and switching magnets , which are listed in the figure .
those in fig.4a were chosen so that the fermi level in the polarizing and switching magnets intersects both the majority- and minority - spin bands ( a weak magnet ) and there is a perfect matching between the bands of the nonmagnetic spacer and one of the ferromagnet bands . in fig.4b
both the polarizing and switching magnets are half - metals , i.e. , the minority - spin band is empty .
it should be noted that , in general , the spin current relevant for current - induced switching has an in - plane ( x ) and out - of - plane ( y ) components @xcite .
however , we show in fig.4 only the in - plane component since it is usually most important in switching . of the cleavage plane in the switching magnet .
the on - site potential parameters in ( a ) are @xmath41 , @xmath42 for the pm , @xmath43 for the spacer , and @xmath44 , @xmath45 for the sm .
the on - site potential parameters in ( b ) are @xmath46 , @xmath47 for the pm , @xmath48 for the spacer , and @xmath46 , @xmath47 for the sm.,scaledwidth=45.0% ] it can be seen from fig.4 that both the cpp and cpip spin currents decrease as the cleavage plane is moved through the switching magnet and become almost zero for a switching magnet of about fifty to hundred chains wide .
the only exception occurs for the aspect ratio @xmath40 for which the spin current is virtually nondecaying .
this will be explained later , once the physical mechanism governing the spin current absorption is clarified .
zero outgoing spin current corresponds to complete absorption of the spin current by the switching magnet , i.e. , maximum spin - transfer torque .
fig.4 demonstrates that almost complete absorption of the spin current is achieved not only in the cpp but also in the cpip geometry .
it should be noted that the rate of decay of the cpip spin current for a half - metallic magnet ( fig.4b ) is comparable to that for a weak magnet ( fig.4a ) but the cpp spin current decays much faster in a half - metallic ferromagnet .
since the results in figs.4a and 4b were obtained for magnets with different band parameters , it is clear that a complete absorption of the spin current by the switching magnet in the cpip geometry is a general phenomenon .
it can also be seen from fig.4 that a switching magnet of height of only one atom has essentially the same absorbing power as that having height of ten atoms . to understand these rather surprising results , we first recall the physical mechanism that governs the absorption of spin current in the cpp geometry @xcite . for noncolinear magnetizations of the polarizing and switching magnets ,
the spin of electrons incident on the switching magnet is at an angle to its exchange field .
it follows that the spin must precess in the exchange field of the switching magnet .
the precession frequency is determined by the components of the wave vectors of majority- and minority - spin electrons parallel to the current flow ( perpendicular to the interfaces ) .
given that the sum of the energies corresponding to perpendicular and parallel motion of electrons is constant ( equal to the fermi energy ) , the perpendicular components of the wave vector , which determine the precession frequency , are functions of the parallel component @xmath2 . since the total spin current involves the sum over @xmath2 , destructive interference of precessions with different frequencies occurs .
the conventional stationary phase argument @xcite then shows that only an extremal frequency of spin current oscillations survives .
the stationary phase argument also predicts that the amplitude of spin current oscillations decays as a function of the distance from the spacer / magnet interface .
such a behaviour of the cpp spin current is clearly seen in fig.4a .
the fast decay of the cpp spin current in the case of a half - metallic switching magnet can be explained as follows .
the wave function of an electron with a spin at an angle to the exchange field of a half - metallic switching magnet is a linear combination of the wave functions with spin parallel and antiparallel to the exchange field .
however , since only electrons with one spin projection on the direction of the exchange field ( magnetization ) exist in a half - metallic magnet the precession amplitude must decay exponentially .
this is the behaviour seen for the cpp spin current in fig.4b .
it is reasonable to assume that spin precession mechanism is also responsible for the decay of the spin current in the cpip geometry .
however , we need to establish that destructive interference of precessing spins can occur in this geometry and also that electrons travelling parallel to the spacer / switching magnet interface do penetrate the switching magnet , so that their spin can precess in the local exchange field . in an inhomogeneous finite sample shown in fig.3 , size quantization occurs and electrons thus travel in discrete size - quantized conductance channels .
this effect combined with the sum over the wave vector @xmath3 in the x direction provides in the cpip geometry the relevant channels for destructive interference .
however , because of the complexity of size quantization both in the y and z directions , a simple stationary phase argument is no longer applicable and an analytical formula for the spin current decay in the cpip geometry is thus not available .
the only exception is the case with an aspect ratio @xmath40 in fig.4a where size quantization is so severe that only one conductance channel is available .
destructive interference then occurs due only to different @xmath3-space channels to which the conventional stationary phase argument is applicable .
in contrast to the planar cpp geometry , the @xmath3-space sum in the cpip geometry is onedimensional and , therefore , the decay of spin current oscillations is much slower then in the planar cpp geometry .
although in the general case of a large number of size - quantized conductance channels we do not have a simple stationary - phase formula for the spin current in the cpip geometry , we can nevertheless make an estimate of the slowest decay of the spin current in a lateral switching magnet .
the spin current in eq.(1 ) is the trace over the real - space position in the vertical ( z ) direction combined with the sum over the wave vector @xmath3 labelling bloch states in the x direction .
the trace in the real space is essentially equivalent to a sum over discrete size - quantized conductance channels . for each conductance channel
the sum over the wave vector @xmath3 can be performed using the conventional stationary phase argument ( see ref . ) . that gives a decay of the spin current in each discrete conductance channel of the form @xmath49 , where n is the position of the cleavage plane in the switching magnet . since this conventional stationary - phase argument
can be applied to each conductance channel , the slowest decay of the spin current must be @xmath49 . in practise ,
destructive interference between different conductance channels also occurs , and that should lead to a faster decay than the most pessimistic estimate @xmath49 .
it remains to demonstrate that transport electrons penetrate the switching magnet despite the fact that they travel parallel to the interface . to show that we have determined the distribution of the local charge current in the switching magnet using the method outlined in section 2 .
the behaviour of the charge current is shown in fig.5 for @xmath50 ( strictly two - dimensional system ) and the aspect ratio @xmath39 .
, @xmath51 for the pm , @xmath52 for the spacer , and @xmath53 , @xmath54 for the sm.,scaledwidth=50.0% ] the orientation of each arrow in fig.5 represents the direction of the current flow and the length of the arrow gives the magnitude of the local charge current flowing between neighboring atoms .
figure 5 demonstrates that there is strong penetration of transport electrons into the switching magnet , and it is the spin precession of these electrons that results in a spin - transfer torque ( spin current absorption ) which is as large as in the cpp geometry .
finally , we need to explain why the decay of the cpip spin current in a half - metallic ferromagnet is slower than in the cpp geometry . in the cpp geometry
all electrons have to pass through the switching magnet and the spin current thus decays exponentially as discussed above . in the cpip geometry
there are many electrons that penetrate only partially the switching magnet and are then reflected back to the spacer .
the spin of such electrons with a shallow penetration can precess in the exchange field of the switching magnet and the decay of the spin current is thus not qualitatively different from that for a weak magnet ( see fig.4a and 4b ) .
the results shown in fig.4 and fig.5 are for structures with perfect interfaces , that are illustrated in fig.2a .
interfaces in real structures may well be rough and it is , therefore , necessary to investigate the effect of interfacial roughness on the absorption of the spin current by the lateral switching magnet . since the systems we consider are `` grown '' in real space it is straightforward to include in our calculations the effect of a random intermixing of atoms in the nonmagnetic spacer and switching magnet .
the effect of an intermixing over two interfacial atomic planes on the absorption of the spin current is shown in fig.6 .
the intermixing was modelled by replacing the two interfacial atomic planes by a 50% alloy of spacer and magnet atoms .
the results for a perfect system are also reproduced in fig.6 .
it can be seen that intermixing does not spoil the strong absorption of the spin current by a lateral switching magnet .
the other interesting feature is that the spin current for a perfect cpip system exhibits oscillations reminiscent of those that are seen in the cpp geometry .
while oscillations of the spin current in the cpp geometry can be explained by the stationary phase theory , a simple stationary - phase argument is not available for the cpip geometry and the precise origin of the oscillations in this geometry is thus not clear .
however , it can be seen in fig.6 that cpip oscillations are removed in a system with rough interface . of the cleavage plane in the switching magnet for a rough and a perfect interface .
the on - site potential parameters are @xmath55 , @xmath56 for the pm , @xmath57 for the spacer , and @xmath55 , @xmath56 for the sm .
the lead / magnet height is 10/10 atomic planes for both the perfect and the rough systems , scaledwidth=45.0% ] we now investigate the cip geometry in which the current flows parallel not only to the interface between the spacer and the switching magnet but also to the interface between the spacer and the polarizing magnet . since the absorbing power of the switching magnet in the cip geometry must clearly be the same as in the cpip geometry , the key question here is the polarizing ability of a polarizing magnet whose interface with the spacer is parallel to the current flow . to determine the spin current , we proceed as in the cpip geometry ( fig.3 ) .
we place a cleavage plane between any two neighboring atomic chains in the switching magnet and determine from eq.(1 ) the local spin current as a function of the position @xmath4 of the cleavage plane in the switching magnet . the continuity of the spin current guarantees that the value of the spin current at the spacer / switching magnet interface is equal to the spin current in the spacer .
it follows that the values of the spin current incident on and leaving the switching magnet can both be determined from the profile of the spin current in the switching magnet .
this is shown in fig.7 for the situation when the polarizing magnet is a half - metal ( the minority - spin band is empty ) but the fermi level in the switching magnet intersects both the majority and minority - spin bands . of the cleavage plane in the switching magnet for a cip and a cpp system .
the on - site potential parameters are @xmath55 , @xmath58 for the pm , @xmath57 for the spacer and @xmath55 , @xmath56 for the sm .
the lead / magnet height is 10/10 atomic planes.,scaledwidth=40.0% ] there are two interesting features seen in fig.7 .
first of all we note that in the cpp geometry only majority - spin carriers can pass through a half - metallic polarizing magnet and , therefore , the spin polarization of the current incident on the switching magnet is 100% and in the direction of the spin of the majority - spin carriers . on the other hand ,
the spin polarization in the cip geometry is much smaller , only about 25% .
the second interesting feature is that the spin polarization of the current in the cip geometry has a sign opposite to that in the cpp geometry .
this can be most easily understood in our special case of a half - metallic polarizing magnet whose majority - spin band matches exactly the bands of either spin in the nonmagnetic spacer .
minority - spin carriers , which can not penetrate the polarizing magnet , travel as if in a perfect slab without being scattered from the region in which the polarizing magnet is located . on the other hand , majority - spin carriers which can easily penetrate the polarizing magnet region
are strongly scattered by the geometrical inhomogeneity of that region , which strongly reduces but does not suppress completely their current flow .
we thus do not expect the spin polarization to be complete .
moreover , the current of the minority - spin carriers is larger than that of the majority - spin carriers and the sign of the spin current polarization is thus reversed .
our model calculations for a single - orbital tight - binding band indicate that the absorption of the spin current by a lateral magnet in the cpip ( cip ) geometry is as efficient as in the standard cpp geometry . to confirm that these results remain valid for a fully realistic system
, we have made calculations of the spin current profile in a cobalt switching magnet whose interface with a nonmagnetic copper spacer is parallel to the current flow ( cpip geometry illustrated in fig.2a ) .
we used in these calculations a semiinfinite fcc co sheet of height 4 and 8 atomic planes as a polarizing magnet .
the switching magnet was a sheet of co of height 4 ( 8) atomic planes deposited on a cu lead whose height was also 4 ( 8) atomic planes .
the crystal orientation of the co and cu sheets was ( 001 ) .
both co and cu sheets were described by a fully realistic multiorbital tight - binding model with tight - binding parameters fitted to the results of first - principles band structure calculations ( see ref . ) .
the magnetization of the polarizing co magnet was taken to be in the x direction and that of the switching co magnet was in the z direction .
as in our one - band model calculations , the co / cu cpip system was grown in real space and the spin current was evaluated without any approximations from the keldysh formula ( 1 ) .
it should be noted that for a system with 8 + 8 atomic sheets , all the matrices in eq.(1 ) have size @xmath59 , which makes the evaluation of the spin current computationally very demanding .
hence our restriction to the maximum size of 8 + 8 atomic sheets .
the dependence of the cpip in - plane spin current on the position @xmath4 of the cleavage plane in the co switching magnet is shown in fig.8 .
for comparison , the cpp spin current is also shown in fig.8 ( continuous line ) . in the cpp geometry , of the cleavage plane in the cobalt switching magnet.,scaledwidth=45.0% ] the co polarizing magnet , the cu spacer , and the co switching magnet were all sheets of 4 atomic planes .
it can be seen from fig.8 that in the case of the 4 + 4 cpip system the absorption of the spin current is as fast as in the conventional cpp geometry .
the long - period oscillations of the spin current in the cpip and cpp geometry are very similar but we can see in the cpp geometry an additional short oscillation period which is not present in the cpip geometry .
the absorption of the spin current for the 8 + 8 cpip system is slower but , nevertheless , more than two thirds of the spin current are absorbed over 50 atomic planes .
our results for realistic co / cu systems thus confirm the viability of a setup with a lateral switching magnet , i.e. the cpip geometry in which the current flows parallel to the spacer / switching magnet interface .
using the nonequilibrium keldysh theory , we have investigated theoretically two geometries for current induced switching of magnetization in which the current flows parallel to the magnet / nonmagnet interface . in the first geometry the current is perpendicular to the polarizing magnet / spacer interface but parallel to the spacer / switching magnet interface ( cpip ) . in the second geometry the current is parallel to both the polarizing magnet / spacer and spacer / switching magnet interfaces ( cip ) . our calculations for a single - orbital tight binding model indicate that the spin current flowing parallel to the switching magnet / spacer interface can be absorbed by a lateral switching magnet as efficiently as in the traditional cpp geometry .
we have confirmed that the results of such model calculations in the cpip geometry are also valid for experimentally relevant co / cu cpip system described by fully realistic tight binding bands fitted to an ab initio band structure .
our results show that almost complete absorption of the incident spin current by a lateral switching magnet ( magnetic dot ) occurs when the lateral dimensions of the switching magnet are of the order of 50 - 100 interatomic distances , i.e. , about 20 nm .
the numerical results are supported by an analytical stationary phase argument which indicates that the decay of the spin current in a lateral switching magnet should not be slower than @xmath60 , where @xmath4 is the lateral size of the magnet measured in the units of interatomic spacing .
hence about 90% spin current absorption should be achieved by a magnet of a lateral size of about 20 nm . moreover , to achieve full absorption of the spin current ( maximum spin - transfer torque ) , the height of a lateral switching magnet can be as small as a few atomic planes .
it follows that the total volume of the switching magnet in the cpip ( cip ) geometry can be even smaller than that in the traditional cpp geometry using magnetic nanopillars .
this indicates that current - induced switching and microwave generation in the cpip geometry should be feasible .
we have also demonstrated that strong spin current absorption in the cpip / cip geometry is not spoilt by the presence of a rough interface between the switching magnet and nonmagnetic spacer .
we find that the polarization achieved using a lateral magnet in the cip geometry is only about 25% of that in the traditional cpp geometry .
the cpip geometry is thus preferable but cip could be still usable with a stronger current .
finally , we wish to make contact with the recent experiment , see ref . , in which the so called pure - spin - current - induced magnetization switching had been demonstrated . in the experimental setup of ref .
the current was spin polarized by passing it through a magnet ( current perpendicular to magnet / spacer interface ) but the resultant spin current was absorbed by a lateral magnet ( current parallel to magnet / spacer interface ) .
the experimental setup of ref .
is thus topologically equivalent to our cpip geometry .
we are grateful to the uk engineering and physical sciences research council for financial support within the framework of the spin@rt consortium and to the members of the consortium for stimulating discussions .
albert , j.a .
katine , r.a .
buhrman , and d.c .
ralph , appl .
. lett . * 77 * , 3809 ( 2000 ) .
katine , f.j .
albert , r.a .
buhrman , e.b .
meyers , and d.c .
ralph , phys .
lett . * 84 * , 3149 ( 2000 ) .
kiselev , j.c .
sankey , i.n .
krivorotov , n.c .
emley , r.j .
schoelkopf , r.a .
buhrman , and d.c .
ralph , nature 425 , 380 ( 2003 ) .
s. urazhdin , norman o. birge , w.p .
pratt , jr . , and j. bass , phys .
lett . * 91 * , 146803 ( 2003 ) .
pufall , w.h .
rippard , shehzaad .
kaka , s.e .
russek , t.j .
silva , jordan .
katine , and matt .
carey , phys .
b. * 69 * , 214409 ( 2004 ) .
d.m . edwards and j. mathon , j. phys . : condens .
matter * 19 * , 165210 ( 2007 ) .
tao yang , takashi kimura , and yoshichika otani , nature physics * 4 * , 851 ( 2008 ) .
a. brataas , g.e.w .
bauer , and p.j .
kelly , phys .
rep . * 427 * , 157 ( 2006 ) .
keldysh , soviet physics jetp * 20 * , 1018 ( 1965 ) . c. caroli , r. combescot , p. nozieres , and d. saint - james , j. phys .
c * 4 * , 916 ( 1971 ) .
edwards , f. federici , j. mathon , and a. umerski , phys .
b * 71 * , 054407 ( 2005 ) .
j. mathon , murielle villeret , a. umerski , r.b .
muniz , j. dalbuquerque e castro , and d.m .
edwards , phys .
b * 56 * , 11797 ( 1997 ) .
a. umerski , phys .
b * 55 * , 5266 ( 1997 ) . m.d . stiles and a. zangwill , phys . rev .
b * 66 * , 014407 ( 2002 ) .
j. mathon , murielle villeret , and h. itoh , phys .
b * 52 * , r6983 ( 1995 ) . | two alternative current - induced switching geometries , in which the current flows parallel to the magnet / nonmagnet interface , are investigated theoretically using the nonequilibrium keldysh theory . in the first geometry , the current is perpendicular to the polarizing magnet / nonmagnet interface but parallel to the nonmagnet / switching magnet interface ( cpip ) . in the second geometry , the current is parallel to both the polarizing magnet / nonmagnet and nonmagnet / switching magnet interfaces ( cip ) .
calculations for a single - orbital tight binding model indicate that the spin current flowing parallel to the switching magnet / nonmagnet interface can be absorbed by a lateral switching magnet as efficiently as in the traditional current - perpendicular - to - plane ( cpp ) geometry .
the results of the model calculations are shown to be valid also for experimentally relevant co / cu cpip system described by fully realistic tight binding bands fitted to an ab initio band structure .
it is shown that almost complete absorption of the incident spin current by a lateral switching magnet occurs when the lateral dimensions of the switching magnet are of the order of 50 - 100 interatomic distances , i.e. , about 20 nm and its height as small as a few atomic planes .
it is also demonstrated that strong spin current absorption in the cpip / cip geometry is not spoilt by the presence of a rough interface between the switching magnet and nonmagnetic spacer .
polarization achieved using a lateral magnet in the cip geometry is found to be about 25% of that in the traditional cpp geometry .
the present cpip calculations of the spin transfer torque are also relevant to the so called pure - spin - current - induced magnetization switching that had been recently observed . |
we are concerned with the following problem : given an equilateral triangle @xmath2 , find all dissections of @xmath2 into smaller nonoverlapping equilateral triangles .
the _ size _ of a dissection is the number of nonoverlapping equilateral triangles .
an example of such a dissection of size @xmath3 is shown in figure [ exdissection ] .
it is well known that in such a dissection all triangles may be regarded as triangles with sides of integer length .
dissections of squares have been studied earlier @xcite as well as dissections of squares into right - angled isosceles triangles @xcite .
recently , laczkovich@xcite studied tilings of polygons by similar triangles .
the earliest study of dissections of equilateral triangles into equilateral triangles is by tutte @xcite .
the problem of dissecting a triangle is different to normal tiling problems where the size of the tiles is known in advance and the tiling area may be infinite .
a naive approach to enumerating dissections is to first fix the sizes and number of the dissecting triangles . observe that in any dissection , some triangles will be oriented in the same way as the triangle @xmath2 ( these are the up - triangles ) while the oppositely oriented triangles are the down - triangles .
let @xmath4 and @xmath5 be the number of up and down triangles of side length @xmath6 , respectively . for any down - triangle the horizontal side is adjacent to the horizontal side of some number of up - triangles .
the up - triangles along the bottom of @xmath2 are not adjacent to any down - triangle .
so if the triangle @xmath2 has side length @xmath7 then @xmath8 a triangle with length @xmath9 has height @xmath10 and so the triangle areas give the relation @xmath11 for small values of @xmath12 we can solve and for the permissible size and number of up and down triangles , and this data may guide an exhaustive search
. we will consider up and down triangles not to be congruent even if they are of the same size .
a _ perfect tiling _ or _ perfect dissection _ has no pair of congruent triangles .
this definition of a perfect dissection arises from the fact that it is impossible to have a perfect dissection if orientation is ignored @xcite .
this fact can also easily derived from the results in @xcite .
tutte conjectured @xcite that the smallest perfect dissection has size @xmath1 ( see also @xcite ) .
unfortunately solving and with @xmath13 is computationally intensive and so another approach is needed . using our enumerative methods we confirm tutte s conjecture and provide all perfect dissections up to size @xmath0 .
lines parallel to the outer sides of the main triangle that are induced by a side of a dissecting triangle are _ dissecting lines_. for any dissecting line @xmath14 , the union of all sides of dissecting triangles that are incident to @xmath14 forms one or more contiguous segments . if there are two or more segments , then on @xmath14 there exist two dissecting vertices such that all triangles in between are cut by the line into two parts .
if such a situation arises for no dissecting line and if no dissecting vertex is incident to six dissecting triangles , then we call the dissection
_ separated_. we enumerate all isomorphism classes of separated and nonseparated dissections up to size @xmath0 .
the connection between equilateral triangle dissections and latin bitrades was first studied in @xcite .
the presentation here follows @xcite .
consider an equilateral triangle @xmath2 that is dissected into a finite number of equilateral triangles .
dissections will be always assumed to be nontrivial so the number of dissecting triangles is at least four .
denote by @xmath15 , @xmath16 and @xmath17 the lines induced by the sides of @xmath2 .
each side of a dissecting triangle has to be parallel to one of @xmath15 , @xmath16 , or @xmath17 .
if @xmath18 is a vertex of a dissecting triangle , then @xmath18 is a vertex of exactly one , three or six dissecting triangles .
suppose that there is no vertex with six triangles and consider triples @xmath19 of lines that are parallel to @xmath15 , @xmath16 and @xmath17 , respectively , and meet in a vertex of a dissecting triangle that is not a vertex of @xmath2 .
the set of all these triples together with the triple @xmath20 will be denoted by @xmath21 , and by @xmath22 we shall denote the set of all triples @xmath19 of lines that are yielded by sides of a dissecting triangle ( where @xmath23 , @xmath24 and @xmath25 are again parallel to @xmath15 , @xmath16 and @xmath17 , respectively ) .
the following conditions hold : 1 . sets @xmath21 and
@xmath22 are disjoint ; 2 . for all @xmath26 and all @xmath27 , @xmath28 ,
there exists exactly one @xmath29 with @xmath30 and @xmath31 ; and 3 . for all @xmath32 and all @xmath27 , @xmath28 ,
there exists exactly one @xmath33 with @xmath34 and @xmath35 .
note that ( r2 ) would not be true if there had existed six dissecting triangles with a common vertex .
conditions ( r13 ) are , in fact , axioms of a combinatorial object called latin bitrades @xcite . a bitrade is usually denoted @xmath36 .
observe that the bitrade @xmath37 associated with a dissection encodes qualitative ( structural ) information about the segments and intersections of segments in the dissection .
the sizes and number of dissecting triangles can be recovered by solving a system of equations derived from the bitrade ( see below ) .
dissections are related to a class of latin bitrades with genus @xmath38 .
to calculate the genus of a bitrade we use a permutation representation to construct an oriented combinatorial surface @xcite . for @xmath39 , define the map @xmath40 where @xmath41 if and only if @xmath42 and @xmath43 for @xmath44 . by ( r1 - 3 ) each @xmath45 is a bijection
. then @xmath46 are defined by @xmath47 we refer to @xmath48 $ ] as the @xmath49 _ representation_. to get a combinatorial surface from a bitrade we use the following construction : [ constructiontauisurface ] let @xmath48 $ ] be the representation for a bitrade where the @xmath49 act on the set @xmath50 .
define vertex , directed edge , and face sets by : @xmath51 where @xmath52 denotes a face with @xmath53 directed edges @xmath54 , @xmath55 , @xmath56 , @xmath57 , @xmath58 for vertices @xmath59 .
the first set in the definition of @xmath60 is the set of _ triangular faces _ , while the other three are the _
@xmath49 faces_. assign triangular faces a positive ( anticlockwise ) orientation , and assign @xmath49 faces negative ( clockwise ) orientation .
now glue the faces together where they share a common directed edge @xmath61 , ensuring that edges come together with opposite orientation
. the orientation of a triangular face is shown in figure [ figtwoorientations ] along with the orientation for a @xmath49 face .
for the sake of concreteness we have illustrated a @xmath62-cycle face due to a @xmath62-cycle of @xmath63 .
figure [ figdrapalrotationscheme ] shows the rotation scheme for an arbitrary vertex in the surface . is a cycle of @xmath49 .
the vertex in the centre is @xmath64 . ] for a bitrade @xmath65 with representation @xmath48 $ ] on the set @xmath50 , define @xmath66 , the total number of cycles , and @xmath67 , the total number of points that the @xmath49 act on . by some basic counting arguments we find that there are @xmath68 vertices , @xmath69 edges , and @xmath70 faces .
then euler s formula @xmath71 gives @xmath72 where @xmath73 is the genus of the combinatorial surface .
we say that the bitrade @xmath74 has genus @xmath73 .
spherical bitrade _ has genus @xmath38 .
[ exspherical ] the following bitrade is spherical : @xmath75 here , the @xmath49 representation is @xmath76 where the triples @xmath77 refer to entries @xmath78 .
we will generally assume that a bitrade is _ separated _ , that is , each row , column , and symbol is in bijection with a cycle of @xmath63 , @xmath79 , and @xmath80 , respectively .
we now describe how to go from a separated spherical latin bitrade to a triangle dissection ( for more details see @xcite ) .
let @xmath81 be a latin bitrade .
it is natural to have different unknowns for rows , columns and symbols , and so we assume that @xmath82 whenever @xmath83 , @xmath84 and @xmath85 .
( if the condition is violated , then @xmath74 can be replaced by an isotopic bitrade for which it is satisfied . )
fix a triple @xmath86 and form the set of equations @xmath87 consisting of @xmath88 , @xmath89 , @xmath90 and @xmath91 if @xmath92 and @xmath84 .
the theorem below shows that if @xmath74 is a spherical latin bitrade then @xmath93 always has a unique solution in the rationals .
the pair @xmath94 will be called a _ pointed _ bitrade .
write @xmath95 , @xmath96 , @xmath97 for the solutions in @xmath93 for row variable @xmath98 , column variable @xmath99 , and symbol variable @xmath100 , respectively .
we say that a solution to @xmath93 is _ separated _ if @xmath101 whenever @xmath102 ( and similar for columns and symbols ) . for each entry
@xmath103 we form the triangle @xmath104 which is bounded by the lines @xmath105 , @xmath106 , @xmath107 .
of course , it is not clear that @xmath108 is really a triangle , i. e. that the three lines do not meet in a single point .
if this happens , then we shall say that @xmath104 _
degenerates_. let @xmath109 denote the subset of @xmath22 such that @xmath104 does not degenerate .
a separated dissection with @xmath110 vertices corresponds to a separated spherical bitrade @xmath36 where @xmath111 .
one of the main results of @xcite is the following theorem : [ sepdissections ] let @xmath112 be a spherical latin bitrade , and suppose that @xmath113 is a triple such that the solution to @xmath114 is separated .
then the set of all triangles @xmath104 , @xmath115 , dissects the triangle @xmath116 @xmath117 , @xmath118 and @xmath119 .
this dissection is separated .
an equilateral dissection can be obtained by applying the transformation @xmath120 .
consider the following spherical bitrade @xmath121 : + @xmath122 let @xmath123 .
then the system of equations @xmath124 has the solution @xmath125 the dissection is shown in figure [ fg2 ] .
entries of @xmath126 correspond to triangles in the dissection .
for example , @xmath127 is the triangle bounded by the lines @xmath128 , @xmath129 , @xmath130 while @xmath131 corresponds to the intersection of the lines @xmath132 , @xmath133 , @xmath134 .
, @xmath99 , @xmath100 , refer to lines @xmath135 , @xmath136 , @xmath137 , respectively .
the trade @xmath138 has @xmath139 entries and the dissection has @xmath140 vertices . applying the transformation @xmath120
gives an equilateral dissection . ]
suppose that the dissection @xmath2 has no vertex of degree @xmath62 .
pick a vertex @xmath18 of degree @xmath141 .
if we move to the right along the row segment to the next vertex @xmath142 then we have @xmath143 .
similarly , moving along the diagonal segments gives the action of @xmath79 and @xmath80 .
if we identify the three vertices of degree @xmath144 then the dissection encodes , geometrically , the permutation representation of the bitrade .
if a dissection has a vertex of degree @xmath62 then the dissecting triangles do not ( uniquely ) define a partial latin square and hence do not encode a latin bitrade .
however , we can recover a separated bitrade by the following procedure .
for each vertex @xmath18 of degree @xmath62 , choose one segment ( say , the @xmath98 segment ) to stay fixed . then for the @xmath99 and @xmath100 segments , label the column segment below @xmath18 with a new name @xmath145 and label the symbol segment below @xmath18 with a new name @xmath146 .
for example , the centre vertex in figure [ figsixway ] results in the new labels @xmath147 and @xmath148 .
the resulting separated bitrade is : @xmath149 this procedure works for any number of vertices of degree @xmath62 , as long as care is taken to only relabel column or symbol segments below a vertex of degree @xmath62 and not to relabel a segment more than once . .
] recently theorem [ sepdissections ] has been strengthened to cover nonseparated as well as separated dissections : [ strongertheorem ] let @xmath112 be a spherical latin bitrade . then for any @xmath113 , the set @xmath109 of non - degenerate triangles dissects the triangle @xmath116 @xmath117 , @xmath118 and @xmath119 .
the dissection may not be separated .
a fundamental consequence of the theorem is that any dissection of size @xmath6 can be derived from a pointed spherical bitrade @xmath150 of size @xmath6 .
note that the systems of equations @xmath93 have also other applications : both @xcite and @xcite use them to show that every spherical latin trade can be embedded into a finite abelian group .
cavenagh and lisonk @xcite showed that spherical bitrades are equivalent to planar eulerian triangulations . to enumerate triangle dissections we use plantri @xcite to enumerate all planar eulerian triangulations up to size @xmath0
( we also note that in @xcite all trades and bitrades have been enumerated up size @xmath151 ) .
we wrote a plugin @xcite to output the equivalent spherical latin bitrade @xmath152 for each triangulation .
for each such @xmath153 we find all solutions @xmath93 for all @xmath154 and compute each dissection ( in practise this is a list of triangles @xmath104 for each @xmath155 ) . to filter out isomorphic dissections
we apply all six elements of the symmetry group for a unit - side equilateral triangle ( identity , two rotations , and three reflections ) .
canonical signature _ of a dissection @xmath109 is the ordered list @xmath156 $ ] .
we repeat the whole process for the bitrade @xmath157 because there are spherical latin bitrades where @xmath153 is not isomorphic to @xmath157 .
the final counts for the number of dissections up to isomorphism are found by simply removing duplicate signatures . while the solutions to @xmath93 exist in the rationals , we find it easier to work with the final equilateral dissections instead .
we use the sympy package @xcite to perform exact symbolic arithmetic on the canonical form of each dissection . using theorems [ sepdissections ] and [ strongertheorem ] we have enumerated the number of isomorphism classes of dissections of size @xmath158 .
we also record @xmath159 , the number of dissections of size @xmath12 with automorphism group of order @xmath53 .
see figures [ figcountseparated ] and [ figcountseparatedandnonseparated ] for the data .
the referee raised the question of asymptotic behaviour .
let us denote by @xmath160 the number of all dissections of size @xmath12 .
thus @xmath161 , @xmath56 , @xmath162 , @xmath56 , @xmath163 .
there are some reasons to believe that @xmath160 can be estimated as @xmath164 , where @xmath165 is a slowly growing function .
the asymptotic behaviour of @xmath165 is not clear yet and is a subject of ongoing research .
note however that @xmath166 for every @xmath167 such that @xmath168 .
if we put @xmath169 , then in this interval the approximate values of @xmath170 are @xmath171 , @xmath172 , @xmath173 , @xmath174 , @xmath175 , @xmath176 , @xmath177 , @xmath178 , @xmath179 , @xmath180 , @xmath181 , @xmath181 and @xmath180 .
the fact that @xmath182 and @xmath183 are slightly greater than @xmath184 and @xmath185 , while @xmath186 is smaller than @xmath187 , seems to be surprising . [ cols="^,^,^,^,^,^",options="header " , ] the dissections that give rise to these maximum ratios are shown below ( sorted by dissection size @xmath12 ) : 10 url # 1`#1`urlprefixhref # 1#2#2 # 1#1 g. brinkmann , b. d. mckay , plantri ( software ) , http://cs.anu.edu.au/bdm/plantri .
g. brinkmann , b. d. mckay , fast generation of some classes of planar graphs , electronic notes in discrete mathematics 3 ( 1999 ) 2831 .
http://cs.anu.edu.au/bdm/plantri .
r. l. brooks , c. a. b. smith , a. h. stone , w. t. tutte , the dissection of rectangles into squares , duke math . j. 7 ( 1940 ) 312340 .
n. j. cavenagh , p. lisonk , planar eulerian triangulations are equivalent to spherical latin bitrades , j. comb .
theory , ser . a 115 ( 1 ) ( 2008 ) 193197 .
n. j. cavenagh , i. m. wanless , latin trades in groups defined on planar triangulations , j. algebraic combin . 30
( 2009 ) 323347 . c. j. colbourn , j. h. dinitz , i. m. wanless , handbook of combinatorial designs , chapman & hall / crc , boca raton , fl , 2007 .
a. drpal , dissections of equilateral triangles and pointed spherical latin bitrades , submitted .
a. drpal , geometrical structure and construction of latin trades , advances in geometry 9 ( 3 ) ( 2009 ) 311348 .
a. drpal , geometry of latin trades , manuscript circulated at the conference loops , prague ( 2003 ) .
a. drpal , hamming distances of groups and quasi - groups , discrete math . 235 ( 1 - 3 ) ( 2001 ) 189197 , combinatorics ( prague , 1998 ) .
a. drpal , v. kala , c. hmlinen , latin bitrades , dissections of equilateral triangles and abelian groups , journal of combinatorial designs , volume 18 issue 1 ( 2010 ) , 124 .
c. hmlinen , latin bitrades and related structures , phd in mathematics , department of mathematics , the university of queensland , http://carlo-hamalainen.net/phd/hamalainen-20071025.pdf ( 2007 ) . c. hmlinen ,
spherical bitrade enumeration code , http://bitbucket.org/carlohamalainen/spherical . c. hmlinen , triangle dissections code , http://bitbucket.org/carlohamalainen/dissections .
m. laczkovich , tilings of polygons with similar triangles , combinatorica 10 ( 3 ) ( 1990 ) 281306 .
j. d. skinner , c. a. b. smith , w. t. tutte , on the dissection of rectangles into right - angled isosceles triangles , j. combin .
theory ser .
b 80 ( 2 ) ( 2000 ) 277319 . , http://www.sympy.org[sympy : python library for symbolic mathematics ] ( 2009 ) .
http://www.sympy.org tiling by similar triangles , http://www.squaring.net/tri/twt.html .
w. t. tutte , the dissection of equilateral triangles into equilateral triangles , proc .
cambridge philos .
44 ( 1948 ) 463482
. ian m. wanless . a computer enumeration of small latin trades . ,
39:247258 , 2007 .
here we present representatives of the isomorphism classes of triangle dissections of size @xmath188 . | we enumerate all dissections of an equilateral triangle into smaller equilateral triangles up to size @xmath0 , where each triangle has integer side lengths .
a _ perfect _ dissection has no two triangles of the same side , counting up- and down - oriented triangles as different .
we computationally prove w. t. tutte s conjecture that the smallest perfect dissection has size @xmath1 and we find all perfect dissections up to size @xmath0 . |
globular clusters ( gcs ) , as single - age , single - metallicity objects , are excellent tracers of the formation history of their host galaxies , through their dynamics , kinematics , metallicities , and ages . for most galaxies within @xmath20 mpc
, gcs can be identified through photometry and image morphology , from which follow - up radial velocity studies can be carried out with multi - object spectroscopy on 4 and 8 m class telescopes . the ability to target hundreds of objects in a single field
has vastly increased the observed samples of gcs confirmed in many galaxies , providing the necessary basis for detailed kinematic and age studies .
another benefit of using gcs as a kinematic tracer is that they provide a useful , independent basis for comparison with results from planetary nebulae ( pne ) .
this is particularly true given the current debate surrounding the use of pn velocities and the implications for low dark matter halos .
for example , @xcite reported , based on pn velocities , that three low - luminosity ellipticals revealed declining velocity dispersion profiles and little or no dark matter .
however , subsequent simulations of merger - remnant ellipticals suggested that the radial anisotropy of intermediate - age pne could give rise to the observed profiles within standard halos of dark matter @xcite .
flattening of the galaxy along the line of sight is another possible explanation .
one of the ellipticals studied by @xcite , ngc 3379 , has also been investigated using gc kinematics @xcite ; both studies found evidence of a dark matter halo .
but the two studies of ngc 3379 suffered from small number statistics . clearly there is a need to directly compare pn and gc kinematics for the same elliptical with sufficiently large numbers of tracer objects .
previous velocity - based studies of globular cluster systems ( gcss ) have shown an intriguing variety of results in their overall kinematics .
prominent recent examples include the following : \1 .
@xcite performed a kinematic analysis of the gcs in m87 ( ngc 4486 ) , the cd galaxy in the virgo cluster . with a sample of @xmath21 gcs
, they showed that the entire gcs rotates on an axis matching the photometric _ minor _ axis of the galaxy , except for the inner metal - poor sample .
inside the onset of the cd envelope , the metal - poor clusters appear to rotate around the _ major _ axis of the galaxy instead .
this study also found evidence for an increase in velocity dispersion , @xmath22 , with radius , due to the larger scale virgo cluster mass distribution .
they also showed no strong evidence for a difference in @xmath22 between the metal - poor and metal - rich groups . using a virgo mass model , they investigated anisotropy and found that as a whole , the gcs had isotropy , but considered separately , the metal - poor and metal - rich subpopulations had slight anisotropy .
@xcite performed a dynamical analysis for m49 ( ngc 4472 ) , the other supergiant member of the virgo cluster . using over @xmath23 gcs
, they found that the metal - rich population shows no strong evidence for rotation , while the metal - poor population does rotate about the minor axis of the galaxy .
in addition , they found that the metal - poor clusters had an overall higher dispersion than the metal - rich population .
@xcite determined the kinematics for the gcs in ngc 1399 , the brightest elliptical in the fornax cluster of galaxies .
armed with a sample of over @xmath24 gcs , they found a marginal rotation signal for the entire gc sample and the outer metal - poor sample , while no rotation was seen for the metal - rich subpopulation .
their projected velocity dispersion showed no radial trend within their determined uncertainty , but the metal - poor clusters had a higher dispersion than the metal - rich clusters .
the kinematics of the pne in ngc 1399 have been most recently studied by @xcite and @xcite using a small sample of 37 pne from @xcite .
these studies do not indicate any significant deviations in velocity dispersion with radius for the pne , yet interestingly , @xcite found stong rotation for the pne in the inner region of the galaxy .
@xcite studied the gc kinematics of the giant elliptical ngc 5128 from a total of @xmath25 gcs .
their results showed definite rotation signals in both metallicity groups beyond a 5 kpc distance from the center of the galaxy , as well as similar velocity dispersions in both the metal - poor and the metal - rich populations within a 20 kpc projected radius .
a similar pn kinematic study in ngc 5128 @xcite showed that the pne population is rotating around a twisted axis that turns just beyond the 5 kpc distance .
these few detailed kinematic studies of gcss in elliptical galaxies , show results that appear to differ on a galaxy - by - galaxy basis , without any clear global trends .
the age distributions of the gc populations in these same galaxies tend to show a consistent pattern in which the blue , or metal - poor , population is found to be universally old .
the red , or metal - rich , population has also been shown to be old in the study of @xcite .
their study found that _ both _ the metal - poor and metal - rich gcs in a sample of eight galaxies ranging from dwarf to massive ellipticals , all have ages as old as their galactic gc counterparts .
conversely , in a small sample of studies , the red gc population has also been found to be 2 - 4 gyr younger than the metal - poor gc population , and with a wider spread in determined ages ( see the studies of * ? ? ?
* ; * ? ? ?
* among others ) , although this is not yet a well - established trend .
the gcs in ngc 5128 appear to be old , with an intermediate - age population @xcite .
ngc 5128 ( centaurus a ) , the central giant in the centaurus group of galaxies at a distance of only @xmath26 mpc , is a prime candidate for both kinematic and age studies .
its gcs has a specific frequency of @xmath27 @xcite , toward the low end of the giant elliptical range but about twice as high as in typical disk galaxies .
its optical features show faint isophotal shells located in the halo ( e.g. * ? ? ?
* ) , the prominent dust lane in the inner 5 kpc , the presence of gas , and star formation , all of which suggest that ngc 5128 could be a merger product . @xcite
first suggested that ngc 5128 could be the result of a merger between two galaxies , a spiral and an elliptical .
this idea was followed by the general formation mechanism of disk - disk mergers proposed by @xcite .
@xcite found that the metallicity distribution function of the halo field stars could be reproduced by a gas - free ( dry ) disk - disk merger scenario .
recent numerical simulations by @xcite also demonstrate that the pn kinematics observed in ngc 5128 @xcite can be reproduced relatively well from a merger of unequal - mass disk galaxies ( with one galaxy half the mass of the other ) colliding on a highly inclined orbital configuration .
alternatively , much of ngc 5128 could be a red and dead galaxy , passively evolving since its initial formation as a large seed galaxy @xcite , while undergoing later minor mergers and satellite accretions .
evidence consistent with this scenario is in the halo population of stars in ngc 5128 , which have a mean age of @xmath28 gyr @xcite .
its metal - poor gc ages have also been shown to have ages similar to milky way gcs , while the metal - rich population appears younger @xcite .
our new spectroscopic study @xcite suggests that ngc 5128 has a trimodal distribution of cluster ages : @xmath29 of metal - rich clusters have ages of @xmath30 gyr , only a small handful of metal - rich clusters have ages of @xmath31 gyr , and a large fraction of both metal - rich and metal - poor clusters have ages of @xmath32 gyr .
lastly , the halo kinematics of ngc 5128 have also been recently shown to match the surrounding satellite galaxies in the low - density centaurus group , suggesting that ngc 5128 acts like an inner component to its galaxy group @xcite . kinematic and age studies with large number of gcs are thus starting to help disentangle the formation of this giant elliptical .
the confirmed gc population in ngc 5128 is now large enough to allow a new kinematic analysis subdivided to explore radial and metallicity dependence , while avoiding small number statistics in almost all regions of the galaxy .
the analysis presented here complements the detailed age distribution study provided by @xcite .
the results provide a broader picture of the formation scenario of ngc 5128 .
the sections of this paper are divided as follows : [ sec : cat ] contains the full catalog of ngc 5128 gcs with known photometry and radial velocities ,
[ sec : kin ] contains the kinematic analysis of gcss , [ sec : kin_pn ] contains the kinematic analysis of the pn population ,
[ sec : dyn ] contains the discussion of the dynamical mass of ngc 5128 , and [ sec : concl ] contains our final discussion , as well as concluding remarks .
finding gcs in ngc 5128 is challenging .
this process begins with photometric surveys of the many thousands of objects projected onto the ngc 5128 field ( e.g. * ? ? ?
* ; * ? ? ?
* ) ; only a few percent of these are the gcs that we seek .
this daunting task is made difficult by the galactic latitude of ngc 5128 ( @xmath33 ) , which means that many foreground stars are present in the field of ngc 5128 .
background galaxies are another major contaminant in the field , forcing the use of search criteria such as magnitude , color , and object morphology to help build the necessary candidate list of gcs .
confirmation of such candidate objects can then be done with spectroscopic radial velocity measurements .
the gcs in ngc 5128 have radial velocities in the range @xmath34 km
s@xmath7 , while most foreground stars have @xmath35 km s@xmath7 . background galaxies have radial velocities of many thousands of km s@xmath7 and can easily be eliminated ( see the recent studies of * ? ? ?
* ; * ? ? ?
* ) . over the past quarter century
, there have been seven distinct radial velocity studies to identify gcs in ngc 5128 @xcite .
although @xcite was not a radial velocity study itself , but rather a ccd photometric study of previously confirmed gcs , it does include the gcs determined spectroscopically from @xcite .
a recent study with measured gc radial velocities published by @xcite has confirmed two new gcs , hch15 and r122 , included in our catalog .
within this catalog are 80 new gcs with radial velocity measurements from @xcite .
the combination of these studies now leads to a confirmed population of 342 gcs .
km s@xmath7 appears to be an m - type star based on the strong molecular bands in its spectrum , see @xcite for further discussion .
objects pff_gc-010 and 114993 are also rejected as gcs because of their starlike appearance under @xmath19 acs imaging ; see @xcite .
however , newly confirmed gcs hch15 and r122 have now been added @xcite .
] also included in our catalog are the new gc candidates from @xmath19 stis imaging from @xcite , labelled c100-c106 , and from @xmath19 acs imaging from @xcite , labelled c111-c179 .
all these previous studies have their own internal numbering systems , which makes the cluster identifications somewhat confusing at this point . here
we define a new , homogeneous listing combining all this material and with a single numbering system .
our catalog of the gcs of ngc 5128 is given in table [ tab : cat_gc ] .
in successive columns , the table gives the new cluster name in order of increasing right ascension ; the previous names of the cluster in the literature ; right ascension and declination ( j2000 ) ; the projected radius from the center of ngc 5128 in arcminutes ; the @xmath36 , @xmath37 , @xmath38 , @xmath39 , and @xmath40 photometric indices and their measured uncertainties ; the @xmath41 , @xmath42 , and @xmath43 photometric indices and their uncertainties ; the colors @xmath44 , @xmath45 , @xmath46 , @xmath47 , @xmath48 , @xmath49 , and @xmath50 ; and , lastly , the weighted mean velocity @xmath51 and its associated uncertainty from all previous studies .
all @xmath52 photometry is from the imaging survey described in @xcite .
the @xmath53 data are from @xcite .
the mean velocities are weighted averages with weights on each individual measurement equal to @xmath54 where @xmath55 is the quoted velocity uncertainty from each study .
the uncertainty in the mean velocity is then @xmath56 .
there are no individual uncertainties supplied for the velocities for clusters studied by @xcite , but their study reports that the mean velocity uncertainty for clusters with @xmath57 is 25 km s@xmath7 and for @xmath58 is 44 km s@xmath7 .
we have adopted these values accordingly for their clusters . , but more recently @xcite claim a distance of @xmath59 , so it has an adopted uncertainty of 44 km s@xmath7 in the weighted mean . ]
the study by @xcite also does not report velocity uncertainties ; however , these clusters have all been recently measured by @xcite .
the rms scatter of the @xcite values from theirs was 58 km s@xmath7 .
this value has been adopted as the velocity uncertainty of the @xcite clusters in the weighted means .
the weighted mean velocity of cluster c10 does not include the measured value determined by @xcite which is significantly different from other measurements .
also , cluster c27 does not include the measurement of @xmath60 km s@xmath7 from @xcite , indicating that this object is a galaxy .
we include c27 as a gc , but with caution . in the weighted velocity calculations ,
the velocities and uncertainties of the 27 gcs from @xcite have been rounded to the nearest whole number , with any velocity uncertainty below 1 km s@xmath7 rounded up to a value of 1 .
lastly , the gc pff_gc-089 overlaps the previously existing confirmed cluster , c49 , within a 0.5 radius ; pff_gc-089 is therefore removed from the catalog of confirmed gcs .
the data in table [ tab : cat_gc ] provide the basis for the kinematic study presented in this paper .
we use them to derive the rotation amplitude , rotation axis and velocity dispersion in the full catalog of clusters , as well as for the metal - poor ( [ fe / h ] @xmath61 ) and metal - rich ( [ fe / h ] @xmath62 ) subpopulations . for this purpose
, we define the metallicity of the gcs by transforming the dereddened colors @xmath63 to [ fe / h ] through the standard conversion @xcite , calibrated through milky way cluster data . a foreground reddening value of @xmath64 for ngc 5128 , corresponding to @xmath65 , has been adopted .
the division of [ fe / h ] = -1 between metal - rich and metal - poor gcs has been shown as a good split between the two metallicity populations from [ fe / h ] values converted from @xmath66 in @xcite and @xcite for ngc 5128 .
if no @xmath41 and/or @xmath43 values are available for the cluster , it is classified as metal - rich or metal - poor through a transformation from @xmath67 to [ fe / h ] from @xcite . in figure
[ fig : position ] we show the spatial distributions of all the gcs from table [ tab : cat_gc ] ( _ left _ ) and the distribution of the known pne ( _ right _ ) . both systems are spatially biased to the major axis of the galaxy because this is where most of the gc and pn searches have concentrated .
for the present discussion we adopt a distance of @xmath68 mpc for ngc 5128 .
this value is based on four stellar standard candles that each have internal precisions near @xmath69 mag : the pn luminosity function , the tip of the old - halo red giant branch , the long - period variables , and the cepheids @xcite .
hch15 and r122 have not been included in our kinematic study , as our study was completed before publication of these velocities .
the weighted velocities used in this kinematic study do not include the most recent 25 velocities of previously known gcs published in @xcite .
note that the velocities published in table [ tab : cat_gc ] do , however , include the @xcite velocities in the quoted final weighted radial velocities for completeness .
the velocity distribution of the entire sample of 340 is shown in figure [ fig : gausfit_all ] ( _ top left _ ) , binned in 50 km s@xmath7 intervals .
a fit with a single gaussian yields a mean velocity of @xmath70 km s@xmath7 , nicely matching the known systemic velocity of @xmath71 km s@xmath7 @xcite . there is a slight asymmetry at the low - velocity end that is likely due to contamination by a few metal - poor milky way halo stars ( also seen in the metal - poor subpopulation in the bottom left panel , which has a mean velocity determined by the gaussian fit as @xmath72 km s@xmath7 ) . selecting the clusters with radial velocity uncertainties less than 50 km
s@xmath7 leaves 226 clusters , plotted in fig .
[ fig : gausfit_all ] ( _ top right _ ) .
the close fit to a single gaussian is consistent with an isotropic distribution of orbits ; the mean velocity is @xmath73 km s@xmath7 .
the metal - rich population , with a mean velocity determined by the gaussian fit of @xmath74 km s@xmath7 , is plotted in the bottom right panel , and also shows no strong asymmetries .
looking closer at the metal - poor velocity asymmetry , we note that the 15 metal - poor clusters between 250 and 300 km s@xmath7 ( in the region where contamination by milky way field stars could occur ) are balanced by only two gcs at the high - velocity end on reflection across the systemic velocity .
the same velocity regions in the metal - rich population are nearly equally balanced with four clusters between 250 and 300 km s@xmath7 with three clusters at the reflected high - velocity range .
interestingly , the four metal - rich clusters between 250 and 300 km s@xmath7 have projected radii @xmath75 kpc even though the metal - rich population is more centrally concentrated than the metal - poor ( see * ? ? ?
* ; * ? ? ?
* among others ) .
the metal - poor clusters between 250 and 300 km s@xmath7 , conversely , are more evenly distributed , with five clusters between projected radii of 5 and 10 kpc , five clusters between 10 and 20 kpc , and five clusters beyond 20 kpc from the center of ngc 5128 .
some of these low - velocity , metal - poor objects could be foreground stars with velocities in the realm of gcs in ngc 5128 ( @xmath76 km s@xmath7 )
. however , with only 340 gcs currently confirmed within @xmath77 from the center of ngc 5128 , out of an estimated @xmath78 total clusters within 25 @xcite , these metal - poor , low - velocity objects could simply be part of a very incomplete gc sample that is also spatially biased .
this potential bias is clearly shown in figure [ fig : gc_thetar ] , which shows the projected radial distribution as a function of azimuthal angle for our gc sample .
beyond 12 kpc , the two voids coincide with the photometric minor axis of the galaxy , attributed at least partly to incomplete cluster surveys in these regions .
these objects should , therefore , not be dropped from the gc catalog without further spectroscopic analysis .
we determine the rotation amplitude and axis of the gcs of ngc 5128 from @xmath79 ( see * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) . in equation [ eqn : kin ] , @xmath51 is the observed radial velocity of the gcs in the system , @xmath80 is the galaxy s systemic velocity , @xmath39 is the projected radial distance of each gc from the center of the system assuming a distance of 3.9 mpc to ngc 5128 , and @xmath81 is the projected azimuthal angle of the gc measured in degrees east of north .
the systemic velocity of ngc 5128 is held constant at @xmath82 km s@xmath7 @xcite for all kinematic calculations .
the rotation axis of the gcs , @xmath83 , and the product @xmath84 , the rotation amplitude of the gcs in the system , are the values obtained from the numerical solution .
we use a marquardt - levenberg non - linear fitting routine @xcite .
[ eqn : kin ] assumes spherical symmetry . while this may be a decent assumption for the inner 12 kpc region ( it has a low ellipticity of @xmath85 ; *
* ) , true ellipticities for the outer regions of the system are not well known because of the sample bias ( see fig . [
fig : gc_thetar ] ) .
future studies to remove these biases are vital to obtaining a sound kinematic solution for the entire system .
[ eqn : kin ] also assumes that @xmath86 is only a function of the projected radius and that the rotation axis lies in the plane of the sky .
it is not entirely clear how these assumptions , discussed thoroughly in @xcite , apply to the gc and pn systems of ngc 5128 .
the @xmath86 we solve for is , therefore , only a lower limit to the true @xmath86 if the true rotation axis is not in the plane of the sky . the projected velocity dispersion is also calculated from the normal condition , @xmath87 where @xmath88 is the number of clusters in the sample , @xmath89 is the gc s radial velocity _ after subtraction of the rotational component determined with eqn .
[ eqn : kin ] _ , and @xmath22 is the projected velocity dispersion . the gcs were assigned individual weights in the sums that combine in quadrature the individual observational uncertainty , @xmath90 , in @xmath51 and the random velocity component , @xmath91 , of the gcs .
the dominance of the latter is evident by the large dispersion in the gc velocities in the kinematic fitting ( see figure [ fig : kin_plot ] ) . in other words ,
the clusters have individual weights , @xmath92 ; the main purpose of this is to assign a bit more importance to the clusters with more securely measured velocities .
this random velocity term dominates in nearly every case , leaving the gcs with very similar base weights in the kinematic fitting .
the three kinematic parameters - rotation amplitude , rotation axis , and velocity dispersion - are determined with three different binning methods .
the first involves binning the gcs in radially projected circular annuli from the center of ngc 5128 .
the chosen bins keep a minimum of 15 clusters in each , ranging as high as 124 clusters .
the bins are 0 - 5 , 5 - 10 , 10 - 15 , 15 - 25 , and 25 - 50 kpc .
also , we include 0 - 50 kpc to determine the overall kinematics of the system .
the second method adopts bins with equal numbers of clusters .
the entire population of clusters had nine bins of 38 clusters each , the metal - poor clusters had nine bins of 20 clusters , and the metal - rich clusters had eight bins of 20 clusters .
the base weighting is applied to the clusters in both the first and second binning methods .
the third method uses an exponential weighting function , outlined in @xcite , to generate a smoothed profile .
this method determines each kinematic parameter at the radial position , @xmath39 , of every gc in the entire sample by exponentially weighting all other gcs surrounding that position based on their radial separation , @xmath93 , following @xmath94.\ ] ] in equation [ eqn : exp_wei ] , @xmath95 is the determined weight on each gc in the sample , and @xmath96 is the half - width of the window size . for this study , @xmath96 is incrementally varied in a linear fashion for the total sample from @xmath97 kpc at the radius of the innermost gc in the sample out to @xmath98 kpc at the radius of the outermost gc , where the population is lowest .
the metal - poor population was given a half - width window of @xmath99 kpc , and the metal - rich population was given a half - width window of @xmath100 kpc , again from the innermost to outermost cluster .
the progressive radial increase in @xmath96 ensured that each point @xmath39 had roughly equal total weights .
the kinematic parameters were determined for the entire sample of 340 gcs , as well as the subpopulations of 178 metal - poor and 158 metal - rich gcs ( four clusters have unknown metallicity ) .
the kinematic results for the entire population of gcs are shown in table [ tab : all_gc ] , reproduced almost in full from @xcite , while the results for the metal - poor and metal - rich clusters are shown in tables [ tab : mp_gc ] & [ tab : mr_gc ] , respectively .
the columns give the radial bin , the mean projected radius in the bin , the radius of the outermost cluster , the number of clusters in the bin , the rotation amplitude , the rotation axis , and the velocity dispersion , with associated uncertainties .
these are followed by the mass correction , the pressure - supported mass , the rotationally supported mass , and the total mass in units of solar mass ( see [ sec : dyn ] for the mass discussion ) .
the results for the alternate two methods , using an equal number of gcs per bin and the exponentially weighted gcs , are not shown in tabular form but are included in all of the figures .
figure [ fig : kin_plot ] shows the sine fit of eqn .
[ eqn : kin ] for the total population and for the metal - poor and metal - rich subpopulations .
all three populations show rotation about a similar axis . as discussed in
[ sec : velfield ] , the metal - poor population has more members with low velocities ( @xmath101 km s@xmath7 ) than the metal - rich population , suggesting possible contamination of milky way foreground stars in the sample . figures [ fig : rotamp_final ] & [ fig : rotamp_metal ] show the rotation amplitude results for the entire population and for the metal - poor and metal - rich subpopulations , respectively .
the three kinematic methods , described in section [ sec : math ] , appear to agree relatively well for all three populations of clusters . while there appears to be no extreme difference in rotation amplitude between the cluster populations , the metal - poor subpopulation of clusters has lower rotation in the inner 5 kpc of ngc 5128 than the metal - rich subpopulation .
the weighted average of the 0 - 5 kpc radial bin and the innermost equal - numbered bin , shows that the entire population has a rotation amplitude of @xmath102 km s@xmath7 , while the metal - poor population has @xmath103 km s@xmath7 and the metal - rich population has @xmath104 km s@xmath7 .
@xcite show in their study that the metal - poor population has very little rotation in the central regions , completely consistent with our findings .
the rotation amplitude does not appear to differ between the two populations outside of 5 kpc .
the results of the rotation axis solutions are shown in figures [ fig : rotaxis_final ] & [ fig : rotaxis_metal ] , again for the entire population and for the metal - poor and metal - rich subpopulations .
the solution for @xmath105 agrees well for all three kinematic methods and all subgroups .
the inner 5 kpc region has a different rotation axis than the outer regions , demonstrated clearly in all three binning methods .
the innermost bin yields weighted averages of @xmath106 , @xmath107 , and @xmath108 , all of which are equal within their uncertainties . beyond 5 kpc ,
the rotation axes for all three populations are in even closer agreement , with averages of @xmath109 , @xmath110 , and @xmath111 for the entire population , the metal - poor subpopulation , and the metal - rich subpopulation , respectively .
the position angle of the photometric major axis of ngc 5128 is @xmath112 and @xmath113 east of north and the photometric minor axis is @xmath114 and @xmath115 east of north @xcite .
it appears the gcs is rotating about an axis similar to the photometric major axis for the full extent of the galaxy , with a possible axial twist or counterrotation within 5 kpc .
figures [ fig : veldisp_final ] & [ fig : veldisp_metal ] show the velocity dispersion for the entire population and for the metal - poor and metal - rich subpopulations .
our results for @xmath22 show no significant differences between the metallicity subpopulations . all three show a relatively flat velocity dispersion ( @xmath116 , @xmath117 , and @xmath118 km s@xmath7 within 15 kpc of the center of ngc 5128 for the entire population and for the metal - poor and metal - rich subpopulations , respectively ) .
these results match the previous study of ngc 5128 by @xcite , whose determined velocity dispersion for the gcs within 20 kpc ranged between 75 and 150 km s@xmath7 . at a larger radius , we find that @xmath22 then slowly increases to @xmath119 km s@xmath7 towards the outer regions of the halo for all populations . the velocity dispersion of the metal - rich gcs , interestingly , appears higher than that of the metal - poor gcs in the outer regions ( although still consistent within the determined uncertainties ) . in most previous studies ,
the velocity dispersion of the metal - poor gcs usually appears higher than that of the metal - rich gcs , if there is a notable velocity dispersion difference between the subpopulations ( see the studies of * ? ? ?
* ; * ? ? ?
* as examples ) . to explore the cause of the distinct rise past 15 kpc a bit further , we have plotted the actual velocity histograms in figure [ fig
: vf_histo ] for the metal - poor and metal - rich subgroups , subdivided further into inner ( @xmath120 kpc ) and outer ( @xmath121 kpc ) regions . in the inner 15 kpc , both samples show histograms strongly peaked near @xmath122 and with at least roughly gaussian falloff to both high and low velocities .
by contrast , the histograms for the outer regions ( @xmath123 kpc ) are noticeably flatter , so that the clusters with larger velocity residuals have relatively more importance to the formal value of @xmath22 .
nominally , the flatter shape of the velocity distribution would mean that the outer - halo clusters display anisotropy in the direction of a bias towards more circular orbits .
however , such a conclusion would be premature at this point for two reasons .
first , the sample size in the outer regions is still too small to lead to high significance , and a direct comparison between the inner and outer histograms ( through a kolmogorov - smirnov test ) does not show a statistically significant difference between them larger than the 70% level .
second , the outer samples may still be spatially biased in favor of objects along the major axis of the halo , as discussed above , and this bias sets in strongly for @xmath124 kpc ( see fig . [
fig : gc_thetar ] ) , very near where we have set the radial divisions in this figure .
this type of velocity distribution can also arise from the accretions of satellite galaxies with their own small numbers of gcs @xcite .
we will need to have a larger sample of the outer - halo clusters , and one in which these potential sample biases have been removed , before we can draw any firmer conclusions .
however , it needs to be explicitly stated that the outermost point in the kinematic plots for the gcs representing @xmath125 kpc suffers from very high spatial biases and low number statistics ( @xmath126 gcs ) and covers a large radial interval .
the rise in velocity dispersion could be driven purely by systematic effects resulting from the radial gradient of the number density of gcs in this outermost bin @xcite .
ngc 5128 has a large number , 780 , of identified pne with measured radial velocity from the studies of @xcite and @xcite ; these pne are projected out to 90 kpc assuming a distance to ngc 5128 of 3.9 mpc . since these are also old objects , it is of obvious interest to compare them with the gcs .
the pne also have the advantage of giving us the best available look at the kinematics of the halo field stars .
the pn kinematic results are listed in table [ tab : pn ] , with the same columns as table [ tab : all_gc ] .
the results are shown in figures [ fig : rotamp_final ] , [ fig : rotaxis_final ] , & [ fig : veldisp_final ] for the rotation amplitude , rotation axis , and velocity dispersion , respectively .
the spatial distribution of the known pne is , like the gcs , biased toward the major axis at large radii ( see fig . [
fig : position ] ) .
nevertheless , their kinematics closely resemble the gcs .
the kinematics of the pn system are very consistent among all three binning methods .
the rotation amplitude and rotation axis show little radial trend , while the velocity dispersion appears relatively flat within the first 15 kpc at @xmath127 km s@xmath7 and then slowly _ decreases _ to @xmath128 km s@xmath7 at large galactocentric radius .
@xcite show that the velocity dispersion of the pne drops from a central value of 140 to 75 km s@xmath7 in the outer regions of the galaxy , consistent with the findings of this study .
their velocity field analysis led to the discovery of a
zero - velocity curve located between the photometric minor axis , @xmath129 east of north @xcite , and the north - south direction , for the innermost region of the galaxy . just beyond 5 kpc ,
the zero - velocity curve turns and follows a straight line at a @xmath130 angle from the photometric major axis , @xmath131 east of north @xcite ( see figure 7 of * ? ? ?
this study does not show a strong change in rotation axis for the pne in the innermost regions of the galaxy .
however , it clearly shows in all three gc populations a significant change in the rotation axis just beyond 5 kpc from the center of the galaxy . a change in axis of 5 kpc
outward ( @xmath10 for the entire population and metal - poor subpopulation and @xmath132 for the metal - rich subpopulation ) has been found , as discussed in
[ sec : rotaxis ] .
similarly , @xcite show from their sample of 215 clusters that a clear sign of rotation beyond 5 kpc about a misaligned axis appears particularly in their metal - rich subpopulation .
the kinematics of the gcs in this study matches the line of zero velocity relatively nicely . within 5 kpc
the rotation axis of the gcs is nearly - parallel to the north - south direction , and beyond 5 kpc the rotation axis is near @xmath133 east of north , which is only @xmath134 from the zero - velocity curve .
however , the velocity field of ngc 5128 is complex and not entirely captured by these approximate solutions . the two - dimensional velocity field shown in @xcite ( see their figure 7 ) , shows that the photometric major axis ( which happens to be very close to our maximum rotation as discussed above ) is only @xmath130 from the line of zero velocity .
this could lead to a very asymmetric velocity profile that may not be well fit by the sine curve described in equation [ eqn : kin ] .
biased kinematics , especially the rotation axis , may develop from the sine fit that could lead to a higher estimated velocity dispersion .
@xcite similarly studied the kinematics of the pn system in ngc 5128 with a sample of 433 pne .
they obtain a rotation axis of @xmath135 .
our result of @xmath136 east of north is consistent with their findings on comparing their sine curve fit of their pn data in their figure 11 to our corresponding fit shown in fig .
[ fig : kin_plot ] for the gcs , which shares a similar axis to our pn sample ( note that in their study , @xmath137 corresponds to our @xmath138 east of north ) .
our fits both correspond to a positive rotation amplitude for a rotation axis near @xmath139 east of north and a negative rotation axis near @xmath140 east of north .
therefore , the rotation axis quoted in @xcite of @xmath135 corresponds to a _ negative _ rotation amplitude of approximately @xmath141 km s@xmath7 ( taken from their figure 11 ) , nicely matching our result of a positive @xmath14 km s@xmath7 about an axis of @xmath136 east of north .
both gcs ( * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* among others ) and pne ( * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* among others ) can be used to estimate the total dynamical mass of their host galaxies .
a variety of tools are in use including derived mass models , the virial mass estimator @xcite , the projected mass estimator @xcite , and the tracer mass estimator @xcite .
ngc 5128 does not have a large x - ray halo ( detected by * ? ? ?
* ; * ? ? ?
* the latter reporting a measurement of log @xmath142 erg s@xmath7 ) , such as is evident in other giant ellipticals such as m87 @xcite or ngc 4649 @xcite .
thus it is difficult to model the dark matter profile of ngc 5128 with _ a priori _ constraints .
without such a mass model , we turn to the tracer mass estimator for the dynamical mass determination .
the tracer mass estimator has the distinct advantage over the virial and projected mass estimators that the tracer population does not have to follow the dark matter density in the galaxy - an extremely useful feature for stellar subsystems such as gcs and pne that might , in principle , have significantly different radial distributions ( see @xcite for extensive discussion ) .
below , we determine the mass of ngc 5128 using the tracer populations of gcs and pne ( our mass estimates do not include stellar kinematics in the inner regions ) .
the mass contributed by the random internal motion of the galaxy ( pressure - supported mass ) is determined from the tracer mass estimator as @xmath143 where @xmath88 is the number of objects in the sample and @xmath89 is the radial velocity of the tracer object _ with the rotation component removed_. for an isotropic population of tracer objects , assumed in this study , the value of @xmath41 is @xmath144 where @xmath145 and @xmath146 are the three - dimensional radii corresponding to the two - dimensional projected radii of the innermost , @xmath147 , and outermost , @xmath148 , tracers in the sample .
the parameter @xmath149 is set to zero for an isothermal halo potential in which the system has a flat rotation curve at large distances . finally , @xmath150 is the slope of the volume density distribution , which goes as @xmath151 , and is found by determining the surface density slope of the sample and deprojecting the slope to three - dimensions .
the tracer mass estimator uses a sample of tracer objects defined between @xmath145 and @xmath146 , yet it is important to emphasize that it determines the _ total _ enclosed mass within @xmath146 .
there is also a contribution to the total mass by the rotational component , as determined in
[ sec : rotamp ] for the gcs and [ sec : kin_pn ] for the pne .
this mass component is determined from the rotational component of the jeans equation , @xmath152 where @xmath148 is the outermost tracer projected radius in the sample and @xmath153 is the rotation amplitude .
therefore , the total mass of ngc 5128 , @xmath154 , is determined by the addition of the mass components supported by rotation , @xmath155 , and random internal motion , @xmath156 , @xmath157 in the determination of the pressure - supported mass , one must estimate values for @xmath145 and @xmath146 knowing @xmath147 and @xmath148 .
@xcite suggest that @xmath158 and @xmath159 for distributions taken over a wide angle .
however , in this study the inner and outer radii of the chosen bins are at intermediate radial values within the distribution . their assumption would therefore lead to an underestimate of the determined mass , since the true @xmath146 can be quite a bit larger than the projected @xmath148 . to correct for this
contributed uncertainty , distributions of sample tracer populations were generated through monte carlo simulations . in the simulations ,
340 gcs were randomly placed in a spherically symmetric system extending out to 50 kpc with an @xmath160 projected density , while 780 pne were placed in the same environment extending out to 90 kpc . from the generated distributions , the value of @xmath41 in eqn .
[ eqn : c ] was determined for both the real and projected positions of the tracer populations in each designated radial bin .
this correction factor , listed in tables [ tab : mp_gc]-[tab : pn ] as @xmath161 , multiplies the pressure - supported mass from eqn .
[ eqn : tme ] .
the same correction was applied to the full gc sample and the corresponding @xmath161 values are listed in table 1 of @xcite .
these values are generally small , but in the worst case they triple @xmath156 . in eqn .
[ eqn : c ] , the value of @xmath150 is determined for the tracer populations by deprojecting the slope of the surface density profile to three - dimensions .
figure [ fig : gc_surden ] shows the surface density profiles for the entire , metal - poor , and metal - rich gc populations , along with the pn profile .
the populations were binned , following @xcite , into circular annuli of equal numbers of objects , providing the same statistical weight to each bin ( although spatial biases may still affect the gc population in the outer regions along the major axis ; see fig .
[ fig : gc_thetar ] ) . in the inner 5 kpc of all tracer populations ,
incompleteness due to the obscuration of the dust lane is evident by the flattening of the surface density profile .
the innermost objects were , therefore , excluded from the surface density profile fittings .
outside of 5 kpc , the surface densities fit well to power laws , leading to @xmath162 , @xmath163 , @xmath164 , and @xmath165 for the entire gc population , the metal - poor and metal - rich subpopulations of gcs , and the pne in ngc 5128 , respectively .
these are all very similar within their uncertainties .
the similar kinematics we find between the metal - poor and metal - rich subpopulations of gcs in this study strongly justifies the combining of the two populations for the mass determination performed in @xcite .
the gc population provides a _ total _ mass estimate of @xmath166 @xmath17 from 340 clusters out to a projected radius of 50 kpc . removing the gcs in our sample with @xmath167 km s@xmath7 , which will remove all possible contamination from foreground stars , discussed in
[ sec : velfield ] , leads to a total mass of @xmath168 @xmath17 .
this mass agrees nicely with our mass determined from our entire gc sample .
the pn population provides a total mass of @xmath169 @xmath17 from 780 pne out to 90 kpc in projected radius , agreeing with the gc value within the uncertainty .
we are also able to generate a mass profile of ngc 5128 from the total gc population and the pne , shown in figure [ fig : mass ] .
the tracer mass estimator determines the _ total _ enclosed mass for ngc 5128 within the outermost radius of a given tracer sample .
it calculates this total mass using a sample of objects defined within the radial range defined by the sample s inner and outermost radii .
it is therefore possible to use a unique set of tracer objects , denoted by the radial bin range , listed in the first column of tables [ tab : all_gc]-[tab : pn ] , to determine a mass profile from independent mass estimates .
the independent binning , leads to sample sizes in the mass determination , in some cases generating higher uncertainties in the total enclosed mass .
the most certain mass is the one determined from the full sample of tracers . in the mass determinations above ,
we have implicity assumed isotropy for the velocity distributions .
but the possibility exists that the pne ( for example ) might have radial anisotropy which would produce their gradually falling @xmath170 curve .
replacing equation [ eqn : c ] in the tracer mass estimator by @xmath171 which includes the anisotropy parameter , @xmath172 , from @xcite , we find that the mass estimate from the pne can be forced to agree with the mass estimate from the gcs for a nominal @xmath173 . for perfect isotropy , @xmath174 .
this would mean roughly 2:1 radial anisotropy for the pne in the outer halo .
however , we find that any @xmath172 in the wide range of @xmath175 would still keep the two methods in agreement within their internal uncertainties , so we are not yet in a position to tightly constrain any anisotropy .
it is possible that the gcs may also have anisotropy ; it may therefore be too simplistic to find a range of @xmath172 for the pne for which the masses of the pne and gcs agree . however ,
the gcs are likelier to be nearly isotropic than the pne ; the gcs are older , hotter subsystems of the halo .
in other studies , the isotropy of the gcs orbits has also been shown to be a good assumption from mass profiles of elliptical galaxies using x - ray observations ( * ? ? ? * ; * ? ? ? * ; * ? ? ? * among others ) .
both mass estimates can be compared to previous studies .
first , we note the total mass determined from the pn data with that of @xcite .
while the rotationally supported mass was determined here with different values of the mean rotational velocity , they calculated the pressure supported mass using the identical tracer mass estimator technique with exactly the same pn population . the total mass estimate given by @xcite
is @xmath176 @xmath17 . subtracting their rotationally
supported mass leaves a pressure supported mass of @xmath177 @xmath17 , quite different from our @xmath178 @xmath17 . recalculating our pressure supported mass estimate with @xmath179 , which was used in @xcite , we are able to reproduce their mass estimate within the uncertainty .
the values of @xmath150 differ between the two studies simply because the @xmath150 used in @xcite was the inverse of the surface density slope instead of the inverse of the volume density slope .
using the correct value of @xmath180 , their pressure supported mass estimate would increase to @xmath181 , matching the mass found in this study .
second , we compare our total mass determined using the gc population with that from @xcite . using 215 gcs out to 40 kpc , they found a pressuresupported mass of @xmath182 @xmath17 , again much different from our pressure supported mass of @xmath183 @xmath17 using 340 clusters out to 50 kpc .
the large difference can again be attributed to their using @xmath184 instead of deprojecting their surface density slope to @xmath185 . using the correct value of @xmath150
, we find a pressure - supported mass of @xmath186 @xmath17 using the same 215 clusters they used in their study . this corrected estimate is closer to the pressure supported mass determined in our study , but it is not necessarily expected to agree with our result , as our sample contains @xmath187 more gcs and uses a slightly different @xmath150 that we have independently redetermined .
third , the mass determined by the h @xmath188 shell study of ngc 5128 by @xcite found a mass of @xmath189 @xmath17 _ within 15 kpc _ assuming a distance of 3.5 mpc . with the distance of 3.9 mpc used in this study , the mass determined in their study would increase to @xmath190 @xmath17 , which is @xmath191 smaller than our total mass of @xmath192 @xmath17 within 15 kpc .
lastly , we compare our determined mass to a recent study by @xcite .
@xcite determined a total mass of ngc 5128 using gcs , pne , and an x - ray data technique .
the galaxy mass determined from the gc and pn data was obtained using the tracer mass estimator and the spherical jeans equation , as performed in our study .
however , @xcite used the volume density slopes determined by @xcite and @xcite , for the pn and gc data , respectively .
they obtained mass estimates for ngc 5128 similar to those of @xcite and @xcite , discussed above , using an identical pn sample and slightly increased gc sample .
they also included an x - ray - modelling mass estimate for ngc 5128 from which they obtained masses of @xmath193 @xmath17 out to 50 and @xmath194 @xmath17 out to 80. this mass estimate is similar to our pne estimate out to the same radial extent , but the author cautions that it is an overestimate of the true mass of ngc 5128 resulting from a lack of hydrostatic equilibrium in the outer region of the galaxy .
we note here that the mass estimates obtained are higher than those from @xcite , and @xcite derived from the pne using a two - component mass model , as well as @xcite , using an x - ray modelling technique .
this discrepancy is not fully understood and possibly lies in the assumptions that go into the mass estimators with a spatially biased sample .
we intend to pursue this issue further with an upcoming larger sample of gcs with less spatial biases .
the mass of ngc 5128 that we find appears to be in the range of other giant elliptical galaxies , such as ngc 1399 ( @xmath195 @xmath17 out to 50 kpc ; * ? ? ?
* ) , m49 ( @xmath195 @xmath17 out to their kinematically studied radius of 35 kpc ; * ? ? ?
* ) , and m87 ( @xmath196 @xmath17 at 20 kpc , the onset of the projected cd envelope ; * ? ? ?
clearly , it is legitimate to say that ngc 5128 is the largest , most massive galaxy in the neighborhood of the local group , and one that can be talked about in the same category as these other giants that reside in larger clusters .
the sample biases mentioned above in our currently available set of both gcs and pne place limitations on how much we can reasonably interpret the kinematic and dynamic data .
we are currently carrying out a set of new spectroscopic programs to increase the tracer sample size and to remove the sample biases , leading to a more complete analysis of the halo velocity field .
angular momentum is an essential quantity for characterizing the sizes , shapes , and formation of galaxies and is often represented as the dimensionless spin parameter , where @xmath198 is the angular momentum , @xmath199 is the binding energy , and @xmath42 is the mass of the galaxy .
the spin parameter is representative of a galaxy s angular momentum compared to the amount of angular momentum needed for pure rotational support : the lower the @xmath200-value , the less rotation and rotational support within the galaxy . for an elliptical galaxy in gravitational equilibrium , the spin parameter simplifies to @xmath201 @xcite , yielding @xmath202 with @xmath203 for the entire population of gcs in ngc 5128 .
table [ tab : spin ] shows the spin parameter for four giant galaxies with large gcs kinematic studies , m87 , m49 , ngc 1399 , and ngc 5128 .
the table columns give the galaxy name , the rotation amplitude , the projected velocity dispersion , and the ratio of the rotation amplitude to the velocity dispersion , followed by the spin parameter .
these quantities are shown for the metal - poor and metal - rich populations .
what is clearly evident in table [ tab : spin ] is the strong galaxy to galaxy differences between these four galaxies , already hinted at in [ sec : intro ] . though the sample is still quite small , no obvious pattern emerges .
there is an indication of metal - poor and metal - rich gcss having similar spin parameters within the same galaxy .
m49 is the only galaxy studied here where this may not be the case .
although the metal - poor and metal - rich cluster spin parameters are consistent within the uncertainties , the metal - rich cluster spin parameter of m49 is also consistent with zero . in the monolithic collapse scenario
, @xcite describes the angular momentum within the galaxy as attributed to the tidal torque transferred from neighbouring proto - galaxies during formation . in this scenario
, @xcite found that a spin parameter of @xmath204 for elliptical systems is expected from simulations of the collapse of an isolated protogalactic cloud . but
ngc 5128 , among many other giant elliptical galaxies , is not in isolation , and therefore not necessarily expected to reproduce such a low spin parameter .
also , the internal rotation axis changes at 5 kpc are not easily explained with only the monolithic collapse scenario . in the monolithic collapse model
, the inner regions would be expected to have more pronounced rotations . yet
all four of the galaxies with major kinematic studies presented here do not show a higher rotational signal in the inner regions .
in fact , for ngc 1399 , the outer region ( @xmath205 ) indicates rotation in the metal - poor population that is not evident in the inner regions . also , a slightly lower rotational signal is present in the inner regions of ngc 5128 for the metal - poor population than in the outer regions .
hierarchical clustering of cold dark matter also relies on angular momentum in a galaxy being produced by gravitational tidal torques during the growth of initial perturbations .
@xcite have demonstrated that the tidal torque theory predicts an increase in angular momentum during the collapse , and with time , the increase in angular momentum slows .
accretion of satellites and/or merger events is therefore a possible culprit for moving the angular momentum outward , as major mergers of disks and bulges suggest that angular momentum resides largely in the outer regions of the galaxy @xcite . alternatively ,
@xcite examine the change in spin parameter in a scenario where the angular momentum in a galaxy is built up by mass accretion .
their results show that the spin parameter changes sharply in major merger events in the galaxy and steadily decreases with small satellite accretion events .
they also show that the spin parameter for a galaxy with a major merger after a redshift of @xmath206 should be notably larger than a galaxy that did not undergo such a major merger .
their study obtains an average of @xmath207 from @xmath200cdm @xmath88-body simulations for galaxies with halo masses of @xmath208 @xmath17 with @xmath209 .
ngc 1399 and m49 , with their weak rotation signals , are consistent with the model predictions discussed above , whereby their major formation events could have occurred at early times and with perhaps only minor accretions happening since then . however , ngc 5128 and m87 have spin parameters 2 - 3 times larger than predicted by the model averages . for ngc 5128 , this relatively large rotation ( which is nearly independent of both metallicity and radius )
may , perhaps , be connected with its history within the centaurus group environment .
the rotation speed and rotation axis for its extended group of satellite galaxies are nearly identical to the ngc 5128 halo @xcite , much as if the accretion events experienced by the central giant have been taking place preferentially along the main axis of the entire group and in the same orientation .
the gcs age distribution discussed by @xcite and the mean age for the halo field stars @xcite strongly suggest that a high fraction of the stellar population in ngc 5128 formed long ago , with particularly large bursts between 8 and 12 gyr .
even if the galaxy underwent a significant merger perhaps a few gyr ago ( the traces of which now appear in the halo arcs and shells ) , the stars in it may already have been old at the time of the merger .
although a very few younger gcs have formed since then , these make up a small minority of what is present , at least for the @xmath210 kpc halo outside the bulge region that now contains gas and dust .
the situation for m87 , with its even larger rotation signature , may require a different sort of individual history .
of the four galaxies compared here , it is at the dynamical center of the richest environment ( virgo ) , has the most extensive cd - type envelope , and sits within the most massive , extended , and dynamically evolved potential well .
a single relatively recent major merger could in principle have caused its present high rotation , but the lack of distinctive tidal features does not necessarily favor such an interpretation and would at least suggest that such a merger should have been with another large elliptical and nondissipative galaxy . @xcite
discuss an interpretation - at least partially resembling what we suggest for ngc 5128 - that stellar material is gradually infalling onto m87 along the so - called principal axis of the virgo cluster .
in conclusion , we have presented a kinematic study of ngc 5128 that makes it now comparable to recent studies of the other giant galaxies , m87 , m49 , and ngc 1399 . using @xmath0 gcs ( @xmath211 metal - rich and @xmath212 metal - poor gcs )
, we have calculated the rotation amplitude , rotation axis , and velocity dispersion and have searched for radial and metallicity dependences .
our findings show that both metallicity populations rotate with little dependence on projected radius , with @xmath213 , 31@xmath21414 , and 47@xmath21415 km s@xmath7 for the total , metal - poor , and metal - rich populations , respectively .
perhaps the inner 5 kpc shows a slower rotation of the metal - poor population , but more clusters would be needed to confirm this finding .
the rotation axis is 189@xmath21412@xmath215 , 177@xmath21422@xmath215 , and 202@xmath21415@xmath215 east of north for the total , metal - poor , and metal - rich populations out to a 50 kpc projected radius , assuming the velocity field is best fit by a sine curve .
the rotation axis does change at 5 kpc , following the zero - velocity curve proposed by @xcite or possibly full - on counterrotation . a study with more gcs and lower uncertainties
is needed to see what is happening in the innermost 5 kpc of ngc 5128 .
the velocity dispersion shows a modest increase with galactocentric radius , although the outer regions ( especially the metal - rich population ) have less reliable statistics ; this increase could be driven purely by statistical effects .
we find the velocity dispersion we find 123@xmath2145 , 117@xmath2147 , and 129@xmath2149 km s@xmath7 for the total , metal - poor , and metal - rich populations , respectively .
the pn data are also used to determine the kinematics of the halo of ngc 5128 .
these show results that are encouragingly similar to those of the gc data , except that no rotation axis change is noted with radius , and a _ decrease _ in velocity dispersion is found with radius , possibly indicating a difference in orbital anisotropy compared with the gcs .
a very similar effect has been noted for the leo elliptical ngc 3379 , although with a much smaller data sample @xcite .
we also determine the total dynamical mass using both the gcs and the pne by separately calculating the pressure supported mass with the tracer mass estimator and the rotationally supported mass using the spherical component of the jeans equation .
the total mass is @xmath166 @xmath17 from the gc population out to a projected radius of 50 kpc , or @xmath169 @xmath17 out to 90 kpc from the pne .
overall , we have enough evidence to cautiously conclude that a major episode of star formation occurred about @xmath216 gyr ago ( corresponding to a redshift z = 1.2 - 1.8 ) and this may have been when the bulk of the visible galaxy was built .
we still do not know just why the most metal - poor clusters show up in such relatively large numbers and appear to have ages of @xmath217 gyr , but this is a common issue in all big galaxies . this kinematic study and the age study of @xcite on the ngc 5128 cluster system
indicate that additional spectroscopic studies to build up both the radial velocity database and age distribution can lead to rich dividends .
large gc samples are clearly needed to remove the current sample biases and to fully understand the complex kinematics and history of this giant elliptical galaxy .
it seems clear as well that each galaxy needs to be individually studied to fully understand the different galaxy formation histories .
we are continuing these studies particularly for ngc 5128 , with the eventual aim of at least doubling the total gc sample size in this unique system .
arnaboldi , m. , freeman , k.c . ,
capaccioli , m. , & ford , h. 1994 , eso messanger , 76 , 40 arnaboldi , m. , freeman , k.c . , gerhard , o. , matthias , m. , kudritzki , r.p .
mndez , r.h . ,
capaccioli , m. , & ford , h. 1998 , , 507 , 759 baade , w. , & minkowski , r. 1954 , , 119 , 215 bahcall , j.n . , & tremaine , s. 1981 , , 244 , 805 barnes , j.e .
1992 , , 393 , 484 beasley , m.a . , forbes , d.a . , brodie , j.p . , & kissler - patig , m. 2004 , , 347 , 1150 beasley , m.a . , bridges , t. , peng , e. , harris , w.e .
, harris , g.l.h . , forbes , d.a . , & mackie , g. 2006 , in preparation bekki , k. , harris , w.e . , & harris , g.l.h .
2003 , , 338 , 587 bekki , k. , forbes , d.a . ,
beasley , m.a . , & couch , w.j . 2003 , , 334 , 1334 bekki , k. , & peng , e.w . 2006 , , 370 , 1737 bergond , g. , zepf , s.e . , romanowsky , a.j . , sharples , r.m . ,
& rhode , k.l .
2006 , , 448 , 155 bridges , t. , gebhardt , k. , sharples , r. , faifer , f.r . , forte , j.c . ,
beasley , m.a .
, zepf , z.e . , forbes , d.a . ,
pierce , m. 2006 , , 373 , 157 ciardullo , r. , jacoby , g.h . , &
dejonghe , h.b .
1993 , , 414 , 454 ct , p. , mclaughlin , d.e .
, hanes , d.a . , bridges , t.j . , geisler , d. , merritt , d. , hesser , j.e . ,
harris , g.l.h . , & lee , m.g .
2001 , , 559 , 828 ct , p. , mclaughlin , d. , cohen , j.g .
, & blakesless , j.p .
2003 , , 591 , 850 dekel , a. , stoehr , f. , mamon , g.a . ,
cox , t.j . ,
novak , g.s .
, & primak , j.r . , 2005 , nature , 437 , 707 dufour , r.j . , harvel , c.a . , martins , d.m . , schiffer , f.h . , talent , d.l . , wells , d.c
. , van den bergh , s. , & talbot , r.j .
1979 , , 84 , 284 efstathiou , g. , & jones , b.j.t .
1979 , , 186 , 133 evans , n.w . ,
wilkinson , m.i .
, perrett , k.m . , & bridges , t.j .
2003 , , 583 , 752 fall , s.m . 1979 , rev .
phys . , 51 , 21 ferrarese , l. , mould , j.r . ,
stetson , p.b .
, tonry , j.l . ,
blakeslee , j.p . , & ajhar , e.a .
2006 , apj , in press ( astro - ph/0605707 ) harris , g.l.h . , geisler , d. , harris , h.c . , & hesser , j.e .
1992 , , 104 , 613 harris , g.l.h . , harris , w.e . , & poole , g. 1999 , aj , 117 , 855 harris , w. e. , & harris , g. l. h. 2002 , , 123 , 3108 harris , w.e . ,
harris , g.l.h . ,
holland , s.t . , & mclaughlin , d.e .
2002 , , 124 , 1435 harris , g.l.h . ,
harris , w.e .
, & geisler , d. 2004 , , 128 , 723 harris , w.e . ,
harris , g.l.h . ,
barmby , p. , mclaughlin , d.e . , & forbes , d. 2006 , , 132 , 2187 heisler , j. , tremaine , s. , & bahcall , j.n .
1985 , , 298 , 8 hernquist , i. 1993 , , 409 , 548 hesser , j.e . ,
harris , h.c .
, van den bergh , s. , & harris , g.l.h .
1984 , , 276,491 hesser , j.e . ,
harris , h.c .
, & harris , g.l.h .
1986 , , 303 , l51 hui , x. , ford , h.c . , freeman , k.c . , & dopita , m.a .
1995 , , 449 , 592 kraft , r.p . , vquez , s.e . ,
forman , w.r . ,
jones , c. , murray , s.s . , hardcastle , m.j . ,
worrall , d.m .
, churazov , e. 2003 , , 592 , 129 larsen , s.s . , brodie , j.p . ,
beasley , m.a .
, & forbes , d.a .
2002 , , 124 , 828 lee , h. , & worthey , g. 2005 , , 160 , 176 maz apellniz , j. & beda , l. 2005 , , 629 , 7873 mamon , g.a . , &
okas , e.l .
2005 , , 362 , 95 napolitano , n.r . ,
arnaboldi , m. , freeman , k.c . , & capaccioli , m. 2001 , , 377 , 784 napolitano , n.r . ,
arnaboldi , m. , & capaccioli , m. 2002 , , 383 , 791 osullivan , e. , forbes , d.a . , & ponman , t.j .
2001 , , 328 , 461 peebles , p.j.e .
1969 , , 155 , 393 peng , e.w . ,
ford , h.c . ,
freeman , k.c .
, & white , r.l .
2002 , , 124 , 3144 peng , e.w . ,
ford , h.c . , & freeman , k.c .
2004 , , 150 , 367 peng , e.w . ,
ford , h.c . , & freeman , k. 2004 , , 602 , 685 peng , e.w . ,
ford , h.c . , & freeman , k. 2004 , , 602 , 705 pierce , m. , beasley , m.a . , forbes , d.a . , bridges , t. , gebhardt , k. , faifer , f.r . , forte , j.c . , zepf , s.e . , sharples , r. , hanes , d.a . ,
& proctor , r. 2006 , , 366 , 1253 press , w.h . ,
teukolsky , s.a . , vetterling , w.t . , & flannery , b.p .
1992 , numerical recipes in fortran ( 2nd ed .
, cambridge : camdridge university press ) puzia , t.h .
, kissler - patig , m. , thomas , d. , maraston , c. , saglia , r.p . ,
bender , r. , goudfrooij , p. , & hempel , m. 2005 , , 439 , 997 reed , l.g . ,
harris , g.l.h . , & harris , w.e .
1994 , , 107 , 555 rejkuba , m. 2001 , , 269 , 812 rejkuba , m. 2004 , , 413 , 903 rejkuba , m. , greggio , l. , harris , w.e . , harris , g.l.h . , & peng , e.w .
2005 , , 631 , 262 rejkuba , m. , dubath , p. , minniti , d. , & meylan , g. 2007 , , in press ( astro - ph/0703385 ) richtler , t. , dirsch , b. , gebhardt , k. , geisler , d. , hilker , m. , alonso , m. v. , forte , j. c , . and grebel , e. k. , infante , l. , larsen , s. , minniti , d. , & rejkuba , m. 2004 , , 127 , 2094 romanowsky , a.j . , douglas , n.g . ,
arnaboldi , m. , kuijken , k. , merrifield , m.r . ,
napolitano , n.r . ,
capaccioli , m. , & freeman , k.c . , 2003 , science , 301 , 1696 saglia , r.p . , kronawitter , a. , gerhard , o. , & bender , r. 2000 , , 119 , 153 samurovi , s. 2006 , serb . astron . j. , 173 , 35 schiminovich , d. , van gorkom , j.h . , van der hulst , j.m . , & kasow , s. 1994 , , 423 , l101 sharples , r. 1988 , iau symp . 126 : the harlow - sharpley symposium on globular cluster systems in galaxies , 126 , 245 strader , j. , brodie , j.p . , cenarro , a.j . , beasley , m.a . , & forbes , d.a .
2005 , , 130 , 1315 sugerman , b. , summers , f.j . , & kamionkowski , m. 2000 , , 311 , 762 thomas , d. , maraston , c. , & korn , a. 2004 , , 351 , 19 toomre , a. , & toomre , j. 1972 , , 178 , 623 van den bergh , s. , hesser , j.e .
, & harris , g.l.h .
1981 , , 86 , 24 vitvitska , m. , klypin , a.a . ,
kravtsov , a.v . ,
wechsler , r.h .
, primack , j.r . , & bullock , j.s .
2002 , , 581 , 799 woodley , k.a . ,
harris , w.e . , & harris , g.l.h .
2005 , , 129 , 2654 woodley , k.a .
2006 , , 132 , 2424 gc0001 & pff_gc-028 & 13 20 01.16 & -42 56 51.5 & 6.46 & 20.11 & 19.68 & 18.75 & 18.20 & 17.60 & 0.07 & 0.02 & 0.01 & 0.01 & 0.03 & 19.92 & 19.04 & 18.22 & 0.01 & 0.01 & 0.01 & 0.43 & 0.93 & 0.55 & 1.15 & 0.82 & 0.88 & 1.70 & 524@xmath21416 + gc0002 & hh-048 & 13 22 45.36 & -43 07 08.8 & 30.26 & - & - & - & - & - & - & - & - & - & - & - & - & - & - & - & - & - & - & - & - & - & - & - & 856@xmath21450 + gc0003 & hh-080 & 13 23 38.33 & -42 46 22.8 & 24.84 & - & - & - & - & - & - & - & - & - & - & 19.45 & 18.72 & 17.98 & 0.02 & 0.01 & 0.01 & - & - & - & - & 0.74 & 0.72 & 1.46 & 497@xmath21433 + gc0004 & hghh-40/hh-044/c40 & 13 23 42.37 & -43 09 37.8 & 21.02 & 19.37 & 19.01 & 18.12 & 17.59 & 17.03 & 0.04 & 0.01 & 0.01 & 0.01 & 0.03 & 20.12 & 19.28 & 18.49 & 0.02 & 0.01 & 0.01 & 0.35 & 0.89 & 0.53 & 1.10 & 0.79 & 0.84 & 1.63 & 365@xmath21424 + gc0005 & hghh-01/c1 & 13 23 44.19 & -43 11 11.8 & 21.41 & - & - & - & - & - & - & - & - & - & - & 18.73 & 18.00 & 17.25 & 0.06 & 0.03 & 0.03 & - & - & - & - & 0.76 & 0.72 & 1.48 & 642@xmath2141 + gc0006 & pff_gc-001 & 13 23 49.62 & -43 14 32.0 & 22.36 & 20.39 & 19.91 & 18.91 & 18.36 & 17.60 & 0.09 & 0.02 & 0.01 & 0.01 & 0.03 & 20.22 & 19.25 & 18.38 & 0.02 & 0.01 & 0.01 & 0.47 & 1.01 & 0.55 & 1.31 & 0.87 & 0.97 & 1.84 & 711@xmath21435 + gc0007 & aat301956 & 13 23 54.52 & -43 20 01.1 & 25.41 & 21.22 & 21.13 & 20.42 & 19.98 & 19.43 & 0.18 & 0.06 & 0.02 & 0.02 & 0.04 & - & - & - & - & - & - & 0.09 & 0.71 & 0.44 & 0.98 & - & - & - & 287@xmath214162 + gc0008 & hh-099 & 13 23 56.70 & -42 59 59.8 & 16.66 & - & - & - & - & - & - & - & - & - & - & 21.72 & 20.90 & 20.01 & 0.09 & 0.09 & 0.06 & - & - & - & - & 0.89 & 0.82 & 1.71 & 798@xmath21449 + gc0009 & aat101931 & 13 23 58.58 & -42 57 17.0 & 16.73 & 20.58 & 20.57 & 19.88 & 19.43 & 18.96 & 0.06 & 0.02 & 0.01 & 0.01 & 0.03 & 20.60 & 20.09 & 19.45 & 0.01 & 0.01 & - & 0.02 & 0.68 & 0.45 & 0.92 & 0.64 & 0.51 & 1.15 & 590@xmath214144 + gc0010 & aat101906 & 13 23 58.76 & -43 01 35.2 & 16.25 & 20.45 & 19.90 & 18.89 & 18.28 & 17.57 & 0.06 & 0.02 & 0.01 & 0.01 & 0.03 & 20.23 & 19.11 & 18.28 & 0.02 & 0.03 & 0.02 & 0.54 & 1.01 & 0.61 & 1.32 & 0.83 & 1.12 & 1.95 & 511@xmath21431 + gc0011 & pff_gc-002 & 13 23 59.51 & -43 17 29.1 & 22.94 & 20.70 & 20.43 & 19.55 & 19.02 & 18.44 & 0.07 & 0.02 & 0.01 & 0.01 & 0.03 & 20.61 & 19.83 & 19.07 & 0.03 & 0.03 & 0.01 & 0.27 & 0.88 & 0.53 & 1.11 & 0.76 & 0.78 & 1.55 & 653@xmath21437 + gc0012 & aat102120 & 13 23 59.61 & -42 55 19.4 & 17.11 & 20.60 & 20.59 & 19.92 & 19.49 & 19.03 & 0.06 & 0.02 & 0.01 & 0.01 & 0.03 & 20.61 & 20.13 & 19.49 & 0.01 & 0.01 & 0.01 & 0.02 & 0.67 & 0.44 & 0.89 & 0.64 & 0.48 & 1.12 & 293@xmath21483 + gc0013 & hh-001 & 13 24 02.67 & -42 48 32.2 & 20.00 & - & - & - & - & - & - & - & - & - & - & 22.52 & 21.20 & 19.89 & 0.08 & 0.05 & 0.03 & - & - & - & - & 1.31 & 1.32 & 2.63 & 775@xmath21474 + gc0014 & pff_gc-003 & 13 24 03.23 & -43 28 13.9 & 31.17 & 20.39 & 20.18 & 19.31 & 18.80 & 18.21 & 0.15 & 0.02 & 0.01 & 0.01 & 0.03 & 20.39 & 19.64 & 18.80 & 0.02 & 0.02 & 0.02 & 0.21 & 0.87 & 0.51 & 1.09 & 0.84 & 0.76 & 1.59 & 704@xmath21422 + gc0015 & pff_gc-004 & 13 24 03.74 & -43 35 53.4 & 37.98 & 21.12 & 20.84 & 20.01 & 19.49 & 18.92 & 0.28 & 0.03 & 0.01 & 0.01 & 0.03 & - & - & - & - & - & - & 0.28 & 0.83 & 0.52 & 1.09 & - & - & - & 571@xmath21438 + gc0016 & aat103195 & 13 24 05.98 & -43 03 54.7 & 15.18 & 21.57 & 21.10 & 20.13 & 19.58 & 18.98 & 0.14 & 0.03 & 0.01 & 0.01 & 0.03 & 21.34 & 20.39 & 19.50 & 0.01 & 0.01 & 0.01 & 0.47 & 0.97 & 0.56 & 1.15 & 0.89 & 0.95 & 1.84 & 277@xmath21467 + gc0017 & aat304867 & 13 24 08.62 & -43 16 27.1 & 21.04 & 22.12 & 21.15 & 19.91 & 19.20 & 18.54 & 0.25 & 0.03 & 0.01 & 0.01 & 0.03 & 21.62 & 20.28 & 19.18 & 0.01 & - & - & 0.97 & 1.24 & 0.71 & 1.36 & 1.10 & 1.34 & 2.44 & 305@xmath21456 + gc0018 & aat305341 & 13 24 10.97 & -43 12 52.8 & 18.27 & 22.13 & 21.54 & 20.51 & 19.88 & 19.19 & 0.23 & 0.04 & 0.01 & 0.01 & 0.03 & 21.87 & 20.80 & 19.90 & 0.01 & 0.01 & 0.01 & 0.59 & 1.03 & 0.63 & 1.31 & 0.89 & 1.07 & 1.97 & 440@xmath214118 + gc0019 & hhh86 - 28/c28 & 13 24 18.06 & -42 49 01.1 & 17.57 & 19.52 & 19.33 & 18.50 & 17.99 & 17.45 & 0.03 & 0.01 & 0.01 & 0.01 & 0.03 & 19.46 & 18.75 & 18.03 & 0.02 & 0.01 & 0.01 & 0.19 & 0.83 & 0.52 & 1.05 & 0.72 & 0.71 & 1.43 & 461@xmath21422 + gc0020 & pff_gc-005 & 13 24 18.92 & -43 14 30.1 & 18.33 & 21.89 & 21.31 & 20.29 & 19.71 & 19.07 & 0.18 & 0.04 & 0.01 & 0.01 & 0.03 & 21.63 & 20.64 & 19.72 & 0.02 & 0.02 & 0.01 & 0.58 & 1.02 & 0.58 & 1.22 & 0.91 & 1.00 & 1.91 & 750@xmath21433 + gc0021 & whh-1/hh-096 & 13 24 21.40 & -43 02 36.8 & 12.19 & 19.14 & 18.85 & 18.01 & 17.48 & 16.98 & 0.03 & 0.01 & 0.01 & 0.01 & 0.03 & 19.11 & 18.33 & 17.63 & 0.02 & 0.02 & 0.03 & 0.29 & 0.84 & 0.53 & 1.03 & 0.70 & 0.78 & 1.48 & 583@xmath21429 + gc0022 & pff_gc-006 & 13 24 23.72 & -43 07 52.1 & 13.48 & 20.22 & 20.00 & 19.22 & 18.70 & 18.20 & 0.05 & 0.02 & 0.01 & 0.01 & 0.03 & 20.10 & 19.46 & 18.73 & 0.01 & 0.01 & 0.01 & 0.21 & 0.78 & 0.52 & 1.02 & 0.73 & 0.65 & 1.37 & 644@xmath21428 + gc0023 & whh-2 & 13 24 23.98 & -42 54 10.7 & 13.56 & 20.57 & 20.45 & 19.67 & 19.17 & 18.65 & 0.06 & 0.02 & 0.01 & 0.01 & 0.03 & 20.54 & 19.94 & 19.27 & 0.03 & 0.03 & 0.02 & 0.12 & 0.78 & 0.50 & 1.02 & 0.68 & 0.59 & 1.27 & 582@xmath21481 + gc0024 & pff_gc-007 & 13 24 24.15 & -42 54 20.6 & 13.45 & 21.85 & 21.16 & 20.12 & 19.53 & 18.82 & 0.17 & 0.03 & 0.01 & 0.01 & 0.03 & 21.54 & 20.50 & 19.60 & 0.02 & 0.02 & 0.01 & 0.69 & 1.04 & 0.59 & 1.30 & 0.90 & 1.04 & 1.94 & 617@xmath21425 + gc0025 & aat308432 & 13 24 25.55 & -43 21 35.6 & 23.38 & 20.98 & 20.43 & 19.40 & 18.80 & 18.11 & 0.27 & 0.02 & 0.01 & 0.01 & 0.03 & 20.83 & 19.78 & 18.84 & 0.03 & 0.01 & 0.02 & 0.54 & 1.04 & 0.60 & 1.29 & 0.94 & 1.05 & 1.99 & 835@xmath21483 + gc0026 & c111 & 13 24 26.97 & -43 17 20.0 & 19.62 & - & - & - & - & - & - & - & - & - & - & 22.54 & 22.25 & 21.36 & 0.04 & 0.04 & 0.04 & - & - & - & - & 0.89 & 0.29 & 1.17 & - + gc0027 & aat106695 & 13 24 28.18 & -42 53 04.3 & 13.54 & 20.95 & 20.75 & 19.63 & 18.96 & 18.30 & 0.09 & 0.03 & 0.01 & 0.01 & 0.03 & 21.16 & 20.17 & 19.19 & 0.06 & 0.06 & 0.04 & 0.20 & 1.12 & 0.67 & 1.33 & 0.98 & 0.99 & 1.96 & 835@xmath21483 + gc0028 & aat106880 & 13 24 28.44 & -42 57 52.9 & 11.30 & 21.11 & 20.94 & 20.14 & 19.61 & 19.11 & 0.10 & 0.03 & 0.01 & 0.01 & 0.03 & 21.06 & 20.43 & 19.65 & 0.01 & 0.01 & 0.01 & 0.18 & 0.80 & 0.53 & 1.03 & 0.79 & 0.63 & 1.41 & 558@xmath21498 + gc0029 & pff_gc-008 & 13 24 29.20 & -43 21 56.5 & 23.38 & 21.40 & 20.93 & 19.94 & 19.41 & 18.75 & 0.35 & 0.03 & 0.01 & 0.01 & 0.03 & 21.24 & 20.31 & 19.43 & 0.02 & 0.01 & 0.01 & 0.47 & 0.99 & 0.54 & 1.19 & 0.87 & 0.93 & 1.81 & 466@xmath21438 + gc0030 & aat107060 & 13 24 29.23 & -43 08 36.6 & 13.02 & 21.81 & 21.26 & 20.26 & 19.63 & 19.00 & 0.19 & 0.04 & 0.01 & 0.01 & 0.03 & 21.60 & 20.62 & 19.69 & 0.01 & 0.01 & 0.01 & 0.55 & 1.01 & 0.63 & 1.25 & 0.93 & 0.99 & 1.92 & 600@xmath21458 + gc0031 & aat107145 & 13 24 29.73 & -43 02 06.5 & 10.62 & 21.20 & 20.93 & 20.08 & 19.50 & 18.97 & 0.11 & 0.03 & 0.01 & 0.01 & 0.03 & 21.16 & 20.33 & 19.55 & 0.03 & 0.02 & 0.04 & 0.27 & 0.85 & 0.58 & 1.11 & 0.78 & 0.82 & 1.60 & 595@xmath214202 + gc0032 & pff_gc-009 & 13 24 31.35 & -43 11 26.7 & 14.55 & 20.82 & 20.58 & 19.77 & 19.25 & 18.70 & 0.07 & 0.02 & 0.01 & 0.01 & 0.03 & 20.73 & 20.05 & 19.26 & 0.03 & 0.02 & 0.01 & 0.23 & 0.81 & 0.52 & 1.07 & 0.80 & 0.68 & 1.47 & 683@xmath21438 + gc0033 & whh-3 & 13 24 32.17 & -43 10 56.9 & 14.10 & 20.56 & 20.32 & 19.49 & 18.97 & 18.43 & 0.06 & 0.02 & 0.01 & 0.01 & 0.03 & 20.46 & 19.76 & 18.99 & 0.02 & 0.02 & 0.01 & 0.25 & 0.83 & 0.52 & 1.06 & 0.76 & 0.71 & 1.47 & 709@xmath21454 + gc0034 & c112 & 13 24 32.66 & -43 18 48.8 & 20.32 & - & - & - & - & - & - & - & - & - & - & 22.98 & 22.22 & 21.31 & 0.03 & 0.02 & 0.02 & - & - & - & - & 0.91 & 0.75 & 1.67 & - + gc0035 & pff_gc-010 & 13 24 33.09 & -43 18 44.8 & 20.23 & 20.45 & 20.36 & 19.68 & 19.26 & 18.84 & 0.06 & 0.02 & 0.01 & 0.01 & 0.03 & 20.45 & 19.91 & 19.24 & 0.01 & 0.01 & 0.01 & 0.09 & 0.68 & 0.42 & 0.85 & 0.67 & 0.54 & 1.21 & 344@xmath21458 + gc0036 & aat107977 & 13 24 34.63 & -43 12 50.5 & 15.18 & 21.73 & 21.18 & 20.18 & 19.57 & 18.89 & 0.17 & 0.04 & 0.01 & 0.01 & 0.03 & 21.52 & 20.54 & 19.59 & 0.02 & 0.02 & 0.01 & 0.54 & 1.01 & 0.60 & 1.28 & 0.95 & 0.98 & 1.93 & 517@xmath214123 + gc0037 & pff_gc-011 & 13 24 36.87 & -43 19 16.2 & 20.36 & 20.01 & 19.82 & 19.03 & 18.53 & 18.00 & 0.13 & 0.02 & 0.01 & 0.01 & 0.03 & 19.95 & 19.33 & 18.55 & 0.01 & 0.01 & 0.01 & 0.18 & 0.79 & 0.50 & 1.04 & 0.78 & 0.62 & 1.40 & 616@xmath21441 + gc0038 & c113 & 13 24 37.75 & -43 16 26.5 & 17.80 & - & - & - & - & - & - & - & - & - & - & 20.50 & 19.87 & 19.12 & 0.02 & 0.02 & 0.01 & - & - & - & - & 0.75 & 0.63 & 1.38 & - + gc0039 & pff_gc-012 & 13 24 38.77 & -43 06 26.6 & 10.38 & 21.24 & 20.51 & 19.45 & 18.83 & 18.15 & 0.13 & 0.02 & 0.01 & 0.01 & 0.03 & 20.92 & 19.78 & 18.85 & 0.01 & 0.01 & 0.01 & 0.73 & 1.06 & 0.62 & 1.30 & 0.93 & 1.14 & 2.07 & 573@xmath21421 + gc0040 & hghh-41/c41 & 13 24 38.98 & -43 20 06.4 & 20.94 & 20.20 & 19.59 & 18.59 & 17.94 & 17.32 & 0.06 & 0.02 & 0.01 & 0.01 & 0.03 & 19.95 & 18.92 & 17.97 & 0.02 & 0.01 & 0.01 & 0.61 & 1.00 & 0.65 & 1.27 & 0.95 & 1.03 & 1.98 & 363@xmath2141 + gc0041 & hghh-29/c29 & 13 24 40.39 & -43 18 05.3 & 19.01 & 19.77 & 19.15 & 18.15 & 17.54 & 16.89 & 0.04 & 0.01 & 0.01 & 0.01 & 0.03 & 19.46 & 18.37 & 17.53 & 0.04 & 0.03 & 0.02 & 0.62 & 1.00 & 0.61 & 1.26 & 0.84 & 1.09 & 1.92 & 726@xmath2141 + gc0042 & pff_gc-013 & 13 24 40.42 & -43 35 04.9 & 35.01 & 20.06 & 19.83 & 19.00 & 18.49 & 17.98 & 0.15 & 0.02 & 0.01 & 0.01 & 0.03 & - & - & - & - & - & - & 0.24 & 0.83 & 0.51 & 1.03 & - & - & - & 727@xmath21431 + gc0043 & c114 & 13 24 40.48 & -42 53 35.3 & 11.46 & - & - & - & - & - & - & - & - & - & - & 23.46 & 22.39 & 21.42 & 0.05 & 0.04 & 0.03 & - & - & - & - & 0.97 & 1.07 & 2.04 & - + gc0044 & whh-4/hh-024 & 13 24 40.60 & -43 13 18.1 & 14.89 & 20.71 & 20.12 & 19.12 & 18.50 & 17.85 & 0.07 & 0.02 & 0.01 & 0.01 & 0.03 & 20.46 & 19.40 & 18.53 & 0.03 & 0.02 & 0.01 & 0.60 & 1.00 & 0.62 & 1.27 & 0.87 & 1.06 & 1.93 & 688@xmath21425 + gc0045 & pff_gc-014 & 13 24 41.05 & -42 59 48.4 & 8.62 & 21.60 & 21.17 & 20.23 & 19.67 & 18.94 & 0.26 & 0.06 & 0.02 & 0.02 & 0.03 & 21.47 & 20.56 & 19.76 & 0.04 & 0.03 & 0.04 & 0.43 & 0.94 & 0.57 & 1.30 & 0.80 & 0.91 & 1.71 & 690@xmath21434 + gc0046 & pff_gc-015 & 13 24 41.20 & -43 01 45.6 & 8.51 & 20.27 & 20.01 & 19.22 & 18.74 & 18.15 & 0.06 & 0.02 & 0.01 & 0.01 & 0.03 & 20.12 & 19.40 & 18.70 & 0.05 & 0.04 & 0.05 & 0.26 & 0.79 & 0.47 & 1.06 & 0.70 & 0.72 & 1.42 & 533@xmath21425 + gc0047 & aat109380 & 13 24 43.58 & -43 08 43.2 & 11.05 & 19.64 & 19.52 & 18.80 & 18.32 & 17.87 & 0.04 & 0.01 & 0.01 & 0.01 & 0.03 & 19.60 & 19.02 & 18.30 & 0.01 & - & - & 0.12 & 0.73 & 0.48 & 0.93 & 0.72 & 0.59 & 1.31 & 465@xmath21438 + gc0048 & pff_gc-016 & 13 24 43.60 & -42 53 07.3 & 11.36 & 21.38 & 20.87 & 19.85 & 19.26 & 18.58 & 0.12 & 0.03 & 0.01 & 0.01 & 0.03 & 21.19 & 20.21 & 19.33 & 0.02 & 0.02 & 0.01 & 0.51 & 1.02 & 0.60 & 1.27 & 0.88 & 0.98 & 1.85 & 505@xmath21417 + gc0049 & pff_gc-017 & 13 24 43.63 & -42 58 16.4 & 8.54 & 20.54 & 20.32 & 19.51 & 18.99 & 18.47 & 0.07 & 0.02 & 0.01 & 0.01 & 0.03 & 20.44 & 19.78 & 19.01 & 0.01 & 0.01 & 0.01 & 0.22 & 0.81 & 0.52 & 1.04 & 0.77 & 0.66 & 1.43 & 963@xmath21450 + gc0050 & whh-5 & 13 24 44.58 & -43 02 47.3 & 8.04 & 20.47 & 20.03 & 19.06 & 18.49 & 17.87 & 0.08 & 0.02 & 0.01 & 0.01 & 0.03 & 20.29 & 19.19 & 18.90 & 0.05 & 0.04 & 0.04 & 0.44 & 0.97 & 0.57 & 1.19 & 0.29 & 1.09 & 1.39 & 671@xmath21441 + gc0051 & hh-054 & 13 24 45.22 & -43 23 19.2 & 23.49 & - & - & - & - & - & - & - & - & - & - & 20.92 & 20.08 & 19.25 & 0.05 & 0.03 & 0.04 & - & - & - & - & 0.82 & 0.84 & 1.66 & 998@xmath214250 + gc0052 & aat109711 & 13 24 45.35 & -42 59 33.5 & 7.89 & 19.88 & 19.87 & 19.30 & 18.90 & 18.33 & 0.06 & 0.02 & 0.01 & 0.01 & 0.03 & 19.84 & 19.48 & 18.91 & 0.02 & 0.02 & 0.04 & 0.02 & 0.56 & 0.40 & 0.98 & 0.58 & 0.36 & 0.94 & 276@xmath21459 + gc0053 & aat109788 & 13 24 45.78 & -43 02 24.5 & 7.75 & 21.82 & 21.02 & 19.89 & 19.23 & 18.56 & 0.25 & 0.04 & 0.01 & 0.01 & 0.03 & 21.39 & 20.17 & 19.43 & 0.02 & 0.02 & 0.03 & 0.80 & 1.14 & 0.65 & 1.33 & 0.74 & 1.22 & 1.96 & 527@xmath21430 + gc0054 & hghh - g019/g19 & 13 24 46.46 & -43 04 11.6 & 8.12 & 20.16 & 19.95 & 19.16 & 18.64 & 18.13 & 0.06 & 0.02 & 0.01 & 0.01 & 0.03 & 20.06 & 19.38 & 18.64 & 0.04 & 0.03 & 0.03 & 0.21 & 0.79 & 0.52 & 1.03 & 0.75 & 0.67 & 1.42 & 710@xmath21422 + gc0055 & hghh-09/c9 & 13 24 46.72 & -43 01 18.4 & 7.48 & 20.00 & 19.76 & 18.95 & 18.40 & 17.90 & 0.05 & 0.02 & 0.01 & 0.01 & 0.03 & 19.87 & 19.13 & 18.42 & 0.04 & 0.04 & 0.04 & 0.24 & 0.81 & 0.56 & 1.05 & 0.70 & 0.74 & 1.45 & 501@xmath21421 + gc0056 & pff_gc-018 & 13 24 47.10 & -43 06 01.7 & 8.87 & 20.90 & 20.03 & 18.91 & 18.25 & 17.54 & 0.10 & 0.02 & 0.01 & 0.01 & 0.03 & 20.45 & 19.23 & 18.28 & 0.02 & 0.01 & 0.01 & 0.87 & 1.12 & 0.66 & 1.37 & 0.95 & 1.22 & 2.17 & 534@xmath21415 + gc0057 & hghh - g277/g277 & 13 24 47.37 & -42 58 29.8 & 7.82 & 20.26 & 19.96 & 19.10 & 18.56 & 18.02 & 0.06 & 0.02 & 0.01 & 0.01 & 0.03 & 20.14 & 19.37 & 18.61 & 0.01 & 0.01 & 0.01 & 0.30 & 0.87 & 0.53 & 1.08 & 0.76 & 0.77 & 1.53 & 714@xmath21433 + gc0058 & whh-6 & 13 24 47.37 & -42 57 51.2 & 8.06 & 21.33 & 20.75 & 19.73 & 19.12 & 18.44 & 0.15 & 0.03 & 0.01 & 0.01 & 0.03 & 21.08 & 20.07 & 19.15 & 0.01 & 0.01 & 0.01 & 0.58 & 1.02 & 0.62 & 1.29 & 0.93 & 1.00 & 1.93 & 685@xmath21443 + gc0059 & aat110138 & 13 24 47.61 & -43 10 48.5 & 12.12 & 20.87 & 20.53 & 19.67 & 19.11 & 18.57 & 0.10 & 0.03 & 0.01 & 0.01 & 0.03 & 20.75 & 19.98 & 19.16 & 0.03 & 0.02 & 0.01 & 0.34 & 0.87 & 0.56 & 1.10 & 0.82 & 0.77 & 1.59 & 358@xmath214137 + gc0060 & hhh86 - 10/c10 & 13 24 48.06 & -43 08 14.2 & 10.13 & 19.87 & 19.44 & 18.49 & 17.91 & 17.33 & 0.05 & 0.01 & 0.01 & 0.01 & 0.03 & 19.66 & 18.75 & 17.93 & 0.03 & 0.03 & 0.01 & 0.44 & 0.95 & 0.58 & 1.16 & 0.81 & 0.91 & 1.73 & 829@xmath21422 + gc0061 & c115 & 13 24 48.71 & -42 52 35.5 & 11.12 & - & - & - & - & - & - & - & - & - & - & 20.88 & 20.23 & 19.51 & 0.02 & 0.02 & 0.01 & - & - & - & - & 0.73 & 0.65 & 1.37 & - + gc0062 & pff_gc-019 & 13 24 48.97 & -42 57 48.4 & 7.81 & 20.10 & 19.87 & 19.05 & 18.53 & 18.00 & 0.05 & 0.02 & 0.01 & 0.01 & 0.03 & 20.00 & 19.31 & 18.57 & 0.03 & 0.03 & 0.01 & 0.23 & 0.81 & 0.53 & 1.05 & 0.74 & 0.69 & 1.43 & 607@xmath21427 + gc0063 & aat110410 & 13 24 49.38 & -43 08 17.7 & 10.00 & 21.47 & 21.18 & 20.34 & 19.79 & 19.23 & 0.16 & 0.04 & 0.02 & 0.01 & 0.03 & 21.37 & 20.66 & 19.81 & 0.03 & 0.02 & 0.01 & 0.29 & 0.84 & 0.55 & 1.11 & 0.85 & 0.71 & 1.56 & 580@xmath21481 + gc0064 & aat110551 & 13 24 50.09 & -43 07 36.2 & 9.42 & 21.43 & 21.08 & 20.44 & 20.04 & 19.49 & 0.16 & 0.04 & 0.02 & 0.02 & 0.03 & 21.19 & 20.70 & 20.03 & 0.02 & 0.02 & 0.01 & 0.35 & 0.64 & 0.41 & 0.96 & 0.67 & 0.49 & 1.16 & 615@xmath21476 + gc0065 & pff_gc-020 & 13 24 50.48 & -42 59 49.0 & 6.92 & 20.57 & 20.33 & 19.53 & 19.00 & 18.47 & 0.09 & 0.03 & 0.01 & 0.01 & 0.03 & 20.42 & 19.58 & 19.68 & 0.05 & 0.04 & 0.04 & 0.24 & 0.80 & 0.53 & 1.06&-0.11 & 0.85 & 0.74 & 331@xmath21447 + gc0066 & hghh-42/c42 & 13 24 50.87 & -43 01 22.9 & 6.72 & 20.53 & 19.91 & 18.86 & 18.24 & 17.53 & 0.09 & 0.02 & 0.01 & 0.01 & 0.03 & 20.21 & 19.11 & 18.21 & 0.04 & 0.03 & 0.05 & 0.62 & 1.05 & 0.63 & 1.33 & 0.90 & 1.10 & 2.00 & 552@xmath21412 + gc0067 & vhh81 - 02/c2 & 13 24 51.49 & -43 12 11.1 & 12.86 & 19.64 & 19.33 & 18.50 & 18.01 & 17.42 & 0.04 & 0.01 & 0.01 & 0.01 & 0.03 & 19.48 & 18.72 & 17.94 & 0.04 & 0.03 & 0.01 & 0.31 & 0.83 & 0.49 & 1.07 & 0.78 & 0.77 & 1.55 & 628@xmath21422 + gc0068 & c100 & 13 24 51.80 & -43 04 33.7 & 7.33 & - & - & 20.08 & - & - & - & - & - & - & - & 20.44 & 19.72 & 19.03 & 0.08 & 0.08 & 0.10 & - & - & - & 1.28 & 0.69 & 0.72 & 1.41 & - + gc0069 & aat111033 & 13 24 52.98 & -43 11 55.8 & 12.50 & 21.17 & 20.93 & 20.12 & 19.66 & 19.13 & 0.11 & 0.03 & 0.01 & 0.01 & 0.03 & 21.10 & 20.40 & 19.64 & 0.01 & 0.01 & 0.01 & 0.24 & 0.81 & 0.46 & 0.99 & 0.76 & 0.70 & 1.47 & 302@xmath214166 + gc0070 & hghh - g302/g302 & 13 24 53.29 & -43 04 34.8 & 7.15 & 20.30 & 20.01 & 19.20 & 18.69 & 18.16 & 0.07 & 0.02 & 0.01 & 0.01 & 0.03 & 20.18 & 19.50 & 18.73 & 0.02 & 0.01 & 0.01 & 0.28 & 0.81 & 0.52 & 1.04 & 0.77 & 0.68 & 1.45 & 558@xmath21443 + gc0071 & aat111185 & 13 24 54.00 & -43 04 24.4 & 6.96 & 20.92 & 20.76 & 19.92 & 19.41 & 18.92 & 0.11 & 0.03 & 0.01 & 0.01 & 0.03 & 20.90 & 20.28 & 19.37 & 0.02 & 0.02 & 0.02 & 0.16 & 0.84 & 0.51 & 1.00 & 0.91 & 0.62 & 1.53 & 466@xmath21487 + gc0072 & pff_gc-021 & 13 24 54.18 & -42 54 50.4 & 8.78 & 20.21 & 20.03 & 19.25 & 18.75 & 18.24 & 0.06 & 0.02 & 0.01 & 0.01 & 0.03 & 20.13 & 19.52 & 18.81 & 0.03 & 0.02 & 0.01 & 0.18 & 0.78 & 0.50 & 1.01 & 0.70 & 0.62 & 1.32 & 594@xmath21450 + gc0073 & pff_gc-022 & 13 24 54.33 & -43 03 15.5 & 6.44 & 20.72 & 20.58 & 19.84 & 19.37 & 18.90 & 0.11 & 0.03 & 0.01 & 0.01 & 0.03 & 20.69 & 20.09 & 19.37 & 0.03 & 0.02 & 0.02 & 0.14 & 0.74 & 0.48 & 0.95 & 0.72 & 0.60 & 1.32 & 619@xmath21444 + gc0074 & hhh86 - 30/c30 & 13 24 54.35 & -42 53 24.8 & 9.84 & 18.76 & 18.24 & 17.25 & 16.67 & 16.02 & 0.03 & 0.01 & 0.01 & 0.01 & 0.03 & 18.47 & 17.49 & 16.68 & 0.03 & 0.04 & 0.02 & 0.52 & 0.98 & 0.58 & 1.23 & 0.81 & 0.98 & 1.79 & 778@xmath21413 + gc0075 & aat111296 & 13 24 54.49 & -43 05 34.7 & 7.50 & 21.70 & 20.99 & 19.96 & 19.31 & 18.61 & 0.23 & 0.04 & 0.01 & 0.01 & 0.03 & 21.40 & 20.28 & 19.33 & 0.02 & 0.01 & 0.01 & 0.71 & 1.03 & 0.65 & 1.35 & 0.96 & 1.11 & 2.07 & 695@xmath21445 + gc0076 & pff_gc-023 & 13 24 54.55 & -42 48 58.7 & 13.59 & 20.55 & 20.30 & 19.44 & 18.90 & 18.37 & 0.06 & 0.02 & 0.01 & 0.01 & 0.03 & 20.48 & 19.74 & 18.97 & 0.01 & 0.01 & 0.01 & 0.25 & 0.86 & 0.54 & 1.08 & 0.77 & 0.74 & 1.51 & 457@xmath21431 + gc0077 & hghh-11/c11 & 13 24 54.73 & -43 01 21.7 & 6.01 & 19.55 & 18.96 & 17.91 & 17.26 & 16.61 & 0.05 & 0.01 & 0.01 & 0.01 & 0.03 & 19.21 & 17.98 & 17.20 & 0.06 & 0.05 & 0.06 & 0.59 & 1.05 & 0.66 & 1.30 & 0.79 & 1.23 & 2.01 & 753@xmath2141 + gc0078 & aat111406 & 13 24 55.29 & -43 03 15.6 & 6.28 & 21.84 & 21.15 & 20.13 & 19.50 & 18.80 & 0.30 & 0.05 & 0.02 & 0.01 & 0.03 & 21.66 & 20.70 & 18.80 & 0.03 & 0.02 & 0.01 & 0.69 & 1.03 & 0.63 & 1.33 & 1.90 & 0.96 & 2.86 & 669@xmath21477 + gc0079 & c116 & 13 24 55.46 & -43 09 58.5 & 10.61 & - & - & - & - & - & - & - & - & - & - & 23.80 & 22.74 & 21.84 & 0.06 & 0.02 & 0.03 & - & - & - & - & 0.90 & 1.07 & 1.97 & - + gc0080 & hh-052 & 13 24 55.71 & -43 22 48.4 & 22.43 & - & - & - & - & - & - & - & - & - & - & 22.58 & 21.61 & 20.42 & 0.05 & 0.03 & 0.03 & - & - & - & - & 1.19 & 0.97 & 2.16 & 921@xmath214146 + gc0081 & pff_gc-024 & 13 24 55.71 & -43 20 39.1 & 20.36 & 21.07 & 20.73 & 19.84 & 19.31 & 18.73 & 0.28 & 0.03 & 0.01 & 0.01 & 0.03 & 21.00 & 20.17 & 19.34 & 0.02 & 0.01 & 0.01 & 0.34 & 0.89 & 0.53 & 1.11 & 0.83 & 0.83 & 1.66 & 279@xmath21438 + gc0082 & c117 & 13 24 56.06 & -42 54 29.6 & 8.80 & - & - & - & - & - & - & - & - & - & - & 21.27 & 20.18 & 19.26 & 0.01 & 0.02 & 0.01 & - & - & - & - & 0.92 & 1.09 & 2.01 & - + gc0083 & aat111563 & 13 24 56.08 & -43 10 16.4 & 10.79 & 21.34 & 21.03 & 20.45 & 20.05 & 19.62 & 0.15 & 0.04 & 0.02 & 0.02 & 0.03 & 21.14 & 20.73 & 20.05 & 0.02 & 0.01 & 0.02 & 0.31 & 0.58 & 0.41 & 0.83 & 0.68 & 0.41 & 1.09 & 649@xmath214102 + gc0084 & hghh - g279 & 13 24 56.27 & -43 03 23.4 & 6.15 & 19.97 & 19.91 & 19.45 & 19.17 & 18.88 & 0.06 & 0.02 & 0.01 & 0.01 & 0.03 & 19.76 & 19.38 & 19.83 & 0.01 & 0.01 & 0.01 & 0.06 & 0.45 & 0.29 & 0.57&-0.45 & 0.38&-0.07 & 366@xmath21434 + gc0085 & c118 & 13 24 57.17 & -43 08 42.6 & 9.39 & - & - & - & - & - & - & - & - & - & - & 22.60 & 21.76 & 20.67 & 0.04 & 0.03 & 0.02 & - & - & - & - & 1.08 & 0.84 & 1.92 & - + gc0086 & hghh-31/c31 & 13 24 57.44 & -43 01 08.1 & 5.52 & 20.09 & 19.42 & 18.38 & 17.75 & 17.06 & 0.08 & 0.02 & 0.01 & 0.01 & 0.03 & 19.73 & 18.59 & 17.71 & 0.04 & 0.03 & 0.04 & 0.67 & 1.04 & 0.62 & 1.31 & 0.88 & 1.14 & 2.02 & 690@xmath21418 + gc0087 & hghh - g369 & 13 24 57.52 & -42 59 23.3 & 5.78 & 19.82 & 19.58 & 18.74 & 18.22 & 17.69 & 0.06 & 0.02 & 0.01 & 0.01 & 0.03 & 19.67 & 18.96 & 18.24 & 0.04 & 0.03 & 0.05 & 0.24 & 0.83 & 0.53 & 1.06 & 0.72 & 0.71 & 1.44 & 512@xmath21417 + gc0088 & pff_gc-025 & 13 24 57.56 & -43 05 32.8 & 7.04 & 21.50 & 21.02 & 20.02 & 19.39 & 18.67 & 0.20 & 0.04 & 0.02 & 0.01 & 0.03 & 21.41 & 20.41 & 19.45 & 0.04 & 0.03 & 0.02 & 0.47 & 1.00 & 0.63 & 1.35 & 0.96 & 1.00 & 1.97 & 923@xmath21430 + gc0089 & c119 & 13 24 57.69 & -42 55 48.4 & 7.64 & - & - & - & - & - & - & - & - & - & - & 21.16 & 21.02 & 20.67 & 0.03 & 0.03 & 0.02 & - & - & - & - & 0.35 & 0.14 & 0.49 & - + gc0090 & c120 & 13 24 57.95 & -42 52 04.9 & 10.56 & - & - & - & - & - & - & - & - & - & - & 23.26 & 22.34 & 21.43 & 0.04 & 0.04 & 0.03 & - & - & - & - & 0.90 & 0.92 & 1.83 & - + gc0091 & vhh81 - 03/c3 & 13 24 58.21 & -42 56 10.0 & 7.33 & 19.34 & 18.73 & 17.71 & 17.10 & 16.44 & 0.04 & 0.01 & 0.01 & 0.01 & 0.03 & 19.02 & 17.88 & 17.08 & 0.04 & 0.03 & 0.02 & 0.61 & 1.02 & 0.61 & 1.26 & 0.80 & 1.14 & 1.94 & 562@xmath2142 + gc0092 & c121 & 13 24 58.42 & -43 08 21.2 & 8.97 & - & - & - & - & - & - & - & - & - & - & 23.53 & 23.01 & 22.45 & 0.05 & 0.04 & 0.09 & - & - & - & - & 0.56 & 0.52 & 1.08 & - + gc0093 & pff_gc-026 & 13 24 58.45 & -42 42 53.3 & 19.02 & 20.44 & 20.33 & 19.52 & 19.07 & 18.50 & 0.08 & 0.02 & 0.01 & 0.01 & 0.03 & 20.47 & 19.82 & 19.09 & 0.02 & 0.01 & 0.01 & 0.12 & 0.81 & 0.45 & 1.02 & 0.73 & 0.65 & 1.37 & 490@xmath21467 + gc0094 & c122 & 13 24 59.01 & -43 08 21.4 & 8.91 & - & - & - & - & - & - & - & - & - & - & - & - & 22.33 & - & - & - & - & - & - & - & - & - & - & - + gc0095 & c123 & 13 24 59.92 & -43 09 08.6 & 9.46 & - & - & - & - & - & - & - & - & - & - & 21.88 & 21.10 & 20.24 & 0.02 & 0.01 & 0.01 & - & - & - & - & 0.86 & 0.78 & 1.64 & - + gc0096 & aat112158 & 13 25 00.15 & -42 54 09.0 & 8.61 & 21.99 & 21.43 & 20.43 & 19.86 & 19.23 & 0.25 & 0.05 & 0.02 & 0.01 & 0.03 & 21.75 & 20.78 & 19.88 & 0.01 & 0.01 & 0.01 & 0.55 & 1.00 & 0.57 & 1.20 & 0.90 & 0.96 & 1.87 & 699@xmath21443 + gc0097 & c124 & 13 25 00.37 & -43 10 46.9 & 10.85 & - & - & - & - & - & - & - & - & - & - & 23.07 & 22.36 & 21.57 & 0.03 & 0.02 & 0.03 & - & - & - & - & 0.78 & 0.71 & 1.50 & - + gc0098 & pff_gc-027 & 13 25 00.64 & -43 05 30.3 & 6.58 & 21.22 & 20.75 & 19.74 & 19.13 & 18.45 & 0.17 & 0.04 & 0.01 & 0.01 & 0.03 & 21.05 & 20.06 & 19.11 & 0.01 & 0.01 & 0.01 & 0.47 & 1.01 & 0.61 & 1.29 & 0.95 & 0.99 & 1.94 & 524@xmath21434 + gc0099 & c125 & 13 25 00.83 & -43 11 10.6 & 11.16 & - & - & - & - & - & - & - & - & - & - & 22.34 & 21.55 & 20.66 & 0.03 & 0.02 & 0.02 & - & - & - & - & 0.89 & 0.78 & 1.67 & - + gc0100 & c126 & 13 25 00.91 & -43 09 14.5 & 9.45 & - & - & - & - & - & - & - & - & - & - & 24.02 & 23.54 & 22.66 & 0.04 & 0.05 & 0.08 & - & - & - & - & 0.88 & 0.48 & 1.35 & - + gc0101 & c127 & 13 25 01.32 & -43 08 43.4 & 8.97 & - & - & - & - & - & - & - & - & - & - & 23.21 & 22.42 & 21.62 & 0.04 & 0.02 & 0.03 & - & - & - & - & 0.80 & 0.79 & 1.59 & - + gc0102 & c128 & 13 25 01.46 & -43 08 33.0 & 8.81 & - & - & - & - & - & - & - & - & - & - & 23.08 & 22.04 & 20.95 & 0.05 & 0.03 & 0.02 & - & - & - & - & 1.09 & 1.04 & 2.13 & - + gc0103 & pff_gc-029 & 13 25 01.60 & -42 54 40.9 & 8.03 & 19.97 & 19.75 & 19.37 & 19.06 & 18.71 & 0.05 & 0.02 & 0.01 & 0.01 & 0.03 & 19.72 & 19.50 & 19.10 & 0.03 & 0.03 & 0.01 & 0.22 & 0.38 & 0.31 & 0.65 & 0.40 & 0.22 & 0.62 & 570@xmath21438 + gc0104 & c129 & 13 25 01.63 & -42 50 51.3 & 11.34 & - & - & - & - & - & - & - & - & - & - & 22.59 & 21.71 & 20.83 & 0.04 & 0.02 & 0.02 & - & - & - & - & 0.88 & 0.88 & 1.76 & - + gc0105 & pff_gc-030 & 13 25 01.73 & -43 00 09.9 & 4.83 & 21.68 & 20.94 & 19.89 & 19.26 & 18.54 & 0.36 & 0.06 & 0.02 & 0.02 & 0.03 & 21.21 & 19.93 & 20.43 & 0.05 & 0.04 & 0.02 & 0.73 & 1.05 & 0.64 & 1.35&-0.50 & 1.29 & 0.78 & 357@xmath21429 + gc0106 & hghh-04/c4 & 13 25 01.83 & -43 09 25.4 & 9.52 & 19.10 & 18.86 & 18.04 & 17.50 & 16.98 & 0.03 & 0.01 & 0.01 & 0.01 & 0.03 & 18.95 & 18.24 & 17.50 & 0.04 & 0.03 & 0.01 & 0.23 & 0.82 & 0.54 & 1.06 & 0.74 & 0.71 & 1.45 & 689@xmath21416 + gc0107 & c130 & 13 25 01.86 & -42 52 27.8 & 9.88 & - & - & - & - & - & - & - & - & - & - & 21.22 & 20.58 & 19.85 & 0.01 & 0.01 & 0.01 & - & - & - & - & 0.73 & 0.64 & 1.37 & - + gc0108 & pff_gc-031 & 13 25 02.76 & -43 11 21.2 & 11.17 & 20.51 & 20.29 & 19.48 & 18.99 & 18.42 & 0.07 & 0.02 & 0.01 & 0.01 & 0.03 & 20.43 & 19.75 & 18.95 & 0.03 & 0.02 & 0.01 & 0.23 & 0.81 & 0.49 & 1.06 & 0.80 & 0.68 & 1.48 & 444@xmath21433 + gc0109 & hghh - g176 & 13 25 03.13 & -42 56 25.1 & 6.51 & 20.47 & 19.94 & 18.93 & 18.32 & 17.67 & 0.09 & 0.02 & 0.01 & 0.01 & 0.03 & 20.24 & 19.23 & 18.36 & 0.01 & 0.02 & 0.01 & 0.54 & 1.01 & 0.61 & 1.26 & 0.87 & 1.01 & 1.88 & 551@xmath21413 + gc0110 & hghh - g066 & 13 25 03.18 & -43 03 02.5 & 4.85 & 20.26 & 19.67 & 18.68 & 18.15 & 17.44 & 0.09 & 0.02 & 0.01 & 0.01 & 0.03 & 19.94 & 18.96 & 18.05 & 0.01 & 0.01 & 0.02 & 0.59 & 0.99 & 0.52 & 1.24 & 0.91 & 0.98 & 1.89 & 576@xmath2148 + gc0111 & pff_gc-032 & 13 25 03.24 & -42 57 40.5 & 5.65 & 20.92 & 20.59 & 19.73 & 19.18 & 18.61 & 0.14 & 0.04 & 0.01 & 0.01 & 0.03 & 20.81 & 20.05 & 19.22 & 0.01 & 0.02 & 0.01 & 0.33 & 0.86 & 0.55 & 1.12 & 0.83 & 0.76 & 1.59 & 648@xmath21429 + gc0112 & pff_gc-033 & 13 25 03.34 & -43 15 27.4 & 14.98 & 21.10 & 20.49 & 19.46 & 18.89 & 18.21 & 0.33 & 0.03 & 0.01 & 0.01 & 0.03 & 20.83 & 19.79 & 18.90 & 0.02 & 0.02 & 0.01 & 0.61 & 1.02 & 0.57 & 1.25 & 0.89 & 1.04 & 1.93 & 531@xmath21418 + gc0113 & hhh86 - 32/c32 & 13 25 03.37 & -42 50 46.2 & 11.28 & 20.22 & 19.50 & 18.44 & 17.81 & 17.15 & 0.05 & 0.01 & 0.01 & 0.01 & 0.03 & 19.86 & 18.71 & 17.85 & 0.03 & 0.03 & 0.01 & 0.72 & 1.06 & 0.63 & 1.29 & 0.85 & 1.15 & 2.01 & 718@xmath21411 + gc0114 & pff_gc-034 & 13 25 03.37 & -43 11 39.6 & 11.41 & 20.98 & 20.70 & 19.91 & 19.35 & 18.84 & 0.09 & 0.03 & 0.01 & 0.01 & 0.03 & 20.89 & 20.19 & 19.38 & 0.03 & 0.02 & 0.01 & 0.28 & 0.80 & 0.56 & 1.06 & 0.81 & 0.69 & 1.50 & 605@xmath21446 + gc0115 & c131 & 13 25 03.67 & -42 51 21.7 & 10.72 & - & - & - & - & - & - & - & - & - & - & 21.57 & 20.98 & 20.20 & 0.02 & 0.02 & 0.01 & - & - & - & - & 0.78 & 0.59 & 1.36 & - + gc0116 & aat112752 & 13 25 04.12 & -43 00 19.6 & 4.37 & 20.51 & 20.16 & 19.34 & 18.88 & 18.44 & 0.15 & 0.04 & 0.02 & 0.01 & 0.03 & 20.35 & 19.58 & 18.87 & 0.02 & 0.02 & 0.03 & 0.35 & 0.82 & 0.46 & 0.90 & 0.72 & 0.76 & 1.48 & 679@xmath21482 + gc0117 & pff_gc-035 & 13 25 04.48 & -43 10 48.4 & 10.54 & 21.32 & 20.73 & 19.71 & 19.07 & 18.44 & 0.14 & 0.03 & 0.01 & 0.01 & 0.03 & 21.13 & 20.07 & 19.13 & 0.03 & 0.02 & 0.01 & 0.59 & 1.02 & 0.64 & 1.27 & 0.94 & 1.06 & 2.00 & 627@xmath21422 + gc0118 & aat112964 & 13 25 04.61 & -43 07 21.7 & 7.50 & 20.45 & 20.39 & 19.61 & 19.13 & 18.68 & 0.08 & 0.03 & 0.01 & 0.01 & 0.03 & 20.47 & 19.90 & 19.11 & 0.01 & 0.01 & 0.01 & 0.06 & 0.78 & 0.49 & 0.94 & 0.78 & 0.57 & 1.35 & 456@xmath214118 + gc0119 & hghh-43/c43 & 13 25 04.81 & -43 09 38.8 & 9.47 & 19.74 & 19.45 & 18.60 & 18.07 & 17.53 & 0.04 & 0.01 & 0.01 & 0.01 & 0.03 & 19.59 & 18.85 & 18.07 & 0.03 & 0.02 & 0.01 & 0.29 & 0.85 & 0.53 & 1.07 & 0.78 & 0.74 & 1.52 & 518@xmath21425 + gc0120 & whh-7 & 13 25 05.02 & -42 57 15.0 & 5.68 & 18.96 & 18.39 & 17.41 & 16.81 & 16.19 & 0.04 & 0.01 & 0.01 & 0.01 & 0.03 & 18.69 & 17.60 & 16.83 & 0.02 & 0.03 & 0.01 & 0.57 & 0.98 & 0.60 & 1.22 & 0.77 & 1.09 & 1.86 & 722@xmath21420 + gc0121 & hghh - g035 & 13 25 05.29 & -42
58 05.8 & 5.09 & 21.02 & 20.59 & 19.62 & 19.01 & 18.41 & 0.17 & 0.04 & 0.01 & 0.01 & 0.03 & 20.87 & 19.95 & 19.08 & 0.01 & 0.02 & 0.01 & 0.43 & 0.97 & 0.61 & 1.21 & 0.87 & 0.92 & 1.80 & 776@xmath21426 + gc0122 & pff_gc-036 & 13 25 05.46 & -43 14 02.6 & 13.52 & 20.09 & 19.83 & 19.04 & 18.52 & 17.91 & 0.05 & 0.02 & 0.01 & 0.01 & 0.03 & 19.96 & 19.27 & 18.49 & 0.02 & 0.02 & 0.01 & 0.26 & 0.79 & 0.52 & 1.13 & 0.78 & 0.69 & 1.47 & 666@xmath21430 + gc0123 & hghh-12/c12/r281 & 13 25 05.72 & -43 10 30.7 & 10.18 & - & - & - & - & - & - & - & - & - & - & 19.34 & 18.23 & 17.36 & 0.04 & 0.03 & 0.02 & - & - & - & - & 0.87 & 1.11 & 1.98 & 440@xmath2141 + gc0124 & hghh - g342 & 13 25 05.83 & -42 59 00.6 & 4.52 & 19.65 & 19.14 & 18.18 & 17.57 & 16.96 & 0.08 & 0.02 & 0.01 & 0.01 & 0.03 & - & - & - & - & - & - & 0.51 & 0.96 & 0.61 & 1.22 & - & - & - & 553@xmath21421 + gc0125 & hhh86 - 13/c13 & 13 25 06.25 & -43 15 11.6 & 14.58 & 20.02 & 19.60 & 18.68 & 18.13 & 17.51 & 0.04 & 0.01 & 0.01 & 0.01 & 0.03 & 19.82 & 18.90 & 18.12 & 0.03 & 0.03 & 0.02 & 0.41 & 0.93 & 0.55 & 1.16 & 0.78 & 0.92 & 1.70 & 601@xmath21412 + gc0126 & r276 & 13 25 07.33 & -43 08 29.6 & 8.23 & 21.93 & 21.62 & 20.78 & 20.26 & 19.73 & 0.26 & 0.07 & 0.02 & 0.02 & 0.04 & 21.88 & 21.16 & 20.32 & 0.03 & 0.03 & 0.02 & 0.31 & 0.83 & 0.52 & 1.05 & 0.84 & 0.71 & 1.55 & 550@xmath21428 + gc0127 & aat113428 & 13 25 07.33 & -43 06 20.6 & 6.38 & 21.25 & 20.92 & 20.04 & 19.49 & 18.93 & 0.16 & 0.04 & 0.01 & 0.01 & 0.03 & 21.14 & 20.36 & 19.51 & 0.02 & 0.01 & 0.01 & 0.33 & 0.88 & 0.54 & 1.11 & 0.85 & 0.78 & 1.63 & 657@xmath21467 + gc0128 & pff_gc-037 & 13 25 07.48 & -43 12 29.4 & 11.93 & 21.60 & 21.29 & 20.42 & 19.91 & 19.33 & 0.55 & 0.05 & 0.02 & 0.01 & 0.03 & 21.46 & 20.73 & 19.90 & 0.02 & 0.01 & 0.01 & 0.31 & 0.87 & 0.50 & 1.08 & 0.83 & 0.73 & 1.56 & 554@xmath21423 + gc0129 & whh-8 & 13 25 07.62 & -43 01 15.2 & 3.66 & 19.83 & 19.16 & 18.12 & 17.46 & 16.82 & 0.13 & 0.03 & 0.01 & 0.01 & 0.03 & 19.48 & 18.32 & 17.45 & 0.03 & 0.03 & 0.04 & 0.67 & 1.04 & 0.66 & 1.30 & 0.87 & 1.16 & 2.03 & 690@xmath21432 + gc0130 & whh-9 & 13 25 08.51 & -43 02 57.4 & 3.93 & 20.59 & 19.95 & 18.91 & 18.37 & 17.64 & 0.17 & 0.03 & 0.01 & 0.01 & 0.03 & 20.28 & 19.17 & 18.30 & 0.05 & 0.04 & 0.05 & 0.64 & 1.04 & 0.54 & 1.27 & 0.87 & 1.11 & 1.98 & 315@xmath214100 + gc0131 & c132 & 13 25 08.79 & -43 09 09.6 & 8.72 & - & - & - & - & - & - & - & - & - & - & 20.38 & 19.69 & 18.90 & 0.02 & 0.02 & 0.01 & - & - & - & - & 0.79 & 0.69 & 1.48 & - + gc0132 & r271 & 13 25 08.81 & -43 09 09.5 & 8.72 & 20.49 & 20.23 & 19.42 & 18.91 & 18.39 & 0.08 & 0.02 & 0.01 & 0.01 & 0.03 & 20.38 & 19.69 & 18.90 & 0.02 & 0.02 & 0.01 & 0.26 & 0.80 & 0.52 & 1.03 & 0.79 & 0.69 & 1.48 & 436@xmath21445 + gc0133 & pff_gc-038 & 13 25 08.82 & -43 04 14.9 & 4.63 & 20.83 & 20.29 & 19.29 & 18.69 & 18.03 & 0.17 & 0.04 & 0.01 & 0.01 & 0.03 & 20.58 & 19.57 & 18.71 & 0.03 & 0.03 & 0.03 & 0.54 & 1.00 & 0.61 & 1.26 & 0.87 & 1.01 & 1.88 & 431@xmath21419 + gc0134 & pff_gc-039 & 13 25 09.10 & -42 24 00.9 & 37.29 & 20.01 & 19.77 & 18.95 & 18.44 & 17.91 & 0.17 & 0.02 & 0.01 & 0.01 & 0.03 & 19.92 & 19.22 & 18.50 & 0.01 & 0.01 & 0.01 & 0.23 & 0.82 & 0.51 & 1.04 & 0.72 & 0.70 & 1.42 & 387@xmath21427 + gc0135 & k-029 & 13 25 09.19 & -42 58 59.2 & 4.00 & 19.13 & 18.65 & 17.71 & 17.13 & 16.55 & 0.08 & 0.02 & 0.01 & 0.01 & 0.03 & - & - & - & - & - & - & 0.48 & 0.94 & 0.58 & 1.16 & - & - & - & 677@xmath21444 + gc0136 & pff_gc-040 & 13 25 09.54 & -42 55 18.5 & 6.71 & 20.09 & 19.89 & 19.12 & 18.60 & 18.11 & 0.06 & 0.02 & 0.01 & 0.01 & 0.03 & 20.00 & 19.37 & 18.66 & 0.02 & 0.02 & 0.01 & 0.20 & 0.76 & 0.52 & 1.01 & 0.71 & 0.62 & 1.34 & 504@xmath21435 + gc0137 & hghh - g329 & 13 25 10.20 & -43 02 06.7 & 3.33 & 21.73 & 20.89 & 19.71 & 19.07 & 18.38 & 0.69 & 0.11 & 0.03 & 0.02 & 0.03 & 21.34 & 20.04 & 19.07 & 0.02 & 0.01 & 0.03 & 0.84 & 1.18 & 0.64 & 1.33 & 0.97 & 1.29 & 2.26 & 502@xmath21416 + gc0138 & k-033 & 13 25 10.25 & -42 55 09.5 & 6.78 & 21.07 & 20.47 & 19.46 & 18.90 & 18.21 & 0.13 & 0.03 & 0.01 & 0.01 & 0.03 & 20.78 & 19.77 & 18.86 & 0.01 & 0.01 & 0.01 & 0.60 & 1.01 & 0.57 & 1.25 & 0.91 & 1.02 & 1.93 & 582@xmath21433 + gc0139 & k-034 & 13 25 10.27 & -42 53 33.1 & 8.23 & 19.33 & 18.79 & 17.80 & 17.26 & 16.55 & 0.04 & 0.01 & 0.01 & 0.01 & 0.03 & 19.14 & 18.03 & 17.25 & 0.03 & 0.03 & 0.02 & 0.55 & 0.99 & 0.54 & 1.25 & 0.79 & 1.11 & 1.89 & 464@xmath21442 + gc0140 & hhh86 - 14/c14 & 13 25 10.49 & -42 44 52.6 & 16.57 & 19.23 & 18.84 & 17.94 & 17.40 & 16.79 & 0.03 & 0.01 & 0.01 & 0.01 & 0.03 & 19.06 & 18.16 & 17.41 & 0.01 & 0.02 & 0.01 & 0.39 & 0.90 & 0.54 & 1.15 & 0.75 & 0.90 & 1.65 & 705@xmath21410 + gc0141 & aat113992 & 13 25 10.51 & -43 03 24.0 & 3.85 & 21.95 & 21.39 & 20.40 & 19.79 & 19.19 & 0.58 & 0.11 & 0.03 & 0.03 & 0.04 & 21.77 & 20.78 & 19.82 & 0.03 & 0.03 & 0.01 & 0.57 & 0.99 & 0.61 & 1.20 & 0.97 & 0.99 & 1.95 & 648@xmath21458 + gc0142 & c133 & 13 25 11.05 & -43 01 32.3 & 3.05 & - & - & - & - & - & - & - & - & - & - & - & - & 19.30 & - & - & - & - & - & - & - & - & - & - & - + gc0143 & hghh - g348 & 13 25 11.10 & -42 58 03.0 & 4.32 & 19.90 & 19.66 & 18.92 & 18.46 & 18.03 & 0.10 & 0.03 & 0.01 & 0.01 & 0.03 & 19.80 & 19.15 & 18.45 & 0.01 & 0.01 & - & 0.24 & 0.74 & 0.46 & 0.89 & 0.70 & 0.65 & 1.35 & 416@xmath21430 + gc0144 & pff_gc-041 & 13 25 11.17 & -43 03 09.6 & 3.62 & 20.57 & 20.32 & 19.51 & 18.99 & 18.48 & 0.19 & 0.05 & 0.02 & 0.02 & 0.03 & 20.48 & 19.77 & 19.01 & 0.02 & 0.02 & 0.02 & 0.24 & 0.81 & 0.52 & 1.03 & 0.76 & 0.71 & 1.47 & 456@xmath21429 + gc0145 & hghh - g327 & 13 25 11.98 & -43 04 19.3 & 4.27 & 21.46 & 20.83 & 19.78 & 19.16 & 18.56 & 0.34 & 0.06 & 0.02 & 0.02 & 0.03 & 21.16 & 20.11 & 19.19 & 0.01 & 0.01 & 0.01 & 0.63 & 1.05 & 0.61 & 1.22 & 0.92 & 1.05 & 1.97 & 608@xmath21419 + gc0146 & hghh - g081 & 13 25 12.11 & -42 57 25.2 & 4.68 & 19.90 & 19.52 & 18.62 & 18.05 & 17.49 & 0.08 & 0.02 & 0.01 & 0.01 & 0.03 & 19.72 & 18.88 & 18.10 & 0.03 & 0.03 & 0.01 & 0.38 & 0.90 & 0.57 & 1.13 & 0.77 & 0.85 & 1.62 & 536@xmath21443 + gc0147 & pff_gc-042 & 13 25 12.21 & -43 16 33.9 & 15.67 & 19.91 & 19.74 & 18.97 & 18.47 & 17.91 & 0.12 & 0.02 & 0.01 & 0.01 & 0.03 & 19.82 & 19.19 & 18.47 & 0.02 & 0.02 & 0.01 & 0.17 & 0.78 & 0.50 & 1.05 & 0.72 & 0.63 & 1.35 & 627@xmath21423 + gc0148 & aat114302 & 13 25 12.34 & -42 58 07.7 & 4.11 & 21.87 & 21.39 & 20.44 & 19.90 & 19.35 & 0.53 & 0.11 & 0.03 & 0.03 & 0.04 & 21.64 & 20.76 & 19.97 & 0.02 & 0.01 & 0.01 & 0.47 & 0.95 & 0.55 & 1.09 & 0.79 & 0.88 & 1.67 & 754@xmath214143 + gc0149 & pff_gc-043 & 13 25 12.45 & -43 14 07.4 & 13.27 & 20.51 & 20.22 & 19.40 & 18.89 & 18.33 & 0.20 & 0.02 & 0.01 & 0.01 & 0.03 & 20.38 & 19.67 & 18.90 & 0.03 & 0.02 & 0.01 & 0.29 & 0.82 & 0.52 & 1.07 & 0.77 & 0.70 & 1.48 & 527@xmath21433 + gc0150 & whh-10 & 13 25 12.84 & -42 56 59.8 & 4.95 & 20.62 & 20.36 & 19.57 & 19.07 & 18.48 & 0.12 & 0.03 & 0.01 & 0.01 & 0.03 & 20.43 & 19.90 & 19.10 & 0.03 & 0.02 & 0.01 & 0.26 & 0.79 & 0.50 & 1.09 & 0.80 & 0.54 & 1.33 & 664@xmath214141 + gc0151 & r261 & 13 25 12.90 & -43 07 59.1 & 7.35 & 19.65 & 19.15 & 18.20 & 17.60 & 17.03 & 0.05 & 0.01 & 0.01 & 0.01 & 0.03 & 19.42 & 18.49 & 17.62 & 0.02 & 0.01 & 0.01 & 0.51 & 0.94 & 0.60 & 1.17 & 0.88 & 0.93 & 1.81 & 615@xmath2144 + gc0152 & pff_gc-044 & 13 25 13.19 & -43 16 35.6 & 15.67 & 20.29 & 20.11 & 19.37 & 18.87 & 18.33 & 0.16 & 0.02 & 0.01 & 0.01 & 0.03 & 20.19 & 19.59 & 18.87 & 0.02 & 0.02 & 0.01 & 0.17 & 0.74 & 0.50 & 1.04 & 0.72 & 0.60 & 1.32 & 568@xmath21453 + gc0153 & c134 & 13 25 13.20 & -43 02 31.3 & 2.97 & - & - & - & - & - & - & - & - & - & - & 22.45 & 21.64 & 20.71 & 0.07 & 0.05 & 0.07 & - & - & - & - & 0.93 & 0.80 & 1.74 & - + gc0154 & pff_gc-045 & 13 25 13.31 & -42 52 12.4 & 9.31 & 21.57 & 20.86 & 19.82 & 19.21 & 18.53 & 0.19 & 0.04 & 0.01 & 0.01 & 0.03 & 21.38 & 20.32 & 19.40 & 0.07 & 0.06 & 0.04 & 0.71 & 1.04 & 0.62 & 1.29 & 0.93 & 1.06 & 1.99 & 474@xmath21423 + gc0155 & r259 & 13 25 13.88 & -43 07 32.5 & 6.87 & 22.13 & 21.70 & 20.70 & 20.11 & 19.48 & 0.37 & 0.08 & 0.02 & 0.02 & 0.03 & 21.96 & 21.07 & 20.10 & 0.02 & 0.01 & 0.01 & 0.43 & 1.00 & 0.60 & 1.23 & 0.96 & 0.89 & 1.85 & 628@xmath21448 + gc0156 & hghh - g271 & 13 25 13.95 & -42 57 42.6 & 4.25 & 19.45 & 19.36 & 18.72 & 18.32 & 17.90 & 0.07 & 0.02 & 0.01 & 0.01 & 0.03 & 19.40 & 18.92 & 18.30 & 0.01 & 0.01 & - & 0.09 & 0.63 & 0.40 & 0.82 & 0.62 & 0.47 & 1.09 & 353@xmath21427 + gc0157 & c135 & 13 25 14.07 & -43 00 51.8 & 2.49 & - & - & - & - & - & - & - & - & - & - & - & - & 19.90 & - & - & - & - & - & - & - & - & - & - & - + gc0158 & c136 & 13 25 14.07 & -43 03 35.0 & 3.47 & - & - & - & - & - & - & - & - & - & - & 22.48 & 21.50 & 21.19 & 0.04 & 0.02 & 0.02 & - & - & - & - & 0.30 & 0.99 & 1.29 & - + gc0159 & whh-11/k-051 & 13 25 14.24 & -43 07 23.5 & 6.71 & 21.26 & 20.63 & 19.55 & 18.90 & 18.21 & 0.18 & 0.04 & 0.01 & 0.01 & 0.03 & 21.02 & 19.95 & 18.95 & 0.03 & 0.02 & 0.01 & 0.64 & 1.07 & 0.65 & 1.34 & 1.00 & 1.06 & 2.07 & 582@xmath21430 + gc0160 & pff_gc-046 & 13 25 14.83 & -43 41 10.6 & 40.10 & 19.91 & 19.72 & 18.93 & 18.43 & 17.95 & 0.11 & 0.02 & 0.01 & 0.01 & 0.03 & 19.87 & 19.14 & 18.37 & 0.01 & - & 0.01 & 0.20 & 0.79 & 0.50 & 0.98 & 0.77 & 0.73 & 1.50 & 532@xmath21421 + gc0161 & pff_gc-047 & 13 25 15.12 & -42 50 30.4 & 10.88 & 20.92 & 20.72 & 19.87 & 19.37 & 18.87 & 0.09 & 0.03 & 0.01 & 0.01 & 0.03 & 20.87 & 20.22 & 19.41 & 0.02 & 0.02 & 0.01 & 0.20 & 0.85 & 0.50 & 0.99 & 0.81 & 0.65 & 1.46 & 717@xmath21455 + gc0162 & aat114769 & 13 25 15.12 & -42 57 45.7 & 4.08 & 21.20 & 21.00 & 20.25 & 19.76 & 19.25 & 0.30 & 0.09 & 0.03 & 0.03 & 0.04 & 21.11 & 20.52 & 19.80 & 0.02 & 0.02 & 0.01 & 0.20 & 0.75 & 0.49 & 1.00 & 0.71 & 0.59 & 1.30 & 650@xmath214169 + gc0163 & r257 & 13 25 15.24 & -43 08 39.2 & 7.84 & 22.29 & 21.77 & 20.78 & 20.18 & 19.58 & 0.35 & 0.07 & 0.02 & 0.02 & 0.03 & 22.02 & 21.09 & 20.16 & 0.02 & 0.02 & 0.02 & 0.51 & 1.00 & 0.60 & 1.20 & 0.92 & 0.93 & 1.85 & 339@xmath21451 + gc0164 & pff_gc-048 & 13 25 15.79 & -42 49 15.1 & 12.09 & 20.92 & 20.67 & 19.85 & 19.30 & 18.78 & 0.09 & 0.03 & 0.01 & 0.01 & 0.03 & 20.85 & 20.13 & 19.41 & 0.02 & 0.02 & 0.02 & 0.25 & 0.82 & 0.55 & 1.07 & 0.71 & 0.72 & 1.44 & 535@xmath21453 + gc0165 & aat114913 & 13 25 15.93 & -43 06 03.3 & 5.35 & 21.69 & 21.33 & 20.39 & 19.88 & 19.29 & 0.34 & 0.08 & 0.02 & 0.02 & 0.04 & 21.60 & 20.72 & 19.86 & 0.02 & 0.02 & 0.02 & 0.36 & 0.94 & 0.52 & 1.10 & 0.86 & 0.88 & 1.74 & 563@xmath21463 + gc0166 & pff_gc-049 & 13 25 16.06 & -43 05 06.5 & 4.49 & 20.74 & 20.45 & 19.59 & 19.05 & 18.50 & 0.15 & 0.04 & 0.01 & 0.01 & 0.03 & 20.64 & 19.85 & 19.07 & 0.03 & 0.03 & 0.03 & 0.29 & 0.86 & 0.54 & 1.09 & 0.78 & 0.79 & 1.57 & 674@xmath21432 + gc0167 & c137 & 13 25 16.06 & -43 02 19.3 & 2.42 & - & - & - & - & - & - & - & - & - & - & - & - & 19.04 & - & - & - & - & - & - & - & - & - & - & - + gc0168 & vhh81 - 05/c5 & 13 25 16.12 & -42 52 58.2 & 8.44 & 18.70 & 18.49 & 17.68 & 17.13 & 16.68 & 0.03 & 0.01 & 0.01 & 0.01 & 0.03 & - & - & - & - & - & - & 0.21 & 0.81 & 0.55 & 1.00 & - & - & - & 555@xmath2143 + gc0169 & hch01 & 13 25 16.22 & -42 59 43.4 & 2.52 & 18.69 & 18.36 & 17.47 & 16.94 & 16.40 & 0.52 & 0.15 & 0.04 & 0.03 & 0.04 & - & - & - & - & - & - & 0.33 & 0.89 & 0.53 & 1.07 & - & - & - & 649@xmath21445 + gc0170 & hhh86 - 33/c33 & 13 25 16.26 & -42 50 53.3 & 10.47 & 19.58 & 19.34 & 18.50 & 18.02 & 17.50 & 0.04 & 0.01 & 0.01 & 0.01 & 0.03 & - & - & - & - & - & - & 0.24 & 0.84 & 0.48 & 1.00 & - & - & - & 596@xmath21418 + gc0171 & aat114993 & 13 25 16.44 & -43 03 33.1 & 3.16 & 22.49 & 21.50 & 20.44 & 19.82 & 19.32 & 1.41 &
0.19&.382 & 0.308 & 0.224 & 21.95 & 20.81 & 19.83 & 0.02 & 0.02 & 0.03 & 0.99 & 1.06 & 0.62 & 1.12 & 0.98 & 1.14 & 2.12 & 352@xmath214136 + gc0172 & hch02 & 13 25 16.69 & -43 02 08.7 & 2.23 & 19.27 & 18.87 & 17.93 & 17.37 & 16.76 & 0.98 & 0.26 & 0.07 & 0.05 & 0.05 & - & - & - & - & - & - & 0.40 & 0.94 & 0.56 & 1.17 & - & - & - & 300@xmath2142 + gc0173 & pff_gc-050 & 13 25 16.73 & -42 50 18.4 & 11.02 & 21.63 & 21.28 & 20.39 & 19.86 & 19.27 & 0.16 & 0.04 & 0.01 & 0.01 & 0.03 & - & - & - & - & - & - & 0.35 & 0.89 & 0.54 & 1.12 & - & - & - & 688@xmath21470 + gc0174 & c138 & 13 25 16.91 & -43 03 08.0 & 2.79 & - & - & - & - & - & - & - & - & - & - & 21.53 & 20.79 & 19.97 & 0.02 & 0.02 & 0.03 & - & - & - & - & 0.81 & 0.74 & 1.55 & - + gc0175 & r254 & 13 25 16.96 & -43 09 28.0 & 8.54 & 20.52 & 20.26 & 19.43 & 18.91 & 18.33 & 0.09 & 0.03 & 0.01 & 0.01 & 0.03 & 20.42 & 19.65 & 18.91 & 0.04 & 0.03 & 0.03 & 0.27 & 0.82 & 0.52 & 1.10 & 0.73 & 0.77 & 1.51 & 561@xmath21427 + gc0176 & c139 & 13 25 17.06 & -43 02 44.6 & 2.50 & - & - & - & - & - & - & - & - & - & - & 20.73 & 19.62 & 18.86 & 0.03 & 0.02 & 0.04 & - & - & - & - & 0.76 & 1.11 & 1.87 & - + gc0177 & hghh - g219 & 13 25 17.31 & -42 58 46.6 & 3.03 & 20.69 & 19.92 & 18.84 & 18.20 & 17.55 & 0.43 & 0.08 & 0.02 & 0.02 & 0.03 & - & - & - & - & - & - & 0.77 & 1.09 & 0.63 & 1.29 & - & - & - & 535@xmath21424 + gc0178 & r253 & 13 25 17.33 & -43 08 39.0 & 7.74 & 20.75 & 20.49 & 19.67 & 19.14 & 18.64 & 0.09 & 0.03 & 0.01 & 0.01 & 0.03 & 20.68 & 20.00 & 19.47 & 0.03 & 0.03 & 0.05 & 0.26 & 0.81 & 0.53 & 1.04 & 0.53 & 0.69 & 1.22 & 486@xmath21452 + gc0179 & c140 & 13 25 17.42 & -43 03 25.2 & 2.94 & - & - & - & - & - & - & - & - & - & - & 21.40 & 20.65 & 19.83 & 0.03 & 0.02 & 0.03 & - & - & - & - & 0.82 & 0.75 & 1.57 & - + gc0180 & c141 & 13 25 18.14 & -43 02 50.9 & 2.43 & - & - & - & - & - & - & - & - & - & - & 22.23 & 21.55 & 20.95 & 0.06 & 0.05 & 0.06 & - & - & - & - & 0.60 & 0.68 & 1.29 & - + gc0181 & whh-12 & 13 25 18.27 & -42 53 04.8 & 8.25 & 20.36 & 20.00 & 19.07 & 18.47 & 17.97 & 0.07 & 0.02 & 0.01 & 0.01 & 0.03 & 20.23 & 19.35 & 18.71 & 0.04 & 0.04 & 0.01 & 0.36 & 0.93 & 0.60 & 1.10 & 0.64 & 0.89 & 1.52 & 558@xmath21499 + gc0182 & aat115339 & 13 25 18.44 & -43 04 09.8 & 3.45 & 21.19 & 20.91 & 20.11 & 19.53 & 19.01 & 0.39 & 0.10 & 0.03 & 0.03 & 0.04 & 21.11 & 20.38 & 19.56 & 0.02 & 0.01 & 0.02 & 0.27 & 0.81 & 0.58 & 1.09 & 0.81 & 0.73 & 1.55 & 618@xmath214108 + gc0183 & c142 & 13 25 18.50 & -43 01 16.4 & 1.67 & - & - & - & - & - & - & - & - & - & - & - & - & 17.64 & - & - & - & - & - & - & - & - & - & - & - + gc0184 & aat320656 & 13 25 19.13 & -43 12 03.8 & 11.03 & 20.61 & 20.50 & 19.81 & 19.35 & 18.89 & 0.08 & 0.03 & 0.01 & 0.01 & 0.03 & 20.57 & 20.01 & 19.35 & 0.02 & 0.01 & 0.02 & 0.10 & 0.70 & 0.46 & 0.92 & 0.66 & 0.56 & 1.22 & 453@xmath214233 + gc0185 & hghh - g331 & 13 25 19.50 & -43 02 28.4 & 1.99 & 20.00 & 19.60 & 18.92 & 18.47 & 17.93 & 0.55 & 0.14 & 0.05 & 0.04 & 0.05 & - & - & - & - & - & - & 0.40 & 0.68 & 0.44 & 0.99 & - & - & - & 371@xmath21422 + gc0186 & aat115561 & 13 25 19.83 & -42 58 27.2 & 3.05 & 22.13 & 21.47 & 20.50 & 19.92 & 19.29 & 1.44 & 0.26 & 0.07 & 0.06 & 0.06 & - & 20.98 & 20.15 & 1.00 & 0.04 & 0.02 & 0.66 & 0.98 & 0.57 & 1.20 & 0.83 & - & - & 514@xmath21446 + gc0187 & r247 & 13 25 19.99 & -43 07 44.1 & 6.73 & 21.86 & 21.20 & 20.11 & 19.40 & 18.62 & 0.28 & 0.05 & 0.02 & 0.01 & 0.03 & 21.73 & 20.65 & 19.75 & 0.05 & 0.04 & 0.04 & 0.66 & 1.10 & 0.70 & 1.49 & 0.90 & 1.08 & 1.98 & 662@xmath21448 + gc0188 & aat115605 & 13 25 20.44 & -42 54 08.5 & 7.13 & 21.44 & 21.22 & 20.47 & 20.00 & 19.50 & 0.18 & 0.05 & 0.02 & 0.02 & 0.03 & 21.36 & 20.76 & 20.11 & 0.01 & 0.03 & 0.01 & 0.22 & 0.75 & 0.47 & 0.97 & 0.66 & 0.60 & 1.25 & 713@xmath214131 + gc0189 & aat115679 & 13 25 20.72 & -43 06 35.9 & 5.60 & 21.93 & 21.49 & 20.44 & 19.80 & 19.15 & 0.40 & 0.09 & 0.02 & 0.02 & 0.03 & 21.71 & 20.79 & 19.84 & 0.02 & 0.02 & 0.03 & 0.44 & 1.05 & 0.64 & 1.30 & 0.95 & 0.92 & 1.87 & 511@xmath214165 + gc0190 & whh-13/hh-090 & 13 25 21.29 & -42 49 17.7 & 11.91 & 20.75 & 20.22 & 19.20 & 18.61 & 17.95 & 0.08 & 0.02 & 0.01 & 0.01 & 0.03 & 20.59 & 19.52 & 18.74 & 0.03 & 0.03 & 0.02&0.53 & 1.02 & 0.59 & 1.25 & 0.78 & 1.08 & 1.85 & 444@xmath21433 + gc0191 & aat321194 & 13 25 21.32 & -43 23 59.3 & 22.87 & 21.46 & 21.03 & 20.06 & 19.48 & 18.82 & 0.38 & 0.03 & 0.01 & 0.01 & 0.03 & 21.33 & 20.42 & 19.53 & 0.03 & 0.02 & 0.01 & 0.43 & 0.97 & 0.58 & 1.24 & 0.90 & 0.90 & 1.80 & 254@xmath214169 + gc0192 & hghh-06/c6 & 13 25 22.19 & -43 02 45.6 & 1.89 & 18.65 & 18.17 & 17.21 & 16.61 & 16.03 & 0.24 & 0.06 & 0.02 & 0.01 & 0.03 & - & - & - & - & - & - & 0.48 & 0.96 & 0.60 & 1.18 & - & - & - & 855@xmath2142 + gc0193 & pff_gc-051 & 13 25 22.35 & -43 15 00.1 & 13.89 & 21.15 & 20.73 & 19.79 & 19.20 & 18.61 & 0.10 & 0.02 & 0.01 & 0.01 & 0.03 & 21.00 & 20.10 & 19.22 & 0.03 & 0.03 & 0.02 & 0.42 & 0.94 & 0.59 & 1.18 & 0.88 & 0.90 & 1.78 & 468@xmath21439 + gc0194 & c143 & 13 25 23.20 & -43 03 12.9 & 2.22 & - & - & - & - & - & - & - & - & - & - & 21.87 & 21.35 & 20.52 & 0.05 & 0.04 & 0.06 & - & - & - & - & 0.82 & 0.52 & 1.34 & - + gc0195 & aat116025 & 13 25 23.46 & -42 53 26.2 & 7.75 & 21.23 & 20.88 & 20.00 & 19.48 & 18.85 & 0.14 & 0.04 & 0.01 & 0.01 & 0.03 & 21.16 & 20.32 & 19.62 & 0.04 & 0.05 & 0.01 & 0.35 & 0.88 & 0.51 & 1.15 & 0.70 & 0.84 & 1.54 & 545@xmath21464 + gc0196 & aat116220 & 13 25 24.40 & -43 07 58.9 & 6.86 & 20.76 & 20.39 & 19.47 & 18.91 & 18.36 & 0.11 & 0.03 & 0.01 & 0.01 & 0.03 & 20.62 & 19.72 & 18.92 & 0.03 & 0.03 & 0.03 & 0.37 & 0.92 & 0.56 & 1.11 & 0.80 & 0.90 & 1.70 & 524@xmath21441 + gc0197 & aat116385 & 13 25 25.39 & -42 58 21.5 & 2.82 & 21.71 & 21.23 & 20.26 & 19.59 & 18.99 & 1.22 & 0.26 & 0.07 & 0.05 & 0.06 & 21.58 & 20.68 & 19.94 & 0.02 & 0.02 & 0.01 & 0.48 & 0.97 & 0.66 & 1.27 & 0.74 & 0.90 & 1.64 & 550@xmath21440 + gc0198 & whh-14 & 13 25 25.49 & -42 56 31.2 & 4.64 & 20.99 & 20.83 & 20.06 & 19.56 & 19.05 & 0.18 & 0.05 & 0.02 & 0.02 & 0.03 & 20.98 & 20.36 & 19.68 & 0.02 & 0.03 & 0.01 & 0.15 & 0.77 & 0.50 & 1.01 & 0.68 & 0.62 & 1.30 & 461@xmath21475 + gc0199 & aat204119 & 13 25 25.70 & -42 37 40.9 & 23.47 & 20.59 & 20.49 & 19.81 & 19.35 & 18.93 & 0.18 & 0.02 & 0.01 & 0.01 & 0.03 & 20.60 & 20.04 & 19.42 & 0.01 & - & - & 0.10 & 0.67 & 0.47 & 0.88 & 0.62 & 0.56 & 1.18 & 404@xmath21474 + gc0200 & pff_gc-052 & 13 25 25.75 & -43 05 16.5 & 4.14 & 20.88 & 20.70 & 19.88 & 19.39 & 18.86 & 0.19 & 0.06 & 0.02 & 0.02 & 0.03 & 20.86 & 20.17 & 19.42 & 0.03 & 0.03 & 0.03 & 0.18 & 0.82 & 0.48 & 1.01 & 0.75 & 0.69 & 1.44 & 462@xmath21452 + gc0201 & hghh-46/c46 & 13 25 25.97 & -43 03 25.7 & 2.30 & 19.74 & 19.71 & 19.10 & 18.74 & 18.30 & 0.37 & 0.12 & 0.04 & 0.04 & 0.05 & 19.77 & 19.30 & 18.64 & 0.01 & 0.02 & 0.22 & 0.03 & 0.61 & 0.36 & 0.80 & 0.66 & 0.47 & 1.13 & 508@xmath21419 + gc0202 & c144 & 13 25 26.28 & -43 04 38.5 & 3.50 & - & - & - & - & - & - & - & - & - & - & 23.22 & 22.36 & 21.96 & 0.06 & 0.05 & 0.05 & - & - & - & - & 0.41 & 0.85 & 1.26 & - + gc0203 & aat116531 & 13 25 26.75 & -43 08 53.4 & 7.74 & 20.92 & 20.65 & 19.83 & 19.34 & 18.83 & 0.10 & 0.03 & 0.01 & 0.01 & 0.03 & 20.77 & 19.95 & 19.94 & 0.07 & 0.06 & 0.02 & 0.27 & 0.82 & 0.49 & 1.00 & 0.01 & 0.82 & 0.83 & 267@xmath21471 + gc0204 & whh-15 & 13 25 26.78 & -42 52 39.9 & 8.48 & 20.73 & 20.56 & 19.81 & 19.36 & 18.87 & 0.09 & 0.03 & 0.01 & 0.01 & 0.03 & 20.68 & 20.80 & 19.53 & 0.03 & 0.02 & 0.01 & 0.16 & 0.75 & 0.46 & 0.94 & 1.27&-0.12 & 1.15 & 513@xmath21453 + gc0205 & r235 & 13 25 26.82 & -43 09 40.5 & 8.53 & 20.55 & 20.30 & 19.53 & 19.02 & 18.51 & 0.08 & 0.02 & 0.01 & 0.01 & 0.03 & 20.42 & 19.73 & 19.01 & 0.04 & 0.03 & 0.04 & 0.25 & 0.77 & 0.51 & 1.03 & 0.72 & 0.69 & 1.41 & 498@xmath21428 + gc0206 & whh-16/k-102 & 13 25 27.97 & -43 04 02.2 & 2.89 & 20.90 & 20.23 & 19.18 & 18.56 & 17.90 & 0.56 & 0.11 & 0.03 & 0.02 & 0.04 & 20.64 & 19.52 & 18.60 & 0.04 & 0.03 & 0.03 & 0.66 & 1.06 & 0.61 & 1.28 & 0.93 & 1.11 & 2.04 & 661@xmath21447 + gc0207 & hch13 & 13 25 28.69 & -43 02 55.0 & 1.78 & 21.32 & 20.82 & 19.89 & 19.30 & 18.63 & 2.53 & 0.55 & 0.15 & 0.12 & 0.12 & - & - & - & - & - & - & 0.50 & 0.93 & 0.59 & 1.26 & - & - & - & 641@xmath21424 + gc0208 & c145 & 13 25 28.81 & -43 04 21.6 & 3.22 & - & - & - & - & - & - & - & - & - & - & 19.32 & 18.54 & 17.81 & 0.01 & 0.01 & 0.02 & - & - & - & - & 0.73 & 0.78 & 1.51 & - + gc0209 & whh-17 & 13 25 29.25 & -42 57 47.1 & 3.38 & 19.68 & 19.39 & 18.57 & 18.06 & 17.48 & 0.15 & 0.04 & 0.02 & 0.01 & 0.03 & 19.54 & 18.77 & 18.13 & 0.02 & 0.03 & - & 0.29 & 0.82 & 0.51 & 1.09 & 0.65 & 0.76 & 1.41 & 619@xmath21445 + gc0210 & aat116969 & 13 25 29.41 & -42 53 25.6 & 7.73 & 21.55 & 21.28 & 20.46 & 20.10 & 19.48 & 0.21 & 0.06 & 0.02 & 0.02 & 0.03 & 21.45 & 20.77 & 20.10 & 0.03 & 0.02 & 0.01 & 0.27 & 0.82 & 0.35 & 0.98 & 0.68 & 0.68 & 1.36 & 446@xmath21461 + gc0211 & hghh - g169 & 13 25 29.43 & -42 58 09.9 & 3.00 & 21.39 & 20.58 & 19.49 & 18.92 & 18.12 & 0.81 & 0.13 & 0.03 & 0.03 & 0.04 & 21.00 & 19.80 & 18.94 & 0.01 & 0.02 & 0.01 & 0.81 & 1.10 & 0.57 & 1.37 & 0.86 & 1.21 & 2.07 & 643@xmath21424 + gc0212 & pff_gc-053 & 13 25 29.62 & -42 54 44.5 & 6.42 & 20.91 & 20.29 & 19.26 & 18.73 & 17.98 & 0.14 & 0.03 & 0.01 & 0.01 & 0.03 & 20.65 & 19.54 & 18.78 & 0.03 & 0.04 & 0.01 & 0.62 & 1.03 & 0.53 & 1.29 & 0.76 & 1.11 & 1.86 & 439@xmath21420 + gc0213 & r229 & 13 25 29.74 & -43 11 42.8 & 10.57 & 20.70 & 20.49 & 19.72 & 19.20 & 18.69 & 0.08 & 0.02 & 0.01 & 0.01 & 0.03 & 20.62 & 19.94 & 19.22 & 0.04 & 0.03 & 0.04 & 0.21 & 0.77 & 0.52 & 1.03 & 0.73 & 0.68 & 1.41 & 517@xmath21466 + gc0214 & hch15 & 13 25 29.80 & -43 00 07.0 & 1.10 & 19.27 & 18.87 & 17.93 & 17.37 & 16.76 & 0.98 & 0.26 & 0.07 & 0.05 & 0.05 & - & - & - & - & - & - & 0.04 & 0.94 & 0.56 & 1.17 & - & - & - & 519@xmath2141 + gc0215 & c146 & 13 25 29.87 & -43 05 09.2 & 4.03 & - & - & - & - & - & - & - & - & - & - & 21.71 & 20.78 & 19.94 & 0.03 & 0.02 & 0.02 & - & - & - & - & 0.84 & 0.93 & 1.77 & - + gc0216 & whh-18 & 13 25 30.07 & -42 56 46.9 & 4.39 & 19.85 & 19.36 & 18.36 & 17.85 & 17.16 & 0.09 & 0.02 & 0.01 & 0.01 & 0.03 & 19.63 & 18.60 & 17.88 & 0.01 & 0.03 & - & 0.50 & 0.99 & 0.51 & 1.20 & 0.72 & 1.03 & 1.75 & 752@xmath21425 + gc0217 & pff_gc-054 & 13 25 30.28 & -43 41 53.6 & 40.75 & 20.65 & 20.60 & 19.99 & 19.57 & 19.09 & 0.19 & 0.03 & 0.01 & 0.01 & 0.03 & - & - & - & - & - & - & 0.05 & 0.62 & 0.41 & 0.89 & - & - & - & 297@xmath21440 + gc0218 & hch16 & 13 25 30.29 & -42 59 34.8 & 1.64 & - & - & - & - & - & - & - & - & - & - & - & - & - & - & - & - & - & - & - & - & - & - & - & 458@xmath2142 + gc0219 & hhh86 - 15/c15/r226 & 13 25 30.41 & -43 11 49.6 & 10.69 & 20.11 & 19.56 & 18.56 & 17.94 & 17.35 & 0.06 & 0.02 & 0.01 & 0.01 & 0.03 & 19.80 & 18.68 & 17.92 & 0.05 & 0.05 & 0.05 & 0.55 & 1.00 & 0.62 & 1.21 & 0.76 & 1.12 & 1.88 & 644@xmath2141 + gc0220 & c147 & 13 25 30.65 & -43 03 47.1 & 2.70 & - & - & - & - & - & - & - & - & - & - & 21.58 & 20.88 & 20.05 & 0.05 & 0.05 & 0.03 & - & - & - & - & 0.83 & 0.69 & 1.52 & - + gc0221 & pff_gc-055 & 13 25 30.72 & -42 48 13.4 & 12.94 & 20.87 & 20.68 & 19.92 & 19.44 & 18.97 & 0.08 & 0.03 & 0.01 & 0.01 & 0.03 & 20.83 & 20.20 & 19.54 & 0.02 & 0.02 & 0.02 & 0.19 & 0.76 & 0.48 & 0.96 & 0.66 & 0.62 & 1.28 & 485@xmath21447 + gc0222 & aat205071 & 13 25 30.74 & -42 30 16.0 & 30.89 & 19.65 & 19.61 & 18.86 & 18.37 & 17.96 & 0.09 & 0.01 & 0.01 & 0.01 & 0.03 & 19.68 & 19.10 & 18.50 & 0.02 & 0.02 & - & 0.04 & 0.74 & 0.49 & 0.90 & 0.60 & 0.58 & 1.18 & 288@xmath21461 + gc0223 & whh-19 & 13 25 31.03 & -42 50 14.9 & 10.92 & 18.47 & 18.16 & 17.29 & 16.78 & 16.24 & 0.02 & 0.01 & 0.01 & 0.01 & 0.03 & 18.08 & 17.67 & 16.91 & 0.06 & 0.06 & 0.02 & 0.31 & 0.87 & 0.51 & 1.05 & 0.76 & 0.42 & 1.18 & 451@xmath21440 + gc0224 & aat117287 & 13 25 31.08 & -43 04 17.0 & 3.20 & 22.20 & 21.44 & 20.36 & 19.82 & 19.13 & 1.22 & 0.20 & 0.05 & 0.04 & 0.05 & 21.82 & 20.56 & 20.45 & 0.04 & 0.03 & 0.03 & 0.76 & 1.08 & 0.54 & 1.23 & 0.11 & 1.27 & 1.37 & 554@xmath21460 + gc0225 & hghh - g292 & 13 25 31.48 & -42 58 08.3 & 3.09 & 21.77 & 21.05 & 19.97 & 19.33 & 18.64 & 1.04 & 0.18 & 0.05 & 0.03 & 0.04 & 21.48 & 20.85 & 19.53 & 0.01 & 0.02 & 0.01 & 0.72 & 1.08 & 0.64 & 1.34 & 1.33 & 0.63 & 1.95 & 655@xmath21443 + gc0226 & hch18 & 13 25 31.60 & -43 00 02.8 & 1.32 & 21.06 & 20.99 & 19.93 & 19.20 & 18.43 & 2.42 & 0.76 & 0.19 & 0.13 & 0.12 & - & - & - & - & - & - & 0.07 & 1.06 & 0.73 & 1.50 & - & - & - & 455@xmath2141 + gc0227 & aat117322 & 13 25 31.73 & -42 55 15.7 & 5.93 & 21.14 & 20.91 & 20.08 & 19.55 & 19.07 & 0.19 & 0.05 & 0.02 & 0.02 & 0.03 & 21.07 & 20.38 & 19.75 & 0.04 & 0.05 & 0.01 & 0.22 & 0.84 & 0.52 & 1.01 & 0.63 & 0.69 & 1.32 & 636@xmath214133 + gc0228 & hghh-44/c44 & 13 25 31.73 & -43 19 22.6 & 18.25 & 19.76 & 19.50 & 18.69 & 18.14 & 17.66 & 0.04 & 0.01 & 0.01 & 0.01 & 0.03 & 19.59 & 18.85 & 18.15 & 0.03 & 0.02 & 0.01 & 0.26 & 0.80 & 0.55 & 1.03 & 0.70 & 0.74 & 1.44 & 505@xmath2141 + gc0229 & c148 & 13 25 31.75 & -43 05 46.0 & 4.68 & - & - & - & - & - & - & - & - & - & - & 21.25 & 20.79 & 20.21 & 0.07 & 0.06 & 0.05 & - & - & - & - & 0.58 & 0.46 & 1.04 & - + gc0230 & r224/c149 & 13 25 32.33 & -43 07 17.1 & 6.20 & 21.07 & 20.90 & 20.15 & 19.64 & 19.19 & 0.15 & 0.04 & 0.02 & 0.02 & 0.03 & 21.02 & 20.42 & 19.68 & 0.03 & 0.02 & 0.02 & 0.16 & 0.76 & 0.50 & 0.96 & 0.74 & 0.60 & 1.34 & 389@xmath21445 + gc0231 & hghh - g359 & 13 25 32.42 & -42 58 50.2 & 2.47 & 20.44 & 19.86 & 18.86 & 18.24 & 17.64 & 0.45 & 0.10 & 0.03 & 0.02 & 0.04 & 20.13 & 19.09 & 18.33 & 0.01 & 0.01 & - & 0.58 & 1.00 & 0.62 & 1.22 & 0.76 & 1.04 & 1.80 & 489@xmath21434 + gc0232 & pff_gc-056 & 13 25 32.80 & -42 56 24.4 & 4.83 & 19.62 & 19.43 & 18.64 & 18.15 & 17.64 & 0.06 & 0.02 & 0.01 & 0.01 & 0.03 & 19.51 & 18.82 & 18.22 & 0.03 & 0.04 & - & 0.19 & 0.79 & 0.48 & 1.00 & 0.60 & 0.69 & 1.29 & 306@xmath21427 + gc0233 & r223 & 13 25 32.80 & -43 07 02.2 & 5.97 & 20.08 & 19.67 & 18.77 & 18.23 & 17.64 & 0.07 & 0.02 & 0.01 & 0.01 & 0.03 & 19.89 & 18.98 & 18.19 & 0.03 & 0.03 & 0.03 & 0.41 & 0.91 & 0.54 & 1.13 & 0.79 & 0.92 & 1.71 & 776@xmath2141 + gc0234 & k-131 & 13 25 32.88 & -43 04 29.2 & 3.48 & 21.23 & 20.44 & 19.37 & 18.77 & 18.12 & 0.38 & 0.07 & 0.02 & 0.02 & 0.03 & 20.84 & 19.65 & 18.74 & 0.03 & 0.02 & 0.03 & 0.79 & 1.08 & 0.59 & 1.24 & 0.91 & 1.20 & 2.10 & 639@xmath21444 + gc0235 & pff_gc-057 & 13 25 33.17 & -42 59 03.2 & 2.33 & 22.21 & 21.38 & 20.35 & 19.75 & 18.95 & 2.11 & 0.37 & 0.09 & 0.07 & 0.07 & 21.67 & 20.75 & 19.86 & 0.03 & 0.02 & 0.01 & 0.82 & 1.04 & 0.60 & 1.40 & 0.90 & 0.91 & 1.81 & 515@xmath21412 + gc0236 & c150 & 13 25 33.82 & -43 02 49.6 & 2.03 & - & - & - & - & - & - & - & - & - & - & - & - & 19.78 & - & - & - & - & - & - & - & - & - & - & - + gc0237 & c151 & 13 25 33.93 & -43 03 51.4 & 2.94 & - & - & - & - & - & - & - & - & - & - & 22.27 & 20.90 & 19.95 & 0.03 & 0.01 & 0.03 & - & - & - & - & 0.95 & 1.37 & 2.32 & - + gc0238 & magjm-11 & 13 25 33.94 & -42 59 39.4 & 1.89 & 19.80 & 19.64 & 18.87 & 18.44 & 17.83 & 0.80 & 0.25 & 0.07 & 0.06 & 0.07 & - & - & - & - & - & - & 0.16 & 0.77 & 0.43 & 1.04 & - & - & - & 444@xmath21417 + gc0239 & hghh - g206 & 13 25 34.10 & -42 59 00.7 & 2.44 & 20.68 & 20.09 & 19.10 & 18.58 & 17.91 & 0.56 & 0.12 & 0.03 & 0.03 & 0.04 & 20.37 & 19.36 & 18.61 & 0.02 & 0.03 & 0.01 & 0.58 & 0.99 & 0.52 & 1.19 & 0.75 & 1.00 & 1.75 & 600@xmath21424 + gc0240 & hghh-45/c45 & 13 25 34.25 & -42 56 59.1 & 4.33 & 19.99 & 19.81 & 19.03 & 18.60 & 18.05 & 0.10 & 0.03 & 0.01 & 0.01 & 0.03 & 19.91 & 19.17 & 18.67 & 0.08 & 0.07 & 0.02 & 0.18 & 0.78 & 0.43 & 0.98 & 0.50 & 0.75 & 1.25 & 612@xmath21434 + gc0241 & whh-20 & 13 25 34.36 & -42 51 05.9 & 10.12 & 20.30 & 19.93 & 19.01 & 18.52 & 17.90 & 0.06 & 0.02 & 0.01 & 0.01 & 0.03 & 20.11 & 19.25 & 18.59 & 0.04 & 0.05 & 0.01 & 0.37 & 0.92 & 0.49 & 1.10 & 0.66 & 0.86 & 1.52 & 259@xmath21433 + gc0242 & c152 & 13 25 34.64 & -43 03 16.4 & 2.48 & - & - & - & - & - & - & - & - & - & - & - & - & 17.82 & - & - & - & - & - & - & - & - & - & - & - + gc0243 & c153 & 13 25 34.64 & -43 03 27.8 & 2.65 & - & - & - & - & - & - & - & - & - & - & - & - & 18.23 & - & - & - & - & - & - & - & - & - & - & - + gc0244 & hch21 & 13 25 34.65 & -43 03 27.7 & 2.65 & - & - & - & - & - & - & - & - & - & - & - & - & - & - & - & - & - & - & - & - & - & - & - & 663@xmath2142 + gc0245 & c154 & 13 25 34.71 & -43 03 30.2 & 2.69 & - & - & - & - & - & - & - & - & - & - & - & - & 19.58 & - & - & - & - & - & - & - & - & - & - & - + gc0246 & hhh86 - 16/c16 & 13 25 35.00 & -42 36 05.0 & 25.10 & 20.01 & 19.57 & 18.59 & 18.00 & 17.41 & 0.12 & 0.01 & 0.01 & 0.01 & 0.03 & 19.85 & 18.83 & 18.12 & 0.03 & 0.03 & 0.01 & 0.44 & 0.98 & 0.59 & 1.19 & 0.70 & 1.02 & 1.73 & 538@xmath21416 + gc0247 & pff_gc-058 & 13 25 35.12 & -42 56 45.3 & 4.60 & 19.67 & 19.47 & 18.68 & 18.19 & 17.70 & 0.07 & 0.02 & 0.01 & 0.01 & 0.03 & 19.55 & 18.85 & 18.29 & 0.03 & 0.03 & 0.01 & 0.20 & 0.78 & 0.49 & 0.99 & 0.56 & 0.70 & 1.26 & 365@xmath21422 + gc0248 & k-144 & 13 25 35.16 & -42 53 01.0 & 8.25 & 22.10 & 21.31 & 20.25 & 19.63 & 18.94 & 0.33 & 0.05 & 0.02 & 0.01 & 0.03 & 21.72 & 20.44 & 19.75 & 0.01 & 0.01 & 0.01 & 0.79 & 1.06 & 0.62 & 1.31 & 0.69 & 1.28 & 1.97 & 593@xmath21443 + gc0249 & whh-21 & 13 25 35.22 & -43 12 01.5 & 10.97 & 19.64 & 19.60 & 19.00 & 18.63 & 18.25 & 0.04 & 0.01 & 0.01 & 0.01 & 0.03 & 19.61 & 19.15 & 18.62 & 0.02 & 0.01 & 0.02 & 0.04 & 0.60 & 0.37 & 0.75 & 0.53 & 0.46 & 0.99 & 243@xmath21461 + gc0250 & whh-22 & 13 25 35.31 & -43 05 29.0 & 4.56 & - & - & - & - & - & - & - & - & - & - & 19.66 & 18.80 & 18.03 & 0.04 & 0.04 & 0.04 & - & - & - & - & 0.77 & 0.86 & 1.63 & 492@xmath214126 + gc0251 & magjm-08 & 13 25 35.50 & -42 59 35.2 & 2.12 & 20.76 & 20.18 & 19.17 & 18.68 & 17.94 & 1.15 & 0.24 & 0.06 & 0.05 & 0.05 & - & - & - & - & - & - & 0.58 & 1.01 & 0.49 & 1.22 & - & - & - & 701@xmath21428 + gc0252 & r215 & 13 25 35.64 & -43 08 36.8 & 7.61 & 22.06 & 21.68 & 20.80 & 20.27 & 19.70 & 0.27 & 0.06 & 0.02 & 0.02 & 0.03 & 21.98 & 21.18 & 20.35 & 0.03 & 0.03 & 0.04 & 0.38 & 0.87 & 0.53 & 1.11 & 0.83 & 0.79 & 1.63 & 543@xmath21457 + gc0253 & r213 & 13 25 35.93 & -43 07 27.9 & 6.50 & 22.70 & 22.03 & 20.92 & 20.35 & 19.62 & 0.56 & 0.10 & 0.03 & 0.02 & 0.03 & 22.34 & 21.26 & 20.34 & 0.03 & 0.01 & 0.02 & 0.67 & 1.11 & 0.58 & 1.30 & 0.92 & 1.08 & 2.00 & 532@xmath21452 + gc0254 & aat117899 & 13 25 36.05 & -42 53 40.3 & 7.63 & 21.70 & 21.17 & 20.17 & 19.62 & 18.97 & 0.23 & 0.05 & 0.02 & 0.01 & 0.03 & 21.51 & 20.51 & 19.70 & 0.03 & 0.03 & 0.01 & 0.53 & 1.01 & 0.54 & 1.20 & 0.81 & 1.01 & 1.81 & 609@xmath214116 + gc0255 & c155 & 13 25 36.47 & -43 08 03.5 & 7.10 & - & - & - & - & - & - & - & - & - & - & 22.98 & 22.21 & 21.36 & 0.06 & 0.05 & 0.05 & - & - & - & - & 0.85 & 0.77 & 1.63 & - + gc0256 & magjm-06 & 13 25 36.69 & -42 59 59.2 & 2.02 & - & - & - & - & - & - & - & - & - & - & - & - & - & - & - & - & - & - & - & - & - & - & - & 443@xmath21415 + gc0257 & aat118198 & 13 25 37.47 & -43 05 44.9 & 4.94 & 21.49 & 20.68 & 19.64 & 18.98 & 18.29 & 0.31 & 0.05 & 0.02 & 0.01 & 0.03 & 21.14 & 19.94 & 19.03 & 0.04 & 0.02 & 0.03 & 0.81 & 1.04 & 0.66 & 1.35 & 0.91 & 1.20 & 2.11 & 575@xmath21436 + gc0258 & aat118314 & 13 25 38.03 & -43 16 59.2 & 15.95 & 20.69 & 20.62 & 19.89 & 19.45 & 18.98 & 0.08 & 0.03 & 0.01 & 0.01 & 0.03 & 20.67 & 20.13 & 19.42 & 0.01 & 0.01 & 0.01 & 0.08 & 0.72 & 0.44 & 0.92 & 0.71 & 0.53 & 1.24 & 257@xmath21459 + gc0259 & r209 & 13 25 38.13 & -43 13 02.2 & 12.04 & 21.92 & 21.51 & 20.62 & 20.05 & 19.43 & 0.20 & 0.05 & 0.02 & 0.01 & 0.03 & 21.91 & 20.95 & 20.13 & 0.04 & 0.04 & 0.04 & 0.41 & 0.90 & 0.57 & 1.19 & 0.83 & 0.96 & 1.79 & 558@xmath21440 + gc0260 & c156 & 13 25 38.43 & -43 05 02.6 & 4.37 & - & - & - & - & - & - & - & - & - & - & 19.76 & 18.55 & 17.65 & 0.02 & 0.02 & 0.03 & - & - & - & - & 0.90 & 1.21 & 2.11 & - + gc0261 & c157 & 13 25 38.45 & -43 03 28.9 & 3.06 & - & - & - & - & - & - & - & - & - & - & - & - & 19.21 & - & - & - & - & - & - & - & - & - & - & - + gc0262 & hghh - g268 & 13 25 38.61 & -42 59 19.6 & 2.71 & 20.23 & 19.83 & 18.93 & 18.36 & 17.79 & 0.34 & 0.08 & 0.03 & 0.02 & 0.04 & 20.05 & 19.15 & 18.44 & 0.02 & 0.02 & - & 0.40 & 0.91 & 0.56 & 1.13 & 0.71 & 0.91 & 1.62 & 436@xmath21443 + gc0263 & c158 & 13 25 38.76 & -43 05 34.5 & 4.88 & - & - & - & - & - & - & - & - & - & - & 21.67 & 20.86 & 20.06 & 0.05 & 0.04 & 0.05 & - & - & - & - & 0.80 & 0.81 & 1.61 & - + gc0264 & c159 & 13 25 39.17 & -43 04 33.8 & 4.02 & - & - & - & - & - & - & - & - & - & - & 21.91 & 20.82 & 19.93 & 0.02 & 0.02 & 0.02 & - & - & - & - & 0.90 & 1.09 & 1.99 & - + gc0265 & magjm-01 & 13 25 39.33 & -43 00 48.8 & 2.17 & 19.90 & 19.52 & 18.61 & 18.06 & 17.50 & 0.32 & 0.08 & 0.02 & 0.02 & 0.03 & - & - & - & - & - & - & 0.37 & 0.91 & 0.55 & 1.11 & - & - & - & 645@xmath21436 + gc0266 & pff_gc-059 & 13 25 39.65 & -43 04 01.4 & 3.62 & 21.53 & 20.97 & 19.96 & 19.35 & 18.73 & 0.55 & 0.11 & 0.03 & 0.02 & 0.04 & 21.42 & 20.33 & 19.48 & 0.05 & 0.04 & 0.04 & 0.56 & 1.01 & 0.61 & 1.23 & 0.86 & 1.09 & 1.94 & 525@xmath21427 + gc0267 & hghh-17/c17 & 13 25 39.73 & -42 55 59.2 & 5.62 & 18.82 & 18.51 & 17.63 & 17.10 & 16.57 & 0.04 & 0.01 & 0.01 & 0.01 & 0.03 & 18.61 & 17.72 & 17.19 & 0.04 & 0.05 & 0.01 & 0.32 & 0.88 & 0.53 & 1.06 & 0.53 & 0.89 & 1.42 & 782@xmath2142 + gc0268 & hhh86 - 18/c18/k-163 & 13 25 39.88 & -43 05 01.9 & 4.49 & 18.79 & 18.42 & 17.53 & 16.97 & 16.43 & 0.04 & 0.01 & 0.01 & 0.01 & 0.03 & 18.49 & 17.51 & 16.89 & 0.05 & 0.04 & 0.05 & 0.38 & 0.89 & 0.56 & 1.10 & 0.62 & 0.99 & 1.60 & 480@xmath2142 + gc0269 & c160 & 13 25 40.09 & -43 03
07.1 & 3.01 & - & - & - & - & - & - & - & - & - & - & - & - & 19.99 & - & - & - & - & - & - & - & - & - & - & - + gc0270 & c101 & 13 25 40.47 & -42 56 02.7 & 5.65 & - & - & 20.34 & - & - & - & - & - & - & - & 21.84 & 20.90 & 20.08 & 0.03 & 0.04 & 0.01 & - & - & - & - & 0.82 & 0.94 & 1.76 & - + gc0271 & r204/c161 & 13 25 40.52 & -43 07 17.9 & 6.59 & 21.58 & 20.84 & 19.80 & 19.16 & 18.41 & 0.21 & 0.04 & 0.01 & 0.01 & 0.03 & 21.21 & 20.15 & 19.26 & 0.04 & 0.03 & 0.03 & 0.74 & 1.05 & 0.63 & 1.39 & 0.89 & 1.06 & 1.95 & 425@xmath21432 + gc0272 & hghh-34/c34 & 13 25 40.61 & -43 21 13.6 & 20.22 & 19.37 & 19.01 & 18.12 & 17.59 & 17.03 & 0.04 & 0.01 & 0.01 & 0.01 & 0.03 & 19.64 & 18.54 & 17.76 & 0.04 & 0.03 & 0.02 & 0.35 & 0.89 & 0.53 & 1.10 & 0.77 & 1.10 & 1.87 & 669@xmath21418 + gc0273 & c162 & 13 25 40.87 & -43 05 00.4 & 4.56 & - & - & - & - & - & - & - & - & - & - & 21.95 & 21.41 & 20.91 & 0.06 & 0.05 & 0.01 & - & - & - & - & 0.51 & 0.54 & 1.05 & - + gc0274 & r203 & 13 25 40.90 & -43 08 16.0 & 7.52 & 20.66 & 20.56 & 19.84 & 19.36 & 18.88 & 0.09 & 0.03 & 0.01 & 0.01 & 0.03 & 20.62 & 20.08 & 19.38 & 0.04 & 0.03 & 0.03 & 0.09 & 0.73 & 0.48 & 0.96 & 0.70 & 0.54 & 1.24 & 455@xmath21435 + gc0275 & aat118874 & 13 25 41.36 & -42
58 08.9 & 3.91 & 21.55 & 21.02 & 19.99 & 19.43 & 18.81 & 0.44 & 0.09 & 0.03 & 0.02 & 0.04 & 21.49 & 20.30 & 19.57 & 0.05 & 0.02 & - & 0.53 & 1.03 & 0.56 & 1.18 & 0.73 & 1.19 & 1.92 & 570@xmath21482 + gc0276 & c163 & 13 25 41.63 & -43 03 45.8 & 3.66 & - & - & - & - & - & - & - & - & - & - & 21.98 & 21.25 & 20.43 & 0.05 & 0.04 & 0.04 & - & - & - & - & 0.81 & 0.73 & 1.55 & - + gc0277 & r202 & 13 25 42.00 & -43 10 42.2 & 9.91 & 20.34 & 20.08 & 19.26 & 18.72 & 18.22 & 0.06 & 0.02 & 0.01 & 0.01 & 0.03 & 20.35 & 19.74 & 18.94 & 0.07 & 0.03 & 0.06 & 0.27 & 0.81 & 0.54 & 1.05 & 0.80 & 0.61 & 1.41 & 286@xmath21445 + gc0278 & c164 & 13 25 42.09 & -43 03 19.5 & 3.43 & - & - & - & - & - & - & - & - & - & - & - & - & 19.60 & - & - & - & - & - & - & - & - & - & - & - + gc0279 & hghh - g370 & 13 25 42.25 & -42 59 17.0 & 3.26 & 19.61 & 19.29 & 18.39 & 17.86 & 17.32 & 0.17 & 0.05 & 0.02 & 0.01 & 0.03 & 19.45 & 18.59 & 17.93 & 0.01 & 0.03 & 0.01 & 0.32 & 0.90 & 0.54 & 1.08 & 0.66 & 0.86 & 1.52 & 507@xmath21434 + gc0280 & pff_gc-060 & 13 25 42.43 & -42 59 02.6 & 3.43 & 20.11 & 19.74 & 18.81 & 18.24 & 17.66 & 0.20 & 0.05 & 0.02 & 0.01 & 0.03 & 19.96 & 19.02 & 18.33 & 0.02 & 0.03 & 0.01 & 0.37 & 0.93 & 0.57 & 1.15 & 0.69 & 0.95 & 1.63 & 890@xmath21419 + gc0281 & aat119058 & 13 25 42.53 & -43 03 41.5 & 3.73 & 21.85 & 21.42 & 20.39 & 19.77 & 19.18 & 0.60 & 0.13 & 0.04 & 0.03 & 0.04 & 21.69 & 20.74 & 19.86 & 0.04 & 0.03 & 0.04 & 0.43 & 1.03 & 0.62 & 1.21 & 0.88 & 0.95 & 1.83 & 385@xmath21487 + gc0282 & pff_gc-061 & 13 25 42.62 & -42 45 10.9 & 16.20 & 21.59 & 21.15 & 20.25 & 19.69 & 19.12 & 0.14 & 0.03 & 0.01 & 0.01 & 0.03 & 21.42 & 20.55 & 19.75 & 0.01 & 0.01 & 0.01 & 0.44 & 0.90 & 0.55 & 1.13 & 0.80 & 0.87 & 1.67 & 500@xmath21448 + gc0283 & pff_gc-062 & 13 25 43.23 & -42 58 37.4 & 3.81 & 21.13 & 20.46 & 19.42 & 18.83 & 18.18 & 0.34 & 0.06 & 0.02 & 0.02 & 0.03 & 20.79 & 19.71 & 18.92 & 0.01 & 0.01 & 0.01 & 0.67 & 1.04 & 0.59 & 1.24 & 0.80 & 1.08 & 1.88 & 697@xmath21437 + gc0284 & hghh-19/c19 & 13 25 43.40 & -43 07 22.8 & 6.87 & 19.37 & 19.01 & 18.12 & 17.59 & 17.03 & 0.04 & 0.01 & 0.01 & 0.01 & 0.03 & 18.96 & 17.83 & 18.64 & 0.05 & 0.04 & 0.01 & 0.35 & 0.89 & 0.53 & 1.10&-0.80 & 1.13 & 0.32 & 632@xmath21410 + gc0285 & c165 & 13 25 43.43 & -43 04 56.5 & 4.77 & - & - & - & - & - & - & - & - & - & - & 20.10 & 18.98 & 18.17 & 0.05 & 0.04 & 0.05 & - & - & - & - & 0.81 & 1.12 & 1.93 & - + gc0286 & pff_gc-063 & 13 25 43.80 & -43 07 54.9 & 7.39 & 20.66 & 20.55 & 19.83 & 19.37 & 18.89 & 0.08 & 0.03 & 0.01 & 0.01 & 0.03 & 20.63 & 20.13 & 19.45 & 0.04 & 0.03 & 0.03 & 0.11 & 0.72 & 0.46 & 0.94 & 0.67 & 0.50 & 1.18 & 554@xmath21475 + gc0287 & pff_gc-064 & 13 25 43.90 & -42 50 42.7 & 10.85 & 21.16 & 20.67 & 19.66 & 19.07 & 18.41 & 0.11 & 0.03 & 0.01 & 0.01 & 0.03 & 20.96 & 19.97 & 19.20 & 0.02 & 0.03 & 0.01 & 0.50 & 1.01 & 0.59 & 1.25 & 0.76 & 0.99 & 1.76 & 560@xmath21424 + gc0288 & hghh-35/c35 & 13 25 44.21 & -42 58 59.4 & 3.72 & 19.92 & 19.51 & 18.58 & 18.01 & 17.43 & 0.13 & 0.03 & 0.01 & 0.01 & 0.03 & 19.74 & 18.79 & 18.11 & 0.02 & 0.02 & 0.01 & 0.41 & 0.93 & 0.57 & 1.14 & 0.68 & 0.94 & 1.63 & 544@xmath21413 + gc0289 & c166 & 13 25 44.90 & -43 04 21.1 & 4.50 & - & - & - & - & - & - & - & - & - & - & - & - & 20.72 & - & - & - & - & - & - & - & - & - & - & - + gc0290 & aat208065 & 13 25 45.77 & -42 34 18.0 & 27.05 & 21.79 & 21.45 & 20.09 & 19.40 & 18.71 & 0.49 & 0.05 & 0.01 & 0.01 & 0.03 & 22.06 & 20.70 & 19.66 & 0.04 & 0.03 & 0.02 & 0.33 & 1.36 & 0.69 & 1.39 & 1.04 & 1.36 & 2.40 & 836@xmath21441 + gc0291 & whh-23 & 13 25 45.90 & -42 57 20.2 & 5.07 & 19.26 & 19.12 & 18.37 & 17.89 & 17.41 & 0.05 & 0.02 & 0.01 & 0.01 & 0.03 & 19.13 & 18.50 & 18.00 & 0.03 & 0.05 & 0.01 & 0.14 & 0.75 & 0.48 & 0.96 & 0.50 & 0.64 & 1.14 & 286@xmath21463 + gc0292 & c167 & 13 25 45.97 & -43 06 45.4 & 6.54 & - & - & - & - & - & - & - & - & - & - & - & - & 20.41 & - & - & - & - & - & - & - & - & - & - & - + gc0293 & whh-24 & 13 25 46.00 & -42 56 53.0 & 5.43 & 20.64 & 20.43 & 19.63 & 19.12 & 18.62 & 0.12 & 0.04 & 0.01 & 0.01 & 0.03 & 20.54 & 19.91 & 19.25 & 0.02 & 0.03 & 0.01 & 0.21 & 0.80 & 0.51 & 1.01 & 0.66 & 0.63 & 1.29 & 566@xmath21448 + gc0294 & aat119596 & 13 25 46.06 & -43 08 24.5 & 8.01 & 21.19 & 20.69 & 19.70 & 19.10 & 18.51 & 0.12 & 0.03 & 0.01 & 0.01 & 0.03 & - & - & - & - & - & - & 0.50 & 1.00 & 0.60 & 1.19 & - & - & - & 573@xmath21453 + gc0295 & hghh - g378 & 13 25 46.13 & -43 01 22.9 & 3.39 & 21.30 & 20.59 & 19.53 & 18.86 & 18.22 & 0.55 & 0.10 & 0.03 & 0.02 & 0.03 & - & - & - & - & - & - & 0.71 & 1.06 & 0.67 & 1.31 & - & - & - & 487@xmath21431 + gc0296 & aat208206 & 13 25 46.49 & -42 34 54.1 & 26.47 & 20.08 & 20.10 & 19.47 & 19.07 & 18.69 & 0.12 & 0.02 & 0.01 & 0.01 & 0.03 & 20.13 & 19.73 & 19.19 & 0.01 & 0.01 & 0.01&-0.02 & 0.63 & 0.40 & 0.78 & 0.55 & 0.40 & 0.94 & 293@xmath214115 + gc0297 & hghh - g284 & 13 25 46.59 & -42 57 03.0 & 5.37 & 21.47 & 20.89 & 19.87 & 19.28 & 18.63 & 0.26 & 0.05 & 0.02 & 0.01 & 0.03 & 21.26 & 20.20 & 19.41 & 0.01 & 0.01 & 0.01 & 0.57 & 1.03 & 0.58 & 1.24 & 0.79 & 1.06 & 1.85 & 479@xmath21412 + gc0298 & aat119697 & 13 25 46.68 & -42 53 48.6 & 8.12 & 21.61 & 21.06 & 20.05 & 19.49 & 18.87 & 0.23 & 0.05 & 0.02 & 0.01 & 0.03 & 21.42 & 20.42 & 19.65 & 0.02 & 0.02 & 0.01 & 0.54 & 1.01 & 0.57 & 1.18 & 0.77 & 1.01 & 1.78 & 633@xmath214108 + gc0299 & pff_gc-065 & 13 25 46.92 & -43 08 06.6 & 7.81 & 20.60 & 20.17 & 19.23 & 18.66 & 18.08 & 0.08 & 0.02 & 0.01 & 0.01 & 0.03 & 20.43 & 19.55 & 18.71 & 0.01 & 0.01 & 0.01 & 0.42 & 0.94 & 0.57 & 1.15 & 0.83 & 0.88 & 1.72 & 450@xmath21444 + gc0300 & hghh - g204 & 13 25 46.99 & -43 02 05.4 & 3.66 & 19.35 & 19.12 & 18.33 & 17.83 & 17.31 & 0.12 & 0.04 & 0.01 & 0.01 & 0.03 & - & - & - & - & - & - & 0.23 & 0.79 & 0.51 & 1.02 & - & - & - & 690@xmath21418 + gc0301 & pff_gc-066 & 13 25 47.14 & -43 06 08.8 & 6.14 & 21.21 & 20.82 & 19.92 & 19.37 & 18.80 & 0.15 & 0.04 & 0.01 & 0.01 & 0.03 & 21.07 & 20.21 & 19.44 & 0.01 & 0.01 & 0.01 & 0.39 & 0.91 & 0.55 & 1.11 & 0.77 & 0.86 & 1.63 & 530@xmath21444 + gc0302 & hghh - g357 & 13 25 47.78 & -43 00 43.4 & 3.71 & 20.87 & 20.55 & 19.71 & 19.24 & 18.71 & 0.26 & 0.07 & 0.02 & 0.02 & 0.04 & 20.71 & 19.96 & 19.31 & 0.01 & 0.02 & 0.01 & 0.33 & 0.83 & 0.47 & 1.00 & 0.66 & 0.75 & 1.41 & 664@xmath21431 + gc0303 & aat119894 & 13 25 47.92 & -42 55 52.4 & 6.45 & 21.45 & 21.30 & 20.25 & 19.59 & 18.92 & 0.23 & 0.07 & 0.02 & 0.02 & 0.03 & 21.68 & 20.75 & 19.85 & 0.05 & 0.04 & 0.03 & 0.15 & 1.05 & 0.66 & 1.33 & 0.90 & 0.93 & 1.84 & 448@xmath21490 + gc0304 & c168 & 13 25 48.46 & -43 07 12.5 & 7.16 & - & - & - & - & - & - & - & - & - & - & - & - & 19.71 & - & - & - & - & - & - & - & - & - & - & - + gc0305 & hghh - g251 & 13 25 48.54 & -42 57 41.2 & 5.16 & 20.50 & 19.93 & 18.93 & 18.33 & 17.69 & 0.13 & 0.03 & 0.01 & 0.01 & 0.03 & 20.23 & 19.22 & 18.46 & 0.02 & 0.01 & 0.01 & 0.57 & 1.00 & 0.60 & 1.24 & 0.76 & 1.01 & 1.77 & 574@xmath21427 + gc0306 & pff_gc-067 & 13 25 48.77 & -43 11 38.7 & 11.19 & 20.51 & 20.35 & 19.60 & 19.08 & 18.60 & 0.07 & 0.02 & 0.01 & 0.01 & 0.03 & 20.47 & 19.87 & 19.14 & 0.01 & 0.01 & 0.01 & 0.16 & 0.75 & 0.52 & 1.00 & 0.74 & 0.60 & 1.33 & 632@xmath21443 + gc0307 & pff_gc-068 & 13 25 49.27 & -43 02 20.4 & 4.13 & 20.73 & 20.41 & 19.55 & 18.99 & 18.48 & 0.20 & 0.05 & 0.02 & 0.02 & 0.03 & 20.65 & 20.17 & 19.13 & 0.03 & 0.04 & 0.01 & 0.32 & 0.85 & 0.56 & 1.08 & 1.04 & 0.48 & 1.52 & 393@xmath21426 + gc0308 & hghh-20/c20 & 13 25 49.69 & -42 54 49.3 & 7.50 & 19.18 & 18.83 & 17.95 & 17.43 & 16.88 & 0.04 & 0.01 & 0.01 & 0.01 & 0.03 & 19.01 & 18.19 & 17.54 & 0.03 & 0.02 & 0.01 & 0.35 & 0.88 & 0.53 & 1.07 & 0.65 & 0.82 & 1.47 & 744@xmath2149 + gc0309 & aat120259 & 13 25 49.73 & -43 05 04.7 & 5.64 & 21.95 & 21.16 & 20.12 & 19.51 & 18.80 & 0.35 & 0.06 & 0.02 & 0.01 & 0.03 & 21.57 & 20.48 & 19.56 & 0.01 & 0.01 & 0.01 & 0.79 & 1.04 & 0.62 & 1.32 & 0.92 & 1.09 & 2.01 & 477@xmath21454 + gc0310 & hghh-48/c48 & 13 25 49.82 & -42 50 15.3 & 11.62 & 19.63 & 19.46 & 18.67 & 18.16 & 17.67 & 0.04 & 0.01 & 0.01 & 0.01 & 0.03 & 19.56 & 18.92 & 18.28 & 0.02 & 0.02 & 0.01 & 0.17 & 0.79 & 0.51 & 1.01 & 0.64 & 0.64 & 1.28 & 547@xmath21419 + gc0311 & pff_gc-069 & 13 25 49.93 & -42 40 08.2 & 21.40 & 21.55 & 20.90 & 19.87 & 19.26 & 18.62 & 0.41 & 0.03 & 0.01 & 0.01 & 0.03 & 21.29 & 20.24 & 19.39 & 0.01 & 0.01 & 0.01 & 0.64 & 1.03 & 0.61 & 1.25 & 0.85 & 1.05 & 1.90 & 518@xmath21431 + gc0312 & hghh-47/c47 & 13 25 49.95 & -42 52 09.4 & 9.87 & 19.80 & 19.55 & 18.69 & 18.18 & 17.65 & 0.05 & 0.02 & 0.01 & 0.01 & 0.03 & 19.70 & 18.97 & 18.30 & 0.03 & 0.02 & 0.01 & 0.26 & 0.86 & 0.51 & 1.04 & 0.67 & 0.73 & 1.40 & 589@xmath21422 + gc0313 & aat120336 & 13 25 50.22 & -43 06 08.5 & 6.48 & 21.66 & 21.14 & 20.17 & 19.57 & 18.96 & 0.23 & 0.05 & 0.02 & 0.01 & 0.03 & 21.48 & 20.50 & 19.67 & 0.02 & 0.01 & 0.01 & 0.52 & 0.97 & 0.60 & 1.21 & 0.83 & 0.97 & 1.81 & 452@xmath21467 + gc0314 & whh-25 & 13 25 50.34 & -43 04 08.2 & 5.12 & 21.33 & 20.93 & 19.97 & 19.42 & 18.83 & 0.22 & 0.05 & 0.02 & 0.01 & 0.03 & 21.19 & 20.27 & 19.57 & 0.02 & 0.02 & 0.01 & 0.40 & 0.95 & 0.55 & 1.14 & 0.71 & 0.92 & 1.62 & 525@xmath21450 + gc0315 & aat120355 & 13 25 50.37 & -43 00 32.6 & 4.20 & 21.26 & 21.14 & 20.42 & 20.02 & 19.58 & 0.28 & 0.09 & 0.03 & 0.03 & 0.04 & 21.32 & 20.75 & 20.17 & 0.03 & 0.03 & 0.01 & 0.12 & 0.72 & 0.39 & 0.84 & 0.59 & 0.56 & 1.15 & 548@xmath21477 + gc0316 & pff_gc-070 & 13 25 50.40 & -42
58 02.3 & 5.20 & 20.39 & 20.23 & 19.42 & 18.94 & 18.45 & 0.12 & 0.04 & 0.01 & 0.01 & 0.03 & 20.35 & 19.67 & 19.06 & 0.01 & 0.01 & 0.01 & 0.16 & 0.81 & 0.48 & 0.97 & 0.61 & 0.68 & 1.29 & 556@xmath21449 + gc0317 & c169 & 13 25 51.01 & -42 55 36.3 & 7.00 & - & - & - & - & - & - & - & - & - & - & 21.60 & 21.00 & 20.39 & 0.03 & 0.03 & 0.02 & - & - & - & - & 0.62 & 0.60 & 1.21 & - + gc0318 & aat120515 & 13 25 51.31 & -42 59 29.4 & 4.64 & 21.11 & 20.81 & 19.92 & 19.38 & 18.81 & 0.21 & 0.06 & 0.02 & 0.02 & 0.03 & 21.04 & 20.24 & 19.46 & 0.04 & 0.04 & 0.01 & 0.30 & 0.90 & 0.53 & 1.11 & 0.77 & 0.81 & 1.58 & 461@xmath214141 + gc0319 & pff_gc-071 & 13 25 51.54 & -42 59 46.8 & 4.58 & 20.79 & 20.59 & 19.82 & 19.34 & 18.83 & 0.16 & 0.05 & 0.02 & 0.02 & 0.03 & 20.71 & 20.07 & 19.45 & 0.01 & 0.01 & 0.01 & 0.20 & 0.77 & 0.48 & 0.99 & 0.62 & 0.64 & 1.27 & 475@xmath21441 + gc0320 & c102 & 13 25 52.07 & -42 59 14.4 & 4.87 & - & - & 21.43 & - & - & - & - & - & - & - & 22.71 & 21.87 & 21.04 & 0.02 & 0.02 & 0.02 & - & - & - & - & 0.83 & 0.84 & 1.67 & - + gc0321 & pff_gc-072 & 13 25 52.14 & -42 58 30.2 & 5.20 & 20.68 & 20.45 & 19.62 & 19.12 & 18.54 & 0.15 & 0.04 & 0.02 & 0.01 & 0.03 & 20.64 & 19.92 & 19.25 & 0.02 & 0.02 & 0.01 & 0.23 & 0.83 & 0.51 & 1.08 & 0.67 & 0.72 & 1.39 & 504@xmath21422 + gc0322 & hghh-21/c21 & 13 25 52.74 & -43 05 46.4 & 6.52 & 19.16 & 18.76 & 17.87 & 17.32 & 16.77 & 0.04 & 0.01 & 0.01 & 0.01 & 0.03 & 18.97 & 18.17 & 17.40 & 0.03 & 0.02 & 0.01 & 0.40 & 0.89 & 0.55 & 1.11 & 0.77 & 0.80 & 1.58 & 462@xmath2142 + gc0323 & pff_gc-073 & 13 25 52.78 & -42 58 41.7 & 5.21 & 21.36 & 20.94 & 19.97 & 19.38 & 18.74 & 0.25 & 0.06 & 0.02 & 0.02 & 0.03 & 21.25 & 20.29 & 19.50 & 0.02 & 0.01 & 0.01 & 0.42 & 0.98 & 0.58 & 1.22 & 0.79 & 0.97 & 1.76 & 401@xmath21429 + gc0324 & hghh - g256 & 13 25 52.88 & -43 02 00.0 & 4.70 & 20.69 & 20.18 & 19.00 & 18.35 & 17.65 & 0.18 & 0.04 & 0.01 & 0.01 & 0.03 & 20.50 & 19.31 & 18.44 & 0.03 & 0.04 & 0.01 & 0.51 & 1.17 & 0.65 & 1.36 & 0.87 & 1.20 & 2.06 & 495@xmath21418 + gc0325 & aat209412 & 13 25 53.30 & -42 30 52.7 & 30.63 & 21.94 & 21.23 & 20.22 & 19.60 & 18.97 & 0.52 & 0.04 & 0.01 & 0.01 & 0.03 & 21.69 & 20.64 & 19.76 & 0.01 & 0.01 & 0.01 & 0.71 & 1.01 & 0.62 & 1.25 & 0.88 & 1.05 & 1.92 & 998@xmath214135 + gc0326 & pff_gc-074 & 13 25 53.37 & -42 51 12.4 & 11.00 & 21.16 & 20.83 & 20.00 & 19.48 & 18.90 & 0.11 & 0.03 & 0.01 & 0.01 & 0.03 & 21.04 & 20.27 & 19.56 & 0.01 & 0.01 & 0.01 & 0.33 & 0.83 & 0.52 & 1.10 & 0.71 & 0.77 & 1.48 & 471@xmath21424 + gc0327 & pff_gc-075 & 13 25 53.50 & -43 03 56.6 & 5.50 & 20.96 & 20.45 & 19.48 & 18.90 & 18.28 & 0.14 & 0.03 & 0.01 & 0.01 & 0.03 & 20.78 & 19.76 & 19.01 & 0.02 & 0.03 & 0.01 & 0.51 & 0.97 & 0.58 & 1.20 & 0.75 & 1.02 & 1.77 & 748@xmath21418 + gc0328 & hghh-22/c22 & 13 25 53.57 & -42 59 07.6 & 5.16 & 19.39 & 19.06 & 18.15 & 17.62 & 17.07 & 0.05 & 0.02 & 0.01 & 0.01 & 0.03 & 19.21 & 18.34 & 17.70 & 0.02 & 0.03 & - & 0.34 & 0.90 & 0.54 & 1.09 & 0.64 & 0.87 & 1.52 & 578@xmath2141 + gc0329 & pff_gc-076 & 13 25 53.75 & -43 19 48.6 & 19.26 & 20.67 & 20.06 & 19.07 & 18.41 & 17.78 & 0.08 & 0.02 & 0.01 & 0.01 & 0.03 & 20.44 & 19.44 & 18.52 & 0.02 & 0.01 & 0.01 & 0.62 & 0.99 & 0.66 & 1.29 & 0.92 & 1.00 & 1.92 & 368@xmath21415 + gc0330 & aat120976 & 13 25 54.28 & -42 56 20.6 & 6.84 & 21.51 & 21.23 & 20.38 & 19.84 & 19.27 & 0.23 & 0.06 & 0.02 & 0.02 & 0.03 & 21.50 & 20.73 & 19.99 & 0.03 & 0.02 & 0.02 & 0.28 & 0.85 & 0.54 & 1.11 & 0.74 & 0.77 & 1.51 & 595@xmath21469 + gc0331 & aat328533 & 13 25 54.39 & -43 18 40.1 & 18.19 & 21.79 & 21.26 & 20.31 & 19.74 & 19.09 & 0.16 & 0.04 & 0.01 & 0.01 & 0.03 & 21.62 & 20.64 & 19.77 & 0.01 & 0.01 & 0.01 & 0.53 & 0.95 & 0.58 & 1.23 & 0.87 & 0.98 & 1.85 & 577@xmath21489 + gc0332 & hghh-23/c23 & 13 25 54.58 & -42 59 25.4 & 5.22 & 18.92 & 18.29 & 17.22 & 16.62 & 15.95 & 0.04 & 0.01 & 0.01 & 0.01 & 0.03 & 18.59 & 17.44 & 16.69 & 0.01 & 0.03 & - & 0.63 & 1.07 & 0.60 & 1.28 & 0.75 & 1.15 & 1.90 & 674@xmath2141 + gc0333 & c103 & 13 25 54.98 & -42 59 15.4 & 5.36 & - & - & 18.88 & - & - & - & - & - & - & - & 20.43 & 20.44 & 18.44 & 0.01 & 0.02 & 0.02 & - & - & - & - & 2.00&-0.01 & 1.99 & - + gc0334 & c170 & 13 25 56.11 & -42 56 12.9 & 7.17 & - & - & - & - & - & - & - & - & - & - & 23.31 & 22.83 & 22.16 & 0.05 & 0.04 & 0.06 & - & - & - & - & 0.67 & 0.48 & 1.15 & - + gc0335 & aat121367 & 13 25 56.26 & -43 01 32.9 & 5.25 & 21.06 & 20.92 & 20.24 & 19.82 & 19.39 & 0.25 & 0.08 & 0.03 & 0.03 & 0.04 & 21.01 & 20.45 & 19.85 & 0.01 & 0.01 & 0.01 & 0.13 & 0.68 & 0.42 & 0.85 & 0.60 & 0.56 & 1.15 & 438@xmath21480 + gc0336 & whh-26 & 13 25 56.59 & -42 51 46.6 & 10.76 & 20.50 & 20.02 & 19.05 & 18.43 & 17.81 & 0.07 & 0.02 & 0.01 & 0.01 & 0.03 & 20.33 & 19.36 & 18.56 & 0.02 & 0.01 & 0.01 & 0.48 & 0.97 & 0.62 & 1.24 & 0.79 & 0.97 & 1.76 & 412@xmath21436 + gc0337 & aat329209 & 13 25 57.28 & -43 41 09.0 & 40.37 & 20.87 & 20.70 & 19.87 & 19.40 & 18.90 & 0.21 & 0.02 & 0.01 & 0.01 & 0.03 & 20.78 & 20.12 & 19.35 & 0.01 & - & 0.01 & 0.17 & 0.82 & 0.47 & 0.97 & 0.77 & 0.66 & 1.43 & 601@xmath21465 + gc0338 & c171 & 13 25 57.78 & -42 55 36.1 & 7.82 & - & - & - & - & - & - & - & - & - & - & 22.25 & 21.45 & 20.67 & 0.04 & 0.03 & 0.02 & - & - & - & - & 0.78 & 0.79 & 1.58 & - + gc0339 & c172 & 13 25 57.95 & -42 53 04.3 & 9.79 & - & - & - & - & - & - & - & - & - & - & 22.15 & 21.45 & 20.91 & 0.02 & 0.01 & 0.01 & - & - & - & - & 0.55 & 0.70 & 1.24 & - + gc0340 & pff_gc-077 & 13 25 58.15 & -42 31 38.2 & 30.03 & 20.57 & 20.32 & 19.50 & 18.96 & 18.47 & 0.16 & 0.02 & 0.01 & 0.01 & 0.03 & 20.55 & 19.83 & 19.09 & 0.02 & 0.01 & 0.01 & 0.24 & 0.83 & 0.53 & 1.03 & 0.74 & 0.71 & 1.45 & 675@xmath21441 + gc0341 & pff_gc-078 & 13 25 58.47 & -43 08 06.3 & 8.96 & 20.93 & 20.34 & 19.30 & 18.69 & 18.02 & 0.10 & 0.02 & 0.01 & 0.01 & 0.03 & 20.70 & 19.68 & 18.75 & 0.02 & 0.01 & 0.01 & 0.59 & 1.03 & 0.62 & 1.28 & 0.93 & 1.02 & 1.95 & 545@xmath21418 + gc0342 & hghh - g143 & 13 25 58.69 & -43 07 11.0 & 8.29 & 20.31 & 20.02 & 19.21 & 18.68 & 18.18 & 0.06 & 0.02 & 0.01 & 0.01 & 0.03 & 20.22 & 19.52 & 18.77 & 0.02 & 0.01 & 0.01 & 0.29 & 0.81 & 0.53 & 1.03 & 0.75 & 0.70 & 1.45 & 503@xmath21438 + gc0343 & pff_gc-079 & 13 25 58.91 & -42 53 18.9 & 9.70 & 20.74 & 20.43 & 19.58 & 19.06 & 18.52 & 0.10 & 0.03 & 0.01 & 0.01 & 0.03 & 20.62 & 19.86 & 19.14 & 0.02 & 0.01 & 0.01 & 0.30 & 0.86 & 0.51 & 1.06 & 0.72 & 0.77 & 1.48 & 410@xmath21421 + gc0344 & aat121826/c104 & 13 25 59.49 & -42 55 30.8 & 8.10 & 20.91 & 20.65 & 19.81 & 19.28 & 18.73 & 0.13 & 0.03 & 0.01 & 0.01 & 0.03 & 20.81 & 20.15 & 19.40 & 0.02 & 0.02 & 0.01 & 0.26 & 0.83 & 0.53 & 1.08 & 0.74 & 0.66 & 1.40 & 448@xmath21498 + gc0345 & pff_gc-080 & 13 25 59.55 & -42 32 39.4 & 29.08 & 21.10 & 20.80 & 19.96 & 19.40 & 18.84 & 0.25 & 0.03 & 0.01 & 0.01 & 0.03 & 21.07 & 20.32 & 19.56 & 0.02 & 0.01 & 0.01 & 0.29 & 0.85 & 0.56 & 1.11 & 0.75 & 0.75 & 1.51 & 598@xmath21456 + gc0346 & c173 & 13 25 59.57 & -42 55 01.5 & 8.46 & - & - & - & - & - & - & - & - & - & - & 23.15 & 22.00 & 21.06 & 0.04 & 0.03 & 0.02 & - & - & - & - & 0.94 & 1.15 & 2.09 & - + gc0347 & c174 & 13 25 59.63 & -42 55 15.7 & 8.30 & - & - & - & - & - & - & - & - & - & - & 23.24 & 22.42 & 21.42 & 0.04 & 0.04 & 0.03 & - & - & - & - & 0.99 & 0.82 & 1.81 & - + gc0348 & pff_gc-081 & 13 26 00.15 & -42 49 00.7 & 13.51 & 20.43 & 20.17 & 19.34 & 18.80 & 18.28 & 0.07 & 0.02 & 0.01 & 0.01 & 0.03 & 20.36 & 19.61 & 18.93 & 0.02 & 0.01 & 0.01 & 0.26 & 0.83 & 0.53 & 1.05 & 0.69 & 0.75 & 1.43 & 304@xmath21431 + gc0349 & k-217 & 13 26 00.81 & -43 09 40.1 & 10.46 & 21.63 & 21.09 & 20.09 & 19.54 & 18.91 & 0.16 & 0.03 & 0.01 & 0.01 & 0.03 & 21.42 & 20.48 & 19.57 & 0.01 & 0.01 & 0.01 & 0.54 & 0.99 & 0.56 & 1.19 & 0.91 & 0.94 & 1.85 & 315@xmath214157 + gc0350 & c175 & 13 26 00.93 & -42 58 28.9 & 6.65 & - & - & - & - & - & - & - & - & - & - & 23.51 & 22.53 & 21.84 & 0.06 & 0.04 & 0.03 & - & - & - & - & 0.69 & 0.97 & 1.67 & - + gc0351 & pff_gc-082 & 13 26 00.98 & -42 22 03.4 & 39.56 & 20.83 & 20.17 & 19.16 & 18.58 & 17.89 & 0.21 & 0.02 & 0.01 & 0.01 & 0.03 & 20.51 & 19.53 & 18.69 & 0.02 & 0.01 & 0.01 & 0.66 & 1.01 & 0.58 & 1.27 & 0.84 & 0.98 & 1.82 & 573@xmath21419 + gc0352 & aat122146 & 13 26 01.00 & -43 06 55.3 & 8.40 & 21.14 & 20.90 & 20.05 & 19.53 & 19.04 & 0.11 & 0.03 & 0.01 & 0.01 & 0.03 & 21.09 & 20.38 & 19.58 & 0.01 & 0.01 & 0.01 & 0.24 & 0.85 & 0.52 & 1.01 & 0.80 & 0.71 & 1.51 & 517@xmath21499 + gc0353 & hghh - g221 & 13 26 01.11 & -42 55 13.5 & 8.52 & 20.87 & 20.24 & 19.30 & 18.75 & 18.13 & 0.10 & 0.02 & 0.01 & 0.01 & 0.03 & 20.54 & 19.60 & 18.83 & 0.01 & 0.01 & 0.01 & 0.62 & 0.94 & 0.55 & 1.16 & 0.77 & 0.94 & 1.72 & 390@xmath21415 + gc0354 & aat329848 & 13 26 01.29 & -43 34 15.5 & 33.68 & 19.44 & 19.19 & 18.39 & 17.89 & 17.37 & 0.08 & 0.01 & 0.01 & 0.01 & 0.03 & - & - & - & - & - & - & 0.25 & 0.79 & 0.50 & 1.03 & - & - & - & 558@xmath21446 + gc0355 & pff_gc-083 & 13 26 01.83 & -42 58 15.0 & 6.89 & 21.23 & 20.80 & 19.86 & 19.30 & 18.72 & 0.16 & 0.04 & 0.01 & 0.01 & 0.03 & 21.12 & 20.21 & 19.44 & 0.02 & 0.02 & 0.01 & 0.42 & 0.95 & 0.55 & 1.13 & 0.77 & 0.91 & 1.68 & 458@xmath21431 + gc0356 & pff_gc-084 & 13 26 02.25 & -43 08 55.6 & 10.03 & 21.31 & 20.80 & 19.84 & 19.24 & 18.63 & 0.12 & 0.03 & 0.01 & 0.01 & 0.03 & 21.13 & 20.19 & 19.32 & 0.02 & 0.01 & 0.01 & 0.51 & 0.97 & 0.59 & 1.21 & 0.87 & 0.93 & 1.80 & 458@xmath21438 + gc0357 & aat122445 & 13 26 02.43 & -43 00 11.7 & 6.43 & 21.13 & 20.99 & 20.28 & 19.81 & 19.42 & 0.15 & 0.05 & 0.02 & 0.02 & 0.03 & 21.08 & 20.50 & 19.92 & 0.01 & 0.02 & 0.01 & 0.14 & 0.71 & 0.47 & 0.86 & 0.58 & 0.58 & 1.16 & 342@xmath21498 + gc0358 & c176 & 13 26 02.79 & -42 57 05.0 & 7.61 & - & - & - & - & - & - & - & - & - & - & 22.35 & 21.92 & 21.30 & 0.02 & 0.02 & 0.03 & - & - & - & - & 0.62 & 0.43 & 1.05 & - + gc0359 & hghh-25/c25 & 13 26 02.85 & -42 56 57.0 & 7.69 & 20.17 & 19.56 & 18.49 & 17.85 & 17.17 & 0.07 & 0.02 & 0.01 & 0.01 & 0.03 & 19.92 & 18.83 & 17.97 & 0.03 & 0.02 & 0.01 & 0.61 & 1.07 & 0.64 & 1.32 & 0.86 & 1.09 & 1.95 & 703@xmath2149 + gc0360 & aat122526 & 13 26 02.90 & -43 05 43.0 & 7.90 & 21.89 & 21.35 & 20.36 & 19.73 & 19.05 & 0.24 & 0.05 & 0.02 & 0.01 & 0.03 & 21.72 & 20.72 & 19.81 & 0.02 & 0.01 & 0.01 & 0.54 & 1.00 & 0.62 & 1.31 & 0.90 & 1.00 & 1.90 & 506@xmath21450 + gc0361 & c177 & 13 26 03.20 & -42 54 30.1 & 9.30 & - & - & - & - & - & - & - & - & - & - & 22.95 & 22.11 & 21.09 & 0.11 & 0.07 & 0.05 & - & - & - & - & 1.02 & 0.84 & 1.87 & - + gc0362 & c178 & 13 26 03.85 & -42 56 45.3 & 7.95 & - & - & - & - & - & - & - & - & - & - & 23.02 & 22.32 & 21.53 & 0.04 & 0.04 & 0.03 & - & - & - & - & 0.79 & 0.69 & 1.48 & - + gc0363 & hghh - g293/g293 & 13 26 04.20 & -42 55 44.7 & 8.60 & 20.11 & 19.90 & 19.10 & 18.61 & 18.11 & 0.06 & 0.02 & 0.01 & 0.01 & 0.03 & 20.04 & 19.36 & 18.69 & 0.01 & 0.01 & 0.01 & 0.21 & 0.80 & 0.49 & 0.99 & 0.67 & 0.69 & 1.35 & 581@xmath21415 + gc0364 & aat122808 & 13 26 04.61 & -43 09 10.2 & 10.49 & 21.05 & 20.87 & 20.12 & 19.66 & 19.18 & 0.10 & 0.03 & 0.01 & 0.01 & 0.03 & 21.00 & 20.45 & 19.69 & 0.01 & 0.01 & 0.01 & 0.18 & 0.75 & 0.46 & 0.94 & 0.76 & 0.55 & 1.31 & 264@xmath214131 + gc0365 & aat122794 & 13 26 04.69 & -42 47 35.1 & 15.16 & 21.89 & 21.15 & 20.00 & 19.29 & 18.69 & 0.21 & 0.04 & 0.01 & 0.01 & 0.03 & 21.60 & 20.37 & 19.38 & 0.01 & 0.01 & 0.01 & 0.74 & 1.15 & 0.71 & 1.31 & 0.98 & 1.23 & 2.22 & 336@xmath214161 + gc0366 & c105 & 13 26 05.12 & -42 55 37.0 & 8.80 & - & - & 22.01 & - & - & - & - & - & - & - & 23.68 & 22.58 & 21.76 & 0.07 & 0.04 & 0.04 & - & - & - & - & 0.82 & 1.10 & 1.92 & - + gc0367 & hghh-07/c7 & 13 26 05.41 & -42 56 32.4 & 8.30 & 18.38 & 18.03 & 17.17 & 16.65 & 16.08 & 0.03 & 0.01 & 0.01 & 0.01 & 0.03 & 18.18 & 17.33 & 16.64 & 0.03 & 0.03 & 0.01 & 0.35 & 0.86 & 0.53 & 1.09 & 0.68 & 0.85 & 1.53 & 595@xmath2141 + gc0368 & c106 & 13 26 06.15 & -42 56 45.4 & 8.32 & - & - & 21.28 & - & - & - & - & - & - & - & 22.54 & 21.46 & 20.66 & 0.03 & 0.01 & 0.01 & - & - & - & - & 0.80 & 1.08 & 1.88 & - + gc0369 & pff_gc-085 & 13 26 06.42 & -43 00 38.1 & 7.11 & 20.39 & 20.20 & 19.43 & 18.99 & 18.47 & 0.07 & 0.02 & 0.01 & 0.01 & 0.03 & 20.29 & 19.64 & 19.03 & 0.02 & 0.02 & 0.01 & 0.20 & 0.76 & 0.44 & 0.96 & 0.61 & 0.65 & 1.26 & 548@xmath21431 + gc0370 & pff_gc-086 & 13 26 06.55 & -43 06 14.5 & 8.75 & 21.48 & 20.93 & 19.95 & 19.37 & 18.76 & 0.16 & 0.03 & 0.01 & 0.01 & 0.03 & 21.29 & 20.32 & 19.45 & 0.02 & 0.01 & 0.01 & 0.56 & 0.97 & 0.59 & 1.20 & 0.87 & 0.97 & 1.84 & 440@xmath21426 + gc0371 & pff_gc-087 & 13 26 06.87 & -42 33 17.3 & 28.77 & 20.26 & 19.96 & 19.11 & 18.56 & 18.05 & 0.14 & 0.02 & 0.01 & 0.01 & 0.03 & 20.17 & 19.42 & 18.70 & 0.02 & 0.01 & 0.01 & 0.30 & 0.86 & 0.55 & 1.05 & 0.73 & 0.75 & 1.47 & 830@xmath21429 + gc0372 & hghh - g170 & 13 26 06.93 & -42 57 35.1 & 8.02 & 20.78 & 20.21 & 19.22 & 18.65 & 17.97 & 0.11 & 0.02 & 0.01 & 0.01 & 0.03 & 20.57 & 19.56 & 18.73 & 0.02 & 0.02 & 0.01 & 0.57 & 0.99 & 0.57 & 1.25 & 0.83 & 1.01 & 1.84 & 636@xmath21427 + gc0373 & aat123188 & 13 26 06.94 & -43 07 52.6 & 9.85 & 21.70 & 21.30 & 20.40 & 19.85 & 19.27 & 0.17 & 0.04 & 0.01 & 0.01 & 0.03 & 21.57 & 20.71 & 19.94 & 0.02 & 0.01 & 0.01 & 0.40 & 0.90 & 0.56 & 1.13 & 0.77 & 0.86 & 1.63 & 364@xmath21456 + gc0374 & hghh-36/c36/r113 & 13 26 07.73 & -42 52 00.3 & 11.72 & 19.42 & 19.19 & 18.35 & 17.81 & 17.33 & 0.04 & 0.01 & 0.01 & 0.01 & 0.03 & 19.32 & 18.61 & 17.94 & 0.02 & 0.01 & 0.01 & 0.23 & 0.84 & 0.54 & 1.03 & 0.67 & 0.71 & 1.38 & 703@xmath2141 + gc0375 & aat123453 & 13 26 08.38 & -42 59 18.9 & 7.67 & 21.29 & 20.95 & 20.09 & 19.55 & 19.03 & 0.17 & 0.04 & 0.02 & 0.01 & 0.03 & 21.20 & 20.40 & 19.67 & 0.03 & 0.02 & 0.01 & 0.34 & 0.86 & 0.54 & 1.06 & 0.73 & 0.80 & 1.53 & 257@xmath214154 + gc0376 & pff_gc-088 & 13 26 08.86 & -43 01 21.4 & 7.54 & 19.99 & 19.79 & 18.99 & 18.46 & 18.00 & 0.05 & 0.02 & 0.01 & 0.01 & 0.03 & 19.86 & 19.14 & 18.59 & 0.03 & 0.03 & 0.01 & 0.20 & 0.80 & 0.53 & 0.99 & 0.55 & 0.72 & 1.27 & 554@xmath21429 + gc0377 & aat123656 & 13 26 09.61 & -43 07 05.9 & 9.71 & 21.77 & 21.13 & 20.08 & 19.46 & 18.79 & 0.19 & 0.04 & 0.01 & 0.01 & 0.03 & 21.57 & 20.51 & 19.59 & 0.03 & 0.02 & 0.02 & 0.64 & 1.05 & 0.63 & 1.29 & 0.92 & 1.07 & 1.99 & 380@xmath21493 + gc0378 & r111 & 13 26 09.71 & -42 50 29.5 & 13.14 & 22.07 & 21.57 & 20.59 & 19.91 & 19.09 & 0.28 & 0.06 & 0.02 & 0.02 & 0.03 & 21.96 & 21.03 & 20.20 & 0.03 & 0.03 & 0.02 & 0.51 & 0.97 & 0.68 & 1.50 & 0.82 & 0.94 & 1.76 & 717@xmath21448 + gc0379 & c179 & 13 26 09.87 & -42 56 36.0 & 8.96 & - & - & - & - & - & - & - & - & - & - & 22.13 & 21.71 & 21.04 & 0.03 & 0.03 & 0.01 & - & - & - & - & 0.67 & 0.42 & 1.09 & - + gc0380 & hghh-37/c37/r116 & 13 26 10.58 & -42 53 42.7 & 10.81 & 19.86 & 19.38 & 18.43 & 17.87 & 17.26 & 0.04 & 0.01 & 0.01 & 0.01 & 0.03 & 19.65 & 18.72 & 17.96 & 0.01 & 0.01 & 0.01 & 0.48 & 0.95 & 0.56 & 1.17 & 0.76 & 0.93 & 1.69 & 612@xmath2141 + gc0381 & whh-27 & 13 26 12.82 & -43 09 09.2 & 11.51 & 20.27 & 19.68 & 18.66 & 18.06 & 17.42 & 0.05 & 0.01 & 0.01 & 0.01 & 0.03 & 20.02 & 19.03 & 18.13 & 0.02 & 0.01 & 0.01 & 0.59 & 1.02 & 0.60 & 1.24 & 0.90 & 0.99 & 1.89 & 545@xmath21460 + gc0382 & whh-28 & 13 26 14.18 & -43 08 30.4 & 11.25 & 20.21 & 19.98 & 19.17 & 18.65 & 18.14 & 0.05 & 0.02 & 0.01 & 0.01 & 0.03 & 20.13 & 19.48 & 18.71 & 0.02 & 0.01 & 0.01 & 0.23 & 0.81 & 0.52 & 1.03 & 0.77 & 0.65 & 1.41 & 506@xmath214123 + gc0383 & hhh86 - 26/c26 & 13 26 15.27 & -42 48 29.4 & 15.36 & 19.99 & 19.26 & 18.13 & 17.47 & 16.77 & 0.05 & 0.01 & 0.01 & 0.01 & 0.03 & 19.66 & 18.45 & 17.59 & 0.04 & 0.03 & 0.02 & 0.73 & 1.13 & 0.66 & 1.35 & 0.86 & 1.21 & 2.07 & 377@xmath21414 + gc0384 & r122 & 13 26 15.95 & -42 55 00.5 & 10.76 & 19.02 & 18.78 & 18.02 & 17.58 & 17.14 & 0.03 & 0.01 & 0.01 & 0.01 & 0.03 & - & - & - & - & - & - & 0.24 & 0.76 & 0.44 & 0.88 & - & - & - & 588@xmath2142 + gc0385 & aat125079 & 13 26 17.29 & -43 06 39.3 & 10.62 & 21.13 & 20.90 & 20.09 & 19.59 & 19.06 & 0.10 & 0.03 & 0.01 & 0.01 & 0.03 & 21.07 & 20.41 & 19.64 & 0.02 & 0.01 & 0.01 & 0.23 & 0.81 & 0.51 & 1.03 & 0.77 & 0.66 & 1.43 & 513@xmath214183 + gc0386 & c50/k-233 & 13 26 19.66 & -43 03 18.6 & 9.76 & 20.09 & 19.68 & 18.74 & 18.17 & 17.57 & 0.05 & 0.02 & 0.01 & 0.01 & 0.03 & 19.91 & 18.92 & 18.28 & 0.04 & 0.04 & 0.01 & 0.41 & 0.93 & 0.58 & 1.17 & 0.64 & 1.00 & 1.63 & 615@xmath21458 + gc0387 & c49/pff_gc-089 & 13 26 20.20 & -43 10 35.7 & 13.48 & 19.90 & 19.73 & 18.95 & 18.45 & 17.95 & 0.12 & 0.02 & 0.01 & 0.01 & 0.03 & 19.85 & 19.30 & 18.53 & 0.03 & 0.02 & 0.01 & 0.17 & 0.78 & 0.50 & 1.00 & 0.77 & 0.56 & 1.32 & 538@xmath21430 + gc0388 & pff_gc-090 & 13 26 20.53 & -43 03 18.5 & 9.91 & 20.90 & 20.34 & 19.33 & 18.74 & 18.10 & 0.10 & 0.02 & 0.01 & 0.01 & 0.03 & 20.70 & 19.61 & 18.85 & 0.03 & 0.03 & 0.01 & 0.56 & 1.00 & 0.60 & 1.24 & 0.75 & 1.09 & 1.85 & 486@xmath21426 + gc0389 & aat215171 & 13 26 20.66 & -42 38 32.0 & 24.60 & 21.03 & 20.63 & 19.69 & 19.11 & 18.53 & 0.27 & 0.03 & 0.01 & 0.01 & 0.03 & 20.97 & 20.06 & 19.26 & 0.02 & 0.02 & 0.01 & 0.40 & 0.94 & 0.58 & 1.15 & 0.80 & 0.90 & 1.70 & 527@xmath21448 + gc0390 & r107 & 13 26 21.11 & -42 48 41.1 & 15.84 & 21.53 & 20.82 & 19.75 & 19.13 & 18.48 & 0.16 & 0.03 & 0.01 & 0.01 & 0.03 & 21.23 & 20.16 & 19.30 & 0.05 & 0.03 & 0.02 & 0.71 & 1.07 & 0.62 & 1.27 & 0.86 & 1.06 & 1.92 & 405@xmath21428 + gc0391 & pff_gc-091 & 13 26 21.14 & -43 42 24.6 & 42.41 & 20.42 & 20.13 & 19.29 & 18.77 & 18.21 & 0.15 & 0.02 & 0.01 & 0.01 & 0.03 & 20.30 & 19.56 & 18.72 & 0.01 & 0.01 & 0.01 & 0.29 & 0.84 & 0.52 & 1.08 & 0.84 & 0.74 & 1.58 & 623@xmath21437 + gc0392 & pff_gc-092 & 13 26 21.31 & -42 57 19.1 & 10.53 & 21.36 & 20.86 & 19.92 & 19.35 & 18.76 & 0.14 & 0.03 & 0.01 & 0.01 & 0.03 & 21.18 & 20.27 & 19.46 & 0.02 & 0.02 & 0.01 & 0.50 & 0.93 & 0.57 & 1.16 & 0.81 & 0.92 & 1.73 & 462@xmath21427 + gc0393 & r117 & 13 26 21.99 & -42 53 45.5 & 12.38 & 20.52 & 20.41 & 19.68 & 19.23 & 18.75 & 0.07 & 0.02 & 0.01 & 0.01 & 0.03 & 20.50 & 19.98 & 19.32 & 0.02 & 0.02 & 0.01 & 0.10 & 0.73 & 0.45 & 0.93 & 0.66 & 0.53 & 1.18 & 484@xmath21426 + gc0394 & r118 & 13 26 22.01 & -42 54 26.5 & 11.99 & 20.73 & 20.60 & 19.87 & 19.38 & 18.93 & 0.08 & 0.03 & 0.01 & 0.01 & 0.03 & 20.72 & 20.19 & 19.50 & 0.03 & 0.02 & 0.01 & 0.13 & 0.73 & 0.49 & 0.95 & 0.70 & 0.52 & 1.22 & 440@xmath21473 + gc0395 & whh-29 & 13 26 22.08 & -43 09 10.7 & 12.79 & 20.97 & 20.70 & 19.82 & 19.28 & 18.73 & 0.08 & 0.02 & 0.01 & 0.01 & 0.03 & 20.92 & 20.17 & 19.38 & 0.03 & 0.02 & 0.01 & 0.27 & 0.88 & 0.54 & 1.09 & 0.79 & 0.75 & 1.54 & 505@xmath21478 + gc0396 & pff_gc-093 & 13 26 22.65 & -42 46 49.8 & 17.50 & 20.58 & 20.39 & 19.62 & 19.13 & 18.66 & 0.08 & 0.03 & 0.01 & 0.01 & 0.03 & 20.52 & 19.91 & 19.23 & 0.02 & 0.01 & 0.01 & 0.20 & 0.77 & 0.48 & 0.96 & 0.69 & 0.61 & 1.30 & 576@xmath21442 + gc0397 & whh-30 & 13 26 23.60 & -43 03 43.9 & 10.55 & 20.56 & 20.25 & 19.38 & 18.83 & 18.30 & 0.10 & 0.03 & 0.01 & 0.01 & 0.03 & 20.45 & 19.62 & 18.96 & 0.03 & 0.03 & 0.01 & 0.31 & 0.87 & 0.55 & 1.08 & 0.65 & 0.83 & 1.49 & 470@xmath21466 + gc0398 & pff_gc-094 & 13 26 23.66 & -43 00 45.6 & 10.25 & 20.99 & 20.77 & 19.97 & 19.49 & 18.93 & 0.09 & 0.03 & 0.01 & 0.01 & 0.03 & 20.92 & 20.24 & 19.57 & 0.03 & 0.02 & 0.01 & 0.22 & 0.80 & 0.47 & 1.04 & 0.67 & 0.68 & 1.35 & 334@xmath21464 + gc0399 & hhh86 - 38/c38/r123 & 13 26 23.78 & -42 54 01.1 & 12.50 & 19.67 & 19.30 & 18.41 & 17.87 & 17.32 & 0.04 & 0.01 & 0.01 & 0.01 & 0.03 & 19.51 & 18.68 & 17.96 & 0.02 & 0.02 & 0.01 & 0.37 & 0.89 & 0.54 & 1.09 & 0.72 & 0.83 & 1.54 & 405@xmath2141 + gc0400 & hghh-51 & 13 26 23.86 & -42 47 17.1 & 17.26 & 19.63 & 19.34 & 18.47 & 17.93 & 17.38 & 0.05 & 0.02 & 0.01 & 0.01 & 0.03 & - & - & - & - & - & - & 0.29 & 0.87 & 0.54 & 1.09 & - & - & - & 343@xmath21443 + gc0401 & aat335187 & 13 26 23.95 & -43 17 44.4 & 19.53 & 21.19 & 21.11 & 20.46 & 20.04 & 19.60 & 0.10 & 0.03 & 0.01 & 0.01 & 0.03 & 21.20 & 20.72 & 20.08 & 0.01 & 0.01 & 0.01 & 0.08 & 0.65 & 0.42 & 0.86 & 0.64 & 0.49 & 1.12 & 277@xmath214158 + gc0402 & pff_gc-095 & 13 26 25.50 & -42 57 06.2 & 11.33 & 21.13 & 20.60 & 19.60 & 19.02 & 18.40 & 0.11 & 0.02 & 0.01 & 0.01 & 0.03 & 20.97 & 19.97 & 19.17 & 0.04 & 0.03 & 0.02 & 0.53 & 1.00 & 0.58 & 1.20 & 0.80 & 0.99 & 1.79 & 405@xmath21418 + gc0403 & r124 & 13 26 28.87 & -42 52 36.4 & 14.08 & 20.32 & 20.17 & 19.42 & 18.96 & 18.43 & 0.05 & 0.02 & 0.01 & 0.01 & 0.03 & 20.28 & 19.69 & 19.02 & 0.02 & 0.02 & 0.01 & 0.15 & 0.75 & 0.46 & 0.99 & 0.68 & 0.58 & 1.26 & 541@xmath21435 + gc0404 & pff_gc-096 & 13 26 30.29 & -42 34 41.7 & 28.83 & 21.24 & 20.73 & 19.73 & 19.10 & 18.42 & 0.31 & 0.03 & 0.01 & 0.01 & 0.03 & 21.08 & 20.09 & 19.24 & 0.01 & 0.01 & 0.01 & 0.52 & 1.00 & 0.63 & 1.30 & 0.85 & 0.99 & 1.83 & 532@xmath21425 + gc0405 & r105 & 13 26 33.55 & -42 51 00.9 & 15.74 & 21.28 & 20.78 & 19.86 & 19.28 & 18.70 & 0.11 & 0.03 & 0.01 & 0.01 & 0.03 & 21.18 & 20.25 & 19.46 & 0.02 & 0.01 & 0.01 & 0.50 & 0.92 & 0.58 & 1.15 & 0.79 & 0.93 & 1.72 & 534@xmath21434 + gc0406 & hghh-27/c27 & 13 26 37.99 & -42 45 49.9 & 20.00 & 19.60 & 19.39 & 18.60 & 18.11 & 17.62 & 0.03 & 0.01 & 0.01 & 0.01 & 0.03 & - & - & - & - & - & - & 0.21 & 0.78 & 0.49 & 0.98 & - & - & - & 492@xmath21437 + gc0407 & whh-31 & 13 26 41.43 & -43 11 25.0 & 16.96 & 20.76 & 20.44 & 19.50 & 18.94 & 18.36 & 0.07 & 0.02 & 0.01 & 0.01 & 0.03 & 20.67 & 19.89 & 19.05 & 0.03 & 0.02 & 0.02 & 0.32 & 0.94 & 0.56 & 1.14 & 0.85 & 0.77 & 1.62 & 573@xmath21467 + gc0408 & hhh86 - 39/c39 & 13 26 42.03 & -43 07 44.8 & 15.12 & 18.72 & 18.33 & 17.43 & 16.91 & 16.42 & 0.02 & 0.01 & 0.01 & 0.01 & 0.03 & 18.57 & 17.73 & 16.92 & 0.01 & 0.01 & - & 0.39 & 0.89 & 0.53 & 1.01 & 0.81 & 0.83 & 1.65 & 271@xmath21420 + gc0409 & pff_gc-097 & 13 26 45.40 & -43 26 34.1 & 29.13 & 19.75 & 19.51 & 18.72 & 18.23 & 17.76 & 0.10 & 0.01 & 0.01 & 0.01 & 0.03 & 19.66 & 19.04 & 18.28 & 0.02 & 0.01 & 0.01 & 0.24 & 0.79 & 0.49 & 0.96 & 0.76 & 0.62 & 1.38 & 599@xmath21430 + gc0410 & hh-017 & 13 26 49.31 & -43 04 57.8 & 15.41 & 23.21 & 22.54 & 20.80 & 19.87 & 19.06 & 0.67 & 0.11 & 0.02 & 0.01 & 0.03 & 23.32 & 21.63 & 20.13 & 0.05 & 0.03 & 0.02 & 0.67 & 1.75 & 0.93 & 1.73 & 1.50 & 1.68 & 3.19 & 839@xmath21463 + gc0411 & pff_gc-098 & 13 26 53.94 & -43 19 17.7 & 24.05 & 19.54 & 19.17 & 18.28 & 17.73 & 17.16 & 0.09 & 0.01 & 0.01 & 0.01 & 0.03 & - & - & - & - & - & - & 0.37 & 0.89 & 0.55 & 1.12 & - & - & - & 631@xmath21418 + gc0412 & aat222977 & 13 26 58.91 & -42 38 53.4 & 27.82 & 21.18 & 21.02 & 20.34 & 19.84 & 19.38 & 0.28 & 0.03 & 0.01 & 0.01 & 0.03 & 21.16 & 20.64 & 20.02 & 0.02 & 0.01 & 0.01 & 0.17 & 0.68 & 0.50 & 0.96 & 0.62 & 0.52 & 1.14 & 655@xmath214112 + gc0413 & hh-060 & 13 26 59.78 & -42 55 26.5 & 17.79 & - & - & - & - & - & - & - & - & - & - & 23.41 & 21.69 & 20.14 & 0.05 & 0.04 & 0.02 & - & - & - & - & 1.55 & 1.72 & 3.26 & 811@xmath21432 + gc0414 & pff_gc-099 & 13 26 59.82 & -42 32 40.5 & 33.09 & 20.96 & 20.72 & 19.94 & 19.40 & 18.92 & 0.23 & 0.03 & 0.01 & 0.01 & 0.03 & 20.92 & 20.29 & 19.59 & 0.02 & 0.01 & 0.02 & 0.25 & 0.78 & 0.53 & 1.02 & 0.70 & 0.63 & 1.33 & 425@xmath21434 + gc0415 & pff_gc-100 & 13 27 03.41 & -42 27 17.2 & 38.12 & 19.92 & 19.37 & 18.41 & 17.81 & 17.17 & 0.11 & 0.01 & 0.01 & 0.01 & 0.03 & - & - & - & - & - & - & 0.56 & 0.96 & 0.60 & 1.24 & - & - & - & 513@xmath21410 + gc0416 & pff_gc-101 & 13 27 21.56 & -42 38 41.4 & 30.63 & 19.89 & 19.51 & 18.72 & 18.20 & 17.74 & 0.10 & 0.01 & 0.01 & 0.01 & 0.03 & 19.74 & 19.02 & 18.39 & 0.01 & 0.01 & - & 0.38 & 0.79 & 0.51 & 0.97 & 0.63 & 0.72 & 1.35 & 263@xmath21416 + gc0417 & pff_gc-102 & 13 28 18.45 & -42 33 12.5 & 41.90 & 20.38 & 20.11 & 19.29 & 18.76 & 18.24 & 0.14 & 0.02 & 0.01 & 0.01 & 0.03 & 20.28 & 19.59 & 18.94 & 0.04 & 0.04 & 0.01 & 0.27 & 0.83 & 0.52 & 1.04 & 0.65 & 0.70 & 1.34 & 428@xmath21440 + 0 - 50 & 12.9 & 48.8 & 340 & 40@xmath21410 & 189@xmath21412 & 123@xmath2145 & 1.0 & [email protected] & [email protected] & [email protected] + 0 - 5 & 3.64 & 4.96 & 54 & 24@xmath21421 & 334@xmath21459 & 120@xmath21412 & - & - & - & - + 5 - 10 & 7.65 & 9.96 & 124 & 43@xmath21415 & 195@xmath21420 & 112@xmath2148 & 2.3 & [email protected] & [email protected] & [email protected] + 10 - 15 & 12.4 & 14.9 & 68 & 83@xmath21425 & 195@xmath21412 & 105@xmath21410 & 1.6 & [email protected] & [email protected] & [email protected] + 15 - 25 & 19.0 & 24.3 & 56 & 35@xmath21426 & 184@xmath21434 & 147@xmath21416 & 1.3 & [email protected] & [email protected] & [email protected] + 25 - 50 & 34.7 & 48.8 & 39 & 96@xmath21445 & 169@xmath21417 & 148@xmath21421 & 1.0 & [email protected] & [email protected] & [email protected] + 0 - 50 & 13.8 & 48.1 & 178 & 31@xmath21414 & 177@xmath21422 & 117@xmath2147 & 1.0 & [email protected] & [email protected] & 95.6 @xmath21458.6 + 0 - 5 & 3.84 & 4.96 & 22 & 16@xmath21438 & 373@xmath214120 & 99@xmath21419 & - & - & - & - + 5 - 10 & 7.76 & 9.96 & 66 & 30@xmath21421 & 223@xmath21439 & 116@xmath21412 & 2.1 & [email protected] & [email protected] & [email protected] + 10 - 15 & 12.7 & 14.8 & 41 & 97@xmath21438 & 199@xmath21413 & 102@xmath21415 & 1.5 & [email protected] & [email protected] & [email protected] + 15 - 25 & 18.8 & 24.3 & 26 & 45@xmath21431 & 139@xmath21449 & 112@xmath21420 & 1.3 & [email protected] & [email protected] & [email protected] + 25 - 50 & 36.7 & 48.1 & 23 & 61@xmath21443 & 137@xmath21450 & 141@xmath21424 & 1.0 & [email protected] & [email protected] & [email protected] + 0 - 50 & 12.0 & 48.8 & 158 & 47@xmath21415 & 202@xmath21415 & 129@xmath2149 & 1.0 & [email protected] & [email protected] & [email protected] + 0 - 5 & 3.61 & 4.98 & 29 & 44@xmath21427 & 308@xmath21447 & 134@xmath21423 & - & - & - & - + 5 - 10 & 7.52 & 9.93 & 58 & 65@xmath21422 & 180@xmath21419 & 105@xmath21411 & 1.5 & [email protected] & [email protected] & [email protected] + 10 - 15 & 12.0 & 14.9 & 27 & 74@xmath21434 & 186@xmath21422 & 108@xmath21417 & 1.2 & [email protected] & [email protected] & [email protected] + 15 - 25 & 19.3 & 24.3 & 30 & 50@xmath21436 & 223@xmath21432 & 168@xmath21424 & 1.1 & [email protected] & [email protected] & [email protected] + 25 - 50 & 31.7 & 48.8 & 15 & 102@xmath21496 & 191@xmath21421 & 146@xmath21465 & 1.0 & [email protected] & [email protected] & [email protected] + 0 - 90 & 14.1 & 88.2 & 780 & 76@xmath2146 & 170@xmath2145 & 118@xmath21413 & 1.0 & [email protected] & [email protected] & [email protected] + 0 - 5 & 3.3 & 4.99 & 184 & 75@xmath21416 & 179@xmath21411 & 131@xmath2147 & - & - & - & - + 5 - 10 & 7.6 & 9.98 & 211 & 82@xmath21412 & 177@xmath2148 & 120@xmath2146 & 3.0 & [email protected] & [email protected] & [email protected] + 10 - 15 & 11.8 & 14.8 & 138 & 76@xmath21415 & 171@xmath21410 & 118@xmath2147 & 2.0 & [email protected] & [email protected] & [email protected] + 15 - 20 & 17.4 & 20.0 & 71 & 96@xmath21424 & 169@xmath21412 & 116@xmath21410 & 1.7 & [email protected] & [email protected] & [email protected] + 20 - 30 & 24.6 & 30.0 & 87 & 76@xmath21417 & 157@xmath21417 & 108@xmath2148 & 1.5 & [email protected] & [email protected] & [email protected] + 30 - 40 & 34.6 & 39.8 & 50 & 44@xmath21415 & 132@xmath21438 & 87@xmath2149 & 1.3 & [email protected] & [email protected] & [email protected] + 40 - 80 & 48.7 & 71.2 & 36 & 61@xmath21445 & 183@xmath21422 & 85@xmath21411 & 1.0 & [email protected] & [email protected] & [email protected] + ngc 5128 & 31@xmath21414 & 117@xmath2147 & [email protected] & [email protected] & 47@xmath21415 & 129@xmath2149 & [email protected] & [email protected] + m87 & 186@xmath218 & 397@xmath219 & 0.47@xmath220 & 0.14@xmath221 & 155@xmath222 & 365@xmath223 & 0.43@xmath224 & 0.13@xmath225 + m49 & 93@xmath226 & 342@xmath227 & 0.27@xmath228 & 0.08@xmath229 & -26@xmath230 & 265@xmath231 & 0.10@xmath232 & 0.03@xmath233 + ngc 1399 & 15@xmath21426 & 291@xmath21414 & [email protected] & [email protected] & 7@xmath21424 & 255@xmath21413 & [email protected] & [email protected] + km s@xmath7 , for 0 - 50 kpc from the center of ngc 5128 .
the top panel shows all 340 gcs with rotation amplitude @xmath234 km s@xmath7 and rotation axis @xmath235 east of north .
the middle panel shows the 178 metal - poor clusters with @xmath236 km s@xmath7 and @xmath237 east of north , and the bottom panel shows the 158 metal - rich clusters with @xmath238 km s@xmath7 and @xmath239 east of north .
the squares represent the weighted velocities in @xmath240 bins . ] | a new kinematic and dynamic study of the halo of the giant elliptical galaxy ngc 5128 is presented . from a spectroscopically
confirmed sample of @xmath0 globular clusters and @xmath1 planetary nebulae , the rotation amplitude , rotation axis , velocity dispersion , and total dynamical mass are determined for the halo of ngc 5128 .
the globular cluster kinematics were searched for both radial dependence and metallicity dependence by subdividing the globular cluster sample into 158 metal - rich ( [ fe / h]@xmath2 ) and 178 metal - poor ( [ fe / h]@xmath3 ) globular clusters .
our results show that the kinematics of the metal - rich and metal - poor subpopulations are quite similar : over a projected radius of @xmath4 kpc , the mean rotation amplitudes are @xmath5 and @xmath6 km s@xmath7 for the metal - rich and metal - poor populations , respectively .
there is a indication within @xmath8 kpc that the metal - poor clusters have a lower rotation signal than in the outer regions of the galaxy .
the rotation axis shows an interesting twist at 5 kpc , agreeing with the zero - velocity curve presented by peng and coworkers . within 5 kpc ,
both metal - rich and metal - poor populations have a rotation axis nearly parallel to the north - south direction , which is @xmath9 , while beyond 5 kpc the rotation axis twists @xmath10 .
the velocity dispersion displays a steady increase with galactocentric radius for both metallicity populations , with means of @xmath11 and @xmath12 km s@xmath7 within a projected radius of 15 kpc for the metal - rich and metal - poor populations ; however , the outermost regions suffer from low number statistics and spatial biases .
the planetary nebula kinematics are slightly different .
out to a projected radius of @xmath13 kpc from the center of ngc 5128 , the planetary nebulae have a higher rotation amplitude of @xmath14 km s@xmath7 , and a rotation axis of @xmath15 east of north , with no significant radial deviation in either determined quantity .
the velocity dispersion decreases with galactocentric distance .
the total mass of ngc 5128 is found using the tracer mass estimator , described by evans et al .
, to determine the mass supported by internal random motions and the spherical component of the jeans equation to determine the mass supported by rotation .
we find a total mass of @xmath16 @xmath17 from the planetary nebula data extending to a projected radius of 90 kpc .
the similar kinematics of the metal - rich and metal - poor globular clusters allow us to combine the two subpopulations to determine an independent estimate of the total mass , giving @xmath18 @xmath17 out to a projected radius of 50 kpc .
lastly , we publish a new and homogeneous catalog of known globular clusters in ngc 5128 .
this catalog combines all previous definitive cluster identifications from radial velocity studies and @xmath19 imaging studies , as well as 80 new globular clusters with radial velocities from a study of m.a .
beasley et al .
( in preparation ) . |
an iterated function system ( ifs ) may be used to construct fractal interpolation functions for some data @xcite .
the simplest example of interpolating a function @xmath0 , given data points ( @xmath1 ) , @xmath2 , starts with an ifs @xmath3 the coefficients @xmath4 , and @xmath5 determined from the conditions , for @xmath6 , @xmath7 which leads to @xmath8 with these , the transformation of eq .
( [ onevar ] ) can be written as @xmath9 in which form it is apparent @xmath10 is determined by a linear ( in @xmath11 ) interpolating function between the points ( @xmath12 ) and ( @xmath13 ) .
graphs of fractal interpolating functions can then be made by applying the random iteration algorithm : * initialize ( @xmath14 ) to a point in the interval of interest * for a set number of iterations * * randomly select a transformation @xmath15 * * plot ( @xmath16 ) @xmath17 * * set ( @xmath14 ) @xmath18 ( @xmath16 ) * end for in this paper we consider two non linear generalizations of such fractal interpolating functions .
the first concerns how to extend the linear interpolation of eq .
( [ linint ] ) to higher degree interpolations .
the second generalization arises when one considers the construction of fractal interpolating functions for functions of two ( or more ) variables in this case , even a linear interpolation of the form of eq .
( [ linint ] ) , when applied to each variable , will result in a non linear interpolating function .
this has an obvious application in representing two dimensional images , where a function of two variables ( the pixel coordinates ) could represent a black and white image ( using a boolean function ) , a gray scale image ( using a scalar function ) , or a colour image ( using a vector valued function of the three _ rgb _ [ red , green , blue ] values ) .
this problem has been examined extensively in the context of image compression @xcite ; in the last section we consider a related problem of using these iterated function systems to rescale images , or portions thereof .
we first consider a function @xmath19 of two variables , and examine the problem of constructing a fractal interpolating function from the data @xmath20 , where @xmath21 , @xmath22 , and @xmath23 . to this end , consider the transformation @xmath24 we then impose , for @xmath25 and @xmath6 the conditions @xmath26 the coefficients turn out to be @xmath27 with these , the transformation of eq .
( [ twovar ] ) can be written as @xmath28 in which form it is apparent @xmath29 is determined by a function implementing a linear interpolation over the grid ( @xmath30 ) , ( @xmath31 ) , ( @xmath32 ) , and ( @xmath33 ) .
the interpolating functions considered up to now have used a linear interpolating formula between adjacent points to construct the ifs . in this section
we indicate how this can be generalized to quadratic interpolations . for a function @xmath0 of one variable , using data points ( @xmath1 ) , @xmath2 , consider the transformations @xmath34 and impose the conditions , for @xmath35 , @xmath36 the point @xmath37 is determined as @xmath38 with corresponding point @xmath39 .
the coefficients of the ifs are determined as @xmath40 with which the transformation can be written as @xmath41 in this form we see that a quadratic ( in @xmath42 ) interpolating function is used between the points ( @xmath13 ) , ( @xmath12 ) , and ( @xmath43 ) .
we next consider a function @xmath19 of two variables , and construct a fractal interpolating function which employs a quadratic interpolation between points . to this end , consider the transformation @xmath44 we then impose , for @xmath45 and @xmath35 , the conditions @xmath46 the points @xmath47 and @xmath48 are determined as @xmath49 along with the corresponding @xmath50 points .
the coefficients of the ifs can then be determined , by which the transformation of eq .
( [ cond2 ] ) can be written as @xmath51 although tedious to work out , the generalization of the preceding considerations to higher order interpolating functions is straightforward in principle .
as an application of the preceding , in this section we consider the task of rescaling a colour image .
this is a natural problem for an interpolating function of two variables ( @xmath52 ) interpreted as pixel coordinates the function @xmath53 in this case will be a vector valued function having three components representing the _ rgb _ value of the pixel specifying the amount of red , green , and blue present .
the procedure used to scale an image of size @xmath54 pixels wide by @xmath55 pixels high is as follows .
we first read in the rgb values of each pixel of the image , and use that as the data to construct a fractal interpolating function @xmath56 , where @xmath57 and @xmath58 .
to then resize the image , so that the resulting image is of size @xmath59 , we construct a new fractal interpolating function @xmath60 . applying the random iteration algorithm to @xmath61 , choosing independently a transformation index ( @xmath62 ) at each stage , will then result in the rescaled image .
the generalization of this procedure to rescale a portion of an image is straightforward . as examples of the results of this procedure ,
consider the figures in the appendix .
we start with the image appearing in fig .
[ lena ] , and zoom in on the area of the face .
the result appears in fig .
[ lena - face ] , together with a comparison done using a simple linear interpolation scheme . zooming further into the area of the eye results in fig .
[ lena - eye ] , again with a comparison of the result of a simple linear interpolation .
generally , the number of iterations needed in the random iteration algorithm to produce acceptable images is of the order of @xmath59 , where the original image is of size @xmath63 . also , while slower , the quadratic fractal interpolating function typically produces , for the same number of iterations , a `` smoother '' looking image than the corresponding linear interpolating function .
however , as with all interpolation schemes , there comes a point where such higher order interpolating formulas actually start to produce worse results due to an artificially high sensitivity to fluctuations in the data .
some informal tests of this procedure seems to indicate that better results are obtained for images of people , natural scenery , etc .
, as opposed to those containing lettering , simple geometric shapes , and similar constructs .
this might be expected , given the general fractal nature of such objects in nature . however , as with all interpolating functions , it is important to remember that no structural information beyond that of the original image is being provided ( for example , one could not zoom in on the face of fig .
[ lena ] to such a degree as to see individual skin pores ) .
the preceding demonstrates that these non linear fractal interpolating functions of two variables can be used in principle to represent images
. it would be interesting to extend these considerations to the case of partitioned iterated function systems , upon which much work has been done with respect to compressing images @xcite .
work along these directions is in progress .
this work was supported by the natural sciences and engineering research council of canada .
99 m. f. barnsley , _ fractals everywhere _ ( academic press , san diego , ca , 1993 ) h. o. peitgen , h. jrgens , and d. saupe , _ chaos and fractals new frontiers of science _ ( springer verlag , new york , 1992 )
. h. o. peitgen and d. saupe , _ the science of fractal images _ ( springer verlag , new york , 1988 ) .
a. jacquin , ieee trans . on image processing , january , 1992 .
m. barnsley , l. hurd , and l. anson , _ fractal image compression _ ( a. k. peters , new york , 1993 )
. y. fisher ( editor ) , _ fractal image compression : theory and application _ ( springer verlag , new york , 1995 ) .
r. massopust , _ fractal functions , fractal surfaces and wavelets _ ( academic press , san diego , ca , 1994 ) . | we consider non linear generalizations of fractal interpolating functions applied to functions of one and two variables . the use of such interpolating functions in resizing images is illustrated . |
@xmath0 meson is the hydrogen atom of quantum chromodynamics ( qcd ) , the simplest nontrivial hadron . in the leading approximation
, the @xmath1 quark in it just seats at rest at the origin and creates chromoelectric field .
light constituents ( gluons , light quarks and antiquarks ) move in this external field .
their motion is relativistic ; the number of gluons and light quark - antiquark pairs in this light cloud is undetermined and varying .
therefore , there are no reasons to expect that a nonrelativistic potential quark model describes @xmath0 meson well enough ( in contrast to the @xmath2 meson , where the nonrelativistic two - particle picture gives a good starting point ) .
similarly , @xmath3 baryon can be called the helium atom of qcd . unlike in atomic physics , where the hydrogen atom is much simpler than helium , @xmath0 and @xmath3 are equally difficult .
both have a light cloud with a variable number of relativistic particles .
the size of this cloud is the confinement radius @xmath4 ; its properties are determined by the large - distance nonperturbative qcd .
the analogy with atomic physics can tell us a lot about hadrons with a heavy quark .
the usual hydrogen and tritium have identical chemical properties , despite the fact that the tritium nucleus is 3 times heavier than the proton .
both nuclei create identical electric fields , and both stay at rest . similarly , @xmath5 and
@xmath0 mesons have identical `` hadro - chemical '' properties , despite the fact that @xmath1 quark is 3 times heavier than @xmath6 .
the proton magnetic moment is of the order of the nuclear magneton @xmath7 , and is much smaller than the electron magnetic moment @xmath8 .
therefore , the energy difference between the states of the hydrogen atom with the total spins 0 and 1 ( hyperfine splitting ) is small ( of the order @xmath9 times the fine structure ) .
similarly , the @xmath1 quark chromomagnetic moment is proportional to @xmath10 by dimensionality , and the hyperfine splitting between @xmath0 and @xmath11 mesons is small ( proportional to @xmath10 ) . unlike in atomic physics ,
both `` gross '' structure intervals and fine structure intervals are just some numbers times @xmath12 , because light components are relativistic ( a practical success of constituent quark models shows that these dimensionless numbers for fine splittings can be rather small , but they contain no small parameter ) . in the limit @xmath13
, the heavy quark spin does not interact with gluon field .
therefore , it may be rotated at will , without changing physics .
such rotations can transform @xmath0 and @xmath11 into each other ; they are degenerate and have identical properties in this limit .
this heavy quark spin symmetry yields many useful relations among heavy - hadron form factors @xcite . not only the orientation , but also the magnitude of the heavy quark spin is irrelevant in the infinite mass limit .
we can switch off the heavy quark spin , making it spinless , without affecting physics .
this trick considerably simplifies counting independent form factors , and we shall use it often . or , if we wish , we can make the heavy quark to have spin 1 ; it does not matter .
this leads to a supersymmetry group called the superflavour symmetry @xcite .
it can be used to predict properties of hadrons containing a scalar or vector heavy quark .
such quarks exist in some extensions of the standard model ( for example , supersymmetric or composite extensions ) .
this idea can also be applied to baryons with two heavy quarks .
they form a small - size bound state ( with the radius of order @xmath14 ) which has spin 0 or 1 and is antitriplet in colour . therefore ,
these baryons are similar to mesons with a heavy antiquark having spin 0 or 1 .
accuracy of this picture can not be high , because even the radius of the @xmath15 diquark is only a few times smaller than the confinement radius .
let s consider mesons with the quark contents @xmath16 , where @xmath17 is a heavy quark with mass @xmath18 ( @xmath6 or @xmath1 ) , and @xmath19 is a light quark ( @xmath20 , @xmath21 , or @xmath22 ) . as discussed above
, the heavy quark spin is inessential in the limit @xmath13 , and may be switched off . in the world with a scalar heavy antiquark , @xmath23-wave mesons have angular momentum and parity @xmath24 ; @xmath25-wave mesons have @xmath26 and @xmath27 .
the energy difference between these two @xmath25-wave states ( fine splitting ) is a constant times @xmath12 at @xmath13 , just like splittings between these @xmath25-wave states and the ground state ; however , this constant is likely to be small . in our real world , the heavy antiquark @xmath28 has spin and parity @xmath29 .
the quantum numbers of the above paragraph are those of the light fields cloud of a meson . adding the heavy antiquark spin , we obtain , in the limit @xmath13 , a degenerate doublet of @xmath23-wave mesons with spin and parity @xmath30 and @xmath31 , and two degenerate doublets of @xmath25-wave mesons , one with @xmath32 and @xmath33 , and the other with @xmath34 and @xmath35 . at a large but finite heavy - quark mass @xmath18
, these doublets are not exactly degenerate .
hyperfine splittings , equal to to some dimensionless numbers times @xmath36 , appear .
it is natural to expect that hyperfine splittings in @xmath25-wave mesons are less than in the ground - state @xmath23-wave doublet , because the characteristic distance between the quarks is larger in the @xmath25-wave case .
note that the @xmath33 mesons from the different doublets do nt differ from each other by any exactly conserving quantum numbers , and hence can mix .
they differ by the angular momenta of the light fields , which is conserved up to @xmath37 corrections ; therefore , the mixing angle should be of order of @xmath38 .
mesons with @xmath39 and @xmath21 form isodoublets ; together with isosinglets with @xmath40 , they form @xmath41 triplets . experimentally observed @xcite mesons containing the @xmath42 antiquark are shown in fig .
the energy scale at the left is in mev , relative to the lowest mass meson .
the mesons @xmath43 and @xmath44 form the doublet with the light fields quantum numbers @xmath45 .
the second @xmath25-wave doublet is suspiciously absent .
it should be close to the @xmath46 one ; it is not more difficult to produce these mesons than the @xmath46 ones .
the problem is that they are too wide , and can not be cleanly separated from the continuum .
( 60,50 ) ( 35.5,29)(0,0),title="fig : " ] ( 10,9)(0,0)[r]0 ( 10,20.29)(0,0)[r]200 ( 10,31.58)(0,0)[r]400 ( 10,42.88)(0,0)[r]600 ( 39,4)(0,0)@xmath47 ( 24,11)(0,0)@xmath48 ( 24,19.02)(0,0)@xmath49 ( 24,38.51)(0,0)@xmath50 ( 24,44.58)(0,0)@xmath51 ( 60,50 ) ( 35.5,29)(0,0),title="fig : " ] ( 10,13.02)(0,0)[r]0 ( 10,24.31)(0,0)[r]200 ( 10,35.60)(0,0)[r]400 ( 10,46.89)(0,0)[r]600 ( 39,4)(0,0)@xmath52 ( 24,11.02)(0,0)@xmath53 ( 24,17.60)(0,0)@xmath54 ( 24,34.61)(0,0)@xmath55 in the leading approximation , the spectrum of @xmath56 containing mesons is obtained from the spectrum of @xmath42 containing mesons simply by the shift by @xmath57 .
experimentally observed mesons containing the @xmath56 antiquark are shown in fig .
the spectrum of @xmath42 containing mesons is shown by dashed lines for comparison .
it is positioned in such a way that the weighted average energies of the ground - state doublets coincide , where the @xmath31 meson has weight 3 and the @xmath58 meson has weight 1 .
the states @xmath59 and @xmath60 are not resolved , and are shown by a single line .
experimentally , it is difficult to measure their masses exactly enough .
the hyperfine splitting of the ground - state doublet is smaller for @xmath0 mesons than for @xmath5 mesons , as expected .
in @xmath23-wave @xmath61 baryons , the light quark spins can add giving @xmath62 or @xmath33 . in the first case their spin wave function is antisymmetric ; the fermi statistics and the antisymmetry in colour require an antisymmetric flavour wave function .
hence the light quarks must be different ; if they are @xmath20 , @xmath21 , then their isospin is @xmath63 .
with the heavy quark spin switched off , this gives the @xmath64 baryon @xmath65 with @xmath63 .
if one of the light quarks is @xmath22 , we have the isodoublet @xmath66 , which forms a @xmath41 antitriplet together with @xmath65 . with
the heavy quark spin switched on , these baryons have @xmath67 . in the @xmath33 case ,
the flavour wave function is symmetric .
if the light quarks are @xmath20 , @xmath21 , then their isospin is @xmath68 .
this gives the @xmath33 isotriplet @xmath69 ; with one @xmath22 quark the isodoublet @xmath70 ; with two @xmath22 quarks the isosinglet @xmath71 .
together they form the @xmath41 sextet . with
the heavy quark spin switched on , we obtain the degenerate doublets with @xmath67 , @xmath46 : @xmath69 , @xmath72 ; @xmath70 , @xmath73 ; @xmath71 , @xmath74 .
the hyperfine splittings in these doublets are of the order of @xmath36 . mixing between @xmath66 and @xmath70
is suppressed both by @xmath37 and by @xmath41 .
there is a large number of @xmath25-wave excited states ; we shall not discuss them here .
( 90,60 ) ( 50.5,33)(0,0 ) quark , title="fig : " ] ( 10,13)(0,0)[r]0 ( 10,28.03)(0,0)[r]200 ( 10,43.07)(0,0)[r]400 ( 27,8)(0,0)@xmath75 $ ] ( 45,8)(0,0)@xmath76 ( 63,8)(0,0)@xmath77 ( 81,8)(0,0)@xmath78 ( 45,4)(0,0)@xmath79 ( 63,4)(0,0)@xmath80 ( 27,17)(0,0)@xmath81 ( 27,32.22)(0,0)@xmath82 ( 27,42.69)(0,0)@xmath83 ( 45,21.63)(0,0)@xmath81 ( 45,34.56)(0,0)@xmath84 ( 63,22.85)(0,0)@xmath81 ( 63,30.92)(0,0)@xmath81 ( 63,44.15)(0,0)@xmath84 ( 63,57)(0,0)@xmath83 ( 81,48.51)(0,0)@xmath81 experimentally observed @xcite baryons containing @xmath6 quark are shown in fig .
the higher states in the first and the third columns are @xmath25-wave . in the third column
, the lowest state @xmath85 is followed by the doublet @xmath86 , @xmath87 . the @xmath88 baryon has not yet been observed .
the only baryon with @xmath1 quark , @xmath89 , has been discovered so far .
in the leading @xmath90 approximation , the masses @xmath91 and @xmath92 are both equal to @xmath93 , where @xmath94 is the energy of the ground state of light fields in the chromoelectric field of the @xmath56 antiquark . this energy @xmath94 is of order of @xmath12 .
excited states of light fields have energies @xmath95 , giving excited degenerate doublets with the masses @xmath96 .
there are two @xmath10 corrections to the masses .
first , the @xmath56 antiquark has an average momentum squared @xmath97 , which is of order of @xmath98 .
therefore , it has kinetic energy @xmath99 .
second , the @xmath56 chromomagnetic moment interacts with the chromomagnetic field created by light constituents at the origin , where @xmath56 stays .
this chromomagnetic field is proportional to the light fields angular momentum @xmath100 .
therefore , the chromomagnetic interaction energy is proportional to @xmath101 = \left\ { \begin{array}{ll } \displaystyle - \frac{3}{4}\ , , & s=0 \\[2 mm ] \displaystyle\phantom{-}\frac{1}{4}\ , , & s=1 \end{array } \right.\ ] ] where @xmath102 is the meson spin .
if we denote this energy for @xmath0 as @xmath103 , then for @xmath11 it will be @xmath104 . here
@xmath105 is of order of @xmath98 .
the @xmath0 , @xmath11 meson masses with @xmath10 corrections taken into account are given by the formulae @xmath106 the hyperfine splitting is @xmath107 taking into account @xmath108 , we obtain @xmath109 the difference @xmath110 is given by a similar formula with @xmath111 instead of @xmath112 .
therefore , the ratio @xmath113 experimentally , this ratio is 0.89 .
this is a spectacular confirmation of the idea that violations of the heavy quark spin symmetry are proportional to @xmath37 .
in fact , the matrix element @xmath105 depends on the normalization scale , and hence is not quite the same for @xmath5 and @xmath0 ; this produces moderate perturbative corrections to ( [ ratio ] ) .
@xmath25-wave excited states decay into the ground state emitting a pion . in the ideal world with an infinitely heavy scalar @xmath42 , the @xmath114 @xmath25-wave meson decays into the @xmath115 ground - state meson plus a pion ( having @xmath30 ) with the orbital angular momentum @xmath116 ( @xmath23-wave ) ; the @xmath46 @xmath25-wave meson decays into the ground - state meson plus a pion with @xmath117 ( @xmath5-wave ) .
the pion momenta @xmath118 in these decays are rather small , and hence the decay of the @xmath46 meson ( whose width is proportional to @xmath119 ) is strongly suppressed .
the decay of the @xmath114 meson is not suppressed , its width is proportional to @xmath118 .
therefore , in our real world , the @xmath64 , @xmath33 doublet mesons are very wide , and difficult to observe .
let s consider decays of @xmath120 , @xmath121 , taking the @xmath42 spin into account , but still in the limit @xmath122 .
the widths of @xmath120 and @xmath121 are equal , because the @xmath42 spin plays no role in the decay .
@xmath120 decays only into @xmath123 , the decay into @xmath124 is forbidden by the angular momentum conservation .
@xmath121 can decay into both @xmath124 and @xmath123 . in order to find the branching ratio @xmath125
, we shall use a simple device known as shmushkevich factory .
let s take a sample of @xmath126 mesons with random polarizations of @xmath42 .
then 3/8 of this sample are @xmath120 , and 5/8 are @xmath121 .
now let s wait for a small time @xmath127 ; a fraction @xmath128 of the sample will decay .
ground - state mesons produced have randomly polarized @xmath42 .
therefore , 1/4 of them are @xmath5 , and 3/4 are @xmath129 .
@xmath5 mesons are only produced from @xmath121 : @xmath130 and hence @xcite @xmath131 in the limit @xmath122 , @xmath5 and @xmath129 are degenerate , and so are @xmath120 and @xmath121 . in the real world , pion momenta in these decays differ .
the widths are proportional to @xmath119 , and even rather small momentum differences produce a drastic effect .
it seems natural to suppose that the heavy - quark spin symmetry predictions hold for the coefficients in front of @xmath119 .
then @xmath132 while the experimental value is @xmath133 @xcite .
formally , the difference of @xmath134 from 1 is a @xmath135 correction .
we can only hope that this kinematical @xmath135 effect , included in the above estimate , is dominant .
let s now discuss the @xmath0-meson leptonic decay constant @xmath136 .
it is defined by @xmath137 where the one - particle state is normalized in the usual lorentz - invariant way : @xmath138 this relativistic normalization becomes senseless in the limit @xmath90 , and the non - relativistic normalization @xmath139 should be used instead .
then , for the @xmath0 meson at rest , @xmath140 denoting this matrix element ( which is mass - independent at @xmath90 ) as @xmath141 , we obtain @xmath142\ , , \label{fb}\ ] ] and hence @xmath143\ , .
\label{fratio}\ ] ] in fact , the matrix element @xmath144 depends on the normalization scale , and hence is not quite the same for @xmath5 and @xmath0 ; this produces moderate perturbative corrections to ( [ fratio ] ) .
lattice simulations and qcd sum rules show that the @xmath135 correction in the formula for @xmath145 similar to ( [ fb ] ) is of order 100% , so that the accuracy of ( [ fratio ] ) is not high .
experimentally @xcite , @xmath146 from the @xmath147 and @xmath148 decays .
the branching @xmath149 should be of order @xmath150 , so that a direct measurement of @xmath151 at @xmath0-factories seems feasible .
theoretical estimates of @xmath136 vary by about a factor 2 .
let s discuss the decay @xmath152 , where @xmath153 is a virtual @xmath154 which decays into @xmath155 . in the limit @xmath90 , @xmath122 , it is enough to consider the case when @xmath56 , @xmath42 , and @xmath153 are scalar .
we concentrate our attention on decays of @xmath0 with 4-velocity @xmath156 into @xmath157 with 4-velocity @xmath158 .
let @xmath159 be the scalar current which replaces a scalar @xmath56 with 4-velocity @xmath156 by a scalar @xmath42 with 4-velocity @xmath158 . with the non - relativistic normalization of the scalar quark wave functions ,
they are just 1 , and the quark decay matrix element is @xmath160 the ground - state @xmath0 meson has @xmath67 ; @xmath157 will be used to denote generically a ground - state or excited @xmath161 meson .
it is convenient to work in the @xmath0 rest frame .
let the @xmath162 axis be in the direction of @xmath157 motion .
angular momentum conservation gives @xmath163 .
reflection in a plane containing the @xmath162 axis transforms a state @xmath164 into @xmath165 .
therefore , the amplitude of the @xmath166 into @xmath166 transition is equal to that of the @xmath167 into @xmath167 transition , up to a phase factor ; an @xmath168 into @xmath169 transition is allowed only when the `` naturalness '' @xmath170 is conserved @xcite .
for example , the transition @xmath171 is described by a single form factor ; @xmath172 is forbidden by `` naturalness '' ( and also suppressed by isospin ) ; @xmath173 is described by two form factors ( @xmath174 and @xmath175 ) @xcite .
the transitions of the ground - state @xmath114 @xmath176 meson into an @xmath23-wave @xmath114 @xmath161 meson , and into a @xmath25-wave @xmath115 or @xmath27 @xmath161 meson , are described by one form factor each @xcite : @xmath177 where @xmath178 , @xmath179 is the minkowski angle between the 4-velocities of @xmath0 and @xmath157 .
the dirac wave function @xmath20 of the initial @xmath114 meson satisfies @xmath180 and is normalized by the non - relativistic condition @xmath181 ; the sum over its two polarizations is @xmath182 the dirac wave function @xmath183 of the final @xmath184 meson has similar properties , with @xmath158 instead of @xmath156 . the rarita
schwinger wave function @xmath185 of the spin @xmath186 meson satisfies @xmath187 , @xmath188 @xmath189 , and is normalized by @xmath190 ; the sum over four polarization of the meson is @xmath191 all the form factors of @xmath0 transitions into @xmath157 , @xmath192 via the vector and axial @xmath193 weak currents are proportional to the isgur wise form factor @xmath194 , with trivial kinematical coefficients . when the current @xmath195 replaces an infinitely heavy @xmath56 by an infinitely heavy @xmath42 with the same 4-velocity and colour , light fields do nt notice it : @xmath196 the maximum @xmath197 accessible in the @xmath198 , @xmath192 decays is about 1.6 ; a rough sketch of @xmath194 as extracted from experimental data is shown in fig .
[ iwexp ] . at @xmath199 ,
the isgur wise form factor behaves as @xcite @xmath200 up to logarithmic factors .
( 66,44 ) ( 35,23)(0,0 ) ( 5,0)(0,0)[b]@xmath201 ( 15,0)(0,0)[b]@xmath202 ( 25,0)(0,0)[b]@xmath203 ( 35,0)(0,0)[b]@xmath204 ( 45,0)(0,0)[b]@xmath205 ( 55,0)(0,0)[b]@xmath206 ( 65,0)(0,0)[b]@xmath197 ( 0,3)(0,0)[l]@xmath207 ( 0,11)(0,0)[l]@xmath208 ( 0,19)(0,0)[l]@xmath209 ( 0,27)(0,0)[l]@xmath210 ( 0,35)(0,0)[l]@xmath211 ( 0,43)(0,0)[l]@xmath201 the @xmath212 form factor is also proportional to @xmath194 . in this case ,
the form factor has a cut in the annihilation channel from @xmath214 to @xmath215 .
therefore , @xmath194 has a cut from @xmath216 to @xmath217 ( fig .
[ iwcomplex ] ) . geometrically speaking
, @xmath218 corresponds to minkowski angles between the world lines of the incoming heavy quark and the outgoing one this is the scattering ( or decay ) channel .
when @xmath219 , the world line is straight , and there is no transition at all ( see ( [ iwnorm ] ) ) . when @xmath220 , the angle is euclidean
when @xmath221 , we have a minkowski angle again , but one of the 4-velocities is directed into the past this is the annihilation channel . at the point @xmath216 ,
the heavy quark returns back along the same world line .
in fact , the very concept of the isgur wise form factor is inapplicable near this point .
the hqet picture is based on the fact that heavy quarks move along straight world lines .
if their relative velocity in the annihilation channel is @xmath222 , they rotate around each other instead .
the @xmath0 meson form factor has poles below the threshold corresponding to @xmath2 mesons with binding energies @xmath223 ; its behaviour in this region is not universal .
the concept of the isgur wise form factor is only applicable at @xmath224 ( fig .
[ iwcomplex ] ) .
( 42,22 ) ( 21,11)(0,0 ) plane , title="fig : " ] ( 31,8)(0,0)1 squaring the matrix elements ( [ iw ] ) , summing over the final meson polarizations ( using ( [ rarita ] ) for the sipn @xmath186 meson ) , averaging over the initial meson polarizations , and normalizing to the quark decay ( [ norm ] ) , we obtain the branching ratios @xmath225 they are the fractions of the number of @xmath226 decays with @xmath227 velocity @xmath158 , where the hadronic system @xmath227 happens to be a single meson .
the decay @xmath228 is @xmath23-wave , and therefore the squared matrix element tends to a constant at @xmath229 .
the decays @xmath230 and @xmath231 are @xmath25-wave , and therefore the squared matrix elements behave as the relative velocity squared , @xmath232 .
similarly , decays into @xmath5-wave mesons @xmath46 , @xmath233 are @xmath5-wave , and behave as @xmath234 at @xmath229 .
if we choose the mass of the virtual @xmath153 larger than @xmath235 , we can consider the channel @xmath236 .
the squared matrix elements are given by the formulae ( [ br ] ) with the extra factor 2 , because we sum over the @xmath5 polarizations now , not average .
the decays of the scalar @xmath153 into @xmath237 , @xmath238 , and @xmath239 are @xmath25-wave , @xmath23-wave , and @xmath5-wave .
therefore , their squared matrix elements behave as the relative velocity to the power 2 , 0 , and 4 , correspondingly .
this explains the behaviour of the formulae ( [ br ] ) at @xmath240 .
the inclusive decay rate @xmath241 can be written as @xmath242 , where @xmath178 , @xmath158 is the @xmath227 4-velocity : @xmath243 , and @xmath244 is the excitation energy ( we are still in the limit @xmath122 , where @xmath245 ) .
the structure function is @xmath246 where the sums run over final states with the indicated quantum numbers , @xmath247 are their excitation energies , the index @xmath248 is not explicitly shown in the form factors @xmath249 and @xmath250 , and the dots mean the contribution of @xmath5-wave and higher states . at @xmath251 , @xmath252 . a qualitative sketch of @xmath253 as a function of @xmath254 at some fixed @xmath255
is shown in fig .
[ struct ] .
it contains a @xmath256 peak at @xmath257 due to the transition into the ground state ( @xmath157 and @xmath192 ) , then some peaks due to excited states which become wider when @xmath254 increases , and the curve becomes smooth . at @xmath258 ,
it is given by the perturbative gluon radiation : @xmath259 its @xmath254 dependence is evident from dimensionality , and the @xmath179 dependence is given by the famous qed soft - photon radiation function .
it is also known from classical electrodynamics : this function is the distribution in the radiation energy when a charge suddenly changes its velocity from @xmath156 to @xmath158 .
( 102,64 ) ( 51,33)(0,0 ) decay structure function , title="fig : " ] ( 1,0)(0,0)[b]@xmath207 ( 21,0)(0,0)[b]@xmath260 ( 101,0)(0,0)[b]@xmath254 ( 2,63)(0,0)[l]@xmath253 the total decay probability is unity : @xmath261 this is the bjorken sum rule @xcite . in particular , the decay rate into the ground - state @xmath114 meson must not exceed the total one : @xmath262 at @xmath199 , the ground state is rarely produced , and the isgur wise form factor ( [ asympt ] ) is much less than the bound ( [ bj1 ] ) .
the bjorken sum rule becomes much simpler in the small @xmath179 limit .
the isgur wise form factor of the transition into the ground state behaves as @xmath263 , and those of transitions into higher @xmath23-wave mesons as @xmath264 .
expanding ( [ bj ] ) up to linear terms in @xmath265 , we obtain @xmath266 ( \cosh\vartheta-1 ) = 1\ , .
\label{bj2}\ ] ] @xmath5-wave final state do nt contribute in this order , as well as higher @xmath23-wave final states .
therefore , the slope of the isgur wise form factor @xmath267 is expressed via the form factors of @xmath25-wave meson production at @xmath219 : @xmath268 in particular , @xmath269 ( this also follows from ( [ bj1 ] ) ) .
we can also consider the inclusive decay of polarized @xmath0 .
its structure function is ( we use ( [ rarita ] ) for the spin @xmath186 meson ) @xmath270 \sum_{\frac{3}{2}^- } \tau_{3/2}^2(\cosh\vartheta ) \delta(\varepsilon-\varepsilon_i)\\ & { } + \cdots\end{aligned}\ ] ] averaging it over polarizations @xmath271 , we reproduce ( [ str ] )
. the decay rate does not depend on the initial meson polarization .
this gives the uraltsev sum rule @xcite @xmath272 it becomes much simpler at @xmath229 .
@xmath5- and higher - wave contributions vanish , and @xmath273 substituting @xmath274 from the this sum rule into ( [ bj3 ] ) , we obtain @xmath275 in particular , @xmath276 this uraltsev boundary is much stronger than the bjorken boundary ( [ bj4 ] ) .
experimantally , @xmath277 .
more sum rules can be obtained from the energy conservation . in the @xmath156 rest frame ,
the light fields in @xmath0 have a definite energy @xmath278 .
they have no definite momentum , because they are in the external chromoelectric field created by the @xmath56 antiquark .
its average is @xmath279 , and the average of its square is @xmath280 ( it is the same as the average squared momentum of the heavy antiquark , see ( [ mass ] ) ) .
the light fields energy in the @xmath158 frame is @xmath281 .
when @xmath56 with the 4-velocity @xmath156 is suddenly transformed into @xmath42 with the 4-velocity @xmath158 , the light fields remain in their original state at the first moment . after that
, the energy @xmath282 is conserved in the field of @xmath42 moving with the 4-velocity @xmath158 .
therefore , the average excitation energy of @xmath227 and the average squared excitation energy are @xmath283 ( because @xmath284 ) .
this gives the voloshin sum rule @xcite and the bgsuv sum rule @xcite : @xmath285 the transition into the ground - state meson with @xmath257 does not contribute here .
expanding these sum rules up to linear terms in @xmath265 , we obtain , similarly to ( [ bj2 ] ) , @xmath286 let @xmath260 be the minimum @xmath25-wave excitation energy .
then , replacing @xmath247 by @xmath260 in the left - hand side of ( [ vol2 ] ) , we make it smaller . after singling out this factor @xmath260 ,
the remaining sum is @xmath287 ( [ bj3 ] ) : @xmath288 this inequality can also be rewritten as an upper bound on @xmath267 @xcite : @xmath289 similarly , replacing @xmath290 by @xmath291 in the left - hand side of ( [ bgsuv2 ] ) , we make it smaller . after singling out the factor @xmath260 ,
the remaining sum is @xmath94 ( [ vol2 ] ) , and @xcite @xmath292 ( at the second step , the inequality ( [ vol3 ] ) was used ) . if the lowest resonances in the @xmath115 and @xmath27 channels with nearly equal energies dominate in ( [ vol2 ] ) , ( [ bgsuv2 ] ) , then the inequalities ( [ vol3 ] ) , ( [ bgsuv3 ] ) should be close to equalities .
this is , probably , the case . strictly speaking
, the minimum excitation energy @xmath260 is equal to @xmath293 , because the ground - state meson plus a soft pion can have the needed quantum numbers .
this @xmath260 is small , and the bounds are very weak .
however , the coupling of a soft pion with a heavy meson is small , and these states contribute little to the sum rules ( [ vol2 ] ) , ( [ bgsuv2 ] ) .
the first important contribution comes from the lowest @xmath25-wave resonances . as you may have noticed
, the integral in the bjorken sum rule ( [ bj ] ) logarithmically diverges at large @xmath254 , due to ( [ pert ] ) .
the integrals ( [ vol ] ) , ( [ bgsuv ] ) diverge even more strongly .
therefore , if we want to take @xmath294 effects into account , we have to cut these integrals somehow .
on the other hand , the form factors @xmath295 , @xmath296 , @xmath297 depend on the normalization point @xmath298 , if perturbative effects are taken into consideration .
it is natural to expect that the sum rules are valid , up to @xmath299 corrections , when @xmath298 is of the order of the cutoff energy ( this is discussed in @xcite in more details ) .
therefore , the anomalous dimension of the hqet heavy - heavy quark current ( which describes the @xmath298-dependence of the formfactors ) is proportional to the soft - photon radiation function ( [ pert ] ) . for more information about inclusive sum rules ,
see @xcite .
it is also possible to establish a bound on the isgur wise form factor at the cut ( fig .
[ iwcomplex ] ) .
the decay rate @xmath236 must be less than the total decay rate @xmath300 : @xmath301 where @xmath302 is the number of light flavours , and @xmath303 is the number of colours .
the left - hand side should be the sum over light flavours , if @xmath41 breaking is taken into account .
the inequality ( [ drt ] ) is applicable at @xmath224 ( outside the pitched region in fig .
[ iwcomplex ] ) , where the isgur wise form factor has sense , and the decay rate @xmath300 is given by the free quark formula . at @xmath304 , production of the pair of ground - state mesons is rare , and the isgur wise form factor ( [ asympt ] ) is much less than the bound ( [ drt ] ) .
the inequality ( [ drt ] ) was proposed in @xcite , where the factor @xmath302 has been erroneously omitted .
it was used , together with analyticity , to obtain bounds on the isgur wise form factor in the physical region .
uraltsev : ` heavy quark expansion in beauty and its decays ' . in _ heavy flavor physics a probe of nature s grand design _ , ed . by i.
bigi , l. moroni ( ios press , amsterdam 1998 ) p. 329 ; + ` topics in the heavy quark expansion ' . in _ at the frontier of particle physics : handbook of qcd _ , ed .
m. shifman ( world scientific , singapore 2001 ) v. 3 , p. 1577 | some topics which can be easily explained to undergraduate students are presented , with elementary derivations . for a more systematic treatment of heavy - quark physics , see the textbook @xcite . |
the carbon star phenomenon , i.e. , the presence of stars for which the abundance of carbon exceeds that of oxygen in the atmospheres , occurs during the final evolutionary stage of low to intermediate mass stars ( @xmath11@xmath810m@xmath4 ; wallerstein & knapp 1998 ) , and it is believed to be due to a dredge - up process driven by quasi - periodic he - shell flashes ( straniero et al .
these stars , located on the asymptotic giant branch ( agb ) , lose copious amounts of matter ( @xmath110@xmath9@xmath810@xmath10m@xmath4yr@xmath5 ) in an intense stellar wind .
in fact , it is the mass loss that determines the agb lifetime and not the nuclear burning processes .
hence , a determination of the mass loss characteristics , i.e. , the dependence on time , mass , chemistry , metallicity , etc .
, is of utmost importance for our understanding of late stellar evolution .
the mass loss creates a circumstellar envelope ( cse ) of gas and dust around the star .
the low temperature of the central star allows the formation of a wide variety of molecular species in its atmosphere , and the expanding gas is therefore mainly in molecular form .
the chemistry in the cse itself can be very rich , and it depends on the c / o - ratio , the thickness of the envelope , and the strength of the ambient ultraviolet radiation field .
presently , @xmath160 different molecular species have been identified in cses around agb - stars ( olofsson 1997 ) .
the study of agb - cses is important for the understanding of the late stages of stellar evolution as mentioned above .
in addition , an understanding of the atmospheric and circumstellar chemistry is required to determine the elemental abundances in these winds , which contribute to the chemical evolution of the interstellar medium ( busso et al .
finally , since agb stars are the progenitors of planetary nebulae ( pne ) , the cses are very likely important ingredients in the formation of pne ( kwok 1993 ) . observations of molecular millimetre - wave line emission have proven to be one of the best tools for studying the structure , kinematics , and chemistry of the cses , and co and oh observations have proven particularly useful for the determination of accurate mass loss rates [ see olofsson @xcite and references therein ] .
while the overall picture of mass loss on the agb is fairly well understood , some of the underlying physical principles are not .
the prevailing theory , which needs to be thoroughly tested observationally , is that the mass loss occurs in a two stage process .
first the pulsations of the star deposit energy in the atmosphere , leading to a considerably increased scale height , and sufficient matter at low enough temperatures for efficient dust formation .
radiation pressure on the dust grains accelerate the dust wind to its terminal expansion velocity in the second stage .
the gas is momentum coupled , through collisions , to the dust , and it will be effectively dragged along .
the mechanisms that cause the mass loss to vary on a wide range of time scales still need to be pinpointed . eventually , the possibility of mass ejected in clumps has to be addressed , although in this paper we will consider only a smooth spherically symmetric wind .
this study of an essentially complete sample of visually bright , relatively unobscured carbon stars will hopefully increase our knowledge of the modelling of circumstellar molecular radio line emission , the mass loss characteristics , the circumstellar chemistry , the elemental abundances in the winds , and possibly the evolutionary status of these objects .
it will also provide us with the tools required to determine the properties of higher mass loss rate objects , for which the central star may be completely obscured . in this paper the basic physical characteristics of the cses ,
e.g. , the stellar mass loss rate , the gas kinetic temperature , and the gas expansion velocity , are determined using a radiative transfer analysis of the observed @xmath11co ( hereafter co ) millimetre - wave line emission .
the co molecule is particularly well suited for this purpose , since it is difficult to photodissociate and easy to excite through collisions , and thus is a very good tracer of the molecular gas density and temperature .
furthermore , the molecular excitation is of a quasi - thermal nature which simplifies detailed modelling .
we have selected a sample of 68 bright n- and j - type carbon stars for which circumstellar co emission was detected by olofsson et al .
( 1993a ) . their original sample consisted of carbon stars brighter than @xmath12@xmath132mag . ,
in total 120 sources . the detection rate for all the carbon stars searched for circumstellar co was as high as 72% .
the original sample is thought to be close to complete to distances below 1kpc [ it is estimated that about a third of the carbon stars within this distance are missing , presumably the ones with thick and dusty cses , i.e. , the high mass loss objects ( olofsson et al . 1993a ) ] , and for distances below 500pc all sources ( a total of 41 ) were detected in co ( note that the distances used in this paper are generally lower than those adopted in olofsson et al.1993a ) .
these facts taken together led to the conclusion that the great majority of all n - type carbon stars are losing mass at a rate higher than about 10@xmath9m@xmath4yr@xmath5 .
they do not , however , constitute a homogeneous group in terms of circumstellar properties .
instead they show a wide range of gas expansion velocities and mass loss rates , presumably due to differences in mass , evolutionary stage , etc .. in addition , the reasonably well studied , high mass loss rate carbon star , lp and ( also known as irc+40540 ) , is included in our analysis , since it provides a good test case for the radiative transfer model . the co ( @xmath14@xmath13@xmath15@xmath16@xmath17 and @xmath14@xmath13@xmath18@xmath16@xmath15 ) observations presented in olofsson et al .
@xcite were obtained using the 20 m telescope at onsala space observatory ( oso ) , sweden , the 15 m swedish - eso sub - millimetre telescope ( sest ) at la silla , chile , and the iram 30 m telescope at pico veleta , spain , during the years 1986@xmath81992 .
llcccccc & & & & & & & + & & & & & & & + & sest & 3@xmath82 & 0.70 & 17.9 & @xmath81.8 & 16.1 & 1.0 + & sest & 3@xmath82 & 0.94 & 8.7 & 1.0 & 7.5 & 2.5 + @xmath19 & nrao & 1@xmath80 & 0.37 & 7.8 & @xmath816.4 & 11.5 & 0.5 + & sest & 3@xmath82 & 0.48 & 12.9 & 12.4 & 19.5 & 1.7 + @xmath19 & sest & 1@xmath80 & 0.06 & 1.2 & 1.6 & 8.6 & @xmath80.7 + & sest & 2@xmath81 & 0.27 & 4.9 & 0.0 & 10.0 & 0.4 + & sest & 3@xmath82 & 0.27 & 4.8 & @xmath81.4 & 11.8 & 1.1 + & sest & 3@xmath82 & 0.24 & 6.3 & 20.7 & 13.9 & 0.4 + & sest & 3@xmath82 & 0.74 & 20.8 & @xmath810.6 & 16.9 & 0.8 + & nrao & 1@xmath80 & 0.06 & 0.7 & @xmath810.3 & 7.6 & 1.3 + & sest & 1@xmath80 & 0.12 & 1.7 & @xmath812.8 & 8.6 & 1.0 + & sest & 2@xmath81 & 0.30 & 3.2 & @xmath814.3 & 7.5 & 1.7 + & nrao & 1@xmath80 & 7.1 & 170.8 & @xmath825.9 & 14.3 & 0.7 + @xmath19 & sest & 3@xmath82 & 0.30 & 5.5 & 25.7 & 13.1 & 1.2 + & nrao & 1@xmath80 & 1.3 & 36.4 & @xmath81.6 & 15.8 & 0.7 + & sest & 3@xmath82 & 1.2 & 15.3 & @xmath82.0 & 8.7 & 1.6 + @xmath19 & nrao & 2@xmath81 & 0.26 & 11.3 & 11.6 & 24.1 & 1.1 + & nrao & 2@xmath81 & 2.9 & 50.1 & 14.2 & 11.0 & 1.1 + & sest & 3@xmath82 & 0.83 & 18.6 & 1.2 & 14.5 & 1.1 + @xmath19 & nrao & 2@xmath81 & 0.28 & 8.7 & 6.3 & 24.1 & 0.0 + & oso & 1@xmath80 & 2.6 & 57.0 & @xmath816.9 & 13.8 & 1.1 + & nrao & 2@xmath81 & 3.1 & 60.7 & @xmath817.1 & 14.0 & 1.7 + @xmath20 & sest & 3@xmath82 & 1.0 & 7.0 & 1.0 & 7.5 & 2.0 + we have supplemented the original data with new co observations of some of the sample stars . in particular , in the @xmath14@xmath13@xmath21@xmath16@xmath18 line using the sest ( august 1992 ) .
also some @xmath14@xmath13@xmath15@xmath16@xmath17 and @xmath14@xmath13@xmath18@xmath16@xmath15 data were obtained at sest ( october 1998 ) .
a few sources were observed in the @xmath14@xmath13@xmath15@xmath16@xmath17 and @xmath14@xmath13@xmath18@xmath16@xmath15 lines with the nrao 12 m telescope at kitt peak , usa ( june 1998 ) , and some @xmath14@xmath13@xmath15@xmath16@xmath17 data were obtained using oso ( 1998 to 1999 ) .
a summary of telescope and receiver data [ the receiver temperature @xmath22 in single sideband mode ( ssb ) , the full half power main beam width @xmath23 , the main beam efficiency @xmath24 , and the corrected main beam efficiency @xmath25 at the observational frequencies are given in table [ teldat ] .
at sest , a dual channel sis mixer receiver was used to simultaneously observe at 115ghz ( the @xmath14@xmath13@xmath15@xmath16@xmath17 line ) and 230ghz ( the @xmath14@xmath13@xmath18@xmath16@xmath15 line ) . at 345ghz ( the @xmath14@xmath13@xmath21@xmath16@xmath18 line ) , a single polarization sis mixer receiver was used . as backends , two acousto - optical spectrometers ( aos ) were used ; one high resolution spectrometer ( hrs ) and one low resolution spectrometer ( lrs ) . the wideband ( 1ghz ) lrs had 1440 channels separated by 0.7mhz , whereas the narrow band ( 84mhz ) hrs used 2000 channels separated by 42khz . for the nrao 12 m observations we used a dual polarization sis receiver . as backends
two filterbanks , each with 256 channels and a channel separation of 1mhz , were used .
the spectra obtained at each of the polarizations were added to increase the signal - to - noise ratio . in addition , for strong sources , a millimetre autocorrelator ( mac ) was used in order to obtain a higher velocity resolution .
the mac was used in a configuration where the usable bandwith was 300mhz and the resolution was 98khz . at oso , a single polarization sis receiver was used for the observations . as backends two filterbanks , with bandwidths of 512mhz ( mul a ) and
64mhz ( mul b ) , were used .
the mul a used 512 channels separated by 1mhz , and the mul b filterbank used 256 channels with a separation of 250khz .
the sest and oso observations were made in a dual beamswitch mode , where the source is alternately placed in the signal and the reference beam , using a beam throw of about 12@xmath26 at sest and 11@xmath26 at oso .
this method produces very flat baselines .
the intensity scales are given in main beam brightness temperature , @xmath27 .
@xmath27@xmath13@xmath28 , where @xmath29 is the antenna temperature corrected for atmospheric attenuation using the chopper wheel method , and @xmath24 is the main beam efficiency , see table [ teldat ] . at the nrao
12 m telescope the observations were carried out using a position switching mode , with the reference position located @xmath3010@xmath26 in azimuth .
this is the preferred observation mode for spectral line observations at the 12 m telescope .
the raw spectra , which are stored in the @xmath31 scale , were converted using @xmath27@xmath13@xmath32 , where @xmath33 is the corrected main beam efficiency given in table [ teldat ] .
the @xmath31 scale is related to @xmath29 through @xmath31@xmath13@xmath29/@xmath34 , where @xmath34 is the forward scattering and spillover efficiency .
regular pointing checks were made on strong sio masers ( oso , sest ) and strong continuum sources ( nrao ) .
the pointing was usually consistent within @xmath13@xmath35 for sest and oso and @xmath15@xmath35 at nrao .
the uncertainties in the absolute intensity scales at the various telescopes are estimated to be about @xmath3615@xmath820% .
however , due to the low efficiency of the sest at 345ghz , and the narrow beam , we consider these data to be particularly uncertain in terms of calibration . the new observational results are summarized in table [ obs ] and the spectra are shown in fig .
[ spectra ] .
the line parameters , i.e. , the main beam brightness temperature at the line centre ( @xmath27 ) , the line centre velocity ( @xmath37 ) , and half the full line width ( @xmath38 ) , are obtained by fitting the following line profile to the data ( olofsson et al .
1993a ) @xmath39 ^{\beta/2},\ ] ] where @xmath40 is a parameter describing the shape of the line ( @xmath40=2 represents a parabolic line shape , and @xmath40@xmath410 means a profile with horns at the extreme velocities ) .
the integrated intensity ( @xmath42 ) is obtained by integrating the emission between @xmath43@xmath38 .
in addition , we have obtained publicly available data from the james clerk maxwell telescope ( jcmt ) at mauna kea , hawaii .
the jcmt data are taken at face value . however , in the cases where there are more than one observation available , the derived line intensities are generally consistent within @xmath3620% . the good agreement with corresponding sest intensities further support the reliability of the jcmt public data .
furthermore , interferometer observations of the co(@xmath14=@xmath15@xmath16@xmath17 ) brightness distribution around some of our carbon stars have been obtained by neri et al .
@xcite using the iram plateau de bure interferometer ( pdbi ) .
the data are publicly available and have been used in this paper .
in order to model the circumstellar molecular line emission and to derive the basic characteristics of the cses we have developed a non - lte radiative transfer code based on the monte carlo method [ bernes @xcite ; see also schier @xcite , for details ] .
an accurate treatment of the molecular excitation is needed for a correct description of the radiative transfer in an expanding circumstellar envelope , e.g. , the sobolev approximation has been shown to introduce significant errors in these circumstances ( e.g. , schnberg 1985 ) .
the monte carlo method is well suited for the study of cses , since it is very flexible and close to the physics and yet simple in principle .
for instance , one may include very complex geometries and velocity fields without being forced away from the physics of the problem .
crosas & menten 1997 have recently used the monte carlo method to model the circumstellar co radio line emission of the prominent carbon star irc+10216 ( from hereon called cw leo ) .
the aim of the radiative transfer is to obtain the steady - state level populations , of the molecule under study , using the statistical equilibrium equations ( se ) . in the monte carlo method information on the radiation field
is obtained by simulating the line photons using a number of model photons , each representing a large number of real photons from all transitions considered .
these model photons , emitted both locally in the gas as well as injected from the boundaries of the cse , are followed through the cse and the number of absorptions are calculated and stored .
photons are spontaneously emitted in the gas with complete angular and frequency redistribution ( crd ) , i.e. , the local emission is assumed to be isotropic and the scattering are assumed to be incoherent .
the weight of a model photon is continuously modified , to take the absorptions and stimulated emissions into account , as it travels through the cse . when all model photons are absorbed in , or have escaped , the cse the se are solved and the whole process is then repeated until some criterion for convergence is fulfilled . once the molecular excitation , i.e. , the level populations , is obtained the radiative transfer equation can be solved exactly .
the main drawback of the monte carlo method is its slow convergence ( @xmath1@xmath44 ) .
this is , however , usually outweighed by its great adaptability .
one of our motivations for choosing the monte carlo method was to be able to treat varying mass loss rates , departures from spherical symmetry , and the possibility of a highly clumped medium .
these complicating issues will be investigated in a future version of our monte carlo code .
the numerical modelling of line emission from cses , where intricate interplays between physical and chemical processes take place , is a challenge .
this applies , in particular , to the important inner regions of the cse model , where very few observational constraints are available and our knowledge is very limited . in this analysis
, we will neglect many of these complexities and assume a relatively simple , yet reasonably realistic , cse model . by applying the model to a large sample
we will derive fairly reliable mass loss rates , and we will also be able to pick out objects for which there appears to be clear deviations from the simple picture . in what will be referred to as `` the standard model '' we assume a spherically symmetric cse that expands at a constant velocity . in our data
we see no evidence for gas acceleration in the co emitting region . the cse is divided into a number of discrete shells , each with its own set of physical parameters describing the state of the molecular gas .
the density structure ( as a function of distance , @xmath45 , from the central star ) can then be derived from the conservation of mass @xmath46 where @xmath47 is the hydrogen gas mass loss rate , and @xmath48 is the gas expansion velocity taken from the line profile fits of olofsson et al .
( 1993a ) or table [ obs ] .
we assume that in the cses of interest here the hydrogen is in molecular form in the region probed by the co emission ( glassgold & huggins , 1983 ) . the turbulent velocity is assumed to be equal to 0.5kms@xmath5 throughout the entire cse .
the excitation of the co molecules were calculated taking into account 30 rotational levels in each of the ground and first vibrational states .
the radiative transition probabilities and energy levels are taken from chandra et al .
@xcite , and the rotational collisional rate coefficients ( co - h@xmath49 ) are based on the results in flower & launay @xcite .
these are further extrapolated for @xmath14@xmath5011 and for temperatures higher than 250k .
collisional transitions between vibrational levels are not important due to the low densities and their short radiative lifetimes . the kinetic gas temperature is calculated after each iteration in a self - consistent way , using the derived level populations , by solving the energy balance equation ( e.g. , goldreich & scoville 1976 ) , @xmath51 where @xmath52 is the adiabatic index [ assumed to be equal to 5/3 , but the results are not very sensitive to its exact value , see also groenewegen @xcite and ryde et al .
@xcite ] , @xmath53 is the boltzmann constant , @xmath54 is the total heating rate per unit volume , and @xmath55 is the total cooling rate per unit volume .
the first term on the right hand side is the cooling due to the adiabatic expansion of the gas .
additional cooling is provided by molecular line emission from co and h@xmath49 .
the molecular cooling due to co is calculated from the derived level populations using the expression of crosas & menten @xcite . for the h@xmath49 cooling we use the approach by groenewegen @xcite .
hcn could be an important coolant in the inner parts of the envelope ( cernicharo et al .
1996 ) , but groenewegen @xcite demonstrates that hcn cooling is only of minor importance . in the present version of the code
hcn cooling is therefore not included ( in a forthcoming paper , schier & olofsson in prep .
, this issue will be addressed ) .
the mechanism responsible for the observed mass loss is probably ( at least for the stars of interest here ) radiation pressure acting on small dust grains , which in turn are coupled to the gas ( e.g. , hfner 1999 ) .
the radiation pressure on the dust grains will give them a drift velocity , @xmath56 , relative to the gas ( gilman 1972 ; goldreich & scoville 1976 ; kwan & hill 1977 ) @xmath57 where @xmath58 is the luminosity , @xmath59 is the averaged momentum transfer efficiency , and @xmath60 is the speed of light [ see sahai ( 1990 ) for a more elaborate treatment ] . as a result of the dust - gas drift , kinetic energy of the order of @xmath61
will be transfered to the gas each time a particle collides with a dust grain .
this is assumed to provide the dominating heating of the gas ( goldreich & scoville 1976 ; kwan & hill 1977 ) , @xmath62 where @xmath63 is the number density of dust grains , and @xmath64 the geometrical cross section of a dust grain . introducing the dust - to - gas mass ratio @xmath65 this equation
can be written @xmath66 where @xmath67 and @xmath68 are the average size and density of a dust grain , respectively . when solving the energy balance equation free parameters describing the dust , i.e. , via the dust - gas collisional heating ,
are introduced .
these are highly uncertain , but affect the derived line intensities . here
we assume that the @xmath59-parameter , i.e. , the efficiency of momentum transfer , to be equal to 0.03 [ for details see habing et al .
( 1994 ) ] , and define a new parameter that contains the other dust parameters , @xmath69 where the normalized values are the ones we used to fit the co line emission of using our model , i.e. , @xmath70@xmath131 for this object .
additional heating is provided by the photoelectric effect ( huggins et al .
1988 ) @xmath71 where @xmath72 is a constant that depends on the properties of the dust grains . following huggins
@xcite we adopt @xmath72@xmath131@xmath210@xmath73ergs@xmath5 .
this heating mechanism is important in the cool , tenuous , outer parts of cses around high mass loss rate stars .
the spatial extent of the molecular envelope is generally an important input parameter , and the derived mass loss rate will depend on this .
the radial distribution of co in the cse was estimated using the modelling presented in mamon et al .
it includes photodissociation , taking into account the effects of dust- , self- and h@xmath74-shielding , and chemical exchange reactions .
mamon et al .
@xcite find that the radial abundance distribution of co with respect to h@xmath74 , @xmath75 , can be described by @xmath76,\ ] ] where @xmath77 is the initial ( photospheric ) abundance , @xmath78 is the photodissociation radius ( where the abundance has dropped to @xmath79 ) , and @xmath80 is a parameter describing the rate at which the abundance declines . both @xmath78 and @xmath80
depend on the mass loss rate ( @xmath47 ) , the expansion velocity ( @xmath38 ) , and @xmath77 . in our modelling
we use analytical fits to these results from stanek et al .
@xcite @xmath81 and kwan & webster @xcite @xmath82 where the units of @xmath47 and @xmath38 are @xmath83 and @xmath84 , respectively .
in what follows we will assume @xmath77=1@xmath210@xmath85 .
this is an average of the @xmath77:s estimated by olofsson et al .
@xcite for this sample of optically bright carbon stars .
the scatter in the estimated @xmath77:s is about 40% .
we note here that a somewhat more sophisticated photodissociation model was developed by doty & leung @xcite in which they include scattering by dust . for the cse of
they derive a co envelope size which is 30% smaller than what is obtained using the mamon et al . model .
it is however not clear by how much the envelope size changes for an object with a significantly lower mass loss rate , and hence we use the results of mamon et al . here . in our models
we include both a central source of radiation and the cosmic microwave background radiation at 2.7k .
the central radiation emanates from the star , and was estimated from a fit to its spectral energy distribution ( sed ) , usually by assuming two blackbodies .
this method is described in kerschbaum @xcite .
a fit to the sed gives the two blackbody temperatures , @xmath86 and @xmath87 , and the relative luminosities of the two blackbodies , @xmath88/@xmath89 .
any dust present around the star will absorb parts of the stellar light and re - emit it at longer wavelenghts .
thus , of the two blackbodies used , one represents the stellar contribution and one represents the dust .
however , it should be noted that , e.g. , the stellar temperature @xmath86 derived in this manner is generally lower than the true temperature of the star ( kerschbaum 1999 ) . in any case , the two blackbodies used give a good description of the radiation that the cse is subjected to .
the temperatures and luminosities used in the modelling are presented in table [ coresult ] . only in the cases where
@xmath88/@[email protected] the dust component was retained .
the inner boundary of the cse , @xmath90 , was set to reside outside the radius of the central blackbody(s ) , but never lower than 1@xmath210@xmath91 cm ( @xmath13r@xmath4 ) , i.e. , generally beyond both the sonic point and dust condensation radius .
the distances , presented in table [ coresult ] , were estimated using one of the following methods : the observed hipparcos parallax , a period - luminosity relation ( groenewegen & whitelock 1996 ) , or an assumed bolometric luminosity . in the two former cases the luminosities were estimated using apparent bolometric magnitudes and the distances . in the rare cases when both the distance obtained from hipparcos and the period - luminosity relation result in extremely high or low luminosities , a value of 4000l@xmath4 was adopted as the luminosity and the distance
was estimated using this value .
for the irregular variables with no hipparcos data we also assumed a luminosity of 4000l@xmath4 .
for a statistical study of a large sample of stars these distance estimates are adequate , although the distance estimate for an individual star has a large uncertainty .
there is no apparent systematic difference between the distances derived from the period - luminosity relation and the hipparcos parallaxes , fig .
the distances in olofsson et al.(1993a ) were estimated from an adopted absolute @xmath12 magnitude .
the distances presented here are systematically lower by , on the average , a factor of 1.4 than those used in olofsson et al .
this was a summary of the assumptions used in `` the standard model '' . in
what follows , tests will be made to see how sensitive the model is to the various assumptions .
llcccccccccc & & & & & & & & & & & + & & & & & & & & & & & + & sra & 369 & 560@xmath20 & 5500 & 2000 & & & 4.0@xmath210@xmath9 & 11.5 & 1.4@xmath210@xmath92 & 0.2@xmath93 + & m & 527 & 970@xmath20 & 7900 & 1800 & 900 & 0.54 & 9.0@xmath210@xmath7 & 18.5 & 6.1@xmath210@xmath92 & 0.5@xmath93 + & & & 670@xmath94 & 4000 & 1300 & 500 & 0.13 & 3.0@xmath210@xmath95 & 28.0 & 1.0@xmath210@xmath96 & 0.2@xmath93 + & srb & 144 & 320@xmath19 & 3100 & 2600 & & & 2.5@xmath210@xmath9 & 3.5 & 1.4@xmath210@xmath92 & 0.2@xmath93 + & m & 389 & 610@xmath20 & 5800 & 1500 & 700 & 0.52 & 1.3@xmath210@xmath95 & 16.5 & 8.0@xmath210@xmath92 & 0.7 + & srb & 158 & 400@xmath19 & 6700 & 2600 & & & 9.0@xmath210@xmath9 & 5.5 & 2.5@xmath210@xmath92 & 0.5 + & m & 535 & 560@xmath20 & 8100 & 1900 & 900 & 10 & 3.5@xmath210@xmath95 & 15.0 & 1.4@xmath210@xmath96 & 0.4 + & sr & & 450@xmath94 & 4000 & 2200 & & & 1.3@xmath210@xmath7 & 9.0 & 2.6@xmath210@xmath92 & 0.2@xmath93 + & srb & 300 & 330@xmath20 & 4400 & 1600 & & & 9.0@xmath210@xmath9 & 9.0 & 2.1@xmath210@xmath92 & 0.2@xmath93 + & srb & 167 & 360@xmath20 & 2400 & 2400 & & & 7.0@xmath210@xmath9 & 5.0 & 2.2@xmath210@xmath92 & 0.2@xmath93 + & m & 467 & 250@xmath19 & 4000 & 1900 & 800 & 0.14 & 7.0@xmath210@xmath7 & 18.0 & 5.3@xmath210@xmath92 & 0.2 + & srb & 212 & 220@xmath19 & 2600 & 2200 & & & 7.0@xmath210@xmath9 & 11.0 & 1.8@xmath210@xmath92 & 0.3 + & sr & 590 & 820@xmath20 & 8900 & 1600 & 700 & 0.32 & 2.2@xmath210@xmath95 & 25.5 & 8.9@xmath210@xmath92 & 0.5@xmath93 + & lb & & 490@xmath94 & 4000 & 2000 & & & 3.0@xmath210@xmath7 & 16.0 & 3.5@xmath210@xmath92 & 0.4 + & srb & 242 & 310@xmath20 & 3500 & 2400 & & & 4.0@xmath210@xmath7 & 11.0 & 4.5@xmath210@xmath92 & 0.1 + & srb & 230 & 330@xmath20 & 3400 & 2400 & & & 2.5@xmath210@xmath7 & 11.5 & 3.4@xmath210@xmath92 & 0.2@xmath93 + & lb & & 360@xmath94 & 4000 & 2700 & & & 1.1@xmath210@xmath7 & 9.0 & 2.4@xmath210@xmath92 & 0.2@xmath93 + & srb & 234 & 260@xmath20 & 6900 & 2500 & & & 3.5@xmath210@xmath7 & 11.0 & 4.2@xmath210@xmath92 & 0.1 + & lb & & 420@xmath19 & 3500 & 2700 & & & 6.5@xmath210@xmath9 & 9.5 & 1.8@xmath210@xmath92 & 0.2@xmath93 + & m & 497 & 770@xmath20 & 7500 & 2000 & 1100&1.4 & 2.2@xmath210@xmath95 & 25.0 & 8.9@xmath210@xmath92 & 0.5@xmath93 + & sra & 456 & 456@xmath19 & 3000 & 2000 & & & 3.5@xmath210@xmath7 & 11.0 & 4.2@xmath210@xmath92 & 0.2@xmath93 + & lb & & 450@xmath94 & 4000 & 2500 & & & 3.0@xmath210@xmath7 & 10.5 & 3.9@xmath210@xmath92 & 0.2@xmath93 + & m & 454 & 730@xmath20 & 6800 & 1300 & 800 & 0.7 & 1.8@xmath210@xmath95 & 18.0 & 9.0@xmath210@xmath92 & 1.0 + & srb & 195 & 280@xmath20 & 2800 & 2200 & & & 1.0@xmath210@xmath7 & 7.0 & 2.4@xmath210@xmath92 & 0.1 + & m & 630 & 120@xmath20 & 9600 & & 510 & & 1.5@xmath210@xmath6 & 14.5 & 3.7@xmath210@xmath96 & 1.0 + & srb & 303 & 350@xmath19 & 4200 & 2600 & & & 1.9@xmath210@xmath7 & 9.0 & 3.2@xmath210@xmath92 & 0.2@xmath93 + & sr & 140 & 310@xmath20 & 2800 & 2200 & & & 1.8@xmath210@xmath7 & 10.0 & 3.0@xmath210@xmath92 & 0.2@xmath93 + & srb & 126 & 580@xmath94 & 4000 & 2400 & & & 2.0@xmath210@xmath95 & 14.0 & 1.1@xmath210@xmath96 & 0.2 + & sra & 640 & 440@xmath20 & 9700 & 1300 & 510 & 6.7 & 6.0@xmath210@xmath95 & 17.0 & 1.9@xmath210@xmath96 & 1.4 + & & & 530@xmath94 & 4000 & 1900 & & & 6.0@xmath210@xmath7 & 14.0 & 5.2@xmath210@xmath92 & 0.2@xmath93 + & m & 442 & 960@xmath20 & 6600 & 1800 & & & 9.0@xmath210@xmath7 & 12.0 & 7.0@xmath210@xmath92 & 0.5@xmath93 + & & & 480@xmath94 & 4000 & 2000 & & & 6.5@xmath210@xmath7 & 17.0 & 5.2@xmath210@xmath92 & 0.2@xmath93 + & srb & 450 & 160@xmath19 & 2500 & 2400 & & & 1.4@xmath210@xmath7 & 7.0 & 2.9@xmath210@xmath92 & 0.2@xmath93 + & lb & & 330@xmath94 & 4000 & 2700 & & & 7.0@xmath210@xmath9 & 6.0 & 2.1@xmath210@xmath92 & 0.2@xmath93 + & sra & 364 & 540@xmath20 & 5400 & 1800 & & & 2.0@xmath210@xmath7 & 12.5 & 3.0@xmath210@xmath92 & 0.2@xmath93 + & srb & 157 & 220@xmath19 & 4400 & 2200 & & & 1.5@xmath210@xmath7 & 8.5 & 2.9@xmath210@xmath92 & 0.2@xmath97 + & srb : & 200 & 490@xmath19 & 10000 & 2500 & & & 3.0@xmath210@xmath7 & 10.0 & 4.0@xmath210@xmath92 & 1.2@xmath97 + & & & 530@xmath94 & 4000 & 2400 & & & 1.0@xmath210@xmath7 & 6.5 & 2.5@xmath210@xmath92 & 0.2@xmath93 + & & & 390@xmath94 & 4000 & 2300 & & & 1.0@xmath210@xmath7 & 11.0 & 2.2@xmath210@xmath92 & 0.2@xmath93 + & lb & & 460@xmath94 & 4000 & 2200 & & & 1.3@xmath210@xmath7 & 8.0 & 2.7@xmath210@xmath92 & 0.4 + & m & 358 & 630@xmath20 & 5300 & 1900 & 740 & 0.15 & 6.0@xmath210@xmath7 & 7.5 & 6.8@xmath210@xmath92 & 0.2 + & srb & 185 & 280@xmath19 & 2700 & 2000 & & & 5.0@xmath210@xmath9 & 7.5 & 1.6@xmath210@xmath92 & 0.2@xmath93 + & m & 422 & 610@xmath20 & 6300 & 1600 & 650 & 0.28 & 1.2@xmath210@xmath95 & 13.5 & 8.0@xmath210@xmath92 & 0.5@xmath93 + & lb & & 340@xmath94 & 4000 & 2000 & & & 7.0@xmath210@xmath9 & 11.5 & 1.8@xmath210@xmath92 & 0.5@xmath97 + & srb & 353 & 370@xmath19 & 6500 & 2100 & & & 3.0@xmath210@xmath7 & 8.5 & 4.2@xmath210@xmath92 & 0.2 + & lb & & 430@xmath94 & 4000 & 2600 & & & 1.6@xmath210@xmath7 & 10.0 & 2.8@xmath210@xmath92 & 0.2@xmath93 + & sra : & 168 & 310@xmath94 & 4000 & 2600 & & & 1.6@xmath210@xmath7 & 4.0 & 4.0@xmath210@xmath92 & 0.2@xmath93 + & srb & 200 & 420@xmath94 & 4000 & 2700 & & & 2.5@xmath210@xmath7 & 10.0 & 3.6@xmath210@xmath92 & 0.2@xmath93 + & srb & 393 & 450@xmath20 & 5900 & 2200 & & & 1.0@xmath210@xmath7 & 8.0 & 2.3@xmath210@xmath92 & 0.2@xmath93 + & m & 463 & 710@xmath20 & 6900 & 2200 & & & 9.0@xmath210@xmath7 & 13.0 & 6.8@xmath210@xmath92 & 0.5@xmath93 + & m & 421 & 370@xmath20 & 6200 & 1500 & 860 & 0.84 & 1.2@xmath210@xmath95 & 11.5 & 8.5@xmath210@xmath92 & 1.2 + & m & 454 & 670@xmath20 & 6800 & 1300 & 620 & 0.46 & 2.5@xmath210@xmath95 & 16.0 & 1.1@xmath210@xmath96 & 0.5 + & srb & 320 & 570@xmath19 & 8800 & 2600 & & & 9.0@xmath210@xmath9 & 6.0 & 2.4@xmath210@xmath92 & 0.5@xmath93 + & srb & 233 & 360@xmath20 & 4100 & 2400 & & & 1.6@xmath210@xmath7 & 8.0 & 3.0@xmath210@xmath92 & 0.2@xmath93 + & m & 470 & 780@xmath20 & 7100 & 1200 & 580 & 0.5 & 1.0@xmath210@xmath6 & 14.0 & 2.9@xmath210@xmath96 & 0.3 + & m & 487 & 340@xmath20 & 7300 & 1400 & & & 1.5@xmath210@xmath95 & 22.0 & 7.5@xmath210@xmath92 & 0.4 + [ coresult ] llcccccccccc & & & & & & & & & & & + & & & & & & & & & & & + & srb & 180 & 230@xmath20 & 2600 & 2800 & & & 1.8@xmath210@xmath7 & 10.0 & 3.0@xmath210@xmath92 & 0.2@xmath93 + & srb & 263 & 310@xmath20 & 3400 & 2100 & & & 4.5@xmath210@xmath7 & 13.5 & 4.5@xmath210@xmath92 & 0.2@xmath93 + & m & & 390@xmath20 & 4000 & 1700 & 840 & 0.13 & 1.4@xmath210@xmath95 & 19.5 & 7.6@xmath210@xmath92 & 0.2@xmath93 + & m & 620 & 630@xmath20 & 9400 & 1100 & 610 & 6.6 & 1.5@xmath210@xmath6 & 14.0 & 4.0@xmath210@xmath96 & 0.7 + & srb & 186 & 290@xmath20 & 2700 & 2500 & & & 6.5@xmath210@xmath3 & 2.5 & 7.4@xmath210@xmath98 & 0.2@xmath93 + [ cols="<,<,<,^,^,^,<,<,<,^,^,^ " , ] lllccc|lllccc & & & & & & & & & & & + & & & & & & & & & & & + & oso & 1@xmath80 & 4.5 & 5.4 & 2 & & oso & 1@xmath80 & 27.5 & 27.2 & 2 + & iram & 1@xmath80 & 10.3 & 11.2 & 2 & & nrao & 2@xmath81 & 50.1 & 36.5 & 1 + & jcmt & 2@xmath81 & 11.9 & 10.9 & 4 & & jcmt & 2@xmath81 & 35.9 & 56.2 & 4 + & iram & 2@xmath81 & 22.7 & 34.2 & 2 & & jcmt & 3@xmath82 & 88.9 & 86.1 & 4 + & jcmt & 3@xmath82 & 20.9 & 15.6 & 4 & & jcmt & 4@xmath83 & 123.4 & 95.2 & 4 + & oso & 1@xmath80 & 2.4 & 3.0 & 2 & & sest & 1@xmath80 & 7.5 & 6.9 & 2 + & iram & 2@xmath81 & 21.6 & 28.3 & 2 & & sest & 2@xmath81 & 18.1 & 17.2 & 2 + & jcmt & 3@xmath82 & 18.9 & 15.8 & 4 & &
sest & 3@xmath82 & 18.6 & 20.5 & 1 + & sest & 1@xmath80 & 1.5 & 1.4 & 2 & & sest & 1@xmath80 & 0.5 & 0.4 & 2 + & sest & 1@xmath80 & @xmath410.7 & 0.6 & 2 & & sest & 2@xmath81 & 1.5 & 1.9 & 2 + & sest & 2@xmath81 & 2.1 & 1.9 & 2 & & sest & 1@xmath80 & 1.6 & 1.4 & 2 + & sest & 1@xmath80 & 2.5 & 2.7 & 2 & & sest & 2@xmath81 & 3.8 & 4.3 & 2 + & sest & 2@xmath81 & 11.8 & 10.2 & 2 & & oso & 1@xmath80 & 17.3 & 17.0 & 2 + & sest & 3@xmath82 & 15.3 & 15.1 & 1 & & sest & 2@xmath81 & 8.7 & 14.3 & 2 + & oso & 1@xmath80 & 2.7 & 3.9 & 2 & & sest & 3@xmath82 & 17.8 & 17.1 & 1 + & iram & 1@xmath80 & 11.4 & 8.4 & 3 & & oso & 1@xmath80 & 16.3 & 19.4 & 2 + & iram & 2@xmath81 & 18.4 & 18.8 & 2 & & iram & 1@xmath80 & 48.5 & 38.1 & 3 + & sest & 1@xmath80 & 1.1 & 0.9 & 2 & & jcmt & 2@xmath81 & 49.3 & 37.3 & 4 + & iram & 2@xmath81 & 8.7 & 10.2 & 2 & & iram & 2@xmath81 & 76.6 & 109.6 & 2 + & oso & 1@xmath80 & 8.8 & 8.4 & 2 & & jcmt & 3@xmath82 & 55.7 & 55.4 & 4 + & iram & 1@xmath80 & 28.6 & 17.7 & 3 & & jcmt & 4@xmath83 & 65.7 & 60.0 & 4 + & iram & 2@xmath81 & 43.7 & 48.7 & 2 & & oso & 1@xmath80 & 4.7 & 4.8 & 2 + & oso & 1@xmath80 & 0.7 & 1.0 & 2 & & iram & 2@xmath81 & 8.5 & 28.4 & 2 + & iram & 2@xmath81 & 5.6 & 9.6 & 2 & & oso & 1@xmath80 & 4.8 & 6.1 & 2 + & jcmt & 3@xmath82 & 6.4 & 4.3 & 4 & & iram & 1@xmath80 & 16.7 & 12.9 & 2 + &
sest & 1@xmath80 & 2.8 & 2.6 & 2 & & iram & 2@xmath81 & 14.4 & 33.5 & 2 + & oso & 1@xmath80 & 3.2 & 4.8 & 2 & & oso & 1@xmath80 & 6.4 & 6.7 & 2 + & sest & 2@xmath81 & 8.1 & 7.0 & 2 & & nrao & 2@xmath81 & 8.7 & 7.4 & 1 + & jcmt & 2@xmath81 & 9.0 & 8.3 & 4 & & oso & 1@xmath80 & 57.0 & 58.8 & 1 + & jcmt & 3@xmath82 & 11.2 & 10.6 & 4 & & iram & 1@xmath80 & 92.7 & 97.4 & 3 + & sest & 1@xmath80 & 0.8 & 0.8 & 2 & & nrao & 2@xmath81 & 60.7 & 46.5 & 1 + & oso & 1@xmath80 & 2.1 & 1.7 & 2 & & jcmt & 2@xmath81 & 90.6 & 68.5 & 4 + & iram & 2@xmath81 & 6.8 & 9.0 & 2 & & iram & 2@xmath81 & 155.8 & 166.8 & 3 + & sest & 1@xmath80 & 1.4 & 1.3 & 2 & & jcmt & 3@xmath82 & 88.0 & 79.1 & 4 + & sest & 2@xmath81 & 4.1 & 4.0 & 2 & & jcmt & 4@xmath83 & 73.3 & 78.6 & 4 + & sest & 1@xmath80 & 0.6 & 0.6 & 2 & & oso & 1@xmath80 & @xmath411.1 & 0.1 & 2 + & sest & 2@xmath81 & 2.0 & 2.1 & 2 & & iram & 1@xmath80 & @xmath412.6 & 0.3 & 2 + & oso & 1@xmath80 & 4.9 & 4.8 & 2 & & iram & 2@xmath81 & 2.6 & 2.8 & 2 + & iram & 2@xmath81 & 9.4 & 31.7 & 2 +
with the assumptions made in the standard model there remains two principal free parameters , the mass loss rate ( @xmath47 ) and the @xmath70-parameter .
these two parameters were allowed to vary simultaneously , @xmath47 in steps of @xmath110% and @xmath70 in somewhat larger steps , until the model with the smallest deviations from the observed intensities was found .
the quality of a particular model with respect to the observational constraints can be quantified using the chi - square statistic defined as @xmath99 ^ 2}{\sigma^2_{i}},\ ] ] where @xmath100 is the total integrated line intensity , @xmath101 is the uncertainty in observation @xmath102 , and the summation is done over all independent observations @xmath103 .
the errors in the observed intensities are generally dominated by the calibation uncertainty of @xmath120% . in some cases
the line profiles were used to discriminate between models . for each new
@xmath47 assigned , a consistent envelope size ( @xmath78 ) was calculated based upon the results of mamon et al .
@xcite .
it is found that the intensity ratios between the different lines are sensitive to various parameters , and hence can be used to constrain them ( sec .
[ mass - loss ] ) .
therefore , in order to accurately determine @xmath70 , and ultimately @xmath47 , multi - transition observations are needed .
observed radial brightness profiles provide additional important information . for a limited number of the sources in our sample
there are enough observational constraints that a reliable @xmath70-parameter may be estimated .
these sources were first modelled , see table [ coresult ] .
the derived @xmath70-parameters , as a function of the adopted luminosity of the central source , are shown in fig .
[ h - para ] . a division between low luminosity objects and those with higher luminosities is evident . for luminosities lower than 6000l@xmath4
the median value of the estimated @xmath70-parameter is 0.2 ( based on 9 sources ) , while for the higher luminosities it is 0.5 ( 12 sources ) ( the j - type stars , i.e. , stars with a high content of @xmath104co , were not included in these estimates since @xmath104co line cooling is not treated in the standard model ) .
the large scatter for the high luminosity sources is unfortunate , since these are generally the sources where the derived mass loss rates are particularly sensitive to the adopted value of @xmath70 . for the majority of objects , where a reliable estimate of the @xmath70-parameter is not possible ,
we have adopted a value of @[email protected] for the low luminosity objects ( below 6000l@xmath4 ) and @[email protected] for the high luminosity objects ( above 6000l@xmath4 ) .
the @xmath70-parameter is related to the dust - to - gas mass ratio , @xmath65 ( see eq .
if we assume that the properties of a dust grain are equal to those used to fit the line emission from , and that they are the same for all stars , an @xmath70-parameter of 0.2 means that @xmath65@xmath132@xmath210@xmath85 .
groenewegen et al .
@xcite have modelled the dust emission towards a large sample of carbon stars and find a value of @xmath65 which is fairly constant at 2.5@xmath210@xmath85 for stars with luminosities up to about 7900l@xmath4 , and which increases drastically with luminosity up to 0.01 and higher .
this is in good agreement with our results and lends further support to our distinction between low- and high - luminosity stars as regards the value of the @xmath70-parameter .
furthermore , hiriart & kwan @xcite derive , for 17 of our sample stars , on the average @xmath65@xmath17@xmath210@xmath10 .
the main reason for this discrepancy is the difference in the adopted dust parameters .
in fact , it can be shown that for @[email protected] we get essentially the same heating term ( cf .
eq . [ h_dg ] ) in the energy balance equation as hiriart & kwan @xcite
. ccccccccccccccccc & & & & & & & + & & & & & & & & & & & & & & & & + @xmath105 & @xmath3050% & & @xmath3025 & @xmath3015 & @xmath3015 & @xmath3010 & & @xmath3050 & @xmath3055 & @xmath3055 & @xmath3055 & & @xmath810 & @xmath3015 & @xmath3040 & @xmath3055 + & @xmath833% & & @xmath825 & @xmath815 & @xmath815 & @xmath815 & & @xmath835 & @xmath835 & @xmath840 & @xmath840 & & @xmath840 & @xmath830 & @xmath825 & @xmath835 + @xmath47 & @xmath3050% & & @xmath3015 & 0 & 0 & @xmath305 & & @xmath3055 & @xmath3040 & @xmath3035 & @xmath3030 & & @xmath3025 & @xmath3030 & @xmath3040 & @xmath3045 + & @xmath833% & & @xmath815 & @xmath85 & @xmath85 & @xmath85 & & @xmath835 & @xmath830 & @xmath830 & @xmath825 & & @xmath840 & @xmath825 & @xmath820 & @xmath830 + @xmath106 & @xmath3050% & & 0 & 0 & 0 & 0 & & 0 & 0 & 0 & 0 & & @xmath3015 & @xmath3010 & @xmath305 & @xmath305 + & @xmath833% & & 0 & 0 & 0 & 0 & & 0 & @xmath85 & 0 & 0 & & @xmath820 & @xmath810 & @xmath85 & @xmath85 + @xmath70 & @xmath3050% & & @xmath3025 & @xmath3035 & @xmath3035 & @xmath3040 & & @xmath3010 & @xmath3015 & @xmath3020 & @xmath3025 & & 0 & @xmath305 & @xmath3010 & @xmath3015 + & @xmath833% & & @xmath820 & @xmath825 & @xmath830 & @xmath830 & & @xmath810 & @xmath815 & @xmath820 & @xmath820 & & 0 & @xmath810 & @xmath810 & @xmath815 + @xmath78 & @xmath3050% & & @xmath305 & 0 & 0 & 0 & & @xmath3015 & @xmath305 & @xmath305 & 0 & & @xmath3020 & @xmath305 & @xmath305 & 0 + & @xmath833% & & @xmath85 & 0 & 0 & 0 & & @xmath820 & @xmath85 & 0 & 0 & & @xmath850 & @xmath830 & @xmath810 & @xmath85 + @xmath107 & @xmath22 & & @xmath305 & 0 & 0 & 0 & & @xmath305&@xmath305 & @xmath305&0 & & @xmath305 & @xmath3015 & @xmath3015 & @xmath3010 + @xmath108 & @xmath22 & & @xmath3010 & 0 & 0 & 0 & & @xmath3015 & @xmath3010 & @xmath305 & @xmath305 & & 0 & @xmath3025 & @xmath3040 & @xmath3030 + [ error ] we have managed to model reasonably well 61 of the 69 sample sources using our standard model and fitting strategy defined above . the reduced @xmath109 for these models is estimated from @xmath110 where @xmath111 is obtained from eq .
[ chi2_eq ] for the best fit model , and @xmath112 is the number of adjustable parameters ( one or two ) in the model .
typically , @xmath113@xmath11@xmath83 assuming a calibration uncertainty of 20% .
the derived mass loss rates and @xmath70-parameters are presented in table [ coresult ] , where also the velocities by which the cses expand , @xmath38 , and their spatial extents , @xmath78 , are given .
the derived mass loss rates presented in table [ coresult ] span more than three orders of magnitude , 6.5@xmath210@xmath3m@xmath4yr@xmath5 to 1.5@xmath210@xmath6m@xmath4yr@xmath5 , over which the physical conditions of the cses vary considerably , posing a challenge to the model .
the intensities , overall line shapes , and when available radial brightness distributions , of the circumstellar co lines produced by the radiative transfer model generally agree well with those observed .
this is illustrated here for three of our sample stars ( fig .
[ lpand ] ) , ( fig .
[ rfor ] ) , and ( fig . [ wori ] ) .
these carbon stars span a large range in mass loss rate and serve to illustrate the various physical conditions in these cses .
a full , detailed , error analysis of the estimated mass loss rates is not possible due to the relatively large number of free parameters entering the model .
instead , we will vary some of the more important parameters in order to illustrate the sensitivity of the model , and to be able to get a rough estimate of the errors involved in the mass loss rate estimates .
the results of these sensitivity tests are shown in table [ error ] for our three example stars .
in addition , in fig .
[ chi2 ] we present chi - square contour plots for these stars , produced by varying the mass loss rate and the @xmath70-parameter .
this will give an estimate of the accuracy in the determination of these two adjustable parameters when all other parameters are held fixed ( the size and shape of the co envelope is however allowed to change in accordance with eqs [ rout ] and [ alpha ] ) . in fig .
[ chi2 ] we indicate various confidence levels with the contour at @[email protected] marking the 68% confidence limit ( i.e. , the `` 1@xmath114 '' level for two adjustable parameters ) .
the information contained in the shape of the line profiles ( to which the @xmath109 defined above is not very sensitive ) have occasionally been used to further constrain the most probable parameter space .
is a high mass loss rate mira variable where the excitation of @xmath11co is dominated by collisions .
consequently , the line intensities are very sensitive to the temperature structure , i.e. , the @xmath70-parameter .
it is interesting to note that the assumed envelope size @xmath78 does not significanly affect the derived mass loss rate in the high mass loss rate regime .
the reason for this being that the density is too low to excite the co molecules effectively ( the lines are sub - thermally excited ) in the cool outer parts of the envelope ( cf . fig . [ lpand ] ) , i.e. , the emission is excitation limited . in the high mass loss rate regime the line intensities are also insensitive to the adopted mass loss rate since an increase in @xmath47 leads to more cooling ( @xmath55@xmath115@xmath47 while @xmath54@xmath115@xmath116 ) , which compensates for the increase of molecular density .
this `` saturation''-effect of the line intensities for high mass loss rates has been noted before in other models where the cooling by co is treated in a self - consistent manner ( sahai 1990 ; kastner 1992 ) .
additional sensitivity tests in the case of a high mass loss rate object ( ) can be found in groenewegen ( 1994 ) .
, on the other hand , is a low mass loss rate object with low optical depths in the lowest rotational transitions .
in fact , some of the low lying transitions are inverted over parts of the cse , i.e. , maser action is taking place .
in particular , the @xmath14@xmath13@xmath15@xmath16@xmath17 is inverted over the entire cse producing a line profile that is clearly double peaked , i.e. , this is due to maser action and not due to resolution effects .
the observations of this transition are complicated by possible contamination by interstellar lines , and the signal - to - noise ratio is not particularly good , erasing any possible signs of maser action . as expected the radiation emitted by the central star plays an important role in the excitation .
the transitions exhibiting population inversion are particularly sensitive ( both the intensity and line profile ) to the physical conditions prevailing in the cse . in this case where radiative excitation is important
the model becomes sensitive to the choice of inner radius , @xmath107 , as well as tubulent velocity , @xmath108 . here ,
also the size of the envelope is important when determining the mass loss rate , i.e. , the emission is photodissociation limited ( at least for the lowest transitions ) .
generally , the line intensities scale roughly linearly with @xmath47 . presents an intermediate case between and
it is also a typical sample star in the sense that the majority of the stars have envelope parameters that resemble those of . from table
[ error ] and fig .
[ chi2 ] we see that shares properties of both the low and the high mass loss rate star .
an intermediate mass loss rate model is generally more sensitive to the temperature structure than the radiation field , thus resembling the high mass loss rate objects .
however , as in the low mass loss rate regime the line intensities scale roughly linearly with @xmath47 .
we also note that the derived line intensities , especially the @xmath14@xmath13@xmath15@xmath16@xmath17 transition , start to become sensitive to the outer radius of the co envelope
. the exact mass loss rate where the transition from photodissociation- to excitation - limited emission occurs is hard to pinpoint , since it depends on the envelope characteristics .
we have made tests which show that for typical envelope parameters the transition lies at @xmath15@xmath210@xmath7m@xmath117yr@xmath5 for the @xmath14@xmath13@xmath15@xmath16@xmath17 transition .
higher @xmath14-lines sample hotter and denser gas , located closer to the star , and are therefore less sensitive to the choice of the envelope size . in summary
, we believe that within the adopted circumstellar model , the estimated mass loss rates are accurate to about @xmath3650% ( neglecting errors introduced by the uncertain co abundance and the distance estimates ) when good observational contraints are available , but one should keep in mind that the causes of the uncertainty varies with the mass loss rate . another rough way to estimate the errors involved is to compare with the results obtained from other self - consistent models . when comparing our mass loss rate estimate for to those obtained from detailed radiative transfer models ( kastner 1992 ; crosas & menten 1997 ; groenewegen et al .
1998a ; skinner et al .
1999 ) we find a very good agreement , within 20% , when adjustments for differences in @xmath77 and distance have been made .
kastner @xcite also modelled the high mass loss rate object ( a.k.a . ) and obtained a mass loss rate in excellent agreement with our estimate ( when corrected for the difference in distance ) .
recently , hiriart & kwan @xcite presented a method to combine a circumstellar dust model with a co model , keeping physical quantities consistent between the models .
they present mass loss rate estimates for 17 of our sample stars .
based upon their eq .
17 and table 6 we derive hydrogen mass loss rates for these stars that are , on the average , 1.3 times larger than those derived from our radiative transfer analysis ( for an individual star , however , the discrepancy can be as large as a factor of 3 ) .
we also compare our derived mass loss rates with those obtained by olofsson et al .
( 1993a ) using a formula based on the work by knapp & morris @xcite , @xmath118 which relates the mass loss rate @xmath119 to easily determined observables ( @xmath27 is given in kelvins , @xmath38 in kms@xmath5 , @xmath120 in pc , and @xmath23 in arcseconds ) . in eq .
[ mdot_simple ] @xmath121 is a factor that depends on the transition ( @xmath14 ) used . for @xmath14@xmath13@xmath15@xmath16@xmath17 @xmath121@xmath131 and for @xmath14@xmath13@xmath18@xmath16@xmath15 @[email protected]
( see the discussion in olofsson et al.1993a ) . when estimating the mass loss rates from eq . [ mdot_simple ]
we have used the observed @xmath14@xmath13@xmath15@xmath16@xmath17 and @xmath14@xmath13@xmath18@xmath16@xmath15 line emission , and averaged the results .
this mass loss rate was compared to that obtained from the radiative transfer analysis , fig .
[ mdot - ratio ] .
it is found that eq .
[ mdot_simple ] significantly underestimates the mass loss rates when compared to those obtained from the radiative transfer analysis by , on the average , a factor of about four .
however , considering the simplicity of eq .
[ mdot_simple ] the agreement is remarkably good .
there is also a trend that the discrepancy increases with lower mass loss rate ( indeed olofsson et al .
suspected that their simple approach led to a systematic underestimate of the mass loss rate that became worse the lower the mass loss rate ) . in this context
it should be noted that eq .
[ mdot_simple ] was derived assuming a fixed co envelope size of 3@xmath210@xmath96 cm , which is appropriate for high mass loss rate objects . in the high mass loss rate regime
the line intensities are not very sensitive to the exact envelope size since the emission is excitation limited .
this is , however , generally not the case for the optically bright carbon stars in our sample , where the line intensities are sensitive to the size of the co envelope set by photodissociation .
a correction for the envelope size [ see neri et al ( 1998 ) and references therein ] would decrease the discrepancy seen in fig .
[ mdot - ratio ] for the stars losing mass at lower rates .
taken at face values the discrepancies between the mass loss rates derived here and those derived by olofsson et al .
( 1993a ) are smaller , but this is due to the fact that on the average olofsson et al .
used larger distances .
this also means that the trend in the dust - to - gas mass ratio , decreasing with the mass loss rate , that olofsson et al .
reported was due to underestimated gas mass loss rates [ see also hiriart & kwan ( 2000 ) ] . indeed , in sec .
[ h - parameter ] we find the opposite behaviour , i.e. , indications of a dust - to - gas mass ratio that increases with the mass loss rate .
the size of the co envelope is an important parameter , and the derived mass loss rate will depend on it to an extent that depends on the mass loss rate ( see discussion in sec .
[ mass - loss ] ) .
neri et al .
@xcite presented iram pdb interferometer ( pdbi ) co(@xmath14@xmath13@xmath15@xmath16@xmath17 ) maps of varying quality for 9 of our sample stars . from their cleaned brightness maps
we have obtained the averaged radial brightness distributions .
in addition , we have obtained two single - dish co(@xmath14@xmath13@xmath15@xmath16@xmath17 ) maps using the oso 20 m telescope . in fig .
[ radial ] we compare the observed radial brightness distributions to those calculated from our model , which uses the mamon et al .
@xcite results to estimate the co envelope size , for the parameters presented in tab .
[ coresult ] .
we find that the estimated co envelope sizes are generally consistent with the observations . in four cases ( , , , and )
the calculated radial brightness distributions fail to reproduce the observations . at least in the case of and
this appears to be due to deviations from a simple @xmath0 density law , in the sense that two distinct components are clearly visible in the brightness maps .
we are not able to distinguish between the different co envelope sizes obtained by mamon et al .
@xcite and doty & leung @xcite for the high mass loss rate object .
this is simply because the emission in this case is excitation limited with the external cool and tenuous parts producing very little emission . as was demonstrated in sec .
[ mass - loss ] the derived mass loss rate can be very sensitive to the kinetic temperature structure of the gas . adiabatic cooling alone results in @xmath122@xmath115@xmath123 , where @xmath124@xmath13@xmath84/3 ( assuming @xmath52@xmath135/3 ) , but from figs .
[ lpand]@xmath8[wori ] it is evident that other cooling processes as well as heating mechanisms are important .
if co line radiation is the dominant cooling mechanism and @xmath125@xmath50@xmath126 one would expect @xmath122@xmath115@xmath127 with a somewhat shallower decline if @xmath125@xmath41@xmath126 ( see doty & leung 1997 ) . generally , it is not possible to assign a simple power law to the radial behaviour of the temperature throughout the entire cse .
however , in the case of we find @xmath124@[email protected] for @xmath45@xmath1285@xmath210@xmath92 cm .
further out in the cool , tenuous , outer layers of this cse , heating due to the photoelectric effect is important and raises the temperature . for and we estimate , over the regions where the observed emissions emanate , i.e. , from @xmath11@xmath210@xmath98 to @xmath15@xmath210@xmath92 cm , @xmath124 to be @[email protected] for both sources . for comparison , doty & leung @xcite derive @xmath124@[email protected] for the high mass loss rate object , while we obtain @xmath124@[email protected] .
from the fact that @xmath55@xmath115@xmath47 and @xmath54@xmath115@xmath116 one would expect high mass loss rate objects to have significantly cooler envelopes than low mass loss rate objects ( e.g. , jura et al . 1988 ) . in our modelling we do not see this trend . the reason being that in addition @xmath54@xmath115@xmath70@xmath129 , where all of the parameters have a mass loss rate dependence in the sense that their values are generally smaller for the low mass loss rate objects
. carbon stars generally have @xmath11c/@xmath104c - ratios in the range 30@xmath870 ( lambert et al .
1986 ; schier & olofsson 2000 ) .
this means that for the large majority of all stars in our sample line cooling by @xmath104co is insignificant , even when optical depth effects are taken into account .
however , there exists a small population of carbon stars , the j - type stars , which have @xmath11c/@xmath104c - ratios of @xmath13 . for these stars
@xmath104co line cooling could be important . to test this we have included @xmath104co line cooling for one of our sample stars , , a known j - type star .
we have used the @xmath104co model as presented in schier & olofsson @xcite to calculate the the contribution from this isotopomer to the total amount of cooling , and used this extra cooling term as input to the standard model .
we find that , after we have iterated between the @xmath11co and @xmath104co models a few times , that @xmath104co line cooling is somewhat less effective as a coolant than expected , based on the isotope ratio ( probably due to slightly different excitation conditions between the two molecules ) .
the effect on the derived line intensities is small ; the largest effect is on the @xmath14@xmath13@xmath21@xmath16@xmath18 line emission which is lowered by @xmath110% .
the changes in line intensity may be compensated for by increasing the @xmath70-parameter . moreover , ryde et al .
@xcite present a model of the high mass loss rate object using the radiative transfer code outlined here .
the @xmath11c/@xmath104c - ratio for this object was estimated to be 5.5 , i.e. , it resembles the j - type stars in this respect .
it was found that @xmath104co line cooling in this case was approximately one third of that of @xmath11co ( due to optical depth effects the @xmath104co line cooling is in a relative sense more important ) .
the decrease in kinetic temperature due to this extra cooling was compensated for by the fact that the @xmath70-parameter ( heating ) had to be somewhat increased to produce the observed line intensities .
consequently , the derived mass loss rate was not affected by introducing the @xmath104co line cooling .
for some stars , where there are enough observational constraints , it is possible to establish deviations from the standard model .
[ dcse ] lllcccccc & & & & & & + & & & & & & & & + & & & & & & & & + @xmath130 & sest & 1@xmath80 & & & & 26.1 & 29.6 & 2 + & iram & 1@xmath80 & & & & 76.2 & 79.2 & 1 + & sest & 2@xmath81 & & & & 47.8 & 49.8 & 1 + & jcmt & 2@xmath81 & & & & 50.6 & 54.8 & 3 + & iram & 2@xmath81 & & & & 67.7 & 72.0 & 1 + & sest & 3@xmath82 & & & & 63.7 & 28.0 & 2 + & jcmt & 3@xmath82 & & & & 62.9 & 30.5 & 3 + & sest & 1@xmath80 & & 0.6 & & 10.0 & 13.4 & 2 + & sest & 2@xmath81 & 3.2 & 2.8 & & 7.6 & 7.3 & 2 + & jcmt & 2@xmath81 & 3.6 & 3.3 & & 7.7 & 7.2 & 3 + & sest & 3@xmath82 & 3.6 & 4.0 & & 5.0 & 3.0 & 2 + & sest & 1@xmath80 & & 0.3 & & 5.3 & 5.6 & 1 + & iram & 1@xmath80 & 2.5 : & 1.3 & & 5.3 & 5.4 & 1 + & sest & 2@xmath81 & 1.1 & 1.3 & & 2.7 & 2.9 & 2 + & jcmt & 2@xmath81 & 0.9 & 1.3 & & 2.4 & 3.0 & 3 + & iram & 2@xmath81 & 1.8 : & 5.4 & & 2.8 & 2.8 & 1 + & sest & 3@xmath82 & 2.1 & 1.7 & & 1.0 & 0.9 & 2 + & jcmt & 3@xmath82 & 2.4 & 2.2 & & 1.0 & 0.9 & 3 + in our sample there exist a number of stars that exhibit a 60@xmath131 m excess , placing them in the regions via and vib in the iras colour - colour diagram ( see olofsson et al.1993a ) , suggesting that their cses are abnormal in some way .
previous co observations of five of these 60@xmath131 m excess stars ( see table [ dcse ] ) have revealed that their cses can not have been produced by a single smooth expanding wind , but rather suggested a drastic modulation of the mass loss on a time scale as short as a thousand years , which produced detached cses ( dcses ; olofsson et al.1996b ) .
presently , these stars are losing matter at a significantly lower rate , forming what is refered to as an attached circumstellar envelope ( acse ) . to be able to put constraints on the mass loss and its variation with time ,
high spatial resolution observations are needed .
such observations have been carried out using the iram pdb interferometer , with an angular resolution of @xmath11@xmath132 , for two of the stars in table [ dcse ] , and .
it was found that the dcse around is fairly young while that around is significantly older .
both dcses show a remarkable overall spherical symmetry , and they are geometrically very thin .
the co line emission towards these two sources , supplemented by a detailed modelling using the radiative transfer code outlined in this paper , were presented in lindqvist et al .
@xcite ( ) and olofsson et al .
( 1998 , 2000 ) ( ) . in this paper
we have attempted to model the remaining three stars with known detached shells ( , , and ) , based upon the results obtained from the high angular resolution observations described above .
the results for both the acses and dcses around these stars are presented in table [ dcse ] ( where also the results for and are included ) . due to the limited observational constraints available ,
the dcses around and were modelled using the same geometrical thickness and kinetic temperature structure as obtained for ( olofsson et al .
for we used the same geometrical thickness as adopted by olofsson et al .
1996b and assumed a kinetic temperature decrasing linearly with radius from a value of 50k at the inner shell radius .
the derived intensities from the model generally agree well with those observed ( table [ dcse - model ] ) .
however , we note that in the case of we only get about half of the observed @xmath14@xmath13@xmath21@xmath16@xmath18 line emission from the model , but here the acse could produce a significant amount of the observed emission .
we find that the estimated shell masses , @xmath133 , are relatively similar , but there is a trend of increasing shell mass with age ( the shell mass may be overestimated due to the difficulty in separating the acse and dcse emission ) , which may indicate that gas is being swept up .
the present mass loss rate was modelled for and using the standard model .
the results are presented in table [ dcse ] and table [ dcse - model ] . for the emission from the acse and dcse
can not be separated .
again , the predicted line intensities agree , with one exception , well with those observed , considering the uncertainties of the observations . for the model of the acse
fails to reproduce the iram co(@xmath14@xmath13@xmath18@xmath16@xmath15 ) observations , in contrast to the dcse model where the fit to the observations is excellent .
an explanation to this inconsistency could be that there are pointing problems in these observations ( the iram observations are more sensitive to the pointing due to its smaller beam ) , which would mainly affect the emission observed from the acse .
we find that the present mass loss rates of the stars with the oldest shells are low , and that the present mass loss rate of ( which has a much younger dcse ) is significantly higher .
the central stars are semi - regular ( sr ) or irregular variables ( lb ) , and they have low present mass loss rates .
they appear to have been subject to drastic modulations of their mass loss rates over relatively short periods of time ( a few thousand years ) .
we estimate that the co dcse emission is observable for @xmath110@xmath93 years ( bergman et al .
if this mass ejection is a repeatable phenomenon , and all our sample stars go through this phase , we estimate that the time scale between mass ejections is about 10@xmath97 years .
therefore , the mass ejections may possibly be linked with the helium - shell flashes ( the process that leads to nuclear - processed matter being transported to the surface of the star ) predicted to occur regularly in agb stars with time intervals of the order of 10@xmath97 years for the relevant masses ( e.g. , blcker 1995 ) .
we note that _ all _ carbon stars that have been found to have clearly detached gas shells are members of our sample , and that no m - type star with a detached co shell has been detected .
the sucessful modelling of the vast majority of our sample stars , assuming a spherically symmetric cse , suggests that axi- or non - symmetric mass loss is not a common phenomenon among these optically bright carbon stars . however , in the case of and , stars with complex radio line profiles , the possibility of bipolar outflows has been considered ( heske et al . 1989 ; kahane et al .
due to the complexity of these outflows , no radiative transfer analysis was attempted here .
kahane et al .
@xcite suggest that , in addition to the high velocity bipolar outflow , has a slowly expanding spherical component .
using a 2d radiative transfer code , based upon the sobolev approximation , they derive a total mass loss rate of @xmath11.5@xmath210@xmath95m@xmath117yr@xmath5 for this object .
the authors further suggest that is a transition object , between carbon stars and pne , having just developed a highly axi - symmetric mass loss .
the slowly expanding spherical component is then interpreted as the fossil cse created during the agb stage .
knapp et al .
@xcite , however , argue , based upon the spectral type , period , colours , and lack of ionizing radiation , that is still on the agb .
moreover , they estimate the total mass loss rate of this object to be @xmath14@xmath210@xmath6m@xmath117yr@xmath5 .
this , in addition to the small dynamical age of the envelope , suggests that has entered its `` superwind '' phase .
the observed line profiles of the co line emission towards are somewhat peculiar ( heske et al .
1989 ; olofsson et al.1993a ) . as a consequence , the expansion velocity derived from the observed radio line profiles range from 7.5 to 12.2kms@xmath5 ( see table [ obs ] and olofsson et al .
1993a ) depending on the transition and the telescope used .
heske et al .
@xcite , mapped the circumstellar co line emission around this object , and interpreted the observed asymmetries as being produced either by a bipolar outflow or a highly clumped wind .
no detailed modelling exist for the cse around , and its evolutionary status remains uncertain .
it might be in a state similar to that of .
[ [ section ] ] has the highest 60@xmath131 m excess of all the sample stars , which places it in the iras two - colour diagram among objects having detached envelopes ( see table [ dcse ] ) .
indeed , kerschbaum @xcite suggests that this might be a detached shell source based upon his modelling of the sed of this object .
however , the co single dish observations reveal no signs of double - peaked profiles ( confusion with interstellar lines complicates the interpretation ) , which are indicative of a detached co shell .
moreover , has a relatively low @xmath11c/@xmath104c - ratio of 6 ( abia & isern 1997 ) .
thus , it resembles the j - type stars . attempting to model this source with a standard model of its cse fails to explain both the observed @xmath14@xmath13@xmath15@xmath16@xmath17 and @xmath14@xmath13@xmath18@xmath16@xmath15 line intensities in that they come out too weak , by more than a factor of two .
this is due to the extremely cold envelope , resulting from the adopted @xmath70-parameter of 0.2 ( due to the luminosity being lower than 6000l@xmath117 ) , in combination with a relatively high mass loss rate @xmath12@xmath210@xmath95m@xmath117yr@xmath5 required to maximize the line intensities . further increasing
the mass loss rate will make the envelope even cooler due to the increase in co line cooling . if one instead assigns an @xmath70-parameter of @xmath11 it is possible to obtain a reasonable fit to the data using a mass loss rate of @xmath15@xmath210@xmath95m@xmath117yr@xmath5 .
this estimate must be regarded as highly uncertain .
fig . [ distr_all ] shows the distribution of the mass loss rates obtained from the radiative transfer model for all sources ( also shown is the distribution for the stars within 500pc of the sun ) , as well as divided into variability groups ( note that objects with known detached shells have not been included in this analysis ) .
since the statistics of the sra stars are poor we have chosen to group them together with the srb stars .
the mass loss rate distribution for all stars is sharply peaked around the median mass loss rate of 2.8@xmath210@xmath7m@xmath4yr@xmath5 ( the median mass loss rate for the stars within 500pc of the sun is 1.6@xmath210@xmath7m@xmath4yr@xmath5 ) .
it is interesting to note that this is roughly the rate at which the core mass is expected to grow due to nuclear burning ( schnberner 1983 ) .
mira variables generally have larger mass loss rates than other variability types , while irregular variables and semiregulars appear to have very similar mass loss rate characteristics .
based on the mass loss rate distribution for the sample of stars within 500pc , which we believe to be close to complete and for which we have detected all sources , we are able to draw some general conclusions .
the sharp decline at mass loss rates below @xmath15@xmath210@xmath9m@xmath117yr@xmath5 is very likely real .
netzer & elitzur @xcite estimate that a mass loss rate in excess of @xmath110@xmath7m@xmath4yr@xmath5 is required to get a dust - driven wind .
the exact limit is , however , sensitive to the adopted stellar and dust parameters .
the drastically decreasing number of high mass loss rate objects is also real , although our selection criterion bias against these objects , and can be explained by the fact that carbon stars of the type discussed here , i.e. , mainly low mass ones ( claussen et al .
1987 ) , only for a limited time , or possibly never , reach high mass loss rates .
the @xmath38-distribution for this sample of stars has already been shown and discussed by olofsson et al .
@xcite . however , a new comparison between the mass loss characteristics @xmath47 and @xmath38 is warranted considering the more reliable mass loss rates obtained in this paper , fig .
[ mdot - ve ] .
we find a clear trend that @xmath38 increases with @xmath47 , @xmath38@xmath115@xmath134 ( with a correlation coefficient of 0.78 ) . in comparison ,
olofsson et al .
@xcite derived @xmath38@xmath115@xmath135 , but their mass loss rates were calibrated using eq . [ mdot_simple ] , which includes a @xmath136-dependence .
the correlation is much tighter than that obtained by olofsson el al .
@xcite , which is reassuring .
thus , the mass loss mechanism operates such that mass loss rate and expansion velocity increase together .
however , the scatter appears larger than the uncertainties in the estimates , and so the mechanism also produces widely different mass loss rates for a given expansion velocity . at high mass loss rates the scatter is larger with a possible divison into objects with high mass loss rates but only moderately high velocities , and objects with moderately high mass loss rates but high velocities .
we note also that for the detached shell sources the mass loss rates and expansion velocities for the two mass loss epochs follow the general trend . habing et al .
@xcite studied the momentum transfer from the photons , via the dust , to the gas in the cses of agb stars .
they found that for mass loss rates close to the minimum mass loss rate for a dust - driven wind , @xmath13@xmath210@xmath9m@xmath4yr@xmath5 , @xmath38 increases linearly with @xmath47 , while for mass loss rates above @xmath110@xmath95m@xmath4yr@xmath5 the dependence weakens considerably . at 10@xmath6m@xmath4yr@xmath5
they found @xmath38@xmath115@xmath137 .
habing et al .
attribute the increase in @xmath38 with mass loss rate ( for low mass loss rates ) to a higher efficiency in the coupling between gas and dust .
this can explain the observed behaviour in fig .
[ mdot - ve ] for mass loss rates below about 10@xmath95m@xmath4yr@xmath5 , but for the higher mass loss rates it appears that a dependence of @xmath38 ( and @xmath47 ) on luminosity gives the most reasonable explanation , see sect . [
s : stellarprop ] . in fig .
[ mdot - stellar ] we plot the circumstellar characteristics , @xmath47 and @xmath38 , against the stellar characteristics luminosity ( @xmath58 ) , period ( @xmath138 ) , effective temperature ( @xmath139 ) , and the photospheric c / o - ratio .
the effective temperatures and c / o - ratios used in this analysis are those presented in olofsson et al .
we have looked for dependences assuming that the ordinate scales as the abscissa to the power of @xmath140 , and a normal correlation coefficient @xmath45 was calculated as an estimate of the quality of the fit , see fig .
[ mdot - stellar ] .
the uncertainties in the estimated quantities are of the order : @xmath3650% ( @xmath47 ) , @xmath362kms@xmath5 ( @xmath38 ) , @xmath3610% ( @xmath138 ; but some periods may be poorly determined ) , a factor of 2 ( @xmath58 ) , and @xmath36200k ( @xmath139 ) . clearly , both the mass loss rate and the expansion velocity increase with the pulsation period of the star .
there is a weaker trend with the luminosity of the star .
the latter dependence is not completely independent of the former , since some of the luminosities are estimated from a @xmath141 relation .
the existence of a @xmath141 relation is usually attributed to a distribution in mass ( jones et al .
1994 ) , i.e. , the higher the mass the longer the period and the higher the luminosity .
when looked at in detail , at least the apparent mass loss rate dependence on period may be attributed to a change from semiregular pulsation at short periods to regular pulsations at longer periods .
we may therefore infer that for the mass lass rate it is not clear whether it is the regularity of the pulsation or the luminosity that causes the increase with period . for the expansion velocity
the increase with period may be a combined effect of increasing mass loss rate with period ( for low mass loss rates ) and an increase in luminosity for the longer periods , e.g. , habing et al . @xcite
derive @xmath48@xmath115@xmath142 in their dust - driven wind model .
in addition , there is evidence of a weak trend in the sense that higher mass loss rate objects have lower effective temperatures .
these correlations are all consistent with a dust - driven wind , where the pulsation may play an important role .
there appears to be no correlation between the mass loss rate and the c / o - ratio , which is surprising considering that the dust - to - gas mass ratio in a c - rich cse should be sensitively dependent on this .
the present mass loss rates for the detached shell sources are typical for their periods but somewhat low for their luminosities .
however , during the formation of the dcses the mass loss rate must have been atypically high for the present periods and luminosities .
we also find that the expansion velocities of the dcses lie at the very high end of expansion velocities found for other stars with the same luminosity , indicating that these stars had higher luminosities during the shell ejection . two other stars , and , stand out in these plots .
the properties of resembles the stars with known dcses , but the co line profiles give no indication of a detached shell .
has by far the lowest estimated mass loss rate in the sample , as well as the lowest c / o - ratio , 1.01 , which classifies it as an sc - star .
it is also a li - rich j - star ( abia & isern1997 ) .
carbon stars on the agb are important in returning processed gas to the interstellar medium ( ism ) .
the total mass loss rate of carbon stars in the galaxy is obtained from @xmath143 where @xmath144 is the surface density of carbon stars , @xmath41@xmath47@xmath50 is the mean mass loss rate of the carbon stars . according to guglielmo et al .
@xcite the infrared carbon stars have a roughly constant surface density out to the galactocentric distance of the sun , after which it follows an exponential decline with a scale length of @xmath12.5kpc .
we approximate this by assuming a constant value , @xmath145 ( the surface density in the solar neighborhood ) , to 10kpc and zero beyond this galactocentric distance . based on the complete sample , i.e. , stars within 500pc from the sun
, we estimate @xmath145@xmath41@xmath47@xmath50 to be @xmath11.7@xmath210@xmath146m@xmath4yr@xmath5pc@xmath147 , where we have included also helium .
this estimate is very sensitive to the number of high mass loss rate objects found within 500pc , e.g. , the high mass loss rate object contributes almost half of our estimate of the total mass returned to the ism .
our estimate of the rate at which matter is returned to the ism by carbon stars is consistent with previous estimates ( considering the large uncertainties ) , e.g. , knapp & morris @xcite and jura & kleinmann @xcite derive values of @xmath12@xmath210@xmath146m@xmath4yr@xmath5pc@xmath147 and @xmath11.5@xmath210@xmath146m@xmath4yr@xmath5pc@xmath147 , respectively .
using eq .
[ totmassloss ] the annual return of matter to the ism in the galaxy is estimated to be @xmath10.05m@xmath4 for the carbon stars considered here .
it is quite possible that a larger mass return from carbon stars is obtained during their final evolution on the agb , a so called superwind phase , as is indicated by the estimated values for infrared carbon stars , @xmath10.5m@xmath4yr@xmath5 ( epchtein et al .
1990 ; guglielmo et al .
1993 ) , and pne , @xmath10.3m@xmath4yr@xmath5 ( gustafsson et al.1999 ) .
this confirms the importance of carbon stars for the cosmic gas cycle in galaxies . by comparison
, high mass stars are estimated to contribute about 0.04m@xmath4yr@xmath5 ( gustafsson et al .
1999 ) .
we estimate the contribution from carbon stars to the carbon enrichment of the ism to be @xmath10.5@xmath210@xmath10m@xmath4yr@xmath5 during the agb stage ( if the subsequent superwind phase is included this value may increase to @xmath15@xmath210@xmath10m@xmath4yr@xmath5 ) .
this corroborates the conclusions by gustafsson et al .
@xcite that ` normal ' carbon stars are not important for the total carbon budget of the galaxy .
according to these authors the main contributors should instead be high mass stars in the wolf - rayet stage , annually supplying the galaxy with @xmath10.01m@xmath4 of carbon .
this is roughly what is required to produce the @xmath110@xmath85@xmath210@xmath148m@xmath4 of carbon present in the milky way over a period of 10@xmath149 years .
we have developed a detailed radiative transfer code to model circumstellar molecular line emission .
the code also solves for the energy balance equation of the gas .
it is found that the mass loss rate determination for low mass loss rate objects depends crucially on a number of assumptions in the cse model , except on the dust properties , since the co molecules are radiatively excited . on the other hand ,
the mass loss rate estimate for a high mass loss rate object depends essentially only on the temperature structure , and hence the uncertain dust parameters .
in addition , for such an object the co emission saturates , and becomes less useful as a mass loss rate measure .
we also find that different lines respond differently to changes in the various parameters .
therefore , a reliable mass loss rate determination requires , in addition to a detailed radiative transfer analysis , good observational constraints in the form of multi - transition observations and radial brightness distributions .
this model has been applied to co radio line observations of a large sample of optically bright n - type and j - type carbon stars on the agb ( 69 objects ) .
the sample is reasonably complete out to about 1kpc , and all stars ( 41 ) within @xmath1500pc of the sun have been detected in circumstellar co line emission .
the derived mass loss rates span almost four orders of magnitude , @xmath15@xmath210@xmath3m@xmath117yr@xmath5 to @xmath12@xmath210@xmath6m@xmath117yr@xmath5 , over which the physical conditions of the cses vary considerably .
the fact that the model can be succesfully applied over such a wide range of environments gives us confidence in the results , even though we are aware of the fact that some assumptions are poorly constrained . the large majority of the stars have mass loss rates in a narrow range centered at @xmath13@xmath210@xmath7m@xmath4yr@xmath5 , and it appears that very few agb carbon stars * ( @xmath1285% ) * lose matter at a rate less than @xmath15@xmath210@xmath9m@xmath4yr@xmath5 .
we find that the mass loss rate and the gas expansion velocity are relatively well correlated , but the scatter is large enough that we may conclude that the mass loss mechanism is able to produce a wide range of mass loss rates for a given expansion velocity .
the mass loss rate is also well correlated with the pulsational period of the star , correlated with the stellar luminosity , and there is a weak trend with the stellar effective temperature , in the sense that the cooler stars tend to have higher mass loss rates .
also the gas expansion velocity is positively correlated with the period and the luminosity .
we conclude that the mass loss rate increases with increased regular pulsation and/or luminosity , and that the expansion velocity increases with mass loss rate ( for low mass loss rates ) and luminosity .
the observed trends are all supporting the common concensus that these winds are driven by radiation pressure on dust grains , and that pulsation may play an important role .
somewhat surprising there appears to be no dependence on the stellar c / o - ratio . our standard cse model , assuming a single smooth expanding wind produced by a continuous mass loss , fails to reproduce the observational data for about 10% of the sample stars .
most notable among these are five stars with detached cses , presumably formed during a period of highly increased mass loss , and low present mass loss rates .
we have found indications that the present mass loss rate is significantly higher for the star with the youngest dcse , and that the shell mass may increase with shell age .
since our sample is reasonably complete , we can estimate that the time scale between mass ejections is about 10@xmath97 years , if it is a repeatable phenomenon .
an association with he - shell flashes is favoured .
these objects have been the targets of a number of extensive , observational and theoretical studies . for some of our sample stars there
exist enough observational constraints to determine a combined dust parameter from the kinetic temperature structure .
this result can be interpreted as a dust - to - gas mass ratio which increases with mass loss rate , but changes in the dust properties may also play a role .
we also find that this means that the gas kinetic temperature in a carbon star cse depends only weakly on the mass loss rate .
the size of the co envelope is an important parameter in the mass loss rate determination , at least for the low mass loss rate objects .
we have used published radial co(@xmath14@xmath13@xmath15@xmath16@xmath17 ) brightness distributions , in combination with the radiative transfer code , to show that the co photodissociation calculation by mamon et al .
@xcite gives reasonably accurate results .
in a few cases the observed radial brightness distributions are clearly different than the model results , suggesting deviations from a simple @xmath0 density law , and hence time - variable mass loss .
we estimate that carbon stars , of the type studied here , return on the order of 0.05m@xmath4yr@xmath5 of gas to the ism making them marginally important for the gas cycle in galaxies .
more extreme carbon stars may contribute an order of magnitude more .
however , they are probably not important as regards the origin of carbon .
we are grateful to dr .
f. kerschbaum for generously providing estimates of some of the input parameters to the co modelling , to dr .
r. liseau for useful comments , and to an anonymous referee for constructive criticism which lead to an improved paper .
financial support from the swedish natural science research council ( nfr ) is gratefully acknowledged . | using a detailed radiative transfer analysis , combined with an energy balance equation for the gas , we have performed extensive modelling of circumstellar co radio line emission from a large sample of optically bright carbon stars , originally observed by olofsson et al .
( apjs , 87 , 267 ) .
some new observational results are presented here .
we determine some of the basic parameters that characterize circumstellar envelopes ( cses ) , e.g. , the stellar mass loss rate , the gas expansion velocity , and the kinetic temperature structure of the gas . assuming a spherically symmetric cse with a smooth gas density distribution , created by a continuous mass loss , which expands with a constant velocity we are able to model reasonably well 61 of our 69 sample stars .
the derived mass loss rates depend crucially on the assumptions in the circumstellar model , of which some can be constrained if enough observational data exist .
therefore , a reliable mass loss rate determination for an individual star requires , in addition to a detailed radiative transfer analysis , good observational constraints in the form of multi - line observations and radial brightness distributions . in our analysis
we use the results of a model for the photodissociation of circumstellar co by mamon et al.(1988 ) .
this leads to model fits to observed radial brightness profiles that are , in general , very good , but there are also a few cases with clear deviations , which suggest departures from a simple @xmath0 density law .
the derived mass loss rates span almost four orders of magnitude , from @xmath15@xmath210@xmath3m@xmath4yr@xmath5 up to @xmath12@xmath210@xmath6m@xmath4yr@xmath5 , with the median mass loss rate being 2.8@xmath210@xmath7m@xmath4yr@xmath5 .
we estimate that the mass loss rates are typically accurate to @xmath150% within the adopted circumstellar model .
the physical conditions prevailing in the cses vary considerably over such a large range of mass loss rates . among other things , it appears that the dust - to - gas mass ratio and/or the dust properties change with the mass loss rate .
we find that the mass loss rate and the gas expansion velocity are well correlated , and that both of them clearly depend on the pulsational period and ( with larger scatter ) the stellar luminosity .
moreover , the mass loss rate correlates weakly with the stellar effective temperature , in the sense that the cooler stars tend to have higher mass loss rates , but there seems to be no correlation with the stellar c / o - ratio .
we conclude that the mass loss rate increases with increased regular pulsation and/or luminosity , and that the expansion velocity increases as an effect of increasing mass loss rate ( for low mass loss rates ) and luminosity .
five , of the remaining eight , sample stars have detached cses in the form of geometrically thin co shells . the present mass loss rates and shell masses of these sources
are estimated .
finally , in three cases we encounter problems using our model . for two of these sources
there are indications of significant departures from overall spherical symmetry of the cses .
carbon stars on the agb are probably important in returning processed gas to the ism .
we estimate that carbon stars of the type considered here annually return @xmath10.05m@xmath4 of gas to the galaxy , but more extreme carbon stars may contribute an order of magnitude more .
however , as for the total carbon budget of the galaxy , carbon stars appear to be of only minor importance . |
the bimodal distribution of galaxy colors at low redshift is well established ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
the blue cloud " , sometimes referred to as the star formation sequence " @xcite , is occupied by star - forming galaxies , while the red sequence " galaxies appear to have little ongoing star formation ( sf ) .
it has been argued that this bimodality suggests that sf of the blue cloud galaxies en route to the red sequence must be turned off promptly to prevent them from lingering in the green valley between the blue and red peaks .
a number of physical mechanisms have been proposed to cause this apparent quenching " of sf .
galaxy mergers have been suggested to trigger strong and rapid sf that subsequently drives gas away and shuts down further sf activities in a sudden fashion .
however , recent observations show that galaxies in the green valley do not show merger signatures , perhaps disfavoring the merger scenario ( e.g. , * ? ? ? * ) .
feedback from active galactic nuclei ( agn ) has also been suggested to provide quenching , but observational evidence for this scenario has been either inconclusive or at best circumstantial ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
some recent studies based on large data sets do not find evidence for agn feedback playing a role in galaxy color migration ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
external , environmental effects may have played an important role in shaping galaxy colors .
high density environments are observed to be occupied primarily by early type ( elliptical and s0 ) red galaxies - the density - morphology relation " ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) - with giant elliptical galaxies anchoring the centers of rich clusters of galaxies ( e.g. , * ? ? ?
this relation is consistent with the larger trend of the galaxy population appearing bluer in more underdense regions in the local universe ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
different types of galaxies are seen to cluster differently and have different environment - dependencies , in the same sense as the density - morphology relation ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
recent quantitative studies have yielded richer details on sf dependence on halo mass and environment , probing their relationships at higher redshifts .
for example , using a large group catalog from the sloan digital sky survey ( sdss ) data release 2 , @xcite find that at fixed luminosity the fraction of early - type galaxies increases with increasing halo mass and this mass dependence is smooth and persists over the entire mass range probed without any break or feature at any mass - scale . from a spectral analysis of galaxies at @xmath6 based on the eso distant cluster survey , @xcite find that the incidence of k+a galaxies increases strongly with increasing velocity dispersion of the environment from groups to clusters . @xcite , examining the sf properties of group and field galaxies from sdss at @xmath7 and from ultraviolet imaging with galex at @xmath8 , find that the fraction of passive galaxies is higher in groups than the field at both redshifts , with the difference between the group and field growing with time and larger at low masses .
with the noao extremely wide - field infrared imager ( newfirm ) survey of the all - wavelength extended groth strip international survey ( aegis ) and cosmic evolution survey ( cosmos ) fields , @xcite show evidence for a bimodal color distribution between quiescent and star - forming galaxies that persists to @xmath9 .
@xcite study the evolution of galaxies located within groups using the group catalog obtained from zcosmos spectroscopic data and the complementary photometric data from the cosmos survey at @xmath10 and find the rate of sf quenching to be faster in groups than in the field .
@xcite analyze galaxy properties at @xmath11 using a spectroscopic sample of 797 cluster and field galaxies drawn from the gemini cluster astrophysics spectroscopic survey , finding that post starburst galaxies with @xmath12 are three times more common in high - density regions compared to low - density regions . based on data from the zcosmos survey @xcite perform an environment study and find that quiescent galaxies prefer more massive systems at @xmath13 .
@xcite , analyzing galex imaging of a statistically representative sample of 23 galaxy groups at @xmath14 , suggest an average quenching timescale of @xmath15gyr .
@xcite , with deep gmos - s spectroscopy for 11 galaxy groups at @xmath16 , show that the strongest environmental dependence is observed in the fraction of passive galaxies , which make up only @xmath17 per cent of the field in the mass range @xmath18 but are the dominant component of groups . using sdss ( @xmath19 ) and
the all - wavelength extended groth strip international survey ( aegis ; @xmath20 ) data , @xcite find a strong environmental dependence of quenching in terms of halo mass and distance to the centrals at both redshifts .
the widespread observational evidence of environment quenching is unsurprising theoretically . in regions of overdensity , whether around a large collapsed halo or unvirialized structure ( e.g. , a zeldovich pancake or a filament ) , gas is gravitationally shock heated when converging flows meet . in regions filled with hot shock - heated gas , multiple gasdynamical processes would occur .
one of the most important gasdynamical processes is ram - pressure stripping of gas , when a galaxy moves through the ambient hot gas at a significant speed , which includes , but is not limited to , the infall velocity of a satellite galaxy .
the theoretical basis for the ram - pressure stripping process is laid down in the seminal work of @xcite .
recent works with detailed simulations of this effect on galaxies ( in non - cosmological settings ) include those of @xcite , @xcite , @xcite , @xcite and @xcite . even in the absence of ram - pressure stripping , ubiquitous supersonic and transonic motions of galaxies of complex acceleration patterns through ambient medium ( intergalactic or circumgalactic medium )
subject them to the raleigh - taylor and richtmyer - meshkov instabilities .
large shear velocities at the interfaces between galaxies and the ambient medium allow the kelvin - helmholtz ( kh ) instability to play an important role .
when these processes work in tandem with ram - pressure displacements , the disruptive effects are amplified .
for example , the kh instability time scale is substantially shorter for a non - self - gravitating gas cloud ( e.g. , * ? ? ? * ) than for one sitting inside a virialized dark matter halo ( e.g. , * ? ? ?
another important process in hot environments is starvation of cold gas that is fuel for sf ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) . in regions with high - temperature and high - entropy cooling of hot gas
is an inefficient process for fueling sf , an important point noted long ago to account for the basic properties ( mass , size ) of galaxies ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
this phenomenon may be understood by considering the dependence of cooling time on the entropy of the gas : the gas cooling time can be written as @xmath21 $ ] @xcite .
it follows that the minimum cooling time of a gas parcel just scales with @xmath22 . as a numerical example , for a gas parcel of entropy @xmath23k @xmath24 ( say , for temperature @xmath25k and density @xmath26@xmath27 ) and metallicity @xmath28 , its cooling time is no shorter than the hubble time at @xmath29 hence the gas can no longer cool efficiently to fuel sf .
it may be that the combination of cold gas removal and dispersal by ram - pressure stripping , hydrodynamic instabilities , and cold gas starvation , all of which are expected to become increasingly important in more massive environments , plays a primary role in driving the color migration from the blue cloud to the red sequence .
in dense environments , gravitational tidal ( stripping and shock ) effects and relatively close fly - bys between galaxies ( e.g. , * ? ? ?
* ) also become important . to understand the overall effect on sf quenching by these external processes in the context of the standard cold dark matter model , a realistic cosmological setting is imperative , in order to capture complex external processes that are likely intertwined with large variations of internal properties of galaxies . in this paper
we perform _ ab initio _ * l*arge - scale * a*daptive - mesh - refinement * o*mniscient * z*oom-*i*n cosmological hydrodynamic simulations , called * laozi simulations * , to obtain a large sample of galaxies to , for the first time , perform a chronological and statistical investigation on a very large scale . the large simulated galaxy sample size and very high resolution of laozi simulations provide an unprecedented opportunity to undertake the study presented .
our study shares the spirit of the work by @xcite , who examine the evolution of a dozen galaxies falling onto a forming group of galaxies , with a substantial improvement in the statistical treatment , the simulation resolution , the range of environment probed , and the analysis scope .
feedback from agn is not included in this simulation , partly because of its large uncertainties and present lack of definitive driving sources and primarily due to our intention to focus on external effects .
internal effects due to sf are automatically included and we find no evidence that sf or merger triggered sf plays a primary role in quenching from our study .
the outline of this paper is as follows . in
2 we detail our simulations ( 2.1 ) , method of making galaxy catalogs ( 2.2 ) , construction of histories of galaxies ( 2.3 ) , tests and validation of simulations ( 2.4 ) and the produced bimodal distribution of galaxies colors ( 2.5 ) .
results are presented in 3 , organized in an approximately chronological order , starting with the ram - pressure stripping effects in 3.1 , followed by the ensuing period of gas starvation in hot environment in 3.2 . in
3.3 we discuss stellar mass growth , evolution of stellar mass function of red galaxies and present galaxy color migration tracks .
3.4 gives an example of consequences of the found color migration picture - galaxy age - mass relation .
we present observable environmental dependence of galaxy makeup at @xmath2 in 3.5 .
conclusions are given in 4 .
we perform cosmological simulations with the amr eulerian hydro code , enzo @xcite .
first we run a low resolution simulation with a periodic box of @xmath30mpc ( comoving ) on a side .
we identify a region centered on a cluster of mass of @xmath31 at @xmath32 .
we then resimulate with high resolution of the chosen region embedded in the outer @xmath33mpc box to properly take into account the large - scale tidal field and appropriate boundary conditions at the surface of a refined region .
the refined region has a comoving size of @xmath34mpc@xmath35 and represents a @xmath36 matter density fluctuation on that volume .
the dark matter particle mass in the refined region is @xmath37 .
the refined region is surrounded by three layers ( each of @xmath38mpc ) of buffer zones with particle masses successively larger by a factor of @xmath39 for each layer , which then connects with the outer root grid that has a dark matter particle mass @xmath40 times that in the refined region .
we choose the mesh refinement criterion such that the resolution is always smaller than @xmath41pc ( physical ) , corresponding to a maximum mesh refinement level of @xmath42 at @xmath32 .
an identical comparison run that has four times better resolution of @xmath43pc / h was also run down to @xmath44 and some relevant comparisons between the two simulations are made to understand effects of limited resolution on our results .
the simulations include a metagalactic uv background @xcite , and a model for self - shielding of uv radiation @xcite .
they include metallicity - dependent radiative cooling @xcite .
our simulations also solve relevant gas chemistry chains for molecular hydrogen formation @xcite , molecular formation on dust grains @xcite , and metal cooling extended down to @xmath45k @xcite .
star particles are created in cells that satisfy a set of criteria for sf proposed by @xcite .
each star particle is tagged with its initial mass , creation time , and metallicity ; star particles typically have masses of @xmath46@xmath47 .
supernova feedback from sf is modeled following @xcite .
feedback energy and ejected metal - enriched mass are distributed into 27 local gas cells centered at the star particle in question , weighted by the specific volume of each cell , which is to mimic the physical process of supernova blastwave propagation that tends to channel energy , momentum and mass into the least dense regions ( with the least resistance and cooling ) .
the primary advantages of this supernova energy based feedback mechanism are three - fold .
first , nature does drive winds in this way and energy input is realistic .
second , it has only one free parameter @xmath48 , namely , the fraction of the rest mass energy of stars formed that is deposited as thermal energy on the cell scale at the location of supernovae .
third , the processes are treated physically , obeying their respective conservation laws ( where they apply ) , allowing transport of metals , mass , energy and momentum to be treated self - consistently and taking into account relevant heating / cooling processes at all times .
we allow the entire feedback processes to be hydrodynamically coupled to surroundings and subject to relevant physical processes , such as cooling and heating .
the total amount of explosion kinetic energy from type ii supernovae with a chabrier initial mass function ( imf ) is @xmath49 ( where @xmath50 is the speed of light ) , for an amount @xmath51 of star formed .
taking into account the contribution of prompt type i supernovae ,
we use @xmath52 in our simulations . observations of local starburst galaxies indicate that nearly all of the sf produced kinetic energy is used to power galactic superwinds ( e.g. , * ? ? ?
supernova feedback is important primarily for regulating sf and for transporting energy and metals into the intergalactic medium .
the extremely inhomogeneous metal enrichment process demands that both metals and energy ( and momentum ) are correctly modeled so that they are transported in a physically sound ( albeit still approximate at the current resolution ) way .
we use the following cosmological parameters that are consistent with the wmap7-normalized @xcite @xmath53cdm model : @xmath54 , @xmath55 , @xmath56 , @xmath57 , @xmath58 and @xmath59 .
these parameters are consistent with those from planck first - year data @xcite if we average planck derived @xmath60 with sn ia and hst based @xmath60 .
we note that the size of the refined region , @xmath34mpc@xmath35 , is still relatively small and the region biased .
this is , of course , designed on purpose .
because of that , however , we are not able to cover all possible environment , such as the center of a void . also because of that
, we have avoided addressing any measures that requires a precise characterization of the abundance of any large galaxy systems , such as the mass function or luminosity function of massive galaxies or groups . despite that , measures that are characterized as a function of environment / system masses should still be valid .
our environment coverage is substantially larger than probed in , for example , @xcite . in @xcite
we show that the present simulation box ( c box ) ( run to @xmath32 with a lower resolution previously ) spans a wide range in environment from rich clusters to the field , and there is a substantial overlap in the field environment with another simulation centered on a void ( v box ) .
it is the density peaks higher than we model here ( i.e. , more massive clusters of galaxies ) that we fail to probe . as it should be clear later
, this shortcoming should not affect any of our conclusions , which may be appropriately extrapolated .
we identify galaxies in our high resolution simulations using the hop algorithm @xcite operating on the stellar particles , which is tested to be robust and insensitive to specific choices of concerned parameters within reasonable ranges .
satellites within a galaxy down to mass of @xmath61 are clearly identified separately in most cases .
the luminosity of each stellar particle in each of the sloan digital sky survey ( sdss ) five bands is computed using the gissel stellar synthesis code @xcite , by supplying the formation time , metallicity and stellar mass .
collecting luminosity and other quantities of member stellar particles , gas cells and dark matter particles yields the following physical parameters for each galaxy : position , velocity , total mass , stellar mass , gas mass , mean formation time , mean stellar metallicity , mean gas metallicity , sfr , luminosities in five sdss bands ( and various colors ) and others . at a spatial resolution of @xmath62pc ( physical ) with thousands of
well resolved galaxies at @xmath63 , the simulated galaxy catalogs present an excellent ( by far , the best available ) tool to study galaxy formation and evolution . when we start the analysis for this paper , the simulation has reached @xmath2 . for each galaxy at @xmath2
a genealogical line is constructed from @xmath2 to @xmath64 by connecting galaxy catalogs at a series of redshifts .
galaxy catalogs are constructed from @xmath2 to @xmath65 at a redshift increment of @xmath66 and from @xmath65 to @xmath64 at a redshift increment of @xmath67 .
the parent of each galaxy is identified with the one at the next higher redshift catalog that has the most overlap in stellar mass .
we call galaxies with @xmath68 blue " , those with @xmath69 green " and those with @xmath70 red " , in accord with the bimodal color distribution that we will show below and with that of observed galaxies ( e.g. , * ? ? ?
* ) , where @xmath71 and @xmath72 are magnitudes of sdss @xmath71 and @xmath72 bands . in subsequent analysis
, we will examine gasdynamic processes , e.g. , cold gas loss or lack of cold gas accretion , under the working hypothesis that ram - pressure stripping and gas starvation are the primary detrimental processes to star formation .
we should assume that other processes , such as hydrodynamical instabilities ( e.g. , rt , kh , tidal shocks , etc ) , may either be lumped together " with ram - pressure stripping or play some role to enhance cold gas destruction that is initiated by ram - pressure stripping .
a word on tidal stripping may be instructive .
it is noted that while tidal stripping would affect both stars and gas , ram - pressure operates only on the latter . as one
will see later , in some cases the stellar masses of galaxies decrease with time , which are likely due to tidal effects .
a simple argument suggests that ram - pressure effects are likely to be more far - reaching spatially and are more consistent with the environment effects becoming effective at 2 - 3 virial radii than tidal stripping that we will show later .
let us take a specific example to illustrate this .
let us assume that the primary and infalling galaxies have a velocity dispersion of @xmath73 and @xmath74 , respectively , and that they both have isothermal sphere density profiles for both dark matter and baryons .
the virial radius is proportional to its velocity dispersion in each case .
under such a configuration , we find that the tidal radius for the satellite galaxy at its virial radius is equal to the virial radius of the primary galaxy . on the other hand ,
the ram - pressure force on the gas in the satellite at its virial radius is already equal to the gravitational restoring by the satellite , when the satellite is ( @xmath75 ) virial radii away from the primary galaxy . in reality , of course , the density profiles for dark matter and baryons are different and neither is isothermal , and the gas may display a varying degree of non - sphericity .
but the relative importance of ram - pressure and tidal strippings is likely to remain the same for relatively diffuse gas .
the relative situation is unchanged , if one allows the gas to cool and condense . as an example
, if the gas within the virial radius of the satellite and the primary galaxies in the above example is allowed to shrink spherically by a factor of 10 in radius ( we will continue to assume that the velocity dispersion or rotation velocity remains flat and at the same amplitude ) , we find that the tidal stripping radius is now a factor of 10 smaller than before ( equal to 0.1 times the virial radius of the primary galaxy ) , while the new ram - pressure stripping radius is @xmath75 times the new tidal stripping radius . as a third example , if the gas within the virial radius of the satellite galaxy in the above example is allowed to shrink spherically by a factor of 10 in radius but the gas in the primary galaxy does not shrink in size , it can be shown that in this case the tidal stripping radius is equal to 0.1 times the virial radius of the primary galaxy , while the new ram - pressure stripping radius is now 0.1 times ( sigma1/sigma2 ) times the virial radius of the primary galaxy .
as the last example , if the gas within the virial radius of the satellite galaxy in the above example is allowed to shrink by a factor of 10 in radius to become a disk but the gas in the primary galaxy does not shrink in size , it can be shown that in this case the tidal stripping radius is equal to 0.1 times the virial radius of the primary galaxy .
the new ram - pressure stripping radius depends on the orientation of the motion vector and the normal of the disk : if the motion vector is normal to the disk , the tidal stripping radius is 0.1 times @xmath75 times the virial radius of the primary galaxy ; if the motion vector is in the plane to the disk , the tidal stripping radius is zero .
-0.0 cm -0.5 cm -0.0 cm -0.5
cm we denote a point in time when the galaxy turns from blue to green as @xmath76 , a point in time when the galaxy turns from green to red as @xmath77 . convention for time
is that the big bang occurs at @xmath78 .
we identify a point in time , searched over the range @xmath792gyr to @xmath801gyr , when the derivative of sfr with respect to time , @xmath81 , is most negative , as @xmath82 ( @xmath83 stands for quenching ) ; in practice , to reduce uncertainties due to temporal fluctuations in sfr , @xmath82 is set to equal to @xmath84 when the sliding - window difference @xmath85 is most negative , where @xmath86 are the times of our data outputs , as noted earlier .
galaxies at @xmath82 are collectively called sfqs for star formation quenching galaxies . to demonstrate the reliability and accuracy of identification of @xmath82 we show in figure [ fig : red0 ]
the histories for a set of four randomly selected red galaxies at @xmath2 of stellar mass @xmath87 .
the vertical dashed line in each panel shows @xmath82 , which is the location of steepest drop of sfr ( solid dots ) . in all four cases ,
our method identifies the location accurately .
figure [ fig : red1 ] is similar to figure [ fig : red0 ] but for galaxies of stellar mass @xmath88 , where we see our method identifies @xmath82 with a similar accuracy .
similarly , we identify a point in time in the range @xmath792gyr to @xmath801gyr , when the derivative of the amount of cold gas , @xmath89 within radial ranges @xmath90kpc , with respect to time is most negative as ( @xmath91 , respectively .
we define cold gas as gas with temperature less than @xmath92k .
the exponential decay time scale of sfr at @xmath93 is defined by @xmath94 .
the exponential decay time scale of @xmath89 at ( @xmath91 are defined by @xmath95 $ ] .
the time interval between @xmath82 and @xmath77 is denoted as @xmath96 , the time duration that the galaxy spends in the green valley before turning red is called @xmath97 .
the time duration the galaxy has spent in the red sequence by @xmath2 is denoted as @xmath98 .
we make a needed simplification by approximating the ram - pressure , denoted as @xmath99 , by @xmath100 , where @xmath101 and @xmath102 , respectively , are the mean density of gas with temperature @xmath103k and @xmath102 the mean mass - weighted gas temperature within a proper radius of @xmath104kpc centered on the galaxy in question .
this tradeoff is made thanks chiefly to the difficulty of defining precisely the motion of a galaxy relative to its ambient gas environment , where the latter often has complex density and velocity structures , and the former has complex , generally non - spherical gas distribution geometry . in a gravitationally shock
heated medium , this approximation should be reasonably good , because the ram - pressure is approximately equal to thermal pressure in post - shock regions .
we define a point in time searched over the time interval between @xmath1052gyr and @xmath1061gyrs , when the derivative of @xmath99 with respect to time is maximum as @xmath107 , intended to serve as the point in time when ram - pressure has the steepest rise , as stated in the introduction , it is convenient to express gas cooling time that is proportional to gas entropy to the power @xmath108 , @xmath22 .
thus , we approximate gas starvation from large scales by the value of environmental entropy @xmath109 , defined to be the average gas entropy within a top - hat sphere of proper radius @xmath104kpc . for convenience , frequently used symbols and their definitions are given in table [ tab : table1 ] .
ll @xmath76 & a point in time when galaxy has @xmath110
+ @xmath77 & a point in time when galaxy has @xmath111 + @xmath112 & amount of cold gas within a radius of 10kpc + @xmath113 & amount of cold gas within a radius of 30kpc + @xmath114 & amount of cold gas within a radius of 100kpc + @xmath82 & a point in time of quenching for sfr + @xmath115 & a point time of quenching for @xmath116 + @xmath117 & a point time of quenching for @xmath118 + @xmath119 & a point time of quenching for @xmath120 + @xmath107 & a point in time of largest first derivative of ram - pressure w.r.t time + @xmath121 & exponential decay time of sfr at @xmath82 + @xmath122 & exponential decay time of @xmath116 at @xmath115 + @xmath123 & exponential decay time of @xmath118 at @xmath117 + @xmath124 & exponential decay time of @xmath120 at @xmath119 + @xmath125 & stellar mass change between @xmath82 and @xmath77 + @xmath126 & @xmath116 mass change between @xmath82 and @xmath77 + @xmath127 & @xmath118 mass change between @xmath82 and @xmath77 + @xmath128 & @xmath120 mass between @xmath82 and @xmath77 + @xmath129 & effective radius of young stars formed within the past 100myr + @xmath130 & environmental temperature within physical radius of @xmath104kpc + @xmath109 & environmental entropy within physical radius of @xmath104kpc + @xmath99 & environmental pressure within physical radius of @xmath104kpc + @xmath131 & environmental overdensity within comoving radius of @xmath132mpc + @xmath133 & distance to primary galaxy in units of virial radius of primary galaxy + @xmath96 & time duration from @xmath82 to @xmath77 + @xmath97 & time spent in green valley + @xmath98 & time spent in red sequence + @xmath134 & halo mass of primary galaxy + @xmath135 & stellar mass ratio of satellite to primary galaxy the galaxy formation simulation in a cosmological setting used here includes sophisticated physical treatment , ultra - high resolution and a very large galaxy sample to statistically address cosmological and astrophysical questions . while this simulation represents the current state - of - the - art in these respects , feedback from sf
is still far from being treated from first principles .
thus , it is necessary that we validate the feedback prescription empirically . in @xcite
we presented an examination of the damped lyman alpha systems ( dlas ) and found that the simulations , for the first time , are able to match all observed properties of dlas , including abundance , size , metallicity and kinematics .
in particular , the metal distribution in and around galaxies over a wide range of redshift ( @xmath136 ) is shown to be in excellent agreement with observations @xcite .
the scales probed by dlas range from stellar disks at low redshift to about one half of the virial radius at high redshift . in @xcite
we further show that the properties of o vi absorption lines at low redshift , including their abundance , doppler - column density distribution , temperature range , metallicity and coincidence between o vii and o vi lines , are all in good agreement with observations @xcite .
the agreement between simulations and observations with respect to o vi lines is recently shown to extend to the correlation between galaxies and o vi lines , the relative incidence ratio of o vi around red to blue galaxies , the amount of oxygen mass around red and blue galaxies as well as cold gas around red galaxies @xcite .
in addition to agreements with observations with respect to circumgalactic and intergalactic medium , we find that our simulations are able to match the global sfr history ( the madau plot ) and galaxy evolution @xcite , the luminosity function of galaxies at high @xcite and low redshift @xcite , and the galaxy color distribution @xcite , within observational uncertainties .
in @xcite we show that our simulations reproduce many trends in the global evolution of galaxies and various manifestations of the cosmic downsizing phenomenon .
specifically , our simulations show that , at any redshift , the specific star formation rate of galaxies , on average , correlates negatively with galaxy stellar mass , which seems to be the primary physical process for driving the cosmic downsizing phenomena observed . smoothed particle hydrodynamic ( sph ) simulations and semi - analytic methods , in comparison , appear to produce a positive correlation between the specific star formation rate of galaxies and galaxy stellar mass , which is opposite to what we find ( e.g. , * ? ? ?
these broad agreements between our simulations and observations indicate that , among others , our treatment of feedback processes from sf , i.e. , the transport of metals and energy from galaxies , from sf sites to megaparsec scale ( i.e. , from interstellar to intergalactic medium ) are realistically modeled as a function of distance and environment , at least in a statistical sense , and it is meaningful to employ our simulated galaxies , circumgalactic and intergalactic medium for understanding physical processes and for confrontations with other , independent observations .
-1.0 cm -0.5 cm in order to determine what galaxies in our simulations to use in our subsequent analysis , we make an empirical numerical convergence test .
top - right panel in figure [ fig : mfr50test ] shows comparisons between galaxies of two simulations at @xmath44 with different resolutions for the luminosity functions in rest - frame @xmath71 and @xmath72 bands .
the fiducial simulation has a resolution of @xmath137pc / h and an identical comparison run has four times better resolution of @xmath43pc / h .
we are not able to make comparisons at redshift substantially lower than @xmath44 at this time . in any case
, we expect that the comparison at @xmath44 is a more stringent test , because the resolution effect is likely more severe at higher redshift than at lower redshift in a hierarchical growth model where galaxies become increasingly larger with time .
the comparisons are best done statistically , because not all individual galaxies can be identified at a one - to - one basis due to resolution - dependent star formation and merging histories . comparisons with respect to other measures , such as stellar mass function , sfr , etc , give comparable convergence . based on results shown ,
we decide to place a lower stellar mass limit of @xmath138 , which is more than @xmath139 complete for almost all relevant quantities , to the extent that we are able to make statistical comparisons between these two runs with respect to the global properties of galaxies ( stellar mass , luminosity , sfr , ssfr , etc ) . in terms of checking the validity and applicability of the simulations , we also make comparisons for the galaxy cumulative mass function at @xmath29 with observations in the top - left panel of figure [ fig : mfr50test ] .
we see that the simulated galaxies have a higher abundance than observed by a factor of @xmath140 in the low mass end and the difference increases towards higher mass end .
this difference is expected , because the simulation volume is an overdense region that has a higher galaxy density overall and progressively higher densities for more rare , higher mass galaxies .
this difference is also borne out in the comparisons between simulated and observed rest - frame @xmath71 band galaxy luminosity function at @xmath2 and @xmath141 band luminosity function at @xmath29 ( lower - left panel of figure [ fig : mfr50test ] ) .
the difference between simulation and observations is also seen to increase with decreasing redshift , as expected .
we note that at the high luminosity end dust effects may become very important , causing the apparent discrepancies between simulations and observations larger than they actually are .
overall , our simulation serves our purposes well for two reasons .
first , it contains massive group / cluster systems as well as field regions that allow us to probe environment effects with enough leverage / range .
second , the faint end slope of the galaxy population matches observations well , albeit with a larger overall amplitude ; as a result , relative , comparative statements that we will make across the stellar mass range @xmath142 and @xmath143 is approximately valid even though the absolute number of galaxies may be off .
as will become clear later , during quenching by ram - pressure stripping and gas starvation , star formation often continues in the central @xmath144kpc regions of galaxies .
therefore , it is useful to have an estimate of the effective resolutions of simulations empirically as follows
. let us denote the actual resolution , given the maximum refinement resolution of @xmath145 , as @xmath146 , where @xmath147 is a contant that is expected to be larger than but of order unity for a mesh code .
we then assume that the computed effective stellar radius ( enclosing @xmath148 of stellar mass ) of any galaxy is @xmath149 where @xmath150 is the actual effective radius from simulation and @xmath151 is the true effective radius if one had infinite resolution .
we have two equations of eq [ eq : re ] for the two simulations with different resolutions ( @xmath152pc / h and @xmath153pc / h ) and two unknowns ( constant @xmath147 and vector @xmath151 ) .
we solve for @xmath147 by requiring the distributions of @xmath151 derived from the two simulations have the closest agreement ; @xmath154 is found .
the bottom - right panel of figure [ fig : mfr50test ] shows the distributions of derived @xmath151 for all galaxies with stellar mass @xmath155 from the two simulations with @xmath154 .
the effective resolutions ( @xmath156 ) for the two runs , @xmath157pc and @xmath158pc , are indicated by the vertical dashed lines corresponding to the histograms with the same colors . as we will show later
, the central gas within @xmath144kpc is largely retained by the galaxy that is being subject to ram - pressure stripping . clearly , our resolution is adequate to resolve gas distribution , where it is demanded to address ram - pressure stripping physics . on the other hand ,
consistent with other panels of figure [ fig : mfr50test ] , we see that at the stellar mass cut of @xmath138 , while the majority of galaxies are effectively resolved with respect to their stellar distribution , a significant fraction of galaxies in our selected sample with stellar mass close to @xmath138 is under - resolved in the central regions .
we expect that , although our simulation may overestimate the ram - pressure stripping effects for the most inner regions for some of the smaller galaxies , the numerical effect on overall ram - pressure stripping process due to this limitation is likely small .
this is because the amount of gas within our resolution limit ( @xmath159pc ) is small compared to the total mass of gas within ( say ) @xmath144kpc , within which , we will show later , gas is little affected by ram - pressure stripping . in other words , an under - resolution should not materially impact the overall the amount of ram - pressure stripped gas , since the gas in the central region much larger than our resolution is not stripped anyway .
-0.0 cm -0.8 cm the main goal of this paper is to identify and understand the physical processes that produce the observed galaxy color bimodality . we first examine if our simulations actually reproduce the observed color bimodality .
figure [ fig : bimodal ] shows @xmath160 color distributions of simulated galaxies in three stellar mass ranges , @xmath161 ( black ) , @xmath162 ( cyan ) and @xmath163 ( magenta ) , at @xmath2 .
the g - r color distributions show clear bimodalities for all three subsets of galaxies , with the red peak becoming more prominent for less luminous galaxies at @xmath2 , consistent with recent observations ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
we also caution that one should not overstate the success in this regard for two reasons .
first , on the simulation side , since our simulation volume does not necessarily represent an average " volume of the universe , a direct comparison to observations would be difficult .
second , observations at high redshift ( i.e. , @xmath164 ) are perhaps less complete than at low redshift , and identification of low mass ( and especially low surface brightness ) galaxies , in particular those that are satellite galaxies and red , may be challenged at present ( e.g. , * ? ? ?
our main purpose is to make a comparative study of galaxies of different types in the simulation and to understand how blue galaxies turn red .
it is intriguing to note that there is no lack of red dwarf galaxies .
while a direct comparison to observations with respect to abundant red dwarf galaxies can not be made at @xmath2 , future observations may be able to check this .
since our simulation does not include agn mechanical feedback , this suggests that the bimodal nature of galaxy colors does not necessarily require agn feedback for galaxies in the mass ranges examined .
this finding is in agreement with @xcite , who find that agn feedback is not an essential ingredient for producing quiescent , red elliptical galaxies in galaxy groups .
while sf feedback is included in our simulation , our subsequent analysis shows that environmental effects play the dominant role in driving galaxy color evolution and consequently color bimodality .
our results do not , however , exclude the possibility that agn feedback may play an important role in regulating larger , central galaxies , such as cd galaxies at the centers of rich clusters of galaxies , for which we do not have a sufficient sample to make a statistical statement .
our earlier comparison between simulated luminosity functions of galaxies at @xmath32 and sdss observations indicates that some additional feedback , likely in the way of agn , may be required to suppress star formation in the most massive galaxies @xcite .
most of our results shown are presented through a variety of comparisons of the dependencies of galaxies of different types on a set of environmental variables , to learn how galaxies change color .
we organize our analysis in an approximately chronological order . in 3.1 we focus on processes around the quenching " time , @xmath82 , followed by the ensuing period of gas starvation in hot environment in 3.2 . in 3.3
we discuss stellar mass growth , evolution of stellar mass function of red galaxies and present galaxy color migration tracks .
3.4 gives an example of consequences of the color migration picture - galaxy age - mass relation .
we present observable environmental dependence of galaxy makeup at @xmath2 in 3.5 .
-0.0 cm -0.5 cm figure [ fig : tauminsfq ] shows the quenching time scale @xmath121 ( star formation rate exponential decay time ) against four environmental variables at the quenching time @xmath82 : ram - pressure @xmath99 , environmental entropy @xmath109 , distance to primary galaxy @xmath165 and environmental overdensity @xmath131 .
it is useful to make clear some nomenclature here .
we have used the distance to the primary galaxy , @xmath165 , as an environment variable , which runs from zero to values significant above unity .
this is merely saying that any galaxy ( except the most massive galaxy in the simulation ) can find a larger galaxy at some distance , not necessarily at @xmath166 .
the definition of satellite galaxies " is reserved only for those galaxies with @xmath166 , shown clearly as red circles in the low - left panel of figure [ fig : tauminsfq ] .
the black circles , labeled as centrals " are galaxies with @xmath167 , i.e. , those that are not not satellite galaxies " .
the observation that galaxies are being quenched at all radii - @xmath167 as well as @xmath168 - indicates that the most likey physical mechanism for the onset of quenching is ram - pressure .
tidal stripping is not expected to be effective at removing gas ( or stars ) at @xmath167 ( see 2.3 for a discussion ) .
the fact that @xmath121 decreases with increasing @xmath99 is self - consistent with ram - pressure being responsible for the onset of quenching .
the outcome that @xmath121 only very weakly anti - correlates with @xmath99 indicates that the onset of quenching is some threshold " event , which presumably occurs when the ram - pressure exceeds gravitational restoring force ( i.e. , the threshold ) , thus strongly re - enforcing the observation that ram - pressure is largely responsible for the onset of quenching .
threshold " type mechanism fits nicely with the fact that the dispersion of @xmath121 at a given @xmath99 is substantially larger than the correlation trend , because galaxies that cross the threshold " are expected to depend on very inhomogeneous internal properties among galaxies ( see figure [ fig : tauminsigma ] below ) .
the weak anti - correlation between @xmath121 and @xmath131 stems from a broad positive correlation between @xmath99 and @xmath131 .
the fact that there is no discernible correlation between @xmath121 and @xmath109 indicates that the onset of quenching is _ not _ initiated by gas starvation .
the most noticeable contrast to the weak trends noted above is the difference between satellite galaxies ( at @xmath168 ) and central galaxies ( at @xmath167 ) , in that @xmath121 of the former is lower than that of the latter by a factor of @xmath169 .
this is naturally explained as follows .
first , at @xmath168 ram - pressure stripping and tidal stripping operate in tandem to accelerate the gas removal process , whereas at @xmath167 ram - pressure stripping operates alone " to remove gas on somewhat longer time scales .
second , at @xmath167 ram - pressure stripping is , on average , less strong than at @xmath168 .
possible internal variables that affect the effectiveness of ram - pressure stripping include the relative orientation of the normal of the gas disk and the motion vector , rotation velocity of the gas disk , whether gas disk spiral arms are trailing or not at time of ram - pressure stripping , gas surface density amplitude and profile , dark matter halo density profile . as an obvious example , galaxies that have their motion vector and disk normal aligned are likely to have maximum ram - pressure stripping effect , everything else being equal . in the other extreme when the two vectors are perpendicular to each other
, the ram - pressure stripping effect may be minimized .
needless to say , given many factors involved , the onset of ram - pressure stripping effect will be multi - variant .
we elaborate on the multi - variant nature of ram - pressure stripping with one example .
the top - panel of figure [ fig : tauminsigma ] shows @xmath121 as a function of the stellar surface density @xmath170 within the effective stellar radius @xmath171 .
we see a significant positive correlation between @xmath121 and @xmath170 in the sense that it takes longer to ram - pressure - remove cold gas with higher central surface density ( hence higher gravitational restoring force ) galaxies . while this positive correlation between @xmath121 and @xmath170 is consistent with observational indications ( e.g. , * ? ? ?
* ) , the underlying physical origin is in a sense subtle . since ram - pressure stripping is a
threshold " event , as noted earlier , when ram - pressure force just exceeds the internal gravitational restoring force , one would have expected that a high surface density would yield a shorter dynamic time hence a shorter @xmath121 .
this is in fact an incorrect interpretation .
rather , the gas in the central regions where @xmath170 is measured is immune to ram - pressure stripping in the vast majority of cases ( see figure ( [ fig : rat ] ) below ) . instead , a higher @xmath170 translates , on average , to a larger scale where gas is removed , which has a longer dynamic time hence a longer @xmath121 .
-0.0 cm -0.5 cm taken together , we conclude that , while a high ram - pressure provides the conditions for ram - pressure stripping to take effect , the effectiveness or timescale for gas removal by ram - pressure stripping also depend on the internal structure of galaxies .
it is very interesting to note that , unlike between @xmath82 and @xmath170 , @xmath96 ( the time interval between the onset of quenching @xmath82 and the time when the galaxy turns red ) and @xmath170 shown in the bottom - panel of figure [ fig : tauminsigma ] , if anything , is weakly anti - correlated .
we attribute this outcome to the phenomenon that galaxies with higher central surface density have a shorter time scale for consuming the existing cold gas hence , once the overall cold gas reservoir is removed .
this explanation will be elaborated more later .
-0.0 cm -0.5 cm -0.0 cm -0.0 cm we have made the case above that ram - pressure stripping is primarily responsible for the onset of quenching process based on evidence on the dependence of exponential decay time of sfr at the onset of quenching on environment variables .
we now make a direct comparison between @xmath82 and @xmath107 .
figure ( [ fig : tqtram ] ) shows the histograms of @xmath172 .
we see that the time difference between the two is centered around zero , indicating a casual connection between the onset of sfr quenching and the rapid rise of ram - pressure .
the width of the distribution of a few hundred myrs reflects the fact that the exact strength of ram - pressure stripping required to dislodge the gas varies greatly , depending on many variables as discussed above .
this is as yet the strongest supporting evidence for ram - pressure stripping being responsible for the onset of quenching , especially considering the conjunctional evidence that the onset of quenching could occur outside the virial radius of a larger neighboring galaxy where tidal stripping is expected to be less effective and yet ram - pressure is expected to become important .
-0.0 cm -1.7 cm -2.15 cm -0.5 cm evidence so far supports the notion that ram - pressure stripping is the initial driver for the decline of sfr in galaxies that are en route to the red sequence .
the immediate question is then : what region in galaxies does ram - pressure stripping affect ? to answer this question , we need to compare the amount of cold gas available at @xmath82 with the amount of star formation that ocurrs subsequently . we compute the following ratios : the ratios of the amount of stars formed during the time interval from @xmath82 to the time the galaxy turns red ( @xmath77 ) to the difference between the amount of cold ( @xmath173k ) gas at @xmath82 and @xmath77 , denoted as ( @xmath174 , @xmath175 , @xmath176 ) within three radii @xmath177kpc .
the minus signs are intended to make the ratios positive , since stellar mass typically increase with time , whereas the cold gas mass for galaxies being quenched decreases .
note that @xmath125 could be negative .
we set a floor value to the above ratios at @xmath26 .
figure ( [ fig : rat ] ) shows the distributions of @xmath178 , where @xmath179 .
we see that for @xmath180kpc the peak of the distribution ( red histograms ) for all stellar masses is at @xmath181 , typically in the range of @xmath182 , with the vast majority of cases at @xmath183 . for a larger radius @xmath184kpc the distribution ( green histograms ) for all stellar masses
is now peaked at @xmath185 with about 50% at @xmath186 , while for a still larger radius @xmath187kpc the distribution ( green histograms ) it is now shifted to @xmath188 and more than 50% at less than @xmath189 for galaxies of stellar mass @xmath190 , respectively .
this is unambiguous evidence that ram - pressure stripping removes the majority of cold gas on scales @xmath191kpc , while the cold gas within @xmath144kpc is unaffected by ram - pressure stripping and consumed by subsequent in situ star formation .
it is noted that the lack of effect on the cold gas at @xmath192kpc by ram - pressure stripping seems universal , the gas removal by ram - pressure stripping at larger radii @xmath193kpc varies substantially from galaxy to galaxy , which we argue is consistent with the large variations of @xmath121 seen in figure [ fig : tauminsfq ] .
we thus conclude that there is continued nuclear sf in the quenching phase .
@xcite , based on a smaller sample of simulated galaxies that form a group of galaxies with a spatial resolution of @xmath104pc ( compared to @xmath194pc here ) , find that in situ star formation is responsible for consuming a substantial fraction of the residual gas on small scales after gas accretion is stopped subsequent to the infall , consistent with our results .
this outside - in ram - pressure stripping picture and continuous sf in the inner region that emerges from the above analysis has important implications and observable consequences , consistent with the latest observations ( e.g. , * ? ? ?
we quantify how centrally concentrated the star formation is at the outset of sf quenching in figure ( [ fig : r50his ] ) , in part to assess our ability to resolve sf during the quenching phase .
the left panel shows the distribution of the effective radius of stars formed in the last @xmath195myrs prior to @xmath82 , denoted as @xmath196 , for galaxies in two stellar mass ranges .
the right panel shows the distribution of the ratio of the decline of @xmath196 with respect to the decline of sfr , @xmath197 at @xmath82 .
it is evident from figure ( [ fig : r50his ] ) that more massive galaxies tend to have larger @xmath196 , as expected .
it is also evident that the recent formation for the vast majority of galaxies occurs within a radius of a few kiloparsecs .
it is noted that ongoing sf in a significant fraction of galaxies with stellar masses @xmath198 is under - resolved , as indicated by the vertical dot - dashed line in the left panel . however , none of our subsequent conclusions would be much altered by this numerical effect , because ( 1 ) all of our conclusions appear to be universal across the stellar mass ranges and ( 2 ) the inner region of @xmath144kpc is not much affected by ram - pressure stripping anyway ( thus under - resolving a small central fraction within @xmath144kpc does not affect the overall ram - pressure stripping effects ) .
what is interesting is that more than 50% of galaxies in both stellar mass ranges have negative values of @xmath197 at @xmath82 , indicating that , when the sfr decreases in the quenching phase , star formation proceeds at progressively larger radii in the central region .
this result , while maybe somewhat counter - intuitive , is physically understandable .
we attribute this inside - out star formation picture to the star formation rate surface density being a superlinear function of gas surface density in the kennicutt - schmidt @xcite law .
the picture goes as follows : when gas supply from large scales ( @xmath199kpc ) is cut off and under the assumption that gas in the central region does not re - distribute radially , the sfr diminishes faster with decreasing radius in the central region where sf occurs , causing the effective sf radius to increase with time and star formation rate to decline faster than cold gas content , while the overall sfr is declining . in summary ,
ram - pressure stripping is ineffective in removing cold gas that is already present on scales of @xmath5kpc but most effective in removing less dense gas on larger scales of @xmath191kpc .
the chief role played by ram - pressure stripping appears to disconnect galaxies from their cold reservoir on scales that are much larger the typical stellar radii .
the time scale in question is then on the order of the dynamical time of galaxies at close to the virial radius .
the previous subsection details some of the effects on galaxies being quenched due to gas removal by ram - pressure stripping ( in conjunction with other hydrodynamical processes ) along with consumption by concurrent sf .
our attention is now turned to the subsequent evolution .
figure [ fig : taubrsfq ] plots @xmath96 against four environmental variables at @xmath82 . from all panels
we consistently see the expected trends : the time interval @xmath96 from onset of quenching @xmath82 to turning red @xmath77 , on average , decreases with increasing environmental pressure , increasing environmental entropy , increasing environmental overdensity and decreasing distance to the primary galaxy .
while there is a discernible difference in @xmath96 between satellite galaxies and central galaxies , the difference is substantially smaller than that in the initial exponential decay time scale of sfr @xmath121 ( see figure [ fig : tauminsfq ] ) .
this observation makes it clear that the onset of quenching initiated by ram - pressure stripping does not determine the overall duration of quenching .
since all the environment variables used tend to broadly correlate with one another - higher density regions tend to have higher temperatures , higher gas entropy and higher pressure - it is not surprising that we see @xmath96 are correlated with all of them in the expected sense .
earlier we have shown that @xmath96 is weakly anti - correlated with the stellar surface density at @xmath200 , @xmath170 ( see bottom - panel of figure [ fig : tauminsigma ] ) .
this suggests that the overall duration from onset of quenching to turning red is not a matter of a galaxy s ability to hold on to its existing cold gas but rather the extent of the external gas supply condition , i.e. , environment .
this hypothesis is significantly affirmed by noticing that the strongest anti - correlation is found between @xmath96 and @xmath109 , among all environment variables examined .
thus , we conclude , given available evidence , that the eventual
push " of galaxies into the red sequence is not as a spectacular event as the initial onset of quenching that is triggered by a cutoff of large - scale gas supply due to ram - pressure stripping , and is essentially the process of gas starvation , when the galaxy has entered a low cold gas density and/or high temperature and/or high velocity dispersion environment .
-0.5 cm -2.55 cm -0.5 cm we present distributions of @xmath96 in figure [ fig : tbrhis ] .
the top - left panel shows the distribution of @xmath96 for satellite galaxies ( those with @xmath201 ) , grouped into three primary halo mass ranges : @xmath202 ( black ) , @xmath203 ( green ) , @xmath204 ( red ) ; the medians of the distributions are ( 1.2,1.3,1.2)gyr , respectively .
the top - right panel shows the distribution of @xmath96 for satellite galaxies grouped into three ranges of the ratio of satellite to cental stellar mass : @xmath205 ( black ) , @xmath206 ( green ) , @xmath207 ( red ) ; the medians of the distributions of the three groups are nearly identical at @xmath208gyr .
the bottom - left panel shows the distribution of @xmath96 for primary galaxies ( those with @xmath209 ) , grouped into two halo mass ranges : @xmath210 ( black ) , and @xmath202 ( green ) .
we see that the medians of the distributions are @xmath211gyr for both mass ranges .
the bottom - right panel plots the distribution of all satellite galaxies and all central galaxies , along with a simple gaussian fit to the combined set .
a look of the bottom - right panel of figure [ fig : tbrhis ] suggests that there is practically no difference between the two distributions . at first sight
, this may seem incomprehensible .
a closer examination reveals the underlying physics .
-1.0 cm -0.5 cm figure [ fig : dissfq ] shows the distributions of @xmath165 at @xmath82 for satellite ( red ) and central ( black ) red galaxies at @xmath2 . while it is not a surprise that the vast majority of the satellite galaxies at @xmath2 have their onset of quenching taking place at @xmath212 at @xmath82 , it is evident that the same appears to be true for the central galaxies at @xmath2 .
this observation supports the picture that both satellite and central red galaxies at @xmath2 have been subject to similar environment effects that turn them red .
it is noted again that this statement that red central galaxies have been subject to similar processes as the red satellite galaxies has been quantitatively confirmed in figure [ fig : tbrhis ] .
the suggestion by @xcite that some central galaxies are ejected satellite galaxies is consistent with our findings here .
our study thus clearly indicates that one should not confuse red central galaxies with their being quenched by processes other than environment .
in fact , all available evidence suggests that it is environment quenching that plays the dominant role for the vast majority of galaxies that turn red , whether they become satellite galaxies at @xmath2 or not .
@xcite , using a much smaller sample of simulated galaxies that form a group of galaxies , find that quenching of gas accretion starts at a few virial radii from the group center , in good agreement with our results .
it is seen in figure [ fig : dissfq ] that only about 20% of the onset of galaxy quenching occurs as satellites , i.e. , within the virial radius of a larger galaxy , consistent with conclusion derived by others ( e.g. , * ? ? ?
* ) . in the bottom - right panel of figure
[ fig : tbrhis ] we provide an approximate fit to the distribution of @xmath96 for all quenched galaxies normalized to galaxies at @xmath2 as @xmath213},\ ] ] where @xmath96 and @xmath214 are in gyr and @xmath215 .
the adopted @xmath216 dependence on @xmath217 is merely an estimate of the time scale , had it scaled with redshift proportional to the dynamical time of the universe .
one is cautioned not to apply this literally .
nevertheless , it is likely that the median quenching time at lower redshift is longer than @xmath218gyr at @xmath2 , perhaps in the range of @xmath219gyr .
incidentally , this estimated quenching time , if extrapolated to @xmath32 , is consistent with theoretical interpretation of observational data in semi - analytic modeling or n - body simulations ( e.g. , * ? ? ?
* ; * ? ? ?
* ) . in semi - analytic modeling ( e.g. , * ? ?
* ) , the quenching time is often taken to be a delta function .
in other words , the satellite quenching process is assumed to be uniform , independent of the internal and external properties of the satellites .
our simulation results ( see eq [ eq : tqr ] ) indicate that such a simplistic approach is not well motivated physically .
we suggest that , if a spread in quenching time is introduced in the semi - analytic modeling , an improvement on the agreement between predictions based on semi - analytic modeling and observations may result in . in summary , we find that , within the environmental sphere of influence , galaxies are disconnected with their large - scale cold gas supply by ram - pressure stripping , and subsequently lack of gas cooling and/or accretion in high velocity environment ensures a prolonged period of gas starvation that ultimately turns galaxies red .
this applies to satellite galaxies as well as the vast majority of apparent " central red galaxies .
the dominance of environment quenching that is found in ab initio cosmological simulations here is in accord with observations ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) . on its way to the red sequence
, a galaxy has to pass through the green valley .
do all galaxies in the green valley migrate to the red sequence ?
we examine the entire population of green galaxies in the redshift range @xmath220 . tracing these green galaxies to @xmath2
, we find that for galaxies with stellar masses greater than ( @xmath142 , @xmath221 , @xmath222 , respectively , ( 40% , 40% , 48% ) of galaxies in the green valley at @xmath220 , do not become red galaxies by @xmath2 .
while this is an important prediction of our simulations , we do not provide more information on how one might tell apart these two different population of galaxies in the green valley , except to point out that attempts to identify galaxies in the green valley as progenitors of red galaxies may generate some confusion .
we examine the distributions ( not shown ) of the time that red galaxies spent in the green valley , @xmath97 , en route to the red sequence .
the trends with respect to @xmath223 and @xmath224 seen are similar to those seen in figure [ fig : tbrhis ] .
no significant differentiation among halo masses of central galaxies is visible , once again supportive of environment quenching .
overall , one may summarize the results in three points .
first , @xmath97 is almost universal , independent of being satellites or not , the mass , or the ratio of masses .
second , the range @xmath225gyr appears to enclose most of the galaxies , although there is a significant tail towards the high end for satellites in low mass central halos .
third , comparing @xmath226gyr to the interval from onset of quenching to the time galaxy turning red of @xmath227gyr , it indicates that , from the onset of quenching to turning red , typical galaxies spend about 25% of the time in the green valley .
cm -0.5 cm let us now examine the migration tracks of galaxies that eventually enter the red sequence .
figure [ fig : migrate ] shows the color - stellar mass diagram for 30 semi - randomly selected red galaxies .
it is striking that the color evolution in the green valley and red sequence is mostly vertical , i.e. , not accompanied by significant change in stellar mass .
this means that the stellar mass growth of most galaxies must occur in the blue cloud .
one can see easily that the blue tracks are mostly moving from lower left to upper right with time for @xmath228 , indicating that galaxies grow when in the blue cloud . in the blue cloud
it is seen that there are occasional horizontal tracks , representing mergers that maintain overall color .
these are mergers that do not result in red galaxies .
the examples of these include the two most massive galaxies in the plot with final stellar masses of @xmath229 , where there is a major binary merger of @xmath230 at @xmath231 .
there are also cases where the tracks temporarily go from north - west to south - east , indicating significant / major mergers that trigger starbursts that render the remnant galaxies bluer .
this anecdotal evidence that galaxies do not significantly grow mass in the red sequence will be confirmed below quantitatively .
@xcite , using a small sample of simulated galaxies that form a group of galaxies , find that mergers and significant mass growth in galaxies occur , prior to their entering groupd environment , consistent with the findings here .
thus , this skyrockets " diagram of color - stellar mass evolution in figure [ fig : migrate ] turns out to be a fair representation of typical tracks of galaxies that become red galaxies .
-0.1 cm -2.1 cm -2.0 cm
-0.5 cm we address the stellar mass growth of red galaxies quantitatively in two different ways .
the left panel of figure [ fig : redmass ] shows the histogram of the ratio of stellar mass of red galaxies at @xmath2 to their progenitor s stellar mass at the onset of quenching @xmath82 .
we see that the overall stellar mass growth of red galaxies since the onset of quenching is relatively moderate , with the vast majority of galaxies gaining less than 30% of their stellar mass during this period , consistent with observations ( e.g. , * ? ? ?
* ; * ? ? ?
there is a non - negligible fraction of galaxies that experience a decline of stellar mass , due to tidal interactions and collisions .
there is @xmath232 of red galaxies that gain more than @xmath233 of their stellar mass during this period , possibly due to mergers and accretion of satellite galaxies .
we do not address red galaxies more massive than @xmath234 because of lack of a statistically significant red sample . since these larger galaxies tend to reside at the centers of groups and clusters ,
there is a larger probability that agn feedback may play a significant role in them .
empirical evidence suggests that radio jets get extinguished in the near vicinity of the central galaxies in groups / clusters ( e.g. , * ? ? ?
* ) , in sharp contrast to agns in isolated galaxies where jets , seen as large radio lobes , appear to deposit most of their energy on scales much larger than the star formation regions .
thus , agn feedback in the central massive galaxies in clusters / groups may be energetically important to have a major effect on gas cooling and star formation in them ( e.g. , * ? ? ?
thus , our neglect of agn feedback in the simulation cautions us to not draw any definitive conclusion with respect to this special class of galaxies at this time .
the stellar mass growth of individual red galaxies shown in the left panel of figure [ fig : redmass ] contains very useful information .
however , it does not address a related but separate question : how does the stellar mass function of red galaxies evolve with redshift ?
we address this question here .
we compute the cumulative stellar mass function of red galaxies at @xmath2 and @xmath235 , separately , and show them in the right panel of figure [ fig : redmass ] .
we see that for red galaxies with stellar masses greater than @xmath236 , when matched in abundance , the stellar masses grow a factor of @xmath237 from @xmath235 to @xmath2 , much larger than @xmath238 ( for about 75% of galaxies ) seen in the left panel of figure [ fig : redmass ] .
we refrain from making a direct comparison to observations in this case , because our limited simulation volume is highly biased with respect to the massive end of the mass function .
we strict ourselves to a comparative analysis of galaxies in our simulation volume and ask the question of how red galaxies in our simulation volume grow with time . the most important point to note
is that this apparent growth of stellar mass of red galaxies based on abundance matching could not be due to growth of individual red galaxies in the red sequence , since the actual stellar mass increase since the onset of quenching is moderate , @xmath239 typically , seen in the left panel of figure [ fig : redmass ] .
physically , this suggests that dry mergers do not play a major role in the apparent " stellar mass growth of red galaxies , consistent with observations ( e.g. , * ? ? ?
rather , galaxies grow their stellar mass when they are still in the blue cloud , illustrated in figure [ fig : migrate ] .
a physical picture of galaxy color migration emerges based on our results . _
the migration from the blue cloud to the red sequence proceeds in a staggered fashion : stellar masses of individual galaxies continuously grow , predominantly in the blue cloud , and blue galaxies over the entire mass range continuously migrate into the red sequence over time .
_ galaxies migrate from the blue cloud to the red sequence almost vertically in the usual color - magnitude diagram ( see figure [ fig : migrate ] ) . for simplicity we will call this type of color migration vertical tracks " , which correspond most closely to
b tracks " proposed by @xcite , with the growth since the onset of quenching being moderate ( @xmath240 ) .
the vertical tracks found have many implications on observables .
the first question one asks is this : if galaxies follow the vertical tracks , is the galaxy age - mass relation consistent with observations ?
we address this question in this subsection .
figure [ fig : redmstarage2panel ] shows a scatter plot of red galaxies in the stellar mass @xmath241-mean galaxy formation time @xmath242 plane at @xmath2 ( top ) and @xmath29 ( bottom ) , where @xmath242 is stellar formation time , not lookback time .
the red galaxies are subdivided into two groups : centrals ( black circles ) and satellites ( red circles ) .
for the purpose of comparison to observations , we only show galaxies with high surface brightness of @xmath243 mag arcsec@xmath244 ( e.g. , * ? ? ? * ) .
several interesting results can be learned .
-0.cm -0.5 cm first , no systematic difference between satellite and central galaxies is visible , supporting earlier findings that there is no appreciable differences between satellites and centrals with respect to duration from quenching to turning red @xmath96 ( figure [ fig : tbrhis ] ) .
second , at any given redshift , the brightest red galaxies are relatively old " ( but not necessarily the oldest ) , of ages of several billion years ( age = @xmath245 and @xmath246gyr for @xmath247 ) , consistent with observations .
third , at stellar masses greater than @xmath248 red galaxies have a nearly uniform mean age ; the age spread at a given stellar mass of @xmath249gyrs is consistent with observations ( e.g. , * ? ? ?
fourth , fainter red galaxies are younger than brighter red galaxies in the mass range @xmath250 ; we see that the age difference between the two ends of the mass range is @xmath251gyr and @xmath252gyr , respectively , at @xmath2 and @xmath29 , suggesting a steepening with decreasing redshift of the age difference between galaxies of different masses in the red sequence .
@xcite find an age difference between the faint and bright ends of red sequence galaxies of @xmath169gyr at @xmath253 , in excellent agreement with our results .
the physical origin for the steepening with decreasing redshift of the age difference between galaxies of different masses in the red sequence is traceable to the steepening of specific sfr with stellar mass with decreasing redshift that is , in a fundamental way , related to the cosmic downsizing phenomenon @xcite .
-0.5 cm it is interesting to note that , in figure [ fig : redmstarage2panel ] , scatters notwithstanding , there appears to be a critical stellar mass of @xmath254 , above which the age ( or formation time ) of red galaxies flattens out to a constant value . at least for the redshift range that we have examined , @xmath255 , this critical stellar mass appears to be redshift independent . at still higher redshift
we do not have enough statistics to see if this critical mass remains the same .
this critical mass is tantalizingly close to the division mass of @xmath256 discovered by @xcite at low redshift , which appears to demarcate a number of interesting trends in galaxy properties .
this physical origin of this mass is unclear and deferred to a future study .
given the
vertical tracks " , i.e. , lack of significant stellar mass growth subsequently to quenching , one may ask this : is the age - mass relation of red galaxies inherited from their blue progenitors ? we will now address this question . to select progenitors of red galaxies at @xmath2 , we note that the majority of galaxies that turn red by @xmath2 have @xmath257gyr .
thus , we choose galaxies in the redshift range @xmath258 ( 8 snapshots with @xmath259 ) , where the hubble time differences between @xmath2 and @xmath260 and @xmath261 are @xmath262gyr , respectively , enclosing the vast majority of blue progenitors of red galaxies at @xmath2 near the onset of quenching .
we separate the blue galaxies into two groups : one group contains the blue progenitors of @xmath2 red galaxies and the other group other blue galaxies that have not turned into red galaxies by @xmath2 .
figure [ fig : rbtf ] shows the stellar mass @xmath241-mean galaxy formation time @xmath242 scatter plot for blue galaxies at @xmath258 that are progenitors of red galaxies at @xmath2 ( top left ) and those that do not become red galaxies ( top right ) .
each small group of mostly linearly aligned circles is one galaxy that appears multiple times ( maximum is 8) . within the scatters we see that the green dashed line , borrowed from figure [ fig : redmstarage2panel ] , provides a good match to the near constant age at the high mass end for the progenitors of red galaxies .
the magenta dots , borrowed from figure [ fig : redmstarage2panel ] , match well the trend for the blue dots in the mass range @xmath250 .
these results are fully consistent with our initial expectation based on the observation ( of our simulation ) of two physical processes : ( 1 ) that stellar mass growth is moderate during @xmath96 hence evolution during @xmath96 does not significantly alter the mean star formation time of each galaxy , ( 2 ) less massive forming galaxies have higher ssfr than massive galaxies , causing a steepening of the age - mass relation at the low mass end .
this explains the physical origin of the age - mass relation seen in figure [ fig : redmstarage2panel ] .
it is prudent to make sure that these important general trends seen in the simulation are robust . in the bottom two panels of figure [ fig : rbtf ]
we make a comparison between blue galaxies of two simulations with different resolutions , at @xmath44 .
the bottom - left panel is from the fiducial simulation with a resolution of @xmath137pc / h and the bottom - right panel is from an identical simulation with four times better resolution of @xmath43pc / h .
we see that both the age - mass trend at low mass end and the near constancy of stellar age at the high mass end are shared by the two simulations , suggesting that results from our fiducial simulation are sufficiently converged for the general trends presented at the level of concerned accuracies . a comparison between the top left and top right panels in figure [ fig : rbtf ] makes it clear that the age - mass relation of the blue progenitors of red galaxies at quenching is , to a large degree , shared by blue galaxies that do not become red galaxies by @xmath2 .
one subtle difference is that the most massive non - progenitor blue galaxies are slightly younger than the most massive progenitors of red galaxies , suggesting that the blue progenitors of red galaxies , on their way " to become red galaxies , have started to foreshadow " quenching effects mildly .
-0.5 cm -0.5 cm -0,5 cm
-0,5 cm at a given redshift the cumulative environmental effects are imprinted on the relative distribution of galaxies of different color types and possibly on the properties of galaxies within each type .
we now present predictions of our simulations with respect to these aspects .
figure [ fig : fracdis ] shows distributions of three types of galaxies as a function of distance to the primary galaxy in units of the virial radius of the primary galaxy at @xmath2 .
all galaxies with distance larger than 4 virial radii of the primary galaxy are added to the bin with @xmath263 .
we use the total galaxy population above the respective stellar mass threshold as a reference sample and distributions in the top ( blue galaxies ) , middle ( green galaxies ) and bottom ( red galaxies ) panels are normalized relative to reference sample . comparing the top ( blue galaxies ) , middle ( green galaxies ) and bottom ( red galaxies ) panels , we see clear differences of environmental dependencies of the three types of galaxies . for blue galaxies
, there is a deficit at @xmath264 , which is compensated by a comparable excess at @xmath265 .
the range @xmath266 seems to mark the region where an excess of green galaxies , about one half of which will become red galaxies during the next @xmath267gyr .
it is useful to recall that not all galaxies in the green valley will turn into red galaxies , which perhaps has contributed in part to some of the irregularities " of the distribution of the green galaxies ( middle panel ) . for red galaxies ,
we see a mirror image of blue galaxies : there is an excess at @xmath264 and a deficit @xmath265 .
this trend is in agreement with observational indications ( e.g. , * ? ? ?
the emerging picture found here that satellite quenching plays a dominant role in quenching galaxies is in accord with observations ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
figure [ fig : fracent ] shows distributions of three types of galaxies as a function of environmental entropy @xmath109 .
we see that the excess of red galaxies starts at @xmath268 and rises toward higher entropy regions for red galaxies .
the trend for blue galaxies is almost an inverted version of that for red galaxies .
the trend for green galaxies lie in - between those for blue and red galaxies , as expected . in @xcite
we put forth the notion that a critical entropy @xmath269 ( at @xmath2 and weakly dependent on redshift ) , marks a transition to a regime of inefficient gas cooling hence cold gas starvation , because above this entropy the gas cooling time exceeds the hubble time .
this is borne out with our more detailed analysis here .
we also plot ( not shown here ) distributions of three types of galaxies as a function of the environmental pressure @xmath99 and environmental overdensity @xmath270 , respectively , and find that the trend is broadly similar to that see in figure [ fig : fracent ] .
overall , our results are in accord with the observed density - morphology relation ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) , and with the general observed trend of galaxy population becoming bluer or mean / median specific star formation rate becoming higher towards underdense regions in the local universe ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
-1.5 cm -1.0 cm having examined the dependencies of three types of galaxies on environmental variables , we now explore the dependencies on two additional variables : the mass of the halo of the primary galaxy and the secondary to primary galaxy stellar mass ratio .
figure [ fig : msat ] shows fractions of three populations of galaxies in terms of color ( red , green , blue ) as a function of secondary to primary galaxy stellar mass ratio .
the ( left , middle , right ) columns are for primary galaxies of halo masses in three ranges ( @xmath271,@xmath272,@xmath273 ) respectively .
the four rows from top to bottom are for secondaries within four different radial shells centered on the primary galaxy ( @xmath274 , @xmath275r_v^c$ ] , @xmath276r_v^c$ ] , @xmath277r_v^c$ ] ) .
we adopt the following language to make comparative statements : the environment quenching is important if the fraction of blue galaxies is less than the fraction of red galaxies and vice versa .
we see two separate trends in figure [ fig : msat ] .
first , more massive environments are more able to quench star formation ; for primary galaxies with halo masses in the range of @xmath278 the quenching appears to extend at least to @xmath276r_v^c$ ] , whereas for primary galaxies with lower halo masses the quenching effect is no longer significant at @xmath276r_v^c$ ] .
second , for any given primary galaxy halo mass , the quenching is more effective when the ratio of secondary to primary stellar mass is smaller .
that the impact of environmental effects increases with decreasing secondary to primary galaxy stellar mass ratio is perhaps not surprising and consistent with observations ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) , but at odds with the assumption of mass - independent environment quenching in empirical modeling ( e.g. , * ? ? ?
our finding that quenching is still effective down to at least halo mass range @xmath279 for the primary is in agreement with observations ( e.g. , * ? ? ?
@xcite , using deep gmos - s spectroscopy for 11 galaxy groups at @xmath16 , show that the strongest environmental dependence is observed in the fraction of passive galaxies , which make up only @xmath17 percent of the field in the mass range @xmath18 but are the dominant component of groups .
if we take the radial range @xmath280 as the field " , we see that the fraction of red galaxies is in the range @xmath281 ( figure [ fig : msat ] ) , reaching @xmath282 within @xmath283 , in agreement with @xcite . in a large galaxy sample study of a @xmath284 cluster
, @xcite find that red galaxy fraction is @xmath285 in the central region of the cluster and declines to a level of @xmath286 at a projected cluster - centric radius of @xmath287mpc . in the right column of figure
[ fig : msat ] , for halos in the mass range of @xmath273 , we see that within the virial radius the red fractions are in the range of @xmath288 , depending on the satellite to primary galaxy mass ratio , with the overall mean of red fraction within the virial radius in the range @xmath289 .
the more central region has a red fraction higher than @xmath289 , consistent with their observation .
@xcite also find that the environmental effect extends even beyond 3mpc , which is about 3 virial radii for their cluster . from the right column in figure
[ fig : msat ] we see that red galaxies dominate up to 3 virial radii , in reasonable accord with observations
. some of our results are not necessarily the most straightforward to imagine physically in the absence of simulations and detailed analysis .
for example , the interpretation by @xcite that the rate of sf quenching is faster in groups than in the field may be interpreted , in a different way , in the context of our results .
the excess of quenched galaxies in group environments is larger than in less dense environments as seen in figures ( [ fig : fracdis],[fig : fracent ] ) , consistent with observations @xcite .
however , we show that the quenching time depends weakly on environment as shown in figure [ fig : tbrhis ] . in other words ,
our simulations suggest the following physical picture : environment quenching of star formation in galaxies is _ more efficient _ in more hostile environment ( closer to a larger galaxy , higher density , high pressure , high entropy , etc ) , but _ not faster , for the most part_. -0.0 cm -0.5 cm given the unambiguous environmental effects on the relative distributions of galaxy colors , we ask the following reverse question : can we decipher environmental effects on a set of blue galaxies alone across different environments at any given epoch ?
figure ( [ fig : sfr ] ) shows the specific sfr ( ssfr ) distributions of blue galaxies as a function of the four environmental variables at @xmath2 . while one sees trends of the overall number of galaxies with respect to each of the environmental variables , as already shown in figures ( [ fig : fracdis],[fig : fracent ] ) , there is no strong visible trend of the mean value of ssfr with respect to environment for blue galaxies . to put it slightly differently ,
the variations of ssfr among blue galaxies at any given mass are sufficiently large and the quenching effects on galaxies at a given ssfr and stellar mass are sufficiently varied that at any environment the scatter in ssfr at a given stellar mass is larger than the difference in average ssfr .
we attribute this result physically to the combined effects of substantial non - uniformity of the properties of blue galaxies prior to entering the environmental spheres " of influence and substantial non - uniformity of the influence of the environment on galaxies illustrated in figures ( [ fig : tauminsfq ] and [ fig : tauminsigma ] ) .
this result - that the intrinsic properties of star - forming galaxies appear independent of environment - is , however , in accord with observations ( e.g. , * ? ? ?
* ; * ? ? ?
utilizing _ ab initio _ * l*arge - scale * a*daptive - mesh - refinement * o*mniscient * z*oom-*i*n cosmological hydrodynamic simulations ( * laozi simulations * ) of the standard cold dark matter model , we perform a chronological and statistical study of formation and evolution of ( 1664 , 367 , 1296 ) ( blue , green , red ) galaxies of stellar mass @xmath290 at @xmath2 .
the simulations have an ultra - high resolution of @xmath291pc / h with an additional calibration run of resolution @xmath43pc / h .
the soundness of treatment of relevant physical processes , especially the uncertain feedback processes from star formation , is checked and confidence built by comparisons with a large set of independent observations over a range of scales and redshift . here
we examine some global trends of formation and evolution of galaxies , focusing on galaxy star formation quenching and color migration .
we do not include agn feedback , in part because of its large uncertainties and primarily because of our intention to focus on external effects .
our main findings may be summarized in several points .
\(1 ) environment quenching is shown to be responsible for making the vast majority , if not all , of red galaxies .
two environmental conditions appear to be necessary in the making of a typical red galaxy : ram - pressure stripping and cold gas starvation .
the ram - pressure stripping disconnects a galaxy from its large - scale cold gas reservoir , while existing cold gas in the central ( @xmath292 kpc ) region is unaffected by ram - pressure but consumed by diminishing sf with an exponential time of several hundred myr . subsequently , gas starvation in the high - velocity environment guards against further significant sf , pushing it through the green valley to the red sequence .
the total duration from the initial sharp drop of sfr to the time of entry into the red sequence is @xmath293gyr ( see eq [ eq : tqr ] for a fitting formula ) with the median of @xmath211gyr for red galaxies at @xmath2 . we suggest that adopting a spread in quenching time in the semi - analytic modeling may result in an improved treatment in that approach .
\(2 ) the radius of a galaxy s quenching sphere depends on properties of the infalling galaxy being quenched , causing large variations in the quenching time scales at a given radius and large variations in the quenching radius at a given mass of infalling galaxy .
nevertheless , the vast majority of red galaxies are found to be within three virial radii of a larger galaxy , at the onset of quenching when the specific star formation rate experiences the sharpest decline to fall below @xmath294gyr@xmath4 ( depending on the redshift when it occurs ) .
the exponential decay time of sfr at the onset of quenching is , on average , about a factor of two shorter for events occurring within the virial radius of the quenching galaxy than for those outside the virial radius of the quenching galaxy , which may be evidence of enhanced quenching with the additional aid of tidal stripping when the onset of quenching takes place within the virial radius .
however , it is stressed that the overall duration from the onset of quenching to the time entering the sequence does not depend much on environment .
\(3 ) not all galaxies in the green valley migrate to the red sequence ; ( 40%,40%,48% ) of green valley galaxies of stellar mass @xmath295 do not proceed to become red galaxies at @xmath2 after having turned green for @xmath15gyrs . for those galaxies that are en route to the red sequence ,
the time spent in the green valley is brief , typically of @xmath104myr .
\(4 ) throughout the quenching period and the ensuing period in the red sequence galaxies follow nearly vertical tracks in the color - stellar - mass diagram , which correspond most closely to c tracks " proposed by @xcite .
in contrast , individual galaxies of all masses grow most of their stellar masses in the blue cloud , prior to the onset of quenching , and progressively more massive blue galaxies with already relatively old mean stellar ages continue to enter the red sequence .
consequently , correlations among observables of red galaxies - such as the age - mass relation - are largely inherited from their blue progenitors .
the age - mass relation of simulated red galaxies is found to be in good agreement with observations .
\(5 ) while environmental effects are responsible for producing the environmental dependence of the color makeup of the galaxy population , the average properties ( e.g. , sfr ) of blue galaxies as a sub - population display little dependence on environment , which is in agreement with observations .
overall , the excess ( deficit ) of red ( blue ) galaxies occurs within about three virial radii , in good agreement with a wide range of observations .
our detailed examination suggests that the excess of quenched galaxies in progressively denser environment ( groups , clusters , etc ) is , for the most part , a result of quenching _ being more efficient _ , _ not faster _ , on average , in denser environment .
physically , this comes about because most of the time it takes to drive a blue star - forming galaxy to the red sequence is spent during the starvation phase , not the initial gas removal phase by ram - pressure stripping that displays a stronger dependence on environment .
i would like to thank claire lackner for providing the sql based merger tree construction software , and drs .
john wise , matthew turk and sam skillmans for help with analysis and visualization program yt @xcite .
i would like to thank michael vogeley for a very careful reading of the manuscript , many colleagues for useful discussions , and an anonymous referee for a critical and constructive report .
computing resources were in part provided by the nasa high- end computing ( hec ) program through the nasa advanced supercomputing ( nas ) division at ames research center .
this work is supported in part by grant nasa nnx11ai23 g .
the simulation data are available from the author upon request . , m. r. , hogg , d. w. , bahcall , n. a. , baldry , i. k. , brinkmann , j. , csabai , i. , eisenstein , d. , fukugita , m. , gunn , j. e. , ivezi , . , lamb , d. q. , lupton , r. h. , loveday , j. , munn , j. a. , nichol , r. c. , okamura , s. , schlegel , d. j. , shimasaku , k. , strauss , m. a. , vogeley , m. s. , & weinberg , d. h. 2003 , , 594 , 186 , m. r. , hogg , d. w. , bahcall , n. a. , brinkmann , j. , britton , m. , connolly , a. j. , csabai , i. , fukugita , m. , loveday , j. , meiksin , a. , munn , j. a. , nichol , r. c. , okamura , s. , quinn , t. , schneider , d. p. , shimasaku , k. , strauss , m. a. , tegmark , m. , vogeley , m. s. , & weinberg , d. h. 2003 , , 592 , 819 , a. , merloni , a. , brusa , m. , magnelli , b. , salvato , m. , mignoli , m. , zamorani , g. , fiore , f. , rosario , d. , mainieri , v. , hao , h. , comastri , a. , vignali , c. , balestra , i. , bardelli , s. , berta , s. , civano , f. , kampczyk , p. , le floch , e. , lusso , e. , lutz , d. , pozzetti , l. , pozzi , f. , riguccini , l. , shankar , f. , & silverman , j. 2012 , , 427 , 3103 , e. , faber , s. m. , koo , d. c. , dutton , a. a. , simard , l. , mcgrath , e. j. , huang , j .- s . , bell , e. f. , dekel , a. , fang , j. j. , salim , s. , barro , g. , bundy , k. , coil , a. l. , cooper , m. c. , conselice , c. j. , davis , m. , domnguez , a. , kassin , s. a. , kocevski , d. d. , koekemoer , a. m. , lin , l. , lotz , j. m. , newman , j. a. , phillips , a. c. , rosario , d. j. , weiner , b. j. , & willmer , c. n. a. 2012 , , 760 , 131 , m. c. , newman , j. a. , croton , d. j. , weiner , b. j. , willmer , c. n. a. , gerke , b. f. , madgwick , d. s. , faber , s. m. , davis , m. , coil , a. l. , finkbeiner , d. p. , guhathakurta , p. , & koo , d. c. 2006 , , 370 , 198 , s. m. , willmer , c. n. a. , wolf , c. , koo , d. c. , weiner , b. j. , newman , j. a. , i m , m. , coil , a. l. , conroy , c. , cooper , m. c. , davis , m. , finkbeiner , d. p. , gerke , b. f. , gebhardt , k. , groth , e. j. , guhathakurta , p. , harker , j. , kaiser , n. , kassin , s. , kleinheinrich , m. , konidaris , n. p. , kron , r. g. , lin , l. , luppino , g. , madgwick , d. s. , meisenheimer , k. , noeske , k. g. , phillips , a. c. , sarajedini , v. l. , schiavon , r. p. , simard , l. , szalay , a. s. , vogt , n. p. , & yan , r. 2007 , , 665 , 265 , p. l. , nichol , r. c. , miller , c. j. , balogh , m. l. , goto , t. , zabludoff , a. i. , romer , a. k. , bernardi , m. , sheth , r. , hopkins , a. m. , castander , f. j. , connolly , a. j. , schneider , d. p. , brinkmann , j. , lamb , d. q. , subbarao , m. , & york , d. g. 2003 , , 584 , 210 , c. m. , alexander , d. m. , mullaney , j. r. , altieri , b. , coia , d. , charmandaris , v. , daddi , e. , dannerbauer , h. , dasyra , k. , del moro , a. , dickinson , m. , hickox , r. c. , ivison , r. j. , kartaltepe , j. , le floch , e. , leiton , r. , magnelli , b. , popesso , p. , rovilos , e. , rosario , d. , & swinbank , a. m. 2012 , , 760 , l15 , c. , lilly , s. j. , kova , k. , peng , y. , bschorr , t. j. , carollo , c. m. , contini , t. , kneib , j .-
p . , le fevre , o. , mainieri , v. , renzini , a. , scodeggio , m. , zamorani , g. , bardelli , s. , bolzonella , m. , bongiorno , a. , caputi , k. , cucciati , o. , de la torre , s. , de ravel , l. , franzetti , p. , garilli , b. , iovino , a. , kampczyk , p. , lamareille , f. , le borgne , j .- f . , le brun , v. , maier , c. , mignoli , m. , pello , r. , perez montero , e. , presotto , v. , silverman , j. , tanaka , m. , tasca , l. , tresse , l. , vergani , d. , zucca , e. , barnes , l. , bordoloi , r. , cappi , a. , cimatti , a. , coppa , g. , koekemoer , a. m. , lpez - sanjuan , c. , mccracken , h. j. , moresco , m. , nair , p. , pozzetti , l. , & welikala , n. 2013 , , 769 , 24 , e. , smith , k. m. , dunkley , j. , bennett , c. l. , gold , b. , hinshaw , g. , jarosik , n. , larson , d. , nolta , m. r. , page , l. , spergel , d. n. , halpern , m. , hill , r. s. , kogut , a. , limon , m. , meyer , s. s. , odegard , n. , tucker , g. s. , weiland , j. l. , wollack , e. , & wright , e. l. 2010 , arxiv e - prints , k. , lilly , s. j. , knobel , c. , bschorr , t. j. , peng , y. , carollo , c. m. , contini , t. , kneib , j .- p . , le fevre , o. , mainieri , v. , renzini , a. , scodeggio , m. , zamorani , g. , bardelli , s. , bolzonella , m. , bongiorno , a. , caputi , k. , cucciati , o. , de la torre , s. , de ravel , l. , franzetti , p. , garilli , b. , iovino , a. , kampczyk , p. , lamareille , f. , le borgne , j .- f . , le brun , v. , maier , c. , mignoli , m. , oesch , p. , pello , r. , perez montero , e. , presotto , v. , silverman , j. , tanaka , m. , tasca , l. , tresse , l. , vergani , d. , zucca , e. , aussel , h. , koekemoer , a. m. , le floch , e. , moresco , m. , & pozzetti , l. 2013 , arxiv e - prints , i. , balogh , m. , de propris , r. , couch , w. , bower , r. , offer , a. , bland - hawthorn , j. , baldry , i. k. , baugh , c. , bridges , t. , cannon , r. , cole , s. , colless , m. , collins , c. , cross , n. , dalton , g. , driver , s. p. , efstathiou , g. , ellis , r. s. , frenk , c. s. , glazebrook , k. , hawkins , e. , jackson , c. , lahav , o. , lumsden , s. , maddox , s. , madgwick , d. , norberg , p. , peacock , j. a. , percival , w. , peterson , b. a. , sutherland , w. , & taylor , k. 2002 , , 334 , 673 , a. , wilson , g. , yee , h. k. c. , gilbank , d. , hoekstra , h. , demarco , r. , balogh , m. , van dokkum , p. , franx , m. , ellingson , e. , hicks , a. , nantais , j. , noble , a. , lacy , m. , lidman , c. , rettura , a. , surace , j. , & webb , t. 2012 , , 746 , 188 , y .- j . , lilly , s. j. , kova , k. , bolzonella , m. , pozzetti , l. , renzini , a. , zamorani , g. , ilbert , o. , knobel , c. , iovino , a. , maier , c. , cucciati , o. , tasca , l. , carollo , c. m. , silverman , j. , kampczyk , p. , de ravel , l. , sanders , d. , scoville , n. , contini , t. , mainieri , v. , scodeggio , m. , kneib , j .-
p . , le fvre , o. , bardelli , s. , bongiorno , a. , caputi , k. , coppa , g. , de la torre , s. , franzetti , p. , garilli , b. , lamareille , f. , le borgne , j .- f . , le brun , v. , mignoli , m. , perez montero , e. , pello , r. , ricciardelli , e. , tanaka , m. , tresse , l. , vergani , d. , welikala , n. , zucca , e. , oesch , p. , abbas , u. , barnes , l. , bordoloi , r. , bottini , d. , cappi , a. , cassata , p. , cimatti , a. , fumana , m. , hasinger , g. , koekemoer , a. , leauthaud , a. , maccagni , d. , marinoni , c. , mccracken , h. , memeo , p. , meneux , b. , nair , p. , porciani , c. , presotto , v. , & scaramella , r. 2010 , , 721 , 193 , b. m. , aragn - salamanca , a. , zaritsky , d. , de lucia , g. , milvang - jensen , b. , desai , v. , jablonka , p. , halliday , c. , rudnick , g. , varela , j. , bamford , s. , best , p. , clowe , d. , noll , s. , saglia , r. , pell , r. , simard , l. , von der linden , a. , & white , s. 2009 , , 693 , 112 , l. , bolzonella , m. , lamareille , f. , zamorani , g. , franzetti , p. , le fvre , o. , iovino , a. , temporin , s. , ilbert , o. , arnouts , s. , charlot , s. , brinchmann , j. , zucca , e. , tresse , l. , scodeggio , m. , guzzo , l. , bottini , d. , garilli , b. , le brun , v. , maccagni , d. , picat , j. p. , scaramella , r. , vettolani , g. , zanichelli , a. , adami , c. , bardelli , s. , cappi , a. , ciliegi , p. , contini , t. , foucaud , s. , gavignaud , i. , mccracken , h. j. , marano , b. , marinoni , c. , mazure , a. , meneux , b. , merighi , r. , paltani , s. , pell , r. , pollo , a. , radovich , m. , bondi , m. , bongiorno , a. , cucciati , o. , de la torre , s. , gregorini , l. , mellier , y. , merluzzi , p. , vergani , d. , & walcher , c. j. 2007 , , 474 , 443 , v. , iovino , a. , scodeggio , m. , cucciati , o. , knobel , c. , bolzonella , m. , oesch , p. , finoguenov , a. , tanaka , m. , kova , k. , peng , y. , zamorani , g. , bardelli , s. , pozzetti , l. , kampczyk , p. , lpez - sanjuan , c. , vergani , d. , zucca , e. , tasca , l. a. m. , carollo , c. m. , contini , t. , kneib , j .-
p . , le fvre , o. , lilly , s. , mainieri , v. , renzini , a. , bongiorno , a. , caputi , k. , de la torre , s. , de ravel , l. , franzetti , p. , garilli , b. , lamareille , f. , le borgne , j .- f . , le brun , v. , maier , c. , mignoli , m. , pell , r. , perez - montero , e. , ricciardelli , e. , silverman , j. d. , tresse , l. , barnes , l. , bordoloi , r. , cappi , a. , cimatti , a. , coppa , g. , koekemoer , a. m. , mccracken , h. j. , moresco , m. , nair , p. , & welikala , n. 2012 , , 539 , a55 , d. j. , mozena , m. , wuyts , s. , nandra , k. , koekemoer , a. , mcgrath , e. , hathi , n. p. , dekel , a. , donley , j. , dunlop , j. s. , faber , s. m. , ferguson , h. , giavalisco , m. , grogin , n. , guo , y. , kocevski , d. d. , koo , d. c. , laird , e. , newman , j. , rangel , c. , & somerville , r. 2013 , , 763 , 59 , s. , rich , r. m. , charlot , s. , brinchmann , j. , johnson , b. d. , schiminovich , d. , seibert , m. , mallery , r. , heckman , t. m. , forster , k. , friedman , p. g. , martin , d. c. , morrissey , p. , neff , s. g. , small , t. , wyder , t. k. , bianchi , l. , donas , j. , lee , y. , madore , b. f. , milliard , b. , szalay , a. s. , welsh , b. y. , & yi , s. k. 2007 , , 173 , 267 , p. , rosario , d. j. , shao , l. , lutz , d. , maiolino , r. , alexander , d. m. , altieri , b. , andreani , p. , aussel , h. , bauer , f. e. , berta , s. , bongiovanni , a. , brandt , w. n. , brusa , m. , cepa , j. , cimatti , a. , daddi , e. , elbaz , d. , fontana , a. , frster schreiber , n. m. , genzel , r. , grazian , a. , le floch , e. , magnelli , b. , mainieri , v. , nordon , r. , prez garcia , a. m. , poglitsch , a. , popesso , p. , pozzi , f. , riguccini , l. , rodighiero , g. , salvato , m. , sanchez - portal , m. , sturm , e. , tacconi , l. j. , valtchanov , i. , & wuyts , s. 2012 , , 540 , a109 , i. , ivezi , . , knapp , g. r. , narayanan , v. k. , strauss , m. a. , gunn , j. e. , lupton , r. h. , schlegel , d. , bahcall , n. a. , brinkmann , j. , brunner , r. j. , budavri , t. , csabai , i. , castander , f. j. , doi , m. , fukugita , m. , gyry , z. , hamabe , m. , hennessy , g. , ichikawa , t. , kunszt , p. z. , lamb , d. q. , mckay , t. a. , okamura , s. , racusin , j. , sekiguchi , m. , schneider , d. p. , shimasaku , k. , & york , d. 2001 , , 122 , 1861 , m. , finoguenov , a. , lilly , s. j. , bolzonella , m. , carollo , c. m. , contini , t. , iovino , a. , kneib , j .-
lamareille , f. , le fevre , o. , mainieri , v. , presotto , v. , renzini , a. , scodeggio , m. , silverman , j. d. , zamorani , g. , bardelli , s. , bongiorno , a. , caputi , k. , cucciati , o. , de la torre , s. , de ravel , l. , franzetti , p. , garilli , b. , kampczyk , p. , knobel , c. , kova , k. , le borgne , j .- f . , le brun , v. , lpez - sanjuan , c. , maier , c. , mignoli , m. , pello , r. , peng , y. , perez - montero , e. , tasca , l. , tresse , l. , vergani , d. , zucca , e. , barnes , l. , bordoloi , r. , cappi , a. , cimatti , a. , coppa , g. , koekemoer , a. m. , mccracken , h. j. , moresco , m. , nair , p. , oesch , p. , pozzetti , l. , & welikala , n. 2012 , , 64 , 22 , k. e. , labb , i. , van dokkum , p. g. , brammer , g. , kriek , m. , marchesini , d. , quadri , r. f. , franx , m. , muzzin , a. , williams , r. j. , bezanson , r. , illingworth , g. d. , lee , k .- s . , lundgren , b. , nelson , e. j. , rudnick , g. , tal , t. , & wake , d. a. 2011 , , 735 , 86 , d. b. , hopkins , a. m. , brough , s. , taylor , e. n. , norberg , p. , bauer , a. , brown , m. j. i. , cameron , e. , conselice , c. j. , croom , s. , driver , s. , grootes , m. w. , jones , d. h. , kelvin , l. , loveday , j. , pimbblet , k. a. , popescu , c. c. , prescott , m. , sharp , r. , baldry , i. , sadler , e. m. , liske , j. , robotham , a. s. g. , bamford , s. , bland - hawthorn , j. , gunawardhana , m. , meyer , m. , parkinson , h. , drinkwater , m. j. , peacock , j. , & tuffs , r. 2012 , , 423 , 3679 , c. n. a. , faber , s. m. , koo , d. c. , weiner , b. j. , newman , j. a. , coil , a. l. , connolly , a. j. , conroy , c. , cooper , m. c. , davis , m. , finkbeiner , d. p. , gerke , b. f. , guhathakurta , p. , harker , j. , kaiser , n. , kassin , s. , konidaris , n. p. , lin , l. , luppino , g. , madgwick , d. s. , noeske , k. g. , phillips , a. c. , & yan , r. 2006 , , 647 , 853 , j. , dekel , a. , faber , s. m. , noeske , k. , koo , d. c. , gerke , b. f. , cooper , m. c. , salim , s. , dutton , a. a. , newman , j. , weiner , b. j. , bundy , k. , willmer , c. n. a. , davis , m. , & yan , r. 2013 , , 428 , 3306 , i. , zheng , z. , weinberg , d. h. , blanton , m. r. , bahcall , n. a. , berlind , a. a. , brinkmann , j. , frieman , j. a. , gunn , j. e. , lupton , r. h. , nichol , r. c. , percival , w. j. , schneider , d. p. , skibba , r. a. , strauss , m. a. , tegmark , m. , & york , d. g. 2011 , , 736 , 59 | an analysis of more than @xmath0 galaxies resolved at better than @xmath1pc / h at @xmath2 in a * laozi * " cosmological adaptive mesh refinement hydrodynamic simulation is performed and insights gained on star formation quenching and color migration .
the vast majority of red galaxies are found to be within three virial radii of a larger galaxy , at the onset of quenching when the specific star formation rate experiences the sharpest decline to fall below @xmath3gyr@xmath4 ( depending on the redshift ) .
we shall thus call this mechanism environment quenching " , which encompasses satellite quenching .
two physical processes are largely responsible : ram - pressure stripping first disconnects the galaxy from the cold gas supply on large scales , followed by a longer period of cold gas starvation taking place in high velocity dispersion environment , during the early part of which the existing dense cold gas in the central region ( @xmath5kpc ) is consumed by _ in situ _ star formation .
quenching is found to be _ more efficient _
( i.e. , a larger fraction of galaxies being quenched ) , _ but not faster _
( i.e. , duration being weakly dependent on environment ) , on average , in denser environment . throughout this quenching period and
the ensuing one in the red sequence galaxies follow nearly vertical tracks in the color - stellar - mass diagram .
in contrast , individual galaxies of all masses grow most of their stellar masses in the blue cloud , prior to the onset of quenching , and progressively more massive blue galaxies _ with already relatively older mean stellar ages _ continue to enter the red sequence .
consequently , correlations among observables of red galaxies - such as the age - mass relation - are largely inherited from their blue progenitors at the onset of quenching . while the color makeup of the entire galaxy population strongly depends on environment , which is a direct result of environment quenching , physical properties of blue galaxies as a sub - population show little dependence on environment .
a variety of predictions from the simulation are shown to be in accord with extant observations .
= 1 |
tropical matroid polytopes have been introduced in @xcite as the tropical convex hull of the cocircuits , or dually , of the bases of a matroid .
the arrangement of finitely many points @xmath1 in the tropical torus @xmath2 has a natural decomposition @xmath3 of @xmath2 into ( ordinary ) polytopes , the tropical complex , equipped with a ( fine ) type @xmath4 , which encodes the relative position to the generating points .
the _ coarse types _ only count the cardinalities of @xmath4 . in @xcite , develin and
sturmfels showed that the bounded cells of @xmath3 yield the tropical convex hull of @xmath1 , which is dual to the regular subdivision @xmath5 of a product of two simplices ( or equivalently due to the cayley trick to the regular mixed subdivisions of a dilated simplex ) .
the authors of @xcite and @xcite use the connection of the cellular structure of @xmath3 or rather of @xmath5 to minimal cellular resolutions of certain monomial ideals to provide an algorithm for determining the facial structure of the bounded subcomplex of @xmath3 .
a main result of @xcite says that the labeled complex @xmath3 supports a minimal cellular resolution of the ideal @xmath6 generated by monomials corresponding to the set of all ( coarse ) types .
the main theme of this paper is the study of the tropical complex of tropical convex polytopes associated with matroids arising from graphs
the _ tropical matroid polytopes_. recall that a _ matroid _ @xmath7 is a finite collection @xmath8 of subsets of @xmath9={1,2,\ldots , n}$ ] , called _ independent sets _ , such that three properties are satisfied : ( i ) @xmath10 , ( ii ) if @xmath11 and @xmath12 then @xmath13 , ( iii ) if @xmath14 and @xmath15 there exists @xmath16 such that @xmath17 . the last one is also called the
_ exchange property_. the maximal independent sets are the _ bases _ of @xmath7 .
a matroid can also be defined by specifying its _ non - bases _ ,
i.e. the subsets of @xmath18 with cardinality @xmath19 that are not bases .
for more details on matroids see the survey of oxley @xcite and the books of white(@xcite , @xcite , @xcite )
. an important class of matroids are the graphic or cycle matroids proven to be regular , that is , they are representable over every field .
a _ graphic matroid _ is associated with a simple undirected graph @xmath20 by letting @xmath18 be the set of edges of @xmath20 and taking as the bases the edges of the spanning forests .
matroid polytopes were first studied in connection with optimization and linear programming , introduced by jack edmonds @xcite .
a nice polytopal characterization for a matroid polytope was given by gelfand et al .
@xcite stating that each of its edges is a parallel translate of @xmath21 for some @xmath22 and @xmath23 . in the case of tropical matroid
polytopes the coarse types display the number @xmath24 of bases @xmath25 of the associated matroid with subsets @xmath26 , where all elements of @xmath6 but none of @xmath27 are contained in @xmath25 .
let @xmath0 be the tropical complex of a tropical matroid polytope with @xmath28 elements and rank @xmath19 .
the set of all coarse types of the maximal cells arising in @xmath0 is given by the tuples @xmath29 with @xmath30 where @xmath31 $ ] and @xmath32 is a sequence of elements such that @xmath33\setminus\{i_1,i_2,\ldots , i_{d'}\}$ ] contains a basis of the associated matroid .
subsequently , we relate our combinatorial result to commutative algebra . for the coarse type @xmath34 of @xmath35 and @xmath36 the monomial ideal @xmath37\ ] ]
is called the _ coarse type ideal_. in @xcite , corollary 3.5 , it was shown that @xmath6 is generated by the monomials , which are assigned to the coarse types of the inclusion - maximal cells of the tropical complex . as a direct consequence of theorem 3.6 in @xcite
, we obtain the generators of @xmath6 .
the coarse type ideal @xmath6 for the tropical complex of a tropical matroid polytope with @xmath28 elements and rank @xmath19 is equal to @xmath38\setminus\{i_1,\ldots , i_{d'}\ } \text { contains a basis } \rangle\ ] ] where @xmath39 .
furthermore , we apply these results to the special case of uniform matroids , introduced and studied in @xcite .
we close this work by stating the minimal tropical halfspaces containing a uniform tropical matroid polytope by using the characterization of proposition 1 in @xcite .
we start with collecting basic facts about tropical convexity and fixing the notation . defining _ tropical addition _ by @xmath40 and _ tropical multiplication _ by @xmath41 yields the _ tropical semi - ring _ @xmath42 .
component - wise tropical addition and _ tropical scalar multiplication _
@xmath43 equips @xmath44 with a semi - module structure . for @xmath45 the set @xmath46_{\mathrm{trop}}:= { \{(\lambda \odot x ) \oplus ( \mu \odot y)\vphantom{\lambda,\mu \in { { \mathbb{r } } } } \mid \vphantom{(\lambda \odot x ) \oplus ( \mu \odot y)}\lambda,\mu \in { { \mathbb{r}}}\}}\ ] ] defines the _ tropical line segment _ between @xmath47 and @xmath48 .
a subset of @xmath44 is _ tropically convex _ if it contains the tropical line segment between any two of its points .
a direct computation shows that if @xmath49 is tropically convex then @xmath50 is closed under tropical scalar multiplication .
this leads to the definition of the _ tropical torus _ as the quotient semi - module @xmath51 note that @xmath2 was called `` tropical projective space '' in @xcite , @xcite , @xcite , and @xcite .
tropical convexity gives rise to the hull operator @xmath52 .
a _ tropical polytope _ is the tropical convex hull of finitely many points in @xmath2 . like an ordinary polytope
each tropical polytope @xmath53 has a unique set of generators which is minimal with respect to inclusion ; these are the _ tropical vertices _ of @xmath53 .
there are several natural ways to choose a representative coordinate vector for a point in @xmath2 .
for instance , in the coset @xmath54 there is a unique vector @xmath55 with non - negative coordinates such that at least one of them is zero ; we refer to @xmath56 as the _ canonical coordinates _ of @xmath57 .
moreover , in the same coset there is also a unique vector @xmath58 such that @xmath59 .
hence , the map @xmath60 is a bijection
. often we will identify @xmath2 with @xmath61 via this map .
the _ tropical hyperplane _
@xmath62 defined by the _ tropical linear form _
@xmath63 is the set of points @xmath64 such that the minimum @xmath65 is attained at least twice .
for @xmath66 the tropical hyperplane is shown in figure [ fig:3hypersimplicesb ] .
the complement of a tropical hyperplane in @xmath2 has exactly @xmath28 connected components , each of which is an _
open sector_. a _ closed sector _ is the topological closure of an open sector .
the set @xmath67 for @xmath68 , is the _
@xmath19-th open sector _ of the tropical hyperplane @xmath69 in @xmath2 defined by the zero tropical linear form .
its closure is @xmath70 we also use the notation @xmath71 for any set @xmath72:=\{1,\dots , d+1\}$ ] . if @xmath73 is an arbitrary tropical linear form then the translates @xmath74 for @xmath75 are the open sectors of the tropical hyperplane @xmath62 .
the point @xmath76 is the unique point contained in all closed sectors of @xmath62 , and it is called the _ apex _ of @xmath62 . for each @xmath72 $ ] with @xmath77
the set @xmath78 is the _ closed tropical halfspace _ of @xmath62 of type @xmath6 .
a tropical halfspace @xmath79 can also be written in the form @xmath80where @xmath6 and @xmath27 are disjoint subsets of @xmath33 $ ] and @xmath81 $ ] .
the tropical polytopes in @xmath2 are exactly the bounded intersections of finitely many closed tropical halfspaces ; see @xcite and @xcite .
we concentrate on the combinatorial structure of tropical polytopes .
let @xmath82 be a sequence of points in @xmath2 .
the _ ( fine ) type _ of @xmath57 with respect to @xmath1 is the ordered @xmath83-tuple @xmath84 where @xmath85 for a given type @xmath86 with respect to @xmath1 the set @xmath87 is a relatively open subset of @xmath2 and is called the _ cell _ of type @xmath86 with respect to @xmath1 .
the set @xmath88 as well as its topological closure are tropically and ordinary convex ; in @xcite , these were called _
polytropes_. with respect to inclusion the types with respect to @xmath1 form a partially ordered set .
the intersection of two cells @xmath89 and @xmath90 with type @xmath91 and @xmath86 is equal to the polyhedron @xmath92 .
the collection of all ( closed ) cells induces a polyhedral subdivision @xmath3 of @xmath2 .
a @xmath93-tropical polytope @xmath94 is the union of cells in the bounded subcomplex @xmath95 of @xmath3 induced by the arrangement @xmath96 of @xmath97-tropical hyperplanes with apices @xmath98 .
a cell of @xmath3 is unbounded if and only if one of its type components is the empty set .
the type of @xmath47 equals the union of the types of the ( maximal ) cells that contain @xmath47 in their closure .
the dimension of a cell @xmath99 can be calculated as the number of the connected components of the undirected graph @xmath100 minus one .
the zero - dimensional cells are called pseudovertices of @xmath53 . replacing the ( fine ) type entries @xmath101 $ ] for
@xmath102 $ ] of a point @xmath103 by their cardinalities @xmath104 we get the _ coarse type _
@xmath105 of @xmath35 .
a coarse type entry @xmath106 displays how many generating points lie in the @xmath19-th closed sector @xmath107 . in @xcite ,
the authors associate the tropical complex of a tropical polytope with a monomial ideal , the coarse type ideal @xmath108.\ ] ] by corollary 3.5 of @xcite , @xmath6 is generated by the monomials assigned to the coarse types of the inclusion - maximal cells of the tropical complex .
the tropical complex @xmath3 gives rise to minimal cellular resolutions of @xmath6 .
[ thm : djs2009 ] the labeled complex @xmath3 supports a minimal cellular resolution of the ideal @xmath6 generated by monomials corresponding to the set of all ( coarse ) types . considering cellular resolutions of monomial ideals , introduced in @xcite and @xcite , is a natural technique to construct resolutions of monomial ideals using labeled cellular complexes and provide an important interface between topological constructions , combinatorics and algebraic ideas .
the authors of @xcite and @xcite use this to give an algorithm for determining the facial structure of a tropical complex .
more precisely , they associate a squarefree monomial ideal @xmath6 with a tropical polytope and calculate a minimal cellular resolution of @xmath6 , where the @xmath22-th syzygies of @xmath6 are encoded by the @xmath22-dimensional faces of a polyhedral complex . a tropical halfspace is called _ minimal _ for a tropical polytope @xmath53 if it is minimal with respect to inclusion among all tropical halfspaces containing @xmath53 .
consider a tropical halfspace @xmath109 with @xmath72 $ ] and apex @xmath110 , and a tropical polytope @xmath111 . to show that @xmath112 is minimal for @xmath53
, it suffices to prove , by proposition 1 of @xcite , that the following three criteria hold for the type @xmath113 of the apex @xmath114 : * @xmath115 $ ] , * for each @xmath116 there exists an @xmath117 such that @xmath118 , * for each @xmath117 there exists @xmath116 such that @xmath119 . here
, we denote the complement of a set @xmath120 $ ] as @xmath121\setminus i$ ] .
obvious minimal tropical halfspaces of a tropical polytope @xmath122 are its cornered halfspaces , see @xcite .
the _ @xmath19-th corner _ of @xmath53 is defined as @xmath123 the tropical halfspace @xmath124 is called the _ @xmath19-th cornered tropical halfspace _ of @xmath53 and the intersection of all @xmath28 cornered halfspaces is the _ cornered hull _ of @xmath53 .
the tropical matroid polytope of a matroid @xmath125 is defined in @xcite as the tropical convex hull of the negative incidence vectors of the bases of @xmath125 . in this paper
, we restrict ourselves to matroids arising from graphs .
the _ graphic matroid _ of a simple undirected graph @xmath126 is @xmath127 . while the forests of @xmath20 form the system of independent sets of @xmath128 its bases are the spanning forests .
we will assume that @xmath20 is connected , so the bases of @xmath128 are the spanning trees of @xmath20 .
furthermore , we exclude bridges , i.e. edges whose deletion increases the number of connected components of @xmath20 , leading to elements that are contained in every basis .
let @xmath28 be the number of elements and @xmath129 be the number of bases of @xmath130 and @xmath131 its bases .
it follows from the exchange property of matroids that all bases of @xmath125 have the same number of elements , which is called the _ rank _ of @xmath125 .
consider the @xmath132-matrix @xmath133 with rows indexed by the elements of the ground set @xmath18 and columns indexed by the bases of @xmath125 which has a @xmath134 in entry @xmath135 if the @xmath22-th element is in the @xmath23-th basis .
the _ tropical matroid polytope _
@xmath53 of @xmath125 is the tropical convex hull of the columns of @xmath7 .
let @xmath136 be the set of generators of @xmath53 .
it turns out that these are just the tropical vertices of @xmath53 , see lemma [ lem : pseudovertices ] .
if the underlying matroid has rank @xmath19 , then the canonical coordinate vectors of @xmath1 have exactly @xmath19 zeros and @xmath137 ones and will be denoted as @xmath138 or for short @xmath139 if the corresponding basis is @xmath140 .
note that with @xmath141 as @xmath97 instead of @xmath93 the generators of a tropical matroid polytope are the positive incidence vectors of the bases of the corresponding matroid . throughout this paper
we write @xmath142 for the set of all tropical matroid polytopes arising from a graphic matroid with @xmath28 elements and rank @xmath19 .
the tropical hypersimplex @xmath143 in @xmath2 studied in @xcite is a tropical matroid polytope of a uniform matroid of rank @xmath19 with @xmath28 elements and @xmath144 bases .
it is defined as the tropical convex hull of all points @xmath145 where @xmath146 is the @xmath22-th unit vector of @xmath44 and @xmath6 is a @xmath19-element subset of @xmath33 $ ] .
the tropical vertices of @xmath143 are @xmath147}{k}\right\}\ , \text { for all } k>0.\ ] ] in @xcite , it is shown that @xmath148 implying that the first tropical hypersimplex contains all other tropical hypersimplices in @xmath2 .
the first tropical hypersimplex @xmath149 in @xmath2 is the @xmath150-dimensional tropical standard simplex which is also a polytrope .
clearly , we have for a tropical matroid polytope @xmath151 the chain @xmath152 . for @xmath66
the three tropical hypersimplices are shown in figure [ fig:3hypersimplices ] . by proposition 3 of @xcite about the shape of a tropical line segment , the only pseudovertex of the tropical line segment between two distinct @xmath134-@xmath156-vectors @xmath157 and @xmath158 in @xmath159 is the point @xmath160 with @xmath161 if @xmath162 or @xmath163 and @xmath164 otherwise .
since every element of @xmath18 is contained in any basis of @xmath128 ( apply any spanning - tree - greedy - algorithm for the connected components of @xmath20 starting from this element ) and by using the previous argument , the origin must be contained in @xmath53 .
let @xmath53 be the corresponding tropical matroid polytope with its generators @xmath176 the type of the origin @xmath169 of @xmath53 is @xmath177 where the @xmath22-th type entry contains all bases using the edge @xmath22 ( italic edge attributes ) .
[ lem : chull of tmp ] the cornered hull of a tropical matroid polytope @xmath151 with generators @xmath1 is the @xmath150-dimensional tropical standard simplex @xmath178 .
the @xmath22-th corner of @xmath53 is the vector @xmath146 .
the type of @xmath146 with respect to @xmath1 is @xmath179 with @xmath180 & \text{if $ j = i$ } \ , , \\ \{l\mid j\in b_l \text { and } i\notin b_l \ } & \text{otherwise } \ , .
\end{cases}\ ] ] for @xmath181 the @xmath22-th ( canonical ) coordinate of @xmath182 is @xmath183 the @xmath23-th coordinate of the @xmath22-th corner @xmath184 of @xmath53 is @xmath185 in canonical coordinates we get @xmath186 , which at the same time is the @xmath22-th apex vertex of the tropical standard simplex @xmath178 .
the type of @xmath146 is @xmath187 , where some index @xmath188 is contained in the @xmath23-th coordinate @xmath189 for @xmath190 if @xmath191 .
this is satisfied by all bases @xmath192 with @xmath193 and @xmath194 .
for @xmath195 all indices @xmath196 $ ] are contained in @xmath197 since the right hand side of @xmath198 is smaller or equal than the left hand side in every case .
let @xmath201 be the type of the pseudovertex @xmath169 with respect to @xmath1 and consider a point @xmath202 .
if the complement @xmath203 of @xmath27 is equal to @xmath204 , then the type @xmath205 of @xmath206 with respect to @xmath1 is given by @xmath207 since the union of the elements of one or more bases of @xmath125 consists of at least @xmath19 elements , the index set @xmath27 has at least @xmath19 elements and thus we have @xmath211 for the cardinality @xmath212 of @xmath203 .
we can assume that @xmath213 .
then some index @xmath188 occurs in the @xmath23-th coordinate @xmath189 if and only if @xmath214 for @xmath215 the left hand side of equation ( [ eq : condhypersimplex : typepseudovert ] ) is @xmath216 . if @xmath193 , we get @xmath217 and this is minimal in ( [ eq : condhypersimplex : typepseudovert ] ) if the coordinates @xmath218 are equal to one for all @xmath219 , i.e. @xmath194 . if @xmath220 , we get @xmath221 . therefore , @xmath189 is equal to @xmath222 .
for @xmath223 the left hand side is @xmath224 . if @xmath193 , we get @xmath225 . if @xmath220 , we get @xmath226 and this is minimal in if the coordinates @xmath218 are equal to one for all @xmath219 , i.e. @xmath227 . therefore , @xmath189 is equal to @xmath228 . if @xmath229 , the pseudovertex @xmath230 is a generator of @xmath53 .
each of its type entries contains the index , which is assigned to a basis @xmath181 . since @xmath25 is the only basis with @xmath231 , its index is the only element of @xmath232 for @xmath233 . for this reason , the generators as defined in ( [ eq : generators ] )
are exactly the tropical vertices of @xmath53 .
now we consider the other points of @xmath234 , i.e. @xmath235 . the intersection of two type entries @xmath236 is equal to @xmath237 in the first case of [ eq : intersection ] , @xmath236 consists of at least one tropical vertex @xmath238 with @xmath239 and @xmath240 for all @xmath219 . in the second case
there are even more tropical vertices allowed and @xmath241 .
hence , proposition 17 of @xcite tells us that the cell @xmath99 has dimension @xmath134 , i.e. the given points really are pseudovertices of @xmath53 . for @xmath242 and @xmath243 with
@xmath244 $ ] the tropical line segment between @xmath245 and @xmath246 is the concatenation of the two ordinary line segments @xmath247 $ ] and @xmath248 $ ] .
the point @xmath249 is again a point of @xmath250
. therefore , there are no other pseudovertices as the given points in @xmath250 .
now we consider the tropical standard simplex @xmath178 .
if the tropical vertex @xmath251 , @xmath252}{1}$ ] , of @xmath178 is given by the vector @xmath253 ( @xmath254 ) , then the type of the origin @xmath169 with respect to @xmath178 is @xmath255 .
therefore , this is an interior point of @xmath178 .
let @xmath245 with @xmath256}{j}$ ] be any pseudovertex of @xmath178 .
since for @xmath257 and @xmath258 , we have and + , respectively , it follows that the index @xmath22 is contained in the @xmath22-th entry of @xmath4 for all @xmath259 , i.e. @xmath260 .
hence , @xmath178 is a polytrope .
let @xmath261 be a pseudovertex of @xmath53 with @xmath242 for @xmath262 $ ] . if the complement @xmath203 of @xmath27 is equal to @xmath263 with @xmath264 , we will denote @xmath245 as @xmath265 and its type with respect to @xmath53 as @xmath266 because of the previous lemma , the @xmath22-th entry of @xmath267 contains all bases using edge @xmath257 that are possible after deleting the edges of @xmath203 in the corresponding graph @xmath20 or , equivalently , all bases that are possible after ( re-)inserting edge @xmath219 into @xmath268 .
we call a sequence of pseudovertices @xmath269 , or rather the set + , _ valid _ if the edge set @xmath270 contains a spanning tree of the underlying graph @xmath20 .
the first point @xmath271 is assigned to the total edge set @xmath18 of @xmath20 .
then we delete edge after edge such that the graph is still connected until the edge set forms a connected graph without cycles .
so the last point of a valid sequence is the tropical vertex @xmath182 of @xmath53 with @xmath272\setminus\{i_1,i_2,\ldots , i_{d - k+1}\}$ ] .
the maximal bounded cells of @xmath151 are of dimension @xmath273 .
they form the tropical convex hull of the pseudovertices of a valid sequence @xmath274 , where the last pseudovertex is a tropical vertex @xmath182 according to the basis @xmath272\setminus\{i_1,i_2,\ldots , i_{d - k+1}\}\in\mathcal{b}$ ] of @xmath125 .
let @xmath275 be the type of the pseudovertex @xmath169 with respect to @xmath53 .
then the type @xmath276 of the interior of the bounded cell @xmath277 is given by @xmath278 and @xmath279 for all @xmath233 .
first , we will show that this sequence really defines a bounded cell of @xmath53 , i.e. @xmath280 for all @xmath281 $ ] .
so consider the type entry at some coordinate @xmath282 @xmath283 the cardinality of @xmath284 is equal to the number of tropical vertices @xmath158 of @xmath53 with @xmath285 and @xmath286 ( in canonical coordinates ) respectively to the number of bases @xmath25 with @xmath287 and @xmath288 , which is greater than @xmath134 since we consider only valid sequences .
so every type coordinate @xmath289 contains at least one entry . in the case of uniform matroids
we have the choice of @xmath290 _ free _ coordinates from which @xmath291 must be equal to @xmath134 , i.e. the cardinality of @xmath289 is equal to @xmath292 .
analogously , the other type entries @xmath293 for @xmath233 and their cardinality @xmath294 can be verified .
furthermore , we have @xmath295 $ ] , because @xmath296 $ ] .
since no type entry of @xmath4 is empty , the cell @xmath297 is bounded .
more precisely , @xmath298 is a partition of the indices of @xmath299 , and the other type coordinates each contain the index of the tropical vertex @xmath182 ; we call this a _ pre - partition_. by proposition 17 in @xcite , the dimension of @xmath297 is @xmath273 . adding a pseudovertex @xmath304 to @xmath297 , @xmath305 with @xmath306 ( @xmath307 ) and @xmath308 for all @xmath309 , we consider @xmath310 . to keep
the status of a maximal bounded cell , the type of the cell still has to be a pre - partition of @xmath9 $ ] without empty type entries .
there are three different cases ( 1)-(3 ) .
\(1 ) for @xmath311 and @xmath312 , there is an index @xmath313 .
we consider the @xmath23-th type entry of @xmath314 that is equal to @xmath315 .
this is an empty set since there are no tropical vertices of @xmath53 with @xmath316 entries equal to one .
the cells with empty type entries are not bounded .
\(2 ) for @xmath311 and @xmath317 , we consider an index @xmath318 that corresponds to a valid sequence with @xmath319 , @xmath320 .
the @xmath23-th type entry of @xmath314 is equal to @xmath321 .
since @xmath322 , the cardinality of @xmath323 is less than @xmath324 , and we get no valid partition of @xmath9 $ ] .
\(3 ) for @xmath325 we have @xmath326 ( otherwise @xmath327 ) .
we choose the smallest index @xmath23 such that @xmath328 .
that means @xmath329 . since we have @xmath330 for all @xmath309 , we know that @xmath331 leading to @xmath332 . as in the other two cases
this is no valid pre - partition of @xmath9 $ ] .
similarly , it is not difficult to see that removing a pseudovertex and adding a new one from another sequence leads to unfeasible types or lower dimensional bounded cells , i.e. mixing of valid sequences is not possible .
altogether , we get the desired maximal bounded cells of @xmath53 .
the tropical matroid polytope @xmath53 from example [ ex : runex1 ] is contained in the @xmath335-dimensional tropical hyperplane with apex @xmath169 .
it is shown in figure [ fig : pvg ] as the abstract graph of the vertices and edges of its bounded subcomplex .
its maximal bounded cells are ordinary simplices of dimension @xmath336 , whose pseudovertices are the tropical vertices @xmath337 ( dark ) , the origin @xmath169 ( the centered point ) and the five corners @xmath338 ( light ) .
the four tropical vertices with indices @xmath339 , @xmath335 , @xmath340 and @xmath341 correspond to the bases that are possible after deleting edge @xmath156 in the underlying graph and therefore adjacent to the point @xmath342 .
one valid sequence @xmath343 leading to a maxim bounded cell is for example the ( tropical / ordinary ) convex hull of @xmath344 and @xmath345 , i.e. @xmath346 and @xmath347 , with interior cell type @xmath348 , representing the basis @xmath349 .
we now state the main theorem about the coarse types of maximal cells in the cell complex of a tropical matroid polytope .
let @xmath24 denote the number of bases @xmath181 with @xmath351 and @xmath352 .
[ thm : coarse types of tmp ] let @xmath0 be the tropical complex induced by the tropical vertices of a tropical matroid polytope @xmath151 .
the set of all coarse types of the maximal cells arising in @xmath0 is given by those tuples @xmath29 with @xmath353 where @xmath354 form a subsequence of a valid sequence of @xmath53 .
the first one , @xmath297 , contains a maximal bounded cell of dimension @xmath273 , which is the tropical convex hull of the pseudovertices of a complete valid sequence @xmath274 where @xmath355 is the complement of a basis of @xmath125 .
to get full - dimensional we have the choice between @xmath291 of @xmath19 free directions @xmath356 , @xmath357 .
so let @xmath358 be the _ extreme rays _ of @xmath297 , and @xmath201 be the type of the pseudovertex @xmath169 with respect to @xmath53 .
then the type @xmath276 of the interior of this unbounded cell @xmath297 is given by the intersection of the types of its vertices and therefore @xmath278 , @xmath359 for @xmath360 and @xmath361 .
choosing @xmath362 and @xmath363 , we get the coarse type entries of equation ( [ eq : coarse types of tmp ] ) .
the second type , @xmath297 , of maximal unbounded cells contains a bounded cell of lower dimension @xmath364 , which is the tropical convex hull of the pseudovertices of some subsequence @xmath365 .
to get full - dimensional we still need the extreme rays @xmath366 for all directions @xmath367 .
then the type @xmath276 of the interior of this unbounded cell @xmath297 is given by @xmath368 , @xmath369 for @xmath370 with the coarse type as given in equation ( [ eq : coarse types of tmp ] ) .
the third and last type of maximal unbounded cells contains a bounded cell of dimension @xmath371 and is assigned to the non - bases of @xmath125 , i.e. to the subsets of @xmath18 with cardinality @xmath19 that are not bases .
let @xmath372 be the complement of a non - basis @xmath373 and @xmath374 a valid subsequence .
then there is an unbounded cell @xmath297 that is the tropical convex hull of the pseudovertices @xmath375 and the extreme rays @xmath376 for all directions @xmath377 and with type entries @xmath278 , @xmath369 for @xmath378 .
choosing @xmath379 and observing that @xmath380 for the non - basis @xmath381 we get the desired result .
the coarse types of the maximal cells in the tropical complex induced by the tropical vertices of the tropical hypersimplex @xmath143 in @xmath2 with @xmath382 are up to symmetry of @xmath383 given by @xmath384 where @xmath385 correlates to the maximal dimension of a bounded cell of its boundary .
now we relate the combinatorial properties of the tropical complex @xmath0 of a tropical matroid polytope to algebraic properties of a monomial ideal which is assigned to @xmath0 . as a direct consequence of theorem [ thm : djs2009 ] and corollary 3.5 in @xcite
, we can state the generators of the coarse type ideal @xmath37,\ ] ] where @xmath34 is the coarse type of @xmath35 and @xmath36 .
the tropical complex @xmath0 of the tropical matroid polytope of example [ ex : runex1 ] has 73 maximal cells .
there are five maximal cells for the case @xmath386 with @xmath387 and @xmath388 for @xmath389 , and 48 for the case @xmath390 according to the 8 bases .
finally , there are 20 maximal cells for the case @xmath391 , where @xmath33\setminus\{i_1\}$ ] contains a basis , but @xmath33\setminus\{i_1,i_2\}$ ] does not necessarily contain a basis .
the coarse type ideal of @xmath0 is given by @xmath392\end{aligned}\]]we obtain its minimal free resolution , which is induced by @xmath0@xmath393where the exponents @xmath22 of the free graded @xmath394-modules @xmath395 correspond to the entries of the @xmath396-vector @xmath397 of @xmath0 . in ( ordinary ) convexity
swapping between interior and exterior description of a polytope is a famous problem known as the _ convex hull problem_. for a uniform matroid it is possible to indicate the minimal tropical halfspaces of its tropical matroid polytope .
let @xmath276 be the type of the apex @xmath169 of @xmath398 .
if a vertex @xmath402 appears in some type entry @xmath197 , then the @xmath22-th ( canonical ) coordinate of @xmath158 is equal to zero .
hence , exactly @xmath19 entries of @xmath4 contain the index of @xmath158 .
since the cardinality of @xmath121\setminus i$ ] is only @xmath291 , every tropical vertex of @xmath143 is contained in some sector @xmath403 with @xmath117 , i.e. @xmath404 . consider the complement @xmath405 of @xmath6 . for all @xmath406 there is a tropical vertex @xmath158 with @xmath407 , i.e. @xmath408 . since the cardinality of @xmath405 is equal to @xmath291 and @xmath158 has @xmath19 entries equal to zero , there must be an index @xmath409 such that @xmath410 .
we can conclude that @xmath411 .
the intersection @xmath412 is not empty for arbitrary @xmath413 $ ] , because its cardinality is equal to the number of tropical vertices @xmath158 with @xmath414 , which is @xmath415 with @xmath416 . for @xmath117 and @xmath116 ,
the set @xmath412 consists of all tropical vertices @xmath158 with @xmath407 and @xmath417 ( in canonical coordinates ) . on the other hand ,
the set @xmath418 contains all tropical vertices @xmath158 with @xmath419 .
so we get @xmath420 .
hence , we obtain that @xmath398 is a minimal tropical halfspace , and @xmath143 is contained in the intersection of its cornered hull @xmath421}h(e_i,\{i\})$ ] with @xmath422}{d - k+2}}h({\mathbf{0}},i)$ ] .
we still have to prove that the intersection of the given minimal tropical halfspaces is contained in @xmath143 .
let us assume that there is a point @xmath423 with @xmath424 .
then for any tropical halfspace @xmath398 , @xmath425}{d - k+2}$ ] , with @xmath406 we obtain @xmath426 .
consequently , the tropical hypersimplex @xmath143 is the set of all points @xmath57 satisfying @xmath427\text { with } \lvert i \rvert = d - k+2\\\text{and } ( -1)\odot x_i&\leq&\bigoplus_{j\neq i } x_j \text { for all } i\in[d+1].\end{aligned}\ ] ] the second tropical hypersimplex @xmath428 in @xmath429 is the intersection of the @xmath335 cornered halfspaces @xmath430 for @xmath431 and the tropical halfspaces @xmath432 , @xmath433 , @xmath434 and @xmath435 with apex @xmath436
. the second tropical hypersimplex @xmath437 in @xmath438 is the intersection of the three cornered halfspaces @xmath430 for @xmath439 and the tropical halfspaces @xmath440 , @xmath441 and @xmath442 with apex @xmath443 , see figure [ fig : delta22 ] . mike develin , francisco santos , and bernd sturmfels , _ on the rank of a tropical matrix _ , combinatorial and computational geometry , math .
52 , cambridge univ . press , cambridge , 2005 , pp . | describing the combinatorial structure of the tropical complex @xmath0 of a tropical matroid polytope , we obtain a formula for the coarse types of the maximal cells of @xmath0 . due to the connection between tropical complexes and resolutions of monomial ideals ,
this yields the generators for the corresponding coarse type ideal introduced in @xcite .
furthermore , a complete description of the minimal tropical halfspaces of the uniform tropical matroid polytopes , i.e. the tropical hypersimplices , is given . |
this paper is the third part of a study of combinatorial quantization of chern - simons theory@xcite .
several ideas developed here are in fact products of those introduced by v.v.fock and a.a.rosly in their study of poisson structures on the moduli space of flat connections@xcite . to understand the motivation of these papers we must recall some general facts about 3d chern - simons theory .
chern - simons theory is a gauge theory in 3 dimensions defined by the action principle _
cs= tr ( _ m ada+2 3a^3 ) where @xmath0 is a 3-manifold , @xmath1 a positive integer and @xmath2 a connection associated to a semisimple lie algebra @xmath3
if we first suppose that the manifold locally looks like a cylinder @xmath4 , @xmath5 considered as the the time direction , we can consider the chern - simons theory in an hamiltonian point of view .
we will denote by @xmath6 the two space - components of the gauge field taken to be the dynamical variables of the theory , and the time component @xmath7 will become a lagrange multiplier . with these notations the action can be written : _ cs= tr ( _ m ( - a _ 0 a + 2 a_0 f ) dt ) .
the first term gives the poisson structure : \
{ a^a_i(x , y ) , a^b_j(x,y ) } = ^ab_ij^(2)((x , y),(x,y ) ) .
[ poisson ] the hamiltonian is a combination of constraints and the second term imposes as a constraint that the curvature of the connection @xmath6 is zero : f = da+a^2=0.[constraint ] computing the poisson brackets of the constraints we obtain that they are first class : \ { f^a(x , y ) , f^b(x,y ) } = f^ab_cf^c(x , y)^(2)((x , y)(x,y ) ) . where the @xmath8 are the structure constants of the lie algebra @xmath9 the constraints ( [ constraint ] ) generate the infinitesimal gauge transformations of the gauge field then the phase space of the hamiltonian chern - simons theory is the space of flat connections modulo gauge transformations .
+ quantizing the latter poisson structure of gauge fields in the usual way , = ^ab_ij^(2)((x , y)(x,y ) ) we are led to work with an infinite dimensional algebra , observables becoming functionals over the elements of this algebra .
the idea of v.v.fock and a.a.rosly is completely different.let us briefly describe it .
+ the space of flat connections modulo the gauge group having only a finite number of freedom degrees , we reduce the connection to live on a graph encoding the topology of the surface and equip this `` graph connection '' with a poisson structure such that the poisson structure induced on the gauge invariant observables is compatible with that induced from the usual symplectic structure .
we obtain a `` simulation '' of hamiltonian chern - simons theory in the sense that the operator algebra derived from both descriptions are the same . + we will consider a graph dividing the surface into contractile plaquettes .
this graph will be equiped with an additionnal structure of `` ciliated fat graph '' , i.e. a graph with a linear order between adjacent links at each vertex .
let @xmath10 be neighbour vertices of the graph , we will denote by @xmath11}$ ] the parallel transport operator associated to the link and the connection .
the gauge group acts in the usual way on this object : u_[x , y ] g_xu_[x , y]g^-1_y we can define a lie poisson structure on such objects @xcite owning the following fascinating property : + the space of flat graph connections modulo graph gauge group is lie poisson isomorphic to the space of flat connections modulo gauge group on the surface .
+ the problem has been reduced to a finite dimensional problem by limiting the gauge group to act at a finite number of sites .
+ the quantization of such a lie poisson structure leads us to an exchange algebra on which acts a quantum group .
the object of @xcite was to define this algebraic structure .
+ in a second time we found a projector in this algebra imposing `` a posteriori '' the flatness condition , the result was then a two dimensional lattice gauge theory based on a quantum group .
the correlation functions associated to gauge invariant objects in this theory being related to expectation values in chern - simons theory .
+ in our second paper @xcite we further investigated the algebra of gauge invariant elements , particularily the algebra associated to loops .
our aim was to describe a new approach to knots invariants , showing in a well defined framework the relation between reshetikhin - turaev invariants and chern - simons theory .
+ the aim of the following paper is double : + * to investigate further the computation of the correlation functions of our theory on any surface and construct the derived invariants associated to the mapping class group , and establish a new description of three manifold invariants using a generalization of our previous construction to the case of roots of unity + * to build up a three dimensional lattice q - gauge theory associated to triangulations of any 3-manifolds which will describe a well defined finite path integral formula for chern - simons theory and the way to compute any correlation functions of this theory . +
this work is a tentativ to revisit the work of e.witten @xcite with a well defined formalism allowing a lot of new computations and specially computations of invariants associated to intersecting loops in chern - simons theory .
in this chapter , after a brief summary of the results of @xcite @xcite , we will further develop the notion of quantum group at root of unity in the dual version and then , using this construction , generalize the results on gauge fields algebra developed in @xcite in a way quite different from that described in @xcite .
we will consider a hopf algebra @xmath12 to simplify we will take @xmath13 with @xmath14 being a complex number different from @xmath15 as usual we will refer to @xmath16 as the universal @xmath17matrix associated to @xmath2 ( we will often write @xmath18 ) , @xmath19 the element defined by @xmath20 verifying the usual properties and @xmath21 the ribbon element defined by @xmath22 ( for details see @xcite ) .
+ depending on whether @xmath14 is a root of unity or not , the representation theory of @xmath2 is completely different
. we will denote by @xmath23 the set of equivalence classes of finite dimensional irreducible representations of @xmath24 in each class @xmath25 we will pick out a representativ @xmath26 and , like often in physics , denote equivalently by @xmath26 or @xmath27the representation space associated to @xmath28 we will denote by @xmath29 ( resp .
@xmath30 ) the right ( resp . the left ) contragredient representation build up from the antipode ( resp .
the inverse of the antipode ) by @xmath31 and @xmath32 the one dimensional representation associated to @xmath33 the tensor product of two representations is defined by the coproduct @xmath34 if q is a root of unity the decomposition of the tensor product of two irreducible representations can involve indecomposable representations ( i.e representations which are not irreducible but can not ye !
t be decomposed in a direct sum of stable @xmath35modules ) .
we are able to introduce @xmath36 and @xmath37 respectively projection of the tensor product of @xmath26 and @xmath38 on the m - th isotypic component @xmath39 and the inclusion of @xmath39 in the tensor product @xmath40 using this notation we will make one more restriction between `` physical '' representations verifying @xmath41 and the other representations , we will denote by @xmath42 this subset of @xmath23 .
we will introduce a new tensor product between elements of @xmath42 simply realizing a truncation of the previous one , defined by : = _ phys(a)n^ _ .
@xmath43 is the fusion matrix of @xmath2 and we will also use the notation @xmath44 to be equal to 1 or 0 depending on whether @xmath39 occurs or not in the decomposition of the tensor product @xmath40 if q is generic the tensor product of two elements of @xmath45 can be decomposed in a direct sum of elements of @xmath46 moreover all irreducible representations are `` physical '' , so we do not have to change the tensor product in this case .
we will associate new projection and inclusion operators @xmath47 and @xmath48 build up from the truncated tensor product .
we will also use the @xmath49 notation @xmath50 defined in the usual way from projection and inclusion operators ( see @xcite for definitions and properties ) .
+ we will use the following notation replacing the coproduct by a truncated coproduct for any element @xmath51 of the algebra @xmath2 : = _ phys(a ) ^_,m ^,m_. the antipode and counity maps do not change through truncation and we will again have : = ^t | the first trivial properties of the projection and inclusion operators are : & & _ ^,m_,m^=_m , m_,()id _ + & & _
phys(a),m ^_,m^,m_= + & & ^0_=^0 _ = _ 0^=_0 ^= _ id_. the essential fact is that when the truncation is not trivial ( i.e in the root of unity case ) the representations @xmath52 and @xmath53 are no more equal but are equivalent , the intertwiner map between them being @xmath54 defined by : = _ , , phys(a ) _ ^_^^_^ _ ( where we have omited the multiplicities to simplify the notation , it will often be the case in the following ) .
we will denote by @xmath55 its quasi - inverse .
we will often use the notation @xmath56 and the coproduct notations @xmath57 + in the case where @xmath14 is generic @xmath54 is simply the identity but more generally it is possible to collect some interesting properties in the root of unity case . using the pentagonal identity and other trivial identities on @xmath49 symbols we can verify : & = & ( ) ( ) [ pentagon ] + = & = & = [ theta0 ] and other similar identities for @xmath58 moreover we have the quasi - inverse properties : = = recalling that , here
, @xmath59 is simply a projector .
let us now define intertwiners between @xmath60 and @xmath61 using our basic objects @xmath62 : & & p_12 = _ phys(a),m _ ^1 2 ^_,m _ ^,m[rdef ] + & & p_12 = _ phys(a),m _ ^-1 2 ^_,m _ ^,m + where @xmath63 and @xmath64 where @xmath65 is the drinfeld casimir , equal to @xmath66 , where @xmath67 is the quadratic casimir .
we will denote in the following @xmath68 and @xmath69 and use sometimes the notation @xmath70 and @xmath71 . using the hexagonal identities on the @xmath49 symbols it can be shown that : & & = + & & = [ quasitriang ] which is simply the analog of the quasitriangularity property of @xmath17matrices .
+ this matrix is no more inversible but we have : = ( ) = .[rinv ] let us now study the properties of the antipodal map and develop the analog of the ribbon properties @xcite
. we will denote by @xmath72 and @xmath73 the matrices defined by @xmath74 and @xmath75 ) where @xmath76 ( resp .
@xmath77 ) is a basis of the representation space of @xmath26 ( resp .
@xmath29 ) and @xmath78 is the duality bracket . to choose the normalisation of @xmath79 and @xmath80s we will impose the ambiant isotopy conditions
i ^(1)_i b s(^(2)_i ) a ^(3)_i = 1 _ i s(^-1^(1)_i ) a ^-1^(2)_i b s(^-1^(3)_i ) = 1 in order to generalize the known properties relative to the antipode , we will also introduce some notations which will be useful in the following : & & g = _ i , j ( ) ( ) ) ( a ) ( ) ( + & & d = _ i , j ( ) ( ) ( b ) ( ) ( ) + & & f = _ i ( ) )
g it can be shown that the latter matrices verify : & & = a = b [ deltaa ] + & & ^_= ^t^|_|| _
^= ^t_|^|| we endly introduce the element @xmath19 associated to the square of the antipode , defined by : u=_i , j s(^-1 ( 2)_i b s(^-1 ( 3)_i))s(b_j)a a_j^-1 ( 1)_i , u is invertible and 1&=&u_i , j s^-1(^-1 ( 1)_i ) s^-1(a d_j)c_j^-1 ( 2)_i b ^-1 ( 3)_i= + & = & s^2(_i , j s^-1(^-1 ( 1)_i ) s^-1(a d_j)c_j^-1 ( 2)_i b ^-1 ( 3)_i)u moreover we have as usual the essential property , s^2()=u u^-1 and the usual corollaries & & s^2(u)=u + & & us(u)=s(u)u + & & _ i s(b_i ) a a_i = s(a)u = s(u)u_i s(c_i)ad_i[contract ] + & & ( u)=1 we will denote by @xmath21 the element satisfying : & & v^2=us(u ) + & & s(v)=v ( v)=1 . and by @xmath81 the element @xmath82 then it can be shown that : = ( ) using the latter notations it can be shown that @xmath83 and @xmath84 in the following the q - dimension of the representation @xmath26 will be defined by @xmath85=tr_{\alpha}(s(a ) \mu b).$ ] now , using the latter framework we can give a well defined construction of the quantum group in the dual version for any value of @xmath14 . as a vector space
this algebra , called @xmath86 , is generated by @xmath87 and the product is simply defined by : _
1 _ 2= _ phys(a)_^^_. this product is not associative but verify : _
123((_1 _ 2)_3)= ( _ 1(_2 _ 3 ) ) _ 123 .
[ quasi - associativity ] moreover we have the exchange relation : _ 12 _ 1 _ 2= _ 2 _ 1 _ 12 .
[ exchange ] this algebra can be equiped with a coproduct and a counity : ( ) = _ i ( ) = ^i_j .
moreover it can be shown that the antipodal map @xmath88 defined to be the linear map verifying @xmath89 owns the properties : & & s(g)^i_k a^k_l g^l_j = a^i_j [ antipode1 ] + & & ( s(a))^k_l g^l_j s(g)^i_k=(s(a))^i_j [ antipode2 ] + & & s ( ) s ( ) = _ 12s ( ) _ 12 + & & s^2 ( ) = our aim is now to define as in our first paper the gauge theory associated to this gauge symmetry algebra .
let @xmath90 be a compact connected oriented surface with boundary @xmath91 and let @xmath92 be a triangulation of @xmath93 let us denote by @xmath94 the oriented faces of @xmath95 by @xmath96 the set of edges counted with their orientation .
if @xmath97 is an interior link , @xmath98 will denote the opposite link .
we have @xmath99 where @xmath100 are respectively the set of interior edges and boundary edges .
finally let us also define @xmath101 to be the set of points ( vertices ) of this triangulation , @xmath102 , where @xmath103 are respectively the set of interior vertices and boundary vertices .
if @xmath97 is an oriented link it will be convenient to write @xmath104 where @xmath105 is the departure point of @xmath97 and @xmath106 the end point of @xmath107 we will write @xmath108 and @xmath109 let us define for @xmath110 the hopf algebra @xmath111 and @xmath112 this hopf algebra was called in @xcite `` the gauge symmetry algebra . ''
if @xmath106 is a vertex we shall write @xmath113 to denote the embedding of the element @xmath114 in @xmath115 in order to define the non commutative analogue of algebra of gauge fields we have to endow the triangulation with an additional structure@xcite , an order between links incident to a same vertex , the _
cilium order_. a ciliation of the triangulation is an assignment of a cilium @xmath116 to each vertex @xmath117 which consists in a non zero tangent vector at z. the orientation of the riemann surface defines a canonical cyclic order of the links admitting @xmath117 as departure or end point .
let @xmath118 be links incident to a common vertex @xmath119 the strict partial cilium order @xmath120 is defined by : @xmath121 if @xmath122 and the unoriented links @xmath123 appear in the cyclic order defined by the orientation .
if @xmath118 are incident to a same vertex @xmath117 we define : @xmath124 the algebra of _ _ gauge fields @xcite@xcite @xmath125 is the non associative algebra generated by the formal variables @xmath126 with @xmath127 and satisfying the following relations : * commutation rules * @xmath128 * decomposition rule * ( l)_1 ( l)_2&=&_,m_,m^ , ( l)_,^,m p_12 , [ ucg ] + 0(l)&=&1,l .
* quasi - associativity * let @xmath129 be a monomial of gauge fields algebra elements with a certain parenthesing @xmath130 for each vertex @xmath106 of the triangulation we construct a tensor product of representations of @xmath2 by replacing each @xmath131 in the monomial by the vector space @xmath26 ( resp .
@xmath29 resp .
@xmath32 ) depending on whether @xmath106 is the endpoint ( resp .
departure point resp .
not element ) of the edge @xmath132 and keeping the previous parenthesing .
let us consider two different parenthesing @xmath133 and @xmath134 of the same monomial .
we can construct for both , as described before , the corresponding vector spaces for each @xmath106 and deduce the intertwiner @xmath135 relating them .
the relation of quasi - associativity is then simply : ( _ x _ x ) m_p_1=m_p_2[assocalg ] @xmath125 is a right @xmath136 comodule defined by the morphism of algebra @xmath137 : ( ( xy))=_x ( xy ) s(_y ) .
the definition relations of the gauge fields algebra are compatible to the coaction of the gauge symmetry algebra .
the subalgebra of gauge invariant elements of @xmath125 is denoted @xmath138 it was shown ( provided some assumption on the existence of a basis of @xmath125 of a special type ) @xcite@xcite that there exists a unique non zero linear form @xmath139 satisfying : 1 .
( invariance ) @xmath140 2 .
( factorisation ) @xmath141 ( we have used the notation @xmath142 for @xmath143 to denote the subalgebra of @xmath125 generated as an algebra by @xmath144 with @xmath145 ) . it can be evaluated on any element using the formula : h((x , y)^i_j)=_,0 where @xmath32 denotes the trivial representation of dimension @xmath146 i.e @xmath32 is the counit .
it was convenient to use the notation @xmath147 instead of @xmath148 we obtained the important formula : h((y , x)_1 ( s id)(_12 ) v_^-1(x , y)_2)= 1p_12b_1 .
[ ortho ] a path @xmath149 ( resp . a loop @xmath149 )
is a path ( resp.a loop ) in the graph attached to the triangulation of @xmath90 , given by the collection of its vertices , it will also denote equivalently the continuous curve ( resp .
loop ) in @xmath90 defined by the links of @xmath130 in this article we will denote by @xmath150 $ ] a link with departure point @xmath151 and end point @xmath152 . following the definition for links , the departure point of @xmath149 is denoted @xmath153 and its endpoint @xmath154 the set of vertices ( resp .
edges ) of the path @xmath149 is denoted by @xmath155 ( resp .
@xmath156 ) , the cardinal of this set is called the `` length '' of @xmath149 and will be denoted by @xmath157 properties of path and loops such as self intersections , transverse intersections will always be understood as properties satisfied by the corresponding curves on @xmath93 let @xmath158 $ ] be a path , we defined the sign @xmath159 to be @xmath160 ( resp .
@xmath161 ) if @xmath162 ( resp .
@xmath163 ) .
if @xmath149 is a simple path @xmath164 $ ] with @xmath165 , we define the holonomy along @xmath149 by _ p = v_^12_i=1^n-1(x_i , p)((x_n x_n-1)a ( x_n-1x_n-2)a ( x_1x_0 ) ) . when @xmath166 is a simple loop @xmath167,$ ] we define the holonomy along @xmath166 by _ c = v_^12(_i=1^n ( x_i , c)- ( x_0,c))((x_0 x_n)a ( x_nx_n-1)a ( x_1x_0 ) ) .
we define an element of @xmath125 , called _ wilson loop _ attached to @xmath168 _
c= tr_(s(a)_c).[wilsonloop1 ] the interior parenthesing being irrelevant because the loop is simple ( the vector space attached to a vertex occurs , first with @xmath26 , second with @xmath29 and the other times in the trivial representation ) and moreover we have the relations ( [ theta0 ] ) .
we will also use the notation @xmath169}.$ ] the properties shown in our first paper are easily generalized : the element @xmath170 is gauge invariant and moreover it does not depend on the departure point of the loop @xmath171 moreover it verifies the fusion equation : _ c _ c=_phys(a ) n_^ _ c the gauge invariance is quite obvious because of relations ( [ antipode1 ] ) , ( [ antipode2 ] ) . to show the cyclicity property we must put our wilson loop in another form called
`` expanded form '' in our first paper . using relations ( [ contract]),([rinv ] )
we easily obtain : & & ( c)=v_^-12(_xc(x , c ) ) tr_^n ( ( s(a))^n_i = n^1 p_ii-1 + & & ( _ i = n^1(x_j+1 x_j)_j ( s i d ) ( r_jj-1^((x_j , c ) ) ) ) ( x_1 x_0)_1)[wilsonloop2].in this form the cyclicity invariance is obvious using the commutation relations ( [ es ] ) . + the fusion relation is less trivial to show .
we first show a lemma describing the decomposition rules for holonomies : ( u_p)_1 ( u_p)_2 = ^__p ^ _ p_12 _ 21 . indeed , using ( [ assocalg])([es ] )
, we easily obtain for a path @xmath172 $ ] : & & ( v_^12(y , p)((x y)a ( yz))_1(v_^12(y , p)((x y)a ( yz))_2= + & & = _ i , j , k , lv_^12(y , p)v_^12(y , p)((x y)_1(x y)_2)(s(^(1)_l)a ^(1)_i b_j ^-1 ( 2)_k ^(31)_l)_1 + & & ( s(^(2)_l ) s(^-1(1)_k)s(a_j)s(^(2)_i ) a ^(3)_i ^-1 ( 3)_k ^(32)_l)_2((y z)_1(y z)_2)= + & & = v_^12(y , p)v_^12(y , p)((x y)_1(x y)_2)_21 ( ( y z)_1(y z)_2 ) + the last equality is obtained by using successively ( [ quasitriang ] ) and ( [ pentagon ] ) . now , using ( [ deltaa])([rdef ] ) , we obtain the announced result for a two links path .
proceeding by induction we can prove it for any simple open path .
+ let us now consider the loop @xmath166 as formed by two pieces @xmath173 $ ] and @xmath174 $ ] , we have , using the same properties as before : & & ( v_^12((y , p)-(x , c))tr_(s(a)(x y ) a ( y x ) ) ) ( v_^12((y , p)-(x , c))tr_(s(a)(x y ) a ( y x)))= + & & = _ ( v_v_)^12((y , p)-(x , c))tr_((s(^-1 ( 2)_p ) s(^(3)_m ) s(c_l ) s ( ^-1 ( 2)_i ) s(a ) ^-1 ( 1)_i ^(1)_m ^-1 ( 11)_p)_1 + & & ( s(^-1 ( 3)_p ) s(a ) ^-1 ( 3)_i d_l ^(2)_m ^-1 ( 12)_p)_2 ( ( x y)_1 ( x y)_2)(s(^-1 ( 11)_q )
s(^(1)_n ) s ( ^-1 ( 1)_j ) + & & a ^-1 ( 2)_j b_k ^(3)_n ^-1 ( 2)_q)_1 ( s(^-1 ( 12)_q ) s(^(2)_n ) s(a_k ) s(^-1 ( 3)_j ) a ^-1 ( 3)_q)_2 ( ( y x)_1 ( y x)_2))= + & & = ( v_v_)^12((y , p)-(x , c))tr_((s s)(g_21r_12 ) ( ) ( ( x y)_1 ( x y)_2 )
( g_21r_21 ^ -1 ) + & & ( ( y x)_1 ( y x)_2 ) ) = + & & = _ n^_v_^12((y , p)-(x , c))tr_(s(a)(x y ) a u(y x ) ) which ends the proof of the theorem .
it can also be shown that @xmath175=0 $ ] for all simple loops @xmath176 without transverse intersections .
although the structure of the algebra @xmath125 depends on the ciliation , it has been shown in @xcite that the algebra @xmath177 does not depend on it up to isomorphism .
this is completely consistent with the approach of v.v.fock and a.a.rosly : in their work the graph needs to be endowed with a structure of ciliated fat graph in order to put on the space of graph connections @xmath178 a structure of poisson algebra compatible with the action of the gauge group @xmath179 however , as a poisson algebra @xmath180 is canonically isomorphic to the space @xmath181 of flat connections modulo the gauge group , the poisson structure of the latter being independent of any choice of r - matrix @xcite .
we introduced a boltzmann weight attached to any simple loop @xmath166 and defined by : _
c=_phys(a)[d _ ] _ c. this element satisfies the flatness relation : @xmath182 moreover we have @xmath183^{2}})^{2}&=&(\frac{(\delta_{c})}{\sum_{\alpha \in phys({\cal a } ) } [ d_{\alpha}]^{2}}).\label{flatness}\end{aligned}\ ] ] using the same properties as in the computation of fusion relations , we obtain : & & _ phys(a)[d_](v_^12((y , p)-(x , c))tr_(s(a)(x y ) a ( y x ) ) ) + & & ( v_^12((y , p)-(x , c))tr _ ( s(a)(x y ) a ( y x)))= + & & = _ [ d_](v_v_)^12((y , p)-(x , c))tr_((s(^-1 ( 2)_p ) s(^(3)_m ) s(c_l ) s ( ^-1 ( 2)_i ) s(a ) ^-1 ( 1)_i ^(1)_m + & & ^-1 ( 11)_p)_1 ( ^-1 ( 3)_i d_l ^(2)_m ^-1 ( 12)_p)_2 ( ( x y)_1 ( x y)_2 ) ( s(^-1 ( 11)_q ) s(^(1)_n ) s ( ^-1 ( 1)_j ) a ^-1 ( 2)_j b_k ^(3)_n ^-1 ( 2)_q)_1 + & & ( s(^-1 ( 12)_q ) s(^(2)_n ) s(a_k ) s(^-1 ( 3)_j ) a ^-1 ( 3)_q)_2 ( ( y x)_1 ( y x)_2)s(^-1 ( 3)_p)_2)= + & & = _ , phys(a)[d _ ] _ p , m tr _ ( ( s(^(3)_m ) s(a ) ^(2)_m ^-1 ( 12)_p)_1 + & & ( ^(1)_m ^-1 ( 11)_p)_2 ^ _ ( x y ) a ( y x ) v_^12((y , p)-(x , c ) ) ^ _ f^-1_21 s(^-1 ( 2)_p)_1 s(^-1 ( 3)_p)_2 ) the last equality uses again the quasitriangularity properties
. then we obtain , for any matrix @xmath184 in @xmath185 : & & _ c tr _ ( s(a ) v _ c ) = + & & = _ phys(a)_p , m tr _ ( ( _ phys(a)[d _ ] ^ _ f^-1_21 ( s(^-1 ( 2)_p ) s(^(3)_m ) s(a ) ^(2)_m ^-1 ( 12)_p)_1 + & & ( s(^-1 ( 3)_p ) s(a ) v ^(1)_m ^-1 ( 11)_p)_2 ^ _ ) ( ( x y ) a ( y x ) v_^12((y , p)-(x , c ) ) ) ) due to the intertwining properties and the normalizations mentioned before of the @xmath186s we can conclude ( see @xcite ) that it does exist some complex coefficients @xmath187 such that : & & id_k ^0_| = _ a ( ) ^ _ ^_| + & & + & & ^_=a()(^_| i d _ ) ( ^|_0id _ ) we obtain & & _ p , m _ phys(a)[d _ ] ^ _ f^-1_21 ( s(^-1 ( 2)_p ) s(^(3)_m ) s(a ) ^(2)_m ^-1 ( 12)_p)_1 + & & ( s(^-1 ( 3)_p ) s(a ) v ^(1)_m ^-1 ( 11)_p)_2 ^ _ = [ d_]id _ tr_(s(a ) v b ) then for any matrix @xmath184 we have @xmath188 , and the linear independance of the generators of our algebra ensures the final result .
+ the last formula of the proposition is a trivial consequence of the last result .
we were led to define an element that we called @xmath189 this element is the non commutative analogue of the projector on the space of flat connections . in @xcite@xcite
it was proved that @xmath190 is a central element of @xmath177 and the algebra @xmath191 was shown to be independant , up to isomorphism , of the triangulation .
the proof is based on the lemma of decomposition rules of the holonomies shown before and on the quite obvious property : let @xmath192 and @xmath193 be two simple contractile loops which interiors are disjoint and with a segment @xmath194 $ ] of their boundary in common : dh(u(xy ) ) _ c_1 _ c_2 = _ , ( c_1 # c_2 ) as a result it was advocated that @xmath195 is the algebra of observables of the chern simons theory on the manifold @xmath196.$ ] this is supported by the topological invariance of @xmath195 ( i.e this algebra depends only on the topological structure of the surface @xmath90 ) and the flatness of the connection .
our aim is now to construct in the algebra @xmath195 the observables associated to any link in @xmath196.$ ] in the following subsection and in the chapter @xmath197 the computations will be made in the case of @xmath14 generic to simplify the notations but the generalization to @xmath14 root of unity can be made exactly in the same way . + we will consider a compact connected surface @xmath90 with boundary @xmath198 .
the boundary is a set of disjoint simple closed curves which are designed to be `` in '' or `` out '' .
let us draw some oriented curves on the surface @xmath90 defining a link @xmath199 , assuming that their boundary is contained in @xmath200 and with simple , transverse intersections , with the specification of over- or undercrossing at each intersection .
we will also consider that representations of the quantum group are attached to connected components of the link .
+ the data ( surface with boundary + colored link ) will be called `` striped surface '' , the data of `` in '' ( resp . `` out '' ) boundary of @xmath90 and @xmath199 with corresponding colors will be called the `` in state '' ( resp . the `` out state '' ) of the striped surface .
+ to describe such objects we will choose a morse function which gives a time direction and the set of `` equitime planes '' @xmath201 cutting the surface .
an equitime plane @xmath202 divides the surface in two parts called respectively `` future '' and `` past '' . on any simple curve drawn on an equitime plane the time direction
give us an orientation of the curve , if we impose moreover a departure point @xmath106 on this curve we are able to decide if a point @xmath117 is on the left ( resp . on the right ) of another point @xmath203 if @xmath204 ( resp .
@xmath205 ) appear in the order given by this orientation .
we will consider the surface in a canonical position defined by the following conditions .
the intersection between the surface and the plane for @xmath206 or @xmath207 is empty .
the `` in '' ( resp . `` out '' )
boundary is contained in the @xmath208 ( resp .
@xmath209 ) plane .
the intersection between the @xmath210 and the surface is a set of disjoint simple closed curves @xmath211 ( not necessary disjoint at the singular times ) , where @xmath212 is the number of connected components of @xmath213 we will call `` @xmath214diagram '' of the surface a graph drawn on it which intersections with @xmath210 determine departure points on each closed curve in these sets .
we impose that the @xmath214diagram never turn around any handle of the surface .
+ our aim is now to define a ciliated fat graph which will encode the topology of the striped surface , i.e .
this decomposition involves only contractile plaquettes , it is sufficiently fine to allow us to put the link on the graph in a generic position and allow us to distinguish two situations related by a dehn twist of the surface .
we then decompose the surface in blocks , their number being chosen with respect to the singularities of the morse function ( considered as a function over the points of the surface and of the link ) .
the information contained in the morse function is not sufficient to deal with the problem of possible non trivial cycles of @xmath199 around handles of @xmath90 .
we will rule out this problem by adding fictively two disjoint @xmath214diagrams of @xmath90 to the link @xmath215 the intersections between the link and the @xmath214diagrams will detect the rotation of the link around an handle of @xmath90 , we will then refine the decomposition with respect to these datas .
we will assume that the singularities of the morse function @xmath216 , considered now as a function of the points of the surface , points of the link and points of the @xmath214diagrams , correspond to different times .
we will denote by @xmath217 the different instants corresponding to the singularities of the morse function .
we will consider a decomposition of the surface and of the link in `` elementary blocks ''
@xmath218 corresponding to the subdivision @xmath219 = [ \tau_0,\tau_1 ] \cup [ \tau_1,\tau_2 ] \cup \cdots \cup [ \tau_n,\tau_{n+1}]$ ] where @xmath220 are called `` cutting times '' .
an example of a striped surface with the block decomposition described before is shown in the following figure : we will consider a triangulation @xmath92 and a ciliation induced by this block decomposition .
the set @xmath221 of vertices of @xmath92 contains all elements of the sets @xmath222 and the singularities of the morse function considered as a function of the link and of the surface .
the edges of the triangulation are either a segment of the link or of the @xmath214diagram between two consecutiv vertices , or a segment drawn on the surface at the same time @xmath223 between two emerging strings .
the plaquettes of the triangulation are the connected regions of @xmath90 surrounded by the edges described before .
let us denote by @xmath224 the intersections of the link with the cutting planes @xmath225 the ciliation is chosen to be : at each vertex @xmath226 , directed to the past and to the left , just `` before '' the equitime line , and at each crossing of the oriented link , between the two outgoing strands .
we will choose a practical indexation satisfying the following properties : @xmath227 for @xmath228 @xmath229 and the @xmath226 belonging to the same @xmath230 are ordered from the departure point and from left to right .
the set of all @xmath224 for a given k is denoted by @xmath231 .
we will say that , at a vertex of @xmath199 , the orientation of the link is `` in the sense of time '' or `` against the sense of time '' according to the position of the link with respect to the cutting plane , the two corresponding subset of @xmath221 will be respectively denoted by @xmath232 and @xmath233 the set of all edges of @xmath92 is again denoted by @xmath234 and we will denote by @xmath235 ( resp .
@xmath236 ) the set of edges contained in @xmath237 ( resp . between @xmath238 and @xmath239 ) .
the elements of @xmath236 will be currently denoted as @xmath240 .
we index the elements of this set with the following rule : begining from the departure point fixed by the @xmath214diagram we order the links in an obvious way from left to right if they do not cross themselves and , at a crossing , the first one is that which overcross the other . at last we will denote by @xmath241 the set of faces of the triangu lation corresponding to the block @xmath242
to each element @xmath97 of @xmath234 we will associate a subset of @xmath94 called @xmath243 defined by the following rule : if @xmath97 belongs to a crossing and is the overcrossing ( resp.undercrossing ) strand , @xmath243 is the set of the four plaquettes surrounding the crossing ( resp .
the set is empty ) , if @xmath97 is an annihilation or a creation , @xmath243 is the set of the two surrounding plaquettes , elsewhere @xmath243 contains only the plaquette just at the left of @xmath97 .
we then define @xmath244 to be the subset of @xmath245 such that @xmath246 if it is on the left of @xmath97 in the same block or anywhere in a past block .
a link in @xmath196 $ ] is an embedding of @xmath247)^{\times p''}$ ] into @xmath196,$ ] with @xmath248 $ ] . on the set of links
we can define a composition law , denoted @xmath249 defined as follows : let @xmath199 and @xmath250 be to links in @xmath251 $ ] considered up to ambiant isotopy .
we define @xmath252 to be the link obtained by putting @xmath199 in @xmath253 $ ] and @xmath250 in @xmath254.$ ] this composition is associative and admit the empty link as unit element .
this composition law is commutative if and only if @xmath90 is homeomorphic to the sphere .
let us denote by @xmath255 the connected components of the link @xmath215 @xmath256 the colour of this component , and by @xmath257 the colored loop obtained by projecting @xmath258 on @xmath93 it is very convenient to associate to the link @xmath199 a coloured chord diagram c @xcite which will encode intersections of the loops
. this chord diagram is constructed as follows : the projection of the link on @xmath90 defines @xmath259 colored loops and @xmath260 coloured open paths ( with boundary on the boundary of @xmath90 ) on @xmath90 with transverse intersections , this configuration of paths defines uniquely a coloured chord diagram by the standard construction .
let us denote by @xmath261 the coloured circles and arcs of the chord diagram corresponding to loops and open paths belonging to the link ( we will call abusively `` circles '' the circles or the arcs of the chord diagrams ) . the family of coloured circles ( resp . arcs ) @xmath262 ( resp .
@xmath263 ) will be denoted @xmath264 ( resp .
@xmath265 ) .
each circle @xmath266 is oriented , we will denote by @xmath267 the intersection points of the circle @xmath266 with the chords .
we will assume that they are labelled with respect to the cyclic order defined by the orientation of the circles .
let @xmath268 , we define a relation @xmath269 on the set @xmath270 by : @xmath271 we will denote by @xmath272 the immersion of the chord diagram in @xmath90 , in particular we have @xmath273 every intersection point of the projection of @xmath199 on @xmath90 have exactly two inverse images by @xmath272 in the chord diagram and these points are linked by a unique chord .
we will denote by @xmath274 the points such that @xmath275\yi_{j}\yi_{j-1}[$ ] and @xmath276 is a vertex of the triangulation corresponding to the cutting time @xmath277 .
we will denote by @xmath278 the set of all points of type @xmath117 in the @xmath279th component , z the union of these sets , @xmath280 the subset of @xmath281 formed by the points of @xmath281 which belong to the boundary of @xmath93 let us denote by @xmath282 the family of segments forming the corresponding circle and @xmath283 the union of these families for all components .
to each segment @xmath284 $ ] we will associate two vector spaces @xmath285 and @xmath286 such that @xmath287 @xmath283 being a finite set , let us choose on it a total ordering .
this ordering allows us to define two vector spaces @xmath288 and @xmath289 @xmath290 and @xmath291 where the order in the tensor product is taken relativ to it .
let @xmath292 and @xmath293 and assume that @xmath294 we will use as a shortcut the notation : @xmath295 we define the space @xmath296 by : @xmath297 if @xmath298 is an element of @xmath282 we denote by @xmath299 the canonical injection @xmath300 let us define two types of holonomy along @xmath298 : @xmath301 is defined by @xmath302 ( the right handside has already been defined so that there is no risk of confusion ) and @xmath303 is defined by @xmath304 let @xmath305 two points of @xmath306 such that @xmath307 and define the endomorphism : @xmath308 ( resp @xmath309 ) if @xmath310 ( resp if @xmath311 by : @xmath312 let @xmath313 be any fixed strict total order on @xmath314 we define a family @xmath315 of elements of @xmath316 as follows : let @xmath317 be any pair of points of @xmath270 such that @xmath318 we can always assume ( otherwise we just exchange @xmath105 and @xmath319 ) that @xmath320 = \ { ll r^(y^-y^+)-1 + r^(y^+y^+ ) r^(y^-y^+)-1 r^(y^-y^+)-1 . = \ { ll r^(y^-y^+)-1 + r^(y^-y^+)-1r^(y^-y^-)r^(y^-y^+)_y^-y^- .
this definition defines completely the elements @xmath321 for @xmath322 similarly if @xmath117 is an element of @xmath281 we will define @xmath323 we have now defined the framework necessary to associate to @xmath199 an element of @xmath125 denoted @xmath324 which generalizes the construction of wilson loops .
we denote by @xmath325 be the strict lexicographic order induced on @xmath270 by the enumeration of the connected components of @xmath199 and a choice of departure point for each of these components , i.e @xmath326 if and only if @xmath327 or ( @xmath328 and @xmath329 ) let @xmath149 be a connected piece of one of the @xmath266 .
let us choose for simplicity @xmath330 $ ] , we will denote the holonomy associated to it , by _
n ] ( < _ l ) ( < _ l ) u_[_1 _ 1 ] , where @xmath331 we will denote by @xmath332 the holonomy associated to the entire circle @xmath266 .
we will also define the permutation operator : @xmath333 ( where the order is given by the order of vertices along @xmath149 ) and @xmath334 will denote @xmath335 to each link in @xmath251 $ ] we associate an element @xmath336 by the following procedure : let us denote by @xmath337 the element _
l= _ _ i=1^p _ i=1^p ; where @xmath338 : the element associated to the link @xmath199 is defined by w_l = tr__x z z^v_x^+w_l where @xmath339 means the partial trace over the space @xmath340 after the natural identification @xmath341 this element satisfies important properties described by the following theorem @xcite :
let @xmath199 be a link satisfying the set of assumptions , then @xmath336 does not depend on the labelling of the components nor does it depend on the choice of departure points of the components . as a result w is a function on the space of links with values in @xmath342 moreover this
mapping is invariant under the coaction of the gauge group at a vertex interior to the surface .
if @xmath199 and @xmath250 are two links , we have the morphism property @xmath343 our principal aim is the computation of the correlation function defined in an obvious way : the correlation function of the link @xmath199 considered as immersed in @xmath196 $ ] is simply defined by : _
q - ym()= _
l ^intdh(u_l ) w_l _ f _ f the observable associated to @xmath199 will be denoted by @xmath344 this element of @xmath195 depends only on the regular isotopy class of the link @xmath215 i.e it satisfies the reidemeister moves of type 0,2,3 .
this fact was established in @xcite .
+ moreover , let @xmath199 be as usual a link in @xmath345 $ ] and @xmath149 the set of projected curves on @xmath90 and let @xmath346 be another link whose projection @xmath347 differs from @xmath149 by a move of type i applied to a curve colored by @xmath26 , we have the following relation : _
l^=v_^1 w_l the expectation value of a wilson loop on a riemann surface can be considered as an invariant associated to a ribbon glued on the surface with the blackboard framing .
our first aim is the computation of the invariants associated to links drawn on a closed riemann surface embedded in @xmath348 to realize this program we want to decompose the computation by introducing surfaces with boundaries and links drawn on them , already called `` striped surfaces '' , and by describing the gluing operation of the latter .
this decomposition allows us to reduce the striped surface to the gluing of the following objects , called `` elementary blocks '' : 1 . * the cups * 2 . * the caps * 3 . * the ( n , m)(n+m ) trinions * 4 . *
the ( n+m)(n , m ) trinions * 5 . * the free propagation of n strands * 6 . *
the propagation of n strands with one overcrossing * 7 . *
the propagation of n strands with one undercrossing * 8 . *
the ( n-2)(n ) creation * 9 . *
the ( n)(n-2 ) annihilation * these objects are described in the following figure + the correlation functions can be put in a more convenient form to reduce the computation to the gluing of elements associated to elementary blocks .
let us consider an element @xmath97 of @xmath234 and an element @xmath149 of @xmath94 then p past(l ) _ p = _
p. this result is a trivial consequence of the choice of ciliation and of the commutation properties developed in @xcite from now the order induced by the orientation of the link will not be convenient anymore , prefering time ordering we will introduce the vector spaces @xmath349 and @xmath350 ( `` after '' and `` before '' ) rather than @xmath351 and @xmath352 let @xmath26 denote the representation associated to the circle where x is taken , then @xmath353 if @xmath106 is in @xmath354 , then @xmath355 and we will introduce the canonical identification maps @xmath356 and @xmath357 if @xmath106 is in @xmath358 , then @xmath359 and we will introduce the canonical maps @xmath360 and @xmath361 we can define a new holonomy by @xmath362}$ ] : @xmath363}&= & id_{(v_{x^a},v_{x^+ } ) } \gua _ { [ x , y ] } id_{(v_{y^b},v_{y^- } ) } , \mbox { if } x \in { \cal v}_+ , y \in { \cal v}_+\\ & = & { \psi_{\bar{\alpha}\alpha}^{0}}_{(v_{x^b},v_{x^+ } ) } { \gua } _ { [ x , y ] } { \phi_{0}^{\bar{\alpha}\alpha}}_{(v_{y^a},v_{y^-})}= { \guabar } _ { [ y , x ] } , \mbox { if } x \in { \cal v}_- , y \in { \cal v}_-\\ & = & { \psi_{\bar{\alpha}\alpha}^{0}}_{(v_{x^b},v_{x^+ } ) } { \gua } _ { [ x , y ] } id_{(v_{y^b},v_{y^- } ) } , \mbox { if } x \in { \cal v}_- , y \in { \cal v}_+\\ & = & id_{(v_{x^a},v_{x^+ } ) } { \gua } _ { [ x , y ] } { \phi_{0}^{\bar{\alpha}\alpha}}_{(v_{y^a},v_{y^- } ) } , \mbox { if } x \in { \cal v}_+ , y \in { \cal v}_- \end{aligned}\ ] ] despite its apparent complexity , this definition has a very simple meaning .
it describes the usual fact that a strand in the direction of the past coloured by a representation @xmath26 can be described by a strand in the direction of the future coloured by a representation @xmath364 we will denote in the following : a_k = _ j=1^card(l__k t _
k+1 ) ( ( _ p present(l^k_j ) _ p ) u^#_l^k_j ) , the elements ( @xmath365 if @xmath97 does not belong to a crossing and @xmath366 if @xmath97 and @xmath367 cross themselves ) will be called `` square plaquettes '' elements in the following .
+ we will also use the following permutation operator : ^_k = _
j=1^card(v__k ) ( _ i = card(v__k+1)^1 p_(x^k_j)^a,(x^k+1_i)^a ) using these definitions , the element associated to the striped surface can be put in a form which respects the ordering induced by the time order : _
l = v_^12_x ^int ( x^b , x^a ) tr__x ^intv_x^a((_k=1^n^_k)(_k=0^na_k ) ) we begin with the ordering of the holonomies attached to the link . using the commutation relations and the properties of the @xmath5 matrix we obtain : @xmath368 now the commutation lemma gives easily : ( _ p _ p)(_k=0^n _ j=1^card(l__k t _
k+1 ) u_l^k_j)=_k=0^n_j=1^card(l__k t _
k+1 ) ( ( _ p present(l^k_j ) _ p ) u_l^k_j ) , and with the definition of @xmath369 we then obtain : @xmath370 this lemma leads us to a new definition of elements associated to `` striped surfaces '' which is based on gluing chronologically ordered elementary blocks .
let us consider a `` striped surface '' @xmath371 .
let us define an element corresponding to @xmath371 ( which will be denoted by @xmath372 ) by the following rules : + * if @xmath371 is an elementary block @xmath373 , the element of the gauge algebra associated to it is : _
b_1=_l _ ] t , t [ dh(u_l ) _
j=1^card(l_1 ) ( ( _ p present(l^1_j ) _ p ) u^#_l^1_j ) v_^12_x ^t ( x^b , x^a ) * if @xmath371 is a disjoint union of @xmath43 elementary blocks placed between @xmath374 and @xmath375 then the element of the algebra associated to @xmath371 is obviously the product of the elements associated to each elementary block , the order between them being irrelevant because they are commuting .
+ * if there exists a time @xmath376 between @xmath374 and @xmath375 such that @xmath371 is obtained by gluing two `` striped surfaces '' @xmath377 and @xmath378 placed respectively between @xmath374 and @xmath376 , and between @xmath376 and @xmath379 the element associated to @xmath371 will be defined by : _
+ l&=&a__1+l_1 _ _ 2+l_2 + & = & _
dh(u_l ) tr__x ^tv_x^a(^ta__1+l_1 a__2+l_2 ) ( the canonical choice of ciliation defined for any striped surface is obviously compatible with the gluing operation ) + these properties give us a new way to compute the invariants associated to links on a closed surface : _ q - ym()= a_+l . from now
the computation of the correlation functions is reduced to the computation of elements associated to elementary blocks .
after some definitions we will give the result of the explicit computation of these elements .
let us consider a striped surface @xmath371 .
a connected component of its `` in state '' is a simple loop @xmath380 $ ] oriented in the inverse clockwise sense with @xmath381 strands going through it at each @xmath382 in the direction of the past with a representation @xmath383 + ( a strand in the direction of the future with a representation @xmath26 is reversed to the past by changing its representation in @xmath29 ) .
the ciliation at each @xmath382 is chosen as in the general construction of striped surfaces . then , choosing @xmath384 other representations @xmath385 we define @xmath386 and @xmath387 : + if there is at least one emerging strand through @xmath388 , @xmath389}\psi^{\beta_1}_{\beta_2 \alpha_2 } \cdots \psi^{\beta_{n-1}}_{\beta_n\alpha_{n}}\guf{\beta_n}_{[x_{n},x_1 ] } \rmff{\beta_n}{\alpha_1}\muf{\beta_n})\\ % % following line can not be broken before 80 char \instate{\beta_n}{\alpha_n}{x_n}{\beta_{n-1}}{\alpha_{n-1}}{x_{n-1}}{\beta_1}{\alpha_1}{x_1}&= & v_{\beta_n}tr_{\v{\beta_n}}(\muf{\beta_n}\rff{\beta_n}{\alpha_1 } \guf{\beta_n}_{[x_1,x_{n}]}\phi^{\beta_n \alpha_{n}}_{\beta_{n-1 } } \cdots \phi^{\beta_2 \alpha_2}_{\beta_1}\guf{\beta_1}_{[x_2,x_{1}]}\phi^{\beta_1 \alpha_1}_{\beta_n } ) \end{aligned}\ ] ] here and in the following we will often forget the multiplicities @xmath390 for readibility .
+ if the connected component of this `` in state '' has no emerging strand we will define @xmath391 and @xmath392 to be : ( _ 0)=w^_0_c^-1(_0)=w^_0_c the properties of the latter objects are described in the following lemma .
the properties of the in and out states are generalizations of those of wilson loops . + cyclicity & = & + & = & + gauge transformation ( ) & = & _ i=1^n
s(_x_i ) + ( ) & = & _ i=1^n _ x_i + scalar product _
l dh(u_l)&&tr__iv_x_i^a(_c + & & ) = _
i=1^n__i,_i the cyclicity property is not completely obvious .
we give here a detailed proof of this fact : & & = + & & = v^-1__ntr_v__n ( ^_n__1 _ 1_[x_1,x_2]^_1__2 _ 2_[x_2,x_3 ] ^_n-1__n_n_[x_n , x_1 ] ) + & & = v^-1__ntr_v__1 ( _ [ x_1,x_2]^_1__2 _ 2 _ [ x_2,x_3 ] ^_n-1__n_n_[x_n , x_1 ] ^_n__1 _ 1 v__2 ) + & & = v^-1__ntr_v__1__1 ( p_v__1,v__1_[x_1,x_2]^_1__2 _ 2 _ [ x_2,x_3 ] ^_n-1__n_n_[x_n , x_1 ] ^_n__1 _ 1 v__2 ) + & & = _
( i),(j)v^-1__ntr_v__1 ( ^_1__2 _ 2 b_(i)^_2 _ [ x_2,x_3 ] ^_n-1__n_n_[x_n , x_1 ] s(a_(j)^_n ) ^_n__1 _ 1 v__2 b_(j)^_1_[x_1,x_2 ] s^2(a_(i)^_1 ) ) + & & = _ ( i)v^-1__ntr_v__1 ( s^2(a_(i)^_1)b_(i)^_1 ^_1__2 _ 2 _ [ x_2,x_3 ] ^_n-1__n_n_[x_n , x_1]^_n__1 _ 1 _ [ x_1,x_2 ] v__n ) + & & = _ ( i)v__1 ^ -1tr_v__1 ( ^_1__2 _ 2 _ [ x_2,x_3 ] ^_n-1__n_n_[x_n , x_1]^_n__1 _ 1 _ [ x_1,x_2 ] _ _ 1 ) + & & =
the gauge transformation is very simple to derive using the decomposition rules of the elements of the group and we can proove the scalar product property using simply the integration formula and the unitarity relations of clebsch - gordan maps . & & _
l dh(u_l)tr__iv_x_i^a(_c + & & ) = + & & = _
l dh(u_l ) v^-1__nv__ntr_v__n v__n ( ^_n m_n__1 _ 1_[x_1,x_2]^_1 m_1__2 _ 2 ^_n-1 m_n-1__n_n _ [ x_n , x_1 ] + & & _ [ x_1,x_n ] ^-1 ^_n _ n__n-1 m_n-1 ^_2 _ 2__1 m_1_[x_2,x_1]^_1 _ 1__n m_n)= + & & = _ l dh(u_l ) tr_v__n ( ^_n m_n__1 _ 1_[x_1,x_2]^_1 m_1__2 _ 2 ^_n-1 m_n-1__n_n ^_n _
n__n-1 m_n-1 + & & ^_2 _
2__1 m_1_[x_2,x_1]^_1 _
1__n m_n)= + & & = tr_v__n ( ) _ i=1^n-1(__i,_i_m_i , m_i)=_i=1^n(__i,_i_m_i , m_i ) .
+ this ends the proof of the lemma .
+ all elements associated to elementary blocks can be computed in terms of `` in '' and `` out '' states of the latter form .
let us give here the expression of the elements associated to the elementary blocks enumerated before : + @xmath393 { \cal o}(\beta_0)\\ & & { \bf { \cal a}^{elem}_{cap}}= \sum_{\beta_0 } [ d_{\beta_0 } ] { \cal i}(\beta_0)\\ & & { \bf { \cal a}^{elem}_{(n , m)(n+m ) tri.}}= \sum_{\beta_1,\cdots , \beta_{n+m } } [ d_{\beta_{n+m}}]^{-1 } % % following line can not be broken before 80 char \intri{{\beta'}_n}{{\alpha'}_n}{{x'}_n}{{\beta'}_1}{{\alpha'}_1}{{x'}_1}\intri{{\beta''}_n}{{\alpha''}_n}{{x''}_n}{{\beta''}_1}{{\alpha''}_1}{{x''}_1 } \times\\ & & \times\outtri{{\beta}_n}{{\alpha}_n}{{x}_n}{{\beta}_1}{{\alpha}_1}{{x}_1 } \delta_{{\beta'}_n , { \beta''}_m , { \beta}_{n+m } , \beta_m } \prod_{k=1}^{n}\delta_{\alpha_{m+k } , { \alpha'}_{k } } \prod_{k=1}^{m}\delta_{\alpha_{k } , { \alpha''}_{k } } \prod_{k=1}^{n-1}\delta_{\alpha_{m+k } , { \alpha'}_{k } } \prod_{k=1}^{m-1}\delta_{\alpha_{k } , { \alpha''}_{k}}\nonumber\\ & & { \bf { \cal a}^{elem}_{(n+m)(n , m ) tri.}}= \sum_{\beta_1,\cdots , \beta_{n+m } } [ d_{\beta_{n+m}}]^{-1 } \intri{{\beta}_n}{{\alpha}_n}{{x}_n}{{\beta}_1}{{\alpha}_1}{{x}_1 } % % following line can not be broken before 80 char \outtri{{\beta'}_n}{{\alpha'}_n}{{x'}_n}{{\beta'}_1}{{\alpha'}_1}{{x'}_1}\times\\ & & \times \outtri{{\beta''}_n}{{\alpha''}_n}{{x''}_n}{{\beta''}_1}{{\alpha''}_1}{{x''}_1 } \delta_{{\beta'}_n , { \beta''}_m , { \beta}_{n+m } , \beta_m } \prod_{k=1}^{n}\delta_{\alpha_{m+k } , { \alpha'}_{k } } \prod_{k=1}^{m}\delta_{\alpha_{k } , { \alpha''}_{k } } \prod_{k=1}^{n-1}\delta_{\alpha_{m+k } , { \alpha'}_{k } } \prod_{k=1}^{m-1}\delta_{\alpha_{k } , { \alpha''}_{k}}\nonumber\\ & & { \bf { \cal a}^{elem}_{free}}=\sum_{\beta_1,\cdots,\beta_n } \intri{\beta_n}{\alpha_{n}}{x_{n}}{\beta_1}{\alpha_1}{x_1 } \outtri{{\beta'}_n}{{\alpha'}_{n}}{{x'}_{n}}{{\beta'}_1}{{\alpha'}_1}{{x'}_1 } % % following line can not be broken before 80 char \prod_{i=1}^{n}\delta_{\beta_i,{\beta'}_i}\prod_{i=1}^{n}\delta_{\alpha_i,{\alpha'}_i}\\ & & { \bf { \cal a}^{elem}_{creation}}=\sum_{{\beta'}_1,\cdots,{\beta'}_n } % % following line can not be broken before 80 char \incross{\beta_n}{\alpha_{k+2}}{x_{k+2}}{\beta_k}{\alpha_{k-1}}{x_{k-1}}{\alpha_1}{x_1 } % % following line can not be broken before 80 char \outtri{{\beta'}_n}{{\alpha'}_{n}}{{x'}_{n}}{{\beta'}_1}{{\alpha'}_1}{{x'}_1}\times\\ % % following line can not be broken before 80 char & & \times\prod_{i = k+2}^{n}(\delta_{\beta_i,{\beta'}_i}\delta_{\alpha_i,{\alpha'}_i } ) % % following line can not be broken before 80 char \delta_{{\beta'}_{k+1},\beta_k,{\beta'}_{k-1}}\delta_{{\alpha'}_{k+1},{\bar{\alpha'}}_{k } } ( v_{{\beta'}_k } v_{{\beta}_k})^{{1\over 2 } } ( \frac{[d_{{\beta'}_k}]}{[d_{{\beta}_k}]})^{{1\over 2 } } n^{{\beta'}_k,{m'}_k}_{{\beta'}_{k-1}\alpha_k } \prod_{i=1}^{k-1 } \delta_{\alpha_i,{\alpha'}_i } \prod_{i=1}^{k-2 } \delta_{\beta_i,{\beta'}_i}\nonumber\\ & & { \bf { \cal a}^{elem}_{annihil . } } = \sum_{{\beta'}_1,\cdots,{\beta'}_n } \intri{{\beta'}_n}{{\alpha'}_{n}}{{x'}_{n}}{{\beta'}_1}{{\alpha'}_1}{{x'}_1 } % % following line can not be broken before 80 char \outcross{\beta_n}{\alpha_{k+2}}{x_{k+2}}{\beta_k}{\alpha_{k-1}}{x_{k-1}}{\alpha_1}{x_1 } \times\\ % % following line can not be broken before 80 char & & \times\prod_{i = k+2}^{n}(\delta_{\beta_i,{\beta'}_i}\delta_{\alpha_i,{\alpha'}_i } ) % % following line can not be broken before 80 char \delta_{{\beta'}_{k+1},\beta_k,{\beta'}_{k-1}}\delta_{{\alpha'}_{k+1},{\bar{\alpha'}}_{k } } ( v_{{\beta'}_k } v_{{\beta}_k})^{-{1\over 2 } } ( \frac{[d_{{\beta'}_k}]}{[d_{{\beta}_k}]})^{{1\over 2 } } n^{{\beta'}_k,{m'}_k}_{{\beta'}_{k-1}\alpha_k } \prod_{i=1}^{k-1 } \delta_{\alpha_i,{\alpha'}_i } \prod_{i=1}^{k-2 } \delta_{\beta_i,{\beta'}_i}\nonumber\\ & & { \bf { \cal a}^{elem}_{overcross.}}=\sum_{\beta_1,\cdots,\beta_n,{\beta'}_k } \incross{\beta_n}{\alpha_{k+1}}{x_{k+1}}{\beta_k}{\alpha_k}{x_k}{\alpha_1}{x_1 } % % following line can not be broken before 80 char \outcross{{\beta'}_n}{{\alpha'}_{k+1}}{{x'}_{k+1}}{{\beta'}_k}{{\alpha'}_k}{{x'}_k}{{\alpha'}_1}{{x'}_1}\times\nonumber\\ & & \times\prod_{i\not= k}\delta_{\beta_i,{\beta'}_i}\prod_{i\not = k , k+1}\delta_{\alpha_i,{\alpha'}_i } \frac % % following line can not be broken before 80 char { tr_q(\psi_{\beta_{k-1}\alpha_{k-1}}^{\beta_{k-2}}\psi_{\beta_{k}\alpha_{k}}^{\beta_{k-1}}\rtff{\alpha_k}{\alpha_{k-1 } } % % following line can not be broken before 80 char \phi^{\beta_{k}\alpha_{k-1}}_{{\beta'}_{k-1}}\phi^{{\beta'}_{k-1}\alpha_{k}}_{\beta_{k-2}})}{[d_{\beta_{k-2}}]v_{{\beta}_k}^{{1\over 2}}v_{{\beta'}_k}^{-{1\over 2 } } } \delta_{\alpha_{k+1},{\alpha'}_k}\delta_{\alpha_{k},{\alpha'}_{k+1}}\\ & & { \bf { \cal a}^{elem}_{undercross.}}=\sum_{\beta_1,\cdots,\beta_n,{\beta'}_k } \incross{\beta_n}{\alpha_{k+1}}{x_{k+1}}{\beta_k}{\alpha_k}{x_k}{\alpha_1}{x_1 } % % following line can not be broken before 80 char \outcross{{\beta'}_n}{{\alpha'}_{k+1}}{{x'}_{k+1}}{{\beta'}_k}{{\alpha'}_k}{{x'}_k}{{\alpha'}_1}{{x'}_1}\times\nonumber\\ % % following line can not be broken before 80 char & & \times\prod_{i\not = k}\delta_{\beta_i,{\beta'}_i}\prod_{i\not = k , k+1}\delta_{\alpha_i,{\alpha'}_i } \frac % % following line can not be broken before 80 char { tr_q(\psi_{\beta_{k-1}\alpha_{k-1}}^{\beta_{k-2}}\psi_{\beta_{k}\alpha_{k}}^{\beta_{k-1}}\rtmff{\alpha_k}{\alpha_{k-1 } } % % following line can not be broken before 80 char \phi^{\beta_{k}\alpha_{k-1}}_{{\beta'}_{k-1}}\phi^{{\beta'}_{k-1}\alpha_{k}}_{\beta_{k-2}})}{[d_{\beta_{k-2}}]v_{{\beta}_k}^{{1\over 2}}v_{{\beta'}_k}^{-{1\over 2 } } } \delta_{\alpha_{k+1},{\alpha'}_k}\delta_{\alpha_{k},{\alpha'}_{k+1}}\end{aligned}\ ] ] + the result for @xmath394 and @xmath395 is clearly given by the boltzmann weight .
the computation of the other elements need a careful description .
the idea is very simple .
we first absorb each link segment in the attached boltzmann weight to put each elements associated to `` square plaquettes '' in a same practical form where all edges of the boundary appear one and only one time and always in the same order .
this form allows us to reduce the gluing of `` plaquettes '' to one commutation plus one integration only .
+ in the case of an empty square plaquette the corresponding boltzmann weight is already in the reduced form . + for example the square plaquette element involved in the computation of a free propagation is given by : & & _ [ x_n , x_n , x_n-1,x_n-1]_[x_n-1,x_n-1]= _ _ n-1_n-2[d__n-1]__n-1 _ n-1_n-2 ^ -1 + & & tr_v__n-1(_[x_n , x_n ] _ [ x_n
, x_n-1]^_n-1 _ n-1__n-2_[x_n-1,x_n-1]__n-1 _ n-1^_n-2 _ [ x_n-1,x_n])where the notations are summarized on the following figure : in the case of a creation element for example , with a similar computation and the same notations , we have : & & _ [ x_n , x_n , x_n-1,x_n-1]_[x_n , x_n-1]^#= + & & = _ _ n,_n ( [ d__n][d__n])^1 2 n^_n _
n__ntr_v__n ( _ [ x_n , x_n]_[x_n , x_n-1]_[x_n-1,x_n-1 ] + & & _ _n _ n^_n _ [ x_n , x_n-1]__n
|_n^_n__n |_n _n ) when the square contains a crossing the reduction is less obvious and is given by the following lemma : with the notations of the figure : + it was shown in our last work @xcite that : & & dh(u_[x_n-1,y])dh(u_[y , x_n])dh(u_[x_n , y])dh(u_[y , x_n-1 ] ) + & & _ [ y x_n x_n-1]_[y x_n-1 x_n-1]_[y x_n-1 x_n ] _ [ y x_n x_n]_[x_n y x_n-1 ] _ [ x_n-1 y x_n]= + & & = _ _ n[d__n]()^1 2tr_q(__k-1_k-1^_k-2__k_k^_k-1 ^_k_k-1__k-1 ^_k-1_k__k-2)[d__k-2 ] tr_v__n ( _ [ x_n x_n ] + & & ^_n _ n__n-1_[x_n x_n-1 ] ^_n-1 _ n-1__n-2_[x_n-1 x_n-1 ] ^_n-2__n-1 _
n_[x_n-1 x_n]__n-1_n ^_n-1 ) and the analog relation for the undercrossing .
now all square plaquettes elements are in the reduced form .
then the problem of computing the whole elementary block element is reduced to the gluing of elements associated to each of the square plaquettes given in the reduced form , i.e. a commutation @xmath396 an integration .
now , let us give a careful computation in the case of @xmath397 and the other ones , very similar to this one , will be led to the reader .
the notations are summarized on the figure : ( * remark : * in the following computation we forget again the multiplicities of representations in all decompositions , but we must take care of them ... ) let us first describe the gluing of two square plaquettes : & & dh(u_[x_n-1,x_n-1 ] ) _ [ x_n , x_n , x_n-1,x_n-1]_[x_n-1,x_n-1 ] _ [ x_n-1,x_n-1,x_n-2,x_n-2]_[x_n-2,x_n-2]= + & & = dh(u_[x_n-1,x_n-1 ] ) _ _ n-1,_n-2,_n-2,_n-3[d__n-1][d__n-2]__n-1 _ n-1_n-2 ^ -1__n-2 _ n-2_n-3 ^ -1 + & & tr_v__n-1(_[x_n , x_n]_[x_n , x_n-1]^_n-1 _ n-1__n-2_[x_n-1,x_n-1]__n-1 _ n-1^_n-2 _ [ x_n-1,x_n ] ) + & & tr_v__n-2(_[x_n-1,x_n-1]_[x_n-1,x_n-2]^_n-2 _ n-2__n-3_[x_n-2,x_n-2]__n-2 _ n-2^_n-3_[x_n-2,x_n-1])= + & & = _ ( i),(j)dh(u_[x_n-1,x_n-1 ] ) _ _ n-1,_n-2,_n-2,_n-3[d__n-1][d__n-2]__n-1 _ n-1_n-2 ^ -1__n-2 _ n-2_n-3 ^ -1 + & & tr_v__n-1v__n-2(_[x_n , x_n]_[x_n , x_n-1]^_n-1
_ n-1__n-2a_(i)^^-1 + & & _ [ x_n-1,x_n-1 ] _ [ x_n-1,x_n-1]_[x_n-1,x_n-2 ] ^_n-2 _ n-2__n-3_[x_n-2,x_n-2]__n-2 _ n-2^_n-3 + & & _ [ x_n-2,x_n-1]s(a_(j)^_n-2 ) _ _ n-1 _ n-1^_n-2 b_(i)^_n-1b_(j)^_n-1_[x_n-1,x_n])= + & & = _ ( i),(j)__n-1,_n-2,_n-2,_n-3[d__n-1][d__n-2]__n-1 _ n-1_n-2 ^ -1__n-2 _ n-2_n-3 ^ -1 + & & tr_v__n-1(_[x_n , x_n]_[x_n , x_n-1]^_n-1 _ n-1__n-2_[x_n-1,x_n-2 ] ^_n-2 _ n-2__n-3_[x_n-2,x_n-2 ] + & & _ _ n-2 _ n-2^_n-3_[x_n-2,x_n-1]s(a_(j)^_n-2 )
s^2(a_(i)^ ) _ _ n-1 _ n-1^_n-2 b_(i)^_n-1b_(j)^_n-1_[x_n-1,x_n])= + & & = _ _ n-1,_n-2,_n-3[d__n-1]__n -1_n-1_n-2 ^ -1__n-2 _ n-2_n-3 ^ -1 tr_v__n-1(_[x_n , x_n]_[x_n , x_n-1 ] + & & ^_n-1 _ n-1__n-2_[x_n-1,x_n-2 ] ^_n-2 _ n-2__n-3_[x_n-2,x_n-2]__n-2 _ n-2^_n-3_[x_n-2,x_n-1 ] + & & _ _ n-1 _ n-1^_n-2_[x_n-1,x_n ] ) + in the same way we can glue @xmath398 boltzmann weights and links . as a result we have obviously : & & _ i=1^n-1 dh(u_[x_i , x_i ] ) _
i=1^n-1(_[x_i+1,x_i+1,x_i , x_i]_[x_i , x_i])= + & & = _ _ n-1,,_2,_1[d__n-1]_i=2^n-1__i _ i_i-1 ^ -1__1 _ 1_n^-1 tr_v__n-1(_[x_n , x_n]_[x_n , x_n-1]^_n-1 _
n-1__n-2 + & & ^_1 _ 1__n_[x_1,x_1]__1 _
1^_n__n-1 _ n-1^_n-2_[x_n-1,x_n ] ) + the computation of @xmath399 can be achieved by the gluing of the last boltzmann weight to obtain the cylinder with @xmath384 strands.this boltzmann weight must be changed to the square plaquette element corresponding to crossings , ... in the computations of the other blocks elements . & & _
i=1^n dh(u_[x_i , x_i ] ) _
i=1^n(_[x_i+1,x_i+1,x_i , x_i]_[x_i , x_i])= + & & = dh(u_[x_n , x_n])dh(u_[x_1,x_1 ] ) + & & ( _ _ n,_n-1[d__n]__n _ n _n-1 ^ -1tr_v__n(_[x_1,x_n ] ^_n _ n__n-1_[x_n , x_n]__n _ n^_n-1_[x_n , x_1]_[x_1,x_1 ] ) ) + & & ( _ _ n-1,,_2,_1[d__n-1]_i=2^n-1__i _ i_i-1 ^ -1__1 _ 1_n^-1 tr_v__n-1(_[x_n , x_n]_[x_n , x_n-1 ] ^_n-1 _ n-1__n-2 + & & ^_1 _ 1__n_[x_1,x_1]__1 _ 1^_n__n-1 _ n-1^_n-2_[x_n-1,x_n]))= + & & = dh(u_[x_n , x_n])dh(u_[x_1,x_1 ] ) _ _n-1,_n,,_2,_1[d__n-1][d__n ] _ i=2^n-1__i
_ i_i-1 ^ -1__1 _ 1_n^-1 _ _ n _
n _n-1 ^ -1 + & & tr_v__n v__n-1(_[x_1,x_n ] ^_n _ n__n-1 a_(j)^_n-1 ^-1 _ [ x_n , x_n]_[x_n , x_n ] _ [ x_n , x_n-1 ] + & & ^_n-1 _ n-1__n-2_[x_1,x_n]b_(i)^_1^_1 _ 1__n _ _ n _ n^_n-1b_(j)^_n_[x_n , x_1 ] _ [ x_1,x_1]_[x_1,x_1]s(a_(i)^_n ) _ _ 1 _ 1^_n + & & _ _ n-1 _ n-1^_n-2_[x_n-1,x_n]))= + & & = _ _ n,,_1[d__n ] _ i=1^nv__i^-1 2tr_v__n(_[x_1,x_n ] ^_n _ n__n-1_[x_n , x_n-1]^_n-1 _ n-1__n-2_[x_1,x_n]b_(i)^_1^_1 _ 1__n + & & ^-1s(a_(i)^_n ) ) tr_v__n-1(s^2(a_(j)^_n-1)__n _
n^_n-1 b_(j)^_n _ [ x_n , x_1 ] _ _ 1 _ 1^_n__n-1 _ n-1^_n-2_[x_n-1,x_n])= + & & = _ _ 1,m_1,,_n , m_n + this concludes the computation of @xmath400 , @xmath401 , @xmath402 , @xmath403 , and @xmath404 + the computations of @xmath405 and @xmath406 need one more step .
it uses naturally the expression of @xmath407 as a basic object .
indeed we compute the element associated to the trinion by gluing one more plaquette to the cylinder with @xmath384 strands as it is shown in the following figure .
the computation is realized by the usual techniques : + & & a_(n , m)(n+m ) tri.=dh(u_[x_1,x_m+n])dh(u_[x_m+1,x_m ] ) _ [ x_1,x_m , x_m+1,x_m+n ] + & & _ _ 1,,_m+n = + & & dh(u_[x_1,x_m+n])dh(u_[x_m+1,x_m ] ) _
_m+n[d__m+n]v__m+n^-1 tr_v__m+n(_[x_m+n , x_1 ] _ [ x_1,x_m ] + & & _ [ x_m , x_m+1 ] _ [ x_m+1,x_m+n ] ) _ _ m+n,,_1 tr_v__n+m ( _ [ x_1,x_n+m]^_m+n_m+n__m+n-1 + & & _ [ x_2,x_1]^_1 _ 1__m+n ) = + & & = dh(u_[x_1,x_m+n])dh(u_[x_m+1,x_m ] ) _ _m+n,_m+n,,_1 [ d__m+n]v__m+n^-1 + & & tr_v__m+nv__n+m ( s^-1(b^_m+n_(j))^-1 ( _ [ x_m+n , x_1])_[x_1,x_n+m ] ) + & & a_(j)^_n+ma_(i)^_n+m _ [ x_1,x_m ] ^_m+n_m+1__m+n-1a_(l)^_n+m-1__m^_m+1_m+1 s^-1(b^_m_(k))^-1 + & & ( _ [ x_m , x_m+1]_[x_m+1,x_m ] ) a_(k)^_n+m_[x_m+1,x_m+n]b_(i)^_m+n s(b^_m+n_(l ) ) _ _ m-1^_m _ m + & & _ [ x_2,x_1]^_1 _ 1__m+n ) = + & & = _ _m+n,_m+n,,_1v__m [ d__m+n]^-1 _ _ m+n,_m,_m+n + & & tr_v__m ( _ [ x_1,x_m ] ^_m_m__m-1_[x_2,x_1 ] _ _ m^_1_1 ) + & & tr_v__m+n(a_(i)^_n+m _ _ m+n-1^_m+n _ m+na_(l)^_n+m-1^_m+1 _ m+1__m+n_[x_m+1,x_m+n ] s(b^_m+n_(l))b_(i)^_m+n ) + & & = + & & = _ _ 1 , , _ n+m
[ d__n+m]^-1 + & & _ _n , _ m , _ n+m , _
k=1^n__m+k , _k _ k=1^m _ _ k , _ k _ k=1^n-1__m+k , _k _ k=1^m-1 _ _ k , _ k + & & = _ _ 1 , , _ n+m [ d__n+m]^-1 + & & _ _n , _ m , _
n+m , _ m _ k=1^n__m+k , _k _ k=1^m _ _ k , _ k _ k=1^n-1__m+k , _k _ k=1^m-1__k , _ k + this ends the proof of the theorem .
+ these results can be easily generalized to the case where @xmath14 is a root of unity .
it suffices to realize the following replacements : + the expression for the in state becomes @xmath408 } s({\buildrel { \beta_n } \over { \theta^{(1)}_m}}){\buildrel { \beta_n } \over { a } } { \buildrel { \beta_n } \over { \theta^{(2)}_m}}\otimes { \buildrel { \alpha_n } \over { \theta^{(3)}_m } } \phi^{\beta_n \alpha_{n}}_{\beta_{n-1 } } \cdots\\ & & \cdots \phi^{\beta_2 \alpha_2}_{\beta_1}\guf{\beta_1}_{[x_2,x_{1 } ] } s({\buildrel { \beta_1 } \over { \theta^{(1)}_i } } ) s({\buildrel { \beta_1 } \over { \theta^{(1)}_j}}){\buildrel { \beta_1 } \over { a}}{\buildrel { \beta_1 } \over { \theta^{(2)}_j}}\otimes { \buildrel { \alpha_1 } \over { \theta^{(3)}_j } } \phi^{\beta_1 \alpha_1}_{\beta_n } { \buildrel { \beta_n } \over { \theta^{(2)}_i } } { \buildrel { \beta_n } \over { b } } s({\buildrel { \beta_n } \over { \theta^{(3)}_i } } ) s({\buildrel { \beta_n } \over { \theta^{-1 ( 3)}_k } } ) s({\buildrel { \beta_n } \over { \theta^{(3)}_l}}))\end{aligned}\ ] ] and the analog formula for the out state . the @xmath49 associated to the crossing
is changed to obtain an intertwiner .
+ the other objects remain unchanged and the summations are restricted to physical representations only .
+ we have the following corollary : if @xmath199 is a link without boundaries in @xmath409 $ ] we have for any value of @xmath14 : = rt_u_q(g)(l ) where @xmath410 is the reshetikhin - turaev s quantum invariant of coloured links .
this result has already been shown in @xcite . and generally if @xmath90 is a closed surface , the invariant associated to @xmath371 is simply a generalization to the case of a surface of the reshetikhin - kirillov invariant in the shadow world @xcite .
this theorem can be considered as a proof of the equivalence of invariants arising from chern - simons theory and reshetikhin - turaev quantum invariants .
we can also generalize our invariant to admit other objects called `` coupons '' defined to be respectively represented on the following figure : and which expressions are given by : _ ( n+k)(n)coupon=_shadow(coupon ) + a_(n)(n+k)coupon=_shadow(coupon ) with @xmath411 being defined as usual by the following rule :
in this chapter @xmath14 will be a root of unity . in the following @xmath0
is a compact orientable 3-manifold given by a simplicial complex @xmath412.let us recall standard definitions that can be found in @xcite .
a * canonical region * @xmath413 of @xmath0 is a region within which there are p non intersecting 2-cells @xmath414(the * canonical cells * ) with boundaries @xmath415 ( the * canonical curves * ) on the boundary @xmath234 of @xmath413 such that we obtain a 3-cell by cutting @xmath413 at each @xmath416 a surface @xmath234 is said to be a * canonical surface * of a 3-manifold @xmath0 if it satisfies these conditions : such a decomposition is called * canonical decomposition*. it is important to recall that , if @xmath420 is the genus of @xmath234 , in this case , @xmath421 and @xmath422 are homeomorphic to a genus @xmath420 handlebody .
* remark : * it is easy to give , for each @xmath0 , at least one canonical decomposition . to this aim ,
let us consider @xmath423 , @xmath424 , @xmath425 , @xmath426 the sets of @xmath427cells of @xmath412 .
let @xmath428 , @xmath429 , @xmath430 be respectively the middle of @xmath424 , @xmath425 , @xmath426 .
the complex @xmath431 obtained by adding the @xmath432 , the @xmath433 and the @xmath434 to the vertices of @xmath412 is called the * first derived complex of k * , its 3-simplexes are of the form @xmath435 .
the * second derived complex of k * , denoted @xmath436 is the complex complex generated from @xmath431 by adding the vertices @xmath437 middle of the @xmath438simplexes of @xmath431 .
let us denote by @xmath421 the set of all @xmath439simplexes of @xmath440 of the type @xmath441 or @xmath442 , by @xmath422 the set of all @xmath439simplexes of @xmath436 of the type @xmath443 or @xmath444 and by @xmath234 the common frontier of @xmath421 and @xmath422 . if we call @xmath445 ( resp.@xmath446 ) the linear graph generated by the 1-simplexes of @xmath412 ( resp .
its dual ) we can see that @xmath447 and @xmath422 are respectively @xmath448neighbourhood of @xmath445 and @xmath446 .
then we have that @xmath417 is a canonical decomposition , it is called the * canonical decomposition derived from the triangulation*. it is easy to check that for each canonical decomposition , there exits a triangulation of @xmath0 such that the decomposition is in fact the canonical decomposition derived from the triangulation . +
a heegaard splitting of a 3-manifold @xmath0 is a set @xmath449 where @xmath420 is a non negative integer and @xmath216 is a diffeomorphism of a genus @xmath420 surface @xmath450 such that @xmath0 is the manifold obtained by gluing two copies of the handlebody @xmath451 ( the interior of @xmath450 ) along their boundaries after having acted on one of them by f : @xmath452 a heegaard diagram is a set @xmath453 where @xmath234 is a compact connected 2-dimensional manifold of genus @xmath420 and @xmath454 ( resp.@xmath455 ) are canonical curves of @xmath421 , the region interior to @xmath234 ( resp .
canonical curves of @xmath422 the exterior of @xmath234).this data is sufficient to reconstruct an element @xmath216 of @xmath456 such that @xmath457 two heegaard diagrams are said to be equivalent if they describe homeomorphic 3-manifolds . let @xmath458 be @xmath420 other canonical curves in @xmath234 such that @xmath459 is a heegaard diagram of the sphere @xmath460 then @xmath461 is said to be an augmented heegaard diagram . we must recall that any element of the moduli space of a surface can be written as the composition of dehn twists .
a dehn twist can be described by the following replacement of a regular neighbourhood of the corresponding curve : * replace a curve by another curve isotopic to it , or to its inverse , or reembedded the canonical surface @xmath234 in a different way in @xmath460 . + * type 1 : solid handlebodies diffeomorphisms * * replace one canonical curve of the set @xmath454 ( resp .
@xmath462 ) by the composition of this curve with another one in this set . * making a dehn twist along one of the @xmath463 . + * type 2 : @xmath464 moves * * add a handle to @xmath234 and define @xmath465 ( resp .
@xmath466 ) to be the @xmath467cycle ( resp .
the @xmath468cycle ) of this handle , or erase a handle with cycles for which @xmath465 is the @xmath467cycle ( resp .
@xmath466 is the @xmath468cycle ) .
if the diagrams @xmath469 and @xmath470 , related by a finite number of singer s moves , give rise to the manifolds @xmath0 and @xmath471 then @xmath0 and @xmath471 are homeomorphic .
conversely , if @xmath469 and @xmath470 are any two heegaard diagrams whatsoever arising from a manifold @xmath0 then @xmath469 and @xmath470 are related by a finite number of singer s moves .
a more generally used description of three manifolds is `` the surgery presentation '' .
let us recall some facts about this description @xcite .
let @xmath472 be a framed link in the oriented sphere @xmath348 we can define a manifold @xmath0 by `` surgery '' from @xmath473 using the following procedure : + remove from @xmath460 pairwise disjoint tubular neighbourhoods @xmath474 of the curves @xmath475 and resew them identifying a meridian @xmath476 in @xmath477 with a curve @xmath478 which links @xmath475 exactly @xmath479 times .
+ moreover , every 3-manifold @xmath0 can be obtained from a certain framed link by this procedure @xcite .
it is relatively easy to relate the heegaard and surgery points of view @xcite .
let us consider a heegaard diagram based on a gluing diffeomorphism @xmath216 described in terms of dehn twists of the surface .
we first remark that splitting @xmath460 along @xmath234 then doing a dehn twist along a certain curve and resewing the handlebody is equivalent to do a surgery along the ribbon glued on the surface along this curve as it can be seen on the figure : let @xmath480 be a set of ribbons trivially embedded on the surface @xmath234 , @xmath481 the corresponding dehn twists .
we want also define the framed link @xmath199 defined to be the set of ribbons @xmath482 for @xmath483 we consider a partition of @xmath460 in three pieces : @xmath484 $ ] , the handlebody @xmath485 interior to @xmath486 and the handlebody @xmath487 exterior to @xmath488 let us consider the manifold @xmath489 obtained by gluing the manifolds @xmath490 , @xmath487 , @xmath491)_{i=1 \cdots n}$ ] with the gluing diffeomorphisms @xmath492 obviously the manifold @xmath489 is the manifold defined by the heegaard data @xmath493 , but it is also obvious that this manifold is that defined by the surgery data @xmath494 we will say that these surgery and heegaard presentation of the same manifold are `` related '' description of @xmath495 [ [ invariants - associated - to - heegaard - diagrams - and - lattice - q - gauge - theory ] ] invariants associated to heegaard diagrams and lattice q - gauge theory ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ let @xmath496 be an augmented heegaard diagram associated to a manifold @xmath0 then the expectation value : _
m= is an invariant of the manifold @xmath495 moreover this value is equal to the reshetikhin - turaev invariant associated to the manifold @xmath495 the normalization by the expectation value associated to the sphere is chosen to obtain an @xmath439manifold invariant equal to @xmath161 for the sphere . * remark 1 : * the latter definition of the correlation function is in fact very natural from the general construction of q - gauge theory .
indeed putting a delta function associated to a plaquette @xmath149 corresponds to imposing that any ribbon , i.e. holonomy defined in terms of the gauge fields algebra , can be displaced through @xmath149 without torsion and without changing the expectation value .
so , adding to the projector associated to the surface some delta functions corresponding to the @xmath382s , i.e. canonical curves of the interior handlebody , and , at a future time , the delta functions of the @xmath497s , i.e. canonical curves of the exterior handlebody , allows us to displace any curve through a handle of any of the two heegaard components.this is exactly what we want to do in the framework of chern - simons theory . + * remark 2 : * using the properties & & ( ) ^2= ( ) + & & ( ) ( ) = ( ) ( ) we can replace easily the correlation function by one where we put all lickorish generators rather than the canonical curves only .
this fact will be useful in the next section .
+ we are going to proove the last theorem through two lemmas describing some properties of this invariant .
the expectation value @xmath498 associated to a heegaard diagram @xmath499 of a manifold @xmath0 is invariant under any singer s
move applied to the diagram . +
* trivial moves : * + it is a fact already established that the expectation value is invariant under any isotopic deformation of any curve in the surface , simply because of the flatness condition .
+ we have the formula @xmath500 , which is exactly the second trivial move . +
* handlebodies diffeomorphisms : * + the flatness condition implies trivially the following property for any curves @xmath501 : _
c_1_c_2=_c_1_c_1 # c_2
moreover the @xmath502s ( resp.@xmath503s ) commute one with the others .
we then obtain the invariance under the first type 1 move .
+ it is easy to see , with the expression of the block element @xmath504 , that we can do the following replacement along any curve @xmath382 .
: which implies the invariance under the second type 1 move .
+ * @xmath464 moves : * + let us consider a heegaard diagram with one handle with its @xmath467 and @xmath468cycles .
we cut the surface along a certain @xmath505cell to obtain a torus with a puncture on which are drawn the two cycles as in the following figure .
we choose the minimal fat graph describing this object to describe the partial integration of the expectation value over the edges of the latter object .
we obtain easily : & & dh(u_[y , t])dh(u_[x , z ] ) _ [ y , t , x , z , t , y , z , x]_[x , y , z , x ] _ [ x , y , t , x ] = + & & = dh(u_[y , t])dh(u_[x , z ] ) _ [ y , t , x , z , t]_[x , y , z , x ] _ [ x , y , t , x ] + & & = dh(u_[y , t])dh(u_[x , z ] ) _ [ t , x , z]_[x , y , z , x ] _ [ x , y , t , x ] + & & = dh(u_[y , t])dh(u_[x , z ] ) _ [ t , x , y , z]_[x , y , z , x ] _ [ x , y , t , x ] + & & = _ [ t , x , y , z ] + the last line is easily obtained by using the property that the integration just `` pick '' the zero component associated to a link .
the latter result establishes the invariance under the type 2 singer move .
this ends the proof of the lemma and shows that the expectation value is an invariant of the manifold @xmath495 for any augmented heegaard diagram @xmath506 describing a manifold @xmath0 there exists a framed link @xmath199 which is a surgery data describing the same manifold @xmath507 and verifying : = rt(m ) where rt is the reshetikhin turaev invariant of the manifold computed from @xmath508
the trick already used in the proof of the invariance under the second singer move can be used also here .
we first use a natural property of delta functions that can be described by the following figure : _ c_2_c_1=_c_2_c_1 to transform the correlation function in a new one + @xmath509)w((r_i,\alpha_i)_{i=1\cdots g})\prod_{i=1}^{g}\delta_{x_i}>_{q - ym({\cal l})}$ ] where the @xmath475s are ribbons glued on the surface with the same framings and knotted in @xmath460 in the same way as the @xmath497s but with a support now included in the area described in the following figure : using now the usual flatness property ( [ flatness ] ) we can deform again the latter knot to put its crossings in the `` discs '' , the rest of the knot being composed of parallel strands along handles zones .
+ then we are able to do the same calculus as in the verification of the invariance under type 2 singer s move .
the integration `` picks '' again the zero component on each segment of the skeleton .
we then obtain the equality : _ q - ym(a)=__1 , , _
g(_i [ d__i])(_j i_disc_j ) with @xmath510 being the invariant associated , by our construction , to the knot contained in the j - th disc , placed on the sphere @xmath511 and with four coupons picking the zero component on the boundary of the disc . + it is then easy to see , using the equivalence already established in section ( 3 ) between our invariant on the sphere and the reshetikhin invariant of link , that the quantity @xmath510 is exactly the reshetikhin invariant associated to this framed link with coupons . + now the proof can be achieved by establishing , using the reshetikhin - turaev framework , that the latter data is a surgery data of the manifold @xmath495 let us first recall that , using the `` related '' surgery and heegaard descriptions , we can replace the set of curves @xmath497 describing the manifold @xmath0 by a link composed of the curves @xmath476 associated to the heegaard description of @xmath460 placed at a time @xmath374 and the curves @xmath512 ( with @xmath513 ) associated to the composition of dehn twists describing the heegaard gluing diffeomorphism encoded in the @xmath497s . if we compute , with the notations of reshetikhin and turaev in @xcite , the invariant associated to the framed link described before , with an insertion of two `` coupons '' for each handle picking the zero component , we obtain easily that this invariant is equal to the invariant associated to the link @xmath514 only .
this property uses trivially the fact that : _ , _1,_ntr_v_(^_n _
n-1__n-1 ^_2 _
1__1^_1 _ 0 _ _1 ^0__2 _ 1^_1__n _
n-1^_n-1)= id_v__1 id_v__n
we then obtain : = _ ( ) ( _ i[d__i ] ) rt((r_i,_i)_i=1 n ) using now the celebrated result of @xcite this object is a non trivial invariant of the 3-manifold @xmath515 it is the reshetikhin - turaev s invariant of three manifolds .
+ [ [ chern - simons - theory - on - a - lattice - and - three - dimensional - lattice - q - gauge - theory ] ] chern - simons theory on a lattice and three dimensional lattice q - gauge theory ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ we will define here a three dimensional gauge theory which extends in some sense the previous construction on a surface .
the definition of this theory is based on a choice of a simplicial presentation of the manifold which exhibits naturally a canonical decomposition of the manifold .
let us consider a 3-manifold @xmath0 given by a complex @xmath412 .
we impose here that all vertices of @xmath412 are tetravalent .
we will denote @xmath516 the dual complex of @xmath412 .
we will denote as before @xmath517 the @xmath427simplexes of @xmath412 , @xmath518 the @xmath427simplexes of @xmath516 and @xmath519 ( resp.@xmath520 ) the middle of the @xmath521 ( resp .
@xmath522 ) . let us define another tetravalent complex @xmath523 build up from the previous one as follows : a couple of @xmath519 and @xmath520 are said to be a couple of neighbours if @xmath519 is the middle of an edge of a certain @xmath524 and @xmath520 is in the middle of this @xmath525 we denote by @xmath526 the set of @xmath527simplexes defined by the set of couples of neighbour points .
we now define the @xmath528 of @xmath529 to be the middles of the elements of @xmath526 .
the @xmath530 of @xmath529 are then given by the set of couples of @xmath528 corresponding to elements of @xmath526 having one vertex in common , if this vertex is a @xmath519 ( resp .
a @xmath520 ) then this @xmath527simplex is said `` of type @xmath412 '' ( resp . `` of type @xmath516 '' ) .
now the @xmath505simplexes are defined to be of three types : one @xmath505simplex is associated to each closed curve formed by type @xmath412 @xmath527simplexes only , one to each closed curve formed by type @xmath516 @xmath527simplexes only , and one to each closed curve formed alternatively by type @xmath412 and type @xmath516 @xmath527simplexes.we will refer us to `` the @xmath531s '' , `` the @xmath532s '' , and `` the @xmath149s '' to denote respectively these three types of @xmath505simplexes . finally the @xmath439simplexes are defined in an obvious way by considering each connected region around the vertices of @xmath412 and @xmath516 .
we will denote by @xmath533 the sets of @xmath427simplexes respectively .
a piece of this new complex is shown in the following figure : if we denote by @xmath421 ( resp .
@xmath422 ) the region defined by the set of @xmath439cells associated to vertices of @xmath412 ( resp.@xmath516 ) and by @xmath234 the surface defined by the set of ps.the decomposition : @xmath534 is a canonical decomposition and @xmath529 is homeomorphic to @xmath412 .
the set formed by the elements of @xmath535,the elements of @xmath536 and all @xmath149s forms the complex l associated to the triangulation of the canonical surface l(for this reason these sets of @xmath537 and @xmath505simplexes will be also denoted respectively by @xmath538 ) as a consequence of the property that @xmath541 and @xmath542 , we can define as before the exchange algebra associated to the elements of @xmath543 by imposing the coaction of the gauge symmetry algebra at each element of @xmath535 and by choosing a cilium order on the surface.we can define as before the wilson loops attached to each closed path formed by elements of @xmath543 ( i.e. drawn on l ) and delta functions associated to each @xmath505cell . in fact we define the yang - mills weight associated to a 2-cell @xmath149 of area @xmath544 to be : ^_p=_phys(a)[d _ ] e^- _ p where @xmath545 is the quadratic casimir of the representation @xmath26 and @xmath38 is a coupling constant of the yang - mills theory .
we define the expectation value associated to any element @xmath2 of @xmath177 in the 3 dimensional q - yang mills theory to be : _ m:=
_ lk^#_1 dh(u_l ) ( _ j ^_f_j)(_pl_2^_p ) ( _ i ^_e_i ) in the limit @xmath546 this theory becomes the well known yang - mills theory on a lattice associated to a manifold @xmath495 let @xmath199 be a link drawn on the 1-skeleton @xmath543 of @xmath0 . using again the properties of the complex @xmath412 ,
@xmath199 is in fact drawn on the canonical surface and we can define @xmath336 in the framework defined in this article .
the correlation function associated to @xmath199 in the limit @xmath547 is then _ =
this formula can be considered as a description of reshetikhin - turaev invariants , i.e. of chern - simons invariants in term of a well defined lattice gauge theory and a definition of the witten s path integral formulas . | a new approach to the quantization of chern - simons theory has been developed in recent papers @xcite .
it uses a `` simulation '' of the moduli space of flat connections modulo the gauge group which reveals to be related to a lattice gauge theory based on a quantum group . after a generalization of the formalism of q - deformed gauge theory to the case of root of unity ,
we compute explicitely the correlation functions associated to wilson loops ( and more generally to graphs ) on a surface with punctures , which are the interesting quantity in the study of moduli space .
we then give a new description of chern - simons three manifolds invariants based on a description in terms of the mapping class group of a surface .
at last we introduce a three dimensional lattice gauge theory based on a quantum group which is a lattice regularization of chern - simons theory . #
1#1v #1#1l #1#1p # 1#1s # 1#1 00u u#1#1u # 1#1 g # 1#1 # 1#1 w # 1#1z # 1r^(#1 ) # 1#2#1 ^(#2 ) # 1u^(#1 ) # 1#2#1 ^(#2 ) # 1#1 # 1#2#1 # 2 r # 1#2#1 # 2 # 1#2#1 # 2 # 1#2#1 # 2 # 1#2#1 # 2 # 1#2#1 # 2 ^-1 # 1 u a # 1_#1 # 1_#1 # 1u(#1 ) # 1u(#1 ) # 1#2#3#4#5#6\{}_q # 1#2#3#4#5#6#7(.)_#7^ # 1#2#3#4#5#6#7(.)_#7^ # 1#2#3#4#5#6#7#8#9 i(#1 # 2 # 3 # 4 # 5 # 6 # 7 # 8 # 9 ) # 1#2#3#4#5#6#7#8#9 o(#1 # 2 # 3 # 4 # 5 # 6 # 7 # 8 # 9 ) # 1#2#3#4#5#6 i(#1 # 2 # 3 # 4 # 5 # 6 ) # 1#2#3#4#5#6 o(#1 # 2 # 3 # 4 # 5 # 6 ) #
1#2#3#4#5#6#7#8 i(#1 # 4 # 5 # 6 ) # 1#2#3#4#5#6#7#8 o(#1 # 4 # 5 # 6 ) |
if we plot the radio - to - optical flux ratio of quasars , we see two populations , the radio quiet and the radio loud quasars with a few sources , radio intermediate quasars ( riqs ) , between them
. whilst in total flux , riqs appear to be part of the radio - loud distribution , their low extended flux indicates that they might rather be part of the radio - weak distribution .
falcke et al .
( 1996 ) and miller et al .
( 1993 ) proposed that the riqs might be relativistically boosted radio - weak quasars .
iii zw 2 ( pg 0007 + 106 , mrk 1501 , @xmath3=0.089 ) is one of the riqs and also is one of the most extremely variable radio sources .
it was initially classified as a seyfert 1 galaxy ( e.g. , arp 1968 ; khachitikian & weedman 1974 ; osterbrock 1977 ) but was later also included in the pg quasar sample ( schmidt & green 1983 ) .
iii zw 2 most likely has a spiral host galaxy ( hutchings & campbell 1983 ; taylor et al .
it is a core - dominated agn with highly inverted synchrotron spectrum with a spectral peak due to self - absorption at 43 ghz ( falcke et al .
1999 ) in outburst and faint extended structure typical for seyfert galaxies .
iii zw 2 is variable up to a factor of 30 within two years with major flares roughly every five years ( see fig .
1 ) . in 1997 , iii zw 2 started a new outburst and we started to monitor this source with the vla and vlba .
we observed iii zw 2 in 27 epochs with the vla in six frequencies ranging from 1.4 to 43 ghz and in 6 epochs with the vlba at 15 and 43 ghz .
the spectral peak stayed constant at 43 ghz during the slow and smooth rise in flux density and we detected no structural change on vlbi scale at 43 ghz during this time ( see first three epochs of fig.3 ) . in december 1998 , the flux density started to drop rapidly , 2.5 times faster than it rose . at the same time the spectral peak dropped quickly from 43 ghz to 15 ghz during a few months ( see fig.2 ) . applying a simple equipartition jet model we predicted a very rapid expansion during this time .
indeed the fifth epoch of vlba observations showed a drastic structural change on milliarcsecond scales with an apparent expansion speed of @xmath4 ( brunthaler et al .
this expansion speed increases to @xmath5 if one considers the time range during which the drop in peak frequency occurred . after this phase of superluminal expansion and rapid spectral evolution
, the expansion stopped and the spectral evolution slowed down .
however , at 15 ghz the picture is completely different .
iii zw 2 shows a slow expansion ( @xmath2 ) during the first four epochs , no expansion during the decrease in flux density and again slow expansion when the expansion at 43 ghz stopped again ( see fig .
this apparent contradiction can be explained by optical depth effects in an inflating - balloon model.
in this model , the initial phase of the flux density rise can be explained by a relativistic jet interacting with the interstellar medium or a torus that creates a shock .
a relativistic shock was proposed earlier by falcke et al .
( 1999 ) due to synchrotron cooling times of 14 - 50 days which is much shorter than the duration of the outburst .
the ultra - compact hotspots are pumped up and powered by the jet and are responsible for the flux - density increase .
the post - shock material expands with the maximum sound speed of a magnetized relativistic plasma of @xmath6 .
since the source is optically thick at 15 ghz , one necessarily observes the outside of the source , i.e. the post - shock material expanding at the sound speed . at 43 ghz ,
the source is optically thin and one can look inside the source and see the stationary hotspots .
the rapid expansion at 43 ghz thereafter has marked the phase where the jet breaks free and starts to propagate relativistically into a lower - density medium until it stopped again .
the unique and simple structure and timescales of such outbursts within 5 years makes iii zw 2 an ideal source to study radio - jet evolution relevant also to radio galaxies , especially those that appear as csos and gpss .
iii zw 2 remains an extremely unusual object .
so far all relativistically - boosted jets with superluminal motion and typical blazars have been detected in early type galaxies ( e.g. , scarpa et al .
we detected for the first time superluminal motion in a spiral galaxy and the good agreement between structural and spectral evolution demonstrates that we are dealing with real physical expansion and not only a phase velocity .
since one has to look very carefully with frequent time sampling to detect this superluminal motion , it is possible that other seyferts and radio quiet quasars also have relativistic jets in their nuclei .
the fact that the sub - pc jet could be relativistic while they appear sub - relativistic at larger scales ( roy et al .
2000 ) , raises the question : are seyfert - jets decelerated on the pc - scale ? | so far , all relativistically boosted jets with superluminal motion have only been detected in typical radio galaxies with early type host galaxies .
we have now discovered superluminal motion in the seyfert i galaxy iii zw 2 , classified as a spiral .
the lower limit for the apparent expansion speed is @xmath0 .
spectral and spatial evolution are closely linked . before and after this rapid expansion
we have seen a period of virtually no expansion with an expansion speed less than @xmath1 . however , at 15 ghz the picture is completely different .
iii zw 2 shows slow expansion ( @xmath2 ) during the time of no expansion at 43 ghz and no expansion during the rapid expansion at 43 ghz . the difference between the two frequencies
is qualitatively explained by optical - depth effects in an inflating- balloon model , describing the evolution of radio lobes on an ultra - compact scale .
the stop - and - go behavior could be explained by a jet interacting with a molecular cloud or the molecular torus .
since iii zw 2 is also part of a sample of so- called radio - intermediate quasars ( riq ) , it confirms earlier predictions of superluminal motion for this source , based on the argument that riqs could be relativistically boosted jets in radio - weak quasars and seyfert galaxies .
psfig.sty # 1#1 0123456789 abcdef @mtlplain@loaded
= cmssi10 = cmssi10 = cmssi10 = cmmib10 = mtssi10 = mtssbx10 = mtssbi10 = mtmib10 @mtlplain@loaded = cmmib10 at 10pt = cmmib10 at 7pt = cmmib10 at 5pt = = = = " 0103 @mtlplain@loaded @mtlplain@loaded = cmbsy10 = cmbsy10 at 7pt = cmbsy10 at 5pt = = = @ = " 0@72 _ in : `` proceedings of the 5th evn symposium '' , eds
. j. conway , a. polatidis , r. booth .
_ _ onsala space observatory , chalmers technical university , gothenburg , sweden ( 2000 ) _ |
for simplicity , we only consider here one dimensional profiles @xmath79 .
we write the slope @xmath3 on scale @xmath12 as @xmath80 , where @xmath81 has long - range correlations described by the exponent @xmath58 and @xmath82 is a gaussian field with logarithmic corrections with slope @xmath83 . then , by definition , @xmath84 we want to compute the various moments of @xmath85 . for even values of @xmath86 , and assuming for simplicity that @xmath81 is gaussian that allows one to use wick s theorem , one has : @xmath87 now , since the @xmath88 s are also gaussian with a logarithmic correlation function , one has : @xmath89 where @xmath90 is a constant that accounts for the diagonal terms @xmath91 , that plays no role in the following .
let us first consider the bacry - delour - muzy ( bdm ) case where @xmath81 has only short range correlation , say even purely local : @xmath92 .
this local contribution will always exist , even in the long - range case , and will eventually have to be compared to the latter contribution in order to determine the dominant one . in the bdm case
, this imposes @xmath93 , and therefore : @xmath94 hence , approximating discrete sums by integrals in the limit @xmath95 , one has : @xmath96 setting @xmath97 , one finds by simple power - counting ( which is correct provided the resulting integral on @xmath98 s converges , i.e. when @xmath99 ; see @xcite ) : @xmath100 or @xmath101 , which is precisely the bdm result @xcite .
now , if one assumes that for large distances @xmath102 , the corresponding contribution to the @xmath51-th moment of @xmath85 reads : @xmath103 power - counting now leads to : @xmath104 as given in the text . however , one has to compare this last contribution to the bmd one , leading to the following inequality : @xmath105 independently of @xmath51 .
this last condition is well satisfied in practice , since @xmath106 and @xmath107 .
for a threshold @xmath108 for the three materials studied . ]
the distributions of height fluctuations in the materials investigated do not follow a mono - affine gaussian behavior at small scale because of the extended fat tails of @xmath23 . as the observation scale @xmath4
is decreased , these tails indeed become more pronounced , resulting in the multi - affine behavior revealed by the spectrum @xmath64 ( see fig .
[ fig_spectrum ] ) . to proceed to a quantitative analysis of the spatial distribution of the steepest slopes that contribute to the distribution tails ,
we introduce @xmath25 defined by eq .
( [ eq_w ] ) as a transformation of the original fracture map that allows for a straightforward localization of these extreme events ( see fig .
[ fig_maps_si ] ) . as the focus
is put on these steepest slopes , the new maps @xmath25 are thresholded , and the fraction @xmath109 of the total number of pixels of the original map is conserved .
we choose @xmath109 in the range @xmath110 .
the position of the steepest slopes is represented on extended data fig .
[ fig_maps_si ] for @xmath111 on the aluminum , mortar and ceramics fracture surfaces .
for the three samples considered .
the length @xmath66 is either the cluster radius of gyration @xmath67 , or its extension along the horizontal @xmath112 and vertical axis @xmath113 .
all lengths are normalized by the crossover length @xmath0 . ]
the statistical characterization of the geometry of these clusters is now detailed .
we investigate first the geometry of the isolated clusters .
figure [ fig_clusters1_si ] shows the variation of cluster area as a function of three typical length scales characterizing the cluster size , namely its extension @xmath112 along the propagation direction , its extension @xmath113 perpendicular to it and its radius of gyration @xmath67 .
these three quantities characterize equivalently the fractal geometry of the clusters since they all scale as @xmath114 , where the fractal dimension @xmath115 irrespective of the material considered .
the distributions of cluster sizes are characterized using their radius of gyration @xmath67 .
they are shown on fig .
[ fig_clusters ] of the main article , and follow power laws with exponent @xmath116 .
we investigate now the robustness of these findings towards our statistical procedure .
the role of @xmath109 on the cluster statistics is investigated on fig .
[ fig_clusters ] of the main article , and does not show any noticeable effect .
similarly , we observe that @xmath109 has a negligible effect on the cluster fractal dimension .
the effect of the scale @xmath12 on the cluster statistical characterization is illustrated on fig .
[ fig_clusters2_si ] in the case of the aluminum fracture surface .
the clusters display robust statistical properties independent of the length @xmath12 as long as one considers sufficiently large cluster sizes @xmath117 . | we show that a length scale @xmath0 can be extracted from the spatial correlations of the `` steep cliffs '' that appear on fracture surface . above @xmath0 , the slope amplitudes are uncorrelated and the fracture surface is mono - affine . below @xmath0 , long - range spatial correlation lead to a multi - fractal behavior of the surface , reminiscent of turbulent flows .
our results support a unifying conjecture for the geometry of fracture surfaces : for scales @xmath1 the surface is the trace left by an elastic line propagating in a random medium , while for scales @xmath2 the highly correlated patterns on the surface result from the merging of interacting damage cavities .
after thirty years of research , it is now well established that fracture surfaces exhibit robust universal fractal statistical properties , first reported in @xcite and recently reviewed in @xcite . yet , identifying the physical mechanisms that lead to such fractal structures is still an open problem @xcite .
the most commonly used approach to characterize the roughness of fractal cracks is to study the scaling of the off - plane height variation @xmath3 of the fracture surface with the observation scale @xmath4 .
the variance of this distribution shows a scaling law @xmath5 where @xmath6 is the so - called roughness exponent . for purely brittle failure
, the roughness exponent is reported to be @xmath7 @xcite whereas for materials that undergo damage during failure , @xmath8 @xcite .
it has been conjectured that these exponents are the signature of the fracture mechanism above and below the size of the process zone @xcite
. however , standard methods for extracting roughness exponents are not able to elicit the differences between the fracture mechanisms in the two regimes . here
we propose a different approach for characterizing crack roughness statistics by focusing on the _ local slopes _ of the fracture surfaces and their spatial correlations .
this allows us to identify unambiguously two scaling regimes : above some length scale @xmath0 , the slope amplitudes are uncorrelated and the fracture surface displays a mono - affine gaussian behavior with a roughness exponent of @xmath7 . below @xmath0
, long - range spatial correlations do appear and lead to a multi - fractal behavior of the surface .
our findings show that the presence of two distinct regimes of roughness first reported in refs .
@xcite is a generic feature of fracture surfaces and is reminiscient of the brittle mode of failure that takes place at large scale and of the damage mechanisms present in the tip vicinity . in addition , it reveals the subtle organization of crack roughness at small length scales @xmath9 , reminiscent of the phenomenology of turbulent flows @xcite .
in particular , we relate quantitatively the multi - fractal spectrum measured at these length scales with the spatial correlations of the local slopes , and show that the largest slopes organize into a network of lines or `` steep cliffs '' that exhibit universal statistics .
this new approach to the characterisation of fracture surfaces brings insights on the microscopic mechanisms at play during material failure , in particular on the mechanism of damage percolation taking place at the tip of cracks
. it also paves the way to a post - mortem measurement of the size of the crack tip damaged zone , as a promising tool to infer material toughness from the statistical analysis of fracture surfaces . and
@xmath10 for the three materials studied .
top : @xmath11 , height of the measured fracture surface .
bottom : @xmath10 transformation providing the field of local slopes computed at a scale @xmath12 . in both cases ,
the quantiles of the distribution are represented by a grey hue , the largest values being represented by the lightest hue .
@xmath10 is computed at the scales @xmath13 , @xmath14 and @xmath15 for the aluminum , mortar and ceramic fracture surfaces , respectively . ] for this work , we have selected three sample materials that show a wide range of fracture behavior , namely an aluminum alloy , a mortar , and a sintered glass beads ceramic .
the aluminum alloy specimens are aluminum 4 wt@xmath16 copper broken under uni - axial mode i tension at @xmath17 , in a semi - solid state @xcite .
the fracture surfaces are observed with a scanning electron microscope at two tilt angles and the elevation map is produced from a cross - correlation surface reconstruction technique .
the mortar fracture surface is obtained by applying four points bending under controlled displacement conditions to a notched beam @xcite .
the topography of the fracture surfaces is recorded using an optical profilometer .
the sintered glass beads ceramic fracture surfaces are obtained with a tapered double cantilever beam broken at constant opening rate @xcite . the roughness of the fractured specimen is measured using a mechanical stylus profilometer . those fracture surfaces are described by their height field @xmath18 , function of a two - dimensional in - plane vector @xmath19 , that are represented in a grayscale on the top panels of fig .
[ fig_maps ] for each material .
a first natural step in the characterization of the roughness statistics is to compute the distribution of height fluctuations at different scales . for a given increment @xmath20 of the coordinates in the average fracture plane ,
we note @xmath21 the probability distribution of an height increment @xmath22 where the sampling of the distribution is done on all admissible coordinates @xmath19 .
we also note @xmath23 the distribution of @xmath3 where the sampling is done on all admissible @xmath19 and @xmath20 such as @xmath24 .
the distribution @xmath23 at different @xmath4 is shows in a semi - logarithmic scale in fig .
[ fig_distribs ] for aluminum , mortar and ceramics fracture surface . in this semi - logarithmic representation ,
the parabolic shape of the distribution obtained for large values of @xmath4 reveals gaussian statistics .
this contrasts with the distributions observed for smaller values of @xmath4 that display fat tails .
this drastic change in the shape of the distribution indicates thata single exponent is unsufficient to describe the variations of the roughness properties with the scale of observation .
fat tail statistics also suggest the presence of over - represented large height variations over small in - plane distances that we would like to analyze further . at various scales @xmath4 for the three samples considered . ] to investigate the spatial distribution of these steep cliffs " , we introduce the quantity @xmath25 that measures the intensity of the local height variations of the fracture surface on a scale @xmath12 : @xmath26 @xmath27 is the local slope of the surface in the direction @xmath20 , and @xmath28 is chosen such that the average of @xmath25 over all @xmath19 is zero .
note that the average of the slopes is done over a circle of radius @xmath12 .
this new field @xmath25 has several interesting properties , like isotropy and robustness to measurement artifacts .
the maps of @xmath10 calculated from the off - plane height maps @xmath11 shown in fig .
[ fig_maps ] are represented on the lower panels in the same figure .
strikingly , the largest values of @xmath10 ( lighter grey ) , corresponding to the steep cliffs that populates the tails of the distribution @xmath23 , are spatially correlated and form a network of rough lines for the aluminum and the mortar fracture surfaces . for the ceramic fracture surface ,
smaller patterns are visible .
for the three materials considered .
the correlations are represented for @xmath10 computed at different scales @xmath12 .
the cut - off length @xmath0 is represented for each case . ]
the visually correlated patterns in fig .
[ fig_maps ] can be quantified by computing the spatial correlations of @xmath29 , which we further average over all directions @xcite , _
i.e. _ @xmath30 .
this quantity is shown in fig .
[ fig_correlations ] as a function of the distance @xmath4 for different observation scales @xmath12 ; @xmath31 is independent of @xmath12 whenever @xmath32 .
for the three materials considered , we clearly observe two regimes . at small @xmath4
, @xmath10 shows strong spatial correlations , which decay _ logarithmically _ with distance , _
i.e. _ @xmath33 , and extrapolates to zero for @xmath34 . for larger distances
, these correlations are zero within statistical noise . both @xmath35 and @xmath0
are found to be , to a good approximation , independent of @xmath12 provided @xmath36 .
note that @xmath35 is a dimensionless number and its value is empirically found to be quite similar for the three materials at hand : @xmath37 ( aluminum ) , @xmath38 ( mortar ) and @xmath39 ( ceramics ) .
the crossover length @xmath0 is found to be @xmath40 , @xmath41 and @xmath42 , respectively .
note that the ratio of @xmath0 to the total map size is @xmath43 and @xmath44 , respectively .
these last values confirms the visual impression conveyed by fig .
[ fig_maps ] where large correlated patterns are observed for aluminum , smaller patterns for mortar and even smaller ones for ceramics . in order to characterize further the two regimes
, we now compute the _ multifractal spectrum _ of the height fluctuations , defined through @xmath45 for the two range of length scales @xmath46 and @xmath47 .
note that the standard roughness exponent @xmath6 corresponds to @xmath48 , and @xmath49 .
for @xmath47 , we observe that @xmath50 is fairly independent of @xmath51 with a value around @xmath52 ( fig .
[ fig_spectrum](right ) ) .
this corresponds to a mono - affine behavior , _
i.e. _ a scaling that preserves the shape of the full distribution @xmath23 of height fluctuations .
this is consistent with the observation of a conserved gaussian distribution at large scales ( see fig .
[ fig_distribs ] ) , and in agreement with previous findings @xcite .
the mono - affine behavior is very clear for the mortar and the ceramic fracture surface . for the aluminum fracture surface , some residual variations of @xmath50 with @xmath51
is observed ; this behavior may be due to the rather limited extension of the large scale regime @xmath47 . for @xmath46 ,
on the other hand , we do observe a significant variation of @xmath50 with @xmath51 .
this multi - affine behavior can in fact be traced back to the logarithmic decay of the spatial correlations @xmath31 of slopes discussed above . indeed , assuming that @xmath10 is a gaussian field , and that the local slope can be written as @xmath53 with @xmath54 and @xmath55
is a _ long ranged correlated _ random variable with unit variance and @xmath56 , one derives , adapting the calculations of @xcite ( see the supplementary material @xcite ) : @xmath57 where @xmath35 is the slope of the logarithmic correlation defined above . as seen in fig .
[ fig_spectrum](left ) , the slope of the multi - fractal spectrum is indeed well captured by this simple model .
we have measured the exponent @xmath58 independently , from the spatial correlations of the _ signs _ of the local slopes in a given direction , with good agreement with the direct estimate of @xmath59 , in particular for aluminum where the scaling region is large .
we therefore claim that fracture surfaces are , on short length scales , bi - dimensional realizations of _ multifractal , persistent _ brownian motions . whereas natural realizations of multifractal brownian motions with @xmath60 have been reported in turbulent flows ( @xmath61 )
@xcite and in financial time series ( @xmath62 ) @xcite , it is to our knowledge the first time that a multifractal signal with @xmath63 is observed . the curvature of the multi - fractal spectrum seen in fig .
[ fig_spectrum ] can not be captured by eq .
( [ eq_multiscaling ] ) .
this can be traced back to the assumption that @xmath10 is a gaussian field . introducing non - trivial higher order correlations of @xmath10
that also decay logarithmically would add higher order contributions to @xmath64 .
however , these higher order correlations are difficult to measure and we lack statistics to test the model beyond the second order correlations reported here @xcite .
( right ) .
the spectrum predicted for @xmath46 by the multifractal model of eq .
( [ eq_multiscaling ] ) is represented by a straight line . ] to characterize further the spatial organization of the steepest regions and its robustness towards material specificity , we study the geometrical properties of the clusters formed by the largest values of @xmath10 , _
i.e. _ the ridge of the cliffs .
the field @xmath10 is thresholded in order to keep only a fraction @xmath65 .
these extreme events organize in a network of disconnected clusters , as suggested by fig .
[ fig_maps](bottom ) ( see also fig .
[ fig_maps_si ] of the supplementary material @xcite ) .
the extension @xmath66 of each cluster can be defined using either its extension along the horizontal or vertical axis , or its radius of gyration @xmath67 .
these three quantities are found to follow the same scaling @xmath68 with the number of pixels ( or area ) @xmath69 contained by the cluster , suggesting that these clusters have fractal geometry with dimension @xmath70 , again independently of the material considered ( see supplementary material @xcite ) . using the values of @xmath67
, we compute on fig . [ fig_clusters ] the distribution of cluster sizes for different values of the threshold @xmath65 .
for scales smaller than @xmath0 , we observe a power law distribution @xmath71 with exponent @xmath72 for all three materials this means that not only the roughness exponent @xmath6 but at least _
three _ other quantities describing the statistics of fracture surfaces are universal in the small @xmath4 regime : @xmath35 that describes the spatial correlation of the amplitude of slopes and the multifractal spectrum , @xmath73 that is the fractal dimension of the ridge of the cliffs on the fracture surface and @xmath74 that characterizes the cluster sizes distribution ( see table [ tab1 ] for a comparison of their value from one material to another ) .
this extended universality is important for at least two reasons : ( i ) it provides additional support for the conjecture that the statistics of fracture surfaces is universal , suggesting a common underlying roughening mechanism ; ( ii ) it provides important further constraints that must be abided by any theory attempting to explain the universal value of the roughness exponent @xmath8 @xcite .
.statistical parameters extracted from the fracture surfaces ( see text for details ) .
[ cols="<,^,^,^,^,^,^,^",options="header " , ] for different thresholds @xmath65 . ] what is the physical interpretation of these observations ?
we believe that the presence of spatially correlated steep cliffs on short length scales @xmath46 is a strong indication that at these scales , fracture proceeds through nucleation and coalescence of microcracks or damage cavities , as previously conjectured @xcite .
as elaborated in the context of planar cracks @xcite , the scale @xmath0 therefore provides the extent @xmath75 of the fracture process zone .
it also reflects the material toughness , since @xmath76 where @xmath77 is the typical failure stress under tension @xcite . our methodology based on the correlation of slopes measured on the fracture surfaces paves the way for a post - mortem characterization of material toughness from the statistical analysis of their fragments .
although this is still an open theoretical issue , the percolation of power law distributed microcracks should provide a rationale for the value of the small scale roughness exponent @xmath8 .
discontinuities along the fracture surface are the stigma of these coalescing cavities @xcite and this picture should be made compatible with values of the new , universal statistical indicators reported here , namely @xmath35 , @xmath73 and @xmath74 . only on large length scales @xmath47 does the notion of a continuous fracture line make sense .
continuum fracture mechanics based models describing crack fronts as an elastic interface driven in a random medium @xcite predict mono - affine gaussian fracture surfaces with @xmath78 @xcite , indeed compatible with our findings .
we thank d. bonamy , e. bouchaud , y. cao , a. hansen and s. morel for fruitful discussions .
the support from the european union through the marie curie integration grant toughbridge " is gratefully acknowledged ( lp ) . 26 natexlab#1#1bibnamefont # 1#1bibfnamefont # 1#1citenamefont # 1#1url # 1`#1`urlprefix[2]#2 [ 2][]#2 , & .
_ _ * * , ( ) . & _ _ * * , ( ) . , & .
_ _ * * , ( ) . , , , & .
_ _ * * , ( ) . , & .
_ _ * * , ( ) . , , &
_ _ * * , ( ) . , & _ _ * * , ( ) . , , & .
_ _ * * , ( ) . , , &
_ _ * * , ( ) .
_ et al . _ . _
_ * * , ( ) .
_ _ ( , ) . _ _ .
ph.d .
thesis , , ( ) . http://library.epfl.ch/theses/?nr=3795 . , & . _ _ * * , ( ) .
, , , & .
_ _ * * , ( ) . .
_ _ * * , ( ) . & & .
_ _ * * , ( ) . & .
_ _ * * , ( ) . , , & .
_ _ * * , ( ) . , , & .
_ _ * * , ( ) .
_ et al . _ . _
_ * * , ( ) . & .
_ _ * * , ( ) . & .
_ _ * * , ( ) . , , , & , .
_ _ * * , ( ) . & .
_ _ * * , ( ) .
_ et al . _ . _
_ * * , ( ) . , & .
_ _ * * , ( ) . .
_ _ * * , ( ) . , , & .
_ _ * * , ( ) . , & .
_ _ * * , ( ) . , , & .
_ _ * * , ( ) . |
luminous infrared galaxies , lirgs , defined as galaxies with @xmath1 , combine very strong star - formation ( sf ) with various levels of nuclear activity .
they are usually more diverse objects regarding morphology and mode of sf than ulirgs ( @xmath2 ) , which in the local universe tend to nearly always be remnants of major mergers with singularly centralised sf .
some lirgs also have centralised sf , but in many cases a significant part of the sf is `` extended '' ( e.g. alonso - herrero et al .
2006 ; rodrguez - zaurn et al .
it appears that high - redshift ulirgs also have a diversity of modes of strong sf present and hence the local lirgs may be a closer representation of the physical sites of dominant sf density at @xmath3 and beyond(e.g .
alonso - herrero et al .
2009 ; rodighiero et al .
2011 ; kartaltepe et al .
2011 ) .
`` extended sf '' is often defined as merely a fraction of mir or hydrogen recombination line emission coming from outside a given nuclear region , and as such it could mean either truly diffuse sf , or rather one or more widely off - nuclear , but clearly localised starbursts .
this latter scenario is playing out for example in the well - studied closest gas - rich merger system , the antennae , where the sf is dominated by an overlap region between the main nuclei .
to study the physics of such cases in detail , including disentangling various kinematic components , requires high spatial resolution , which has not often been available except for a handful of local galaxies , .
we are conducting an ongoing large survey of lirgs and ulirgs using adaptive optics nir imaging mainly with vlt / naco , and also gemini / altair / niri .
both instruments deliver images with a spatial resolution of @xmath4 .
the data are very well complemented with existing hst optical data for many of the targets , and there is also an on - going optical spectroscopic campaign with salt / rss .
the galaxies are at distances ranging from 40 to 180 mpc , span the range from starbursts to lirgs , and are at various stages of merging , interaction , or isolation .
some galaxies have been studied in more detail already here we will present analysis of two such objects , and will finish with a brief discussion of some initial results from the larger sample .
ngc 1614 is a lirg at a distance of 64 mpc ( @xmath5 ) .
optically it is classified as barred and interacting ( fig . 1 ) , and as a borderline case between a liner and hii - dominated galaxy .
it has also been classified as a composite starburst + agn .
previous studies have identified a strong nuclear star forming ring of approximately 600 pc diameter ( kotilainen et al 2001 ; alonso - herrero et al .
we observed the central 1.5 kpc region with ukirt / uist integral field unit in the l - band ( visnen et al .
this wavelength region has been shown to be ideal for disentangling agn signatures from starbursts , using both the continuum slope and the 3.3 and 3.4 @xmath0 m pah band features ( e.g. imanishi & dudley 2000 ; risaliti et al .
the method has not , however , been much used with both spatial and spectroscopic information simultaneously , i.e. with ifu , due to the rarity of suitable instrumentation .
archival acs / hst images of ngc 1614 .
north is up and east left in each image , the tick - marks are 5 `` in the top - left image and 1 '' in the right side panels . _
top - left _ : an @xmath6-band ( fw814 ) image showing the outer structures of ngc 1614 at 20 kpc scales .
the white rectangle shows the area of the same image zoomed and re - scaled at _ top - right _ , showing the more regular spiral pattern .
the higher surface density structure 9 `` sw of the nucleus is pointed out , which is the likely remnant of a companion galaxy , and possibly the trigger for the starburst core of the whole system .
the _ bottom - right _ panel shows the acs / hst @xmath7 colour map of the same region as the panel above it , where darker areas mean redder colours , i.e. higher extinction ; the main nucleus lies behind a dust feature .
the innermost 1 kpc area is marked with the white rectangles in both right - hand - side panels and is shown zoomed - in at _ bottom - left_. the left one is the @xmath6-band image , and at right we show our uist @xmath8-band continuum image , equivalent of fig . 2 . , right panel .
the bright very blue point source @xmath9 '' ne of the nucleus visible in the i - band and in the colour - map is a likely young super star cluster . ]
we detect a strong central continuum source which is slightly resolved ( @xmath10 80 pc ; fig . 1 lower left , fig .
2 ) , and a clear signature of a pah 3.3 @xmath0 m ring around the nucleus ( fig . 3 ) . parameterising the pah flux and the l - band continuum following risaliti et al .
( 2006 ) we find no evidence at all for an agn , obscured or otherwise : the continuum is flat , the pah feature is strong even within the central 30 pc resolution element , and there is no 3.4 @xmath0 m absorption ( fig . 2 )
. @xmath8-band continuum map of the ngc 1614 nuclear regions from the ifu observations showing the strong marginally resolved nuclear source .
_ right : _
l - band spectrum of the central @xmath10 700 pc area showing a flat continuum and an ew @xmath11 70 nm pah feature , typical in starbursts .
the spectrum is similar within 0.3 " of the core as well.,title="fig : " ] @xmath8-band continuum map of the ngc 1614 nuclear regions from the ifu observations showing the strong marginally resolved nuclear source . _
l - band spectrum of the central @xmath10 700 pc area showing a flat continuum and an ew @xmath11 70 nm pah feature , typical in starbursts .
the spectrum is similar within 0.3 " of the core as well.,title="fig : " ] the ngc 1614 3.3 @xmath0 m pah emission ew shows a wide ring around the nucleus ; the contours show pa@xmath12 emission_right : _ the ew of the 3.3 pah ( blue squares ) , the pah feature flux ( green circles ) , and pa@xmath12 flux ( red stars ) shown as a function of radius .
each profile is normalised to unity , though the actual values of the 3.3 pah ew can be read from the right - side axis .
pah ew clearly peaks outward of the main strong sf ring as defined by recombination line strengths .
, title="fig : " ] the ngc 1614 3.3 @xmath0 m pah emission ew shows a wide ring around the nucleus ; the contours show pa@xmath12 emission_right : _ the ew of the 3.3 pah ( blue squares ) , the pah feature flux ( green circles ) , and pa@xmath12 flux ( red stars ) shown as a function of radius .
each profile is normalised to unity , though the actual values of the 3.3 pah ew can be read from the right - side axis .
pah ew clearly peaks outward of the main strong sf ring as defined by recombination line strengths .
, title="fig : " ] the pah ring is especially clear when inspected in terms of its equivalent width ( ew ) and more or less coincides with , but also extends well outside of , the star - formation ring detected with recombination - lines and radio continuum .
figure 3 shows the pah distribution compared to the sf ring measured from archival nicmos pa@xmath12 data .
we note that the pah flux extends to the core regions , where pa@xmath12 is not seen , while the pah ew strength simultaneously extends quite far outside of the star formation ring . using optical and nir colour information of the core , and the pah ew strengths in the various regions , and comparing to previous studies linking pah strengths with ages of sf regions ( tacconi - garman et al .
2005 ; daz - santos et al .
2008 ) , we find the central kpc of ngc 1614 to be consistent with the following scenario : the core has experienced a starburst some 20 to 100 myr ago , has no more ongoing sf and much of the initial ism has evacuated the region .
the ring at 300 pc radius is the site of strongest current sf , and the age of the sf is continuously decreasing as one moves outwards from the core to mid - ring , judging by the change of the pah - to - pa@xmath12 ratio .
this ratio grows again outward of the ring , but there the level of star - formation is likely ( still ) too low to be detected in recombination lines , while the strong pah ew might instead be a signal of the very youngest sf ( 2 - 4 myr time scale ; beiro et al 2009 ) .
it thus appears that ngc 1614 is experiencing an outward propagating starburst , the detected pah ring highlighting the location where the sf is just propagating into the `` new '' molecular gas clouds .
moreover , we detect a likely remnant of the companion galaxy and likely trigger of the lirg phenomenon in the linear tidal tail extending ssw of the nucleus , @xmath109 " away . comparing to dynamical models ( johansson et al .
2009 ) , this feature is consistent with a remnant of a smaller spiral galaxy which is now falling towards the main component for the second time , and has likely lost much of its mass , and perhaps is the source of the whole linear tidal tail .
the timescales would put the first passage somewhere at 50 myr in the past , consistent with the estimates of the age of the central starburst .
visnen et al . ( 2008 ) presented nir ao - imaging and optical spectroscopy of the system iras19115 - 2124 . the galaxy , which we dubbed the bird ,
was expected to consist of a pair - interaction based on optical hst imaging .
the vlt / naco imaging and salt spectroscopy , however , suggested a third component .
( u)lirgs are complicated dusty objects , and optical data alone , even of high spatial resolution , often can miss whole nuclei and major components of the interaction ( fig . 4 . ; see also haan et al .
moreover , there was evidence that this third , least massive and irregular component dominates the current star - formation output of the whole system .
a similar case was recently presented by inami et al .
this is in contrast to the widely held picture that tidal interactions are expected to drive large quantities of gas into the _ central regions _ of the interaction resulting in nuclear starbursts .
we have now observed the bird system in more detail using mir imaging in the 11 @xmath0 m band with vlt / visir and jk ifu observations with vlt / sinfoni .
: a bik - three colour image of the bird , where optical bands are from hst and the k - band is our vlt / naco ao - data .
two major nuclei are seen in the centre .
while the optical hst images show spectacular tidal tails extending both vertically and horizontally , the larger southern nucleus ( `` body '' ) is virtually invisible in the hst b - band image , and also the third component ( `` head '' of the bird ) north - east of the two central nuclei is also totally inconspicuous .
the white box shows the region observed with vlt / sinfoni .
_ right _ : the sinfoni pa@xmath13 map of the bird .
the current sf of the bird is dominated by the small and irregular `` head '' component , which has an estimated mass of 1/10 of the two other components .
there is some sf also happening in the smaller of the major nuclei ( `` heart '' ) , which still resembles a spiral galaxy in the k - band images regardless of the fairly advanced interaction .
the tick marks are in arcsec , and 1 arcsec corresponds to approximately 1 kpc.,title="fig : " ] : a bik - three colour image of the bird , where optical bands are from hst and the k - band is our vlt / naco ao - data .
two major nuclei are seen in the centre .
while the optical hst images show spectacular tidal tails extending both vertically and horizontally , the larger southern nucleus ( `` body '' ) is virtually invisible in the hst b - band image , and also the third component ( `` head '' of the bird ) north - east of the two central nuclei is also totally inconspicuous .
the white box shows the region observed with vlt / sinfoni .
_ right _ : the sinfoni pa@xmath13 map of the bird .
the current sf of the bird is dominated by the small and irregular `` head '' component , which has an estimated mass of 1/10 of the two other components .
there is some sf also happening in the smaller of the major nuclei ( `` heart '' ) , which still resembles a spiral galaxy in the k - band images regardless of the fairly advanced interaction .
the tick marks are in arcsec , and 1 arcsec corresponds to approximately 1 kpc.,title="fig : " ] figure 4 .
, right panel , shows the pa@xmath13 flux over the bird system , which clearly locates the current sf into the `` head '' component .
about 3/4 of the current sfr originates from this minor component . while the analysis of the whole dataset is still ongoing , several physical characteristics are immediately obvious from the sinfoni ifu spectra .
the br@xmath14 ew is an order of magnitude larger in the `` head '' than in the `` body '' , and mid - way in the `` heart '' , indicating youngest sf in the head .
interestingly the [ feii]/pa@xmath13 ratio in the head component is larger in its outer parts compared to its inner regions , suggesting the current sf is more compact than the somewhat older sf of few 10s of myr on the outer rim
. ratios of molecular hydrogen to hydrogen recombination lines show liner - like values over the body component while being pure starburst - like elsewhere .
figure 5 . shows the 11.8 @xmath0 m contours from visir data , corresponding to 11.2 @xmath0 m pah - feature at z=0.049 , overlaid on the k - band naco data .
the strongest pah emission is coming from the `` spiral galaxy '' at the heart of the bird .
while there is significant pah flux from the head as well , it does appear that the flux is depleted there , possibly by the destruction of pah carriers or e.g. by dilution of hotter dust continuum .
the spitzer / irac @xmath15 m colour of the head , in fact , is very red , typical to dwarf galaxies ( e.g. smith et al .
2009 ) .
some example velocity slices extracted from the pa@xmath12 and h@xmath16 lines are shown in fig . 6 to show the quality of the data and the main kinematic components .
the velocity of the head component is quite much higher than the others and actually extends to 15200 km / s .
when examining the full data - cube it appears that there are ionised gas _ outflows _ from both the two main galaxy nuclei
cool gas outflows were already detected by the optical salt spectroscopy in nad absorption .
in addition , there are also indications of _ inflows _ of gas at the body and head components .
vlt / visir 11.8 @xmath0 m contours on the k - band image .
there appears to be a depletion of pah strength at the location of the hottest component , the head of the bird . ]
pa@xmath12 maps at selected velocities from our sinfoni data , with the corresponding h@xmath16 emission shown as contours . the latter follow the pa@xmath12 distribution very well over the head component , but the distributions are more varied over the main nuclei
. the positions of the two main k - band nuclei seen in fig . 5 above , are marked with crosses .
, title="fig : " ] pa@xmath12 maps at selected velocities from our sinfoni data , with the corresponding h@xmath16 emission shown as contours .
the latter follow the pa@xmath12 distribution very well over the head component , but the distributions are more varied over the main nuclei
. the positions of the two main k - band nuclei seen in fig . 5 above , are marked with crosses .
, title="fig : " ] pa@xmath12 maps at selected velocities from our sinfoni data , with the corresponding h@xmath16 emission shown as contours .
the latter follow the pa@xmath12 distribution very well over the head component , but the distributions are more varied over the main nuclei .
the positions of the two main k - band nuclei seen in fig . 5 above , are marked with crosses .
, title="fig : " ] pa@xmath12 maps at selected velocities from our sinfoni data , with the corresponding h@xmath16 emission shown as contours .
the latter follow the pa@xmath12 distribution very well over the head component , but the distributions are more varied over the main nuclei .
the positions of the two main k - band nuclei seen in fig . 5 above , are marked with crosses .
, title="fig : " ]
sf dominated ( u)lirgs are expected to hide in their central regions large numbers of undetected core - collapse supernovae ( ccsne ) , i.e. stars more massive than @xmath108 m@xmath17 exploding at the end of their ( short ) lives .
such sne can not be detected at optical wavelengths , even in the local universe , because of severe ( @xmath18 mag ) dust extinction .
recently , horiuchi et al .
( 2011 ) indicated that the measured cosmic ccsn rate is a factor @xmath102 smaller than that predicted from the measured cosmic sfr , suggesting there is a `` sn rate problem '' .
therefore , an accurate correction for the number of sne lost in starbursts , lirgs and ulirgs will be crucial for the present and future ccsn surveys at high-@xmath19 .
our sn search using gemini / altair / niri has so far produced the detections of 6 sne .
the most recent ones , sne 2010cu and 2011hi , both occurred in the same host galaxy ic 883 and were located at only @xmath10200 - 400 pc projected galacto - centric distances . both the sne seem to suffer from relatively low host galaxy extinction @xmath20 and @xmath21 , respectively ; kankare et al .
2012 ) suggesting they are not deeply buried in the nuclear regions of ic 883 . despite the rather low extinctions such sne occurring close to the lirg nuclei have remained undiscovered by the current sn searches and will therefore contribute to the `` sn rate problem '' . moreover ,
some of the other sne we have discovered have had significantly higher extinctions ( e.g. kankare et al .
2008 ) .
based on the ir - luminosities for our sample of 7 lirgs in the multi - epoch gemini observations , the intrinsic number of ccsne exploding over the program is about 15 ccsne .
we expect about one third of these to be detectable depending on the assumed distribution of extinctions and concentration of sne within the innermost nuclear regions .
although the number statistics are arguably still low and the nature of ( u)lirgs at higher redshifts not precisely known , it already appears clear that a significant fraction of ccsne at intermediate and high-@xmath19 remains undetected in these systems .
therefore , ao - assisted nir observations provide an important window to detect and study sne in the obscured nuclear regions of the local counterparts of the galaxies dominating the sf at higher-@xmath19 .
c young and massive super star clusters ( sscs ) are found whenever very active star formation is going on , such as that in interacting lirgs ( e.g. whitmore et al .
we are studying the luminosity and mass functions of sscs as a function of sfr and type of host galaxy , interaction stage , etc .
, in an effort to constrain the cluster mass function ( cmf ) , study the formation , evolution , and disruption of sscs , which all have relevance also to star formation laws in general ( escala & larson 2008 ) . as a first step ,
we have studied the k - band lfs of the sample of galaxies from our gemini observations . according to theoretical expectations and many optical studies of nearby star - forming galaxies , ssc lfs
are well fitted by a single power - law index of @xmath22 in a @xmath23 distribution .
we , in contrast , find a surprising variety of slopes , ranging from much shallower ones of @xmath24 , to more or less the `` nominal '' @xmath25 case . in addition , intriguingly , most of the galaxies in the initial sample have clear break - points in their ssc lf slope , where the bright section is significantly steeper than the faint magnitude ranges , or we find distributions which might be better fit with e.g. gaussian distributions .
7 shows three examples of the lfs , one with a clear break - point , and others with the opposite ends of the range of slopes , with @xmath24 and @xmath26 , respectively , as fitted to the bright part of the lf .
these effects persist through rigorous photometric , blending , and detection / completeness analyses and simulations in the complex background conditions of the target galaxies .
a range of lf slopes and/or breaks in the slopes could indicate mass - dependent disruption of the sscs , or environment and/or host sfr dependent cluster cmfs ( e.g. bastian 2008 ; chandar et al .
2010 ; larsen 2009 ) .
most previous determinations are from less luminous ir - galaxies , while recently relatively shallow slopes for ssc lfs have been found in blue compact dwarfs ( e.g. adamo et al . 2011 and references therein ) .
our analysis is still ongoing , and will be presented in randriamanakoto et al .
( 2012 , in preparation ) , and the full sample of 40 galaxies will also be added to the ssc analysis to get a handle on how universal cfls are , and how they depend on host galaxy properties .
we are studying a sample of several dozen strongly star - forming galaxies , mostly lirgs , using nir adaptive optics imaging , and optical spectroscopy , and ifu observations for selected targets .
adaptive optics and near- and mid - ir observations are often necessary to uncover how many nuclei are involved in the interactions , and to characterise the components , and how star formation is triggered and is propagating in the systems .
we find the l - band , and in particular the 3.3 @xmath0 m pah emission , to trace well various levels of star formation in a lirg nucleus .
we also find hundreds of super star clusters , and detect a population of dust - obscured core - collapse sne in our lirgs . | we study star formation ( sf ) in very active environments , in luminous ir galaxies , which are often interacting .
a variety of phenomena are detected , such as central starbursts , circumnuclear sf , obscured sne tracing the history of recent sf , massive super star clusters , and sites of strong off - nuclear sf .
all of these can be ultimately used to define the sequence of triggering and propagation of star - formation and interplay with nuclear activity in the lives of gas rich galaxy interactions and mergers . in this paper
we present analysis of high - spatial resolution integral field spectroscopy of central regions of two interacting lirgs .
we detect a nuclear 3.3 @xmath0 m pah ring around the core of ngc 1614 with thermal - ir ifu observations .
the ring s characteristics and relation to the strong star - forming ring detected in recombination lines are presented , as well as a scenario of an outward expanding starburst likely initiated with a ( minor ) companion detected within a tidal feature .
we then present nir ifu observations of iras 19115 - 2124 , aka the bird , which is an intriguing triple encounter .
the third component is a minor one , but , nevertheless , is the source of 3/4 of the sfr of the whole system .
gas inflows and outflows are detected at the locations of the nuclei . finally , we briefly report on our on - going nir adaptive optics imaging survey of several dozen lirgs .
we have detected highly obscured core - collapse sne in the central kpc , and discuss the statistics of `` missing sne '' due to dust extinction .
we are also determining the characteristics of hundreds of super star clusters in and around the core regions of lirgs , as a function of host - galaxy properties . |
the first subject of this article is the embedded resolutions of `` simple '' singularities , called also rational double point surface singularities .
they are called simple because they have no moduli ( i.e. a hypersurface singularity with the same topological type of one of these singularities is analytically isomorphic to it ) @xcite .
but at the same time , they are simple from a resolution of singularities point of view , i.e. easy to resolve .
the traditional approach to resolve singularities is to blow up at smooth centers in order to make an invariant drop . this invariant takes values in a discrete ordered set with a smallest element ( which detects smoothness ) .
it should not only detect smoothness , but also be easy to compute so that its behavior can be followed after the blowing ups . in this article
, we adopt a different strategy to resolve a simple singularity @xmath1 we construct an embedded toric resolution of @xmath2 this construction is based on a deep invariant , the set * ec * ( essential components ) of irreducible components of the jet schemes satisfying some natural properties .
recall that ( see section 2 ) for @xmath3 the @xmath4th jet scheme of a variety @xmath5 parametrizes morphisms @xmath6/(t^{m+1})\longrightarrow x.$ ] these are finite dimensional approximations of the space of arcs which parametrize germs of curves drawn on @xmath7 actually the geometry of the @xmath4th jet schemes is intimately related with the geometry of the set of arcs in a smooth ambient space containing @xmath5 and which have `` contact '' with @xmath5 larger than @xmath8 this explains in parts why while the arc space of @xmath5 detect information about abstract resolution of singularities @xcite , jet schemes detect information about embedded resolution of singularities @xcite,@xcite,@xcite .
the other subject that this article considers is the minimality of embedded resolutions of singularities .
in contrast with the abstract resolution case , there is no `` universal '' minimal embedded resolution for surface singularities .
therefore we need to make more precise this notion .
actually , the embedding resolution of @xmath9 that we construct is toric and is obtained by considering , using jet schemes , a particular regular subdivision of the newton dual fan @xmath10 associated with the defining equation of @xmath5 at the singularity(see section 3 for the definition of @xmath10 ) . since the singularities that we are considering are newton non - degenerate ( see section 3 ) , regular subdivisions of @xmath10 give toric embedded resolutions of @xmath9 ( @xcite,@xcite ) .
this is equivalent to saying that an abstract resolution of singularities of the toric variety @xmath11 defined by the fan @xmath10 gives an embedded toric resolution of @xmath2 but the toric variety @xmath11 is of dimension @xmath12 and hence thanks to a theorem by bouvier and gonzalez - sprinberg @xcite , we know that there exists a toric resolution of singularities of @xmath11 where all the irreducible divisors of the exceptional locus are essential , i.e. the centers of the divisorial valuations associated with these divisors give irreducible components of the exceptional locus of any other toric resolution of singularities of @xmath13 we call such a resolution of singularities a minimal toric resolution ; in @xcite , it is called @xmath14desingularization ; we call a toric embedded resolution of @xmath9 minimal if it corresponds to a minimal toric resolution of @xmath13 note that in general , a minimal toric embedded resolution of singularities is not unique , but the divisorial valuations associated with the irreducible divisors of its exceptional locus are the same for all minimal toric embedded resolution of singularities .
we say that these last divisorial valuations are embedded essential .
+ we will prove that the toric embedded resolutions of @xmath9 that we construct from jet schemes is minimal for all simple singularities except for @xmath15 in particular the set of essential components of the jet schemes of simple singularities ( except for @xmath0 ) is in a bijective correspondence with the set of embedded essential divisorial valuations .
this can be thought as a solution of an embedded nash problem . +
the choice of this class of singularities is related to the following reasons : + first , for @xmath16 big enough the number of irreducible components of their @xmath4th is constant , this simplifies the classification of the irreducible components we are interested in , when @xmath16 varies .
this is not the case in general @xcite , @xcite .
these singularities are newton non - degenerate with respect to their newton polyhedron , and therefore they have a toric embedded resolution , see @xcite,@xcite,@xcite
. defining the class of irreducible components of jet schemes , mentioned above , is more subtle for newton degenerate singularities @xcite .
this is related to a conjecture of teissier on embedding any singularity in such way that it can be resolved by a toric morphism @xcite,@xcite,@xcite .
+ we would like to thank shihoko ishii , monique lejeune - jalabert , patrick popescu - pampu and bernard teissier for several discussions about this article .
in this section , after giving some preliminaries on jet schemes , we recall from @xcite the structure of jet schemes of rational double point singularities and we extract from it some informations about particular irreducible components of these jet schemes . these informations will be used in the next section to obtain the canonical resolution of these singularities .
we will treat all the cases .
the case of @xmath17 ( see below the defining equation ) has been treated in @xcite , but we consider it here briefly for the convenience of the readers .
+ let @xmath18 be an algebraically closed field of arbitrary characteristic and @xmath5 be a @xmath18-algebraic variety
. for @xmath19 the functor @xmath20 @xmath21/(t^{m+1}),x)\ ] ] where @xmath22 is a @xmath23algebra , is representable by a @xmath23scheme @xmath24 @xcite .
@xmath24 is the m - th jet scheme of @xmath5 , and @xmath25 is isomorphic to its functor of points .
in particular the closed points of @xmath24 are in bijection with the @xmath26/(t^{m+1})$ ] points of @xmath5 . + for @xmath27 , the truncation homomorphism @xmath28/(t^{m+1 } ) \longrightarrow a[t]/(t^{p+1})$ ] induces a canonical projection @xmath29 these morphisms verify @xmath30 for @xmath31 , and they are affine morphisms , they define a projective system whose limit is a scheme that we denote @xmath32 this is the arc space of @xmath5 .
+ note that @xmath33 .
we denote the canonical projection @xmath34 by @xmath35 and by @xmath36 the canonical morphisms @xmath37 + we now assume that @xmath9 is defined by one of the following equations : @xmath38 @xmath39 @xmath40 @xmath41 @xmath42 we denote by @xmath43 where @xmath44 is the origin of @xmath45 it also is the singular locus of @xmath7 we will now associate a divisorial valuation over @xmath46 with the irreducible component of @xmath47 satisfying the property @xmath48 that for every irreducible component @xmath49 @xmath50 for @xmath51 let @xmath52 be the canonical morphism , here the exponent @xmath53 stands for ambient . for @xmath54 we consider the following cylinder in the arc space @xmath55 since @xmath56 is a trivial fibration , for every irreducible component @xmath57 we have that @xmath58 is an irreducible component of @xmath59 whenever this last intersection is non - empty . but from the property @xmath48 we have that @xmath60
this implies that @xmath61 + let @xmath62 be the generic point of @xmath63 then for every @xmath64,$ ] we define the map @xmath65\longrightarrow \mathbb{n},$ ] as follows @xmath66 it follows from corollary @xmath67 in @xcite , that @xmath68 is a divisorial valuation ( see also @xcite , @xcite , prop . 3.7 ( vii ) applied to @xmath69 ) .
+ given @xmath70 with an irreducible component @xmath71 of @xmath72 we associate the following vector that we call the weight vector : @xmath73 in the following definition , we consider the irreducible components of @xmath74 that will be meaningful for the problem we are considering . we define the following set of irreducible components of @xmath75 @xmath76 @xmath77 where @xmath78 stands for essential components .
+ we also consider the following set of associated valuations : @xmath79 where @xmath80 stands for essential embedding , to say that these valuations are supposed to appear on every embedded toric resolution .
note that the definitions of the sets @xmath81 and @xmath82 are ad - hoc to the type of singularities we are considering . indeed the fact that these singularities are non - degenerate with respect to their newton polyhedron , implies that the valuations that will be meaningful are monomial , this what motivates the above definition . from on other point of view , if we check carefully the equations , we will figure out that the components which does not belong to the set @xmath83 are at their generic points trivial fibrations above components which belong to @xmath84 that is why they are not useful and this phenomena reflects the fact that the singularities we consider are newton non - degenerate .
+ note also that on one hand we are interested in the embeddings of the singularities in @xmath46 that are defined by the equations we gave at the beginning of this section ; this includes that we consider the variables @xmath85 as given and hence the embedding of the torus in @xmath46 is also given .
on the other hand , since we are interested in toric resolution , the exceptional locus of such a resolution will be defined as the inverse image of the complement of the torus , i.e the coordinate hyperplanes .
so the fact that the @xmath86 and @xmath87 axis belong to the singularities @xmath88 the @xmath87 axis belongs to the singularities @xmath89 and the @xmath90 axis belongs to the singularity @xmath91 suggests that if we want to find a toric resolution , we need also to consider the jets which go through these axis .
but the family of @xmath4th jets whose center is a generic point on one of these axis is irreducible because our surfaces have isolated singularities at the origin .
so the components which come from those families are easy to determine ; again since our singularities are newton non - degenerate , we are only interested in the vectors associated with these components , so we only consider such components for small @xmath92 whenever the associated vector changes . for @xmath16 large enough , the associated vector stabilizes .
+ a careful reading of @xcite ( resp .
@xcite ) permits to determine @xmath78 for a rational double point singularitiy @xmath93 that we will denote @xmath94 let @xmath95 $ ] be the defining equation of a rational double point singularity @xmath7 we write @xmath96 then the @xmath16-th jet scheme @xmath24 of @xmath5 is naturally embedded in @xmath97\ ] ] and is de fined by the ideal @xmath98 from section @xmath12 in @xcite and section @xmath99 in @xcite , we are able to determine the set @xmath81 , this is the subject of the following lemma .
we do not treat the @xmath0 singularity ( which will be in some sense an exception to part of our formulation ) because it takes too much space ; we only give the weight vectors associated with the components in @xmath100 in the corollary that follows the lemma .
[ ec ] for an @xmath101 singularity , we have that @xmath102 is given by the components centered at the origin @xmath103 and the components centered at the @xmath86 respectively @xmath87 axis are @xmath104 respectively @xmath105 for a @xmath106 singularity , we have that @xmath107 is given by the components centered at the origin @xmath108 @xmath109 @xmath110 @xmath111 @xmath112 and the components centered at the @xmath87-axis are @xmath113 for an @xmath17 singularity we have that @xmath114 @xmath115 @xmath116 for an @xmath117 singularity we have that @xmath118 is given by the components centered at the origin @xmath119
@xmath120 @xmath121 @xmath122 @xmath123 and those centered above the @xmath90 axis @xmath124 from lemma [ ec ] , we deduce directly what is the set @xmath125 first recall that a monomial valuation , defined on the ring @xmath126 $ ] and associated with a vector @xmath127 is defined as follows : if @xmath128 $ ] then @xmath129 from lemma [ ec ] we deduce the following proposition : the valuations that belong to @xmath130 are monomial .
they are associated with the vectors @xmath131 where @xmath132 these vectors are @xmath133 @xmath134 @xmath135 the valuations that belong to @xmath136 are monomial .
they are associated with the vectors @xmath131 where @xmath137 these vectors are + @xmath138
@xmath139 @xmath140 @xmath141 the valuations that belong to @xmath142 are monomial .
they are associated with the vectors @xmath131 where @xmath143 : + @xmath144 the valuations that belong to @xmath145 are monomial .
they are associated with the vectors @xmath131 where @xmath146 + @xmath147 @xmath148 the valuations that belong to @xmath149 are monomial .
they are associated with the vectors @xmath131 where @xmath150 + @xmath151 @xmath152 @xmath153 let @xmath154 assume that @xmath155 is an irreducible component of @xmath156 which belongs to @xmath94 then , a generic point of @xmath157 is of the shape @xmath158 where @xmath159 is generic . for @xmath160,$ ]
it follows now from the definition of @xmath161 ( see equation ( [ val ] ) ) and from the definition of @xmath162 ( see equation ( [ mon ] ) ) that @xmath163 and the proposition follows .
in this section , we define minimal toric embedded resolutions , when they exist , and we give the motivations for this definition .
+ let @xmath164 be a lattice and let @xmath165 be the strictly convex cone generated by the vectors @xmath166 , i.e. @xmath167 let @xmath168 be a polynomial in @xmath126 $ ] with @xmath169 .
we denote by @xmath170 its newton polyhedron at the origin : this is the convex hull in @xmath171 of the set @xmath172 where we have supposed that @xmath173 is given by its taylor expansion at the origin and let @xmath174 be the union of the compact boundaries of @xmath170 , called newton boundary .
this polyhedron and its boundary depend on the choice of coordinates .
let @xmath175 be its dual fan ; we will call covectors vectors which lie in @xmath176 a covector defines a linear map on @xmath177 for a positive covector @xmath178 we define the distance @xmath179 as the minimal value of the linear map @xmath180 at @xmath181 let @xmath182 be the dual face to @xmath183 we can then define @xmath184 as the restriction of @xmath173 to the dual face of @xmath180 , @xmath185 that is @xmath186 we can find more details about newton polyhedrons , their dual fans and newton non - degeneracy in @xcite(see also @xcite ) . + let @xmath187 $ ] be such that @xmath188 we say that the hypersurface @xmath189 is non - degenerate with respect to a covector @xmath180 if the variety @xmath190 is a reduced smooth hypersurface in the torus @xmath191 we say that @xmath173 is non degenerate at the origin @xmath192 with respect to its newton polyhedron @xmath174 if it is non degenerate for each covector . a very useful result about toric embedded resolutions of this type of singularities goes as follows . [ nnd](@xcite,@xcite,@xcite )
we consider a pair @xmath9 where @xmath193 is a newton non - degenrate singularity , then the folllowing properties are equivalent : + 1 ) a subdivison @xmath194 of @xmath175 is regular , + 2 ) the proper birational morphism @xmath195 $ ] induced by @xmath194 is an embedded resolution of singularities of the pair . here
@xmath196 is the toric variety associated with the fan @xmath197 recall that a fan @xmath194 ( subdividing @xmath175 ) is said to be regular
if every cone @xmath198 is regular , i.e. the vectors generating @xmath199 are part of a basis of @xmath200 as a @xmath201module , or equivalently ( for @xmath199 a cone maximal dimension ) , if @xmath202 then the absolute value @xmath203 note that the morphism @xmath204 is explicit in term of @xmath205 more precisely , a regular cone @xmath206 of maximal dimension determines an affine chart of @xmath196 which is isomorphic to the affine space @xmath207 $ ] and the rectriction of @xmath204 to this chart is given by @xmath208 @xmath209 @xmath210 where @xmath211 moreover , an edge of @xmath194 ( a cone of dimension @xmath212 ) determines an orbit of @xmath196 whose zariski closure defines a divisor on @xmath196 which is an irreducible component of the exceptional divisor of @xmath204 ( @xcite ) . and the strict transform of @xmath193 is defined as follows :
[ st ] let @xmath194 be a regular subdivision of @xmath176 the strict transform of @xmath189 by @xmath204 is the zariski closure of @xmath213 we now recall the notion of `` essential divisor '' which is one of the motivations of this work .
for a singular variety , we can find infinitely many resolutions of singularities ; the notion of essential divisor searches for intrinsic data in all resolutions of singularities of a given singular variety .
( essential divisors ) + let @xmath5 be a singular variety and @xmath214 be a resolution of singularities of @xmath215 let @xmath216 be a divisor which is an irreducible component of the exceptional locus @xmath217 .
we say that @xmath218 is an essential divisor if for any other resolution of singularities @xmath219 the center on @xmath220 of the divisorial valuation determined by @xmath218 is an irreducible component of @xmath221 .
a divisor @xmath218 is said inessential if it is not essential .
it should be noticed that the notion of essential divisor given above , is associated with abstract resolutions of singularity and not with embedded resolutions of singularities .
a good example for such a divisior , is an essential divisor on a surface singularity : in this case , a divisor is essential if it appears on the minimal resolution of singularities .
note that a minimal resolution does not exist in general for a singular variety of dimension larger than @xmath222 for normal toric singularities , it is possible to determine essential divisors . for
that we need the following definition of a minimal system of generators of a cone .
we give the statements for toric varieties in dimension @xmath12 since this is the case which interests us .
( @xcite , see also @xcite ) let @xmath199 be a strictly convex cone .
the minimal system of generators of @xmath199 is by definition the set : @xmath223 an element in @xmath224 is also called irreducible .
it is primitive by definition .
this implies ( @xcite , propostion @xmath225 ) that the elements of @xmath226 appear on every regular fan which subdivides @xmath199 as an extremal vector , i.e. a primitive generator of a dimension @xmath212 cone of the fan . note that such a subdivision determines a ( toric ) resolution of singularities of the toric variety defined by @xmath227 and dimension @xmath212 cones determine the irreducible components of the exceptional divisors .
this gives a feeling of the following theorem which characterizes essential divisors on toric varieties .
[ bg](@xcite,@xcite ) + let @xmath199 be a strongly convex cone .
the minimal system of generators @xmath226 is in bijection with the set of essential divisors of the toric variety @xmath228 .
this implies that a primitive vector is in @xmath226 if and only if it is an extremal vector of any regular subdivision of the cone @xmath199 .
actually , it follows from @xcite the fact that the divisors which correpond to elements of @xmath226 are toric essential ( i.e. essential for toric resolutions of singularities ) and from @xcite the fact that they are essential for all resolutions of singularities .
it follows also from @xcite that for a toric variety defined by a cone @xmath199 and which is of dimension @xmath229 there exists a toric resolution of singularities which is associated with a subdivision of @xmath199 that makes intervene only elements of @xmath230 such a resolution is called a @xmath226-resolution .
we think of such a resolution as a minimal toric resolution .
note that such a resolution does not exist in general for normal toric varieties of dimension larger than @xmath231 this discussion joint to theorems [ nnd ] and [ bg ] motivate the following two definitions : ( toric embedded essential divisors ) + we consider a pair @xmath9 where @xmath193 is a newton non - degenrate singularity .
a toric embedded essential divisor @xmath217 is a divisor which is essential for every toric resolution of singularities of @xmath5(such a resolution exists thanks to theorem [ nnd ] ) .
this is equivalent to say that @xmath217 corresponds to an element of @xmath226 for some @xmath199 a cone of maximal dimension belonging to @xmath176 ( minimal toric embedded resolution ) a toric embedded resolution of newton non - degenerate pair @xmath9 where @xmath232 associated with a subdivision @xmath194 of @xmath175 is minimal if the only edges which intervene in @xmath194 are elements of @xmath226 for some @xmath199 a cone of maximal dimension belonging to @xmath176
in this section , we construct from the data of the jet schemes of simple singularities that we gave in section 2 , toric embedded resolution .
we also prove that for all singularities of this this type , except for the @xmath0 singularity , the essential components of the jet schemes ( introduced in section 2 ) correspond to toric embedded essential divisors .
we will treat each type of singularities apart and the strategy of the proofs is the same : first we construct a toric resolution of singularities : using the weight vectors that we associated in section 2 with essential components , we construct a regular subdivision of the dual fan of the singularity embedded in @xmath233 and then by using a theorem from @xcite ( see also @xcite ) which characterizes toric essential divisor for toric varieties , we are able ( except for the @xmath0 ) singularity to prove the minimality of our toric embedded resolution of singularities .
weight vectors of @xmath234 give a toric minimal embedded resolution of the singularity .
in particular the set of essential components @xmath235 is in bijection with the divisors which appear on every toric minimal embedded resolution of these singularities .
the dual fan is not simplicial ; a resolution of this singularity is given by a regular and simplicial decomposition of the newton polyhedron , containing it as a sub - fan .
+ studying the paper of the first author @xcite , one can compute the weight vectors @xmath236 @xmath237 @xmath238 to which we add , if @xmath239 is even , @xmath240 et @xmath241 . + as the @xmath87 axis and @xmath86 axis are in the surface and contains the singularity , it also contains the vectors @xmath242 and @xmath243 .
the weight vectors are positioned on the following figure : + + * claim : these vectors are irreducible , i.e. they represent essential components in the resolution . *
+ before proving this claim , we study the position of the vectors in the basis faces of dual fan , and show that they provide a toric embedded resolution . +
* some remarks on the vectors of @xmath130 * : + ( with abuse of notations , the vectors are considered as points on the basis face of the cone @xmath244 $ ] ) .
we first prove that the vectors are rational sums of @xmath245 . by the symmetry of the singularity , we only have to look at half of the vectors : + we have : @xmath246 for all @xmath247 and @xmath248 .
+ + moreover @xmath249 .
+ this implies that they all belong to the cone @xmath250 .
2 . then , one can remark the symmetry one obtains .
it will follow that we have to do only half of the computations ; 3
. the vectors @xmath251 are on the line @xmath252 as @xmath253 .
idem for @xmath254 for @xmath255 ; 4 .
homogeneous vectors @xmath256 such that @xmath257 , for @xmath258 are also one the same line .
all these remarks show us a way to obtain an almost symmetric resolution from the vectors
. many algorithms work , we propose to the lector one of them .
+ + * algorithm * + first place @xmath259 and make the triangle @xmath260 simplicial .
+ then place successively the vectors @xmath261 , lying this new vector to @xmath259 , to make the resolution simplicial .
do the same symmetrically with @xmath262 .
+ one has @xmath263 thus these vectors are primitive vectors .
+ for @xmath264 : + now place the vectors @xmath265 and @xmath266 and the corresponding edges.first one can remark that the vectors @xmath267 lie on the line @xmath268 .
it is at this point that the resolution is not symmetrical : if one choose first @xmath265 , then one makes the trapezium simplicial by adding the edges @xmath269 and @xmath270 .
so in the toric resolution , one will have the edges relying the vectors on @xmath271 to @xmath272 .
conversely if one first choose @xmath266 , one will have the edges relying the vectors on @xmath271 to @xmath273 .
+ for @xmath274 , one had @xmath275 $ ] first , the trapezium thus becomes divided into three parts and the resolution become symmetrical . +
then add alternatively @xmath276,and if @xmath239 is even @xmath277,@xmath241 and obtain : * let us look at triangles @xmath278 $ ] and @xmath279 $ ] for @xmath247 .
+ we have : @xmath280 * we also have the triangles @xmath281 $ ] et @xmath282 $ ] of determinant @xmath212 .
* now , we look at @xmath283 $ ] . + we have @xmath284 + if @xmath239 is even , we have the triangles
t @xmath285 $ ] and if @xmath239 is odd , we have the triangles @xmath286 $ ] et @xmath287 $ ] with determinant @xmath212 . *
the cones @xmath288 $ ] and @xmath289 $ ] are non singular . *
finally we have to look at the triangles @xmath290 $ ] and @xmath291 $ ] for all @xmath292 and @xmath293(and to the symmetrical one too ) .
+ one has + @xmath294 now we have to prove that in fact this resolution is a @xmath295 ( according to c.bouvier and g.gonzalez-sprinberg ) . +
first of all , the weight vectors @xmath296 are clearly primitive as they have @xmath212 as third component .
+ consider the simplicial cone generated by @xmath297 ; all the weight vectors belongs to this cone .
+ let us call the same way the points in @xmath298 , end of the vectors starting from @xmath192 and the vectors themselves
. the points in @xmath298 @xmath299 belong to the face @xmath250 of the thetahedron @xmath300 $ ] .
this implies that the set @xmath299 is free over @xmath301 , i.e. none vector is reducible into other vectors of this set .
suppose the lemma is shown , then the only primitive vectors that could generate the characteristic vectors should be vectors inside the thetahedron @xmath300 $ ] .
suppose that there is such a vector . from th .
1.10 of @xcite , it corresponds to an essential divisor in the toric resolution of the cone @xmath302 $ ] , i.e. it appears in every regular subdivision as an extremal vector . but with the vectors @xmath303
, we get such a regular subdivision , so there is no primitive vector in the thetahedron @xmath300 $ ] .
+ so let us prove the lemma : weight vectors of @xmath306 give a " _ canonical " toric minimal embedded resolution of the singularity . in particular the set of essential components @xmath307 is in bijection with the divisors which appear on every toric minimal embedded resolution of these singularities .
_ first we make the dual polyhedron simplicial by adding the edge @xmath308 $ ] .
let us call @xmath309 be the cone @xmath310 $ ] , @xmath311 be the cone @xmath312,$ ] @xmath313 be the cone @xmath314 $ ] , @xmath315 be the cone @xmath316 $ ] .
first of all , to obtain a toric resolution of singularities of @xmath106 with equation @xmath317 , as it is not commode , one has to blow up the @xmath87 axes , i.e. to add the vector @xmath318 .
+ let us now consider the weight vectors and show some of their properties : * the vectors @xmath319 ( @xmath320 ) are primitives and belong to the cone @xmath309
. more precisely they lie onto the line @xmath321 : + in fact we have the following equalities : @xmath322 and if we note @xmath323 , then for @xmath320 , @xmath324 * the vectors @xmath325 ( @xmath320 ) are primitives and belong to the cone @xmath313 .
more precisely they lie onto the line @xmath326 : + in fact we have the following equalities : @xmath327 and if we note @xmath328 , then for @xmath320 , @xmath329 * the vectors @xmath330 ( @xmath331 ) are primitives and belong to the cones @xmath311 and @xmath313 . more precisely they lie onto the line @xmath332 : + in fact we have the following equalities : @xmath333 and , if we note @xmath334 , then for @xmath335 , @xmath336 * claim : the simplicial decomposition of the dual newton cone of @xmath106 obtained by adding , the edge @xmath308 $ ] , the vector @xmath318 , then the vectors @xmath337 to @xmath338 in the natural order , the vectors @xmath339 to @xmath340 in the natural order and the vectors @xmath341 to @xmath342 in the natural order too , give a toric minimal embedded resolution of @xmath106*. + the resolved dual cone is of the form : proof of the claim : by the previous observations , the above vectors are primitive ; thus it remains to prove that the cones obtained are non singular , i.e. they have determinant @xmath343 or @xmath344 and using the same proof as for @xmath101 , we prove then that the vectors are irreducible . + + we add first the vector @xmath318 , which corresponds to the blowing up of the @xmath87-axes .
then , the cone @xmath313 is decomposed into two cones : @xmath345 and @xmath346 . by the above observation
, we have now that each group of characteristic vectors are in different cones .
* decomposition of the cone @xmath309 : + let us note @xmath347 .
+ we add the vectors from @xmath337 to @xmath338 and show by induction that we obtain at the end regular cones . at each step
the cone where lie the new @xmath348 is decomposed into three cones : @xmath349 , @xmath350 and @xmath351 .
that last one contains @xmath352 .
@xmath353 + that last determinant is not equal to @xmath343 or @xmath344 , so we add @xmath352 and so on . + for @xmath354 the last determinant is equal to one .
* decomposition of the cone @xmath355 : + let us note @xmath328 .
+ we add the vectors from @xmath339 to @xmath340 and show by induction that we obtain at the end regular cones . at each step the cone where lie
the new @xmath356 is decomposed into three cones : @xmath357 , @xmath358 and @xmath359 . that last one contains @xmath360 for @xmath361 .
+ we have : @xmath362 that last determinant is not equal to @xmath343 or @xmath363 so we add @xmath364 and so on . + for @xmath354 the last determinant is equal to one . *
decomposition of the cones @xmath365 and @xmath311 : + let us note @xmath334 .
+ we add the vectors from @xmath341 to @xmath342 and show by induction that we obtain at the end regular cones . at each step
the two cones where lie the new @xmath366 are decomposed into two cones themselves : @xmath367 , @xmath368 and @xmath369 , @xmath370 .
those last ones contain @xmath371 and are not regular .
@xmath372 + @xmath373 + and + @xmath374 + for @xmath375 the two last determinants are equal to @xmath343 or @xmath376 moreover we have obtained a @xmath377-resolution : + as for @xmath101 singularity , one can show that @xmath356 ( resp.@xmath378 and @xmath379 ) belongs to the base of the cone @xmath380 ( resp .
@xmath381 , and @xmath382 , whose equation is @xmath383 ( resp . @xmath384 and @xmath385 ) . with these vectors
we obtain a regular subdivision of each cone , which implies that there are no other primitive vectors inside each cone and that the characteristic vectors are irreducible .
so we also obtain a @xmath377-resolution for @xmath106 .
the set of weight vectors @xmath142 is @xmath144 call them in alphabetic order .
+ weight vectors of @xmath17 give a " _ canonical " toric minimal embedded resolution of the singularity .
in particular the set of essential components @xmath388 is in bijection with the divisors which appear on every toric minimal embedded resolution of these singularities . _
the vectors are in different sub cones of the dual fan . to make the fan regular , we divide each cone thanks to the vectors .
so first we have to look at the positions of each vector .
one has : @xmath389 @xmath390 @xmath391 as for the previous singularities , one can prove that adding the weight vectors in each cone ( in the alphabetic order , see fig.5 ) , gives a minimal resolution for each of them .
see fig.8 for the picture .
weight vectors of @xmath117 give a " _ canonical " toric minimal embedded resolution of the singularity .
in particular the set of essential components @xmath118 is in bijection with the divisors which appear on every toric minimal embedded resolution of these singularities . _
the vectors are in different sub cones of the dual fan ( see fig.6 for denomination of the cones ) . to make the fan regular , we divide each cone thanks to the vectors .
so first we have to look at the positions of each vector .
one has : @xmath398 @xmath399 @xmath400 @xmath401 one can show that we get a regular subdivision of each cone thanks to the following characteristic vectors @xmath402 @xmath403 moreover they are irreducible ( can show as above ) , so we get a minimal embedded toric resolution for @xmath0 .
+ in fact we have [ [ remark - on - abstract - resolution - of - singularities - of - simple - singularities ] ] remark on abstract resolution of singularities of simple singularities ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ it worths noticing that , for a simple singularity , the restriction of a minimal embedded resolution to the strict transform of the singularitiy gives its minimal abstract resolution ; note that not all the divisors that appear on the minimal embedded resolutions cut the strict transform , only those which corresponds to weight vectors which belong to the two dimensional cones of the newton dual fan .
moreover , these last divisors may cut the strict tansform along more than one irreducible component .
s. aguzzoli , d. mundici , an algorithmic desingularization of 3-dimensional toric varieties .
tohoku math .
j. ( 2 ) 46 ( 1994 ) , no . 4 ,
557 - 572 .
f. aroca , fuensanta , m. gomez - morales , k. shabbir , torical modification of newton non - degenerate ideals .
exactas fs .
ser . a math .
racsam 107 ( 2013 ) , no . 1 , 221 - 239 . c. bouvier , diviseurs essentiels , composantes essentielles des varits toriques singulires , duke math . j.
* 91 * ( 1998 ) , 609 - 620 . c. bouvier , g. gonzalez - sprinberg , systme gnrateur minimal , diviseurs essentiels et g - dsingularisations des varits toriques , tohoku math . j. ( 2 ) * 47 * ( 1995 ) , 125 - 149 . v. cossart , o. piltant , olivier resolution of singularities of threefolds in positive characteristic .
j. algebra 321 ( 2009 ) , no .
7 , 1836 - 1976 .
l. ein , r. lazarsfeld , m. mustata , contact loci in arc spaces .
( 2004 ) , no . 5 , 1229 - 1244 . t. de fernex , l. ein , s. ishii , divisorial valuations via arcs .
inst . math .
44 ( 2008 ) , no . 2 , 425 - 448 .
g.gonzalez-sprinberg , m. lejeune - jalabert modles canoniques plongs .
i. kodai math .
j. 14 ( 1991 ) , no .
2 , 194209 . doi:10.2996/kmj/1138039394 .
h. hironaka , resolution of singularities of an algebraic variety over a field of characteristic zero .
ann . of math .
( 2 ) 79 ( 1964 ) , 109 - 203 ; ibid . ( 2 )
79 1964 205 - 326 .
a.g.hovanskii , newton polyhedra , and toroidal varieties .
( russian ) funkcional .
i prilozen .
11 ( 1977 ) , no . 4 , 56 - 64 , 96 . s. ishii , jet schemes , arc spaces and the nash problem , c.r.math . rep
canada , 29 ( 2007 ) 1 - 21 .
s. ishii , j. kollr , the nash problem on arc families of singularities .
duke math .
j. 120 ( 2003 ) , no . 3 , 601 - 620 .
l , m. tosun , simple singularities and simple lie algebras .
twms j. pure appl .
math . 2 ( 2011 ) , no . 1 , 97 - 111 . m. lejeune jalabert , dsingularisation explicite des surfaces quasi - homognes dans @xmath411 , nova acta leopoldina , * nf 52 * , nr 240 ( 1981 ) , 139160 .
m. lejeune - jalabert , h. mourtada , a. reguera jet schemes and minimal embedded desingularization of plane branches.revista de la real academia de ciencias exactas , fisicas y naturales .
serie a. matematicas : volume 107 , issue 1 ( 2013 ) , page 145 - 157 .
h. mourtada , jet schemes of rational double point singularities , valuation theory in interaction , ems ser .
soc . , sept .
2014 , pp : 373 - 388 , doi : 10.4171/149 - 1/18 . h. mourtada , jet schemes of complex branches and equisingularity , ann . inst .
fourier 61 , no . 6 , 2313 - 2336 ( 2011 ) .
h. mourtada , jet schemes of toric surfaces , c. r. , math .
paris 349 , no .
9 - 10 , 563 - 566 ( 2011 ) .
h. mourtada , jet schemes of normal toric surfaces , submitted .
h. mourtada , jet schemes and minimal generating sequences of divisorial valuations in dimension @xmath412 submitted .
j. f. nash , arc structure of singularities , duke math .
j. 81 ( 1995 ) , no . 1 , 31 - 38 ( 1996 ) .
t. oda , convex bodies and algebraic geometry .
an introduction to the theory of toric varieties .
ergebnisse der mathematik und ihrer grenzgebiete ( 3 ) [ results in mathematics and related areas ( 3 ) ] , 15 .
springer - verlag , berlin , 1988 .
m. oka , on the resolution of the hypersurface singularities .
complex analytic singularities , 405 - 436 , adv . stud .
pure math .
, 8 , north - holland , amsterdam , 1987 . m. oka , non - degenerate complete intersection singularity .
actualits mathmatiques .
hermann , paris , 1997 .
a. reguera , towards the singular locus of the space of arcs .
amer . j. math . 131 ( 2009 ) , no . 2 , 313 - 350 . b. teissier , valuations , deformations , and toric geometry .
valuation theory and its applications , vol .
ii ( saskatoon , sk , 1999 ) , 361 - 459 , fields inst . commun . , 33 , amer . math .
soc . , providence , ri , 2003 .
b. teissier , overweight deformations of affine toric varieties and local uniformization , to appear in proceedings of the second international conference on valuation theory .
varchenko , zeta - function of monodromy and newton s diagram .
( 1976 ) , no . 3 , 253 - 262 .
hussein mourtada , + equipe gomtrie et dynamique , + institut mathmatique de jussieu - paris rive gauche , + universit paris 7 , + btiment sophie germain , case 7012 , + 75205 paris cedex 13 , france .
+ email : [email protected] + camille plnat + ( cmi ) , aix marseille universit + technople chteau - gombert + 39 , rue f. joliot curie , 13453 marseille cedex 13 + email : camille.plenat at univ-amu.fr | using the structure of the jet schemes of rational double point singularities , we construct `` minimal embedded toric resolutions of these singularities '' .
we also define , for these singularities , a correspondence between a natural class of irreducible components of the jet schemes centered at the singular locus and essential divisors for `` toric embedded resolutions '' : these are divisors which appear on every `` minimal embedded toric resolution '' .
we prove that this correspondence is bijective except for the @xmath0 singulartiy .
this can be thought as an embedded nash correspondence for rational double point singularities . |
each of the known amsps have now been well - studied in followup observations , and a review of their observational properties can be found in @xcite .
six of the seven are transients with outburst intervals of 2 or more years .
however , activity of the most recently - discovered example , hete j1900.1@xmath02455@xcite , has continued long beyond the usual active interval for the other six sources .
the latest observation , on 2006 february 2 , indicates that the source is still at approximately the same flux level as it was throughout the second half of 2005 .
the source is also unusual in that the pulsations are not detected consistently while the source is active @xcite . here
we present a summary of the outburst history of the accretion - powered msps , in order to compare the long - term accretion rates .
we also introduce a new method which can give an upper limit on the distance for sources where no bursts have been detected .
we analysed observations of the amsps made with the _ rossi x - ray timing explorer _ ( _ rxte _ ) .
we used measurements of the persistent flux and peak flux of thermonuclear ( type i ) bursts ( where available ) tabled in the catalog of galloway et al .
( 2006a , in preparation ) .
the data were analysed with lheasoft version 5.3 , released 2003 november 17 .
the persistent flux was measured by averaging the integrated flux from the best - fit absorbed blackbody plus power - law model in the energy range 2.525 kev , to spectra extracted separately for each pcu .
we used a bolometric correction to the 2.525 kev flux based on absorbed comptt model fits to combined pca and hexte data .
the x - ray colors for individual amsps were relatively constant over each outburst , and so we adopted a constant correction for each source : xte j1814@xmath0338 , 1.86 ; sax j1808.4@xmath03658 , 2.12 ; xte j0929@xmath0314 , 1.80 ; xte j1751@xmath0305 , 1.66 ; xte j1807@xmath0294 , 1.57 ; and igr j00291 + 5934 , 2.54 .
we estimated the fluence for each outburst using public _ rxte _
pca and asm measurements . for intervals where the outburst was not covered by pca observations , we integrated the asm intensities instead , using a linear cross - calibration between the pca and the nearest 1-day average 210 kev asm intensities .
xte j1807@xmath0294 and xte j1751@xmath0305 lie towards the galactic center , and a cataclysmic variable is within the @xmath1 _ rxte _ field - of - view centred on igr j00291 + 5934 .
thus , for those sources we subtracted out a baseline level of 10 , 5 and @xmath2 respectively , which we attribute to contributions from diffuse background and/or unrelated field sources .
the fluences derived by this method , scaled to give the estimated bolometric values , are listed in table [ outbursts ] .
we propagated the errors from the uncertainties on individual asm / pca measurements .
we note that the calculated fluences were generally consistent with prior estimates , to within the uncertainties . the time - averaged accretion rate driven by angular momentum loss arising from gravitational radiation from the binary
is given by @xcite @xmath3 where @xmath4 is the minimum companion mass , @xmath5 the neutron star mass , and @xmath6 the binary orbital period .
because pulse timing allows measurement only of the projected semimajor axis @xmath7 , only a lower limit on @xmath4 is available .
thus , on equating the time - averaged x - ray flux @xmath8 and @xmath9 , we derived lower limits on the distance @xmath10 for the interval prior to each outburst ( table [ outbursts ] ) .
lcccccccc & & & & & & + xte j1807@xmath0294 & feb 2003 & 52681 & @xmath11 & @xmath12 & @xmath13 & 4.7 + xte j1751@xmath0305 & jun 1998 & 50978 & @xmath14 & @xmath15 & ( @xmath16 ) & ( 6.2 ) + & apr 2002 & 52363 & 3.8 & @xmath17 & 1.9 & ( 7.8 ) + xte j0929@xmath0314 & apr 2002 & 52376 & @xmath18 & @xmath19 & @xmath20 & ( 3.6 ) + sax j1808.4@xmath03658 & sep 1996 & 50333 & @xmath21 & @xmath22 & @xmath23 & 1.4 + & apr 1998 & 50911 & 1.58 & @xmath24 & 10 & 2.5 + & jan 2000 & @xmath25 & 1.74 & @xmath26 & 9.8 & 2.6 + & oct 2002 & 52559 & 2.8 & @xmath27 & 7.0 & 3.1 + & june 2005 & 53522 & 2.6 & @xmath28 & 5.9 & 3.4 + igr j00291 + 5934 & nov 1998 & 51143 & @xmath29 & @xmath15 & ( @xmath30 ) & ( 4.3 ) + & sep 2001 & 52163 & 2.8 & @xmath15 & ( 1.8 ) & ( 4.2 ) + & dec 2004 & 53341 & 3.2 & @xmath31 & 1.6 & 4.5 + xte j1814@xmath0338 & jun 2003 & 52789 & @xmath32 & @xmath33 & @xmath34 & 10.5 + for the three sources with thermonuclear bursts , independent estimates of the distance can be made from the peak burst flux . based on bursts observed by _
bepposax _ , @xcite estimated @xmath35 kpc for sax j1808.4@xmath03658or up to 3.3 kpc for a pure he burst . similarly , @xcite estimated @xmath36 kpc for hete j1900.1@xmath02455 based on a burst observed with _
hete - ii_. while the brightest burst from xte j1814@xmath0338 was not conclusively shown to exhibit radius - expansion , the implied @xmath37(10 ) kpc for @xmath38(0.0 ) @xcite .
based on these distances , or the limits from table [ outbursts ] for those sources with no detected bursts , we estimated the long - term averaged @xmath39 for each of the amsps ( fig .
[ mdot ] ) .
we note that the measured outburst fluences and intervals for sax j1808.4@xmath03658 indicate that @xmath8 ( and hence @xmath39 ) is decreasing steadily . since the distance depends on the flux only to the @xmath40 power , the derived limit varied only by 30% , up to a maximum of 2.9 kpc ( table [ outbursts ] ) .
igr j00291 + 5934 is the only other source for which multiple outbursts have been identified , and the increasing outburst interval suggests that @xmath39 may also have been decreasing with time .
markedly different behaviour has been exhibited by hete j1900.1@xmath02455 since its discovery in 2005 june @xcite .
although the source was too close to the sun for observations during 2005 december and 2006 january , activity has apparently continued for more than 8 months . while the estimated @xmath39 in outburst ( based on the approximately constant flux level of @xmath41 since 2005 june 14 @xcite ) is just 2% @xmath42 ( for @xmath36 kpc )
, continuing activity would make this the amsp with the highest average @xmath39 by far ( fig .
[ mdot ] ) . [ [ nobursts ] ] distance upper limits for non - bursting amsps + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + while thermonuclear bursts have not been detected from four of the amsps , we expect that this is because they have been missed in data gaps rather than being absent altogether , as in ( e.g. ) the high - field pulsars .
_ rxte _ is in a low - earth orbit with a period of @xmath43 min , and suffers regular interruptions when observing most of the sky due to earth occultations , as well as observations of other sources and passages through regions of high particle density , which introduce additional gaps .
the amsps be arbitrarily distant , because the implied @xmath39 would exceed the eddington limit ; we may however infer a lower limit , at which point the implied @xmath39 would be high enough to produce sufficiently frequent x - ray bursts that it would be highly improbable that they would all be missed by the _
observations .
the key factors to determine the likelihood of burst detection are the time density of observations ( duty cycle ) and the underlying burst rate , which depends in turn on @xmath39 and the h - fraction in the accreted fuel , @xmath44 .
three of the four amsps in which no bursts have been detected are in `` ultracompact '' binaries with @xmath45 min .
the roche lobes in such tiny binaries can not contain a main - sequence companion , indicating that the mass donors are evolved and ( probably ) h - poor ( e.g. @xcite ) .
the expected burst recurrence times are thus very long due to the absence of heating from persistent h - burning between 2 + * figure : distribution of time - averaged @xmath39 for the amsps .
the shaded histogram shows the distribution for sources with distances measured from the peak flux of thermonuclear bursts ; the other values are from lower limits on @xmath10 .
the estimated @xmath39 for hete j1900.1@xmath02455 in outburst , based on the reported source flux and distance , is indicated .
the top @xmath46-axis is in units of g s@xmath47 .
[ mdot ] * + * figure : combined distance - hydrogen fraction limits based on the december 2004 outburst of igr j00291 + 5934 .
the dashed line indicates the lower limit on the distance based on measurements of the time - averaged flux ( table [ outbursts ] .
the contours indicate the estimated likelihood that all the bursts were missed for each combination of @xmath44 and @xmath10 .
[ ratelim ] * the bursts ( e.g. @xcite ) .
the combination of very low expected burst rates for these sources , and low duty cycles for the pca observations ( e.g. 6.6% for xte j1807@xmath0294 ) makes it difficult to constrain the distances .
the duty cycle for the pca observations of igr j00291 + 5934 , on the other hand , was higher , at 27% . since the mass donor in this source
is also thought to be h - rich , we expect the highest burst rate of all four sources with no detected bursts , and thus is the least likely to have missed all the bursts .
we generated plausible burst sequences for igr j00291 + 5934 based on the cubic - spline interpolated flux evolution measured during the 2004 december outburst , with burst ignition conditions calculated as in @xcite , to which we refer the reader for further details .
we adopted a grid of distances beginning at the lower limits in table [ outbursts ] .
we assumed a @xmath48 neutron star with radius @xmath49 , giving a surface gravity @xmath50 and redshift @xmath51 .
we generated @xmath52 burst sequences for each value of @xmath10 and @xmath53 , 0.3 , 0.5 and 0.7 .
we fixed @xmath54 ( equivalent to solar metallicity ) throughout .
we varied the start time of the first burst evenly within the first predicted burst interval , and also introduced a modest degree of scatter on the burst times , with a standard deviation of 0.13 hr we then checked how many of the predicted burst times fell within the intervals during which the pca was observing the source .
we interpreted the fraction of trials which resulted in one or more detected bursts , as the probability that we could reject that set of parameters . at the lowest value of @xmath53 , the predicted burst rate was sufficiently low that the likelihood of missing any bursts present was high , even for distances as large as 8 kpc
however , for higher values of @xmath44 , the higher @xmath39 implied by such large distances made it increasingly unlikely that we would have missed all the bursts . thus , the likely distance limit became smaller .
since we expect the mass donor in igr j00291 + 5934 is h - rich , we expect a source distance of no more than 6 kpc ( fig .
[ ratelim ] ) .
we have derived distances or limits to the known amsps via analysis of _ rxte _ observations . based on the outburst history of the amsps ,
the next outburst expected is from xte j1751@xmath0305 , early in 2006 . in both cases where more than two outbursts are known ( sax j1808.4@xmath03658 and igr j00291 + 5934 )
, we find evidence that the long - term averaged flux is decreasing .
for the two transients with independent constraints on the distance from the peak flux of photospheric radius - expansion bursts , sax j1808.4@xmath03658 and xte j1814@xmath0338 , the maximum lower distance limit derived from equating the average flux and @xmath9 is just above the distance range derived from the bursts .
we also derived an upper limit on the distance to igr j00291 + 5934 of 6 kpc , based on the predicted burst rate and the duty cycle for the _ rxte _ observations .
given a sufficiently high observational duty cycle , this method may be used to derive distance limits on other lmxbs where bursts have not been detected . finally , we note the discovery of hete j1900.1@xmath02455 , the first `` quasi - persistent '' amsp . as of 2006 february
this source is still active , more than 8 months after its discovery .
although the estimated @xmath55% @xmath42 is significantly lower than the peak reached by most of the other transient amsps , the fact that activity is continuous indicates that the long - term time - averaged @xmath39 of this source may exceed all the others . in that case ,
hete j1900.1@xmath02455 is the best candidate for the detection of gravitational waves from an amsp .
we thank lars bildsten and philip podsiadlowski for useful discussions .
this research has made use of data obtained through the heasarc online service , provided by nasa / gsfc .
this work was supported in part by the nasa ltsa program under grant nag 5 - 9184 . | the population of accretion - powered millisecond pulsars ( amsps ) has grown rapidly over the last four years , with the discovery of six new examples to bring the total sample to seven . while the first six discovered are transients active for a few weeks every two or more years , the most recently - discovered source hete j1900.1@xmath02455 , has been active for more than 8 months .
we summarise the transient behaviour of the population to estimate long - term time - averaged fluxes , and equate these fluxes to the expected mass transfer rate driven by gravitational radiation in order to constrain the distances .
we also estimate an upper limit of 6 kpc to the distance of igr j00291 + 5934 based on the non - detection of bursts from this source .
address = school of physics , university of melbourne , vic 3010 australia |
the process of photon propagation in a strong external electromagnetic field remains topical , despite an extensive literature on this subject @xcite-@xcite .
the first theory of this process has been developed in refs .
@xcite-@xcite using polariztion operator method , and the refractive index of polarized vacuum has been found . in external electromagnetic field , vacuum becomes an optically active medium , and the effect similar to light birefringence can be observed . in the works
@xcite-@xcite considerable attention was paid to vacuum polarization and the singularity of the polarization operator in a constant magnetic field .
also , the influence of vacuum polarization on the propagation and absorption of electromagnetic waves near pulsars was considered .
it should be emphasized that in the resonant case the intermediate electron - positron pair still can annihilate to the final photon instead of producing a real electron - positron pair .
however , the divergence of the polarization operator is usually associated only with one - photon pair creation . the cascade process of pair production with subsequent annihilation requires more detailed study .
the similar process of production of an electron - positron pair and successive annihilation has been recently considered also in a plane - wave field in @xcite .
also , there is another resonance in the problem , which is the result of the formation of a bound state between the created electron and positron @xcite .
it should be noted , that the resonant divergences are typical for two vertex qed processes in external electromagnetic field . in this case , the intermediate particle goes to the mass shell @xcite-@xcite . in this work
the process of electron - positron pair production by a photon with subsequent annihilation to a single photon in a magnetic field is studied using diagram technique .
magnetic field is considered to be comparable with the critical schwinger one , @xmath0 g. such fields are believed to exist at neutron stars @xcite-@xcite .
subcritical magnetic field can be generated in heavy ion collisions with the impact parameter comparable to the compton wavelength of an electron @xcite .
the process rate has been found in the low landau levels ( lll- ) approximation and the polarization degree of the final photon has been considered in resonant and nonresonant cases . in should be noted , that the process rate contains the divergence ( dirac delta - function ) , since there is a single particle in the final state .
nevertheless , it is still possible to consider photon polarization because the delta function cancels in the expression of polarization degree .
let us consider the propagation of an arbitrary polarized photon in a uniform magnetic field .
the landau gauge is chosen of electromagnetic 4-potential , which is @xmath1 .
the feynman diagram of the process is shown in fig.[diagramma ] .
the wavy line is the photon wave function and the internal double lines are the green s functions of the virtual electron and the positron in a magnetic field .
the wave function of a photon has a standard form ( we will use natural units , where @xmath2 , throughout ) @xcite : @xmath3 where @xmath4 is the normalization volume , @xmath5 is the polarisation 4-vector of the photon : @xmath6 here , @xmath7 and @xmath8 are the azimuthal and polar angles , @xmath9 and @xmath10 are the parameters that determine a photon polarization .
the green s function of an electron in a magnetic field has the following form @xcite-@xcite @xmath11 @xmath12 + \nonumber\\ + im\sqrt{2nh}\gamma^1 \left[\tau u_{n-1 } u'_{n}-\tau^ * u_n u'_{n-1}\right ] , \end{aligned}\ ] ] where @xmath13 is magnetic field strength in the units of the quantum critical magnetic field of @xmath14 , @xmath15 are the dirac gamma matrices , @xmath16 is the hermite function , @xmath17 is the argument of @xmath16 and the primed functions in the formula ( [ gh ] ) depend on @xmath18 , @xmath19 according to standard rules of quantum electrodynamics @xcite , the amplitude of the process of photon propagation in vacuum in the presence of a magnetic field can be written as @xmath20,\end{aligned}\ ] ] where the primed wave function of the final photon depends on @xmath18 .
taking into account the expressions for the green s functions ( [ g ] ) and the photon wave functions ( [ a ] ) and carrying out corresponding mathematical transformations , it is easy to obtain the following expression for the general amplitude of the process , @xmath21 where @xmath22 have the form :
@xmath23 @xmath24 @xmath25 @xmath26 @xmath27 @xmath28 here , @xmath29 @xmath30 are the landau levels of the intermediate particles , @xmath31 are the parallel to the field components of the polarization vectors of the photons , @xmath32 is the known functions appearing in quantum electrodynamics in a magnetic field @xcite-@xcite .
the following notation is introduced : @xmath33 note that the functions @xmath34 determining the polarization properties of photons coincide with the similar functions in the process of two - photon electron - positron pair production in a magnetic field @xcite . after carrying out the integration ( [ sfig ] ) over @xmath35 ,
the amplitude of the process has the form @xmath36 note that the amplitude contains @xmath37 .
in other words , in the considered process both energy and momentum conserve , despite the presence of a magnetic field .
it should be emphasized that this is not the case for most qed processes in a magnetic field , and the amplitude normally contains only 3-delta function ( @xmath38 is absent for landau gauge of the potential ) . in the considered process
the dispersion law is satisfied for the both initial and final photons , @xmath39 and @xmath40 .
taking into account the 3-delta functions @xmath41 it is easy to find that @xmath42 .
the case @xmath43 should be rejected as non - physical , since it corresponds to reflection of the photon regardless of the magnetic field strength and the photon frequency .
hereinafter , without loss of generality , we will use a system of reference in which the parallel to the field photon momentum is absent , @xmath44 since the lorentz transformation along @xmath45 does not change the magnetic field .
it is known that the amplitude of this process has the divergence .
therefore , it is necessary to carry out the procedure of regularization or renormalization .
we will use the bogolubov regularization method @xcite .
namely , the denominator in the expression for the green s function should be replaced as follows : @xmath46 where @xmath47 is the additional mass . in the bogolubov @xmath9 representation @xcite ( which is similar to the schwinger s proper - time method @xcite ) , the integrations in ( [ sfig ] )
can be performed using the relation @xmath48 after carrying out the procedure of regularization , the general amplitude of the process is given by the formula @xmath49 here , @xmath50 is the positive small quantity .
the following notation is used : @xmath51 @xmath52 @xmath53 @xmath54)+4h^2 n_-^2},\end{aligned}\ ] ] @xmath55 and @xmath56 is the frequency of the initial photon .
let us consider the resonant case when virtual particles become real ones . in this case , both first - order poles in ( [ sfig2 ] ) coincide with each other : @xmath57 equation ( [ res ] ) has two solutions , one of which is spurious .
indeed , @xmath58 does not belong to the integration interval ( [ sfig2 ] ) , hence , the singularity in the integrand vanishes .
therefore , we will use the other solution , which has the form @xmath59 thus , the resonant frequency is equal to the sum of the electron and the positron energies with zero longitudinal momenta .
it should be noted that the same condition was obtained in @xcite when analyzing the polarization operator in the presence of a magnetic field . in the resonant case ( [ wres ] )
the amplitude of the process ( [ sfig2 ] ) takes on the form @xmath60 where @xmath61 is the width of the resonance .
hereinafter , we will use lll - approximation , @xmath62 these conditions are common in subcritical magnetic field , when the field strength approaches the critical value of @xmath14 . in this approximation in the first order in @xmath13 the resonant photon frequency
can be rewritten as : @xmath63 in the view of the above expressions , the resonant amplitude in the linear approximation on @xmath13 has the form @xmath64+hn_+ t_1 ^ - t_2^+ ) , \end{aligned}\ ] ] where @xmath65 here , @xmath66 it should be noted that based on the formula for the amplitude ( [ sfi5 ] ) the optical theorem can be derived . in other words
the rate of single - photon electron - positron pair production is determined by the imaginary part of the amplitude of the studied process . in the case of unpolarized photon ,
the optical theorem looks like @xmath67 where @xmath68 is the probability of the process of one - photon electron - positron pair creation ( see for example @xcite ) , @xmath69 is the detuning from the threshold of the one - photon pair creation . it should be emphasized that in the formula ( [ op ] ) @xmath69 goes to zero if the pair is created on fixed landau levels with zero longitudinal momenta .
it should be noted that the methods of quantum field theory ( qft ) @xcite-@xcite , in particular the optical theorem , can be applied to the problem of heavy charged particle passing through the magnetized electron gas , which is related to the method of electron cooling . in this method
the emittance of heavy charged particle beam is reduced due to collisions with electrons which have a small velocity spread @xcite-@xcite .
the energy loss of a charged particle in the first born approximation is determined by the imaginary part of the polarization operator .
it should be noted that despite the widespread usage of the electron cooling method in accelerating technology , yet there is no complete theoretical description of the difference in the friction force for positive and negative particles , which is experimentally observed @xcite .
this problem will be important for electron cooling of antiproton beams at the hight energy storage ring ( hesr ) @xcite . the difference in cooling of antiprotons and protons
can be described using qft methods in the second born approximation .
let us find the resonance rate of the process of photon propagation in a strong magnetic field .
according to the well - known qed rules the differential rate is determined as @xmath70 the integration over @xmath71 can be carried out using @xmath72-function properties .
the rate accurate to the first corrections in @xmath13 can be obtained in the form @xmath73+(\xi_1
\xi'_1+\xi_2 \xi'_2)n_+ h\right),\end{aligned}\ ] ] where @xmath74 are the stokes parameters of photons @xcite : @xmath75 @xmath76 @xmath77 it follows from ( [ wres ] ) that the rate substantially dependent on the initial photon polarization and vanish in the case of normal linear polarization ( @xmath78 ) within the accuracy of the approximation .
also , the suppression of process rate occurs in the case @xmath79 in other processes in a magnetic field as well ( for example , in the process of electron - positron pair by a single photon @xcite .
the account of the next correction in @xmath13 gives a small but non - zero rate , @xmath80 as it is known , the two vertex processes have the resonant divergences . in this case , the intermediate particle goes to the mass shell and second - order process can be represented by two subsequent first - order processes .
the resonant process of photon propagation in a magnetic field can be represented as a sequence of processes of one - photon pair creation and annihilation of pair in a single photon .
according to @xcite , in the case when the photon has normal linear polarization ( @xmath79 ) , the rate of the process of pair production by a photon is suppressed .
thus , the rate of the process ( fig .
1 ) is also suppressed when the initial photon has normally linear polarized .
it can be concluded from the expressions ( [ wres ] ) and ( [ wxi ] ) that the rate is considerably less in the later case . fig.([whres ] ) shows the dependence of the dimensionless process rate on the magnetic field in the units of the critical strength for various initial photon polarizations .
it should be noted that the width of resonance has the form @xcite
let us consider the process in the nonresonant case when the condition ( [ res ] ) is not fulfilled . in this case
, the initial photon frequency looks like @xmath82 where @xmath83 belongs to the interval @xmath84 . taking into account the general amplitude and the condition ( [ wnotres ] ) the nonresonant amplitude in lll - approximation ( [ lll ] ) takes on the form @xmath85 where @xmath86 @xmath87 similarly to the previous section , one can find the rate of the nonresonant process : @xmath88 in the same way , when the initial photon has normal linear polarization the nonresonant process rate is @xmath89 the formulas ( [ wnres ] ) , ( [ wnresxi ] ) show that the rate in the case of normal linear polarization of the initial photon is much less then in other cases . according to the expressions ( [ wres ] ) , ( [ wnres ] )
, the ratio of the resonant and nonresonant rates can be found as @xmath90 it follows from the above expressions that the process rate is substantially greater in the resonant case compared to the nonresonant one .
let us find the polarization of the final photon when the resonance condition ( [ res ] ) is fulfilled . by the definition
, the degree of polarization of the final photon can be written as @xmath91 after developing eq .
( [ p1 ] ) into a power series and keeping terms linear in @xmath13 , the degree of polarization takes on the form @xmath92 equality ( [ p2 ] ) shows that the degree of polarization is almost independent on the polarization of the initial photon .
taking into account the relation ( [ p2 ] ) , the stokes parameters of the final photon can be written as @xmath93 thus , the final photon polarization is almost always linear anomalous and @xmath94 .
the case @xmath95 is the only exception . according to the expression ( [ wxi ] ) , the polarization of the final photon is @xmath96 .
therefore , the photon propagates in a magnetic field without changing the polarization when the initial polarization is normal linear one .
it should be noted that the photon polarization shows similar behavior in the nonresonant case .
thus , if the photon is polarized as an eigenmode ( normal and anomalous mode polarizations are @xmath79 and @xmath97 , respectively ) , its polarization does not change after traveling through the magnetic field , in accordance with known results [ 1 - 3 ] .
in the present work the process of photon propagation in a subcritical magnetic field has been considered .
the photon polarization is assumed to have arbitrary values .
the obtained expressions of the process rate have been analyzed in the lll - approximation .
considerable attention has been paid to the resonant case ( [ wres ] ) when the virtual electron - positron pair goes to the mass shell .
it should be emphasised that the resonant photon propagation with production and annihilation of electron - positron pair is possible when the condition ( [ wres ] ) is true . in the previous works ,
however , the resonance was associated only to the single - photon production of a real pair .
the expressions for the process rate in the resonant and nonresonant cases substantially depend on the initial photon polarization . in the case of normal linear polarization ( @xmath78 )
the rate has the minimum value .
it was found from the consideration of the polarization degree , that the final photon has anomalous linear polarization ( @xmath98 ) in most cases .
the exception is the case of normal linear polarization @xmath95 , when the photon propagates without changing of the polarization ( @xmath99 ) .
thus , vacuum in a magnetic field does not show optical activity when the photon polarization @xmath100 . | the process of electron - positron pair production by a photon with subsequent annihilation to a single photon in a strong magnetic field has been studied .
the general amplitude has been calculated and the process rate have been found in the low landau levels approximation ( resonant and nonresonant cases ) .
the comparison of resonant and nonresonant cases shows a significant excess of the resonant rate .
the polarization of the final photon in a strong magnetic field has also been found .
it has been shown that polarizations of the initial and the final photons are independent except for the case of normal linear polarization of the initial photon . _
2016 laser phys .
* 26 * 066001 ; _ * * doi:**10.1088/1054 - 660x/26/6/066001 |
two states , @xmath7 and @xmath8 , have been discovered recently both in continuum @xmath1 annihilation near @xmath9 gev/@xmath10 and in @xmath11 meson decays @xcite .
their spin - parities are , respectively , @xmath12 and @xmath13 @xcite , and they are presumed to be p - wave excited @xmath14 states with @xmath15 . here , @xmath16 is the orbital angular momentum and @xmath17 is the spin of the light antiquark . however , their masses are unexpectedly low @xcite .
this has renewed interest in measurements of p - wave excited charm mesons .
we report the first observation of the decay @xmath18 .
( the inclusion of charge - conjugate modes is implied throughout this paper . )
the @xmath19 pair in the final state is the only @xmath20 combination that can not come from a @xmath21 resonance : @xmath22 mesons can only be produced virtually here since @xmath23 .
the new @xmath24 mode reported here is only the second observed three - body decay of the @xmath25 , after @xmath26 @xcite . in addition , we have performed an angular analysis of the @xmath27 mode .
heavy quark effective theory ( hqet ) predicts that for an infinitely heavy @xmath28-quark this decay of a @xmath29 , @xmath30 state should proceed via a pure d - wave @xcite .
the corresponding decay of its partner , the @xmath8 , which is believed to be a @xmath13 , @xmath31 state is energetically forbidden , but if it were allowed it would proceed via a pure s - wave . since heavy quark symmetry is not exact , the two @xmath13 states can mix with each other , @xmath32 where @xmath33 and @xmath34 denote the states with @xmath31 and @xmath30 , respectively . note that the coupling via common decay channels can give a contribution to the mixing that might not be well represented by an orthogonal rotation @xcite .
we neglect this possibility in the expression above .
if @xmath35 , an s - wave component can appear in the decay @xmath36 .
moreover , even if @xmath37 is small , the s - wave component can give a sizeable contribution to the width because the d - wave contribution is strongly suppressed by the small energy release in the @xmath36 decay .
the first attempt to decompose s- and d - waves in the analogous decays of the non - strange mesons @xmath38 and @xmath39 was reported more than ten years ago by cleo @xcite ; currently , no results on the @xmath25 exist .
moreover , cleo s method did not allow the measurement of the ratio of partial widths : it only determined the relation between this ratio and the relative phase between the s- and d - wave amplitudes .
some information on @xmath37 is obtained from the ratio of electromagnetic decay rates @xmath40 , since only the @xmath41 state in @xmath8 undergoes an e1 transition to @xmath42 and only the @xmath43 state to @xmath44 @xcite .
the bases @xmath45 and @xmath46 are related by the rotation angle @xmath47 , where @xmath48 .
the angle between the bases @xmath49 and @xmath46 is @xmath50 .
the belle collaboration studied @xmath40 decays using @xmath8 from both @xmath11 decays @xcite and from @xmath1 annihilation @xcite , and determined the ratio of decay rates to be @xmath51 and @xmath52 , respectively . according to ref .
@xcite , the average ratio of @xmath53 gives the constraint @xmath54 .
detailed knowledge of the mixing is important to test different theoretical models @xcite , to fix their parameters and to understand better the nature of @xmath55 mesons .
finally , using the @xmath56 mode , we have measured the spin alignment of high momentum @xmath25 mesons produced in @xmath1 annihilation .
production of excited mesons in the hqet framework is described in ref .
the fragmentation process is assumed to be so rapid that the color magnetic forces do not have time to act and thus the spin of the light antiquark in the produced meson is uncorrelated with that of the heavy quark .
one consequence of this is that @xmath21 mesons with @xmath57 are produced unpolarized .
this was confirmed with good accuracy by cleo @xcite in @xmath58 events at @xmath59 gev and was also checked by other experiments @xcite .
another prediction is that @xmath21 and @xmath60 mesons are produced according to the number of available helicity states in a 3:1 ratio .
however , experimental data from several different production mechanisms ( @xmath1 , hadroproduction , photoproduction , etc . ) give an average probability for an s - wave meson to be produced in a vector state of @xmath61 @xcite , which is much smaller than the expected value of 0.75 .
there are no similar measurements for the p - wave states .
contrary to the @xmath62 case , hqet predicts that the members of @xmath30 doublet can be produced aligned .
the probabilities for the light degree of freedom to have helicity @xmath63 are expressed via one parameter @xmath64 as @xmath65 , respectively . by adding the @xmath28-quark spin and resolving the @xmath14 system into @xmath13 and @xmath66 states
, one can calculate their alignment .
for @xmath25 the probability of zero helicity is @xmath67 .
a calculation based on perturbative qcd and a nonrelativistic quark model gives @xmath68 @xcite and @xmath69 .
this calculation also predicts the dependence of @xmath64 on the longitudinal momentum fraction and on the transverse momentum of the meson relative to the heavy quark jet .
the argus analysis of the angular distributions in @xmath70 decay @xcite gives an upper limit @xmath71 at 90% cl @xcite .
once @xmath64 is measured , one can make definite predictions for the angular distributions of the remaining @xmath30 meson decays and check the validity of hqet .
this study is based on a data sample of 462 fb@xmath0 collected near the @xmath72 resonance with the belle detector at the kekb asymmetric - energy @xmath1 ( 3.5 on 8 gev ) collider @xcite .
the belle detector is a large - solid - angle magnetic spectrometer that consists of a silicon vertex detector , a 50-layer central drift chamber , an array of aerogel threshold cherenkov counters , a barrel - like arrangement of time - of - flight scintillation counters , and an electromagnetic calorimeter comprised of csi(tl ) crystals located inside a superconducting solenoid coil that provides a 1.5 t magnetic field .
an iron flux - return located outside the coil is instrumented to detect @xmath73 mesons and to identify muons .
the detector is described in detail elsewhere @xcite .
two inner detector configurations were used . a 2.0 cm beampipe and a 3-layer silicon vertex detector
were used for the first sample of 155 fb@xmath0 , while a 1.5 cm beampipe , a 4-layer silicon detector and a small - cell inner drift chamber were used to record the remaining 307 fb@xmath0 @xcite .
@xmath74 and @xmath75 candidates are required to originate from the vicinity of the event - dependent interaction point . to identify kaons , we combine the ionization energy loss ( @xmath76 ) from the central drift chamber , time of flight and cherenkov light yield information for each track to form kaon and pion likelihoods @xmath77 and @xmath78 , respectively , @xcite and then impose the requirement @xmath79 .
this requirement has 98% ( 97% ) efficiency for a kaon from @xmath25 ( kaon from @xmath60 ) and a 12% ( 17% ) misidentification probability for a pion with the same momentum .
all unused tracks , whether identified as a kaon or not , are treated as pion candidates in what follows .
@xmath80 candidates are reconstructed via the @xmath81 decay channel , with a mass within @xmath82 of the nominal @xmath80 mass ( among other quality requirements ) . @xmath83 and @xmath84 mesons
are reconstructed using @xmath85 , @xmath86 , @xmath87 and @xmath88 , @xmath89 decay modes , respectively .
all combinations with masses within @xmath90 mev/@xmath10 of the nominal @xmath60 mass are selected ( 99% efficiency ) ; a mass and vertex constrained fit is then applied .
@xmath91 mesons are reconstructed using the @xmath92 mode .
the slow @xmath93 momentum resolution is degraded by multiple scattering , but is improved by a track refit procedure in which the @xmath93 origin point is constrained by the intersection of the @xmath83 momentum and the known @xmath1 interaction region .
the @xmath92 mass is required to be within @xmath94 mev/@xmath10 of the @xmath91 nominal value , which corresponds to 98% efficiency .
a @xmath91 mass constraint is not imposed .
instead , we characterize the @xmath95 candidate using the mass difference @xmath96 , where the error in the @xmath91 momentum nearly cancels out . for the @xmath97 decay mode ,
the track refit procedure described above is applied to the pion and kaon momenta , and the @xmath95 is characterized by the @xmath98 mass .
it is known that the momentum spectrum of the excited charm resonances from continuum @xmath1 annihilation is hard .
in addition , due to the strong magnetic field in the belle detector , the reconstruction efficiency for slow @xmath75 and @xmath99 mesons rises with @xmath25 momentum .
therefore , we require @xmath100 for the scaled momentum @xmath101 , defined as the ratio @xmath102 . here , @xmath103 is the momentum of the @xmath25 candidate in the @xmath1 center - of - mass frame , while @xmath104 is the momentum in this frame for a candidate carrying all the beam energy .
this selection also removes @xmath25 mesons produced in the decays of @xmath11 mesons . in monte carlo ( mc ) simulation ,
@xmath25 mesons from @xmath1 annihilation , particle decays and the detailed detector response are simulated using the pythia , evtgen and geant packages @xcite , respectively .
the @xmath83 and @xmath84 decay modes used in reconstruction are generated with their resonant substructures taken from the particle data group ( pdg ) compilation @xcite but neglecting any interference effects .
the @xmath25 momentum spectrum as measured with the @xmath56 decay mode is used for mc generation .
as shown below , no clear resonant substructure is visible in the decay @xmath18 .
therefore , this mode is simulated as a three - body phase space decay .
@xmath56 decays are generated according to our measured @xmath25 polarization and d / s - wave interference .
the mass @xmath105 ( upper plot ) and the mass difference @xmath106 ( lower plot ) for all accepted combinations are shown in fig .
the pdg superscript denotes the nominal mass value from ref .
a clear peak for the new decay channel @xmath18 is visible .
the mass spectrum of the wrong sign combinations @xmath107 shown by the hatched histogram has no enhancement in the @xmath25 region .
mass spectra for @xmath108 ( top ) and @xmath109 ( bottom ) decay modes .
the hatched histogram in the top plot shows the corresponding spectrum of wrong sign @xmath107 combinations .
the fit is described in the text .
the fit results are listed in table [ yields].,scaledwidth=54.0% ] to obtain the number of @xmath25 decays , each of the distributions in fig .
[ m ] is fit to the sum of two gaussians with a common mean ( but not necessarily common between the two decay modes ) . to ensure that the second gaussian is always wider than the first one , its width is chosen to be of the form @xmath110 .
the position of the peak , @xmath111 , @xmath112 , the fraction of events in the first gaussian and the total number of events in two gaussians are allowed to vary in the fit .
the background for the three - body @xmath108 ( two - body @xmath113 ) mode is parameterized by a second ( first ) order polynomial multiplied by the threshold function @xmath114 ( @xmath115 ) , where @xmath116 is the sum of the nominal masses of final state particles @xmath117 @xcite .
table [ yields ] contains the fit results together with the parameters of the gaussians obtained from mc simulation .
there is a small fraction of events that contribute two entries to the @xmath25 signal region in the mass plot .
the excess of such events in comparison with the same number averaged over the left and the right sideband is @xmath118 and @xmath119 for the @xmath120 and @xmath113 modes , respectively .
these values are estimates of the contribution of double counted signal events . in the branching ratio calculation
they are subtracted from the yields given in table [ yields ] . the signal and the sidebands are defined as @xmath121 mev/@xmath10 , @xmath122 mev/@xmath10 @xmath123 mev/@xmath10 , respectively , where @xmath124 is measured relative to the peak position obtained from the fit . @l@@c@@c@@c@ @c@@c@ & yield & narrow gaussian & @xmath111 & @xmath112 & @xmath125 + & ( _ efficiency _ ) & fraction & @xmath126 & @xmath126 & @xmath126 + @xmath120 , data & @xmath127 & @xmath128 & @xmath129 & @xmath130 & @xmath131 + @xmath120 ,
mc & @xmath132 & @xmath133 & @xmath134 & @xmath135 & @xmath136 + @xmath113 , data & @xmath137 & @xmath138 & @xmath139 & @xmath140 & @xmath141 + @xmath113 , mc & @xmath142 & @xmath143 & @xmath144 & @xmath145 & @xmath146 + the ratio of branching fractions is found to be @xmath147 where the first error is statistical and the second is systematic .
this ratio takes into account the partial branching fractions for @xmath148 mesons to the selected final states and of neutral kaons to @xmath81 .
the systematic error receives contribution from the sources listed in table [ syst ] .
a possible difference between the data and mc simulation in evaluation of the tracking efficiency is estimated using partially reconstructed @xmath91 decays .
the reconstruction efficiency errors for slow @xmath99 , @xmath75 and @xmath149 are added linearly .
the uncertainty in the kaon particle identification is estimated using @xmath91 decays .
uncertainty in the ratio of @xmath84 and @xmath83 efficiencies is conservatively estimated by a comparison of different decay modes used in the reconstruction .
one of the largest contributions to the systematic uncertainty arises due to the model of the background .
it is evaluated by fitting the wrong sign @xmath107 subtracted spectrum , which contains almost no background . for the @xmath108 mode , the efficiency is almost independent of the @xmath19 , @xmath150 masses and the angular distribution of decay products . therefore , the possible difference between the simplified phase space mc model and the real @xmath24 decay results in a small uncertainty in the efficiency determination .
it is estimated by comparing the yields of events using either an average or differential efficiency in the @xmath25 decay angles and the @xmath19 and @xmath150 masses .
the total systematic error is found to be 11.2% ( table [ syst ] ) .
@l@ @c@ source & uncertainty , % + reconstruction efficiencies of slow @xmath75 , @xmath99 and @xmath151 from @xmath152 and @xmath25 & @xmath153 + particle identification of slow @xmath99 from @xmath25 & 1.2 + ratio of @xmath84 and @xmath83 efficiencies & @xmath154 + background model in @xmath155 spectrum & 6.5 + efficiency dependence on @xmath19 , @xmath150 masses and & + angular distribution of decay products in @xmath120 decay & 1.2 + branching ratio of intermediate resonances@xcite & 4.1 + total & 11.2 + to cross - check the results , the @xmath84 mass spectrum is plotted in fig . [ dp ] for the @xmath25 signal and sidebands .
the latter is normalized to the area of the signal interval .
the sideband subtracted plot shown in the bottom of fig .
[ dp ] is fit to a double gaussian as above and a constant background .
the resulting yield @xmath156 is consistent with the yield @xmath157 obtained from the fit of the @xmath25 mass spectrum .
the constant background level is found to be @xmath158 , which is consistent with zero .
the enhancement in the @xmath84 mass region observed in the @xmath25 sidebands ( top plot of fig .
[ dp ] ) is due to combinations of a real @xmath84 with a random @xmath159 pair in the event .
mass spectrum for the @xmath25 signal band ( @xmath160 mev/@xmath10 , open histogram in the top plot ) and the sidebands ( @xmath161 mev/@xmath10 , normalized to the signal interval , hatched histogram ) .
@xmath162 is measured relative to the peak position in the top plot of fig .
the bottom plot shows the sideband subtracted distribution .
the solid curve shows the results of the fit described in the text.,scaledwidth=54.0% ] the @xmath19 and @xmath150 mass distributions for the @xmath18 decay are shown in fig .
[ mdpi_mkpi ] .
the @xmath25 signal yield is obtained from fits to the @xmath108 mass distribution in bins of @xmath19 and @xmath150 mass .
all gaussian parameters except the total number of events are fixed in the fit to the values listed in table [ yields ] .
the position of the threshold used for the background description depends on the chosen bin .
the areas under the histograms have been normalized to unity .
the spectra are not efficiency corrected .
the dashed histograms show the corresponding mc spectra for @xmath24 decays simulated according to a phase space distribution . from this plot
, the data points do not appear to be entirely consistent with a phase space distribution , but neither do they exhibit any clear dominant resonant substructure . (
top ) and @xmath150 ( bottom ) pairs from @xmath18 decay obtained from fits to the @xmath108 mass distributions in different @xmath19 or @xmath150 mass bins .
the dashed histograms show the corresponding mc distributions for @xmath24 decays simulated according to a phase space distribution.,scaledwidth=54.0% ]
the @xmath163 decay kinematics can be described by three angles @xmath164 , @xmath165 and @xmath166 defined as shown in fig .
[ angles_scheme ] .
the angles @xmath164 and @xmath165 are measured in the @xmath167 rest frame : @xmath164 is the angle between the boost direction of the @xmath1 center - of - mass and the @xmath80 momentum , while @xmath165 is the angle between the plane formed by these two vectors and the @xmath167 decay plane . the third angle @xmath166 is defined in the @xmath91 rest frame between @xmath93 and @xmath149 . ,
@xmath165 and @xmath166 .
the first two are measured in the @xmath25 rest frame , the third in the @xmath91 frame .
`` boost '' refers to the direction of the @xmath1 center - of - mass in the @xmath25 rest frame.,scaledwidth=54.0% ] @xmath25 polarization can be described in terms of its helicity density matrix @xmath168 .
the contribution of the element @xmath168 to the decay amplitude is proportional to @xmath169 , where @xmath170 is the azimuthal rotation angle around the @xmath1 boost direction in the @xmath167 rest frame .
after integration over @xmath170 the contribution of off - diagonal elements vanishes . due to parity conservation , the three diagonal elements can be expressed in terms of the longitudinal polarization @xmath171 , i.e. , the probability that the @xmath95 helicity is zero .
the other two probabilities are both equal to @xmath172 . in the helicity formalism , the angular distribution in the decay chain @xmath56 , @xmath173 is given by @xmath174\right.\ ] ] @xmath175 \right.\ ] ] @xmath176 the formula depends on three variables : @xmath171 , @xmath177 and @xmath178 .
here @xmath179 , where @xmath180 and @xmath181 are the helicity amplitudes corresponding to the @xmath91 helicities @xmath182 and 0 , respectively .
they are related to s- and d - wave amplitudes in @xmath25 decay by @xmath183 , @xmath184 .
equation ( [ formula ] ) allows one to extract @xmath171 and @xmath185 from the @xmath25 angular distributions and to obtain @xmath186 , where @xmath187 are the partial widths of @xmath25 and @xmath188 is the phase between d- and s - amplitudes . the last interference term in eq .
( [ formula ] ) , with phase @xmath178 , vanishes after integration over any angle .
in particular , it does not appear in ref .
@xcite in the formulas for the two- and one - dimensional distributions @xmath189 and @xmath190 .
therefore , in refs .
@xcite , only @xmath191 is measured for the @xmath192 meson . this only constrains the possible ranges of @xmath193 and the phase @xmath188 . to determine them unambiguously
, one needs to measure the phase @xmath178 and to fit the whole three - dimensional @xmath194 distribution .
the probability density function ( pdf ) for the unbinned maximum likelihood fit has the form @xmath195 it includes the efficiency corrections and the contribution of the background .
the background fraction @xmath196 is estimated as the ratio of the number of entries in the sidebands and in the signal region , respectively .
the signal and the sideband regions , defined as @xmath197 mev/@xmath10 and 10 mev/@xmath19817 mev/@xmath10 , respectively , are wider than in the @xmath120 case since the background is lower .
the pdf @xmath199 , which is normalized to unity , is modelled using the sideband event distribution and the procedure described below .
@xmath200 is the mc - determined efficiency .
the average efficiency @xmath201 normalizes to unity the signal part of the pdf .
it is recalculated in every iteration of the fit procedure as @xmath202 .
the sum is taken over @xmath203 `` bins '' in a three - dimensional @xmath204 space .
the efficiency map @xmath205 is determined from mc simulation , while the integral @xmath206 of @xmath194 over each bin volume is calculated analytically .
the density @xmath207 of sideband events in the vicinity of @xmath208 is calculated as follows .
first , the three - dimensional @xmath204 space is rescaled along each axis to the unit cube .
this ensures that in the case of uniform distributions all three variables have the same `` weight '' .
then , for the given point @xmath209 , we find the volume @xmath210 ( @xmath211 ) of the smallest cube centered at this point and containing 10 ( 11 ) sideband events .
the 10th ( 11th ) event bisects a face of @xmath210 ( @xmath211 ) , so the half - weight of event 10 outside @xmath210 and the half - weight of event 11 inside @xmath211 occupy the volume @xmath212 .
therefore , we assign 10 full events to the volume @xmath213 and estimate @xmath214 . here
, @xmath215 is the original volume of the @xmath204 space and 528 is the total number of sideband events .
the resulting @xmath216 is thus normalized . to determine a systematic uncertainty due to this procedure
, we use 20 or 50 points instead of 10 .
the changes are found to be negligible compared to statistical errors ( see below ) .
the advantage of this procedure is that for any one signal event @xmath216 is always determined from 10 ( or 20 , or 50 ) sideband events .
therefore the sideband fluctuations are much smaller than fluctuations of one signal event .
this ensures the necessary degree of `` smoothness '' of the @xmath216 distribution . on the other hand ,
the typical volume @xmath217 is about @xmath218 of the whole @xmath204 space volume .
@xmath216 thus reproduces the background behaviour at this level of granularity .
a similar method is used to construct the efficiency function @xmath219 . due to the much larger mc sample , instead of @xmath217
we use the volume with 100 mc reconstructed events @xmath220 .
we then determine the number of mc events generated there , @xmath221 , and calculate the efficiency as @xmath222 .
as in the previous case , usage of the 50 or 200 closest events instead of 100 reproduces the same results within the statistical errors , and is used to determine the systematic uncertainty of this method due to the efficiency .
the three - dimensional fit of all @xmath25 signal entries to @xmath223 gives @xmath224 note that the angular distributions are sensitive only to @xmath225 , not to @xmath178 itself .
therefore @xmath178 has a @xmath226 ambiguity , and @xmath227 is determined up to complex conjugation .
the average @xmath25 longitudinal polarization in the region @xmath228 is measured to be @xmath229 .
systematic uncertainties are calculated as a sum in quadrature of the contributions listed in table [ ang_err ] .
mc simulation shows that the detector resolution in @xmath164 , @xmath165 , @xmath166 not only increases the final errors but also effectively decreases the parameter @xmath177 by 0.13 .
the corresponding correction has already been applied to the above result .
the systematic uncertainties in modelling @xmath216 and @xmath230 are estimated by varying the number of closest points as explained above , and by using different sidebands : @xmath231 plus @xmath232 ( instead of @xmath233 ) .
the errors due to statistical fluctuation of the mc sample , which is 15.4 times larger than data and has almost the same @xmath234 and @xmath171 , are calculated as @xmath235 of the statistical errors .
@l@ @c@ @c@ @c@ @c@ source & @xmath177 & @xmath178 & @xmath171 + angular resolution & 0.05 & 0.02 & 0.001 + modeling of @xmath216 and efficiency & 0.04 & 0.00 & 0.001 + different sidebands & 0.03 & 0.02 & 0.002 + mc statistics & 0.07 & 0.04 & 0.003 + total & 0.10 & 0.05 & 0.004 + the ratio of d- and s - wave amplitudes is found to be @xmath236 .
the relative phase is close to @xmath237 , @xmath238 .
one can see that , contrary to the hqet prediction , the s - wave dominates .
its contribution to the total width is @xmath239 .
the background - subtracted efficiency corrected and normalized one - dimensional projections of @xmath194 distribution , together with the fit results , are shown in fig .
the @xmath240 difference between the points and the projected fit results corresponds to a goodness - of - fit probability of about 60% . as mentioned earlier , one - dimensional projections are not sensitive to the phase @xmath178 .
they are described instead by the following formulas derived from eq .
( [ formula ] ) : @xmath241 + \cos^2\alpha(r_{\lambda}-1)(1 - 3\rho_{00})\},\ ] ] @xmath242 + 2r_{\lambda}(3\rho_{00}-1)\cos^2\beta\},\ ] ] @xmath243 , @xmath165 , @xmath244 angular distributions . projected results of the three - dimensional fit are shown by the solid curves.,scaledwidth=54.0% ] in spite of the complexity of eq .
( [ formula ] ) , it depends only quadratically on @xmath245 and only linearly on @xmath225 and @xmath171 . the efficiency entering the pdf in eq .
( [ pdf ] ) is almost constant in all three projections . when it is set to a constant value in the fit , the results change by less than 1/3 of the statistical error .
the background fraction @xmath246 is small
. therefore one does not expect any significant biases of the fit results . to quantify this statement , 1000 samples of events
are generated according to pdf eq .
( [ pdf ] ) .
each sample contains the same number of events as observed in data .
the parameters @xmath234 and @xmath171 are set to the values determined from data .
a three - dimensional fit is performed for each sample ; it is verified that the fit results are not systematically biased and the errors are estimated correctly .
the value of the overall likelihood function is measured to be worse than the one observed in data in 33% of cases . as a final check
, the fit to data is repeated in different bins of the mass recoiling against the @xmath25 , defined as @xmath247 , where all quantities are measured in the @xmath1 center - of - mass frame .
the parameters @xmath177 and @xmath178 are found to be independent of @xmath25 momentum or recoil mass within statistical errors .
the recoil mass spectrum is shown in the top half of fig .
the resolution is about 70 mev/@xmath10 at 2 gev/@xmath10 and is approximately inversely proportional to the recoil mass .
there is an indication of two - body contributions from @xmath248 where @xmath249 , @xmath44 and higher @xmath250 resonances .
this agrees with the @xmath25 polarization spectrum , shown in the bottom half of fig .
[ rm ] , which also exhibits some structure at low recoil masses .
this spectrum is obtained when @xmath177 and @xmath178 are fixed to their values determined from the overall fit . at @xmath42 mass , as one expects , the longitudinal polarization is low .
it then rises rapidly at the @xmath44 mass and eventually reaches a plateau at @xmath251 .
recoil mass spectrum ( top ) , probability @xmath171 that @xmath25 helicity is zero ( bottom ) .
the decrease at low values in the bottom plot can be attributed to the contribution of two - body @xmath248 reactions.,scaledwidth=54.0% ]
in conclusion , a new decay channel @xmath18 is observed .
the @xmath19 pair is the only @xmath252 combination that can not come from a @xmath21 resonance .
it can be produced only through the virtual @xmath22 , broad @xmath253 or @xmath254 resonances .
in addition , the @xmath108 final state can be formed by two - body decays to a @xmath84 and a virtual @xmath255 or higher @xmath256 resonance .
no clear resonant substructure is found in the @xmath108 system .
the ratio of branching fractions @xmath257 is measured to be @xmath4 .
an angular analysis of the decay @xmath163 is also performed .
since the @xmath28-quark is not infinitely heavy , hqet is violated and the @xmath25 can contain an admixture of another @xmath29 state with @xmath31 and can decay in an s - wave .
the energy release in this reaction is small .
therefore the d - wave is suppressed by the barrier factor @xmath258 , where @xmath259 is the relative momentum of @xmath25 decay products in the @xmath25 rest frame , and @xmath260 is a momentum scale characteristic of the decay .
the s - wave contribution to the total width is proportional to @xmath261 and can be sizeable even if the mixing is small . using an unbinned maximum likelihood fit to the three angles in the @xmath56 , @xmath173 decay chain , we measure ( up to a complex conjugation ) the ratio of s- and d - wave amplitudes : @xmath262 .
the s - wave dominates , and its contribution to the total width is @xmath6 .
this result allows to calculate the mixing angle in the theoretical models with a known value of parameter @xmath260 @xcite .
the spin of high momentum @xmath25 mesons produced in @xmath1 annihilation prefers to align transversely to the momentum .
the probability that a @xmath25 with @xmath228 has zero helicity is found to be @xmath229 . assuming the hqet relation @xmath67 , @xcite this implies a value of the falk
peskin parameter , @xmath263 , in this momentum region .
this value is close to the prediction of ref .
@xcite , @xmath264 , obtained for the entire momentum region , although the applicability of the perturbative qcd fragmentation model for @xmath250 mesons is questionable .
we thank the kekb group for the excellent operation of the accelerator , the kek cryogenics group for the efficient operation of the solenoid , and the kek computer group and the national institute of informatics for valuable computing and super - sinet network support .
we acknowledge support from the ministry of education , culture , sports , science , and technology of japan and the japan society for the promotion of science ; the australian research council and the australian department of education , science and training ; the national science foundation of china and the knowledge innovation program of the chinese academy of sciences under contract no .
10575109 and ihep - u-503 ; the department of science and technology of india ; the bk21 program of the ministry of education of korea , the chep src program and basic research program ( grant no .
r01 - 2005 - 000 - 10089 - 0 ) of the korea science and engineering foundation , and the pure basic research group program of the korea research foundation ; the polish state committee for scientific research ; the ministry of education and science of the russian federation and the russian federal agency for atomic energy ; the slovenian research agency ; the swiss national science foundation ; the national science council and the ministry of education of taiwan ; and the u.s.department of energy .
b. aubert _ et al . _
( babar collaboration ) , phys .
* 90 * , 242001 ( 2003 ) ; + d. besson _ et al . _
( cleo collaboration ) , phys . rev .
d * 68 * , 032002 ( 2003 ) ; + a. drutskoy _ et al . _
( belle collaboration ) , phys .
lett . * 94 * , 061802 ( 2005 ) .
s. abachi _ et al . _
( hrs collaboration ) , phys .
b * 199 * , 585 ( 1987 ) ; + s. aihara _ et al .
_ ( tpc collaboration ) , phys .
d * 43 * , 29 ( 1991 ) ; + k. ackerstaff _ et al . _
( opal collaboration ) , z. phys .
c * 74 * , 437 ( 1997 ) ; + k. abe _ et al .
_ ( sld collaboration ) , presented at the international symposium on lepton and photon interactions ( lp97 ) , hamburg , germany , 1997 , report no . slac - pub 7574 .
evtgen home page http://www.slac.stanford.edu/@xmath265lange/evtgen , + see also d. lange , nucl .
instr . and meth . a * 462 * , 152 ( 2001 ) ; + pythia , t. sjstrand _ et al .
_ , computer physics commun . * 135 * , 238 ( 2001 ) , hep - ph/0308153 ; + geant , r. brun _ et al .
_ , geant 3.21 , cern report dd / ee/84 - 1 ( 1984 ) . | using 462 fb@xmath0 of @xmath1 annihilation data recorded by the belle detector , we report the first observation of the decay @xmath2 . the ratio of branching fractions @xmath3 is measured to be @xmath4 .
we also study the angular distributions in the @xmath5 decay and measure the ratio of d- and s - wave amplitudes .
the s - wave dominates , with a partial width of @xmath6 . |
valence fluctuation phenomena in heavy - fermion ( hf ) compounds have attracted much attention because the quantum critical point ( qcp ) of the valence transition can play an important role the novel phenomena such as the superconductivity and the non - fermi liquid behaviors @xcite .
@xmath0-ybalb@xmath1 is theoretically proposed to be located very close to the qcp of the valence transition @xcite .
actually , considerable valence fluctuation was observed experimentally @xcite .
moreover , since the valence state directly reflects the degree of localization of the @xmath5-electrons in hf systems , the temperature and the magnetic field dependences of the valence are intriguing research subjects .
there are several techniques to measure the valence state , e.g. , photoemission spectroscopy , x - ray absorption spectroscopy , x - ray emission spectroscopy , and mssbauer spectroscopy .
@xmath6-edge x - ray absorption spectroscopy on the rare - earth ions is very suitable for experiments at high pressures @xcite and/or high magnetic fields @xcite , because high - energy x - rays have strong penetrating power , so an ultrahigh - vacuum condition is not necessary .
determination of the valence state can be done in very high magnetic fields of up to 40 t by x - ray absorption measurement @xcite . in the present study , we have studied the temperature dependence of yb valence in @xmath0-ybalb@xmath1 @xcite and @xmath2-ybalb@xmath1 @xcite by using @xmath6-edge x - ray absorption and emission spectroscopy .
the x - ray absorption spectra were measured in high magnetic fields up to 32 t and a small , but finite valence increase was found at 40 k.
the experiment was done at the beamline bl39xu of spring-8 , japan .
pulsed high magnetic fields were generated by using a miniature magnet @xcite . @xmath0- and @xmath2-ybalb@xmath1 crystals were powdered and mixed with epoxy resin so that the effective thickness was appropriate for obtaining an x - ray absorption intensity of @xmath7 , where @xmath8 and @xmath9 are the absorption coefficient and the thickness of the sample , respectively .
a single crystal of @xmath0-ybalb@xmath1 was also used for x - ray absorption and x - ray emission spectroscopy .
the details of the experimental techniques using a pulsed high field are described elsewhere @xcite .
figure 1 shows the x - ray absorption spectrum at the @xmath6-edge ( from the @xmath10 to @xmath11 transition ) in a powdered @xmath0-ybalb@xmath1 at 10 k. the spectrum of a single crystal of @xmath0-ybalb@xmath1 was also measured and found to be very similar to that of the powdered sample . the solid curve is the result of the curve - fitting analysis .
we analyzed the x - ray absorption spectra ( xas ) by using a fitting function that consists of a lorentzian and an arctangent function .
this is a standard method for analyzing the absorption edge and has previously been used in an x - ray absorption study of ybincu@xmath1 @xcite . the dashed and dot - dashed curves are the results of the curve - fitting for yb@xmath12 and yb@xmath13 states , respectively .
background is represented by a dotted line .
the main absorption peak around 8.950 kev due to the yb@xmath12 state is distinct . on the other hand ,
the absorption band around 8.945 kev due to the yb@xmath13 state is not clearly seen in the spectrum .
since the valence ( @xmath14 ) of @xmath0-ybalb@xmath1 was reported to be 2.75 at 20 k @xcite , about a 25% spectrum weight is expected to be observed .
the lifetime broadening effect of the @xmath10 core hole can be the reason the yb@xmath13 band at the low energy is not very clear .
however , as we show later , the yb@xmath13 band is confirmed to be located around 8.945 kev by the magnetic field dependence of xas .
although the @xmath14 is deduced to be 2.84 from the fitting results shown in fig . 1 , the error bar can be as large as 0.05 because of the broadness of the xas .
the solid curve is the result of the curve - fitting analysis . the dashed and dot - dashed curves are the results of the curve - fitting for the yb@xmath12 and yb@xmath13 states , respectively .
background is represented by a dotted line .
, width=377 ] therefore , we determined @xmath14 at low temperatures by another method , i.e. , x - ray emission spectroscopy that enables us to observe the xas with smaller broadening effect of the core hole @xcite .
the x - ray emission spectra ( xes ) with different excitation energies at 2 and 280 k are shown in fig .
it is found that the emission band due to yb@xmath13 state is larger at 2 k compared to that at 280 k , indicating that the valence decreases with decreasing temperature .
x - ray emission spectra at different excitation energies in a @xmath0-ybalb@xmath1 single crystal at 2 and 280 k.,width=321 ] from the analysis of the xes @xmath14 was found to be @xmath15 and @xmath16 at 2 and 280 k , respectively .
@xcite since the optical transition process of xes is more complicated than that of xas , the valence state can not be directly obtained by using the spectral weight .
the detailed analysis is explained in elsewhere @xcite and is beyond the scope of this paper .
it is found that the xes show distinct emission bands due to the yb@xmath13 and yb@xmath12 contributions .
this is because the lifetime of the @xmath17 core hole that exists in the final state of the emission process ( from the 3@xmath18 to 2@xmath19 @xmath20 transition ) is expected to be much longer than that of the @xmath10 core hole .
this high - resolution spectroscopy is more reliable than xas when the xas is broad , making it difficult to distinguish each valence state separately .
hence , we take @xmath15 for the valence value of @xmath0-ybalb@xmath1 at 2 k , and the temperature dependence of @xmath14 obtained from xas is used as the relative change . the relative change in the spectrum
is determined precisely , and the valence change is determined with much smaller error bar compared to the error bar for determination of the absolute value .
the temperature variation of @xmath14 is shown in fig .
-ybalb@xmath1 and @xmath2-ybalb@xmath1 .
the closed squares are the results of the single crystal of @xmath0-ybalb@xmath1 , and closed and open circles are the results from the powdered @xmath0- ybalb@xmath1 and @xmath2-ybalb@xmath1 samples , respectively .
open diamonds denote the results of xes in the single crystal of @xmath0-ybalb@xmath1 .
the dashed curves are guides for the eyes .
( b ) temperature dependences of the yb valence in the single crystal @xmath0-ybalb@xmath1 ( open squares ) and powdered @xmath2-ybalb@xmath1 ( open circles ) are shown together with that of the yb valence in ybagcu@xmath1 ( open triangles ) @xcite .
, width=491 ] @xmath14 of @xmath2-ybalb@xmath1 is also plotted as a function of temperature .
because we did not measure xes of @xmath2-ybalb@xmath1 , we used the literature value at 20 k ( 2.73 ) @xcite for the @xmath14 value at 5 k , and the relative change in @xmath14 with increasing temperature was deduced by the change in the xas in this work .
it is found that the valence gradually increases with increasing temperature , and the increasing trends for @xmath0- and @xmath2-ybalb@xmath1 look similar .
the occupation number of the hole in the 4@xmath5-shell ( @xmath21 ) is deduced from @xmath22 . according to the gtzwiller approximation , the deviation of @xmath21 from unity ( @xmath23 )
can be expressed by @xmath24 , where @xmath25 and @xmath26 correspond to the kinetic energy of the @xmath5 electrons with the electron interaction and that without the interaction , respectively @xcite .
an itinerant character manifests itself in the deviation of the occupation number of the @xmath5 hole from unity @xmath23 .
if we define @xmath27 as the difference of @xmath21 from that at the lowest temperature , the value @xmath27/(@xmath28 ) corresponds to the relative change in the degree of itinerancy when the temperature is increased . in fig .
3(b ) , the temperature dependences of @xmath27/(@xmath28 ) in single crystal @xmath0-ybalb@xmath1 ( open squares ) and powdered @xmath2-ybalb@xmath1 ( open circles ) are shown .
@xmath27/(@xmath28 ) in another heavy fermion compound ybagcu@xmath1 @xcite is also plotted ( open triangles ) .
ybagcu@xmath1 is one of the materials that are expected to be located in the vicinity of the qcp of the valence transition @xcite .
the scale of the horizontal axis is set differently for ybalb@xmath1 ( bottom ) and ybagcu@xmath1 ( top ) so that the temperature variation of @xmath27/(@xmath28 ) in both materials almost overlap .
the valence transition temperature ( @xmath29 ) was determined to be 42.3 k in ybagcu@xmath1 in our previous work @xcite .
if we simply assume that the temperature dependence of the valence in ybalb@xmath1 is scaled to that in ybagcu@xmath1 , the hypothetical valence transition temperature ( @xmath30 ) is 288 k for ybalb@xmath1 .
this value seems to be consistent with the fact that @xmath0-and @xmath2-ybalb@xmath1 have a large characteristic temperature @xmath31 @xmath4 250 k below which the magnetic moments are screened to form the hf state @xcite .
@xmath14 becomes about 2.96 and the local magnetic moments are restored in ybagcu@xmath1 at about 70 k @xcite . according to fig .
3(b ) , the localized magnetic moments are expected to be restored in @xmath32 and @xmath33ybalb@xmath1 at corresponding temperature around 480 k. since the characteristic temperature @xmath31 of ybalb@xmath1 is as high as 250 k @xcite , a strong magnetic field is necessary to see the effect of the magnetic field on the valence state .
figure 4 shows xas at 0 and 32 t in the lower panel and the difference spectrum between 0 and 32 t ( dxas ) in the upper panel .
the measurement temperature was 40 k. the xas seem to be almost identical at 0 and 32 t. however , the positive peak around 8.950 kev and negative peak around 8.945 kev in dxas are clear evidence of an increase in the yb@xmath12 component and a decrease in the yb@xmath13 component , respectively .
the solid curve is the result of the curve fitting .
the agreement of the fitting and the experimental data confirms that a small valence increase , @xmath34 , is induced by a magnetic field of 32 t. although a similar measurement was made at 5 k for the same sample no clear valence change was observed .
therefore , the observed field - induced small valence change was assisted by a thermal excitation at 40 k. -ybalb@xmath1 sample at 0 and 32 t ( lower panel ) .
the difference spectrum between the xas at 0 and 32 t is shown in the upper panel , together with the result of the curve fitting .
a magnetic field is applied parallel to the c - axes of the powders that were aligned by means of a steady magnetic field of 14 t when the epoxy resin was solidified .
the measurement temperature was 40 k. , width=377 ]
xas and xes in @xmath0- and @xmath2-ybalb@xmath1 were investigated at several temperatures and a high magnetic field of 32 t. the observed yb valence is 2.78 @xmath3 0.01 in @xmath0-ybalb@xmath1 at 2 k by using xes .
the yb valences of these materials are found to increase with increasing temperature and by applying a magnetic field . by comparison with another heavy fermion compound ybagcu@xmath1 , the hypothetical valence transition temperature in @xmath0- and @xmath2-ybalb@xmath1
is expected to be about 288 k ; this value is close to the characteristic temperature @xmath31 of these materials as heavy fermion systems .
experiments at fields above 100 t are required to observe more distinct field - induced valence changes .
magnetization experiments in the 100-t range by means of a single - turn coil technique @xcite are currently ongoing at the institute for solid state physics , university of tokyo .
+ we thank h. toyokawa and t. uruga of japan synchrotron radiation research institute / spring-8 for technical support of the pilatus detector . for the analysis of the xes results , we thank n. kanai and h. hayashi of japan women s university .
this work was partly supported by grants - in - aid for scientific research b ( 22340091 ) , scientific research a ( 22244047 ) , scientific research c ( 24540389 ) and young scientists b ( 22740208 ) provided by the ministry of education , culture , sports , science and technology ( mext ) , japan .
the synchrotron radiation experiments were performed with the approval of the japan synchrotron radiation research institute ( jasri ) ( proposal no .
2011b1090 , 2011b2097 and 2012a1173 ) .
m. okawa , m. matsunami , k. ishizaka , r. eguchi , m. taguchi , a. chainani , y. takata , m. yabashi , k. tamasaku , y. nishino , t. ishikawa , k. kuga , n. horie , s. nakatsuji and s. shin , phys .
lett . * 104 * , 247201 ( 2010 ) .
n. kanai , h. hayashi , n. kawamura , m. mizumaki , y. h. matsuda , k. kuga and s. nakatsuji , abstracts of the 15th international conference on x - ray absorption fine structure ( xafs15 ) , beijing , china , 22 - 28 july 2012 . | the valence state of yb ions in @xmath0-ybalb@xmath1 and its polymorph @xmath2-ybalb@xmath1 has been investigated by using x - ray absorption and emission spectroscopy in spring-8 at temperatures from 2 to 280 k. the observed yb valence is 2.78 @xmath3 0.01 in @xmath0-ybalb@xmath1 at 2 k by using the x - ray emission spectroscopy .
the valence is found to gradually increase with increasing temperature toward the trivalent state , and the characteristic temperature of the valence fluctuation is expected to be about 290 k. we also found a small increase in the yb valence ( @xmath4 0.002 ) by applying a magnetic field of 32 t at 40 k to @xmath0-ybalb@xmath1 . |
in recent years , there has been a remarkable development in hadron spectroscopy .
one of the most interesting observations is the evidence of the exotic baryon @xmath0 , reported first by the leps collaboration @xcite .
subsequently , a signal of another exotic state @xmath8 was observed @xcite .
the spin and parity of @xmath0 and @xmath8 are not yet determined experimentally . since these states can be constructed minimally with five valence quarks , they are called pentaquarks .
evidences of the exotic pentaquarks have been stimulating many theoretical studies @xcite . in the study of these exotic particles
, it should be important to identify other members with nonexotic flavors in the same su(3 ) multiplet which the exotic particles belong to .
this is naturally expected from the successes of @xmath9 flavor symmetry with its breaking in hadron masses and interactions @xcite . in other words
, the existence of exotic particles would require the flavor partners , if the flavor su(3 ) symmetry plays the same role as in the ordinary three - quark baryons .
an interesting proposal was made by jaffe and wilczek @xcite , based on the assumption of the strong diquark correlation in hadrons and the representation mixing of an octet ( @xmath10 ) with an antidecuplet ( @xmath11 ) .
the attractive diquark correlation in the scalar - isoscalar channel leads to the spin and parity @xmath12 for the @xmath0 . with the ideal mixing of @xmath10 and @xmath11 , in which states are classified by the number of strange and antistrange quarks , @xmath13 and @xmath14 resonances are well fit as members of the multiplet together with the @xmath0
however , it was pointed out that mixing angles close to the ideal one encountered a problem in the decay pattern of @xmath15 and @xmath16 . rather , their decays implied a small mixing angle @xcite .
this is intuitively understood by observing the broad decay width of @xmath16 and the narrow widths of @xmath15 and @xmath17 @xcite .
employing the @xmath10-@xmath11 mixing scenario , here we examine the possibilities to assign other quantum numbers , such as @xmath18 , @xmath19 , @xmath20 , and search the nucleon partners among the known resonances . for convenience , properties of relevant resonances
are summarized in appendix [ sec : expinfo ] .
the present study is based on the flavor su(3 ) symmetry , experimental mass spectra and decay widths of the @xmath0 , the @xmath8 and known baryon resonances .
hence , our analysis presented here is phenomenological , but does not rely upon any specific models .
for instance , we do not have to specify the quark contents of the baryons .
although the exotic states require minimally five quarks , nonexotic partners do not have to .
instead , we expect that the resulting properties such as masses and decay rates reflect information from which we hope to learn internal structure of the baryons .
first we briefly discuss the case where the @xmath0 belongs to the pure @xmath11 without mixing with other representations . in this case , the masses of particles belonging to the @xmath11 can be determined by the gell - mann okubo ( gmo ) mass formula with equal splitting @xmath21 where @xmath22 is the hypercharge of the state , and @xmath23 denotes the mass matrix .
note that at this point the spin and parity @xmath24 are not yet specified .
this will be assigned as explained below . in eq .
, there are two parameters , @xmath25 and @xmath26 , which are not determined by the flavor su(3 ) symmetry .
however , we can estimate the order of these parameters by considering their physical meanings .
for instance , in a constituent quark model , @xmath11 can be minimally expressed as four quarks and one antiquark .
therefore , @xmath25 should be larger than the masses of three - quark baryons , such as the lowest - lying octet baryons . in this picture
, the mass difference of @xmath27 and @xmath28 , namely @xmath29 , should be the constituent mass difference of the @xmath30 and the @xmath31 quarks , which is about 100 - 250 mev @xcite .
on the other hand , in the chiral quark soliton model , @xmath29 is related to the pion nucleon sigma term @xcite . in this picture @xmath29
can take values in the range of 300 - 400 mev , due to the experimental uncertainty of the pion nucleon sigma term @xmath3264 - 79 mev @xcite .
note that in the chiral quark model , spin and parity are assigned as @xmath12 for the antidecuplet .
taking into account the above estimation , we test several parameter sets fixed by the experimentally known masses of particles .
the results are summarized in table [ tbl : bar10result ] .
first , we determine the parameters by accommodating @xmath2 and @xmath33 in the multiplet . in this case
we obtain the mass of the @xmath34 and @xmath35 states at 1647 and 1753 mev , respectively .
since these values are close to the masses of the @xmath18 baryons @xmath36 and @xmath37 , we expect their spin and parity to be @xmath38 .
for @xmath12 , we adopt the @xmath13 as the nucleon partner , and predict the @xmath35 and @xmath39 states .
this assignment corresponds to the original assignment of the prediction @xcite . for @xmath40 ,
we pick up @xmath41 , and for @xmath4 , n(1700 ) . in the three cases of @xmath42 , the exotic @xmath39 resonance
is predicted to be higher than 2 gev , and the inclusion of @xmath33 in the same multiplet seems to be difficult .
furthermore , the @xmath35 states around 1.8 - 1.9 gev are not well assigned ( either two - star for @xmath12 , or not seen for @xmath43 ) .
therefore , fitting the masses in the pure antidecuplet scheme seems to favor @xmath38 .
next we study the decay width of the @xmath44 resonances with the above assignments . for the decay of a resonance @xmath45
, we define the dimensionless coupling constant @xmath46 by @xmath47 where @xmath48 is the relative three momentum of the decaying particles in the resonance rest frame , @xmath49 and @xmath50 are the decay width and the mass of the resonance @xmath45 .
@xmath51 is the isospin factor , which takes the value 2 for @xmath5 and 3 for @xmath52 . assuming flavor @xmath9 symmetry , a relation between the coupling constants of @xmath53 and @xmath52 is given by : @xmath54 here we adopt the definition of the coupling constant in ref .
note that this definition is different from refs .
@xcite , in which @xmath55 is used . with these formulae and
, we calculate the decay width of the @xmath0 from those of @xmath52 of the nucleon partner .
results are also shown in table [ tbl : bar10result ] .
we quote the errors coming from experimental uncertainties in the total decay widths and branching ratios , taken from the particle data group @xcite .
it is easily seen that as the partial wave of the two - body final states becomes higher , the decay width of the resonance becomes narrower , due to the effect of the centrifugal barrier .
considering the experimental width of the @xmath0 , the results of @xmath4 , @xmath19 , @xmath56 are acceptable , but the result of the @xmath38 case , which is of the order of hundred mev , is unrealistic .
.summary of section [ sec : bar10 ] .
mass spectra and @xmath0 decay width are shown for several assignments of quantum numbers .
for @xmath18 the masses of @xmath58 and @xmath39 are the input parameters , while for @xmath59 , the masses of @xmath58 and @xmath34 are the input parameters
. values in parenthesis are the predictions , and we show the candidates to be assigned for the states .
all values are listed in units of mev . [ cols="^,^,^,^,^,<",options="header " , ] leps , t. nakano _ et al .
lett . * 91 * , 012002 ( 2003 ) .
na49 , c. alt _ et al .
rev . lett . * 92 * , 042003 ( 2004 ) .
d. p. roy , j. phys .
g * 30 * , r113 ( 2004 ) .
zhu , int .
a * 19 * , 3439 ( 2004 ) .
m. oka , prog .
. phys . * 112 * , 1 ( 2004 ) .
d. diakonov , hep - ph/0406043 .
r. l. jaffe , hep - ph/0409065 .
s. sasaki , hep - lat/0410016 .
k. goeke , h .- c .
kim , m. praszalowicz , and g .- s .
yang , hep - ph/0411195 [ prog .
( to be published ) ]
. j. j. de swart , rev .
* 35 * , 916 ( 1963 ) .
r. l. jaffe and f. wilczek , phys .
91 * , 232003 ( 2003 ) .
t. d. cohen , phys .
d * 70 * , 074023 ( 2004 ) .
s. pakvasa and m. suzuki , phys .
d * 70 * , 036002 ( 2004 ) .
m. praszalowicz , ann .
( berlin ) * 13 * , 709 ( 2004 ) .
a. hosaka , t. hyodo , f.j .
llanes - estrada , e. oset , j. r. pelez , + and m. j. vicente vacas , hep - ph/0411311 [ phys .
c ( to be published ) ] .
p. schweitzer , eur .
j. a * 22 * , 89 ( 2004 ) .
d. diakonov and v. petrov , phys .
d * 69 * , 094011 ( 2004 ) .
m. praszalowicz , hep - ph/0410241 . j. r. ellis , m. karliner , and m. praszalowicz , j. high energy phys .
05 ( 2004 ) 002 .
d. diakonov , v. petrov , and m. v. polyakov , z. phys .
a * 359 * , 305 ( 1997 ) .
s. h. lee , h. kim , and y. oh , hep - ph/0402135 .
particle data group , s. eidelman _
et al . _ ,
b * 592 * , 1 ( 2004 ) .
v. guzey and m. v. polyakov , hep - ph/0501010 .
f. stancu , phys .
b * 595 * , 269 ( 2004 ) , erratum - ibid , b * 598 * , 295 ( 2004 ) a. hosaka , m. oka , and t. shinozaki , hep - ph/0409102 . s. takeuchi and k. shimizu , hep - ph/0410286 .
t. inoue , v. e. lyubovitskij , t. gutsche , and a. faessler , hep - ph/0407305 .
y. kanada - enyo , o. morimatsu , and t. nishikawa , hep - ph/0404144 [ phys .
d ( to be published ) ] .
d. e. kahana and s. h. kahana , phys .
d * 69 * , 117502 ( 2004 ) .
s. sarkar , e. oset , and m. j. v. vacas , nucl - th/0404023 .
t. nishikawa , y. kanada - enyo , y. kondo , and o. morimatsu , hep - ph/0411224 . n. isgur and g. karl , phys .
d * 18 * , 4187 ( 1978 ) .
s. sarkar , e. oset , and m. j. v. vacas , nucl - th/0407025 .
e. e. kolomeitsev and m. f. m. lutz , phys .
b * 585 * , 243 ( 2004 ) .
e. jenkins and a. v. manohar , j. high energy phys .
06 ( 2004 ) 039 .
m. e. wessling , phys .
b * 603 * , 152 ( 2004 ) .
d. pirjol and c. schat , phys .
d * 71 * , 036004 ( 2005 ) .
t. hyodo , a. hosaka , and e. oset , phys .
b * 579 * , 290 ( 2004 ) .
k. nakayama and k. tsushima , phys .
b * 583 * , 269 ( 2004 ) .
a. w. thomas , k. hicks , and a. hosaka , prog .
. phys . * 111 * , 291 ( 2004 ) .
kim , and s. h. lee , nucl .
phys . * a745 * , 129 ( 2004 ) . c. hanhart
_ et al . _ ,
b * 590 * , 39 ( 2004 ) .
s. i. nam , a. hosaka , and h. c. kim , phys .
b * 602 * , 180 ( 2004 ) .
y. n. uzikov , phys .
b * 595 * , 277 ( 2004 ) .
k. nakayama and w. g. love , phys .
c * 70 * , 012201(r ) ( 2004 ) . c. hanhart , j. haidenbauer , k. nakayama , and u. g. meissner , phys .
b * 606 * , 67 ( 2005 ) .
w. roberts , phys .
c * 70 * , 065201 ( 2004 ) .
y. n. uzikov , nucl - th/0411113 .
y. oh , k. nakayama , and t. s. h. lee , hep - ph/0412363 . | we examine several assignments of spin and parity for the pentaquark @xmath0 state ( @xmath1 ) in connection with known baryon resonances . assuming that the @xmath0 belongs to an antidecuplet representation which mixes with an octet , we calculate the mass spectra of the flavor partners of the @xmath0 based on the su(3 ) symmetry .
the decay widths of the @xmath0 and nucleon partners are analyzed for the consistency check of the mixing angle obtained from the masses .
it is found that a suitable choice of the mixing angle successfully reproduces the observed masses of @xmath2 and @xmath3 , when their spin and parity are assigned to be @xmath4 , together with other @xmath4 resonances .
the decay widths of @xmath5 , @xmath6 , and @xmath7 are also reproduced simultaneously . |
Subsets and Splits
No saved queries yet
Save your SQL queries to embed, download, and access them later. Queries will appear here once saved.